The

Psychology
of Safety
HANDBOOK
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page ii
 The
Psychology
of Safety
HANDBOOK
E. SCOTT GELLER

LEWIS PUBLISHERS
Boca Raton London New York Washington, D.C.
820049_CRC1_L1540_FM 11/13/00 10:08 AM Page iv

Library of Congress Cataloging-in-Publication Data

Geller, E. Scott, 1942-

The psychology of safety handbook / E. Scott Geller.--2nd ed.
p. cm.
Includes bibliographical references and index.
ISBN 1-56670-540-1
1. Industrial safety-Psychological aspects. I. Title.

T55.3.B43 G45 2000

658.3′82′01—dc21 00-063750

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with per-
mission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reli-
able data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for
the consequences of their use.

Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, includ-
ing photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in
writing from the publisher.

The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or
for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying.

Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identifi-
cation and explanation, without intent to infringe.

© 2001 by CRC Press LLC

Lewis Publishers is an imprint of CRC Press LLC

No claim to original U.S. Government works

International Standard Book Number 1-56670-540-1
Library of Congress Card Number 00-063750
Printed in the United States of America 1 2 3 4 5 6 7 8 9 0
Printed on acid-free paper
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page v

Dedication

Past
To my mom (Margaret J. Scott) and dad (Edward I. Geller) who taught me
the value of learning and reinforced my need to achieve.
To B. F Skinner and W. Edwards Deming who developed and researched
the most applicable principles in this text and inspired me to teach
them.

Present
To my wife (Carol Ann) and mother-in-law (Betty Jane) whose continuous
support for over 30 years made preparation to write this book possible.
To the students and associates in our university Center for Applied
Behavior Systems whose data collection and analysis provided
practical examples for the principles.

Future
To my daughters (Krista and Karly) who I hope will someday experience
the sense of accomplishment I feel by completing this Handbook.
To my eight associates at Safety Performance Solutions, Inc., who I hope
will continuously improve their ability to assist others worldwide in
achieving a Total Safety Culture.
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page vi
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page vii

Preface
Psychology influences every aspect of our lives, including our safety and health; and psy-
chology can be used to benefit almost every aspect of our lives, including our safety
and health.
So what is “psychology” anyway?
My copy of The American Heritage Dictionary (Second College Edition, Copyright 1991
by Houghton Mifflin Company) defines psychology as

1. The science of mental processes and behavior.

2. The emotional and behavioral characteristics of an individual, group, or activity
(page 1000).

Similarly, the two definitions in the New Merriam-Webster Dictionary (Copyright 1989 by
Merriam-Webster, Inc.) are

1. The science of mind and behavior.

2. The mental and behavioral characteristics of an individual or group (page 587).

In both dictionaries, the first definition of “psychology” uses the term “science” and
refers to behavioral and mental processes. Behaviors are the outside, objective, and observ-
able aspects of people; mental or mind reflects our inside, subjective, and unobservable
characteristics. Science implies the application of the scientific method or the objective and
systematic analysis and interpretation of reliable observations of natural or experimental
phenomena.
So what should you expect from a Handbook on the Psychology of Safety? Obviously, such
a book should show how psychology influences the safety and health of people. To be use-
ful, it should explain ways to apply psychology to improve safety and health. This is, in
fact, my purpose for writing this text—to teach you how to use psychology to both explain
and reduce personal injury.
As a science of mind and behavior, psychology is actually a vast field of numerous sub-
disciplines. Areas covered in a standard college course in introductory psychology, for
example, include research methods, physiological foundations, sensation and perception,
language and thinking, consciousness and memory, learning, motivation and emotion,
human development, intelligence, personality, psychological disorders, treatment of men-
tal disorders, social thought and behavior, environmental psychology, industrial/organi-
zational psychology, and human factors engineering. This book does not cover all of these
areas of psychology, only those directly relevant to understanding and influencing safety-
related behaviors and attitudes. In addition, my coverage of information within any one
subdiscipline of psychology is not comprehensive but focuses on those aspects directly rel-
evant to reducing injury in organizational and community settings.
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page viii

This information will help you improve safety and health in any setting, from your
home to the workplace and every community location in between. You can apply the knowl-
edge gained from reading this book in all aspects of your daily life. Most organized safety-
improvement efforts occur in work environments, however, because that is where the
exposure to hazardous conditions and at-risk behavior is most obvious. As a result, most
(but not all) of my illustrations and examples use an industrial context. My hope is that you
will see direct relevance of the principles and procedures to domains beyond the workplace.
A psychology of safety must be based on rigorous research, not common sense or intu-
ition. This is what science is all about. Much of the psychology in self-help paperbacks,
audiotapes, and motivational speeches is not founded on programmatic research but is
presented because it sounds good and will “sell.” The psychology in this Handbook was not
selected on the basis of armchair hunches but rather from the relevant research literature.
In sum, the information in this Handbook is consistent with a literal definition of its title—
the psychology of safety.
The human element of occupational health and safety is an extremely popular topic at
national and regional safety conferences. Safety leaders realize that reducing injuries below
current levels requires increased attention to human factors. Engineering interventions and
government policy have made their mark. Now, it is time to include a focus on the human
dynamics of injury prevention—the psychology of safety.
Most attempts to deal with the human aspects of safety have been limited in scope.
Many trainers and consultants claim to have answers to the human side of safety, but their
solutions are too often impractical, shortsighted, or illusory. To support their particular
program, consultants, authors, and conference speakers often give unfair and inaccurate
criticism of alternative methods.
Tools from behavior-based safety have been criticized in an attempt to justify a focus
on people’s attitudes or values. In contrast, promoters of behavior-based safety have
ridiculed a focus on attitudes as being too subjective, unscientific, and unrealistic. Both
behavior- and attitude-oriented approaches to injury prevention have been faulted in order
to vindicate a systems or culture-based approach. The truth of the matter is that both
behaviors and attitudes require attention in order to develop large-scale and long-term
improvement in people’s safety and health.
There are a number of books on the market that offer advice regarding the human ele-
ment of occupational safety. Unfortunately, many of these texts offer a limited perspective.
I have found none comprehensive and practical enough to show how to integrate behav-
ior- and attitude-based perspectives for a system-wide total culture transformation. This
Handbook was written to do just that and, in this regard, it is one of a kind.
Simply put, behavioral science principles provide the basic tools and procedures for
building an improved safety system. However, the people in a work culture need to accept
and use these behavior-based techniques appropriately. This is where a broader perspec-
tive is needed, including insight regarding more subjective concepts like attitude, value,
and thought processes. Recall that psychology includes the scientific study of both mind
and behavior. Therefore, a practical handbook on the psychology of safety needs to teach
science-based and feasible approaches to change what people think (attitude) and do
(behavior) in order to achieve a Total Safety Culture.
I refer to a Total Safety Culture throughout this text as the ultimate vision of a safety-
improvement mission. In a Total Safety Culture, everyone feels responsible for safety
and pursues it on a daily basis. At work, employees go beyond “the call of duty” to iden-
tify environmental hazards and at-risk behaviors. Then, they intervene to correct them.
Safe work practices are supported with proper recognition procedures. In a Total Safety
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page ix

Culture, safety is not a priority that gets shifted according to situational demands. Rather,
safety is a value linked to all situational priorities.
Obviously, building a Total Safety Culture requires a long-term continuous improve-
ment process. It involves cultivating constructive change in both the behaviors and atti-
tudes of everyone in the culture. This book provides you with principles and procedures to
make this happen. Applying what you read here might not result in a Total Safety Culture.
However, it is sure to make a beneficial difference in your own safety and health, and in the
safety and health of others you choose to help.
I refer to helping others as “actively caring.” This book shows you how to increase
the quality and quantity of your own and others’ actively caring behavior. Indeed, actively
caring is the key to safety improvement. The more people actively caring for the safety
and health of others, the less remote is the achievement of our ultimate vision—a Total
Safety Culture.

Who should read this book?

My editor has warned me that one book can serve only a limited audience. I know he is
right but, at the same time, a practical book on reducing injuries is relevant for everyone.
All of us risk personal injury of some sort during the course of our days, and all of us can
do something to reduce that risk to ourselves and others. Therefore, a book that teaches
practical ways to do this is pertinent reading for everyone.
The average person, however, will not spend valuable time reading a handbook on
ways to reduce personal risk for injury. In fact, most people do not believe they are at risk
for personal injury, so why should they read a book about improving safety? While I
believe everyone should read this Handbook, a text on the psychology of safety is destined
for a select and elite audience—people who are concerned about the rate of injuries in their
organization or community and want to do something about it.
This Handbook represents an extensive revision of my 1996 book, The Psychology of
Safety. Every chapter in the earlier edition has been updated and expanded, and three
new chapters have been added—one on behavioral safety analysis, another on intervening
with supportive conversation, and a third on promoting high-performance teamwork. As
a result, this edition is substantially longer than the first.
This is the first time I have prepared a second edition of a textbook, and I was sensitive
to the fact that new editions should justify their existence. I believe it is unfair to prepare
another edition of a book that is not a significant improvement over an earlier edition,
although I have seen this happen many times. I have often purchased a follow-up edition
to a book only to find very little difference between the two versions. This is frequently the
case with college textbooks.
This book offers significantly more information than the 1996 version. Thus, readers of
the first edition will not be disappointed if they purchase this Handbook. Plus, there are
many potential applications of this text. It is a comprehensive source of psychological prin-
ciples and practical applications for the safety professional or corporate safety leader. It
could also be used as required or recommended reading in a number of undergraduate or
graduate courses. More specifically, this Handbook is ideal for courses on human factors
engineering, safety management, or organizational performance management.
Many engineering and psychology departments do not offer courses with safety or
human factors in their titles. However, this Handbook is quite suitable for such standard
courses as applied psychology, organizational psychology, management systems, engi-
neering psychology, applied engineering, and even introductory psychology.
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page x

Fun to read
The writing style and format of this handbook are different from any professional text I
have written or read. Most authors of professional books, including me, have been taught
a particular academic or research style of writing that is not particularly enjoyable to read.
When did you last pick up a nonfiction technical book for recreational or “fun” reading.
To attract a larger readership, this text is written in a more exciting style than most pro-
fessional books, thanks to invaluable editorial coaching by Dave Johnson, editor of
Industrial Safety and Hygiene News. Each chapter includes several original drawings by
George Wills to illustrate concepts and add some humor to the learning process. I inter-
sperse these drawings in my professional addresses and workshops, and audiences find
them both enjoyable and enlightening.
I predict some of you will page through the book and look for these illustrations. That
is a useful beginning to learning concepts and techniques for improving the human
dynamics of safety. Then, read the explanatory text for a second useful step toward mak-
ing a difference with this information. If you, then, discuss the principles and procedures
with others, you will be on your way to putting this information to work in your organi-
zation, community, or home.

A testimony
Throughout this book, I include personal anecdotes to supplement the rationale of a prin-
ciple or the description of a technique or process. I would like to end this preface with
one such anecdote. In August 1994, the Hercules Portland Plant stopped chemical produc-
tion for two consecutive days so all 64 employees at the facility could receive a two-day
workshop on the psychology of safety.
Management had received a request for this all-employee workshop from a team of
hourly workers who previously attended my two-day professional development confer-
ence sponsored by the Mt. St. Helena Section of the American Society of Safety Engineers.
Rick Moreno, a Hercules warehouse operator and hazardous materials unloader for more
than 20 years, wrote the following reaction to my workshop. He read it to his coworkers at
the start of the Hercules workshop. It set the stage for a most constructive and gratifying
two days of education and training. If you approach the information in this Handbook with
some of the enthusiasm and optimism reflected in Rick’s words, you cannot help but make
a difference in someone’s safety and health.

Knowledge is precious. It is like trying to carry water in your cupped

hands to a thirsty friend. Ideas that were crystal clear upon hearing
them, tend to slip from your memory like water through the creases
of your hands, and while you may have brought back enough water
to wet your friend’s lips, he will not enjoy the full drink that you
were able to take.
So it is with this analogy of the Total Safety Culture. Those who
were there can only wet your lips with this new concept. Not a class
or a program, but a safe well way to live your life that spills into other
avenues of our environment.
It has no limit or boundaries as in this year, this plant. It is more
like we are on our way and something wonderful is going to happen.
Even though no answers are promised or given, the avenues in
which to find our own answers for our own problems will be within
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xi

our reach . . . That is why it is important that everyone has the

opportunity to take a full drink of the Total Safety Culture instead of
having our lips wet. Something wonderful is going to happen.

This Handbook is for you—Rick Moreno—and the many others who want to under-
stand the psychology of safety and reduce personal injuries. Hopefully, this material will
be used as a source of principles and procedures that you can return to for guidance and
benchmarks along your innovative journey toward building a safer culture of more
actively caring people.

E. Scott Geller
October, 2000
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xii
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xiii

The Author
E. Scott Geller, Ph.D. is a senior partner of Safety Performance
Solutions, Inc.—a leading edge organization specializing in behav-
ior-based safety training and consulting. Dr. Geller and his part-
ners at Safety Performance Solutions (SPS) have helped companies
across the country and around the world address the human
dynamics of occupational safety through flexible research-founded
principles and industry-proven tools. In addition, for more than
three decades, Professor E. Scott Geller has taught and conducted
research as a faculty member in the Department of Psychology at
Virginia Polytechnic Institute and State University, better known as
Virginia Tech. In this capacity, he has authored more than 300
research articles and over 50 books or chapters addressing the development and evaluation
of behavior-change interventions to improve quality of life.
His recent books in occupational health and safety include: The Psychology of Safety;
Working Safe; Understanding Behavior-Based Safety; Building Successful Safety Teams; Beyond
Safety Accountability: How to Increase Personal Responsibility; The Psychology of Safety
Handbook; and the primer: What Can Behavior-Based Safety Do For Me?
Dr. Geller is a Fellow of the American Psychological Association, the American
Psychological Society, and the World Academy of Productivity and Quality Sciences. He is
past editor of the Journal of Applied Behavior Analysis (1989 –1992), current associate editor
of Environment and Behavior (since 1982), and current consulting editor for Behavior and
Social Issues, the Behavior Analyst Digest, the Journal of Organizational Behavior Management,
and the International Journal of Behavioral Safety.
Geller earned a teaching award in 1982 from the American Psychological Association
and every university teaching award offered at Virginia Tech. In 1983 he received the
Virginia Tech Alumni Teaching Award and was elected to the Virginia Tech Academy of
Teaching Excellence; in 1990 he was honored with the university Sporn Award for distin-
guished teaching of freshman level courses; and in 1999 he was awarded the prestigious
W.E. Wine Award for Teaching Excellence.
To date, Dr. Geller has written almost 100 articles for Industrial Safety and Hygiene News,
a trade magazine disseminated to more than 75,000 companies. Dr. Geller has been the
principal investigator for more than 75 research grants that have involved the application
of behavioral science for the benefit of corporations, institutions, government agencies, and
communities.
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xiv
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xv

Acknowledgements
In December 1992, I purchased an attractive print of a newborn colt from an artist at Galeria
San Juan, Puerto Rico. While the artist—Jan D’Esopo—was signing my print, I asked her
how long it took to complete the original. “It took 25 minutes or 25 years,” she replied,
“depending on how you look at it.” “What do you mean,” I asked. “Well, it took me only
25 minutes to fill the canvas, but it took me 25 years of training and experience to prepare
for the artistry.”
I feel similarly about completing this Handbook which is an extensive revision and
expansion of my earlier book—The Psychology of Safety—published in 1996. While writing
the first edition and this revision took substantial time, the effort pales in comparison to
the many years of preparation supported by invaluable contributions from teachers,
researchers, consultants, safety professionals, university colleagues, and countless univer-
sity students.
Actually, I have been preparing to write this text since entering the College of Wooster
in Wooster, OH, in 1960. Almost all exams at this small liberal arts college required written
discussion (rather than selecting an answer from a list of alternatives). Therefore, I received
early experience and feedback at integrating concepts and research findings from a variety
of sources. I was introduced to the scientific method at Wooster and applied it to my own
behavioral science research during both my junior and senior years.
Throughout five years of graduate education at Southern Illinois University in
Carbondale, IL, I developed sincere respect and appreciation for the scientific method as
the key to gaining profound knowledge. My primary areas of graduate study were learn-
ing, personality, social dynamics, and human information processing and decision making.
The chairman of both my thesis and dissertation committees (Dr. Gordon F. Pitz) gave me
special coaching in research methodology and data analysis and refined my skills for pro-
fessional writing.
In 1968, I was introduced to the principles and procedures of applied behavior analy-
sis (the foundation of behavior-based safety) from one graduate course and a few visits to
Anna State Hospital in Anna, IL, where two eminent scholars, Drs. Ted Ayllon and Nate
Azrin, were conducting seminal research in this field. Those learning experiences (brief in
comparison with all my other education) convinced me that behavior-focused psychology
could make large-scale improvements in people’s lives. This insight was to have dramatic
influence on my future teaching, research, and scholarship.
I started my professional career in 1969 as assistant professor of Psychology at Virginia
Polytechnic Institute and State University (Virginia Tech). With assistance from under-
graduate and graduate students, I developed a productive laboratory and research pro-
gram in cognitive psychology. My tenure and promotion to associate professor were based
entirely upon my professional scholarship in this domain. However, in the mid-1970s I
became concerned that this laboratory work had limited potential for helping people. This
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xvi

conflicted with my personal mission to make beneficial large-scale differences in people’s

quality of life. Therefore, I turned to another line of research.
Given my conviction that behavior-based psychology has the greatest potential for
solving organizational and community problems, I focused my research on finding ways
to make this happen. Inspired by the first Earth Day in April 1970, my students and I devel-
oped, evaluated, and refined a number of community-based techniques for increasing
environment-constructive behaviors and decreasing environment-destructive behaviors.
This prolific research program culminated with the 1982 Pergamon Press publication of
Preserving the Environment: New Strategies for Behavior Change, which I co-authored with
Drs. Richard A. Winett and Peter B. Everett.
Besides targeting environmental protection, my students and I applied behavior-based
psychology to a number of other problem areas, including prison administration, school
discipline, community theft, transportation management, and alcohol-impaired driving. In
the mid-1970s we began researching strategies for increasing the use of vehicle safety belts.
This led to a focus on the application of behavior-based psychology to prevent uninten-
tional injury in organizational and community settings.
Perhaps this brief history of my professional education and experience legitimizes my
authorship of a handbook on the psychology of safety. However, my purpose for provid-
ing this information is not so much to provide credibility but to acknowledge the vast num-
ber of individuals—teachers, researchers, colleagues, and students—who prepared me for
writing this book. Critical for this preparation were our numerous research projects (since
1970), and this could not have been possible without dedicated contributions from hun-
dreds of university students. My graduate students managed most of these field studies,
and I am truly grateful for their valuable talents and loyal efforts.
Financial support from a number of corporations and government agencies made our
30 years of intervention research possible. Over the years, we received significant research
funds from the Alcohol, Drug Abuse, and Mental Health Administration, the Alcoholic
Beverage Medical Research Foundation, Anheuser-Busch Companies, Inc., Centers for
Disease Control and Prevention, Domino’s Pizza, Inc., Exxon Chemical Company, General
Motors Research Laboratories, Hoechst-Celanese, the Motor Vehicle Manufacturers
Association, the Motors Insurance Corporation, the National Highway Traffic Safety
Administration, the National Institute on Alcohol Abuse and Alcoholism, the National
Institute for Occupational Safety and Health, the National Science Foundation, Sara Lee
Knit Products, the U.S. Department of Education, the U.S. Department of Energy, the U.S.
Department of Health, Education, and Welfare, the U.S. Department of Transportation, and
the Virginia Departments of Agriculture and Commerce, Litter Control, Motor Vehicles,
and Welfare and Institutions. Profound knowledge is only possible through programmatic
research, and these organizations made it possible for my students and me to develop and
systematically evaluate ways to improve attitudes and behaviors throughout organiza-
tions and communities.
I am also indebted to the numerous guiding and motivating communications I have
received from corporate and community safety professionals worldwide. Daily contacts with
these individuals shaped my research and scholarship and challenged me to improve the
connection between research and application. They also provided valuable positive rein-
forcement to prevent “burnout.” It would take pages to name all of these friends and
acquaintances, and then I would necessarily miss many. You know who you are—thank you!
The advice, feedback, and friendship of two individuals—Harry Glaser and Dave
Johnson—have been invaluable for my preparation to write this text. I first met Harry
Glaser in September 1992 after I gave a keynote address at a professional development
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xvii

conference for the American Society of Safety Engineers. As Executive Vice President of
Tel-A-Train, Inc., Harry decided that a video-training series on the human dynamics I pre-
sented in my talk would be useful. That was the start of ongoing collaboration in develop-
ing videotape scripts, training manuals, and facilitator guides. This was invaluable
preparation for writing this text. In particular, my relationship with Harry Glaser improved
my ability to communicate the practical implications of academic research and scholarship.
Also vital to bridging the gap between research and application has been my long-term
alliance and synergism with Dave Johnson, editor of Industrial Safety and Hygiene News
(ISHN). Dave and I began learning from each other in the spring of 1990 when I submitted
my first article for his magazine. That year I submitted five articles on the psychology of
safety, and Dave did substantial editing on each. Every time one of my articles was pub-
lished, I learned something about communicating more effectively the bottom line of a psy-
chological principle or procedure.
As an author of more than 300 research articles and former editor of the premier
research journal in the applied behavioral sciences (Journal of Applied Behavior Analysis), I
knew quite well how to write for a research audience in psychology. However, Dave
Johnson showed me that when it comes to writing for safety professionals and the general
public, I had a lot to learn. In this regard, I continue to learn from him. Beginning in 1994,
I have written an article for a “Psychology of Safety” column. As a result, I have submitted
97 articles to ISHN and each profited immensely from Dave’s suggestions and feedback.
Preparing these articles laid the groundwork for this Handbook. Dave served as editor of the
first edition of this text, dedicating long hours to improving the clarity and readability of
my writing. Thus, the talent and insight of Dave Johnson have been incorporated through-
out this Handbook, and I am eternally beholden to him.
The illustrations throughout this handbook were drawn by George Wills (Blacksburg,
VA), which I think add vitality and fun to the written presentation. I hope you agree.
However, without the craft and dedication of Brian Lea, the illustrations could not have
been combined with the text for use by the publisher. In fact, Brian coordinated the final
processing of this entire text, combining tables and diagrams (which he refined) with
George Wills’ illustrations and the word processing from Gayle Kennedy, Nick Buscemi,
and Cassie Wright.
I also sincerely appreciate the daily support and encouragement I received from my
graduate students in 2000: Rebecca Click, Chris Dula, Kelli England, Jeff Hickman, and
Angie Krom; my colleagues at Safety Performance Solutions: Susan Bixler, Anne French,
Mike Gilmore, Molly McClintock, Sherry Perdue, Chuck Pettinger, Steve Roberts, and Josh
Williams; and from Kent Glindemann—research scientist for the Center for Applied
Behavior Systems. All of these people, plus many, many more, have contributed to 40 years
of preparation for this Psychology of Safety Handbook. I thank you all very much. I am hope-
ful the synergy from all your contributions will help readers make rewarding and long-
term differences in people’s lives.

E. Scott Geller
October, 2000
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xviii
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xix

Contents

Section one: Orientation and alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

Chapter 1 Choosing the right approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

Selecting the best approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
Behavior-based programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
Comprehensive ergonomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
Engineering changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Group problem solving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Government action (in Finland) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Management audits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Stress management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7
Poster campaigns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7
Personnel selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7
Near-miss reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
The critical human element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
The folly of choosing what sounds good . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9
The fallacy of relying on common sense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
Relying on research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16
Start with behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

Chapter 2 Starting with theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

The mission statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21
Theory as a map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22
Relevance to occupational safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24
A basic mission and theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25
Behavior-based vs. person-based approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26
The person-based approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
The behavior-based approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28
Considering cost effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29
Integrating approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
Chapter 3 Paradigm shifts for total safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33
The old three Es . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33
Three new Es . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xx

Ergonomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
Empowerment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36
Shifting paradigms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36
From government regulation to corporate responsibility . . . . . . . . . . . . . . . . . . . .37
From failure oriented to achievement oriented . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38
From outcome focused to behavior focused . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39
From top-down control to bottom-up involvement . . . . . . . . . . . . . . . . . . . . . . . . .40
From rugged individualism to interdependent teamwork . . . . . . . . . . . . . . . . . . .41
From a piecemeal to a systems approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41
From fault finding to fact finding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42
From reactive to proactive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45
From quick fix to continuous improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46
From priority to value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46
Enduring values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48

Section two: Human barriers to safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51

Chapter 4 The complexity of people . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53

Fighting human nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53
Learning to be at-risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .54
Dimensions of human nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55
Cognitive failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57
Capture errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59
Description errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60
Loss-of-activation errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60
Mode errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60
Mistakes and calculated risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61
In summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61
Interpersonal factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62
Peer influence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63
Power of authority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .64
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67

Chapter 5 Sensation, perception, and perceived risk . . . . . . . . . . . . . . . . . . . . . . . . . . . .69

An example of selective sensation or perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70
Biased by context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71
Biased by our past . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .73
Relevance to achieving a Total Safety Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75
Perceived risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76
Real vs. perceived risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76
The power of choice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77
Familiarity breeds complacency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78
The power of publicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78
Sympathy for victims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78
Understood and controllable hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78
Acceptable consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79
Sense of fairness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxi

Risk compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80

Support from research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .82
Implications of risk compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .84
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86
Chapter 6 Stress vs. distress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89
What is stress? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90
Constructive or destructive? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .91
The eyes of the beholder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92
Identifying stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93
Work stress profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .94
Coping with stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98
Person factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99
Fit for stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .100
Social factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101
Attributional bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102
The fundamental attribution error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103
The self-serving bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .106
Section three: Behavior-based psychology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .107
Chapter 7 Basic principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .109
Primacy of behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110
Reducing at-risk behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111
Increasing safe behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
Direct assessment and evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .114
Intervention by managers and peers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115
Learning from experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115
Classical conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .116
Operant conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .117
Observational learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119
Overlapping types of learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .126
Chapter 8 Defining critical behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .129
The DO IT process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .130
Defining target behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134
What is behavior? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .136
Outcomes of behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .136
Person–action–situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .137
Describing behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .137
Interobserver reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .138
Multiple behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .138
Observing behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .139
Properties of behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .140
Measuring behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .140
Recording observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .140
A personal example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .142
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxii

Using the critical behavior checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .143

Two basic approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .147
Starting small . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .148
Observing multiple behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1149
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151

Chapter 9 Behavioral safety analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .153

Reducing behavioral discrepancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155
Can the task be simplified? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155
Is a quick fix available? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .156
Is safe behavior punished? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .157
Is at-risk behavior rewarded? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .158
Are extra consequences used effectively? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159
Is there a skill discrepancy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .160
What kind of training is needed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .160
Is the person right for the job? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161
In summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .162
Behavior-based safety training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .162
Safety training vs. safety education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .163
Different teaching techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .164
An illustrative example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .165
In summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .165
Intervention and the flow of behavior change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .166
Three types of behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .166
Three kinds of intervention strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .167
The flow of behavior change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169
Accountability vs. responsibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .170
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .171
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .172

Section four: Behavior-based intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .173

Chapter 10 Intervening with activators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .175

Principle #1: Specify behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177
Principle #2: Maintain salience with novelty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177
Habituation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .178
Warning beepers: a common work example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .181
Principle #3: Vary the message . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .182
Changeable signs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .182
Worker-designed safety signs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183
Principle #4: Involve the target audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183
Safe behavior promise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .184
The “Flash for Life” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .186
The Airline Lifesaver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .188
Principle #5: Activate close to response opportunity . . . . . . . . . . . . . . . . . . . . . . . . . . .190
Point-of-purchase activators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .191
Activating with television . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .191
Buckle-up road signs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192
Principle #6: Implicate consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195
Incentives vs. disincentives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxiii

Setting goals for consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197

In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .199
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .200

Chapter 11 Intervening with consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .203

The power of consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .204
Consequences in school . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .205
Intrinsic vs. extrinsic consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .207
Internal vs. external consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .209
An illustrative story . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .210
Four types of consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .211
Managing consequences for safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .212
Four behavior-consequence contingencies for motivational intervention . . . . .213
The case against negative consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .214
Discipline and involvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .216
“Dos” and “don’ts” of safety rewards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .222
Doing it wrong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .222
Doing it right . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .223
An exemplary incentive/reward program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .226
Safety thank-you cards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .226
The “Mystery Observee” program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .228
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .229
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .230

Chapter 12 Intervening as a behavior-change agent . . . . . . . . . . . . . . . . . . . . . . . . . . . .233

Selecting an intervention approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .233
Various intervention approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .234
Multiple intervention levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236
Increasing intervention impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .238
Intervening as a safety coach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .239
Athletic coaching vs. safety coaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .239
The safety coaching process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .240
“C” for care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .241
“O” for observe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .242
“A” for analyze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .250
“C” for communicate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .250
“H” for help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
“H” for humor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
“E” for esteem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
“L” for listen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
“P” for praise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .256
What can a safety coach achieve? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .257
Self-appraisal of coaching skills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .259
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .259
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .261

Chapter 13 Intervening with supportive conversation . . . . . . . . . . . . . . . . . . . . . . . . . .265

The power of conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .266
Building barriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267
Resolving conflict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267
Bringing tangibles to life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxiv

Defining culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267

Defining public image and self-esteem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .268
Making breakthroughs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269
In summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269
The art of improving conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .270
Do not look back . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .270
Seek commitment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .271
Stop and listen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .271
Ask questions first . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .271
Transition from nondirective to directive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .273
Beware of bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .274
Plant words to improve self-image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .275
In summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .275
Conversation for safety management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .276
Coaching conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .277
Delegating conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .277
Instructive conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .277
Supportive conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .278
Recognizing safety achievement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .278
Recognize during or immediately after safe behavior . . . . . . . . . . . . . . . . . . . . . .279
Make recognition personal for both parties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .280
Connect specific behavior with general higher-level praise . . . . . . . . . . . . . . . . .281
Deliver recognition privately and one-on-one . . . . . . . . . . . . . . . . . . . . . . . . . . . . .281
Let recognition stand alone and soak in . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .281
Use tangibles for symbolic value only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .282
Secondhand recognition has special advantages . . . . . . . . . . . . . . . . . . . . . . . . . .282
Receiving recognition well . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .283
Avoid denial and disclaimer statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .283
Listen attentively with genuine appreciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . .284
Relive recognition later for self-motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .284
Show sincere appreciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Recognize the person for recognizing you . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Embrace the reciprocity principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Ask for recognition when deserved but not forthcoming . . . . . . . . . . . . . . . . . . .285
Quality safety celebrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .286
Do not announce celebrations for injury reduction . . . . . . . . . . . . . . . . . . . . . . . .286
Celebrate the outcome but focus on the journey . . . . . . . . . . . . . . . . . . . . . . . . . . .287
Show top-down support but facilitate bottom-up involvement . . . . . . . . . . . . . .287
Relive the journey toward injury reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .287
Facilitate discussion of successes and failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . .288
Use tangible rewards to establish a memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .288
Solicit employee input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .288
Choosing the best management conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .289
The role of competence and commitment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .289
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .290
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .291
Section five: Actively caring for safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .293

Chapter 14 Understanding actively caring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .295

820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxv

What is actively caring? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .296

Three ways to actively care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .297
Why categorize actively caring behaviors? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .298
An illustrative anecdote . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .300
A hierarchy of needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .302
The psychology of actively caring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .304
Lessons from research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .305
Deciding to actively care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309
Step 1. Is something wrong? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309
Step 2. Am I needed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .311
Step 3. Should I intervene? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .312
Steps 4 and 5. What should I do? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .313
Summary of the decision framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .314
A consequence analysis of actively caring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .314
The power of context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .317
Experiencing context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .317
An illustrative anecdote . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .318
Context at work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .319
Summary of contextual influence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .320
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .320
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .321

Chapter 15 The person-based approach to actively caring . . . . . . . . . . . . . . . . . . . . . . .325

Actively caring from the inside . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .326
Person traits vs. states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .328
Searching for the actively caring personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .328
Actively caring states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .329
Measuring actively caring states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .337
A safety culture survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .338
Support for the actively caring model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .338
Check your understanding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .338
Theoretical support for the actively caring model . . . . . . . . . . . . . . . . . . . . . . . . . . . . .339
Research support for the actively caring model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .341
Self-esteem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .342
Personal control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .342
Optimism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .342
Belonging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .343
Direct test of the actively caring model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .344
Actively caring and emotional intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .345
Safety, emotions, and impulse control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .346
Nurturing emotional intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .347
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .348
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349

Chapter 16 Increasing actively caring behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .353

Enhancing the actively caring person states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .353
Self-esteem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .354
Self-efficacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .357
Personal control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .361
The power of choice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .364
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxvi

Optimism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .366
Belonging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .367
Directly increasing actively caring behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .371
Education and training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372
Consequences for actively caring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .373
The reciprocity principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .374
Reciprocity: “Do for me and I’ll do for you” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .374
Commitment and consistency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .377
Some influence techniques can stifle trust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .380
Reinforcers vs. rewards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .380
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .381
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .382

Section six: Putting it all together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .385

Chapter 17 Promoting high-performance teamwork . . . . . . . . . . . . . . . . . . . . . . . . . . . .387
Paradigm shifts for teamwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .388
From individual to team performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .388
From individual jobs to team tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .388
From competitive rewards to rewards for cooperation . . . . . . . . . . . . . . . . . . . . .388
From self-dependence to team-dependence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .389
From one-to-one communication to group interaction . . . . . . . . . . . . . . . . . . . . .389
When teams do not work well . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .389
Group gambles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .390
Overcoming groupthink . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .392
Cultivating high-performance teamwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .392
Selecting team members . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .393
Clarify the assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .394
Establish a team charter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .395
Develop an action plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .399
Make it happen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .400
Evaluate team performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .403
Disband, restructure, or renew the team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405
In summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .408
The developmental stages of teamwork . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .409
Forming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .409
Storming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .410
Norming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .410
Performing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .411
Adjourning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .412
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .412
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .414

Chapter 18 Evaluating for continuous improvement . . . . . . . . . . . . . . . . . . . . . . . . . . .415

Measuring the right stuff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .415
Limitations of performance appraisals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .416
What is performance improvement? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .417
Developing a comprehensive evaluation process . . . . . . . . . . . . . . . . . . . . . . . . . . . . .420
What to measure? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .422
Evaluating environmental conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .423
820049_CRC1_L1540_FM 11/10/00 1:00 PM Page xxvii

Evaluating work practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .427

Evaluating person factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .427
Reliability and validity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .430
Cooking numbers for evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .435
What do the numbers mean? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .436
An exemplar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .438
Evaluating costs and benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .439
You cannot measure everything . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .441
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .442
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .443
Chapter 19 Obtaining and maintaining involvement . . . . . . . . . . . . . . . . . . . . . . . . . . .445
Starting the process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .446
Management support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .446
Creating a Safety Steering Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .446
Developing evaluation procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .447
Setting up an education and training process . . . . . . . . . . . . . . . . . . . . . . . . . . . . .447
Sustaining the process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .450
Awareness support—activators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .450
Performance feedback—consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .450
Tangible consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .451
Ongoing measurement and evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .451
Follow-up instruction/booster sessions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .451
Involvement of contractors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .452
Trouble shooting and fine-tuning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .452
Cultivating continuous support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .453
Where are the safety leaders? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .453
Safety management vs. safety leadership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .455
Communication to sell the process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .459
Overcoming resistance to change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .462
Planning for safety generalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .467
Building and sustaining momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .470
Relevance to industrial safety and health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .470
Achievement of the team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .471
Atmosphere of the culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .471
Attitude of the leaders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .472
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .473
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .474
Chapter 20 Reviewing the principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .477
The 50 principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .478
In conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .497
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .498
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .501
Name Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .523
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 1

section one

Orientation and alignment

820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 2
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 3

chapter one

Choosing the right approach

The basic purpose of this book is outlined in this chapter to explore the human dynamics of occupa-
tional health and safety, and to show how they can be managed to significantly improve safety per-
formance. The principles and practical procedures you will learn are not based on common sense nor
intuition, but rather on reliable scientific investigation. Many recommendations seem counter to
“pop psychology” and traditional approaches to safety. So keep an open mind while you read about
the psychology of safety.

“Organizations learn only through individuals who learn.”—Peter Senge

Safety professionals, team leaders, and concerned workers today scramble to find the
“best” safety approach for their workplace. Typically, whatever offers the cheapest “quick
fix” sells. This is not surprising, given the “lean and mean” atmosphere of the times.
Programs that offer the most benefit with least effort sound best, but will they really work
to improve safety over the long term?
This text will help you ask the right questions to determine whether a particular approach
to safety improvement will work. More importantly, this text describes the basic ingredients
needed to improve organizational and community safety. In fact, you will find sufficient
information to improve any safety process. Learning the principles and procedures described
here will enable you to make a beneficial, long-term difference in the safety and health of
your workplace, home, and community. The information is relevant for most other perform-
ance domains, from increasing the quantity and quality of productivity in the workplace to
improving quality of life in homes, neighborhoods, and throughout entire communities.

Selecting the best approach

With so many different approaches to safety improvement available, how can we select the
best? My first thought is to ask, “What does the research indicate?” In other words, are there
objective data available from program comparisons to shed light on our dilemma?
Unfortunately, there are few systematic comparisons of alternative safety interventions.
However, this does not stop consultants from showing us impressive results regarding the
success of their approaches. Nor does it prevent them from implying (or boldly stating) that
we can obtain similar fantastic results by simply following their patented “steps to success.”
Keep in mind this marketing information usually comes from selected client case stud-
ies. Very few of these “success stories” were collected objectively and reliably enough to
meet the rigorous standards of a professional research journal. When consultants try to sell
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 4

4 Psychology of safety handbook

Figure 1.1 Some research is not worth considering.

an approach to safety with this kind of data to you, ask them if they have published their
results in a peer-reviewed journal. If they can show you a published research report of their
impressive results or a professional presentation of a program very similar to theirs, then
give their approach special consideration in your selection process. The validity and appli-
cability of even published research varies dramatically. Figure 1.1 depicts the low end of
research quality.
Most of the published research on safety improvement systematically evaluates
whether a particular program worked in a particular situation, but it does not compare one
approach with another. In other words, this research tells us whether a particular strategy
is better than nothing, but offers no information regarding the relative impact of two or
more different strategies on safety improvement. Such research has limited usefulness
when selecting between different approaches.
An exception can be found in a 1993 review article in Safety Science, where Stephen
Guastello (1993) summarized systematically the evaluation data from 53 different research
reports of safety programs. Guastello provided rare and useful information for deciding
how to improve safety. You can assume the evaluations were both reliable and valid,
because each report appeared in a scientific peer-reviewed journal. All of the studies
selected for his summary were conducted in a workplace setting since 1977, and each study
evaluated program impact with outcome data (including number and severity of injuries).
As listed in Figure 1.2, 10 different approaches to safety improvement were repre-
sented in the 53 research articles summarized by Guastello. They are ranked according to
the mean percentage decrease in injury rates as detailed by Guastello in his careful analy-
sis of the published reports. Because one-half of the percentages were based on three or
fewer research reports, the program ranking should be considered preliminary. More
research on program impact is clearly needed, as are systematic comparisons.
From my reading of Guastello’s article, I believe it is safe to say the behavior-based
and comprehensive ergonomics approaches lead the field. Personnel selection, the most
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 5

Chapter one: Choosing the right approach 5

Figure 1.2 Research comparisons reveal informative ranking of approaches to reduce

work injuries. (Adapted from Guastello, 1993. With permission.)

popular method (26 studies targeted a total of 19,177 employees), is among the least effec-
tive. With the exception of “near miss” reporting, the other program techniques are clearly
in the middle of the ranking, with insufficient evidence to favor one over another. To appre-
ciate this ranking of program effectiveness, it is helpful to define the program labels given
in Figure 1.2. Here are brief descriptions of these approaches to reduce workplace injuries.

Behavior-based programs
Programs in this category consisted of employee training regarding particular safe and at-
risk behaviors, systematic observation and recording of the targeted behaviors, and feed-
back to workers regarding the frequency or percentage of safe vs. at-risk behavior. Some of
these programs included goal setting and/or incentives to encourage the observation and
feedback process. See Petersen (1989) for a comprehensive review of behavior-based stud-
ies in the research literature and for more evidence that this approach to industrial safety
deserves top billing.

Comprehensive ergonomics
The ergonomics (or human factors) approach to safety refers essentially to any adjustment
of working conditions or equipment in order to reduce the frequency or probability of an
environmental hazard or at-risk behavior (Kroemer, 1991). An essential ingredient in these
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 6

6 Psychology of safety handbook

programs was a diagnostic survey or environmental audit by employees which led to spe-
cific recommendations for eliminating hazards that put employees at risk or promoted at-
risk behaviors. Guastello noted a direct relationship between injury reductions and the
amount of time devoted to dealing with the ergonomic recommendations of a diagnostic
survey. See Guastello (1989) for further discussion of the development and application
of an ergonomic diagnostic survey.

Engineering changes
This category includes the introduction of robots or the comprehensive redesign of facili-
ties to eliminate certain at-risk behaviors. It is noted, however, that the robotic interven-
tions introduced the potential for new types of workplace injuries, like a robot catching an
operator in its work envelope and impaling him or her against a structure. Thus, robotic
innovations usually require additional engineering intervention such as equipment
guards, emergency kill switches, radar-type sensors, and workplace redesign to prevent
injury from robots. Behavioral training, observation, and feedback (as detailed in Section 4
of this Handbook) are also needed following engineering redesign.

Group problem solving

For this approach, operations personnel met voluntarily to discuss safety issues and prob-
lems, and to develop action plans for safety improvement (Saarela, 1990). This approach is
analogous to quality circles where employees who perform similar types of work meet reg-
ularly to solve problems of product quality, productivity, and cost.

Government action (in Finland)

In Finland, two government agencies that are responsible for labor production target
the most problematic occupational groups and implement certain action strategies. These
include

1. Disseminating information to work supervisors regarding the cause of workplace

injuries and methods to reduce them.
2. Setting standards for safe machine repair and use.
3. Conducting periodic work site inspections.

See Bjurstrom (1989) for more specifics regarding this Finnish national intervention.

Management audits
For the programs in this category, designated managers were trained to administer a stan-
dard International Safety Rating System (ISRS). This system evaluates workplaces based
on 20 components of industrial safety. These include leadership and administration, man-
agement training, planned inspections, task and procedures analysis, accident investiga-
tions, task observations, emergency preparedness, organizational rules, accident analysis,
employee training, personal protective equipment, health control, program evaluation,
engineering controls, and off-the-job safety.
Managers conduct the comprehensive audits annually to develop improvement strate-
gies for the next year. Specially certified ISRS personnel visit target sites and recognize a
plant with up to five “stars” for exemplary safety performance. See Eisner and Leger (1988)
or Pringle and Brown (1990) for more specifics on application and impact of the ISRS.
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 7

Chapter one: Choosing the right approach 7

Stress management
These programs taught employees how to cope with stressors or sources of work stress
(Ivancevich et al., 1990; Murphy, 1984). Exercise was often a key action strategy promoted
as a way to prevent stress-related injuries in physically demanding jobs (Cady et al., 1985).
I shall discuss the topic of stress as it relates to injury prevention in Chapter 6.

Poster campaigns
The two published studies in this category evaluated the accident reduction impact of post-
ing signs that urged workers at a shipyard to avoid certain at-risk behaviors and to follow
certain safe behaviors. Most signs were posted at relevant locations and gave specific
behavioral instructions like “Take material for only one workday,” “Gather hoses immedi-
ately after use,” “Wear your safety helmet,” and “Check railing and platform couplings (on
scaffolds).”
For one study, safety personnel at the shipyard gave work teams weekly feedback
regarding compliance with sign instructions (Saarela et al., 1989). In the other study, envi-
ronmental audits, group discussions, and structured interviews were used to develop the
poster messages (Saarela, 1989). Thus, it is possible that factors other than the posters them-
selves contributed to the moderate short-term impact of this intervention approach. All of
these factors are covered in this Handbook, including ways to maximize the beneficial effects
of safety signs (in Chapter 10).

Personnel selection
This popular but ineffective approach to injury prevention is based on the intuitive notion
of “accident proneness.” The strategy is to identify aspects of accident proneness among
job applicants and then screen out people with critical levels of certain characteristics.
Accident proneness characteristics targeted for measurement and screening have included
anxiety, distractibility, tension, insecurity, beliefs about injury control, general expectancies
about personal control of life events, social adjustment, reliability, impulsivity, sensation
seeking, boredom susceptibility, and self-reported alcohol use.
Although measuring and screening for accident proneness sounds like a “quick fix”
approach to injury prevention, this method has several problems you will readily realize as
you read more in this Handbook about the psychology of safety. Briefly, this technique has
not worked reliably to prevent workplace injuries because

1. The instruments or procedures available to measure the proneness characteristics

are unreliable or invalid.
2. The characteristics do not carry across settings, so a person might show them at
home but not at work or vice versa.
3. A person with a higher desire to take risks (such as a sensation seeker) might be
more inclined to take appropriate precautions (like using personal protective equip-
ment) to avoid potential injury.

Also, although individuals have demonstrated different risk levels, many have exhib-
ited these inconsistently—risk taking is likely to be influenced by environmental condi-
tions. Additionally, finding correlations between certain personal characteristics and injury
rates does not mean the proneness factors caused changes in the injury rate (Rundmo,
1992). Other factors, including cultural factors or environmental events, could cause both
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 8

8 Psychology of safety handbook

the personal characteristics and the accident proneness. See Geller (1994a,b) for additional
details regarding problems with this approach to injury prevention.

“Near-miss” reporting
This approach involved increased reporting and investigation of incidents that did not
result in an injury but certainly could have under slightly different circumstances. One pro-
gram in this category increased the number of corrective suggestions generated but did not
reduce injury rate. The other scientific publication in this category reported a 56 percent
reduction in injury severity as a result of increased reporting of near hits,* but the overall
number of injuries did not change.

The critical human element

Every safety approach listed in Figure 1.2 requires that you consider the human element or
the psychology of safety. Indeed, the most successful approaches, behavior-based safety
and comprehensive ergonomics, directly address the human aspects of safety. The bottom
line is illustrated in Figure 1.3. The three employees here are looking at a contributing fac-
tor in almost every injury—the human factor. Thus, any safety intervention that improves
the safety-related behaviors of workers will prevent workplace injuries.
The behavior-based approach targets human behavior and relies on interpersonal
observation and feedback for intervention. The success of comprehensive ergonomics
depends on employees observing relationships between behaviors and work situations,

Figure 1.3 Human dynamics contribute to almost every injury.

* “Near miss” is used routinely in the workplace to refer to an incident that did not result in an injury. Because a
literal translation of this term would mean the injury actually occurred, “near hit” is used throughout this text
instead of “near miss.”
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 9

Chapter one: Choosing the right approach 9

and then recommending feasible changes in behavior, equipment, or environmental con-

ditions to make the job more “user friendly” and safe. See the text edited by Oborne et al.
(1993) for a comprehensive discussion of the psychological aspects of ergonomics.
Today, achieving success in safety requires concerted efforts in the realm of psychology.
Safety professionals are hungry for insights. In recent years, many seminars at national and
regional safety conferences purporting to teach aspects of the psychology of safety have
attracted standing-room only crowds. Just look at these titles from recent conferences of the
National Safety Council or the American Society of Safety Engineers.

• “Managing Safe Behavior for Lasting Change”

• “Humanizing the Total Safety Program”
• “The Human Element in Achieving a Total Safety Culture”
• “The Psychology of Injury Prevention”
• “Behavior-Based Safety Management: Parallels with the Quality Process”
• “Behavioral Management Techniques for Continuous Improvement”
• “Improving Safety Through Innovative Behavioral and Cultural Approaches”
• “Safety Leadership Power: How to Empower All Employees”
• “Moving to the Second Generation in Behavior-Based Safety”
• “Potholes in the Road to Behavioral Safety”
• “Implementing Behavior-Based Safety on a Large Scale”
• “Motivating Employees for Safety Success”
• “Integrating Behavioral Safety into Other Safety Management Systems”
• “From Knowing to Doing: Achieving Safety Excellence”
• “Safety and Psychology: Where Do We Go From Here?”

I attended each of these presentations and found numerous inconsistencies be-

tween presentations dealing with the same topic. Sometimes, I noted erroneous and
frivolous statements, inaccurate or incomplete reference to psychological theory or
research, and invalid or irresponsible comparisons between various approaches to dealing
with the psychology of safety. It seemed a primary aim of several presentations was
to “sell” their own particular program or consulting services by overstating the benefits
of their approach and giving an incomplete or naive discussion of alternative methods or
procedures.

The folly of choosing what sounds good

The theory, research, and tools in psychology are so vast and often so complex that it can
be overwhelming to decide which particular approach or strategy to use. As a result, we
are easily biased by common-sense words that sound good. Valid theory, principles, and
procedures founded on solid research evidence are often ignored. Today, there is an appar-
ent endless market of self-help books, audiotapes, and videotapes addressing concepts
seemingly relevant to the psychology of injury prevention. In recent years, I have listened
to the following audiotapes—representing only a fraction of “pop psychology” tapes with
topics relevant to the psychology of safety.

• “Coping with Difficult People” by R. M. Branson

• “Personal Excellence” by K. Blanchard
• “How to Build High Self-Esteem” by J. Canfield
• “The Seven Habits of Highly Effective People” by S. R. Covey
• “First Things First” by S. R. Covey, A. R. Merrill, and R. R. Merrill
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 10

10 Psychology of safety handbook

• “The Science of Personal Achievement” by N. Hill

• “Increasing Human Effectiveness” by R. Moawad
• “Lead the Field” by E. Nightingale
• “Unlimited Power” by A. Robbins
• “The Psychology of Achievement” by B. Tracey
• “The Psychology of Success” by B. Tracey
• “The Universal Laws of Success and Achievement” by B. Tracy
• “The Psychology of Winning” by D. Waitley
• “Self-Esteem” by J. White
• “Goal Setting” by Z. Ziglar
• “Top Performance” by Z. Ziglar
• “The Secrets of Power Persuasion” by R. Dawson
• “The 12 Life Secrets” by R. Stuberg
• “The Courage to Live Your Dreams” by L. Brown
• “The New Dynamics of Goal Setting” by D. Waitley
• “Transforming Stress into Power” by M. J. Tazer and S. Willard

Which, if any, of these pop psychology audiotapes gives safety professionals the
“truth”—the most effective and practical tools for dealing with the human aspects of
safety? Many of the strategies to promote personal growth and achievement, including atti-
tude and behavior change, were selected and listened to with trust and optimism because
they sound good—not because there is solid scientific evidence that the strategies work.
Many of the same anecdotes and quotes from famous people are repeated across audio-
tapes. Also, strategies suggested for developing self-esteem and building personal success
are quite similar, with goal-setting and self-affirmations (such as repeating “I am the great-
est” to oneself) leading the list. Does this repetition of good-sounding self-help strategies
make them right?
Some of the most cost-effective strategies for managing behaviors and attitudes at the
personal and organizational level are not even mentioned in many of the pop psychology
books, audiotapes, and videotapes. This might be the case not only because authors and pre-
senters are unaware of the latest research, but also because many of the best techniques for
individual and group improvement do not sound good—at least at first. The primary pur-
pose of this text is to teach the most effective approaches for dealing with the human aspects
of occupational safety and health. These principles and procedures were not selected because
they sound good, but because their validity has been supported with sound research.

The fallacy of relying on common sense

Since we live psychology every day, there is no doubt that good common sense can go a
long way in selecting effective techniques for benefiting human achievement. I have met
many people, including supervisors, line workers, safety professionals, and motivational
speakers who seem to have special intuition or common sense for selecting approaches to
help people improve. Indeed, Tom Peters, Anthony Robbins, Brian Tracey, Denis Waitley,
and many others who successfully market techniques for increasing human potential and
achievement are particularly skillful at selecting those principles and procedures backed
by research. But as depicted in Figure 1.4, common sense is subjective, based on a person’s
everyday selective experiences and biased interpretation of those experiences.
As mentioned, I prefer principles and procedures based on scientific knowledge,
which comes from the experience of the researcher. At a four-day Quality Enhancement
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 11

Chapter one: Choosing the right approach 11

Figure 1.4 Without science, decision making is a biased shot in the dark.

Seminar in 1991, I heard W. Edwards Deming assert, “Experience teaches us nothing; that’s
why American business is in such a mess.” Deming called for theory to guide objective and
reliable observations, and to integrate the results from these systematic data-based experi-
ences. Thus, while common sense is gained through biased subjective experience, scientific
knowledge is gained through theory-driven objective experimentation.
Dr. Aubrey C. Daniels, a world-renowned educator and consultant in the field of orga-
nizational performance management, asserts in his book, Bringing out the Best in People, that
he is “on a crusade to stamp out the use of common sense in business. Contrary to popu-
lar belief there is not too little common sense in business, there is too much.” Daniels lists
the following distinctions between common sense and scientific knowledge, reflecting the
need to be cautious when relying on only common sense to deal with human aspects of
occupational health and safety.

• Common sense knowledge is acquired in ordinary business and living, while scien-
tific knowledge must be pursued deliberately and systematically.
• Common sense knowledge is individual; scientific knowledge is universal.
• Common sense knowledge accepts the obvious; scientific knowledge questions the
obvious.
• Common sense knowledge is vague; scientific knowledge is precise.
• Common sense cannot be counted on to produce consistent results; applications of
scientific knowledge yield the same results every time.
• Common sense is gained through uncontrolled experience; scientific knowledge is
gained through controlled experimentation.

I have heard or read a number of psychology-related statements from motivational

speakers and consultants that sound like good common sense but in fact contradict scien-
tific knowledge. Some of these statements appear so many times in the pop psychology lit-
erature that they are accepted as basic truths, when in fact they cannot be substantiated
with objective evidence. Consider, for example, the following 15 myths which are com-
monly stated but make no sense.
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 12

12 Psychology of safety handbook

Myth 1. “Reprimand privately but recognize publicly.”

Sound familiar? Actually whether correcting behavior or giving recognition, it is always

better to communicate one-to-one in private. Never recognize a person in front of a group
without that person’s permission. Some people are embarrassed by public commendation;
others fear verbal harassment from peers. They imagine someone saying, “Why did you
deserve that ‘safe-employee-of-the-month’ award? I’ve done as much around here for
safety as you. Have you been kissing up to the boss again?”

Myth 2. “We learn more from our mistakes than our successes.”

Think about what is being said here. What do you learn when you make a mistake? You
learn what not to do. That’s something worth learning, but consider how much more you
learn when you do something correctly and receive feedback that you are correct. You learn
what you need to continue in order to be successful.

Myth 3. “77 percent of our mental thoughts are negative.”

I have heard more than one pop psychologist make this statement to justify the need to give
more positive than negative feedback. I like the conclusion. We do need more recognition
for correct behavior than correction for incorrect behavior, primarily because we learn
more from correct than incorrect performance. But to claim that a certain proportion of neg-
ative thoughts pervade the minds of human beings is absurd. Ask yourself the question
“How could they know?” and you will see that this statement is ridiculous.

Myth 4. “Do something 21 times and it becomes a habit.”

I am sure you have heard this foolish assertion. It is even the title of a self-help book. For
years I have wondered where the “21” came from. Then a local farmer reminded me. It
takes 21 days for an egg to hatch into a chicken.
Behavior needs to be repeated many times to develop fluency and then a habit. That is
why it is important to support the safe work practices of our friends and teammates. But to
presume there is a set number of repetitions needed for habit formation is downright silly,
and frankly insults the intelligence of most listeners.
Many people have developed the habit of safety-belt use, for example, but the number
of times belt use occurred before it became automatic varied dramatically across individu-
als, and depended partly on the strength of the old bad habit. Specifically, how ingrained
was a person’s routine of entering and starting a vehicle without buckling up? How incon-
venient is the simple buckle-up behavior for a certain individual in a particular vehicle?
Some behaviors are so complex or inconvenient they never become habitual. Consider
the chain of behaviors needed to lock out a power source, complete a behavioral audit, or
follow a stretch and exercise routine. These important safety-related behaviors are never
likely to become automatic. Their occurrence will probably always require some deliberate
motivating influence, whether external or self-imposed.

Myth 5. “We can only motivate ourselves, not others.”

It is a good thing this frequent pop psychology statement is untrue, or we could not
motivate others to choose the safe way of doing something when the at-risk alternative is
more comfortable, efficient, convenient, or perhaps habitual. In Section 4 of this Handbook,
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 13

Chapter one: Choosing the right approach 13

I discuss the role of external motivational intervention to increase safe behavior and decrease
at-risk behavior. Here, I only want to discredit the myth that only self-motivation works.

Myth 6. “Incentives and rewards are detrimental to self-motivation.”

This unfounded statement is related to the previous pop psychology myth, and is fre-
quently used by safety professionals to criticize all applications of incentives to improve
safety performance (Krause, 2000). A popular book entitled Punished by Rewards (Kohn,
1993) has been read by too many safety professionals. Now, this myth is getting dispersed
at safety conferences and in safety magazines.
Yes, you can insult one’s intelligence or self-motivation by using the wrong type of
intervention to improve a particular behavior (Geller, 1998), and this can have a temporary
detrimental effect on individual performance. Research, however, indicates this negative
impact is relatively infrequent and if it does occur, it is short-lived (Carton, 1996; Cameron
and Pierce, 1994; Eisenberger and Camerson, 1996). Incentives and rewards are far more
likely to benefit desired performance and even self-motivation for long-term behavior
change if they are used correctly. The key is to use a behavior-based approach to incen-
tive/reward programs, as I detail later in Chapter 11 of this Handbook.

Myth 7. “People can perform any job they really want.”

I am sure you have heard something like, “You can do anything you want, if you just per-
sist long and hard enough.” Perhaps you have even made a similar assertion to motivate
someone to try harder. Of course, words like these sound good, but surely they cannot be
true. Few of us can become the professional athlete, entertainer, or movie star we would
like to be. Most of us could not even become President.
Environmental, physical, and psychological factors limit our potential and narrow the
range of things we can do with our lives. As illustrated in Figure 1.5, “trying harder” can-
not substitute for talent, equipment, and method, but this should not lead to despair.
Rather, we should attempt to become the best we can be within our limitations. We try to
find our niche.
By the time we reach employment age, there is a finite range of jobs we can perform
effectively. Of course, people can learn new tasks and thereby expand their possibilities,
but there is a limit. It is very important to recognize and understand our limitations, as well
as to realize our special interests and skills.

Myth 8. “Brainpower, experience, and desire make the difference.”

Perhaps the most common theme among the research-based management/leadership

books I have read is that talent is the key to success, not brainpower, motivation, or desire
(see especially Buckingham and Coffman, 1999). Naturally, these factors contribute to an
individual’s interests and abilities, which are largely determined when a person applies
for a job. One’s talent is the ultimate observable factor determining whether a particular
task is done well. So a manager’s critical challenge is to select the right (talented) person
for a job.
Several selection techniques can be used to match talent with job, from interviews and
interest questionnaires to abilities tests and behavioral observation. Such assessment tools
help determine people’s special interests and skills, which together define talent. Long-
term success of a performance team, whether in sports or industry, is dependent on match-
ing talents with tasks or functions.
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 14

14 Psychology of safety handbook

Figure 1.5 Motivation cannot substitute for equipment and method.

Myth 9. “Identify people’s weaknesses and fix them.”

This seems to be a common theme in supervisory counseling and performance appraisal ses-
sions. However, few, if any, managers or supervisors have the talent and time to “fix” an
employee’s weaknesses. They can provide objective feedback, of course, to support desired
(e.g., safe) behavior and correct undesired (e.g., at-risk) behavior. Here, I mean more than a
list of job components on a behavioral checklist. I am referring to a person’s relative ability to
perform a certain job. It is far more cost effective to identify people’s strengths and give them
the kind of job opportunities that benefit from their talents and enable them to flourish.

Myth 10. “Spend more time with the least productive workers.”

Attempting to fix people’s weaknesses takes time; the more weaknesses people have, the
more time it will take. Managers who believe this is their job will naturally spend more
time with the least productive employees. This also follows from an assumption
that the more productive workers know what they are doing and do not need supervisory
attention. However, this discussion leads to an opposite assumption and action plan.
Those employees more successful at a particular task are more talented and, therefore,
contribute more to the successful performance of a team or business unit. These employees
need supportive feedback and recognition. Managers need to assure these people have
what it takes to make the best of their talents.
In many if not most job settings, resources and managerial support are limited. It is far
more cost effective to keep talented personnel working at optimal levels than helping those
less talented at a particular role improve their effectiveness.
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 15

Chapter one: Choosing the right approach 15

Myth 11. “Don’t play favorites.”

This myth is of course inconsistent with the prior myth that managers should spend more
time with less productive employees. My refutation of Myth 9 leads to an obvious contra-
diction of this myth. Specifically, managers should play favorites. As I indicated previously,
managers need to spend more time with their more talented employees.

Myth 12. “Promotion is the best way to reward excellence.”

This myth is obviously detrimental to optimizing a system, given the need to keep people
working at tasks that use their talents. Yet it seems commonplace to offer promotion,
including a higher salary and more authority, to individuals who excel at a particular job.
In fact, with continued promotions people can be promoted to jobs at which they are actu-
ally incompetent. This was termed “The Peter Principle” by Lawrence Peter (1969).
After finding the best talent for a job and enabling that talent to flourish and improve,
managers need to keep that talent on the job, not see it leave for another position. For this
to happen, however, recognition and promotion need to occur within the same job assign-
ment. It must be possible for an employee to be promoted to a higher status level by doing
the same job rather than leaving for another. How much of a paradigm shift would this be
for your company? Can your culture develop heroes at every role?

Myth 13. “There is one best way to perform every job.”

This is a common belief in many traditional hierarchical organizations. I am sure you see
how this myth can stifle creativity and the ability to keep talents thriving on the same job.
Sustaining talent on a particular job requires workers to feel like heroes. They need to
believe they are contributing individually and creatively to their team. This is unlikely if
they feel they are only following someone else’s protocol or assigned checklist. Allowing
for and expecting people to find better ways to perform a job will lead to continuous
improvement. When you give talent an opportunity to improve, it does.

Myth 14. “Follow The Golden Rule: treat others as you want to be treated.”

I bet this statement is heavily ingrained in your belief system and it is difficult for you to
consider this a myth. Okay, you followed my logic up to this point, with perhaps some min-
imal acceptance, but now I have gone too far. How can anyone refute “The Golden Rule”?
I question the literal translation of this rule.
When working with people to identify job-related talents, and then helping the best tal-
ents succeed where needed, should you really treat people the way you want to be treated?
What kind of job recognition would you want—public or private? Would you want a man-
ager to emphasize competition or co-operation when requesting more applications of your
talent? Would an incentive to work overtime insult you or demotivate you? Would you
appreciate an extrinsic reward for your extra effort? Do you want cash, a meal for two at a
local restaurant, or two tickets to the football game on Saturday?
Many more questions could be asked related to desired ways to select, support, and
maintain talent. But I hope I have made my point. We are all unique, and have different
needs, desires, and interests. It is simply not wise nor valid to assume everyone else wants
to be treated as you want to be treated. Of course, it is safe to make some generalizations.
For example, few people like criticism and most people enjoy genuine praise. But effective
managers do not assume others like what they like, but rather listen empathetically and
observe carefully to find out how to treat others.
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 16

16 Psychology of safety handbook

Myth 15. “For every 300 unsafe acts there are 29 minor injuries and 1 major injury.”

I could not end this brief list of unfounded claims without adding the most popular myth
in the field of industrial safety. This ratio between at-risk behavior and injury was first pro-
posed in the 1930s by H. W. Heinrich (1931). It has been repeated so often, some safety pros
refer to it as “Heinrich’s Law.” It started as a mere estimate, and after years of use in safety
speeches and publications, without any empirical verification, its status was elevated to
“basic principle” or “natural law.”
This statement does show the connection between behavior and injury, and implies
that a pro-active approach to injury prevention requires attention to behaviors and near
hits. However, the number of at-risk behaviors per injury is much larger than 300, as veri-
fied empirically by Frank Bird in 1966 and 1969, who also found property damage to be a
reliable predictor or leading indicator of personal injury (see Bird and Davies, 1996).
My point here, however, is not about a pro-active vs. re-active approach to safety, but
rather how the long-term repetition of an unfounded proclamation that sounds good can
become a common myth with presumed validity. This happens all too often when dealing
with a topic like psychology about which everyone has an opinion from their biased com-
mon sense.

Relying on research
This Handbook teaches research-based psychology related to occupational safety. Thus, by
reading this text you will improve your common sense about the psychology of safety. At
this point, I hope you are open to questioning the validity of good-sounding statements
that are not supported by sound research.
Research in psychology, for example, does not generally support the following com-
mon statements related to the psychology of occupational health and safety.

• Practice makes perfect.

• Spare the rod and spoil the child.
• Attitudes need to be changed before behavior will change.
• Human nature motivates safe and healthy behavior.
• People will naturally help in a crisis.
• Rewards for not having injuries reduce injuries.
• All injuries are preventable.
• Zero injuries should be a safety goal.
• Manage only that which can be measured.
• Safety should be considered a priority.

These and other common safety beliefs will be refuted in this book, with reference to
scientific knowledge obtained from systematic research. Sometimes, case studies will illus-
trate the practicality and benefits of a particular principle or procedure, but the validity of
the information was not founded on case studies alone. The approaches presented in this
text were originally discovered and verified with systematic and repeated scientific
research in laboratory and field settings.

Start with behavior

Many pop psychology self-help books, audiotapes, and motivational speeches give mini-
mal if any attention to behavior-based approaches to personal achievement. “Behavioral
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 17

Chapter one: Choosing the right approach 17

control” and “behavior modification” do not sound good. The term “behavior” has nega-
tive connotations, as in “let’s talk about your behavior at the party last night.” Dr. B. F.
Skinner, the founder of behavioral science and its many practical applications, was one of
the most misunderstood and underappreciated scientists and scholars of this century, pri-
marily because the behavior management principles he taught did not sound good. Two
particularly insightful but underappreciated and misunderstood books by Skinner are
Walden Two (1948) and Beyond Freedom and Dignity (1971).
Professor Skinner and his followers have shown over and over again that behavior is
motivated by its consequences, and thus behavior can be changed by controlling the events
that follow behavior. But this principle of “control by consequences” does not sound as
good as “control by positive thinking and free will.” Therefore, the scientific principles and
procedures from behavioral science have been underappreciated and underused.
This Handbook teaches you how to apply behavioral science for safety achievement.
The research recommends we start with behavior. But the demonstrated validity of a
behavior-based approach does not mean the better-sounding, personal approaches should
not be used. It is important to consider the feelings and attitudes of employees, because it
takes people to implement the tools of behavior management.
This Handbook will teach you how certain feeling states critical for safety achievement—
self-esteem, empowerment, and belonging—can be increased by applying behavioral
science. It is possible to establish interpersonal interactions and behavioral consequences in
the workplace to increase important feelings and attitudes. I will show you how increasing
these feeling states benefits behavior and helps to achieve safety excellence.
As illustrated in Figure 1.6, an attitude of frustration or an internal state of distress
can certainly influence driving behaviors, and vice versa. Indeed, internal (unobserved)
personal states of mind continually influence observable behaviors, while changes in
observable behaviors continually affect changes in person states or attitudes. Thus, it is

Figure 1.6 Behavior influences attitude and attitude influences behavior.

820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 18

18 Psychology of safety handbook

possible to “think a person into safe behaviors” (through education, coaching, and con-
sensus-building exercises), and it is possible to “act a person into safe thinking” (through
behavior management techniques).
In an industrial setting, it is most cost effective to target behaviors first through behav-
ior management interventions (described in this text) implemented by employees them-
selves. Small changes in behavior can result in attitude change, followed by more behavior
change and more desired attitude change. This spiraling of behavior feeding attitude, atti-
tude feeding behavior, behaviors feeding attitudes and so on can lead to employees becom-
ing totally committed to safety achievement, as reflected in their daily behavior. And all
of this could start with a relatively insignificant behavior change in one employee—
a “small win.”

In conclusion
In this initial chapter, I have outlined the basic orientation and purpose of this text, which
are to teach principles for understanding the human aspects of occupational health and
safety, and to illustrate practical procedures for applying these principles to achieve signif-
icant improvements in organizational and community-wide safety.
The principles and procedures are not based on common sense nor intuition, but rather
on reliable scientific investigation. Some will contradict common folklore in pop psychol-
ogy and require shifts in traditional approaches to the management of organizational
safety. Approach this material with an open mind. Be ready to relinquish fads, fancies, and
folklore for innovations based on unpopular but research-supported theory.
I promise this “psychology of safety” is based on the latest and most reliable scientific
knowledge, and I promise greater safety achievement in your organization if you follow
the principles and procedures presented here. Reference to the research literature is given
throughout this text to verify the concepts, principles, and procedures discussed. Read
some of these yourself to experience the rewards of scientific inquiry and distance yourself
from the frivolity of common sense.

References
Bird, F. E., Jr. and Davies, R. J., Safety and the Bottom Line, Institute Publishing, Inc., Loganville, GA,
1996.
Bjurstrom, L. M., Priority to key areas and management by results in the national accident preven-
tion policy, in Advances in Industrial Ergonomics and Safety, Mital, A., Ed., Taylor & Francis,
London, 1989.
Buckingham, M. and Coffman, C., First, Break All the Rules: What the World’s Greatest Managers Do
Differently, Simon & Schuster, New York, 1999.
Cady, L. D., Thomas, P. C., and Karwasky, R. J., Program for increasing health and physical fitness
of fire fighters, J. Occup. Med., 27 110, 1985
Carton, J. S., The differential effects of tangible rewards and praise on intrinsic motivation: a com-
parison of cognitive evaluation theory and operant theory, Behav. Anal., 19, 237, 1996.
Cameron, J. and Pierce, W. D., Reinforcement, reward, and intrinsic motivation: a meta-analysis,
Rev. Educ. Res., 64, 363, 1994.
Daniels, A. C., Bringing out the Best in People, 2nd ed., McGraw-Hill, New York, 2000.
Deming, W. E., Quality, productivity, and competitive position, workshop presented by Quality
Enhancement Seminars, Inc., Cincinnati, OH, May 1991.
Eisenberger, R. and Cameron, J., Detrimental effects of reward: reality or myth?, Am. Psychol., 51,
1153, 1996.
820049_CRC1_L1540_CH01 11/11/00 11:41 AM Page 19

Chapter one: Choosing the right approach 19

Eisner, H. S. and Leger, J. P., The international safety rating system in South African mining, J.
Occup. Accid., 10, 141, 1988.
Geller, E. S., What’s in a perception survey?, Ind. Saf. Hyg. News, 28(11), 11, 1994a.
Geller, E. S., Survey reliability vs. validity, Ind. Saf. Hyg. News, 28(12), 12, 1994b.
Geller, E. S., How to select behavioral strategies, Ind. Saf. Hyg. News, 32(9), 12, 1998.
Guastello, S. J., Catastrophe modeling of the accident process: evaluation of an accident reduction
program using the Occupational Hazards Survey, Accid. Anal. Prev., 21, 61, 1989.
Guastello, S. J., Do we really know how well our occupational accident prevention programs
work?, Saf. Sci., 16, 445, 1993.
Heinrich, H. W., Industrial Accident Prevention, a Scientific Approach, McGraw-Hill, New York, 1931.
Ivancevich, J. M., Matteson, M. T., Freedman, S. M., and Phillips, J. S., Worksite stress management
interventions, Am. Psychol., 45, 252, 1990.
Kohn, A., Punished by Rewards: The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other
Bribes, Houghton Mifflin, New York, 1993.
Krause, T. R., Motivating employees for safety success, Prof. Saf., 45(3), 22, 2000.
Kroemer, K. H., Ergonomics. Encycl. Hum. Biol., 3, 473, 1991.
Murphy, L. R., Occupational stress management: a review and appraisal, Occup. Psychol., 57, 1,
1984.
Oborne, D. J., Branton, R., Leal, F., Shipley, P., and Stewart, T., Eds., Person-Centered Ergonomics: a
Brontonian View of Human Factors, Taylor & Francis, Washington, D.C., 1993.
Peter, L. J., The Peter Principle, Morrow, New York, 1969.
Petersen, D., Safe Behavior Reinforcement, Aloray, New York, 1989.
Pringle, D. R. S. and Brown, A. E., International safety rating system: New Zealand’s experience
with a successful strategy, J. Occup. Accid., 12, 41, 1990.
Rundmo, T., Risk perception and safety on offshore petroleum platforms. Part II. Perceived risk, job
stress and accidents, Saf. Sci., 15, 53, 1992.
Saarela, K. L., A poster campaign for improving safety on shipyard scaffolds, J. Saf. Res., 20, 177,
1989.
Saarela, K. L., An intervention program utilizing small groups: a comparative study, J. Saf. Res., 21,
149, 1990.
Saarela, K. L., Saari, J., and Aaltonen, M., The effects of an informal safety campaign in the ship
building industry, J. Occup. Accid., 10, 255, 1989.
Skinner, B. F., Beyond Freedom and Dignity, Alfred A. Knopf, New York, 1971.
Skinner, B. F., Walden Two, MacMillan, New York, 1948.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 21

chapter two

Starting with theory

In this chapter we consider the value of theory in guiding our approaches to safety and health
improvement. You will see how a vision for a Total Safety Culture is a necessary guide to achieve
safety excellence. A basic principle here is that safety performance results from the dynamic interac-
tion of environment, behavior, and person-based factors. Achieving a Total Safety Culture requires
attention to each of these. I make a case for integrating person-based and behavior-based psychology
in order to address most effectively the human dynamics of injury prevention.

“There’s nothing so practical as a good theory.”—Kurt Lewin

As you know, some safety efforts suffer from a “flavor of the month” syndrome. New pro-
cedures or intervention programs are tried seemingly at random, without an apparent
vision, plan, or supporting set of principles. When the mission and principles are not clear,
employees’ acceptance and involvement suffer.
Without a guiding theory or set of principles, it is difficult to design and refine proce-
dures to stay on course. This was the theme of Deming’s four-day workshops on Quality,
Productivity, and Competitive Position. Covey (1989, 1990) emphasizes the same point in
his popular books The Seven Habits of Highly Effective People, Principle-Centered Leadership
and First Things First co-authored by Merrill and Merrill (1994).
A theory or set of guiding principles makes it possible to evaluate the consistency and
validity of program goals and intervention strategies. By summarizing the appropriate the-
ory or principles into a mission statement, you have a standard for judging the value of
your company’s procedures, policies, and performance expectations.
It is important to develop a set of comprehensive principles on which to base safety
procedures and policies. Then teach these principles to your employees so they are under-
stood, accepted, and appreciated. This buy-in is certainly strengthened when employees or
associates help select the safety principles to follow and summarize them in a company
mission statement.
This is theory-based safety. A critical challenge, of course, is to choose the most relevant
theories or principles for your company culture and purpose, and develop an appropriate
and feasible mission statement that reflects the right theory.

The mission statement

Several years ago I worked with employees of a major chemical company to develop the
general mission statement for safety given in Figure 2.1. This vision for a Total Safety
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 22

22 Psychology of safety handbook

Mission Statement
A Total Safety Culture continually improves safety
performance. To that end, a TSC:
Promotes a work environment based on
employee involvement, ownership, teamwork,
education, training, and leadership.
Builds self-esteem, empowerment, pride,
enthusiasm, optimism, and encourages
innovation.
Reinforces the need for employees to actively
care about their fellow coworkers.
Promotes the philosophy that safety is not a
priority that can be reordered, but is a value
associated with every priority.
Recognizes group and individual achievement.

Figure 2.1 The principles and procedures covered in this Handbook are reflected in this
safety achievement mission statement.

Culture serves as a guideline or standard for the material presented throughout this book,
in the same way a corporate mission statement serves as a yardstick for gauging the devel-
opment and implementation of policies and procedures.
This mission statement might not be suited for all organizations, but it is based on
appropriate and comprehensive theory, supported by scientific data from research in psy-
chology. Before developing this statement, employees learned basic psychological theories
most relevant to improving occupational safety. These principles are illustrated through-
out this text, along with operational (real-world) definitions.

Theory as a map
I would like to relate an experience to show how a theory can be seen as a map to guide us
to a destination. The mission statement in Figure 2.1 reflects a destination for safety within
the realm of psychology. This story also reflects the difficulty in finding the best theory
among numerous possibilities.
I had the opportunity to conduct a training program at a company in Palatka, FL. My
client sent me step-by-step instructions to take me from Interstate 95 to Palatka. That was
my map, limited in scope for sure, but sufficient I presumed to get the job done—to get me
to the Holiday Inn in Palatka. But while at the National Car Rental desk, an attendant said
my client’s directions were incorrect and showed me the “correct way” with National’s
map of Jacksonville.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 23

Chapter two: Starting with theory 23

Figure 2.2 Start your journey with the right theory.

I was quick to give up my earlier theory (from my client’s handwritten instructions) for
this more professional display. After all, I now had a professionally printed map and direc-
tions from someone in the business of helping customers with travel plans—a consultant,
so to speak. But National’s map showed details for a limited area, and Palatka was not on
the map. I could not verify the attendant’s directions with the map, nor could I compare
these directions with my client’s very different instructions. Without a complete perspec-
tive, I chose the theory that looked best. And I got lost. As depicted in Figure 2.2, it is criti-
cal to start out with the right map (or theory).
After traveling 15 miles, I began to question the “National theory” and wondered
whether my client’s scribbling had been correct after all. But I stuck by my decision, and
drove another ten miles before exiting the highway in search of further instruction. I cer-
tainly needed to reach my destination that night, but motivation without appropriate direc-
tion can do more harm than good. In other words, a motivated worker cannot reach safety
goals with the wrong theory or principles.
It was late Sunday night and the gas station off the exit ramp was closed, but another
vehicle had also just stopped in the parking lot. I drove closer and announced to four
tough-looking, grubby characters in a pick-up truck loaded with motorcycles that I was
lost and wondered if they knew how to get to Palatka. None of these men had heard of
Palatka, but one pulled out a detailed map of Florida and eventually found the town of
Palatka. I could not see the details in the dark, but I accepted this new “theory” anyway,
with no personal verification.
The packaging of this theory was not impressive, but my back was against the wall. I
was desperate for a solution to my problem and had no other place to turn. As I left the
parking lot with a new theory, I wondered whether I was now on the right track. Perhaps,
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 24

24 Psychology of safety handbook

the theory obtained from the National Car rental attendants was correct and I had missed
an exit. Whom should I believe? Fortunately, I looked beyond the slick packaging and went
with the guy who had the more comprehensive perspective (the larger map). This theory
got me to the Holiday Inn Palatka.

Relevance to occupational safety

That evening I thought about my experience and its relevance to safety. It reminded me of
the dilemma facing many safety professionals when they choose approaches, programs,
and consultants to help solve people problems related to safety. Theories, research, and
tools in psychology are so vast and often so complex that it can be an overwhelming task
to select a theory or set of principles to follow. As discussed in Chapter 1, there is a huge
market of self-help books, audiotapes, and videotapes addressing concepts seemingly rel-
evant to understanding and managing the human dynamics of safety. Many of the anec-
dotes, principles, and procedures given in pop psychology books and audiotapes are
founded on limited or no scientific data. In fact, more of the material was probably used
because it sounded good rather than because systematic research found it valid.
The theory that got me to Palatka was not the most professional or believable, nor was
it “packaged” impressively. This does not mean you should avoid the slick, well-marketed
approaches to occupational safety. I only wish these factors were given much less weight
than scientific data.
It is relevant, though, that the more comprehensive map enabled me to find my desti-
nation. I have found that many of the human approaches to improving safety are limited
in scope or theoretical foundation. Many are sold or taught as packaged programs or step-
by-step procedures for any workplace culture.
In the long run, it is more useful to teach comprehensive theory and principles. On this
foundation, culture-relevant procedures and interventions can be crafted by employees
who will “own” and thus follow them. As the old saying goes, “Give a man a fish and you
feed him for a day, teach him how to fish and you feed him for a lifetime.”
At breakfast, I told the human resource manager and the safety director, the one who
gave me the handwritten instructions, about my problems finding Palatka. Interestingly,
each had a different theory on the best way to travel between the Jacksonville airport and
Palatka. The safety director stuck with his initial instructions. The human resource man-
ager recommended the route I eventually took. Their discussion was not enlightening. In
fact it got me more confused, because I did not have a visual picture or schema (a compre-
hensive map) in which to fit the various approaches (or routes) they were discussing. In
other words, I did not have a framework or paradigm to organize their verbal descriptions.
Without a relevant theory my experience taught me nothing, except the need for an appro-
priate theory—in this case a map.
A theory should serve as the map that provides direction to meet a specific safety chal-
lenge. Obviously, it is important to teach the basic theory to everyone who must meet the
challenge. Then it is a good idea to have an employee task force summarize the theory in a
safety mission statement. When the workforce understands the theory and accepts the
summary mission statement, intervention processes based on the theory will not be viewed
as “flavor of the month,” but as an action plan to bring the theory to life.
When employees appreciate and affirm the theory, they will get involved in designing
and implementing the action steps. They will also suggest ways to refine or expand action
plans and theory on the basis of systematic observations or scientific evidence. This is the
best kind of continuous improvement.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 25

Chapter two: Starting with theory 25

A basic mission and theory

The mission statement in Figure 2.1 reflects the ultimate in safety—a Total Safety Culture.
In a Total Safety Culture,

• Everyone feels responsible for safety and does something about it on a daily basis.
• People go beyond the call of duty to identify unsafe conditions and at-risk behav-
iors, and they intervene to correct them.
• Safe work practices are supported intermittently with rewarding feedback from
both peers and managers.
• People “actively care” continuously for the safety of themselves and others.
• Safety is not considered a priority that can be conveniently shifted depending on the
demands of the situation; rather, safety is considered a value linked with every pri-
ority of a given situation.

This Total Safety Culture mission is much easier said than done, but it is achievable
through a variety of safety processes rooted in the disciplines of engineering and psychol-
ogy. Generally, a Total Safety Culture requires continual attention to three domains.

1. Environment factors (including equipment, tools, physical layout, procedures,

standards, and temperature).
2. Person factors (including people’s attitudes, beliefs, and personalities).
3. Behavior factors (including safe and at-risk work practices, as well as going beyond
the call of duty to intervene on behalf of another person’s safety).

This triangle of safety-related factors has been termed “The Safety Triad” (Geller, 1989;
Geller et al., 1989) and is illustrated in Figure 2.3.
These three factors are dynamic and interactive. Changes in one factor eventually
impact the other two. For example, behaviors that reduce the probability of injury often
involve environmental change and lead to attitudes consistent with the safe behaviors. This
is especially true if the behaviors are viewed as voluntary. In other words, when people
choose to act safely, they act themselves into safe thinking. These behaviors often result in
some environmental change.

Person Environment
Knowledge, Skill, Abilities, Equipment, Tools, Machines
Intelligence, Motives, Housekeeping, Heat/Cold
Personality Safety Engineering, Standards,
Culture Operating Procedures

Behavior
Complying, Coaching,
Recognizing, Communicating,
Demonstrating ''Actively Caring''

Figure 2.3 A Total Safety Culture requires continual attention to three types of contributing
factors.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 26

26 Psychology of safety handbook

Figure 2.4 Performance results from the dynamic interaction of environment, behavior, and
person factors.

The behavior and person factors represent the human dynamics of occupational safety
and are addressed in this book. The basic principle here is that behavior-based and person-
based factors need to be addressed in order to achieve a Total Safety Culture. These two
divergent approaches to understanding and managing the human element represent the
psychology of injury prevention.
Paying attention to only behavior-based factors (the observable activities of people) or
to only person-based factors (unobservable feeling states or attitudes of people) is like
using a limited map to find a destination, as with my attempt to find Palatka, FL. The mis-
sion to achieve a Total Safety Culture requires a comprehensive framework—a complete
map of the relevant psychological territory. Figure 2.4 illustrates the complex interaction of
environment, person, and behavior factors.

Behavior-based vs. person-based approaches

There are numerous opinions and recommendations on how the psychology of safety can be
used to produce beneficial changes in people and organizations. Most can be classified into
one of two basic approaches: person-based and behavior-based. In fact, most of the numerous
psychotherapies available to treat developmental disabilities and psychological disorders,
from neurosis to psychosis, can be classified as essentially person-based or behavior-based.
That is, most psychotherapies focus on changing people either from the inside (“think-
ing people into acting differently”) or from the outside (“acting people into thinking dif-
ferently”). Person-based approaches attack individual attitudes or thinking processes
directly. They teach clients new thinking strategies or give them insight into the origin of
their abnormal or unhealthy thoughts, attitudes, or feelings. In contrast, behavior-based
approaches attack the clients’ behaviors directly. They change relationships between
behaviors and their consequences.
Many clinical psychologists use both person-based and behavior-based techniques
with their clients, depending upon the nature of the problem. Sometimes the same client is
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 27

Chapter two: Starting with theory 27

treated with person-based and behavior-based intervention strategies. I am convinced both

are relevant in certain ways for improving health and safety. This text will show you how
to integrate relevant principles from these two psychological approaches in order to
achieve a Total Safety Culture.

The person-based approach

Imagine you see two employees pushing each other in a parking lot as a crowd gathers
around to watch. Is this aggressive behavior, horseplay, or mutual instruction for
self-defense? Are the employees physically attacking each other to inflict harm or does
this physical contact indicate a special friendship and mutual understanding of the line
between aggression and play? Perhaps if you watch longer and pay attention to verbal
behavior, you will decide whether this is aggression, horseplay, or a teaching/learning
demonstration.
However, a truly accurate account might require you to assess each individual’s per-
sonal feelings, attitudes, or intentions. It is possible, in fact, that one person was being hos-
tile while the other was just having fun or the contact started as horseplay and progressed
to aggression.
This scenario illustrates a basic premise of the person-based approach. Focusing only
on observable behavior does not explain enough. People are much more than their behav-
iors. Concepts like intention, creativity, intrinsic motivation, subjective interpretation, self-
esteem, and mental attitude are essential to understanding and appreciating the human
dynamics of a problem. Thus, a person-based approach in the workplace applies surveys,
personal interviews, and focus-group discussions to find out how individuals feel about
certain situations, conditions, behaviors, or personal interactions.
A wide range of therapies fall within the general framework of person-based, from the
psychoanalytic techniques of Sigmund Freud, Alfred Adler, and Carl Jung to the client-
centered humanism developed and practiced by Carl Rogers, Abraham Maslow, and
Viktor Frankl (Wandersman et al., 1976). Humanism is the most popular person-based
approach today, as evidenced by the current market of pop psychology videotapes, audio-
tapes, and self-help books. Some current popular industrial psychology tools—such as the
Myers-Briggs Type Indicator and other trait measures of personality, motivation, or risk
taking propensity—stem from psychoanalytic theory and practice.
The key principles of humanism found in most pop psychology approaches to increase
personal achievement are

1. Everyone is unique in numerous ways. The special characteristics of individuals

cannot be understood or appreciated by applying general principles or concepts,
such as the behavior-based principles of performance management or the perma-
nent personality trait perspective of psychoanalysis.
2. Individuals have far more potential to achieve than they typically realize and
should not feel hampered by past experiences or present liabilities.
3. The present state of an individual in terms of feeling, thinking, and believing is a
critical determinant of personal success.
4. One’s self-concept influences mental and physical health, as well as personal effec-
tiveness and achievement.
5. Ineffectiveness and abnormal thinking and behavior result from large discrep-
ancies between one’s real self (“who I am”) and ideal self (“who I would like
to be”).
6. Individual motives vary widely and come from within a person.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 28

28 Psychology of safety handbook

Readers familiar with the writings of Deming (1986, 1993) and Covey (1989, 1990) will
recognize these eminent industrial consultants as humanists, or advocates of a person-
based approach.

The behavior-based approach

The behavior-based approach to applied psychology is founded on behavioral science as
conceptualized and researched by Skinner (1938, 1974). In his experimental analysis of
behavior, Skinner rejected for scientific study unobservable factors such as self-esteem,
intentions, and attitudes. He researched only observable behavior and its social, environ-
mental, and physiological determinants. The behavior-based approach starts by identify-
ing observable behaviors targeted for change and the environmental conditions or
contingencies that can be manipulated to influence the target behavior(s) in desired direc-
tions. (Contingencies are relationships between designated target behaviors and their sup-
porting consequences).
The basic idea is that behavior can be objectively studied and changed by identifying
and manipulating environmental conditions (or stimuli) that immediately precede and fol-
low a target behavior. The antecedent conditions (which I call “activators”) signal when
behavior can achieve a pleasant consequence (a reward) or avoid an unpleasant conse-
quence (a penalty). Therefore, activators direct behavior, and consequences determine
whether the behavior will recur. Accordingly, people are motivated by the consequences
they expect to receive, escape, or avoid after performing a target behavior.
Humanists maintain that this ABC (activator—behavior—consequence) analysis is
much too simple to explain human behavior. For many applications, they are right.
However, as shown in Figure 2.5 any of our daily behaviors are directed by preceding acti-
vators and motivated by ensuing consequences. I have much more to say about this ABC
approach to understanding and improving behavior in Sections 3 and 4 of this Handbook.
Then in Section 5, I explain how the person-based approach of the humanists can be inte-
grated with the behavior-based approach to bring out the best in people and their organi-
zations for the sake of achieving a Total Safety Culture.

Figure 2.5 It is an S-R world after all.

820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 29

Chapter two: Starting with theory 29

Considering cost effectiveness

When people act in certain ways, they usually adjust their mental attitude and self-talk to
parallel their actions (Festinger, 1957); when people change their attitudes, values, or think-
ing strategies, certain behaviors change as a result (Fishbein and Ajzen, 1975). Thus,
person-based and behavior-based approaches to changing people can influence both atti-
tudes and behaviors, either directly or indirectly. Most parents, teachers, first-line supervi-
sors, and safety managers use both approaches in their attempts to change a person’s
knowledge, skills, attitudes, or behaviors.

• When we lecture, counsel, or educate others in a one-on-one or group situation, we

are essentially using a person-based approach.
• When we recognize, correct, or discipline others for what they have done, we are
operating from a behavior-based perspective.

Unfortunately, we are not always effective with our person-based or behavior-based

change techniques, and often we do not know whether our intervention worked as
intended. In order to apply person-based techniques to psychotherapy, clinical psycholo-
gists receive specialized therapy or counseling training for four or more years, followed by
an internship of at least one year. This intensive training is needed because tapping into an
individual’s perceptions, attitudes, and thinking styles is a demanding and complex
process. Also, internal dimensions of people are extremely difficult to measure reliably,
making it cumbersome to assess therapeutic progress and obtain straightforward feedback
regarding therapy skills. Consequently, the person-based therapy process can be very time-
consuming, requiring numerous one-on-one sessions between professional therapist and
client.
In contrast, behavior-based psychotherapy was designed to be administered by indi-
viduals with minimal professional training. From the start, the idea was to reach people
where problems occur—in the home, school, rehabilitation institute, and workplace, for
example—and to teach parents, teachers, supervisors, friends, or coworkers the behavior-
change techniques most likely to work under the circumstances (Ullman and Krasner, 1965).
More than three decades of research have shown convincingly that this on-site
approach is cost effective, primarily because behavior-change techniques are straightfor-
ward and relatively easy to administer and because intervention progress can be readily
monitored by the ongoing observation of target behaviors. By obtaining objective feedback
on the impact of intervention techniques, a behavior-based process can be continually
refined or altered.
Behavior-based methods are especially cost effective for large-scale applications. Much
community-based and organizational research has shown substantial improvements in
environmental, transportation, production, and health-related problems as a direct result
of this approach to intervention (e.g., see comprehensive research reviews by Elder et al.,
1994; Geller et al., 1982; Goldstein and Krasner, 1987; Greene et al., 1987). And there is
plenty of evidence that the behavior-based approach can dramatically improve an organi-
zation’s safety performance (e.g., DePasquale and Geller, 1999; McSween, 1995; Petersen,
1989; Ward, 2000).

Integrating approaches
A common perspective, even among psychologists, is that humanists and behaviorists are
complete opposites (Newman, 1992; Wandersmann et al., 1976). Behaviorists are consid-
ered cold, objective, and mechanistic, operating with minimal concern for people’s feel-
ings. In contrast, humanists are thought of as warm, subjective, and caring, with limited
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 30

30 Psychology of safety handbook

concern for directly changing another person’s behavior or attitude. The basic humanistic
approach is termed “nondirective” or “client-centered.” Therapists, counselors, or coaches
do not directly change their clients, but rather provide empathy and a caring and support-
ive environment for enabling clients to change themselves, from the inside out.
Given the foundations of humanism and behaviorism, it is easy to build barriers
between person-based and behavior-based perspectives and assume you must follow one
or the other when designing an intervention process. In fact, many consultants in the safety
management field market themselves as using one or the other approach, but not both. It
is my firm belief that these approaches need to be integrated in order to truly understand
the psychology of safety and build a Total Safety Culture.

In conclusion
Theory or basic principles are needed to organize research findings and guide our
approaches to improve the safety and health of an organization. Similarly, a vision for a
Total Safety Culture incorporated into a mission statement is needed to guide us in devel-
oping action plans to achieve safety excellence.
When employees understand and accept the mission statement and guiding princi-
ples, they become more involved in the mission. The action plan will not be viewed as one
more flavor of the month, but as relevant to the right principles and useful for achieving
shared goals. Indeed, the workforce will help design and implement the action plans. This
is crucial for cultivating a Total Safety Culture.
A basic principle introduced in this chapter is that the safety performance of an organ-
ization results from the dynamic interaction of environment, behavior, and person factors.
The behavior and person dimensions represent the human aspect of industrial safety and
reflect two divergent approaches to understanding the psychology of injury prevention.
Figure 2.6 summarizes the distinction between person-based and behavior-based psy-
chology and shows that both approaches contribute to understanding and helping people.
Both the internal and external dimensions of people are covered in this Handbook as they relate
to improving organizational safety. Profound knowledge on the person side comes from cog-
nitive science, whereas the behavior-based approach is founded on behavioral science.
The best I can do is provide education by improving your knowledge and thinking
about the human dynamics of safety improvement. You will do the training by using the
observation and feedback techniques detailed later in Section 4 to improve your own or
someone else’s behavior.
Taken alone, the behavior-based approach is more cost effective than the person-based
approach in affecting large-scale change. But it cannot be effective unless the work culture
believes in the behavior-based principles and willingly applies them to achieve the mutual
safety mission. This involves a person-based approach. Therefore, to achieve a Total Safety
Culture we need to integrate person-based and behavior-based psychology. This text
shows you how to meet this challenge.

References
Covey, S. R., The Seven Habits of Highly Effective People: Restoring the Character Ethic, Simon &
Schuster, New York, 1989.
Covey, S. R., Principle-Centered Leadership, Simon & Schuster, New York, 1990.
Covey, S. R., Merril, A. R., and Merril, R. R., First Things First, Simon & Schuster, New York, 1994.
Deming, W. E., Out of the Crisis, Massachusetts Institute of Technology, Center for Advanced
Engineering Study, Cambridge, MA, 1986.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 31

Chapter two: Starting with theory 31

People

Internal External
States Traits: Behaviors:
attitudes, beliefs, coaching,
feelings, thoughts, recognizing,
personalities, complying,
perceptions, and communicating, and
values, intentions actively caring

* Education * Training
* Person Based
* Behaviour-Based
* Cognitive Science
* Behavioral Science
Behavioral Audits
* Perception Surveys *

Figure 2.6 The internal and external aspects of people determine the success of a safety
process.
Deming, W. E., Quality, productivity, and competitive position, workshop presented by Quality
Enhancement Seminars, Inc., Cincinnati, OH, May 1991.
Deming, W. E., The New Economics for Industry, Government, Education, Massachusetts Institute of
Technology, Center for Advanced Engineering Study, Cambridge, MA, 1993.
DePasquale, J. D. and Geller, E. S., Critical success factors for behavior-based safety: a study of
twenty industry-wide applications, J. Saf. Res., 30, 237, 1999.
Elder, J. P., Geller, E. S., Hovell, M. F., and Mayer, J. A., Motivating Health Behavior, Delmar
Publishers, New York, 1994.
Festinger, L., A Theory of Cognitive Dissonance, Stanford University Press, Stanford, CA, 1957.
Fishbein, M. and Ajzen, I., Belief, Attitude, Intention, and Behavior: an Introduction to Theory and
Research, Addison-Wesley, Reading, MA, 1975.
Geller, E. S., Managing occupational safety in the auto industry, J. Organ. Behav. Manage., 10(1), 181,
1989.
Geller, E. S., Lehman, G. R., and Kalsher, M. R., Behavior Analysis Training for Occupational Safety,
Make-A-Difference, Inc., Newport, VA, 1989.
Geller, E. S., Winett, R. A., and Everett, P. B., Preserving the Environment: New Strategies for Behavior
Change, Pergamon Press, Elmsford, NY, 1982.
Goldstein, A. P. and Krasner, L., Modern Applied Psychology, Pergamon Press, New York, 1987.
Greene, B. F., Winett, R. A., Van Houten, R., Geller, E. S., and Iwata, B. A., Eds., Behavior Analysis in
the Community: Readings from the Journal of Applied Behavior Analysis, University of Kansas Press,
Lawrence, KS, 1987.
McSween, T. E., The Values-Based Safety Process: Improving Your Safety Culture with a Behavioral
Approach, Van Nostrand Reinhold, New York, 1995.
820049_CRC1_L1540_CH02 11/10/00 1:30 PM Page 32

32 Psychology of safety handbook

Newman, B., The Reluctant Alliance: Behaviorism and Humanism, Prometheus Books, Buffalo, NY,
1992.
Petersen, D., Safe Behavior Reinforcement. Alorey, Inc., New York, 1989.
Skinner, B. F., The Behavior of Organisms, Copley Publishing Group, Acton, MA, 1938.
Skinner, B. F., About Behaviorism, Alfred A. Knopf, New York, 1974.
Ullman, L. P. and Krasner, L., Eds., Case Studies in Behavior Modification, Holt, Rinehart, & Winston,
New York, 1965.
Wandersman, A., Popper, P., and Ricks, D., Humanism and Behaviorism: Dialogue and Growth,
Pergamon Press, New York, 1976.
Ward, S., One size doesn’t fit all: customizing helps merge behavioral and traditional approaches,
Prof. Saf., 45(3), 33, 2000.
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 33

chapter three

Paradigm shifts for total safety

This chapter outlines ten new perspectives we need to adopt in order to exceed current levels of safety
excellence and reach our ultimate goal—a Total Safety Culture. The traditional three Es of safety
management—engineering, education, and enforcement—have only gotten us so far. A Total Safety
Culture requires understanding and applying three additional Es—empowerment, ergonomics, and
evaluation.

“Mindsets are yesterday—mind growth is tomorrow.”—Joe Batten

Safety in industry has improved dramatically in this century. Let us examine the evolution
of accident prevention to see how this was accomplished. The first systematic research
began in the early 1900s and focused on finding the psychological causes of accidents. It
assumed people were responsible for most accidents and injuries, usually through mental
errors caused by anxiety, attitude, fear, stress, personality, or emotional state (Guarnieri,
1992). Reducing accidents was typically attempted by “readjusting” attitude or personal-
ity, usually through supervisor counseling or discipline (Heinrich, 1931).
This so-called “psychological approach” held that certain individuals were “accident
prone.” By removing these workers from risky jobs or by disciplining them to correct their
attitude or personality problems, it was thought that accidents could be reduced. As I dis-
cussed in Chapter 1, this focus on accident proneness has not been effective, partly because
reliable and valid measurement procedures are not available. Also, the person factors con-
tributing to accident proneness are probably not consistent characteristics or traits within
people, but vary from time to time and situation to situation.

The old three Es

Enthusiasm for the early “psychological approach” waned because of the difficulty meas-
uring its impact (Barry, 1975). In addition, the seminal research and scholarship of Haddon
(1963, 1968) suggested that engineering changes held the most promise for large-scale,
long-term reductions in injury severity.
As the first administrator of the National Highway Safety Bureau [now the National
Highway Traffic Safety Administration (NHTSA)], Haddon was able to turn his theory and
research into the first federal automobile safety standards. Haddon believed injury is
caused by delivering excess energy to the body, and that injury prevention depends on con-
trolling that energy. The prevention focus now shifted to engineering and epidemiology,
and resulted in developing personal protective equipment (PPE) for work and recreational
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 34

34 Psychology of safety handbook

environments, as well as standards and policy regarding the use of PPE. In vehicles,
Haddon’s basic theory eventually led to collapsible steering wheels, padded dashboards,
head restraints, and air bags in automobiles.
This brief history of the safety movement in the United States explains why engineer-
ing is the dominant paradigm in industrial health and safety (Petersen, 1991; Winn and
Probert, 1995), with secondary emphasis on two additional “Es”—education and enforce-
ment. Over the past several decades, the basic protocol for reducing injury has been to

1. Engineer the safest equipment, environmental settings, and protective devices.

2. Educate people regarding the use of the engineering interventions.
3. Use discipline to enforce compliance with recommended safe work practices.

Thanks to this paradigm most safety professionals are safety engineers who commonly
advocate that “Safety is a condition of employment.”
The three Es have dramatically reduced injury severity in the workplace, at home, and
on the road. Let us take a look at motor vehicle safety for a minute. The Government
Accounting Office has estimated conservatively that the early automobile safety standards
ushered through Congress by Haddon had saved at least 28,000 American lives by 1974
(Guarnieri, 1992). In addition, the state laws passed in the 1980s requiring use of vehicle
safety belts and child safety seats have saved countless more lives. Many more lives would
be saved and injuries avoided if more people buckled up and used child safety seats for
their children.
The current rate of safety belt use in the United States is about 70 percent (NHTSA,
2000), a dramatic improvement from the 15 percent prior to statewide interventions,
including belt-use laws, campaigns to educate people about the value of safety-belt use,
and large-scale enforcement blitzes by local and state police officers.
There is still much room for improvement, especially considering that most of the riski-
est drivers still do not buckle up (Evans et al., 1982; Wagenaar, 1984; Waller, 1987). Each
year since 1990, the U.S. Department of Transportation has set nationwide belt-use goals of
70 percent, but to date this goal has not been met—at least over the long term. It seems the
effectiveness of current methods to increase the use of this particular type of PPE has
plateaued or asymptoted around 70 percent. Recently, President Clinton set the U.S.
buckle-up goal at 85 percent by the year 2005. We just cannot get there with the same old
intervention approaches.
Now, let us turn our attention to industry. I have worked with many corporate safety
professionals over the years who say their plant’s safety performance has reached a
plateau. Yes, their overall safety record is vastly better than it once was, but continuous
improvement is elusive. A frantic search for ways to take safety to the next level has not
paid off. The old “three Es” paradigm will not get us there. A certain percentage of people
keep falling through the cracks. Keep on doing what you are doing and you will keep on
getting what you are getting. As I heard Deming (1991) say many times, “Goals without
method, what could be worse?”

Three new Es
This book discusses the three new Es—ergonomics, empowerment, and evaluation. I cer-
tainly do not suggest abandoning tradition. We need to maintain a focus on engineering,
education, and enforcement strategies. But to get beyond current plateaus and reach new
heights in safety excellence, we must attend more competently to the psychology of injury
prevention. These three new Es suggest specific directions or principles.
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 35

Chapter three: Paradigm shifts for total safety 35

Ergonomics
As discussed in Chapter 1, ergonomics requires careful study of relationships between envi-
ronment and behaviors, as well as developing action plans (such as equipment work
orders, safer operating procedures, training exercises) to avoid possible acute or chronic
injury from the environment –behavior interaction. This requires consistent and voluntary
participation by those who perform the behaviors in the various work environments. These
are usually line operators or hourly workers in an organization, and their participation will
happen when these individuals are empowered. Throughout this book, I discuss ways to
develop an empowered work culture and I explain procedures for involving employees in
ergonomic interventions.

Empowerment
Some operational definitions of the traditional three Es for safety (especially enforcement)
have been detrimental to employee empowerment. Many supervisors have translated
“enforcement” into a strict punishment approach, and the result has turned off many
employees to safety programs. These workers may do what is required, but no more. Some
individuals who feel especially controlled by safety regulations might try to beat the sys-
tem, and success will likely bring a sense of gratification or freedom. This is predictable
from theory and research in the area of psychological reactance (Brehm, 1966, 1972) and is
illustrated in Figure 3.1.
I discuss this principle in more detail later, especially as it relates to developing behav-
ior change interventions. At this point, I want you to understand that some types of
enforcement are likely to inhibit empowerment and should be reconsidered and refined.
Paradigms must change—the theme of this chapter.

Figure 3.1 Some top-down rules have undesirable side effects.

820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 36

36 Psychology of safety handbook

Evaluation
The third new E word essential to achieving a Total Safety Culture is evaluation. Without
appropriate feedback or evaluation, practice does not make perfect. Thus, we need the
right kind of evaluation processes. Later in this book, especially Chapter 15, I detail proce-
dures for conducting the right kind of comprehensive evaluation. Right now, what is
important to understand is that some traditional methods of evaluation actually decrease
or stifle empowerment. This calls for changing some safety measurement paradigms.
Remember the need for a guiding theory or set of principles? Basic theory from person-
based and behavior-based psychology suggests shifts to new safety paradigms. These par-
adigm shifts provide a new set of guiding principles for achieving new heights in safety
excellence.

Shifting paradigms
I have heard many definitions of “paradigm,” some humorous, some academic, and some
practical. From my perspective, this is one of those superfluous academic terms that is com-
pletely unnecessary. We have simple and straightforward words in the English language to
cover every definition of paradigm. Perhaps that is why I often get humorous or sarcastic
reactions from audiences when I ask, “What’s a paradigm?”

“Isn’t that 20 cents?” shouts one participant. (Get it—”pair-a-dimes.”)

Another participant replies, “A paradigm is what I use on the farm to dig post holes.”
(Get it—“pair-a-dig-ems.”)

When I was a graduate student of psychology in the mid-1960s, paradigm was used to
refer to a particular experimental procedure or methodology in psychological research. For
this discussion, I consulted three different dictionaries (Webster’s New Universal Unabridged,
The American Heritage Dictionary, and The Scribner-Bantam English Dictionary) and came up
with a consensus definition for paradigm. It is a pattern, example, or model. However,
words can change their meaning through usage, as discussed brilliantly by Hayakawa
(1978) in his instructive and provocative text Language in Thought and Action. In business,
paradigm has been equated with psychological terms such as perception, attitude, cogni-
tion, belief, and value.
The popular 1989 video Discovering the Future: The Business of Paradigms by Joel Barker
(see also Barker, 1992) was certainly responsible for some of the new applications of the
term “paradigm.” A number of articles and speeches in the safety field have supported and
precipitated this change. Indeed, Dan Petersen’s keynote speech at the 1993 Professional
Development Conference of the American Society of Safety Engineers was entitled,
“Dealing with Safety’s Paradigm Shift,” and followed up his earlier 1991 article in
Professional Safety entitled “Safety’s Paradigm Shift.” Here, Petersen (1991) claimed that
safety has shifted its focus to large-scale culture change through employee involvement.
Some safety professionals, however, assert that a revolutionary change in ideas, beliefs,
and approaches—the new definition of paradigm—has not yet occurred in safety (Winn,
1992; Winn and Probert, 1995).
The aim of this chapter is not to dissect the meaning of paradigm nor to debate whether
one or more paradigm shifts have occurred in industrial safety. Instead, I want to define ten
basic changes in belief, attitude, or perception that are needed to develop the ultimate Total
Safety Culture. These shifts require new principles, approaches, or procedures, and will
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 37

Chapter three: Paradigm shifts for total safety 37

result in different behaviors and attitudes among top managers and hourly workers.
Empowerment will increase throughout the work culture.
The shift in how paradigm is commonly defined does contain an important lesson.
When we adopt and use new definitions, our mindset or perception changes. In other
words, as I indicated in the previous chapter, we act ourselves into a new way of thinking
or perceiving. This is a primary theme of this book. When employees get involved in more
effective procedures to control safety, they develop a more constructive and optimistic
attitude toward safety and the achievement of a Total Safety Culture. Let us consider the
shifts in principles, procedures, beliefs, attitudes, or perceptions needed for the three new
Es—ergonomics, empowerment, and evaluation—and for achieving a Total Safety Culture.

From government regulation to corporate responsibility

Many safety activities and programs in U.S. industry are driven by OSHA (the U.S.
Occupational Safety and Health Administration) or MSHA (the Mine Safety and Health
Administration) rather than by the employers and employees who can benefit from a
safety process. In other words, many in industry do “safety stuff” because the government
requires it—not because it was their idea and initiative.
People are more motivated and willing to go beyond the call of duty when they are
achieving their own self-initiated goals. Ownership, commitment, and proactive behaviors
are less likely when you are working to avoid missing goals or deadlines set by someone
else. This statement is intuitive and reflected in Figure 3.2. Just compare your own motiva-
tion when working for personal gain vs. someone else’s gain or when working to earn a
reward vs. to avoid a penalty.
The language used to define safety programs and activities influences personal partic-
ipation. Remember, we can act ourselves into an attitude. So it makes sense to talk about
safety as a company mission that is owned and achieved by the very people it benefits. A

Figure 3.2 Top-down control stifles creativity.

820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 38

38 Psychology of safety handbook

safety process is not intended to benefit federal regulators. Let us work to achieve a Total
Safety Culture for the right reasons.

From failure oriented to achievement oriented

If you strive to meet someone else’s goals rather than your own, you will probably develop
an attitude of “working to avoid failure” rather than “working to achieve success.” We are
more motivated by achieving success than avoiding failure. If you have a choice between
earning positive reinforcers (rewards) or avoiding negative reinforcers (punishers), you
will probably choose the positive reinforcement situation. Moreover, if you feel controlled
by negative reinforcement, you often procrastinate and take a reactive rather than a pro-
active stance (Skinner, 1971).
Figure 3.3 illustrates what I mean. The runner will surely start running, but how long
will he run? When the coach is not around to threaten a negative consequence for not run-
ning, will he keep going? Will he practice on his own to improve his running skills? Will he
hold himself accountable to be the best he can be on the running track? A “yes” answer to
these questions will only occur if the runner can put himself in an achievement-oriented
mindset. This is difficult in the enforcement context established by the coach.
This principle helps explain why more continuous and proactive attention goes to pro-
ductivity and quality than to safety. Productivity and quality goals are typically stated in
achievement terms, and gains are tracked and recorded as individual or team accomplish-
ments, sometimes followed by rewards or recognition awards.
In contrast, safety goals are most often stated in negative reinforcement terms. How many
times have you heard, “We will reach our safety goal after another month without a lost-
time injury,” and “keeping score” in safety means tracking and recording losses or injuries?

Figure 3.3 Working to avoid failure is not fun.

820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 39

Chapter three: Paradigm shifts for total safety 39

Measuring safety with only records of injuries not only limits evaluation to a reactive
stance, but it also sets up a negative motivational system that is apt to take a back seat to
the positive system used for productivity and quality. Giving safety an achievement per-
spective (like production and quality) requires a different scoring system, as indicated by
the next paradigm shift.

From outcome focused to behavior focused

Companies are frequently ranked according to their OSHA recordables and lost-time
injuries. Within companies, work groups or individual workers earn safety awards accord-
ing to outcomes—those with the lowest numbers win. Offering incentives for fewer
injuries, for instance, can often reduce the reported numbers while not improving safety.
Pressure to reduce outcomes without changing the process (or ongoing behaviors) often
causes employees to cover up their injuries. How many times have you heard of an injured
employee being driven to work each day to sign in and then promptly returned to the hos-
pital or home to recuperate? This keeps the outcome numbers low, but does more harm
than good to the corporate culture. Likewise, failure to report even a minor first-aid case
prohibits key personnel from correcting the factors that led to the incident.
A misguided emphasis on outcomes rather than process is illustrated in Figure 3.4.
Although the idea of a dead person receiving a safety reward is clearly ridiculous, this type
of incentive/reward process is quite common in American industry. A 1993 survey of more
than 400 companies in Wisconsin revealed 58 percent used rewards to motivate safety. Of
these, more than 85 percent based rewards on outcomes such as OSHA recordables rather
than process (Koepnick, 1993). These programs often bring down numbers by influencing
the reporting of injuries, but rarely do they benefit the safety processes which control results.
A scoring system based on what people do for safety (as in a behavior-based process)
not only attacks a contributing factor in most work injuries, it can also be achievement ori-
ented. This puts safety in the same motivational framework as productivity and quality.
In Chapter 11, I explain principles for establishing an incentive/reward process to
motivate the kinds of safety processes that influence outcomes. For now, just recognize and
appreciate the advantage of focusing on achieving process improvements over working to

Figure 3.4 Safety reward programs should pass the “dead-man’s test.”
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 40

40 Psychology of safety handbook

avoid failure. This is especially true if a failure-oriented goal is remote, such as a plant-wide
reduction in injuries, and thus might be perceived as uncontrollable.
Safety can be on equal footing with productivity and quality if it is recorded and
tracked with an achievement score perceived by employees as directly controllable and
attainable. This occurs with a focus on the safety processes that can decrease an organiza-
tion’s injury rate, as well as an ongoing measurement system that continuously tracks
safety accomplishments and displays them to the workforce.
As discussed in Chapter 2, safety accomplishments occur in three general areas, envi-
ronment, behavioral, and personal, with environmental successes easiest to record and track.
Environmental achievements for safety range widely, from purchasing safer equipment, to
correcting environmental hazards and demonstrating improved environmental audits.
Person factors are influenced by numerous situations, such as safety education, safety
celebrations, and increased safety personnel. It is possible to estimate achievements in this
domain by counting the occurrences of these events. A more direct assessment can occur
through periodic perception surveys, interviews, or focus-group discussions (as detailed
later in Chapter 15). These measurements can be rather time-consuming, though, and the
reliability and validity of results from intermittent subjective surveys are equivocal.
Moreover, finding an improvement in perceptions does not necessarily imply an increase
in safe work practices—the human dynamic most directly linked to reducing work injuries.
Work practices can be observed, recorded, and tracked objectively (Geller, 1998b;
Geller et al., 1989; Krauss et al., 1996; McSween, 1995). When daily displays of behavioral
records show increases in safe behaviors and decreases in at-risk behaviors, the workforce
can celebrate the success of an improved safety process. In Section 4 of this book, I detail
principles and procedures for accomplishing this.

From top-down control to bottom-up involvement

As I discussed when introducing three new Es, a Total Safety Culture requires continual
involvement from operations personnel, such as hourly workers. After all, these are the
people who know where safety hazards are located and when the at-risk behaviors occur.
Also, they can have the most influence in supporting safe behaviors and correcting at-risk
behaviors and conditions. In fact, the ongoing processes involved in developing a Total
Safety Culture need to be supported from the top but driven from the bottom. This is more
than employee participation; it is employee ownership, commitment, and empowerment.
As discussed in Chapter 1, research has shown that safe work practices can be
increased and work injuries decreased with behavior-based interventions (Geller, 1990;
Komaki et al., 1980; Sulzer-Azaroff, 1982, 1987). This research invariably involved outside
agents such as consultants to help implement and evaluate the tactics, and the projects
were usually short-term and small-scale. Large-scale and long-term behavior change
requires employees themselves to apply the techniques throughout their workplace. For
this to happen, employees must understand the relevant behavioral science principles and
feel good about using them to prevent work injuries.
Understanding and feeling good about something brings us to considering again those
person factors such as knowledge, intentions, attitudes, expectancies, and mood states.
Certain dispositions or mood states, for example, influence an individual’s propensity to
help another person, and it is possible to increase these personal factors through changing
environmental and behavioral factors (see reviews by Carlson et al., 1988; and Geller, 1994,
1998a,b).
This supports the general principle I introduced in Chapter 2. A Total Safety Culture
requires integrating both behavior-based and person-based approaches to understand and
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 41

Chapter three: Paradigm shifts for total safety 41

influence the human dynamics of a corporation. To show you how to do this is my primary
aim with this book.

From rugged individualism to interdependent teamwork

An employee-driven safety process requires teamwork founded on interpersonal trust,
synergy, and win–win contingencies. However, from childhood most of us have been
taught an individualistic, win–lose perspective, supported by such popular slogans as
“You have to blow your own horn,” “Nice guys finish last,” “No one can fill your shoes like
you,” and “It’s the squeaky wheel that gets the grease.” Furthermore, as shown in Figure
3.5, grades in school, the legal system, and many sports orient us to think win–lose inde-
pendence rather than win –win interdependence. This is why a true team approach to
safety does not come easily.
Figure 3.6 illustrates a competitive situation quite common in the workplace. Although
some office environments were originally designed to promote more open communication
and group interaction, physical and psychological barriers have often been erected to
maintain privacy and an individualistic atmosphere. This partially results from work sys-
tems that offer more rewards for individual and group achievement. Processes and systems
can be implemented to promote group behaviors and interdependence over individual
behaviors and independence. These processes and contingencies are emphasized through-
out this book, because a Total Safety Culture requires more interdependent teamwork than
rugged individualism.

From a piecemeal to a systems approach

The long-term improvements of a Total Safety Culture can only be achieved with a systems
approach, including balanced attention to all aspects of the corporate culture. Deming

Figure 3.5 U.S. culture promotes more independence than interdependence.

820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 42

42 Psychology of safety handbook

Figure 3.6 Some work environments create barriers to synergy.

(1986, 1993) emphasized that total quality only can be achieved through a systems
approach, and of course the same is true for safety. As I discussed earlier in Chapter 2, three
basic domains need attention when designing and evaluating safety processes and when
investigating the root causes of near hits and injuries.

1. Environment factors such as equipment, tools, machines, housekeeping, heat/cold,

and engineering.
2. Person factors such as employees’ knowledge, skills, abilities, intelligence, motives,
and personality.
3. Behavior factors such as complying, coaching, recognizing, communicating, and
“actively caring.”

Two of these system variables involve human factors. Each generally receives less
attention than the environment, mostly because it is more difficult to visibly measure the
outcomes of efforts to change the human factors. Some human factors programs focus on
behaviors (as in behavior-based safety); others focus on attitudes (as in a person-based
approach). A Total Safety Culture integrates these two approaches.

From fault finding to fact finding

Blaming an individual or group of individuals for an injury-producing incident is not con-
sistent with a systems approach to safety. Instead, an injury or near hit provides an oppor-
tunity to gather facts from all aspects of the system that could have contributed to the
incident. However, most evaluations of near hits or injuries are incomplete, and are much
less informative than they could be. Part of the problem here is the very term we use to
describe the process—accident investigation.
Accident investigation is a common phrase in industrial safety and health, but what
does it mean? Or more to the point, what does it imply? It is a basic job requirement for
safety pros, and they attend professional development workshops to improve their skills
in this area. Really, what is your assignment when investigating an accident? Let us look at
the language we are using.
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 43

Chapter three: Paradigm shifts for total safety 43

The word “accident” implies “a chance occurrence” outside your immediate control.
The first definition of “accident” in my New Merriam-Webster Dictionary (1989) is “an event
occurring by chance or unintentionally” (page 23). When a child has an “accident” in his
pants, we presume he was not in control. He could not help it.
And what about the word “investigation?” Does this term not imply a hunt for some
single cause or person to blame for a particular incident, as in “criminal investigation?”
How can we promote fact-finding over fault-finding with a term like “investigation” defin-
ing our job assignment?
Truly, to learn more about how to prevent injuries from an analysis of an incident,
we need to approach the task with a different mindset. It is not “accident investigation”—
it is “incident analysis.” This simple substitution of words can have great impact. We
can get more employee participation in the process and reap more benefits. I suggest
the following shifts in perspective and approach toward the evaluation of a near hit or
injury.
Gain a broader understanding. A common myth in the safety field holds that
injuries are caused by one critical factor—the root cause. “Ask enough questions,” advises
the safety consultant, “and you’ll arrive at the critical factor behind an injury.” Do you
really believe there is a single root cause?
Consider the three sides of “The Safety Triad” I introduced in Chapter 2 (see Figure 2.2).
One side is for environment, including tools, equipment, engineering design, climate, and
housekeeping factors. Another side of this triangle stands for behavior, the actions every-
one did or did not perform related to an incident. The third side represents person factors,
or the internal feeling states of the people involved in the incident—their attitudes, percep-
tions, and personality characteristics.
Given the dynamic interdependency of environmental, behavioral, and personal fac-
tors in everyday events, how can anyone expect to find one root cause of an incident?
Instead, take the systems approach and search for a variety of contributory factors within
the environment, behavior, and person domains. Then, decide which of these factors can
be changed to reduce the chance of another unfortunate incident. Environmental factors
are usually easiest to define and improve, followed by behavioral factors. Most difficult to
define and change directly are the person factors, but many of these can be affected posi-
tively by properly influencing behaviors.
Improve communication. Interpersonal conversation is key to finding and correct-
ing the potential contributors to an incident. People need to talk openly about the various
environment, behavior, and person factors related to a near hit, injury, or damage to prop-
erty. But this will not happen if the focus of an “investigation” is to find a single reason for
the “failure.” People want nothing to do with a failure.
“It is human nature to deny personal influence in a loss. Kids blame the other kid—“he
made me do it.” Adults just keep their mouths shut. To get people to open up, we need to
approach incident analysis as an opportunity for success. Such a mindset is really right.
Incident analysis is an opportunity to prevent future mishaps, perhaps a much greater loss
than the one precipitating the current analysis.
Let us get away from the perspective of incident equals failure. The focus should be on
how an incident gives us the chance to learn and improve. This can lead to more reports of
personal near hits and property damage to correct problems before a major injury to a
friend or coworker occurs.
Increase involvement. You can expect more participation in incident reporting and
analysis if you involve workers in the actual correction phase of the process. People will
contribute more if they have a say in the outcome. Of course, management needs to
approve and support the corrections recommended by the workforce. Workers know more
than anyone else about what it will take to make environmental, behavioral, and personal
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 44

44 Psychology of safety handbook

factors safer. Use their critical expertise and you will motivate more ownership and
involvement in the entire process.
Apply systems solutions. Traditionally, the corrective action following an incident is
not only designed narrowly, it is also applied narrowly. A safety leader presents a report to
management, and then the recommended solution to eliminating the “root cause” is imple-
mented in the work area where the incident occurred. An equipment guard might be
replaced, more comfortable personal protection equipment ordered, or a certain employee
might be “retrained” or even punished (incorrectly referred to as “discipline” in the safety
literature).
You will get broader interest and involvement in an incident analysis process if correc-
tive action plans are applied to all relevant work areas. This promotes a systems perspec-
tive rather than the piecemeal “band-aid” approach common to so many work cultures.
Look at the bigger picture. Use the results of an incident analysis to improve relevant
environment behavior and person factors plantwide. This sends the kind of actively
caring message that not only promotes participation but also makes that participation more
constructive.
Promote accountability. Both the quantity and quality of participation in an incident
analysis process depend on the numbers you use to evaluate success or failure. The success
of any safety effort is ultimately determined by the bottom line outcome—the total record-
able injury rate (TRIR), but this index provides no instructive guidance nor motivation to
continue a particular safety process.
Instead, keep track of the various components of an incident analysis. Monitor the
number of near hit, property damage, and injury reports. Track the number of corrective
actions implemented for environment, behavior, and person-based factors. Now, you have
an accountability system that facilitates participation. Of course, the focus needs to be on

Accident
Investigation
• A Safety Professional
Investigates
• Reactive: Investigate
Serious Injuries
• Fault Finding
• One Root Cause
• Piecemeal Approach
• Avoid Failure
• Conversation Stifled
• Management Corrects the
Environment
• Management Punishes the
Behavior
• Solution Applied Narrowly
• Evaluation Focuses on
Injury Rate
Incident
Analysis
• A Safety Team Analyzes
• Proactive: Analyze Near
Hits and First Aid Cases
• Fact Finding
• Many Contributing Factors
• Systems Approach
• Achieve Success
• Conversation Encouraged
• Workers Recommend
Environment Change
• Workers Encourage
Behavior Change
• Solution Applied Broadly
• Evaluation Focuses on
Participation

Figure 3.7 Accident investigation is not the same as incident analysis.

820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 45

Chapter three: Paradigm shifts for total safety 45

successfully completing the various steps of the process rather than avoiding a penalty for
not completing the process. Keep score of your process achievements rather than only wait-
ing to see a reduction in injuries. This is a more valid and instructive measure of your suc-
cess than any outcome measure currently available.
Figure 3.7 reviews the points given here to encourage more and better involvement
in defining and correcting factors contributing to a near hit, property damage, or per-
sonal injury. The differences between traditional “accident investigation” and the pro-
posed “incident analysis” also summarize the paradigm shifts needed to empower
more involvement in safety and achieve a Total Safety Culture. The key is increased
participation, and the principles given here for encouraging more people to contri-
bute to incident analysis are relevant for motivating more activity in any worthwhile
endeavor.

From reactive to proactive

Analyzing events preceding an incident, be it a near hit or an injury, demonstrates the need
to think and act proactively. Unfortunately, a proactive stance is extremely difficult to
maintain, especially in a corporate culture that is increasingly complex and demanding.
There is a higher and higher price tag on “free time.” With barely enough time to react suf-
ficiently to crises each day, how can we find time to be proactive?
Proactivity is especially challenging within the context of downsizing, disguised as
“reengineering” in many work cultures. The worker in Figure 3.8 is barely able to react
effectively to daily crises. How can he be expected to think ahead and be proactive? There
are no quick-fix answers, but injury prevention requires us to find solutions. This text pro-
vides theory, procedures, and tools to guide long-term continuous improvement. Thus, we
need to accept the next paradigm shift.

Figure 3.8 Technology cannot always substitute for personnel.

820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 46

46 Psychology of safety handbook

From quick fix to continuous improvement

“Proactive” can be substituted for “reactive” only with a systems perspective and an opti-
mistic attitude of continuous improvement through increased employee involvement.
Understanding the psychology of safety can be a great aid here. The principles and proce-
dures described in this book will enable you to influence incremental changes in work
practices and attitudes that can prevent an injury. This represents a proactive, continuous
improvement paradigm, which will surely improve your safety performance.

From priority to value

“Safety is a priority.” This is probably the most common safety slogan found in workplaces
and voiced by safety leaders. I have seen signs, pens, buttons, hats, T-shirts, and note pads
with this message. No wonder safety and health professionals are surprised when I say that
safety should not be a priority. To justify my case, I offer the following explanation.
Think about a typical workday morning. We all follow a prioritized agenda, often a
standard routine, before traveling to work. Some people eat a hearty breakfast, read the
morning newspaper, take a shower, and wash dishes. Others wake up early enough to go
for a morning jog before work. Some grab a roll and a cup of coffee, and leave their home
in disarray until they get back in the evening.
In each of these scenarios the agenda—the priorities—are different. Yet, there is one
common activity. It is not a priority but a basic value. Do you know what it is?
One morning you wake up late. Perhaps your alarm clock failed. You have only 15 min-
utes to prepare for work. Your morning routine changes drastically. Priorities must be
rearranged. You might skip breakfast, a shower, or a shave. Yet every morning schedule
still has one item in common. It is not a priority, capable of being dropped from a routine
owing to time constraints or a new agenda. No, this particular morning activity represents
a value which we have been taught as infants, and it is never compromised. Have you
guessed it by now? Yes, this common link in everyone’s morning routine, regardless of time
constraints, is “getting dressed.”
This simple scenario shows how circumstances can alter behavior and priorities.
Actually, labeling a behavior a “priority” implies that its order in a hierarchy of daily
activities can be rearranged. How often does this happen at work? Does safety some-
times take a “back seat” when the emphasis is on other priorities such as production quan-
tity or quality?

Enduring values
It is human nature to shift priorities, or behavioral hierarchies, according to situational
demands or contingencies. But values remain constant. The early morning anecdote illus-
trates that the activity of “getting dressed” is a value that is never dropped from the rou-
tine. Should “working safely” not hold the same status as “getting dressed”? Safe work
practices should occur regardless of the demands of a particular day.
Safety should be a value linked with every activity or priority in a work routine. Safe
work should be the enduring norm, whether the current focus is on quantity, quality, or
cost effectiveness as the “number one priority.”
The ultimate aim of a Total Safety Culture is to make safety an integral aspect of all per-
formance, regardless of the task. Safety should be more than the behaviors of “using per-
sonal protective equipment,” more than “locking out power” and “checking equipment for
potential hazards,” and more than “practicing good housekeeping.” Safety should be an
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 47

Chapter three: Paradigm shifts for total safety 47

Activator

Culture

Intention

Behavior Value

Consequence Attitude

Figure 3.9 Behavior change interventions influence intentions, attitudes, values, and
culture.

unwritten rule, a social norm, that workers follow regardless of the situation. It should
become a value that is never questioned—never compromised.
This, of course, is much easier said than done. How do you even begin to work for such
lofty aims? Figure 3.9 summarizes the relationships between intentions, behaviors, atti-
tudes, and values. It outlines a starting point and general process for developing safety as
a corporate value. A key point is that attitudes and values follow from behavior. This brings
us to behavior management techniques. They are the starting point for acting a person into
safe thinking.
This is how it works. When you follow safe procedures consistently for every job and
attribute your behavior to a voluntary decision, you begin thinking safe. Eventually, work-
ing safe becomes part of your value system.
Figure 3.9 illustrates how attitudes and values influence intentions and behaviors
directly. But, as discussed in Chapter 2, it is not cost effective to manage attitudes and val-
ues directly to “think people into safe acting.” Notice in the figure the different thickness
of rectangles enclosing the terms. The thicker the border, the more measurable and man-
ageable the concept. Activators (antecedent conditions which direct behavior), behaviors,
and consequences (events which follow and motivate behaviors and influence attitudes)
are easiest to define, measure, and manage.
In contrast, values and culture are the most difficult to measure reliably and influence
directly. This book gives you specific techniques for managing behaviors to promote sup-
portive safety attitudes and values. Put them all together and eventually you will construct
an integrated Total Safety Culture.
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 48

48 Psychology of safety handbook

In conclusion
This chapter describes ten shifts in perspective needed to go beyond current levels of safety
excellence. The first nine could be considered goals for achieving a Total Safety Culture.
The tenth—making safety a value—is not something that can be measured and tracked. It
is the ideal vision for our safety mission.
Here is how the new paradigms fit together. Your safety achievement process should
be considered a company responsibility, not a regulatory obligation. It should be achieve-
ment oriented with a focus on behaviors, supported by all managers and supervisors but
driven by the line workers or operators through interdependent teamwork. A systems
approach is needed, which leads to a fact-finding perspective, a proactive stance, and a
commitment to continuous improvement.
These new perspectives reflect new principles to follow, new procedures to develop
and implement. This “new safety work” will lead to different perceptions, attitudes, and
even values. Ultimately, the tenth paradigm shift can be reached. When safety goes from
priority to value, it will not be compromised at work, at home, or on the road. Naturally,
numerous injuries will be prevented and lives saved everyday. This vision should moti-
vate each of us to be active in the safety achievement process. This book helps you define
your role.

References
Barry, P. Z., Individual versus community orientation in the prevention of injuries, Prev. Med., 4, 47,
1975.
Brehm, J. W., A Theory of Psychological Reactance, Academic Press, New York, 1966.
Brehm, J. W., Responses to Loss of Freedom: A Theory of Psychological Reactance, General Learning
Press, New York, 1972.
Carlson, M., Charlin, V., and Miller, N., Positive mood and helping behavior: a test of six hypothe-
ses, J. Person. Soc. Psychol., 55, 211, 1988.
Deming, W. E., Out of the Crisis, Massachusetts Institute of Technology, Center for Advanced
Engineering Study, Cambridge, MA, 1986.
Deming, W. E., Quality, productivity, and competitive position, workshop presented by Quality
Enhancement Seminars, Inc., Cincinnati, OH, May 1991.
Deming, W. E., The New Economics for Industry, Government, Education, Massachusetts Institute of
Technology, Center for Advanced Engineering Study, Cambridge, MA, 1993.
Evans, L., Wasielewski, P., and von Buseck, C. R., Compulsory seat belt usage and driver risk-tak-
ing behavior, Hum. Fact., 24, 41, 1982.
Geller, E. S., Performance management and occupational safety: start with a safety belt program, J.
Organ. Beh. Manage., 11(1), 149, 1990.
Geller, E. S., The human element in integrated environmental management, in Implementing
Integrated Environmental Management, Cairns, J., Crawford, T. V., and Salwaster, H., Eds.,
University Press, Blacksburg, VA, 1994.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998a.
Geller, E. S., Understanding Behavior-Based Safety: Step-by-Step Methods to Improve Your Workplace, 2nd
ed., J. J. Keller & Associates, Inc., Neenah, WI, 1998b.
Geller, E. S., Lehman, G. R., and Kalsher, M. R., Behavior Analysis Training for Occupational Safety,
Make-A-Difference, Inc., Newport, VA, 1989.
Guarnieri, M., Landmarks in history of safety, J. Saf. Res., 23, 151, 1992.
Haddon, W., Jr., A note concerning accident theory and research with special reference to motor
vehicle accidents, Ann. NY Acad. Sci., 107, 635, 1963.
820049_CRC1_L1540_CH03 11/10/00 1:31 PM Page 49

Chapter three: Paradigm shifts for total safety 49

Haddon, W., Jr., The changing approach to the epidemiology, prevention, and amelioration of
trauma: the transition to approaches etiologically rather than descriptively based, Am. J. Publ.
Health, 58, 1431, 1968.
Hayakawa, S. I., Language in Thought and Action, 4th ed., Harcourt Brace Jovanovich, New York,
1978.
Heinrich, W. W., Industrial Accident Prevention, McGraw-Hill, New York, NY, 1931.
Koepnick, W., Do safety incentive programs really work?, presented at the National Safety Council
Congress and Exposition, Chicago, IL, October 1993.
Komaki, J., Heinzmann, A. T., and Lawson, L., Effective training and feedback: component analysis
of a behavioral safety program, J. Appl. Psychol., 65, 261, 1980.
Krause, T. R., Hidley, J. H., and Hodson, S. J., The Behavior-Based Safety Process: Managing
Involvement for an Injury-Free Culture, 2nd ed., Van Nostrand Reinhold, New York, 1996.
McSween, T. E., The Values-Based Safety Process: Improving Your Safety Culture with a Behavioral
Approach, Van Nostrand Reinhold, New York, 1995.
New Merriam-Webster Dictionary, Merriam-Webster, Inc, Springfield, MA, 1989.
NHTSA Traffic Safety Facts 2000, DOT HS 808 954, National Center for Statistics and Analysis,
Washington, DC, 2000.
Petersen, D., Safety’s paradigm shift, Prof. Saf., 36(8), 47, 1991.
Skinner, B. F., Beyond Freedom and Dignity, Alfred A Knopf, New York, 1971.
Sulzer-Azaroff, B., Behavioral approaches to occupational health and safety, in Handbook of
Organizational Behavior Management, Frederickson, L. W., Ed., Wiley & Sons, (1982).
Sulzer-Azaroff, B., The modification of occupational safety behavior, J. Occup. Accid., 9, 177, 1987.
Wagenaar, A. C., Restraint usage among crash-involved motor vehicle occupants, Report UMTRI-
84-2, University of Michigan Transportation Research Institute, Ann Arbor, MI, 1984.
Waller, J. A., Injury: conceptual shifts and prevention implication, Ann. Rev. Publ. Health, 8, 21, 1987.
Winn, G. L., In the crucible: testing for a real paradigm shift, Prof. Saf., 37(12), 30, 1992.
Winn, G. L. and Probert, L. L., Philosopher’s stone: it may take another Monongah, Prof. Saf., 40(5),
18, 1995
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 51

section two

Human barriers to safety

820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 52
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 53

chapter four

The complexity of people

Safety is usually a continuous fight with human nature. This chapter explains why. Understanding
this basic point will lead to less victim blaming and fault finding when investigating an injury.
Instead, we will be able to find factors in the system that can be changed in advance to prevent
injuries at work, at home, and throughout the community.

“What lies behind us and what lies before us are small matters compared to what lies
within us.”—Ralph Waldo Emerson

“All injuries are preventable.”

“It is human nature to work safely.”
“Safety is just common sense.”
“Safety is a condition of employment.”

Read these familiar statements and you get the idea that working safely is easy or nat-
ural. Nothing could be further from the truth.
In fact, it is often more convenient, more comfortable, more expedient, and more com-
mon to take risks than to work safely. And past experience usually supports our decisions
to choose the at-risk behavior, whether we are working, traveling, or playing. So, we are
often engaged in a continuous fight with human nature to motivate ourselves and others
to avoid those risky behaviors and maintain safe ones.
Let us consider what holds us back from choosing the safe way, whether it is following
safe operating procedures, driving our automobile, or using personal protective equipment.

Fighting human nature

When I ask safety professionals, corporate executives, or hourly workers what causes
work-related injuries, I get long and varied lists of factors. Actually, each list is quite simi-
lar. After all, everyone experiences events, attitudes, demands, distractions, responsibili-
ties, and circumstances that get in the way of performing a task safely.

Most of us have been in situations where we were not sure how to perform safely.
Perhaps we lacked training.
Maybe the surrounding environment was not as safe as it could be.
Demands from a supervisor, coworker, or friend put pressure on us to take a short cut
or risk.
Maybe, it was inconvenient or uncomfortable to follow all of the safety procedures.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 54

54 Psychology of safety handbook

It is possible our physical condition—fatigue, boredom, drug impairment—influenced

at-risk performance.
There are other factors. Have you ever been unsafely distracted by external stimuli, like
another person’s presence or by an internal state, like personal thoughts or emotions?
Can you remember a time when you just did not feel like taking the extra time to be
safe?
I am sure you have experienced the “macho” attitude, from yourself or others that, “It
will never happen to me.”
Fortunately, it is rare that an injury follows unsafe behavior. The attitude, “It won’t
happen to me,” is usually supported or rewarded by our actual experiences. Risk taking is
rarely punished with an injury or even a near hit, instead it is consistently rewarded with
convenience, comfort, or time saved.
This creates something of a vicious cycle. The rewards of risky behavior mean you are
likely to take more chances. As you gain experience at work you often master dangerous
shortcuts. Because these at-risk behaviors are not followed by a near hit or injury, they
remain unpunished, and they persist.
This basic principle of human nature reinforced throughout our lives runs counter to
the safety efforts of individuals, groups, organizations, and communities. It explains why
promoting safety and health is the most difficult ongoing challenge at work. The reality is
that injuries really do happen to the “other guy.”

Learning to be at-risk
Remember when you first learned to drive a car? I bet this was an important but stressful
occasion. Even with the right amount of driver training from your parents or a professional
instructor, you felt a bit nervous getting behind the wheel for the first time.
At first, you were probably very careful to follow all the safe procedures you learned.
Both hands on the wheel—the nine o’clock and three o’clock positions. Both eyes on the
road at all times. You always used your turn signal; always stopped when traffic lights
turned yellow. If a safety belt was in the car, you used it. Conversations with passengers
were avoided, as well as distractions from a radio or cassette tape. This was all right and
proper, of course, because driving is a relatively complex and risky task, requiring the dri-
ver’s undivided attention. This time human nature was on the safe side.
But how quickly you took driving for granted! Your complete concentration was no
longer needed—tasks became automatic and “second nature.” Many precautionary behav-
iors fell by the wayside. You began driving with one hand on the wheel. Your other hand
held a drink, a cigarette, or a passenger’s hand. Distractions were soon permitted—loud
music, emotional conversations (sometimes on a telephone), and even “love making.” I
have even seen some people read a book, a letter, or a map while driving. All this while
blowing past the speed limit, running a “yellow” traffic light, or following too close behind
another vehicle.
We continue these risky driving behaviors every day because they are “cool”—they are
fun, convenient, and save us time. We never think of crashing, and thank goodness it usu-
ally does not happen to us.
In a short time behind the wheel, we have gone from controlled processing to auto-
matic processing (Schneider and Shiffrin, 1977; Shiffrin and Dumais, 1981). Various risky
practices are adopted for fun, comfort, or convenience. These consequences reward the
risky behavior and sustain it. This is human nature on the side of at-risk behavior and can
be explained by basic principles of behavioral science. Thus, risky driving behaviors like
those shown in Figure 4.1 occur quite often.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 55

Chapter four: The complexity of people 55

Figure 4.1 Natural learning experiences result in at-risk driving.

Dimensions of human nature

The factors contributing to a work injury can be categorized into three areas.

1. Environment factors.
2. Person factors.
3. Behavior factors.

This is the “Safety Triad” (Geller et al., 1989) introduced in Chapter 2. The most common
reaction to an injury is to correct something about the environment—modify or fix equip-
ment, tools, housekeeping, or an environmental hazard.
Often the incident report includes some mention of personal factors, like the
employee’s knowledge, skills, ability, intelligence, motives, or personality. These factors
are typically translated into general recommendations.

“The employee will be disciplined.”

“The employee will be retrained.”

This kind of vague attention to critical human aspects of a work injury shows how frus-
trating and difficult it is to deal with the psychology of safety—the personal and behavioral
sides of the Safety Triad. The human factors contributing to injury are indeed complex,
often unpredictable and uncontrollable. This justifies my conclusion that all injuries cannot
be prevented.
The acronym BASIC ID reflects the complexity and uncontrollability of human nature.
As depicted in Figure 4.2, each letter represents one of seven human dimensions of an indi-
vidual. Many clinical psychologists use a similar acronym as a reminder that helping peo-
ple improve their psychological state requires attention to each of these areas (Lazarus,
1971, 1976). Here is a simple scenario that underscores the need in safety to understand
personal factors.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 56

56 Psychology of safety handbook

Figure 4.2 The acronym BASIC ID reflects the complexity of people and potential contri-
butions to injury.

Dave, an experienced and skilled craftsman, works rapidly to make

an equipment adjustment while the machinery continues to operate.
As he works, production-line employees watch and wait to resume
their work. Dave realizes all too well that the sooner he finishes his
task, the sooner his coworkers can resume quality production. So, he
does not shut down and lock out the equipment power. After all, he
has adjusted this equipment numerous times before without locking
out and he has never gotten injured.
A morning argument with his teenage daughter pervades
Dave’s thoughts as he works, and suddenly he experiences a near hit.
His late timing nearly results in his hand being crushed in a pinch
point.
Removing his hand just in time, Dave feels weak in his knees and
begins to perspire. This stress reaction is accompanied by a vivid
image of a crushed right hand. After gathering his composure, Dave
walks to the switch panel, shuts down and locks out the power, then
lights up a cigarette. He thinks about this scary event for the rest of the
day and talks about the near hit to fellow workers during his breaks.

This brief episode illustrates each of the psychological dimensions represented by

BASIC ID (see Figure 4.2) and demonstrates the complexity of human activity.
Behavior is illustrated by observable actions such as adjusting equipment, pulling a
hand away from the moving machinery, lighting up a cigarette, and talking to coworkers.
Dave’s attitude about work was fairly neutral at the start of the day, but immediately
following his near hit he felt a rush of emotion. His attitude toward “energy control and
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 57

Chapter four: The complexity of people 57

power lockout” changed dramatically, and his commitment to locking out increased after
relating his close call to friends.
Sensation is evidenced by Dave’s dependence on visual acuity, hand–eye coordi-
nation, and a keen sense of timing when adjusting the machinery. His ability to react
quickly to the dangerous situation prevented severe pain and potential loss of valuable
touch sensation.
Imagery occurred after the near hit when Dave visualized a crushed hand in his
“mind’s eye,” and this contributed to the significance and distress of the incident. Dave will
probably experience this mental image periodically for some time to come. This will moti-
vate him to perform appropriate lockout procedures, at least for the immediate future.
Cognition or “mental speech” about the morning argument with his daughter may
have contributed to the timing error that resulted in the near hit. Dave will probably
remind himself of this episode in the future, and these cognitions may help trigger proper
lockout behavior.
Interpersonal refers to the other people in Dave’s life who contributed to his near hit
and will be influential in determining whether he initiates and maintains appropriate lock-
out practices.
For example, it was the interpersonal discussion with his daughter that occupied his
thoughts or cognitions before the near hit. The presence of production-line workers influ-
enced Dave through subtle peer pressure to quickly adjust equipment without lockout
practices. These onlookers may have distracted Dave from the task, or they could have
motivated him to show off his adjustment skills. After Dave’s near hit, his interpersonal
discussions were therapeutic, helping him relieve his distress and increase his personal
commitment to occupational safety.
Drugs in the form of caffeine from morning coffee may have contributed to Dave’s tim-
ing error. The extra cigarettes Dave smoked as a “natural” reaction to distress also had
physiological consequences, which could have been reflected in Dave’s subsequent behav-
ior, attitude, sensation, or cognition.
Dave’s lesson shows how human nature interacts with environmental factors to cause
at-risk work practices, near hits, and sometimes personal injuries. It is relatively easy to
control the environmental factors. As I will explain in Section 3 on behavior-based safety, it
is feasible to measure and control the behavioral factors. However, the complex personal
factors, described by the BASIC ID acronym, are quite elusive. The field of psychology
provides insights here, and this information can benefit occupational safety and health
programs.
Let us further discuss aspects of human nature that can make safety achievement so
challenging.

Cognitive failures
“All injuries are preventable.” I have heard this said so many times that it seems to be a slo-
gan or personal affirmation that safety pros use to keep themselves motivated. I suspect
some readers will resist any challenge to this ideal. I certainly appreciate their optimism,
and there is no harm if such perfectionism is kept to oneself. But sharing this belief with
others can actually inhibit achieving a Total Safety Culture.
You see, if a common workplace slogan declares all injuries preventable, workers may
be reluctant to admit they were injured or had a near hit. After all, if all injuries are pre-
ventable and I have an injury, I must be a real “jerk” for getting hurt.
Combine this slogan with a goal of zero injuries and a reward for not having an injury
and human nature will dictate covering up an injury if possible. Also, as I will discuss in
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 58

58 Psychology of safety handbook

the next chapter, claiming that all injuries are preventable can reduce the perceived risk of
the situation. This can create the notion that “it will not happen to me,” a perception that
can increase the probability of at-risk behavior and an eventual injury.
Eliminate the “all injuries are preventable” slogan from your safety discussions. The
most important reason to drop it is that most people do not believe it anyway. They have
been in situations where all the factors contributing to the near hit or injury could not have
been anticipated, controlled, or prevented. The most uncontrollable factors are the personal
or internal subjective dimensions of people. Consider, for example, the role of cognitive
failures.
Just what is a “cognitive failure”? Some people call it a “brain cramp.” Research by
Broadbent et al. (1982) demonstrated that people who report greater frequency of “cogni-
tive failures” are more likely to experience an injury. The items listed in Figure 4.3 were
used by Broadbent and his associates to measure cognitive failures. Respondents were
merely asked to indicate on a 5-point scale the extent to which they agreed with each state-
ment (from “strongly disagree” to “strongly agree”). Broadbent’s measurement instrument
offers an operational definition for this person dimension that apparently influences injury
frequency.
Scientific protocol will not allow us to conclude from the research by Broadbent et al.
that cognitive failures cause injury but, based on personal experience, it sure seems rea-
sonable to interpret a cause-and-effect relationship. It is likely every reader has experienced
one or more of the “brain cramps” listed in Figure 4.3. Surely you have walked into a room
to get something and forgotten why you were there. And how often have you left home for
work more than once in a single morning because you forgot something? The same sort of
breakdown in cognitive functioning can cause an injury. Does Figure 4.4 reflect potential
reality?
In his classic book, The Psychology of Everyday Things, Norman (1988) classifies various
types of cognitive failure according to a particular stage of routine thinking and decision

• I sometimes forget why I went from one part of the house to another.
• I often fail to notice signposts on the road.
• I sometimes bump into things or people.
• I often forget whether I’ve turned off a light or the coffeepot, or
locked the door.
• I sometimes forget which way to turn on a road I know well but
rarely use.
• I sometimes fail to see what I want in a supermarket (although it’s
right there).
• I often forget where I put something like a newspaper or a book.
• I often daydream when I ought to be listening to something.
• At home, I often start doing one thing and get distracted into doing
something else (unintentionally).
• I sometimes forget what I came to the store to buy.
• I often drop things.
• I often find myself putting the wrong things in the wrong place when
I’m done with them – like putting milk in the cereal cupboard.
• I frequently confuse right and left when giving directions.

Figure 4.3 Broadbent et al. (1982) used these survey items to measure an individual’s
propensity for cognitive failure.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 59

Chapter four: The complexity of people 59

Figure 4.4 A cognitive failure or “brain cramp” can cause a workplace injury.

making. More specifically, consider that we continually take in, process, and react to infor-
mation in our surroundings. Almost everything we do results from this basic information
processing cycle.
We sense a stimulus (input), we evaluate the stimulus and plan a course of action
(interpretation and decision making), and then we execute a response (output).
Unintentional cognitive errors usually occur at the input and output stage of information
processing. Judgment errors and calculated risks occur at the middle cognitive stage—
interpretation and decision making.

Capture errors
Have you ever started traveling in one direction (like to the store) but suddenly find your-
self on a more familiar route (like on the way to work)? How many times have you bor-
rowed someone’s pen to write a note or sign a form, and later found the pen in your
pocket? Norman calls these “capture errors,” because a familiar activity or routine seem-
ingly “captures” you and takes over an unfamiliar activity. This error seems to occur at the
execution stage of information processing, but it also involves misperception or inattention
to relevant stimuli, as well as the absence of conscious judgment or decision making.
How does this error slip into the work routine? Have you ever started a new task and
found yourself using old habits? Has a change in PPE requirements influenced this kind of
human error? It seems reasonable that a routine way of doing something (even at home)
could “capture” your execution of a new work process and lead to this type of cognitive
failure and an injury.
This is one reason to get in the habit of practicing the safe way of doing something,
regardless of the situation. Then your safe behavior is put into automatic mode, and a cap-
ture error can actually be to your advantage. This happens when you reach for the shoul-
der belt in the back seat of a vehicle because of your habitual buckle-up behavior as a
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 60

60 Psychology of safety handbook

driver. When you use your vehicle turn signal at every turn, this safe behavior becomes
habitual. The “capture” is beneficial to your safety. When basic safety-related behaviors
become habits, we have more mental capability for higher level thinking, and the proba-
bility of a cognitive failure is reduced.

Description errors
These “brain cramps” occur when the descriptions or locators of the correct (safe) and
incorrect (at-risk) executions are similar. For example, I periodically throw a tissue in our
clothes hamper instead of the waste can, even though our clothes hamper is not next to a
trash receptacle. On a few occasions, I have actually thrown dirty clothes in the trash can,
and once I threw a sweaty T-shirt in the toilet. According to Norman, the similar charac-
teristic of these three items—a large oval opening—led to these errors.
Do you have any switches in your work setting which are similar and nearby but con-
trol different functions? How unsafe would it be to throw the wrong switch? Many control
panels are designed with this error in mind. Switches or knobs controlling incompatible
functions are not located in close proximity with one another and often look and feel dis-
tinctly different for quick visual and tactile discrimination. Thus, it might be useful to eval-
uate your work setting with regard to the need for different behaviors with similar
descriptions.

Loss-of-activation errors
Have you ever walked into a room to do something or to get a certain object, but when you
got there you forgot what you were there for? You think hard but just cannot remember.
Then, you go back to the first room and suddenly you remember what you wanted to do
or get in the other room. What happened here?
This cognitive failure is commonly referred to as “forgetting.” Norman refers to it as
“loss-of-activation,” because the cue or activator that got the behavior started was lost or
forgotten. This happens whenever you start an activity with a clear and specific goal, but
after you get engaged in the task you lose sight of the goal. You might, in fact, continue the
task but with little awareness of the rationale for progress toward a goal. With regard to the
three stages of information processing discussed previously, this error starts in Stage 1—
input—but eventually affects the output stage when you cannot complete the task without
more information. Stage 2 is involved because lapses in memory occur during interpreta-
tion and decision making.
When people tell you they already know what to do with statements like “Stop harp-
ing on the same old thing,” you can say you are just actively caring by trying to prevent a
“loss-of-activation” error. You will never know how many of these cognitive failures you
will prevent, but you can motivate yourself to keep activating by reflecting on your own
experiences with this sort of “brain cramp.” Then, consider the large number of people in
your work setting who have made similar unintentional errors every day.

Mode errors
Mode errors are probable whenever we face a task involving multiple options or modes of
operation. These errors are inevitable when equipment is designed to have more functions
than the number of control switches available. In other words, when controls are designed
for more than one mode of operation, you can expect occurrences of this error.
Over the years, I have owned a variety of digital watches with a stopwatch mode. Each
one has had a different arrangement of switches designed to provide more functions than
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 61

Chapter four: The complexity of people 61

control buttons. Therefore, the meaning of a button press depends upon the position of a
mode switch. So, guess how many times I have pressed the wrong button and illuminated
the dial or reset the digital readout when I only wanted to stop the timer? Have you expe-
rienced the same kind of mode error, if not with a stop watch, perhaps with the text editor
of a personal computer? Airline pilots must be especially wary of this kind of error.
This type of cognitive failure is essentially one of execution, but these errors often occur
because we forget the mode we are in. This involves memory and the interpretation
and decision-making phase of information processing. Equipment design is certainly
important here, along with proper training and the behavior-based tools detailed later in
Section Three.

Mistakes and calculated risks

The four types of cognitive failures discussed so far—capture, description, loss-of-activa-
tion, and mode errors—are unintentional. Their sources are mostly at the input and out-
put stages of information processing. The middle interpretation and decision-making
stage is essentially uninvolved. Thus, the at-risk behavior resulting from these errors is
unintentional. The person meant well. The plan was good, but the execution was uninten-
tionally flawed.
Mistakes and calculated risks occur at the interpretation and decision-making stage of
information processing. Here is where we interpret our sensory input and decide on a
course of action. With mistakes, the individual was well-intentioned regarding the ultimate
outcome of getting the job done, but used poor judgment in getting there.
While driving, have you ever turned right on to a main road into the path of an oncom-
ing vehicle you had not seen, or whose speed you had misjudged? Have you ever miscal-
culated a parking space and scraped an adjoining vehicle? How many times have you
planned a bad travel route and got caught in traffic congestion you could have avoided?
Have you ever pressed the brake too quickly on a slippery road or pumped the brakes in
an antilock system? Parking and braking are frequent and intentional driving behaviors,
but under certain circumstances they are mistakes.
Now suppose you do not buckle your safety belt. Perhaps you divide your attention
between the road and some other task like map reading, phone dialing, or cassette select-
ing. You know this behavior is unsafe, but you decide to take a calculated risk. Your judg-
ment is faulty, as in a mistake, but unlike a mistake, your at-risk behavior is deliberate. Such
behavior does seem rational because it is not followed by a negative consequence and it is
supported with perceived comfort, convenience, or efficiency.

In summary
Human error is caused by a cognitive failure at one or more of the three basic stages of
information processing. Understanding the difference between the various types of cogni-
tive failure can help us predict when one type of at-risk behavior is more likely. For exam-
ple, the probability of an input or output error increases with more experience and
perceived proficiency on the job. That is, as people become more competent and confident,
they pay less deliberate and conscious attention to what they are doing. They automatically
filter out certain stimulus inputs, they do less interpretation and decision making, and they
resort to automatic modes of execution.
Mistakes and calculated risks are possible among both beginning and experienced
workers. New hires make safety-related mistakes when they do not know the safe way to
perform a task or when they do not understand the need for special safety precautions.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 62

62 Psychology of safety handbook

They take calculated risks when the actions and conversations from others favor the at-risk
alternatives.
Experienced workers make mistakes when they take safety for granted and fail to
consider the injury potential of a certain at-risk behavior. They take calculated risks when they
feel especially skilled at a task, realize they have never been seriously injured at work, and con-
sider the soon and certain benefits of an at-risk behavior to outweigh the improbable costs.
Understanding the variety of potential cognitive failures in the workplace leads to a
realization that most of these are unnoticed or ignored. In other words, when our at-risk
behaviors do not lead to personal injury, we just forget them or explain them away. This is
basic human nature. Who likes to talk about their errors?
We feel much better talking about our good times than our bad times, but I am sure you
recognize injury prevention requires a shift in perspective. Only through open and fre-
quent conversation about our cognitive failures can we alter the environmental conditions
that can reduce them. This Handbook shows you how to make this happen. You will learn
how to develop and sustain the kinds of interpersonal interaction and intervention needed
for a Total Safety Culture. First, let us see how some interpersonal aspects of human nature
can be a barrier to safety. This is the second “I” of BASIC ID.

Interpersonal factors
Our interpersonal relationships dramatically influence our thoughts, attitudes, and
actions. How much of your time each day is dedicated to gaining the approval of others?
Of course, we sometimes attempt to avoid the disapproval of others, be they a parent,
spouse, work supervisor, or department head. As discussed in Chapter 3, we do not feel
as good—or as “free”—when working to avoid failure or disapproval as when working

Figure 4.5 Authority can be taken too far.

820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 63

Chapter four: The complexity of people 63

to achieve success or approval. In both cases, other people are the cause of our motivation
and behavior.
Is the scenario depicted in Figure 4.5 completely ridiculous and unrealistic? Well, it is
a bit extreme, but consider for a moment the adversity many people go through to impress
others. And have you ever followed orders you know put yourself or others at some degree
of risk? If something went wrong, it would not be your fault. It was not your responsibil-
ity; you were just following orders. Just like when we were kids and we got into trouble,
we were quick to say, “It’s not my fault, he told me to do it.”
It is not hard to see what all of this has to do with safety in the workplace. People take
risks on the job because others do the same, and sometimes workers blindly follow a super-
visor’s orders that could endanger them, other coworkers, or the environment. This reflects
the interpersonal power of two principles of social influence—conformity and authority.
Let us examine these interpersonal phenomena more closely to understand exactly how
they can be human barriers to safety. Then we can consider ways to turn these social influ-
ence factors around and use them to benefit safety. Indeed, the right kind of interpersonal
influence is critical for achieving a Total Safety Culture.

Peer influence
Research conducted by Asch and associates in the mid-1950s found more than one out of
three intelligent and well-intentioned college students were willing to publicly deny real-
ity in order to follow the obviously inaccurate judgments of their peers. Asch’s classic stud-
ies of conformity (1955, 1956, 1958) involved six to nine individuals sitting around a table
judging which of three comparison lines was the same length as the standard. Figure 4.6
depicts one of these judgment situations.
All but the last individual to voice an opinion were research associates posing as sub-
jects. Sometimes the research associates uniformly gave obviously incorrect judgments.
The last person to decide was the real subject of the experiment. About one-third of the time
the subject denied the obvious truth in order to go along with the group consensus.

Figure 4.6 Asch used stimulus comparisons like these to study conformity.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 64

64 Psychology of safety handbook

Figure 4.7 Peer pressure drives social conformity.

This and similar procedures were used in numerous social psychology experiments to
study factors influencing the extent of conformity. For example, a subject’s willingness to
deny reality in order to conform to the group was bolstered by increasing group size (Asch,
1955) and the apparent competence or status of group members (Crutchfield, 1955; Endler
and Hartley, 1973). On the other hand, the presence of one dissenter or nonconformist in
the group was enough to significantly decrease conformity—it increased a subject’s will-
ingness to choose the correct line even when only 1 of 15 prior decisions reflected the cor-
rect choice (Nemeth, 1986).
The phenomenon of social conformity depicted humorously in Figure 4.7 is certainly
not new to any reader. We see examples of it every day, from the clothes people wear, to
how they communicate both verbally and in writing. We cannot overlook the power of con-
formity in influencing at-risk behavior. We have learned that peer pressure increases when
more people are involved and when the group members are seen as relatively competent
or experienced.
It is important to remember, though, that one dissenter—a leader willing to ignore peer
pressure and do the right thing—is often enough to prevent another person from suc-
cumbing to potentially dangerous conformity at work.

Power of authority
Imagine you are among nearly 1000 participants in one of Milgram’s 20 obedience studies
at Yale University in the 1960s. You and another individual are led to a laboratory to
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 65

Chapter four: The complexity of people 65

Figure 4.8 Subjects experienced distress when giving electric shocks to peers.

participate in a human learning experiment. First, you draw slips of paper out of a hat to
determine randomly who will be the “teacher” and the “learner.”
You get to be the teacher; the learner is taken to an adjacent room and strapped to a
chair wired through the wall to an electric shock machine containing 30 switches with
labels ranging from 15 volts—light shock to 450 volts—severe shock. You sit behind this
shock generator and are instructed to punish the learner for errors in the learning task by
delivering brief electric shocks, starting with the 15-volt switch and moving up to the next
higher voltage with each of the learner’s errors. The scenario is depicted in Figure 4.8.
Complying with the experimenter’s instructions, you hear the learner moan as you
flick the third, fourth, and fifth switches. When you flick the eighth switch (labeled 120
volts), the learner screams, “These shocks are painful,”and when the tenth switch is acti-
vated, the learner shouts, “Get me out of here!”
At this point, you might think about stopping, but the experimenter prompts you with
words like, “Please continue—the experiment requires that you continue.”
Increasing the shock intensity with each of the learner’s errors, you reach the 330-volt
level. Now you hear shrieks of pain—the learner pounds on the wall, then becomes silent.
Still, the experimenter urges you to flick the 450-volt switch when the learner fails to
respond to the next question.
At what point will you refuse to obey the instructions?
Milgram asked this question of a group of people, including 40 psychiatrists, before
conducting the experiment. They thought the sadistic game would stop soon after the
learner indicated the shock was painful. So Milgram and his associates were surprised that
65 percent of his actual subjects, ranging in age from 20 to 50, went along with the experi-
menter’s request right up to the last 450-volt switch (Milgram, 1963, 1974).
Why did they keep following along? Did they figure out the learner was a confederate
of the experimenter and did not really receive the shocks? Did they realize they were being
deceived in order to test their obedience?
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 66

66 Psychology of safety handbook

No, the subjects sweated, trembled, and bit their lips when giving the shocks, as shown
in Figure 4.7. Some laughed nervously. Others openly questioned the instructions, but most
did as they were told.
Milgram and associates learned more about the power of authority in further studies.
Full obedience exceeded 65 percent, with as many as 93 percent flicking the highest shock
switch, when

• The authority figure—the one giving the orders—was in the room with the subject.
• The authority was supported by a prestigious institution, such as Yale University.
• The shocks were given by a group of “teachers” in disguise and remaining anony-
mous (Zimbardo, 1970).
• There was no evidence of resistance—no other subject was observed disobeying the
experimenter.
• The victim was depersonalized or distanced from the subject in another room.

Milgram drew this lesson from the research: “Ordinary people, simply doing their jobs
and without any particular hostility on their part, can become agents in a terrible destruc-
tive process” (Milgram, 1974).
Let us apply this research to the workplace. As a result of social obedience or social con-
formity, people might perform risky acts or overlook obvious safety hazards, and put
themselves and others in danger. To say, “I was just following orders,” reflects the obedi-
ence phenomenon, and “Everyone else does it!” implies social conformity or peer pressure.
To achieve a Total Safety Culture, we need to realize the power of these two interper-
sonal factors. Interventions capable of overcoming peer pressure and blind obedience are
detailed in Section 4 of this book. What I want to stress at this point is the vital role of lead-
ership. One person can make a difference—decreasing both destructive conformity and
obedience—by deviating from the norm and setting a safe example. And when a critical
mass of individuals boards the “safety bandwagon,” you get constructive conformity and
obedience that supports a Total Safety Culture.

In conclusion
We need to understand a problem as completely as possible and from many perspectives
before we can solve it. In this chapter, we explored dimensions of the safety problem by
considering the complexity of people. I attempted to convince you that human nature does
not usually support safety. The natural relationships between behavior and its motivating
consequences usually result in some form of convenient, time-saving—and risky—behav-
ior. Consequently, to achieve a Total Safety Culture, you should prepare for an ongoing
fight with human nature.
Human barriers to safety are represented by a popular acronym from clinical psychol-
ogy (BASIC ID). The “C” (cognitions) and second “I” (interpersonal) dimensions of this
acronym, in particular, explain the special challenges of achieving a Total Safety Culture.
The phenomenon of cognitive failures shows the shallowness—in fact, the potential dan-
ger—of the popular safety slogan, “All injuries are preventable.” Conformity and obedi-
ence, two powerful phenomena from social psychological research, further help us to
understand the individual, group, and system factors responsible for at-risk behavior and
injury. Both of these social influence phenomena influence the kind of at-risk behavior
depicted in Figure 4.9.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 67

Chapter four: The complexity of people 67

Figure 4.9 Social conformity and obedience can inhibit safety-related behavior.

The human barriers to safety discussed here should lead us to be more defensive and
alert in hazardous environments. They also show how difficult it is to find the factors con-
tributing to a “near hit” or injury. Another psychological challenge to safety is explored in
the next chapter when we discuss the “S” (sensation) of BASIC ID.

References
Asch, S. E., Opinions and social pressure, Sci. Am., 193, 31, 1955.
Asch, S. E., Studies of independence and conformity: a minority of one against a unanimous major-
ity, Psychol. Monogr., 70, 1, 1956.
Asch, S. E., Effects of group pressure upon modification and distortion of judgments, in Readings in
Social Psychology, 3rd ed. Maccoby, E. E., Newcomb, T. M., and Hartley, E. L., Eds., Holt,
Rinehart & Winston, New York, 1958.
Broadbent, D., Cooper, P. F., Fitzgerald, P., and Parker, K., The cognitive failures questionnaire
(CFQ) and its correlates, Br. J. Clin. Psychol., 21, 1, 1982.
Crutchfield, R. S., Conformity and character, Am. Psychol., 10, 191, 1955.
Endler, N. S. and Hartley, S., Relative competence, reinforcement and conformity, Eur. J. Soc.
Psychol., 3, 63, 1973.
Geller, E. S., Lehman, G. R., and Kalsher, M. J., Behavior Analysis Training for Occupational Safety,
Make-A-Difference, Inc., Newport, VA, 1989.
Lazarus, A. A., Behavior Therapy and Beyond, McGraw-Hill, New York, 1971.
Lazarus, A. A., Multimodal Behavior Therapy, Springer, New York, 1976.
Milgram, S., Behavioral studies of obedience, J. Abnorm. Soc. Psychol., 67, 371, 1963.
820049_CRC1_L1540_CH04 11/10/00 1:32 PM Page 68

68 Psychology of safety handbook

Milgram, S., Obedience to Authority, Harper Collins, New York, 1974.

Norman, D. A., The Psychology of Everyday Things, Basic Books, New York, 1988.
Nemeth, C., Differential contribution of majority and minority influence, Psychol. Rev., 93, 23, 1986.
Schneider, W. and Shiffrin, R. M., Controlled and automatic information processing. I. Detection,
search, and attention, Psychol. Rev., 84, 1, 1977.
Shiffrin, R. M. and Dumais, S. T., The development of automatism, in Cognitive Skills and Their
Acquisition, Anderson, J. R., Ed., Erlbaum, Hillsdale, NJ, 1981.
Zimbardo, P. G., The human choice: individuation, reason, and order versus deindividuation,
impulse, and chaos, in Nebraska Symposium on Motivation, Arnold, W. J. and Levine, D., Eds.,
University of Nebraska Press, Lincoln, NB, 1970.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 69

chapter five

Sensation, perception,
and perceived risk
It is critically important to understand that perceptions of risk vary among individuals. We cannot
dramatically improve safety until people increase their perception of risk in various situations and
reduce their overall tolerance for risk.
In this chapter we shall explore the notion of selective sensation or perception, and then relate this
concept to perceived risk and injury control. Several factors will be discussed that impact whether
employees react to workplace hazards with alarm, apathy, or something in between. Taken together,
these factors shape personal perceptions of risk and illustrate why the job of improving safety is so
challenging.

“What we see depends mainly on what we look for.”—John Lubbock

The “S” of the BASIC ID acronym introduced in Chapter 4 refers to sensation—a human
dimension that influences our thinking, attitudes, emotions, and behavior. In grade school,
we learned there are five basic senses we use daily to experience our world (we see, hear,
smell, taste, and touch). Later we learned that our senses do not take in all of the informa-
tion available in our immediate surroundings. Instead, we intentionally and unintention-
ally tune in and tune out certain features of our environment; thus, some potential
experiences are never realized.
This is a complex process. To experience life on a selective basis, we begin by using our
five senses, but from there, we

• Define (or encode) the information received.

• Interpret its meaning or relevance to us.
• Decide whether the information is worth remembering or responding to.
• Plan and execute a response (if called for).

At any time in this chain of information processing and decision making, we can—and
do—impose our own individual bias, which is shaped by our past experiences, personal-
ity, intentions, aspirations, and expectations. You can see how our everyday sensations are
dramatically influenced consciously and unconsciously by a number of person factors
unique to the situation and the individual sensing the situation. Psychologists refer to such
biased sensation as perception.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 70

70 Psychology of safety handbook

There is also a term called “selective perception” that is commonly used to refer to our
biased sensations. Because all perceptions result from our intentional or unintentional
distortion of sensations, adding the adjective “selective” to perception is unnecessary and
actually redundant. We experience our surroundings through the natural selection of our
sensations. This process is simply referred to as perception.

An example of selective sensation or perception

At the 1994 Professional Development Conference of the American Society of Safety
Engineers (ASSE), the following instructions were printed in one-half of the 40-page hand-
outs distributed to the audience of more than 350 individuals at the start of my two-hour
presentation.

You are going to look briefly at a picture and then answer some ques-
tions about it. The picture is a rough sketch of a poster of a couple at
a costume ball. Do not dwell on the picture. Look at it only long
enough to “take it all in” at once. After this, you will answer “yes” or
“no” to a series of questions.

After the participants read the instructions, I presented the illustration depicted in
Figure 5.1 for about five seconds. If you would like to experience the biased visual sensation

Figure 5.1 Selective sensation can be demonstrated with this ambiguous drawing. Please
read instructions on the prior page, and then look at the drawing for five seconds.
Afterwards, answer the five questions on the following page.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 71

Chapter five: Sensation, perception, and perceived risk 71

(or perception) demonstrated to the ASSE audience, please read the instructions previously
given and then look at Figure 5.1 for approximately five seconds. Then, answer the ques-
tions that I asked my ASSE audience.

1. Did you see a man in the picture?

2. Did you see a woman in the picture?
3. Did you see an animal in the picture? If so, what kind of animal did you see?
What other details did you detect in the brief exposure to the drawing?

___ A woman’s purse?

___ A man’s cane?
___ A trainer’s whip?
___ A fish?
___ A ball?
___ A curtain?
___ A test?

Practically everyone in the audience raised a hand to answer “yes” to the first question,
and I suspect you also see a man in Figure 5.1. But only about one-half of the audience
acknowledged seeing a woman in the drawing, and many said they had seen an animal.
When I asked what type of animal, the common response heard across the room was
“seal.” This drew many laughs, and the laughter got louder when I asked what else was
quickly perceived in the illustration.
Several people saw a woman’s purse and a man’s cane; others said they had seen a
trainer’s whip, a fish, and a beach ball. Some remembered seeing a curtain. Others saw part
of a circus tent. What did you see in Figure 5.1?
“What’s going on here?” I asked the ASSE audience. Why are we getting these diverse
reactions to one simple picture? Some people speculated about environmental factors in
the seminar room, including lighting, spatial orientation, and visual distance from the pres-
entation screen. Others thought individual differences, including gender, age, occupation,
and personal experiences even “last night” could be responsible. Finally, someone asked
whether the instructions printed in their handouts could have influenced the different per-
ceptions. This was, in fact, the case.
Every handout included the same exact instructions except for a few words. Included
in one-half were all the words given above; the rest had the words “trained seal act” sub-
stituted for “couple at a costume ball.” This was enough to make a marked difference in
perceptions. Perhaps, this makes perfect sense to you. Critical words in the instructions cre-
ated expectations for a particular visual experience. I had set up my audience. Was your
perception of Figure 5.1 influenced by this “set up”?

Biased by context
Now take a look at Figure 5.2. I suspect you have no difficulty reading the sentence as “The
cat sat by the door,” even though the symbol for “H” is exactly the same as the symbol for
“A.” It is a matter of context. The symbol was positioned in a way that influenced your
labeling (or encoding) of the symbol. Likewise, the context or environmental surroundings
in our visual field influence how we see particular stimuli.
The same is true for our other senses—hearing, smelling, tasting, and touching. How
we experience food, which involves the sensations of smell, touch, and taste, can be dra-
matically influenced by the atmosphere in which it is served. This is a basic rule of the
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 72

72 Psychology of safety handbook

T E C T S T
BY T E DOOR
Figure 5.2 The context or circumstances surrounding a stimulus can influence how we
perceive it.

restaurant business. Of course, other factors also bias food sensations, including hunger
and past experiences with the same and similar food.
Let us take a look at Figure 5.3. What label do you give the man in the drawing on the
left? The setting or context certainly influences your decision. The sign, keys, and uniform
are cues that the man is probably a doorman. The environmental context in the drawing on
the right leads to a different perception and label for the same person. Here, he is a police-
man enforcing a safety policy.
Now let us take our discussion of perception and apply it to the workplace. Here, per-
ceptions of people can be shaped by equipment, housekeeping, job titles, and work attire.
In fact, our own job title or work assignment can influence perceptions of ourselves, as well
as affect our perceptions of others. This can dramatically influence how we interact with
others if, indeed, we choose to interact at all.
It is important to recognize this contextual bias. Pick out someone you communicate
with at work, and think how your relationship would be different in another setting.
Would you still feel superior or inferior? Also, as depicted in Figure 5.4, the work setting
has a way of turning individuals into numbers, depersonalizing them. This impression cer-
tainly can be misleading and might cause you to overlook someone’s potential. In another
setting, the same individual might feel empowered and be perceived as a leader.

Figure 5.3 The environment context influences personal perception of the man in
uniform.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 73

Chapter five: Sensation, perception, and perceived risk 73

Figure 5.4 The environmental context influences perception and behavior.

Biased by our past

Perhaps every reader realizes that our past experiences influence our present perceptions.
In Chapter 3, we considered shifts in methods and perceptions needed to achieve a
Total Safety Culture. When I give workshops on paradigm shifts, someone invari-
ably expresses concern about resistance. “He (or she) keeps playing old tapes and is
not open to new ideas,” is a common refrain. Past experiences are biasing present percep-
tions. Actually, there is a long trail of intertwined factors here. Past experiences filter
through a personal evaluation process that is influenced by person factors, including
many past perceived experiences. The cumulative collection of these previous perceived
experiences biases every new experience and makes it indeed difficult to “teach an old dog
new tricks.”
Some participants arrive at my seminars and workshops with a “closed mind” and a
“have to be here” attitude. Others start with an “open mind” and an “opportunity to learn”
outlook. This is another example of the power of personal perception—how much one
learns at these seminars depends on perceptions going in.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 74

74 Psychology of safety handbook

Figure 5.5 Prior perceptual experience influences current perception. Adapted from
Bugelski and Alimpay (1961). With permission.

Perhaps you will find it worthwhile to copy Figure 5.5 and use it for a group demon-
stration. You can show how current impressions are affected by prior perceptions by asking
participants to call out what they see as you reveal each drawing. The drawings must be
uncovered in a particular order. Show the top row of pictures first, revealing each successive
one from left to right. The last picture will probably be identified as the face of an elderly man.
With the top row covered, then show the successive animal pictures of the second row.
Now, the last picture will likely be identified as a rat or mouse. Even after knowing the pur-
pose of the demonstration, you can view serially the row pictures in Figure 5.5 and see how
your perception of the last drawing changes depending on whether you previously looked
at human faces or animals.
Now I would like you to read the sentence given in Figure 5.6 with the intent of under-
standing what it means. The sentence might seem to make little sense, but treat it as a sen-
tence in a memo you have received from a colleague or supervisor. Some of those memos
seem meaningless, too. Your past experience at reading memos, as well as your mood at the
time, can influence how you perceive and react to a memo.
After reading the sentence in Figure 5.6, go back and quickly count the number of let-
ter “Fs” in the sentence. Record your answer. When I show this sentence to workshop par-
ticipants and ask the same question, most will answer “three.” A few will shout out “six,”
usually because they have seen the demonstration before. “Six” is actually the correct
answer, but even after knowing this, a number of people cannot find more than three “Fs”
in the sentence. Why?
When I was first introduced to this exercise many years ago, I showed the sentence to
my two young daughters, and they both found six “Fs” immediately. Karly was in kinder-
garten and Krista was a second grader. Neither could understand the words.
My wife had the same difficulty as I, and could only see three “Fs.” I remember look-
ing at the sentence over and over trying to find six “Fs” but to no avail. My past experience

FINISHED FILES ARE THE RESULT OF

YEARS OF SCIENTIFIC STUDY COMBINED
WITH THE EXPERIENCE OF MANY
YEARS.

Figure 5.6 Past experience can teach us to overlook details.

820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 75

Chapter five: Sensation, perception, and perceived risk 75

Figure 5.7 Viewing this face from a different orientation (by turning the book upside
down) will influence a different perception.

at speed reading had conditioned me to simply overlook small, unessential words like
“of,” and so I simply did not perceive the “Fs” in the three “ofs.” My history had biased my
perception. I bet you had a similar experience if you had not seen this demonstration
before. If you had seen it, then that experience biased your current perception of Figure 5.6.
Finally, take a look at the woman in Figure 5.7. Notice anything strange, other than the
picture is upside down? Is this face relatively attractive, or at least normal? Now turn the
book upside down and view the woman’s face from the normal orientation. Has your per-
ception changed? Why did you not notice her awkward (actually ugly) mouth when the
picture was upside down? Perhaps both context and prior experience (or learning) biased
your initial perception. I bet this perceptual bias will persist even after you realize the cause
of the distortion, and after viewing the face several times in both positions. A biased per-
ception can be difficult to correct. It is not easy to fight human nature.

Relevance to achieving a Total Safety Culture

Is the relevance of this discussion to occupational safety and health obvious? Perhaps by
understanding factors that lead to diverse perceptions, we can become more tolerant of
individuals who do not appear to share our opinion or viewpoint. Perhaps the person fac-
tors discussed here increase your appreciation and respect for diversity and support the
basic need to actively listen. “Seek first to understand, before being understood” is Covey’s
fifth habit for highly effective people (Covey, 1989).
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 76

76 Psychology of safety handbook

It is also possible that this discussion and the exercises on personal perception have
reduced your tendency to blame individuals for an injury or to look for a single root cause
of an undesirable incident. Before we react to an incident or injury with our own viewpoint,
recommendation, or corrective action, we need to ask others about their perceptions.
I hope I have not reduced your optimism toward achieving a Total Safety Culture.
Maybe I have alerted you to challenges not previously considered. If I have not convinced
you yet to stop claiming “All injuries are preventable,” the next section should do the trick.

Perceived risk
People are generally underwhelmed or unimpressed by risks or safety hazards at work.
Why? Our experiences on the job lead us to perceive a relatively low level of risk. This is
strange. After all, it is quite probable someone will eventually be hurt on the job when you
factor in the number of hours workers are exposed to various hazards.
In Chapter 4, I discussed one major reason for low perceptions of risk in the workplace.
It is elemental, really—we usually get away with risky behavior. As each day goes by with-
out receiving an injury, or even a near hit, we become more accepting of the common belief,
“It is not going to happen to me.” Now, let us further explore why we are generally not
impressed by safety hazards at work.

Real vs. perceived risk

The real risk associated with a particular hazard or behavior is determined by the magni-
tude of loss if a mishap occurs, and the probability that the loss or accident will indeed
occur. For example, the risk that comes from driving during any one trip can be estimated
by calculating the probability of a vehicle crash on one trip and multiplying this value
by the magnitude of injury from a crash. Of course, the injury potential or mortality rate
from a vehicle crash is influenced by many other factors, including size of vehicle(s)
involved, speed of vehicle(s), road conditions, and whether the vehicle occupants were
using safety belts.
On any single trip, the chance of a vehicle crash is minuscule; however, in a lifetime of
driving the probability is quite high, varying from 0.30 to 0.50 depending upon factors such
as geographic location, trip frequency and duration, and characteristics of the driver such
as age, gender, reaction time, or mental state (Evans, 1991). Obviously, the risk of driving
an automobile is difficult to assess, although it has been estimated that 55 percent of all
fatalities and 65 percent of all injuries would have been prevented if a combination shoul-
der and lap belt had been used (Federal Register, 1984).
Estimating the risk of injury from working with certain equipment is even more diffi-
cult to determine, because work situations vary so dramatically. Plus the risk can be elimi-
nated completely by the use of appropriate protective clothing or equipment. Still, many
people do not appreciate the value of using personal protective equipment or following
safe operating procedures. Their perception of risk is generally much lower than actual
risk. This thinking pervades society.
Automobile crashes are the nation’s leading cause of lost productivity, greater than
AIDS, cancer, and heart disease (National Academy Press, 1985; Waller, 1986), but how
many of us take driving for granted?
The risk of a fatality from driving a vehicle or working in a factory is much higher than
from the environmental contamination of radiation, asbestos, or industrial chemicals. Yet,
look at the protests over asbestos in schools and neighborhood chemical plants.
Researchers of risk communication have found that various characteristics of a hazard,
irrelevant to actual risk, influence people’s perceptions (Covell et al., 1991). It is important
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 77

Chapter five: Sensation, perception, and perceived risk 77

Lower Risk Higher Risk

• exposure is voluntary
• hazard is familiar
• hazard is forgettable
• hazard is cumulative
• collective statistics
• hazard is understood
• hazard is controllable
• hazard affects anyone
• preventable
• consequential
• exposure is mandatory
• hazard is unusual
• hazard is memorable
• hazard is catastrophic
• individual statistics
• hazard is unknown
• hazard is uncontrollable
• hazard affects vulnerable
people
• only reducible
• inconsequential

Figure 5.8 Factors on the left reduce perception and are generally associated with the
workplace. Adapted from Sandman (1991). With permission.

to consider these characteristics, because behavior is determined by perceived rather than

actual risk.
Figure 5.8 shows factors that influence our risk perceptions. It is derived from research
by Sandman (1991), Slovic (1991), and their colleagues. The factors listed on the left reduce
perceptions of risk and are typically associated with the workplace. The opposing factors
in the right-hand column have been found to increase risk perception, and these are not
usually experienced in the work setting. As a consequence, our perception of risk on the job
is not as high as it should be and, therefore, we do not work as defensively as we should.
Discussing some of these factors will reveal strategies for increasing our own and others’
perception of risk in certain situations.

The power of choice

Hazards we choose to experience (like driving, skiing, and working) are seen as less risky
than ones we feel forced to endure (like food preservatives, environmental pollution, and
earthquakes). Of course, the perception of choice is also subjective, varying dramatically
among individuals. For example, people who feel they have the freedom to pull up stakes
and move whenever they want would likely perceive less risk from a nearby nuclear plant
or seismic fault. Likewise, employees who feel they have their pick of places to work
generally perceive less risk in a work environment. They are typically more motivated and
less distressed. In the next chapter, I discuss relationships among perceived choice, stress,
and distress.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 78

78 Psychology of safety handbook

Familiarity breeds complacency

Familiarity is probably a more powerful determinant of perceived risk than choice. The
more we know about a risk, the less it threatens us. Remember how attentive you were
when first learning to drive, or when you were first introduced to the equipment in your
workplace? It was not long before you lowered your perceptions of risk, and changed your
behavior accordingly. When driving, for example, most of us quickly shifted from two
hands on the wheel and no distractions to steering with one hand while turning up the
radio and carrying on a conversation.

The power of publicity

It is so easy to tune out the familiar hazards of the workplace. Safety professionals respond
by constantly reminding employees of risks with a steady stream of memos, newsletters,
safety meetings, and signs. Still, these efforts cannot compete with the impact of unusual,
catastrophic, and memorable events broadcast by the media and dramatized on television
and in the movies. Publicity of memorable injuries, like those suffered by John Wayne
Bobbitt and Nancy Kerrigan in 1994, influences misperception of actual risk.

Sympathy for victims

Many people feel sympathy for victims of a publicized incident, even vividly visualizing
the injury as if it happened to them. Personalizing these experiences increases perceived
risk. At work, employees show much more attention and concern for hazards when injuries
or “near hits” are discussed by the coworkers who experienced them, compared to a pres-
entation of statistics. The average person cannot relate to group numbers, but there is
power in personal stories. I have met many people over the years who accepted individual
accounts in lieu of convincing statistics—“The police officer told Uncle Jake he would have
been killed if he had been buckled up”; “Aunt Martha is 91 years old and still smokes two
packs of cigarettes a day.”
This suggests that we should shift the focus of safety meetings away from statistics,
emphasizing instead the human element of safety. Safety talks and intervention strategies
should center on individual experiences rather than numbers. This might be easier said
than done. Encouraging victims to come forward with their stories is often stifled by man-
agement systems in many companies that seem to value fault finding over fact finding,
piecemeal rather than system approaches to injury investigation, and enforcement more
than recognition to influence on-the-job behaviors.

Understood and controllable hazards

Hazards we can explain and control cause much less alarm than hazards that are not
understood and, thus, perceived as uncontrollable. This points up a problem with many
employee safety education and training programs. Workplace hazards are explained in a
way that creates the impression they can be controlled. Indeed, safety professionals often
state a vision or goal of “zero injuries,” implying complete control over the factors that
cause injuries. This actually lowers perceived risk by convincing people the causes of occu-
pational injuries are understood and controllable.
Perhaps it would be better for safety leaders to admit and publicize that only two of the
three types of factors contributing to workplace injuries can be managed effectively—
environmental/equipment factors and work behaviors. As I have already discussed in
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 79

Chapter five: Sensation, perception, and perceived risk 79

preceding chapters, the mysterious inside, unobservable, and subjective world of people
dramatically influences the risk of personal injury. These attitudes, expectancies, percep-
tions, and personality characteristics cannot be measured, managed, or controlled reliably.
Internal human factors make it impossible to prevent all injuries. By discussing the com-
plexity of people and their integral contribution to most workplace hazards and injuries,
you can increase both the perceived value of ongoing safety interventions and the belief
that a Total Safety Culture requires total commitment and involvement of all concerned.

Acceptable consequences
We are less likely to feel threatened by risk taking or a risk exposure that has its own
rewards. But if few benefits are perceived by an at-risk behavior or environmental condi-
tion, outrage—or heightened perceived risk—is likely to be the reaction, along with a con-
certed effort to prevent or curtail the risk.
Some people, for example, perceive guns, cigarettes, and alcohol as having limited
benefit and, thus, lobby to restrict or eliminate these societal hazards. The availability of
and exposure to these hazards will continue, though, as long as a significant number of
individuals perceive the risk benefits to outweigh the risk costs. Cost–benefit analyses are
subjective and vary widely as a function of individual experience. For example, the two
women in Figure 5.9 obviously perceive the consequences of smoking very differently.
On the other hand, the benefits of risky work behaviors are generally obvious to every-
one. For example, it is cooler and more comfortable to work without a respirator. It is also
convenient and enables a worker to be more productive. The costs of not wearing the mask
might be abstract and delayed (if the exposure is not immediately life threatening).
Statistics might point out a chance of getting a lung disease, which will not surface for

Figure 5.9 The perceived consequences of at-risk behavior can vary widely from one per-
son to another.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 80

80 Psychology of safety handbook

decades, if ever. Decisions about risk taking are made every day by workers. By playing the
odds and shooting for short-term gains, risky work practices are often accepted and not
perceived to be as dangerous as they really are.

Sense of fairness
Most people believe in a just and fair world (Lerner, 1975, 1977). In other words, most peo-
ple generally perceive the world as the large rather than the small fish in Figure 5.10. “What
goes around comes around.” “There’s a reason for everything.” “People generally get what
they deserve.”
When people receive benefits like increased productivity from their risky behavior, the
outrage, public attention, or perceived risk is relatively low. On the other hand, when haz-
ards or injuries seem unfair, as when a child is molested or inflicted with a deadly disease,
special attention is given. This increased attention results in more perceived risk.
This makes it relatively easy to obtain contributions or voluntary assistance for pro-
grams that target vulnerable populations, like learning-disabled children. It is fair and just
for the small fish in Figure 5.10 to obtain special assistance. The victims of workplace
injuries, however, are not perceived as weak and defenseless. Occupational injuries are
indiscriminately distributed among employees who take risks, and they deserve what they
get. This is a common perception or attitude and it lowers the outrage we feel when some-
one gets injured on the job. And lower outrage translates into lower perceived risk.

Risk compensation
A discussion of risk perception would not be complete without examining one of the most
controversial concepts in the field of safety. In recent years, it has been given different
labels, including risk homeostasis, risk or danger compensation, risk-offsetting behavior, and per-
verse compensation. Whatever the name, the basic idea is quite simple and straightforward.

Figure 5.10 Justice is a matter of personal perspective.

820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 81

Chapter five: Sensation, perception, and perceived risk 81

Figure 5.11 Personal protective equipment can reduce the perception of risk.

People are presumed to adjust their behavior to compensate for changes in perceived risk.
If a job is made safer with machine guards or the use of personal protective equipment,
workers might reduce their perception of risk and, thus, perform more recklessly. For
example, if the individual depicted in Figure 5.11 is taking a risk owing to the perceived
security of fall protection, we would have support for the risk compensation theory
(Peltzman, 1975).
The notion of taking more risks to compensate for lower risk perception certainly
seems intuitive. I bet every reader has experienced this phenomenon. I clearly remember
taking more risks after donning a standard high school football uniform. With helmet and
shoulder pads, I would willingly throw my body in the path of another player or leap to
catch a pass. I did not perform these behaviors until perceiving security from the personal
protective equipment (PPE).
Today, I experience risk compensation of a different sort on the tennis court. If I get
ahead of my opponent by a few games, I take more chances. I will hit out for a winner or
go to the net for a volley. When I get behind my opponent by two or more games, I play
more conservatively from the base line. I adjust the risk level of my game depending on the
circumstances—my opponent’s skills and the score of the match.
Risk compensation has seemingly universal applications. How can the phenomenon
be denied?
Figure 5.12 depicts a workplace situation quite analogous to my teenage experiences
on the football field. There appear to be limited scientific data to support the use of com-
mercially available back belts (Metzgar, 1995). Could this be partly owing to risk compen-
sation? If the use of a back belt leads to employees lifting heavier loads, then the potential
protection from this device could be offset by greater risk taking. The protective device
could give a false sense of security and reduce one’s perception of being vulnerable to back
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 82

82 Psychology of safety handbook

Figure 5.12 Back belts can give a false sense of security.

injury. The result could be more frequent and heavier lifting, and greater probability of
injury. This is why back belt suppliers emphasize the need for training and education in the
use of belts.
Obviously, the notion that an individual’s behavior could offset the safety benefits of
PPE is extremely repugnant to a safety professional. Could this mean that efforts to make
environments safer with engineering innovations are useless in the long run? Are safety
belts and air bags responsible for increases in vehicle speeds? Does this mean laws and pol-
icy to enforce safe behavior actually provoke offsetting at-risk behavior?
Some researchers and scholars are convinced risk compensation is real and detrimen-
tal to injury prevention (Adams, 1985a,b; Peltzman, 1975; Wilde, 1994); others contend the
phenomenon does not exist. Lehman and Gage (1995) proclaim, for example, that “this
alleged theory (risk compensation) has neither experimental nor analytical scien-
tific basis”; Dr. Leonard Evans of General Motors Research Laboratories is quoted as say-
ing “ . . . there are no epicycles and there is no phlogiston . . . similarly, there is no risk
homeostasis” (Wilde, 1994).

Support from research

In fact, there is scientific evidence that risk compensation, or risk homeostasis, is real, as our
intuition or common sense tells us, but the off-setting or compensating behavior does not
negate the benefits of intervention. Although football players increase at-risk behaviors
when suited up, for example, they sustain far fewer injuries than they would without the
PPE. This is true even if a lack of protection reduced their risk taking substantially. More
important, if people lower the level of risk they are willing to accept (as promoted in a Total
Safety Culture), then risk compensation or risk homeostasis is irrelevant. I shall explain this
“good news” further, but first let us look more closely at research evidence supporting the
phenomenon.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 83

Chapter five: Sensation, perception, and perceived risk 83

Comparisons between people. The notion of risk compensation made its debut
among safety professionals following the theorizing and archival research of University of
Chicago economist, Dr. Sam Peltzman (1975). Peltzman systematically compared vehicle
crash statistics before (1947–1965) vs. after (1966-1972) the regulated installation of safety
engineering innovations in vehicles, including seat belts, energy-absorbing steering
columns, padded instrument panels, penetration-resistant windshields, and dual braking
systems. As predicted by risk compensation theory, Peltzman found that these vehicle-
manufacturing safety standards had not reduced the frequency of crash fatalities per miles
driven. Perhaps the most convincing evidence of risk compensation was that the cars
equipped with safety devices were involved in a disproportionately high number of
crashes.
Peltzman’s article has been criticized on a number of counts, primarily statistical but it
did stimulate follow-up investigations. Dr. John Adams of University College, London,
UK, for example, compared traffic fatality rates between countries with and without safety-
belt use laws. His annual comparisons (from 1970 to 1978) showed dramatic reductions in
fatal vehicle crash rates after countries introduced seat-belt use laws. Taken alone this data
would lend strong support to seat-belt legislation. But the drop in fatality rates was even
greater in countries without safety-belt use laws (Adams, 1985b). Apparently, the large-
scale impact of increased use of vehicle safety belts has not been nearly as beneficial as
expected from laboratory crash tests. Risk compensation has been proposed to explain this
discrepancy.
There are obviously other possible explanations for the fluctuations in large data bases
compiled and analyzed by Peltzman (1975) and Adams (1985a,b)—changes in the econ-
omy, improvements in vehicle performance, and media promotion of particular life styles,
to name a few. Regarding safety-belt mandates, for example, it is generally believed that
the safest drivers are the first to buckle up and comply, meaning the most prominent
decrease in injuries from vehicle crashes will not occur until the remaining 30 percent
buckle up—those currently resisting belt-use laws (Campbell et al., 1987). In other words,
“those segments of the driving population who are least likely to comply with safe driving
laws are precisely those groups that are at highest risk of serious injury” (Waller, 1987).
Research supports this presumed direct relationship between at-risk behavior and
noncompliance with safety policy. Young males (Preusser et al., 1985), persons with ele-
vated blood alcohol levels (Wagenaar, 1984), and “tailgaters” who drive dangerously close
to the vehicles they follow (Evans et al., 1982) are less likely to comply with a belt-use law.
These findings could certainly have implications for occupational safety. If the riskiest
workers are least likely to comply with rules and policy, traditional top-down enforcement
and discipline are not sufficient to achieve a Total Safety Culture. Of course, this is a pri-
mary theme of this book. But let us get back to the issue of risk compensation.
Studies that compared risk behaviors across large data sets and found varying charac-
teristics among people who complied with a safety policy vs. those who did not certainly
weaken the case for risk compensation. Behavioral scientists call this between groups
research, and it can only indirectly test the occurrence of risk compensation. Because risk
compensation theory predicts that individuals increase their risky behavior after perceiv-
ing an increase in safety or security, the theory can only be tested by comparing the same
group of individuals under different conditions. Behavioral-science researchers call this a
within subjects design.
Within subject comparisons. Most within subject tests of risk compensation
theory have been restricted to simulated laboratory investigations (Wilde et al., 1985).
These observations of different risk conditions are time consuming and quite difficult
to pull off in a real-world situation. Dr. Fredrick Streff and I conducted one such study in
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 84

84 Psychology of safety handbook

1987. We built an oval clay go-cart track about 100 m in circumference and equipped a
5-horsepower go-cart with an inertia reel-type combination shoulder–lap safety harness.
Subjects were told to drive the go-cart around the track “quickly, but at a speed that is
comfortable for you” (Streff and Geller, 1988). The 56 subjects were either buckled or
unbuckled in the first of two phases of driving trials. After the first phase, the safety con-
dition was switched for one-half the subjects. That is, for these subjects the safety belt was
no longer used if the drivers had previously been buckled up, or the belt was used by driv-
ers who previously did not use it. The speed and accuracy of each subject’s driving trial
were systematically measured. Following the first and second phases (consisting of 15 tri-
als each), the subjects completed a brief questionnaire to assess their perceived risk while
driving the go-cart.
The between subject comparisons showed no risk compensation. Subjects who used
the safety belt for all trials did not drive faster than subjects who never used the safety belt.
Perceptions of risk were not different across these groups of subjects, either.
On the other hand, the within subject differences did show the predicted changes in
risk perception and significant risk compensation. Subjects reported feeling safer when
they buckled up, and subsequently drove the go-cart significantly faster than subjects who
used the safety belt during both phases. Those who took off their safety belts reported a sig-
nificant decrease in perceived safety, but this change in risk perception was not reflected in
slower driving speeds, compared to drivers who never buckled up in the go-cart.
Our go-cart study was later followed up in the Netherlands using a real car on real
roads. Convincing evidence of risk compensation was found (Jansson, 1994). Specifically,
habitual, “hard-core” nonusers of safety belts buckled up at the request of the experi-
menter. Compared to measures taken when not using a safety belt, these buckled-up driv-
ers drove faster, followed more closely behind vehicles in front of them, changed lanes at
higher speeds, and braked later when approaching an obstacle.

Implications of risk compensation

I am convinced from personal experience and reading the research literature that risk com-
pensation is a real phenomenon. What does this mean for injury prevention? Wilde (1994)
says it means safety excellence cannot be achieved through top-down rules and enforce-
ment. Some people only follow the rules when they are supervised and might take greater
risks when they can get away with it. This behavior is not only predicted by risk compen-
sation theory, but also by the theory of psychological reactance (Brehm, 1966) discussed in
Chapter 3 (see Figure 3.1). According to reactance theory, some people feel a sense of free-
dom or accomplishment when they do not comply with top-down regulations. They like
to beat the system. Skinner (1971) referred to reactive behavior as countercontrol—a means
by which some people attempt to assert their freedom and dignity when feeling controlled.
Whether dangerous behavior results from psychological reactance or risk compensa-
tion, our risk reduction attempts are the same. As the title of his book Target Risk indicates,
Wilde advocates that safety interventions need to lower the level of risk people are willing
to tolerate. This requires a change in values. Wilde (1994) asserts that improvements in
safety cannot be “achieved by interventions in the form of training, engineering or enforce-
ment” (page 213). “The extent of risk taking with respect to safety and health in a given
society, therefore, ultimately depends on values that prevail in that society, and not on the
available technology” (Wilde, 1994, page 223).
I hope it is obvious that Wilde’s position is consistent with the theme of this text. When
people understand and accept the paradigm shifts needed for a Total Safety Culture (see
Chapter 3), they are on track to reduce their tolerance for risk. Next, they need to believe in
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 85

Chapter five: Sensation, perception, and perceived risk 85

the vision of a Total Safety Culture and buy into the mission of achieving it. Then, they need
to understand and accept the procedures that can achieve this vision. These methods are
explained in Section 3 of this text. Through a continuous process of applying the right pro-
cedures, the workforce will feel empowered to actively care for a Total Safety Culture.
Finally, they will come to treat safety as a value rather than a priority. I discuss these con-
cepts more fully in Section 4.

In conclusion
This chapter explored the concept of selective sensation or perception, and related it to per-
ceived risk and injury control. Visual exercises illustrated the impact of past experience and
contextual cues on present perception. This allows us to appreciate diversity and realize the
value of actively listening during personal interaction. We need to work diligently to
understand the perceptions of others before we impulsively jump to conclusions or
attempt to exert our influence.
It is important to realize, however, that people often hold on stubbornly to a precon-
ceived notion about someone or something. As illustrated in Figure 5.13, this bias is often
caused by prior experience, and it can dramatically affect perception. Perhaps you know
this phenomenon as prejudice, one-sidedness, history, discrimination, pigheadedness, or
just plain bias. I like the label Langer (1989) uses for this kind of mindlessness—premature
cognitive commitment.
I like the term “premature cognitive commitment” because it makes me mindful of the
various ingredients of inflexible prejudice. First, it is premature, meaning it is accom-
plished before adequate diagnosis, analysis, and consideration. Second, it is cognitive,
meaning it is a mental process that influences our perceptions, our attitudes, and our
behaviors. Finally, it is a commitment. It is not just a fleeting notion or temporary opinion.

Figure 5.13 We all have premature cognitive commitment.

820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 86

86 Psychology of safety handbook

It is a solid, relatively permanent position or sentiment that affects what information a per-
son seeks, attends to, understands, appreciates, believes, and uses.
Premature cognitive commitment is the root cause of much, if not most, interpersonal
conflict. And it is a barrier we must overcome to develop the interdependent teamwork
needed for a Total Safety Culture. Being mindful of premature cognitive commitment in
ourselves and others will not stop this bias, but it is a start.
We must realize that perceptions of risk vary dramatically among individuals. And we
cannot improve safety unless people increase their perception of, and reduce their toler-
ance for, risk. Changes in risk perception and acceptance will occur when individuals get
involved in achieving a Total Safety Culture with the principles and procedures discussed
in this Handbook.
Several factors were discussed in this chapter that affect whether employees react to
workplace hazards with alarm, apathy, or something in between. Taken together, these fac-
tors shape personal perceptions of risk and illustrate why the job of improving safety is so
daunting. This justifies more resources for safety and health programs, as well as inter-
vention plans to motivate continual employee involvement. I discuss various intervention
approaches in Section 4. But before discussing strategies to fix the problem, we need to
understand how stress, distress, and personal attributions contribute to the problem. That
is our topic for the next chapter.

References
Adams, J. G. U., Risk and Freedom, Transport Publishing Projects, London, 1985a.
Adams, J. G. U., Smead’s law, seat belts and the emperor’s new clothes, in Human Behavior and
Traffic Safety, Evans, L. and Schwing, R. C., Eds., Plenum, New York, 1985b.
Brehm, J. W., A Theory of Psychological Reactance, Academic Press, New York, 1966.
Bugelski, B. R. and Alimpay, D. A., The role of frequency in developing perceptual sets, Can. J.
Psychol., 15, 205, 1961.
Campbell, B. J., Stewart, J. R., and Campbell, F. A., 1985–1986 Experiences with Belt Laws in the United
States, UNC Highway Safety Research Center, Chapel Hill, NC, 1987.
Covell, V. T., Sandman, P. M., and Stovie, P., Guidelines for communicating information about
chemical risks effectively and responsibly, in Acceptable Evidence: Science and Values in Risk
Management, Mayo, D. G. and Hollander, R. D., Eds., Oxford University Press, New York,
1991.
Covey, S. R., The Seven Habits of Highly Effective People: Restoring the Character Ethic, Simon &
Schuster, New York, 1989.
Evans, L., Traffic Safety and the Driver, Van Nostrand Reinhold, New York, 1991.
Evans, L., Wasielawski, P., and von Buseck, C. R., Compulsory seat belt usage and driver risk-tak-
ing behavior, Hum. Fact., 24, 41, 1982.
Federal Register, Federal motor vehicle safety standards: occupant crash protection, Final Rule, U.S.
Department of Transportation, Washington, DC., 48(138), July 1984.
Janssen, W., Seat belt wearing and driving behavior: an instrumented-vehicle study, Accid. Anal.
Prev., 26, 249, 1994.
Langer, E., Mindfulness, Addison-Wesley, Reading, MA, 1989.
Lehman, B. J. and Gage, H., How much is safety really worth? Countering a false hypothesis, Prof.
Saf., 40(6), 37, 1995.
Lerner, M. S., The justice motive in social behavior J. Soc. Iss., 31(3), 1, 1975.
Lerner, J. J., The justice motive: some hypotheses as to its origins and forms, J. Person., 45, 1, 1977.
Metzgar, C. R., Placebos, back belts and the Hawthorne effect, Prof. Saf., 40(4), 26, 1995.
National Academy Press, Injury in America: A Continuing Public Health Problem, National Academy
Press, Washington, D.C., 1985.
Peltzman, S., The effects of automobile safety regulation, J. Pol. Econ., 83, 677, 1975.
820049_CRC1_L1540_CH05 11/10/00 1:34 PM Page 87

Chapter five: Sensation, perception, and perceived risk 87

Preusser, D. F., Williams, A. F., and Lund, A. K., The Effect of New York’s Seat Belt Law on Teenage
Drivers, Insurance Institute for Highway Safety, Washington, D.C., 1985.
Sandman, P. M., Risk  Hazard + Outrage: a Formula for Effective Risk Communication, videotaped
presentation for the American Industrial Hygiene Association, Environmental Communication
Research Program, Cook College, Rutgers University, New Brunswick, NJ, 1991.
Skinner, B. F., Beyond Freedom and Dignity, Alfred A. Knopf, New York, 1971.
Slovic, P., Beyond numbers: a broader perspective on risk perception and risk communication, in
Deceptable Evidence: Science and Values in Risk Management. Mayo, D. G. and Hollander, R. D.,
Eds., Oxford University Press, New York, 1991.
Streff, F. M. and Geller, E. S., An experimental test of risk compensation: between-subject versus
within-subject analysis, Accid. Anal. Prev., 20, 277, 1988.
Wagenaar, A. C., Restraint usage among crash-involved motor vehicle occupants. Report VMTRI-
84-2, University of Michigan Transportation Research Institute, Ann Arbor, MI, 1984.
Waller, J. A., State liquor laws as enablers for impaired driving and other impaired behaviors, Am. J.
Publ. Health, 76, 787, 1986.
Waller, J. A., Injury: conceptual shifts and prevention implications, Ann. Rev. Publ. Health, 8, 21,
1987.
Wilde, G. J. S., Target Risk. PDE Publications, Toronto, Ontario, Canada, 1994.
Wilde, G. J. S., Claxton-Oldfield, S. P., and Platenius, P. H., Risk homeostasis in an experimental
context, in Human Behavior and Traffic Safety. Evans, L. and Schwing, R. C., Eds., Plenum Press,
New York, 1985.
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 89

chapter six

Stress vs. distress

Stressors can contribute to a near hit or an injury; they are barriers to achieving a Total Safety
Culture. However, stressors can provoke positive stress rather than negative distress, which can lead
to constructive problem solving rather than destructive, at-risk behavior. This chapter explains the
important distinction between stress and distress, and defines factors, which determine the occur-
rence of one or the other.
The concept of “attribution” is introduced as a cognitive process we use to turn stressors into pos-
itive stress or negative distress. Attribution bias can reduce distress, but it can also prevent a con-
structive analysis of an injury or property damage incident. This chapter explains the benefits and
liabilities of such bias and shows its role in shifting stress to distress or vice versa.

“Even if you’re on the right track, you’ll get run over if you just sit there.”—Will Rogers

Judy was tired and worried. She had just left her six-year-old son at
her sister’s house with instructions for her to take him to Dr.
Slayton’s office for a 10:30 a.m. appointment. She had been up much
of the night with Robbie, attempting to comfort him. With tears in his
eyes, he had complained of a “hurt” in his stomach. This was the
third night his cough had periodically awakened her, but last night
Robbie’s cough was deeper, seemingly coming from his lungs.
Judy arrived at her workstation a little later than normal and
found it more messy than usual. Grumbling under her breath that
the night shift had been “careless, sloppy, and thoughtless,” she
downed her usual cup of coffee and waited for the production line to
crank up. She did not clean the work area. After all, it was not her
mess. The graveyard shift is not nearly as busy as the day shift. How
could they be so sloppy and inconsiderate?
Judy was ready to start her inspection and sorting when she
noticed the “load cart” was misaligned. She inserted a wooden han-
dle in the bracket and pulled hard to jerk the cart in place. Suddenly
the handle broke, and Judy fell backward against the control panel.
Fortunately, she was not hurt, and the only damage was the broken
handle. Judy discarded it, inserted another one, and put the cart in
place.
During lunch Judy called the doctor’s office and learned that her
son had the flu, and would be fine in a day or two. She completed the
day in a much better mood, and without incident.
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 90

90 Psychology of safety handbook

At the end of her shift, Judy filled out a near-hit report on her
morning mishap. She wrote that someone on the graveyard shift had
left her work area in disarray, including a misaligned load tray. She
also indicated that the design of the cart handle made damage likely;
in the past other cart handles had been broken. She recommended a
redesign of the handle brackets and immediate discipline for the
graveyard shift in her work area.

The fact that Judy filled out a near-hit report is certainly good news, but was this a com-
plete report? Were there some personal factors within Judy that could have influenced the
incident? Was Judy under stress or distress and, if so, could this have been a contributing
factor? It has been estimated that from 75 to 85 percent of all industrial injuries can be par-
tially attributed to inappropriate reactions to stress (Jones, 1984). Furthermore, stress-
related headaches are the leading cause of lost-work time in the United States (Jones, 1984).
Judy’s near-hit report was also clearly biased by common attribution errors researched
by social psychologists and used by all of us at some time to deflect potential criticism and
reduce distress. Attribution errors, along with stress and distress, represent potential bar-
riers to achieving a Total Safety Culture.

What is stress?
In simple terms, stress is a psychological and physiological reaction to events or situations
in our environment. Whatever triggers the reaction is called a stressor. So stress is the reac-

Figure 6.1 Certain environmental conditions and personality states contribute to stress
and distress.
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 91

Chapter six: Stress vs. distress 91

tion of our mind and body to stressors such as demands, threats, conflicts, frustrations,
overloads, or changes.
Figure 6.1 depicts a scene that might seem familiar—perhaps too familiar. So many
people with so much to do and not enough time to do it. Then our goals are thwarted, and
our stress turns to distress. Such frustration can lead to aggression and a demeanor that
only increases our distress. It is a vicious cycle, and it certainly increases our propensity for
personal injury. Certain personality characteristics referred to as “Type A” are more likely
to experience the time urgency and competitiveness depicted in Figure 6.1, and these are
associated with higher risk for coronary disease (Rhodewalt and Smith, 1991).
Let us return to the basics of stress. When you interpret a situation as being stressful,
your body prepares to deal with it. This is the fight-or-flight syndrome, a process controlled
through our sympathetic nervous system (see Figure 6.2). Adrenaline rushes into the
bloodstream, the heart pumps faster, and breathing increases. Blood flows quickly from
our abdomen to our muscles—causing those “butterflies in our stomach.” We can also feel
tense muscles or nervous strain in our back, neck, legs, and arms (Selye, 1974, 1976).

Constructive or destructive?
We usually talk about stress in negative terms. It is unwanted and uncomfortable, but the
first definition of stress in my copy of The American Heritage Dictionary (1992) is “impor-
tance, significance, or emphasis placed on something” (page 1205). Similarly, The New
Merriam-Webster Dictionary (1989) defines stress as “a factor that induces bodily or mental
tension . . . a state induced by such a stress . . . urgency, emphasis” (page 701).
The bad state is distress. Distress is defined as “anxiety or suffering . . . severe strain
resulting from exhaustion or an accident” (The American Heritage Dictionary, 1991, page 410)
or “suffering of body or mind: pain, anguish: trouble, misfortune . . . a condition of des-
perate need” (The New Merriam-Webster Dictionary, 1989, page 224).
Psychological research supports these distinctions between stress and distress. Stress
can be positive, giving us heightened awareness, sharpened mental alertness, and an
increased readiness to perform. Certain psychological theories presume that some stress is

Figure 6.2 An initial reaction to a stressor is fight or flight.

820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 92

92 Psychology of safety handbook

Figure 6.3 Arousal from external pressure (or a stressor) improves performance to an opti-
mal point.

necessary for people to perform. The person who asserts, “I work best under pressure,”
understands the motivational power of stress, but can too much pressure, too many dead-
lines, be destructive?
I am sure most of you have been in situations—or predicaments—where the pres-
sure to perform seemed overwhelming. This is the point where too much pressure
can hurt performance, where stress becomes distress. The relationship between external
stimulation or pressure to perform and actual performance is depicted in Figure 6.3.
This inverted U-shaped function is known as the Yerkes-Dodson Law (Yerkes and
Dodson, 1908).
The Yerkes-Dodson law states that, up to a point, performance will increase as arousal,
or pressure to perform well, increases, but the best performance comes when arousal is
optimum rather than maximum. Push a person too far and his performance starts to dete-
riorate. In fact, at exceptionally high levels of pressure or tension a person might perform
as poorly as when he is hardly stimulated at all. Ask someone who is hysterical and some-
one who is about to fall asleep to do the same job, and you will not be pleased by either
one’s results.
Hans Selye, the Austrian-born founder of stress research said, “Complete freedom
from stress is death” (Selye, 1974). It is extreme, disorganizing stress we need to avoid.
Watch out for distress.

The eyes of the beholder

Perceptions play an important role in stress and distress. The boss gives a group of employ-
ees a deadline; some tighten up inside, others take it in stride. Some circle their calendars
and cannot take their minds off the due date. Others seem to pay it no mind.
When a stressor is noticed and causes a reaction, the result can be constructive or
destructive. If we believe we are in control—that we can deal with the overload, frustra-
tions, conflicts, or whatever is triggered by the stressor—we become aroused and
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 93

Chapter six: Stress vs. distress 93

Figure 6.4 Our butterflies are aligned for stress and misaligned for distress.

motivated to go beyond the call of duty. We actually achieve more. On the other hand,
when we believe we cannot handle the demands of the stressor, the resulting psychologi-
cal and physiological reactions are likely to be detrimental to our performance and our
health and safety.
Why does a manager’s deadline motivate one person and distress another? It depends
on a number of internal person factors. These include the amount of arousal already pres-
ent in the individual and the person’s degree of preparedness or self-confidence. The dif-
ference between arousal leading to stress vs. distress is illustrated in Figure 6.4.
Those “butterflies” we feel in our stomach can help or hinder performance, depending
on personal perception. When the “butterflies” are aligned for goal-directed behavior,
we feel in control of the situation. The stress is positive—it arouses or motivates perform-
ance improvement. Such is the case for the runner on the right side of Figure 6.4. The run-
ner on the left, though, does not feel prepared. The “butterflies” are misaligned and
scattered in different directions. Stress is experienced as distress. This arousal can divert
our attention, interfere with thinking processes, disrupt performance, and reduce our abil-
ity and overall motivation to perform well. The result can be at-risk behavior and a serious
injury.

Identifying stressors
Stress or distress can be provoked by a wide range of demands and circumstances. Some
stressors are acute, sudden life events, such as death or injury to a loved one, marriage,
marital separation or divorce, birth of a baby, failure in school or at work, and a job pro-
motion or relocation. Other stressors include the all-too-frequent minor hassles of every-
day life, from long lines and excessive traffic (Figure 6.1) to downsized work conditions
and worries about personal finances (Holmes and Masuda, 1974).
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 94

94 Psychology of safety handbook

Prolonged, uncontrollable stressors can lead to “burnout.” Common symptoms of

burnout include

1. Physical exhaustion resulting in lack of energy, headaches, backaches and general

fatigue.
2. Emotional exhaustion manifested by loss of appetite, feelings of helplessness, and
depression.
3. Mental or attitudinal exhaustion revealed through irritability, cynicism, and a neg-
ative outlook on life (Baron, 1998).

Obviously, burnout puts people at risk for causing injury to themselves or others.
Our jobs or careers are filled with stressors. Consider for a moment how much time
you spend working, or thinking about your work. Some workplace stressors are obvious;
others might not be as evident but are just as powerful. Work overload can obviously
become a stressor and provoke either stress or distress, but what about “work underload”?
Being asked to do too little can produce profound feelings of boredom, which can also lead
to distress. Performance appraisals are stressors that can be motivational if perceived as
objective and fair, or they can contribute to distress and inferior performance if viewed as
subjective and unfair.
Other work-related factors that can be perceived as stressors and lead to distress and
eventual burnout include: role conflict or ambiguity; uncertainty about one’s job responsi-
bilities; responsibility for others; a crowded, noisy, smelly, or dirty work environment; lack
of involvement or participation in decision making; interpersonal conflict with other
employees; and insufficient support from coworkers (Maslach, 1982).
We are at a good point in our discussion for you to complete the questionnaire in Figure
6.5. It was developed by Rice (1992) to identify individuals’ various workplace stressors.
Instructions for scoring your distress profile are included. The survey addresses three
domains of work distress—interpersonal relations, physical demands, and level of mental
interest—and these are totaled for an overall distress score.
The interpersonal scale measures distress related to personal relationships, or lack
thereof, at work. Stressors in this category can emanate from communication breakdowns
with coworkers or supervisors, lack of appropriate job training or recognition processes,
infrequent opportunities for personal choice in work assignments or work processes, and
insufficient social support from colleagues or team members.
The second scale of the distress questionnaire estimates the physical demands of work
that can wear on an individual day after day. This includes environmental stressors such
as noise, crowded conditions, and incessant work demands; personal stressors such as feel-
ing overworked or ineffective; and interpersonal stressors like insufficient team support.
When evaluating your score for this scale, it is important to understand the critical rela-
tionship between the outside world and your inside world—the world of your own per-
ceptions. Remember, the same work demands and interpersonal stressors can result in
stress and increased productivity for some employees but lead to distress and burnout
for others.

Work stress profile

The interest scale reflects one’s personal reaction to the stressors of his or her workplace. A
high score reflects a low level of personal interest, commitment, and involvement for your
job. This may indicate a need to change jobs or perhaps alter the way you view your work
situation. You might perceive it as an opportunity rather than a necessity. A job should be per-
ceived as something you get to do, not something you got to do. Interacting more effectively with
work associates, especially through more active listening, can readily turn job distress into
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 95

Chapter six: Stress vs. distress 95

Never Rarely Sometimes Often Most Times

1. Support personnel are incompetent
1 2 3 4 5
or inefficient.
2. My job is not very well defined. 1 2 3 4 5
3. I am not sure about what is expected of me. 1 2 3 4 5
4. I am not sure what will be expected of me
1 2 3 4 5
in the future.
5. I cannot seem to satisfy my superiors. 1 2 3 4 5
6. I seem to be able to talk with my superiors. 5 4 3 2 1
7. My superiors strike me as incompetent,
1 2 3 4 5
yet I have to take orders from them.
8. My superiors seem to care about me as a
5 4 3 2 1
person.
9. There is a feeling of trust, respect, and 5 4 3 2 1
friendliness between me and my superiors.
10. There seems to be tension between
1 2 3 4 5
management and operators.
11. I have a sense of individuality in carrying
5 4 3 2 1
out my job duties.
12. I feel as though I can shape my own
5 4 3 2 1
destiny in this job.
13. There are too many bosses in my area. 1 2 3 4 5
14. It appears that my boss has "retired on
1 2 3 4 5
the job."
15. My superiors give me adequate feedback
5 4 3 2 1
about my job performance.
16. My abilities are not appreciated by my
1 2 3 4 5
superiors.
17. There is little prospect of personal or
1 2 3 4 5
professional growth in this job.
18. The level of participation in planning and
5 4 3 2 1
decision making is satisfactory.
19. I feel that I am over-educated for this job. 5 4 3 2 1
20. I feel that my educational background is just
1 2 3 4 5
right for this job.
21. I fear that I will be laid off or fired. 1 2 3 4 5
22. In-service training for my job is inadequate. 1 2 3 4 5
23. Most of my colleagues seem uninterested
1 2 3 4 5
in me as a person.
24. I feel uneasy about going to work. 1 2 3 4 5
25. There is no release time for personal affairs
1 2 3 4 5
or business.
26. There is obvious sex/race/age discrimination
1 2 3 4 5
in this job.
NOTE: Complete the entire questionnaire first. Then add all the values you circled
for questions 1- to 26 and enter here.
Total 1 to 26 _______
27. The physical work environment is crowed,
1 2 3 4 5
noisy, or dreary.
28. Physical demands of the job are unreasonable
(heavy lifting, extraordinary concentration 1 2 3 4 5
required, etc.).
29. My workload is never-ending. 1 2 3 4 5
30. The pace of work is too fast. 1 2 3 4 5
31. My job seems to consist of responding to
1 2 3 4 5
emergencies.

Figure 6.5a Various workplace stressors are identified in this survey.

820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 96

96 Psychology of safety handbook

32. There is no time for relaxation , coffee breaks,

1 2 3 4 5
of lunch breaks on the job.
33. Job deadlines are constant and unreasonable. 1 2 3 4 5
34. Job requirements are beyond the range of my
1 2 3 4 5
ability.
35. At the end of the day, I am physically
1 2 3 4 5
exhausted from work.
36. I can’t even enjoy my leisure because of the
1 2 3 4 5
toll my job takes on my energy.
37. I have to take work home to keep up. 1 2 3 4 5
38. I have responsibility for too many people. 1 2 3 4 5
39. Support personnel are too few. 1 2 3 4 5
40. Support personnel are incompetent or
1 2 3 4 5
inefficient.
41. I am not sure about what is expected of me. 1 2 3 4 5
42. I am not sure what will be expected of me in
1 2 3 4 5
the future.
43. I leave work feeling burned out. 1 2 3 4 5
44. There is little prospect for personal or
1 2 3 4 5
professional growth in this job.
45. In-service training for my job is inadequate. 1 2 3 4 5
46. There is little contact with colleagues on the
1 2 3 4 5
job.
47. Most of my colleagues seem uninterested in
1 2 3 4 5
me as a person.
48. I feel uneasy about going to work. 1 2 3 4 5
NOTE: Complete the entire questionnaire first. Then add all the values you circled
for questions 27- to 48 and enter here.
Total 27 to 48 _______
49. The complexity of my job is enough to keep
5 4 3 2 1
me interested.
50. My job is very exciting. 5 4 3 2 1
51. My job is varied enough to prevent boredom. 5 4 3 2 1
52. I seem to have lost interest in my work. 1 2 3 4 5
53. I feel as though I can shape my own destiny
5 4 3 2 1
in this job.
54. I leave work feeling burned out. 1 2 3 4 5
55. I would continue to work at my job even if I
5 4 3 2 1
did not need the money.
56. I am trapped in this job. 1 2 3 4 5
57. If I had it to do all over again, I would still
5 4 3 2 1
choose this job.

NOTE: Now go back and add the values for questions 1- to 26. Do the same for questions
27- to 48. Enter the values where indicated. Then add all the values circled for questions
49 to 57.
Total 49 to 57 _______

Last, enter those sums for each of the following groups of questions and add them all
to get a cumulative total.
1- to 26 27 to -48 49 to -57
QUESTIONS: 1 to 57
Interpersonal Physical Conditions Job Interest
TOTALS: ________ + _________ + ________ + _______

Figure 6.5b Various workplace stressors are identified in this survey.

820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 97

Chapter six: Stress vs. distress 97

⇐Distress⇒ ⇐Medium Distress⇒ ⇐High Distress⇒

Interpersonal …39…. 43…. 46 ….. 51….. 54….. 57….. 62 ….. 68….. 75
Physical …35…. 40…. 44 ….. 48….. 52….. 55….. 58 ….. 62….. 67
Interest …13…. 15…. 17 ….. 18….. 19….. 21….. 23 ….. 25….. 27
Total …91.…101….111 …..117…..123…..134…..141 …..151…..167
Percentile …10…. 20…. 30 ….. 40….. 50….. 60….. 70 ….. 80….. 90

Figure 6.6 Normative data from a sample of 275 school psychologists can be used to com-
pare individual scores from the “Work Distress Profile.” Adapted from Rice (1992). With
permission.
productive stress. More suggestions for decreasing personal distress are offered in the next
section of this chapter.
After obtaining your survey totals for each of the three scales and adding these subto-
tals for an overall distress score, you can compare your results with the values given in
Figure 6.6. The numbers here were obtained from a sample of 275 school psychologists
(Rice, 1992) whose job responsibilities might be very different from yours. It might be more
useful to compare your results with others in your work culture.
I urge you not to take your score, or relative rankings with others, too seriously. Surveys like
this are only imperfect estimates of your perceptions and feeling states at the time you
respond to the questions. Answering the questions and deriving a personal score will
surely increase your understanding of both job stress and distress from an environmental,
interpersonal, and personal perspective, but do not get discouraged by a high distress
score. Your distress state can be changed with strategies we shall soon discuss. There is
much you can do on your own, for yourself.
Let us review the key points about stress and distress. Actually, the flow schematic in
Figure 6.7 says it all. First an environmental event is perceived and appraised as a stressor
to be concerned about or as a harmless or irrelevant stimulus. Lazarus (1966, 1991) refers
to this stage of the process as primary appraisal. According to Lazarus, an event is per-
ceived as a stressor if it involves harm or loss that has already occurred, a threat of some
future danger, or a challenge to be overcome.
Harm is how we appraise the impact of an event. For example, if you oversleep and
miss an important safety meeting, the damage is done. In contrast, threat is how we assess
potential future harm from the event. Missing the safety meeting could lower your team’s
opinion of you and reduce your opportunity to get actively involved in a new safety
process. Challenge is our appraisal of how well we can eventually profit from the damage
done. You could view missing the safety meeting as an opportunity to learn from one-on-
one discussions with coworkers. This could demonstrate your personal commitment to the
safety process and allow you to collect diverse opinions.
In this case, you are perceiving the stressor as an opportunity to learn and show com-
mitment. This evaluation occurs during the secondary appraisal stage (Lazarus and
Folkman, 1984), and the result can be positive and constructive. On the other hand, your
appraisal could be downbeat—you see no recourse for missing the safety meeting, and so
you do nothing about it. Now, coworkers might think you do not care about the new safety
process and withdraw their support. In turn, you might give up or actively resist partici-
pating. This outcome would be nonproductive, of course, and possibly destructive.
The secondary appraisal stage, as depicted in Figure 6.7, determines whether the stres-
sor leads to positive stress and constructive behavior or to negative stress and destructive
behavior. The difference rests with the individual. Does he or she assess the stressful
820049_CRC1_L1540_CH06 11/10/00 1:35 PM Page 98

98 Psychology of safety handbook

Environmental
Event

Primary
Appraisal
Unnoticed
Denied
Noticed Harmless
Accepted No
Important Stressor

Stressor

Secondary
Appraisal
In Control No Control
Optimistic Pessimistic

Positive Negative
Stress Distress
Motivating Exhaustion
Energy Burnout
Constructive Destructive
Safe At-Risk
Behavior Behavior

Figure 6.7 Through personal appraisal, people transform stressors into positive stress or
negative distress.

situation as controllable and, thus, remain optimistic during attempts to cope with the
stressor? As illustrated in Figure 6.8, when people judge their stressors as uncontrollable
and unmanageable, a helpless or pessimistic attitude can prevail and lead to distress and
destructive or even at-risk behavior. Several personal, interpersonal and environmental
factors influence whether this secondary appraisal leads to constructive or destructive
behavior. This is the theme of the next section.

Coping with stressors

Seek first to understand, says Covey (1989). This applies not only to relationships with peo-
ple but with stressors as well. Understanding the multiple causes of conflict, frustration,
overload, boredom, and other potential stressors in our lives can sometimes lead to effec-
tive coping mechanisms. These include

• Revising schedules to avoid hassles like traffic and shopping lines.

• Refusing a request that will overload us.
• Finding time to truly relax and recuperate from tension and fatigue.
• Communicating effectively with others to clarify work duties, reduce conflict, gain
support, or feel more comfortable about added job duties.
• Getting reassigned to a task that better fits our present talents and aspirations.

The fact is, though, it is often impossible to avoid sudden (acute) or continual (chronic)
stressors in our lives. We need to deal with these head-on. Believing you can handle the
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 99

Chapter six: Stress vs. distress 99

Figure 6.8 Lack of perceived control can lead to distress.

harm, threat, or challenge of stressors is the first step toward experiencing stress rather
than distress and acting constructively rather than destructively.

Person factors
Certain personality characteristics make some people more resistant to distress. Individuals
who believe they control their own destinies and generally expect the best from life are, in
fact, more likely to gain control of their stressors and experience positive stress rather than
distress, according to research (Bandura, 1982, 1986; Scheier and Carver, 1988, 1992). It is
important to realize that these person factors—self-mastery and optimism—are not per-
manent inborn traits of people. They are states of mind or expectations derived from per-
sonal experience, and they can be nurtured. It is possible to give people experiences that
increase feelings of being “in control”—experiences that lead people to believe something
good will come from their attempts to turn stress into constructive action.
Learning to feel helpless. When I help clients assess the safety climate of their work-
places, I often uncover an attitude among hourly workers, and some managers as well, that
reflects an important psychological concept called “learned helplessness.” For instance,
when I ask workers what they do regularly to make their workplace safer, I often hear:

“Besides following the safety procedures there’s not much I can do for safety around
here.”
“It really doesn’t matter much what I do, whatever will be will be.”
“There’s not much I can do about reducing work injuries; if it’s my time, it’s my time.”

This is learned helplessness. The concept was labeled more than 20 years ago by
research psychologists studying the learning process of dogs (Maier and Seligman, 1976;
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 100

100 Psychology of safety handbook

Seligman, 1975). They measured the speed at which dogs learned to jump a low barrier sep-
arating two chambers in order to avoid receiving an electric shock through the grid floor.
A tone or light would be activated, then a shock was applied to the grid floor of the cham-
ber. At first, the dogs did not jump the hurdle until receiving the shock, but after a few tri-
als the dogs learned to avoid the shock by jumping into the other chamber as soon as the
warning signal was presented.
Some dogs experienced shocks regardless of their behavior before the regular shock-
avoidance learning trials. These dogs did not learn to jump the barrier to escape the shock.
Instead, they typically just laid down in the shock chamber and whimpered. The earlier
bad experience with inescapable shocks had taught the dogs to be helpless. Seligman and
associates coined the term “learned helplessness” to describe this state. Their finding has
been demonstrated in a variety of human experiments as well (Albert and Geller, 1978;
Seligman and Garber, 1980).
Note how prior failures conditioned experimental subjects ranging from dogs to
humans to feel helpless, in fact to be helpless. It is rather easy to assume that workers
develop a “helpless” perspective regarding safety as a result of bad past experiences. If
safety suggestions are ignored, or policies and procedures always come from management,
workers might learn to feel helpless about safety. It is also true, however, that life experi-
ences beyond the workplace can shape an attitude of learned helplessness. Certain indi-
viduals will come to work with a greater propensity to feel helpless in general, and this can
carry over to feelings regarding occupational safety and health.
Learned optimism. A bad experience does not necessarily lead to an attitude of
learned helplessness. You probably know people who seem to derive strength or energy
from their failures, and try even harder to succeed when given another chance. Similarly,
Seligman and colleagues found that certain dogs resisted learned helplessness if they pre-
viously had success avoiding the electric shock. So it is that some people tend to give up in
the face of a stressor, while others fight back.
You probably recognize this difference between learned helplessness and learned opti-
mism as the more popular pessimist vs. optimist distinction. As you have heard it asked
before, “How do you see the glass of water?” Is it half full or half empty? We see it differ-
ently, depending on our current state of optimism or pessimism. This contrast in personal
perception is illustrated humorously in Figure 6.9. The point is that our personality, past
experience, and current situation influence whether we feel optimistic and in control or
pessimistic and out of control.
What can be done to help those who feel helpless? How can we get them to commit to
and participate in the proactive processes of injury prevention? The work climate can play
a critical role here. This happens when employees are empowered to make a difference and
perceive they are successful.
When workers believe through personal experience that their efforts can make a dif-
ference in safety, they develop an antidote for learned helplessness. This has been termed
“learned optimism” (Scheier and Carver, 1992; Seligman, 1991). If the corporate climate
empowers workers to take control and manage safety for themselves and their coworkers,
they can legitimately attribute safety success to their own actions. This bolsters learned
optimism and feelings of being in control. Besides seeing the glass as half full, optimistic
people under stress find ways to fill the rest of the glass.

Fit for stressors

Fitness is another way to increase our sense of personal control and optimism. Being phys-
ically fit increases our body’s ability to cope with the fight-or-flight syndrome discussed
earlier. You probably know the basic guidelines for improving fitness, which include stop
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 101

Chapter six: Stress vs. distress 101

Figure 6.9 Perception affects expectation which affects behavior which, in turn, affects
perception.

smoking; reduce or eliminate alcohol consumption; exercise regularly, at least 3 times a

week for about 30 minutes per session; eat balanced meals with decreased fat, salt, and
sugar; do not skip breakfast; and obtain enough sleep (usually 7 or 8 hours per 24-hour
period for most people). Some of us find that following these guidelines over the long haul
is easier said than done. We need support and encouragement to break a smoking or drink-
ing habit or to maintain a regular exercise routine.
Figure 6.10 illustrates the type of behavior that has come with the computer revolution.
Low physical activity has become the way of work life for many of us. Often this inactivity
spills over into home life. Survey research has shown that only one in five Americans exer-
cises regularly and intensely enough to reduce the risk of stressor-induced heart disease
(Dubbert, 1992). Figure 6.10 also depicts smoking behavior, considered to be the largest
preventable cause of illness and premature death (before age 65) in the United States, and
accounting for approximately 125,000 deaths each year (American Cancer Society, 1989).
Also portrayed in Figure 6.10 is the positive influence of perceived control. Although
the behavior is essentially the same at work (10 to 5) and at home (5 to 10), the individual
is seemingly much happier at home. Why? Because at home he holds the remote control
and therefore perceives more personal control.
But personal control is truly in the eyes of the beholder. Figure 6.11 depicts legitimate
perceptions of control from the subjects of an experiment. These rodents are not usually
considered “in control” of the situation but in many ways they are. By simply changing our
perspective, we can often perceive and accept more personal control at work and this can
turn negative distress into positive stress.

Social factors
A support system of friends, family, and coworkers can do wonders at helping us reduce
distress in our lives (Coyne and Downey, 1991; Janis, 1983; Lieberman, 1983). Social
support can motivate us to do what it takes to stay physically fit and the people around
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 102

102 Psychology of safety handbook

Figure 6.10 Becoming a “mouse potato” by day and a “couch potato” by night can reduce
one’s physical ability to cope with stressors.

us can make a boring task bearable and even satisfying. Of course, they can also turn a
stimulating job into something dull and tedious. It works both ways. People can motivate
us or trigger conflict, frustration, hostility, a win–lose perspective, and distress. It is up to
us to make the most of the people around us. We can learn from those who take effective
control of stressful situations and expect the best or we can listen to the complaining, back-
stabbing, and cynicism of others and fuel our own potential for distress.
It is obviously important to interact with those who can help us build resistance against
distress and help us feel better about potential stressors. We can also set the right example
and be the kind of social support to others that we want for ourselves. The good feelings of
personal control and optimism you experience from reaching out to help others can do
wonders in helping you cope with your own stressors; this actively caring stance builds
your own support system, which you might need if your own stressors get too over-
whelming to handle yourself.
The next section of this chapter introduces another means of reducing distress. It is a
phenomenon that has particular implications for safety. In the aftermath of an injury or
near hit, it can distort reports and incident analyses. This results in inappropriate or less-
than-optimal suggestions for corrective action. This phenomenon of attributional bias can
also create communication barriers between people and limit the co-operative participa-
tion needed to achieve a Total Safety Culture.

Attributional bias
Think back to the anecdote at the start of this chapter. I suggested that Judy’s near-hit report
was incomplete or biased. Specifically, Judy did not report the potential influence of her own
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 103

Chapter six: Stress vs. distress 103

Figure 6.11 Even the most obvious top-down situation allows for perceptions of bottom-
up control.

distress on the incident. Rather, she focused on factors outside her immediate control—the
poor bracket design for the wooden handle and the messy work area left by others. Giving
up personal responsibility eliminated the incident as a stressor for her. She did not have to
deal with any guilt for almost hurting herself and damaging property. Her denial eased her
distress but biased the near-hit report. Psychologists refer to this as an attributional bias. By
understanding when and how this phenomenon occurs, we can focus injury analysis on
finding facts—not faults. This is a paradigm shift needed to achieve a Total Safety Culture.

The fundamental attribution error

Every day, we struggle to explain the actions of others. Why did she say that to me? Why
did the job applicant refuse to answer that question? Why did Joe leave his work station in
such a mess? Why did the secretary hang up on me? Why did Gayle take sick leave? Why
does she allow her young children to ride in the bed of her pick-up truck? Why did the
motorist pull a gun out of his glove compartment to shoot someone in the next car? Why
were Nicole Brown-Simpson and Ronald Goldman murdered so brutally? In trying to
answer questions like these, we point to external, environmental factors, such as equip-
ment malfunctioning, excessive traffic, warm climate, and work demands; or to internal,
person factors, such as personality, intelligence, attitude, or frustration.
Social psychologists have discovered a fundamental attribution error when systematically
studying how people explain the behavior of others (Ross, 1977; Ross et al., 1977). When
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 104

104 Psychology of safety handbook

evaluating others, we tend to overestimate the influence of internal factors and underesti-
mate external factors. We are more apt to judge the job applicant as rude or unaware (inter-
nal factors) than caught off guard by a confusing or unclear question (external factor). Joe
was sloppy or inconsiderate rather than overwhelmed by production demands. The injured
employee was careless rather than distracted by a sudden environmental noise.
That is how we see things when we are judging others. It is different when we evalu-
ate ourselves. The individuals performing the behaviors in the previous paragraph would
say the causes were owing more to external than internal factors.
Here is an example. My university students are quick to judge me as being an extro-
vert—outgoing and sociable. When I lecture in large classes of 600 to 800 students I am ani-
mated and enthused, and they attribute my performance to internal personality traits, but
I know better. I see myself in many different situations and realize just how much my
behavior changes depending on where I am. In many social settings I am downright shy
and reserved. I am very sensitive to external influences.

The self-serving bias

Students who flunk my university exams are quick to blame external factors, like tricky
questions, wrong reading material assigned, and unfair grading. In contrast, students who
do well are quite willing to give themselves most of the credit. It was not that I taught them
well or that the exam questions were straightforward and fair; rather the student is intelli-
gent, creative, motivated, and prepared. This real-world example, which I bet most readers
can relate to, illustrates another type of attributional distortion, referred to as the self-serv-
ing bias (Harvey and Weary, 1984; Miller and Ross, 1975).
How does this bias affect incident or injury analysis? Think of Judy’s near-hit experi-
ence. She protected her self-esteem by overestimating external causes and underplaying
internal factors. This aspect of the self-serving bias is illustrated by the list of explanations
for vehicle crashes given in Figure 6.12. The external and situational excuses given in
Figure 6.12 were taken from actual insurance forms submitted by the drivers.

• The other car collided with mine without giving warning of its intentions.
• A pedestrian hit me and went under my car.
• The guy was all over the road. I had to swerve a number of times before I
hit him.
• I had been shopping for plants all day and was on my way home. As I
reached an intersection, a hedge sprang up, obscuring my vision. I did not
see the other car.
• As I approached the intersection, a stop sign suddenly appeared in a
place where no stop sign had ever appeared before. I was unable to stop
in time to avoid the accident.
• An invisible car came out of nowhere, struck my vehicle, and vanished.
• My car was legally parked as it backed into the other vehicle.
• The pedestrian had no idea which direction to go, so I ran over him.
• The telephone pole was approaching fast. I was attempting to swerve out
of its path when it struck my front end.

Figure 6.12 People are reluctant to admit personal blame for their vehicle crashes.
Excerpted from the Toronto Sun (1977). With permission.
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 105

Chapter six: Stress vs. distress 105

People will obviously go to great lengths to shake blame for unintentional property
damage or injury. This reduces negative stress or distress. No one wants to feel responsible
for a workplace injury, especially if the company puts heavy emphasis on reducing “the
numbers,” such as the plant’s total recordable injury rate.
You can see how a focus on outcome statistics, perhaps supported with rewards for not
having an injury, can motivate people to cover up near hits and injuries whenever possible.
It also motivates a self-serving bias during injury investigations. Actually, the term “inves-
tigation,” as in “criminal investigation,” encourages the self-serving bias. A better term is
“incident analysis,” as I discussed in Chapter 3.
There is good news here. By accentuating outside causes, victims remind us that
behavior is, in fact, influenced by many external factors and, compared to internal factors,
these are more readily corrected.
It is important for us to acknowledge how perceptions can be biased. Outsiders tend to
blame the victim; victims look to extenuating circumstances. We should empathize with
the self-serving bias of the victim because it will reduce the person’s distress. It will shift
attention to external factors that can be controlled more easily than internal factors related
to a person’s attitude, mood, or state of mind.

In conclusion
In this chapter, I explained the difference between stress and distress and discussed
some strategies for reducing distress or turning negative distress into positive stress.
Stress and distress begin with a stressor which can be a major life event or a minor
irritation of everyday living. You can evaluate or appraise the stressor in a way that is
constructive, resulting in safe behavior; or destructive, causing at-risk behavior. When
people are physically fit, in control, optimistic, and able to rely on the social support of
others, they are most likely to turn a stressor into energy for achieving success. This is
positive stress.
When stressors are perceived as insurmountable and unavoidable, distress is likely.
Without adequate support from others, this condition can lead to physical and mental
exhaustion, at-risk behavior, and unintentional injury to oneself or others. We need to
become aware of the potential stressors in our lives and in the lives of our coworkers. In
addition, we need to develop personal and interpersonal strategies to prevent distress in
ourselves and others.
Victims of a near hit or injury will likely feel stressed during their primary appraisal. If
their secondary appraisal clarifies the incident as an uncontrollable failure, negative stress
or distress is likely, but they could interpret it as an opportunity to collect facts, learn, and
implement an action plan to prevent a recurrence. Now, the victim is experiencing positive
stress and constructive behavior is likely.
The work culture, including policies, paradigms, and personnel, can have a dramatic
impact on whether victims of near hits, injuries, or other adversities experience stress or
distress. The fundamental attribution error, where we overestimate personal factors to
explain others’ behavior (“Judy broke the handle because she was tired, stressed out, and
careless”), can provoke distress and pinpoint the very aspects of an incident most difficult
to define and control.
A victim’s natural tendency to reveal a self-serving bias when discussing an incident—
by putting more emphasis on external, situational causes—should be supported by the
work culture. This reduces the victim’s distress and puts the focus on the observable fac-
tors, including behavior, most readily defined and influenced. I detail processes for doing
that in Section 3.
820049_CRC1_L1540_CH06 11/10/00 1:36 PM Page 106

106 Psychology of safety handbook

References
Albert, M. and Geller, E. S., Perceived control as a mediator of learned helplessness, Am. J. Psychol.,
91, 389, 1978.
American Cancer Society, Cancer Facts and Figures —1989, Atlanta, GA, 1989.
The American Heritage Dictionary, 2nd College ed., Houghton Mifflin, New York, 1991.
Bandura, A., Self-efficacy mechanism in human agency, Am. Psychol., 37, 122, 1982.
Bandura, A., Social Foundations of Thought and Action: A Social Cognitive Theory, Prentice-Hall,
Englewood Cliffs, NJ, 1986.
Baron, R. A., Psychology, 4th ed., Allyn & Bacon, Boston, 1998.
Covey, S. R., The Seven Habits of Highly Effective People: Restoring the Character Ethic, Simon &
Schuster, New York, 1989.
Coyne, J. C. and Downey, G., Social factors and psychopathology: stress, social support, and coping
processes, Ann. Rev. Psychol., 42, 401, 1991.
Dubbert, P. M., Exercise in behavioral medicine, J. Consult. Clin. Psychol., 60, 613, 1992.
Harvey, J. H. and Weary, G., Current issues in attribution theory, Ann. Rev. Psychol., 35, 427, 1984.
Holmes, T. H. and Masuda, M., Life change and illness susceptibility, in Dohrenwend, B. S. and
Dohrenwend, B. P., Eds., Stressful Life Events: Their Nature and Effects, Wiley, New York, 1974.
Janis, I. L., The role of social support in adherence to stressful decisions, Am. Psychol., 38, 143, 1983.
Jones, J. W., Cost evaluation for stress management, EAP Dig. 34, 1984.
Lazarus, R. S., Psychological Stress and the Coping Process, McGraw-Hill, New York, 1966.
Lazarus, R. S., Emotion and Adaptation, Oxford University Press, New York, 1991.
Lazarus, R. S. and Folkman, N., Stress Appraisal and Coping, Springer, New York, 1984.
Lieberman, M. A., The effects of social support on response to stress, in Handbook of Stress
Management, Goldberg, G. and Bresnitz, D. S., Eds., Free Press, New York, 1983.
Maier, S. F. and Seligman, M. E. P., Learned helplessness: theory and evidence, J. Exp. Psychol. Gen.,
105, 3, 1976.
Maslach, C., Burnout: The Cost of Caring, Prentice-Hall, Englewood Cliffs, NJ, 1982.
Miller, D. T. and Ross, M., Self-serving biases in the attribution of causality: fact or fiction?, Psychol.
Bull., 82, 213, 1975.
The New Merriam-Webster Dictionary, Merriam-Webster, Springfield, MA, 1989.
Rhodewalt, F. and Smith, T. W., Current issues in Type A behavior, coronary proneness, and coro-
nary heart disease, in Handbook of Social and Clinical Psychology, Snyder, C. R. and Forsyth, D. R.,
Eds., Pergamon Press, New York, 1991.
Rice, P. L., Stress and Health, 2nd ed., Brooks/Cole Publishing, Pacific Grove, CA, 1992.
Ross, L., The intuitive psychologist and his shortcomings: distortions in the attribution process, in
Advances in Experimental Social Psychology, Vol. 10, Berkowitz, L., Ed., Academic Press, New
York, 1977.
Ross, L. D., Amabile, T. M., and Steinmetz, J. L., Social roles, social control, and biases in social-
perception processes, J. Personal. Soc. Psychol., 35, 485, 1977.
Scheier, M. F. and Carver, C. S., Perspectives on Personality, Allyn & Bacon, Boston, MA, 1988.
Scheier, M. F. and Carver, C. S., Effects of optimism on psychological and physical well-being: theo-
retical overview and empirical update, Cognit. Ther. Res., 16, 201, 1992.
Seligman, M. E. P., Helplessness: On Depression Development and Death, Freeman, San Francisco, CA,
1975.
Seligman, M. E. P., Learned Optimism, Alfred A. Knopf, New York, 1991.
Seligman, M. E. P. and Garber, J., Eds., Human Helplessness: Theory and Application, Academic Press,
New York, 1980.
Selye, H., Stress without Distress, Lippincott, Philadelphia, PA, 1974.
Selye, H., The Stress of Life, 2nd ed., McGraw-Hill, New York, 1976.
Toronto Sun, Reports from insurance/accident forms, Toronto, Ontario, Canada, July 26, 1977.
Yerkes, R. M. and Dodson, J. D., The relation of strength of stimulus to rapidity of habit formation,
J. Comp. Neurol. Psychol., 18, 459, 1908.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 107

section three

Behavior-based psychology
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 108
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 109

chapter seven

Basic principles
To achieve a Total Safety Culture, we need to integrate behavior-based and person-based psychology
and effect large-scale culture changes. The five chapters in Section 3 explain principles and proce-
dures founded on behavioral research which can be applied successfully to change behaviors and atti-
tudes throughout organizations and communities. This chapter describes the primary characteristics
of the behavior-based approach to the prevention and treatment of human problems and shows their
special relevance to occupational safety. The three basic ways we learn are reviewed and related to
the development of safe vs. at-risk behaviors and attitudes.

“One can picture a good life by analyzing one’s feelings, but one can achieve it only by
arranging environmental contingencies.”—B. F. Skinner

Specific safety techniques can be viewed as possible routes to reach a destination, in our
case, a Total Safety Culture. A particular route may be irrelevant or need to be modified sub-
stantially for a given work culture. The key is to begin with a complete and accurate map.
In other words, it is most important to start with an understanding of the basic principles.
If you recall, our overall map or guiding principle is represented by the Safety Triad
(Figure 2.3). Its reference points are the three primary determinants of safety perform-
ance—environment, person, and behavior factors. To achieve a Total Safety Culture, we
need to understand and pay attention to each.
In Section 2, I addressed a number of person-based factors that can contribute to
injuries, including cognitions, perceptions, and attributions. The BASIC ID acronym was
introduced in Chapter 4 to express the complexity of human dynamics and the special chal-
lenges involved in preventing injuries. Behavior was the first dimension discussed, and it
is implicated directly or indirectly in each of the other dimensions. Attitudes, sensations,
imagery, and cognitions—the thinking person side of the Safety Triad (Geller et al., 1989)—
are each influenced by behavior. That is what is meant by the phrase, “Acting people into
changing their thinking.” When we change our behaviors, such as adopting a new strategy
or paradigm, certain person factors change, too.
The reverse is also true. Changes in attitudes, sensations, imagery, and cognitions can
alter behaviors. However, considerable research has shown that it is easier and more cost
effective to “act people into changing their thinking” than the reverse, especially in organ-
izations and community settings (Glenwick and Jason, 1980, 1993; Goldstein and Krasner,
1987; Greene et al., 1987).
In Chapter 1, I justified a behavior-based approach to industrial health and safety by
citing the research review article by Guastello (1993) that evaluated a variety of procedures.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 110

110 Psychology of safety handbook

The two with the greatest impact on injury reduction, behavior-based and ergonomics, use
principles and procedures from behavioral psychology. Actually, Guastello’s review sup-
ports the power of behavior-based problem solving. Over the past 30 years, I have person-
ally witnessed the large-scale effectiveness of this approach to

• Treat agoraphobia (Brehony and Geller, 1981)

• Improve the teaching/learning potential in elementary schools (Geller, 1992c) and
universities (Geller, 1972; Geller and Easley, 1986).
• Manage maximum security prisons efficiently and safely (Geller et al., 1977).
• Improve the impact of a community mental health center (Johnson and Geller, 1980).
• Control litter (Geller, 1980b) and increase community recycling (Geller, 1980a, 1981).
• Reduce excessive use of transportation energy (Reichel and Geller, 1981; Mayer and
Geller, 1982).
• Prevent community crime (Geller et al., 1983; Schnelle et al., 1987).
• Improve sanitation during food preparation (Geller et al., 1980).
• Increase the use of vehicle safety belts in community and industrial settings (Geller,
1984, 1988, 1992a, 1993).
• Reduce alcohol abuse and the risk of alcohol-impaired driving (Geller, 1990; Geller
and Lehman, 1988; Geller et al., 1991).
• Improve the effectiveness of child dental care (Kramer and Geller, 1987).
• Increase the immunization of children in Nigeria (Lehman and Geller, 1987).
• Protect the environment (Geller, 1987, 1992a,b; Geller et al., 1982).
• Increase safe driving practices among pizza deliverers (Ludwig and Geller, 1991,
1997, 2000).
• Increase the use of personal protective equipment (Streff et al., 1993, Williams and
Geller, 2000).
• Improve pedestrian safety throughout a university campus (Boyce and Geller, 2000).
• Increase interpersonal recognition at an industrial site (Roberts and Geller, 1995)
and throughout a university campus (Boyce and Geller, in press).

Given these testimonials, let us examine the fundamental characteristics of the behavior-
based approach.

Primacy of behavior
Whether treating clinical problems (such as drug abuse, sexual dysfunction, depression,
anxiety, pain, hypertension, and child or spouse abuse) or preventing any number of health,
social, or environmental ills (from developing healthy and safe lifestyles to improving edu-
cation and protecting the environment), overt behavior is the focus. Treatment or preven-
tion is based on three basic questions.

1. What behaviors need to be increased or decreased to treat or prevent the problem?

2. What environmental conditions, including interpersonal relationships, are cur-
rently supporting the undesirable behaviors or inhibiting desirable behaviors?
3. What environmental or social conditions can be changed to decrease undesirable
behaviors and increase desirable behaviors?

Thus, behavior change is both the outcome and the means. It is the desired outcome of
treatment or prevention, and the means to solving the identified problem.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 111

Chapter seven: Basic principles 111

Reducing at-risk behaviors

Heinrich’s well-known Law of Safety implicates at-risk behavior as a root cause of most
near hits and injuries (Heinrich et al., 1980). Over the past 20 years, various behavior-based
research studies have verified this aspect of Heinrich’s Law by systematically evaluating
the impact of interventions designed to lower employees’ at-risk behaviors. Feedback from
behavioral observations was a common ingredient in most of the successful intervention
processes, whether the feedback was delivered verbally, graphically by tables and charts,
or through corrective action. See, for example, the comprehensive review by Petersen, 1989,
or individual research articles by Chhokar and Wallin, 1984; Geller et al., 1980; Komaki et
al., 1980; and Sulzer-Azaroff and De Santamaria, 1980.
The behavior-based approach to reducing injuries is depicted in Figure 7.1. At-risk
behaviors are presumed to be a major cause of a series of progressively more serious inci-
dents, from a near hit to a fatality. According to Heinrich’s Law, there are numerous risky
acts for every near hit, and many more near hits than lost-time injuries. This is fortunate
news, but let us not forget that timing or luck is usually the only difference between a near
hit and a serious injury.
Typically, behavior change techniques are applied to specific targets. It is necessary, of
course, that participants know why targeted behaviors are undesirable and have the phys-
ical ability to avoid them. Education and engineering interventions are sometimes needed
to satisfy the physical and knowledge factors of Figure 7.1. The execution factors represent
the motivational aspect of the problem, and usually require the most attention. In other
words, people usually know what at-risk behaviors to avoid and have the ability to do so,
but their motivation might be lacking or misdirected. Behavior change techniques are used
to align individual and group motivation with avoiding the undesired at-risk behavior.

Top 5 are Indices

of Failure Leading
to Reactive Action.
Fatality
Lost
Workday
FAILURES
Recordable Injury

First Aid Case

Near Hit Property

Damage

CAUSE At-Risk Behavior

Behavior Change Process

CONTROL Behavior Change Techniques

Physical Knowledge Execution

Factors Factors Factors

Attitudes

Values

Figure 7.1 Behavior-based safety can decrease at-risk behavior in order to avoid failure.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 112

112 Psychology of safety handbook

Values and attitudes form the foundation of the pyramid in Figure 7.1. These obviously
critical person factors need to support the safety process. Remember our discussion about
risk compensation in Chapter 5, and Wilde’s warning that it is more important to reduce
risk tolerance than increase compliance with specific safety rules (Wilde, 1994)? This hap-
pens when people believe in the safety process and help to make it work. Behavior helps
to make the process work and, if involvement is voluntary and appropriately rewarded, it
will lead to supportive attitudes and values to keep the process going.
Du Pont STOP. One popular behavior-based safety intervention is Du Pont’s STOP (for
Safety Training and Observation Program). Employees are given STOP cards to record the
occurrence of at least one at-risk behavior or work condition each workday, along with their
corrective action. At the end of the day the STOP cards are collected, compiled, and recorded
in a data log. Sometimes the data are transferred to a display chart or graph for feedback.
I have seen Du Pont STOP work well in some plants; in other plants I have noted sub-
stantial resistance. Why? Interviews I have conducted with employees illustrate some
important reminders for rolling out a behavior-based process. In some cases, employees
felt like the program was not theirs, that it was forced on them by top management. I also
talked to employees who did not understand the rationale or underlying principles behind
the program. There was also concern about its negativity. Behavior observation programs
cannot succeed if they are viewed as “gotcha” or “rat-on-your-buddy” campaigns.
Employees will refuse to record the at-risk behaviors of their peers or focus only on envi-
ronmental conditions.
It should be noted that Du Pont has released an “Advanced STOP for Safety Auditing”
program that the company says encourages the recording of safe work practices as well as
unsafe acts. It is indeed important that observations offer positive as well as negative rein-
forcement. Remember in Chapter 3 we discussed the need to shift our orientation regard-
ing safety from failure thinking to an achievement mind-set. People have a more positive
attitude when working to achieve rather than trying to avoid failure. This explains why
employees might criticize and resist an intervention process that targets only failures.
The behavior-based approach illustrated in Figure 7.1 is failure oriented. It is also more
reactive than proactive. The outcome measures are failures—fatalities, lost workdays, and
the like—that require a fix.
The reactive and punitive approach is typical for government agencies. The most
convenient way to control behavior is to pass a law and enforce it. In fact, as depicted in
Figure 7.2, this is the standard government approach to safety improvement. When agents
of the Occupational Safety and Health Administration (OSHA) visit a site for inspection,
they expect to write citations. They look for mistakes or failures, thereby hoping to improve
behavior through negative reinforcement. Unfortunately, this perspective can promote
negative attitudes about the whole process.
It is usually better to focus on increasing safe behaviors. This is being proactive; when
safe behaviors are substituted for at-risk behaviors, injuries will be prevented. By empha-
sizing safe behaviors, employees feel more positive about the process and are more willing
to participate.

Increasing safe behaviors

Figure 7.3 illustrates a positive and proactive behavior-based model. I do not recommend
this instead of the corrective action approach depicted in Figure 7.1. A complete behavior-
based process should target both what is right and wrong about a particular work routine,
but, again, more employees will participate with a positive attitude and remain committed
over time if there is more recognition of achievements than correction of failures.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 113

Chapter seven: Basic principles 113

Figure 7.2 A reactive and punitive approach to safety promotes avoiding failure rather
than achieving success.

The Top 5 are Indices

Injury of Success from a
Pre-
vention
Proactive Approach.

Involvement
Peer Support
ACHIEVEMENTS Records
of Safe Behavior

Corrective
Action

Near Hit Safety

Reporting Share

CAUSE Safe Behavior

Behavior Change Process

CONTROL Behavior Change Techniques

Physical Knowledge Execution

Factors Factors Factors

Attitudes

Values

Figure 7.3 Behavior-based safety can increase safe behavior in order to achieve success.

Monitoring achievement. The indices of achievement in Figure 7.3 are generally

more difficult to record and track than those in Figure 7.1. Actually, the failure outcomes in
Figure 7.1 are observed and recorded quite naturally. Except for near hits and first-aid
cases, the failures in Figure 7.1 have traditionally resulted in systematic investigation and
formal reports. In contrast, the achievements in Figure 7.3 are somewhat difficult to define
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 114

114 Psychology of safety handbook

and record. In fact, it is impossible to obtain an objective record of the number of injuries
prevented. A reasonable estimate of injuries prevented can be calculated, though, after
you achieve a consistent decrease in injuries as a result of a proactive, behavior-based
process.
It is possible to derive direct and objective definitions of the other success indices in
Figure 7.3 and to use these to estimate overall achievement. Involvement, for example,
can be defined by recording participation in voluntary programs, and incidents of correc-
tive action can be counted in a number of situations. You can chart the number of safety
work orders turned in and completed, the number of safety audits conducted and safety
suggestions given, and the number of safety improvements occurring as a result of near-
hit reports.
Throughout Section 3, suggestions for monitoring achievements are offered as I explain
particular intervention strategies for teaching and motivating safe behavior. It is also possi-
ble to use surveys periodically and estimate successes from employee reactions to certain
questions. A distress survey was presented in Chapter 6, for example, and a lower score on
this survey would suggest improvement. In Section 4, I show you how to measure an indi-
vidual’s propensity to “actively care” or go beyond the call of duty for another person’s
safety. Increases in these measures indicate safety success.
Safety share. The “safety share” noted in Figure 7.3 is a simple behavior-focused
process that reflects my emphasis on achievement. At the start of group meetings, the
leader asks participants to report something they have done for safety during the past
week or since the last meeting. Because the “safety share” is used to open all kinds of meet-
ings, safety is given special status and integrated into the overall business agenda. My
experience is that people come to expect queries about their safety accomplishments, and
many go out of their way to have an impressive safety story to share.
This simple awareness booster—“What have you done for safety?”—helps teach an
important lesson. Employees learn that safety is not only loss control, an attempt to avoid
failure, but can be discussed in the same terms of achievement as productivity, quality, and
profits. As a measurement tool, it is possible to count and monitor the number of safety
shares offered per meeting as an estimate of proactive safety success in the work culture.

Direct assessment and evaluation

The roots of behavior-based interventions are in clinical psychology, because the focus on
outward behavior allows for an empirical assessment of therapeutic outcome. Today, this
approach is the leading strategy for program evaluation, in part because of the research
rigor of experimental behavior analysis (Skinner, 1938), and also because the focus is on
behavior rather than the internal subjective concepts of the psychoanalytic, humanistic,
and cognitive approaches to therapy. There is more solid research support for the validity
of behavior-based approaches to solve diverse human problems than for all the other
approaches combined, even though the behavior-based focus is one of the “youngest kids
on the block.”
Baseline measures. Typically, the impact of an intervention is evaluated in three
stages. First, the behavior to be influenced is systematically assessed through direct obser-
vation in naturalistic settings, such as at home, school, or work. This is done by “relevant”
observers such as parents, spouses, teachers, supervisors, or coworkers. Often question-
naires are given to both those being observed and those doing the observing to obtain
opinions, perceptions, and attitudes regarding the targeted problem and relevant environ-
mental factors. Baseline information collected in this stage is used to set intervention goals
and design ways to achieve them.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 115

Chapter seven: Basic principles 115

Monitoring during intervention. The targeted behavior is carefully monitored,

sometimes by those responsible for the behavior, throughout the intervention process.
Desired change often occurs as a result of feedback. The feedback needs to be a frequent
and objective assessment of the target behavior and the circumstances where it occurs too
frequently or not enough. Observing progress in changing a behavior is a powerful reward
for all parties involved—those whose behavior is being changed and those helping facili-
tate the change. Such a reward motivates continuous efforts from all involved. If feedback
data indicate the intervention process is not working as expected, appropriate adjustments
are made. Sometimes entirely different behavior-change techniques will be substituted.
Follow-up measures. In a clinical sense, after the client is presumed “cured” and the
intervention program is withdrawn, follow-up measurement occurs. This is how the long-
term effectiveness of an intervention is assessed. Without an appropriate support system
in the environmental settings where the problem behavior occurred or where a desirable
behavior should occur more often, the client’s problem is likely to resurface. Intermittent
follow-up evaluations check for evidence of this support and indicate whether additional
intervention is needed.

Intervention by managers and peers

The remarkable success of the behavior-based approach to solve people’s problems
changed dramatically the role of the clinical psychologist. Therapists had been spending
most of their days in the office applying psychotherapies to clients. Because behaviors are
triggered by certain environmental circumstances, behavioral improvement requires
changes in those settings. This means the therapist needs to work with clients and poten-
tial support personnel where the problem exists.
Designing and refining an intervention process requires profound understanding of
the problem’s context, including the environment and the people in it. Target behaviors are
observed systematically in the field, and individuals are interviewed close to the problem.
The most cost-effective way to implement on-site intervention is to teach the natural
managers of the setting—parents, teachers, supervisors, prison guards, peers—how to
implement the process. After all, these people deal with the target behavior on an ongoing
basis. So clinical psychologists using behavioral techniques spend significant time in the
field customizing site-specific plans and teaching others how to execute and evaluate a
behavior-change process.

Learning from experience

A key assumption of behavior-based theory is that behavior (desirable and undesirable) is
learned and can be changed by providing people with new learning experiences. Diverse
cultural, social, environmental, and biological factors interact to influence our readiness to
learn behaviors. These factors also support or hinder behaviors once they are learned. We
do not understand exactly how the particulars work—diverse factors interacting to influ-
ence behaviors for each individual. However, basic ways to develop behavioral patterns
have been researched, and it is possible to identify principles behind learning and main-
taining human behavior.
Psychologists define learning as a change in behavior, or potential to behave in a certain
way, resulting from direct and indirect experience. In other words, we learn from observ-
ing and experiencing events and behaviors in our environment (Bandura, 1986). While the
effects of learning are widespread and varied, it is generally believed there are three basic
models: classical conditioning, operant conditioning, and observational learning.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 116

116 Psychology of safety handbook

Laboratory methods have been developed to systematically study each type of learn-
ing. Although it is possible to find pure forms of each in real-world situations, it is likely
we learn simultaneously from different models as they overlap to influence what we do
and how we do it.

Classical conditioning
This form of learning became the subject of careful study in the early 20th century with the
seminal research of Pavlov (1849–1936), a Nobel Prize-winning physiologist from Russia.
Pavlov (1927) did not set out to study learning, rather his research focused on the process
of digestion in dogs. He was interested in how reflex responses were influenced by stimu-
lating a dog’s digestive system with food. During his experimentation, he serendipitously
found that his subjects began to salivate before actually tasting the food. They appeared to
anticipate the food stimulation by salivating when they saw the food or heard the
researchers preparing it. Some dogs even salivated when seeing the empty food pan or the
person who brought in the food. Through experiencing the relationship between certain
stimuli and food, the dogs learned when to anticipate food. Pavlov recognized this as an
important phenomena and shifted the focus of his research to address it.
The top half of Figure 7.4 depicts the sequence of stimulus–response events occurring
in classical conditioning. Actually, before learning occurs, the sequence includes only three
events—conditioned stimulus (CS), unconditioned stimulus (UCS), and unconditioned
response (UCR). The UCS elicits a UCR automatically, as in an autonomic reflex. That is,
the food (UCS) elicited a salivation reflex (UCR) in Pavlov’s dogs. In the same way, the
smell of popcorn (UCS) might make your mouth water (UCR), a puff of air to your eye

Figure 7.4 The stimulus–response relationships in classical and operant conditioning

overlap.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 117

Chapter seven: Basic principles 117

(UCS) would result in an automatic eyeblink (UCR), ingesting certain drugs (like
Antabuse) would elicit a nausea reaction, and a state trooper writing you a speeding ticket
is likely to influence an emotional reaction (like distress, nervousness, or anger).
If a particular stimulus (CS) consistently precedes the UCS on a number of occasions,
the reflex (or involuntary response) will become elicited by the CS. This is classical condi-
tioning and occurred when Pavlov’s “slobbering dogs” salivated when they heard the bell
that preceded food delivery. Classical conditioning would also occur if your mouth
watered when you heard the bell from the microwave oven tell you the popcorn was ready,
if you blinked your eye following the illumination of a dim light that consistently preceded
the air puff, if you felt nauseous after seeing and smelling the alcoholic beverage that pre-
viously accompanied the ingestion of Antabuse, and if you got nervous and upset after see-
ing a flashing blue light in your vehicle’s rearview mirror.

Operant conditioning
As shown in the lower portion of Figure 7.4, the flashing blue light on the police car might
influence you to press your brake pedal and pull over. This is not an automatic reflex
action, but is a voluntary behavior you would perform in order to do what you consider
appropriate at the time. In other words, you have learned (perhaps indirectly from watch-
ing or listening to others or from prior direct experience) to emit certain behaviors when
you see a police car with a flashing blue light in your rearview mirror. Of course, you have
also learned that certain driving behaviors (like pressing the brake pedal) will give a
desired consequence (like slowing down the vehicle), and this will enable another behav-
ior (like pulling over for an anticipated encounter with the police officer).
Selection by consequences. B. F. Skinner (1904–1990), the Harvard professor who
pioneered the behavior-based approach to solving societal problems, studied this type of
learning by systematically observing the behaviors of rats and pigeons in an experimental
chamber referred to as a “Skinner Box” (much to Skinner’s dismay). Skinner termed the
learned behaviors in this situation operants because they were not involuntary and reflex-
ive, as in classical conditioning, but instead they operated on the environment to obtain a
certain consequence. A key principle demonstrated in the operant learning studies is that
voluntary behavior is strengthened (increased) or weakened (decreased) by consequences
(events immediately following behaviors).
The relationship between a response and its consequence is a contingency, and this
relationship explains our motivation for doing most everything we choose to do. Thus, the
hungry rat in the Skinner Box presses the lever to receive food and the vehicle driver
pushes the brake pedal to slow down the vehicle. Indeed, we all select various responses
to perform daily—like eating, walking, reading, working, writing, talking, and recreat-
ing—to receive the immediate consequences they provide us. Sometimes, we emit the
behavior to achieve a pleasant consequence, such as a reward. Other times, we perform a
particular act to avoid an unpleasant consequence—a punisher, and we usually stop per-
forming behaviors that are followed by punishers.
The ABC (activator–behavior–consequence) contingency is illustrated in Figure 7.5.
The dog will move if he expects to receive food after hearing the sound of the can opener.
In other words, the direction provided by an activator is likely to be followed when it is
backed by a consequence that is soon, certain, and significant. This is operant conditioning.
Might the dog in Figure 7.5 salivate when hearing the can opener? If the sound of the
can opener elicits a salivation reflex in the dog, we have an example of classical condition-
ing. In this case, the can-opener sound is a CS (conditioned stimulus) and the salivation is
the CR (conditioned response). What is the UCS (unconditioned stimulus) in this example?
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 118

118 Psychology of safety handbook

Figure 7.5 Activators direct behavior change when backed by a consequence.

Right on, the food which previously followed the sound of the electric can opener is the
UCS, which elicits the UCR (unconditioned response) of salivating without any learning
experience. This UCS –UCR reflex is natural or ”wired in” the organism.
So, an important point illustrated in Figure 7.5 is that operant and classical condition-
ing often occur simultaneously. While we operate on the environment to achieve a positive
consequence or avoid a negative consequence, emotional reactions are often classically
conditioned to specific stimulus events in the situation. We learn to like or dislike the envi-
ronmental context and/or the people involved as a result of the type of ABC contingency
influencing ongoing behavior. This is how attitude can be negatively affected by enforce-
ment techniques.
Emotional reactions. As I discussed earlier in Chapter 3, we feel better when work-
ing for pleasant consequences than when working to avoid or escape unpleasant conse-
quences. This can be explained by considering the classical conditioned emotions that
naturally accompany the agents of reward vs. punishment. How do you feel, for example,
when a police officer flashes a blue light to signal you to pull over? When the instructor
asks to see you after class? When you enter the emergency area of a hospital? When the
dental assistant motions that “you’re next”? When your boss leaves you a phone message
to “see him immediately”? When you see your family at the airport after returning home
from a long trip?
Your reactions to these situations depend, of course, on your past experiences. Police
officers, teachers, doctors, dentists, and supervisors do not typically elicit negative emo-
tional reactions in young children. However, through the association of certain cues with
the consequences we experience in the situation, a negative emotional response or attitude
can develop, and you do not have to experience these relationships directly. In fact, we
undoubtedly learn more from observing and listening to others than from first-hand
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 119

Chapter seven: Basic principles 119

experience. This brings us to the third way in which we learn from experience—observa-
tional learning.

Observational learning
A large body of psychological research indicates that this form of learning is involved to
some degree in almost everything we do (Bandura, 1977, 1986). Whenever you do some-
thing a particular way because you saw someone else do it that way, or because someone
showed you to do it that way, or because characters on television or in a video game did it
that way, you are experiencing observational learning. Whenever you attribute “peer pres-
sure” as the cause of someone taking up an unhealthy habit (like smoking cigarettes or
drinking excessive amounts of alcohol) or practicing an at-risk behavior (like driving at
excessive speeds or adjusting equipment without locking out the energy source), you are
referring to observational learning. When you remind someone to set an example for oth-
ers, you are alluding to the critical influence of observational learning.
Vicarious consequences As children, we learned numerous behavioral patterns by
watching our parents, teachers, and peers. When we saw our siblings or schoolmates
receive rewards like special attention for certain behaviors, we were more likely to copy
that behavior, for instance. This process is termed vicarious, or indirect, reinforcement. At
the same time, when we observed others getting punished for emitting certain behaviors,
we learned to avoid these behaviors. This is referred to as vicarious punishment.
As adults, we teach others by example. As illustrated in Figure 7.6, our children learn
new behavior patterns, including verbal behaviors, by watching us and listening to us. In
this way they learn what is expected of them in various situations. I have talked with many
parents of teenagers who are nervous about their son or daughter getting a driver’s license.
For some, their concern goes beyond the numerous dangers of real-world driving situa-
tions. They realize that several of their own driving behaviors, practiced regularly in front
of their children for years, have not been exemplary.
How can we expect our teenagers to practice safe driving and keep their emotions
under control if we have shown them the opposite throughout their childhood? Of course,
we are not the only role models who influence our children through observational learn-
ing, but we can make a difference.
The prominent role of parental modeling in socializing children is illustrated by this
humorous but true story of a father tucking his six-year-old daughter into bed.

When he came into her bedroom, his daughter requested, “Daddy,

would you tuck me in like you do mommy every night?” He said,
“Sure, honey,” and pulled the covers up and underneath his daugh-
ter’s chin. As he left the room his daughter called after him saying,
“Wait daddy, would you give me a good-night kiss like you do
mommy every night?” Dad said, “Sure honey,” and he kissed his
daughter on her cheek. As he was leaving the room again, his daugh-
ter called after him with one more request, “Daddy . . . Daddy
wait . . . would you whisper in my ear like you do mommy every
night?” “Sure honey,” he replied, and he leaned down and went
“Bzzz, Bzzz, Bzzz,” in his daughter’s ear. Then she said, “Not
tonight, daddy, I have a terrible headache.”

Our actions influence others to a greater extent than we realize. Without our being
aware of our influence, children learn by watching us at home; our coworkers are
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 120

120 Psychology of safety handbook

Figure 7.6 Children learn a lot from their parents through observational learning.

influenced by our practices at work. Not only does the occurrence of safe acts encourage
similar behavior by observers, but verbal behavior can also be influential. If a supervisor is
observed commending a worker for her safe behavior or reprimanding an employee for an
at-risk practice, observers may increase their performance of similar safety behaviors
(through vicarious reinforcement) or decrease the frequency of similar at-risk behavior
(through vicarious punishment). Indeed, to make safe behavior the norm—rather than the
exception—we must always set an example both in our own work practices and in the ver-
bal consequences we offer coworkers following their safe and at-risk behaviors. Figure 7.7
offers a memorable pictorial regarding the influence of example setting on observational
learning.
Observational learning and television. Behaviors by television performers can have
a dramatic impact on viewers’ behaviors. Research has shown, for example, that aggres-
sion can be learned through television viewing (Bandura et al., 1963; Baron and
Richardson, 1994; Synder, 1991). When children and adults were exposed to certain ways
of fighting they had not seen before, they later performed these aggressive behaviors when
frustrated, irritated, or angry. In addition to teaching new forms of aggression, movies and
television programs often convey the message that aggression is an acceptable means of
handling interpersonal conflict. In other words, violence on television gives the impression
that interpersonal aggression is much more common than it really is and, thus, it reduces
our tendency to hold back physical acts of aggression toward others (Berkowitz, 1984).
Given the potential for observational learning from television viewing, our compre-
hensive and systematic observations of violence and unsafe sex on television during the
1994 –1995 and 1997–1998 seasons were disappointing and alarming. In the fall of 1994, my
students coded 297 violent scenes from 152 prime-time episodes over a nine-week period.
The FOX Television Network showed the most violence, with an average of almost three
violent scenes per episode. The most commonly used weapon was the hand or fist (36.2
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 121

Chapter seven: Basic principles 121

Figure 7.7 Intentionally and unintentionally we teach through our example.

percent). A gun was used in 29.6 percent of the violent acts, and for only 21 percent of the
scenes was a negative consequence indicated for initiating the violence. Furthermore, most
of the negative consequences involved the court systems and were therefore delayed.
For 14 percent of the scenes, the instigator of the violence received immediate positive
consequences.
In 242 television episodes coded by England et al. (2000), 219 violent scenes were
observed. As in 1994, the FOX shows displayed, by far, the most violence (58 percent of 73
shows). The percentages of episodes showing violent scenes were lower and similar for
other networks (i.e., 43 percent of 55 shows for CBS, 39 percent of 61 shows for NBC, and
38 percent of 53 shows for ABC).
My students’ coding of sexual behavior on prime-time television in 1994 revealed per-
vasive portrayal of irresponsible sexual behavior. As with violence, the FOX Television
Network showed the most sexual behavior. Of the 81 scenes coded, 82.7 percent showed or
clearly implied sexual intercourse. In only 2 of the 81 scenes was there any discussion of
contraception. A negative consequence for the irresponsible sexual behavior was rarely
shown. A low 7 percent of the characters showed guilt after the sexual act, only 4 percent
appeared to have less respect for themselves, and a mere 2 percent showed less respect for
their partner.
In 1997 –1998, a total of 111 scenes from the 242 prime time episodes coded depicted
sexual intercourse explicitly or implicitly, and most of this sex was at-risk. Specifically,
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 122

122 Psychology of safety handbook

condom use, discussions of sexual history, and communication regarding the potential
negative consequences of unsafe sex were portrayed in only 2.7, 2.8, and 5.5 percent of the
scenes, respectively.
Learning safety from television. Now consider the potential observational learning
in showing television stars using vs. not using vehicle safety belts. When seeing a televi-
sion hero buckle up, some viewers, mostly children, learn how to put on a vehicle safety
belt; others are reminded that they should buckle up on every trip; still others realize that
safety belt use is an acceptable social norm. On the other hand, the frequent nonuse of
safety belts on television teaches the attitude that certain types of individuals, perhaps
macho males and attractive females, do not use safety belts.
As depicted in Figure 7.8, safety-belt use on prime-time television clearly increased
across the first four years, averaging 8 percent of 2094 driving scenes observed in 1984,
15 percent of 1478 driving scenes in 1985, 22 percent of 927 driving scenes observed in 1986,
and 29 percent of 96 driving scenes monitored in 1995.
Our most recent television monitoring during the 1997–1998 season (England et al.,
2001) showed a slight decrease to an average of 26 percent safety belt use across 168 driv-
ing scenes observed on prime-time shows for ABC, CBS, NBC, and FOX. NBC displayed
the highest belt use (41 percent), whereas CBS showed the lowest (14 percent). These lev-
els of safety-belt use on prime-time television were well below the real-world average at
the time.
Figure 7.9 compares the safety-belt use percentages observed on television with
national safety-belt use percentages during the same time period. A steady and steep
increase in the safety-belt use of actual U.S. drivers is shown, but only a slight increase in
the safety-belt use of television characters is evident, with a leveling off from 1995 to 1998.
This is clearly irresponsible broadcasting and calls for social action. What can we do about
such inappropriate observational learning generated by prime-time television shows?
In 1984, my students and I conducted a nationwide campaign to bring public attention
to the nonuse of vehicle safety belts by television performers. We circulated a petition

Figure 7.8 The percentage of driving scenes with a front-seat passenger using a safety-belt
on prime time television is disappointingly low.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 123

Chapter seven: Basic principles 123

80%
Television Characters
70% 67% 69%

Percentage Safety Belt Use

U.S. Residents Nationwide
60%

50%

40% 37%

30% 27% 26%

21% 22%
20% 14% 15%

8%
10%

0%
1984 1985 1986 1994 1998
Observation Years

Figure 7.9 Safety-belt use on prime time television is substantially below the national
average. Adapted from England et al. (2000). With permission.

throughout the United States that described the detrimental observational learning effects
of low safety-belt use on television. We received approximately 50,000 signatures from res-
idents in 36 states. In addition, we distributed a list of 30 names and addresses of television
stars along with instructions to write letters requesting safety-belt use by those who did not
buckle up and to write thank-you notes to those who already buckle up on television.
As part of a creative writing assignment in elementary schools in Olympia, WA, more
than 800 third and fourth graders wrote a buckle-up request to Mr. T, a star on a popular
action program at the time called “The A-Team.” We believe it was no coincidence that
Mr. T increased his use of safety belts from no belt use in 1984 to over 70 percent belt use
in 1985, following the letter-writing campaign (Geller, 1988, 1989). Actually, Mr. T was the
only A-Team member to buckle up during the 1985 season. In 1986 (the last year of this
prime-time show), the entire A-Team was more likely to buckle up (39 percent of all driv-
ing scenes).
With graphs of the low use of safety belts on television from 1984 to 1986, I traveled to
Hollywood and gave a special workshop to producers, writers, and actors on the need to
buckle up on television and in the movies. The workshop was sponsored and marketed by
the National Highway Traffic Safety Administration. The feedback graphs proved to be an
influential means of convincing the large audience of a problem needing their attention.
My point here is that there are a number of things we can do to promote responsible
broadcasting on television. If everyone contributes a “small win,” the benefits can add up
to a big difference. Considering the substantial influence of observational learning on
behaviors and attitudes, and the millions of daily viewers of prime-time television shows,
efforts to depict exemplary behavior among network stars—like safe driving practices—
could potentially prevent millions of injuries and save thousands of lives. Television shows
clearly influence our culture. Thus, to achieve a true societal Total Safety Culture, the
behavior depicted on television needs to be consistent with such a vision.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 124

124 Psychology of safety handbook

The eye's a better teacher and more willing than the ear;
Fine counsel is confusing, but example's always clear;
And the best of all the preachers are the one's who live their
creeds.
For to see the good in action is what everybody needs.

I can soon learn how to do it if you'll let me see it done;

I can watch your hands in action, but your tongue too fast may
run;
And the lectures you deliver may be very wise and true.
But I'd rather get my lesson by watching what you do.

For I may not understand you and the high advice you give.
There's no misunderstanding how you act and how you live.

Figure 7.10 This poem, written by Forrest H. Kirkpatrick, illustrates the power of obser-
vational learning. With permission.

Setting examples. The poem “Setting Examples” by Forrest H. Kirkpatrick says it all.
This poem is presented in Figure 7.10, and I recommend copying it and posting it for oth-
ers to read. It is so easy to forget the dramatic influence we have on others by our own
behaviors. Obviously, we need to take the slogan “walk the talk” very seriously. In fact, if
we are not convinced a particular safeguard or protective behavior is necessary for us (“It’s
not going to happen to me”), we need to realize, at least, that our at-risk behaviors can
endanger others.
For example, I never get in my vehicle believing a crash will happen to me, so my
rationale for buckling my combination lap and shoulder belt is to set the right example for
others, whether they are in the car with me or not. Understanding the powerful influence
of observational learning, we should feel obligated to set the safe example whenever some-
one could see us.

Overlapping types of learning

Laboratory methodologies have been able to study each type of learning separately, but the
real world rarely offers such purity. In life, the usual situation includes simultaneous influ-
ence from more than one learning type. The operant learning situation, for example, is
likely to include some classical (emotional) conditioning. As I indicated earlier, this is one
reason rewarding consequences should be used more frequently than punishing conse-
quences to motivate behavior change.
Remember, a rewarding situation (unconditioned stimulus) can elicit a positive emo-
tional experience (unconditioned response), and a punitive situation (UCS) can elicit a neg-
ative emotional reaction (UCR). With sufficient pairing of rewarding or punishing
consequences with environmental cues (such as a work setting or particular people), the
environmental setting (conditioned stimulus) can elicit a positive or negative emotional
reaction or attitude (conditioned response). This can in turn facilitate (if it is positive) or
inhibit (if it is negative) ongoing performance.
Figure 7.11 depicts a situation in which all three learning types occur at the same time.
As discussed earlier and diagrammed in Figure 7.4, the blue flashing light of the police car
signals drivers to press the brake pedal of their car and pull over. In this case, the blue light
is considered a discriminative stimulus because it tells people when to respond in order to
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 125

Chapter seven: Basic principles 125

Figure 7.11 Three types of learning occur in some situations.

receive or avoid a consequence. Actually, drivers would apply their brakes to avoid puni-
tive consequences, so this situation illustrates an avoidance contingency where drivers
respond to avoid failure.
The flashing blue light might also serve as a conditioned stimulus eliciting a negative
emotional reaction. This is an example of classical conditioning occurring simultaneously
with operant learning. Our negative emotional reaction to the blue light might have been
strengthened by prior observational learning. As a child, we might have seen one of our
parents pulled over by a state trooper and subsequently observed a negative emotional
reaction from our parent. The children in Figure 7.11 are not showing the same emotional
reaction of the driver. Eventually, they will probably do so as a result of observational
learning. Later, their direct experience as drivers will strengthen this negative emotional
response to a flashing blue light on a police car.

In conclusion
In this chapter, I reviewed the basic principles underlying a behavior-based approach to the
prevention and treatment of human problems. The variety of successful applications of this
approach was discussed, based on my personal experiences. The behavior-based princi-
ples—the primacy of behavior, direct assessment and evaluation, intervention by man-
agers and peers, and three types of learning—were explained with particular reference to
reducing personal injury.
Because at-risk behaviors contribute to most if not all injuries, a Total Safety Culture
requires a decrease in at-risk behaviors. Organizations have attempted to do this by
targeting at-risk acts, exclusive of safe acts, and using corrective feedback, reprimands, or
disciplinary action to motivate behavior change. This approach is useful, but less pro-
active and less apt to be widely accepted than a behavior-based approach that emphasizes
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 126

126 Psychology of safety handbook

Figure 7.12 Negative consequences can have an undesirable impact on attitude.

recognition of safe behaviors. It will be easier to get employees involved in safety achieve-
ment if credit is given for doing the right thing more often than reprimands for doing
wrong. Excessive use of negative consequences can lead to the feelings or attitude
expressed by the wolf in Figure 7.12.
The three types of learning are relevant for understanding safety-related behaviors and
attitudes. Most of our safe and at-risk behaviors are learned operant behaviors, performed
in particular settings to gain positive consequences or to avoid negative consequences.
Classical conditioning often occurs at the same time to link positive or negative emotional
reactions with the stimulus cues surrounding the experience of receiving consequences.
These cues include the people who deliver the rewards or punishment. We often learn what
to do and what not to do by watching others receive recognition or correction for their
operant behaviors. This is observational learning, an ongoing process that should motivate
us to try to set the safe example at all times.

References
Bandura, A., Social Learning Theory, Prentice-Hall, Englewood Cliffs, NJ, 1977.
Bandura, A., Social Foundations of Thought and Action: A Social Cognitive Theory, Prentice-Hall,
Englewood Cliffs, NJ, 1986.
Bandura, A., Ross, D., and Ross, S. A., Imitation of film-mediated aggressive models, J. Abnorm. Soc.
Psychol., 66, 3, 1963.
Baron, R. A. and Richardson, D., Human Aggression, 2nd ed., Plenum, New York, 1994.
Berkowitz, L., Some effects of thoughts on anti- and pro-social influences of media events: a
cognitive-neoassociation analysis, Psychol. Bull., 95, 410, 1984.
Boyce, T. E. and Geller, E. S., A community-wide intervention to improve pedestrian safety:
guidelines for institutionalizing large-scale behavior change, Environ. Behav., 32, 502–520.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 127

Chapter seven: Basic principles 127

Boyce, T. E. and Geller, E. S., Attempts to increase prosocial behavior: a comparison of reinforce-
ment and intrinsic motivation theory, Environ. Behav., in press.
Brehony, K. A. and Geller, E. S., Agoraphobia: a behavioral perspective and critical appraisal of
research, in Hersen, M., Eisler, R. M., and Miller, P. M., Eds., New York, Progress in Behavior
Modification, Vol. 8, Academic Press, 1981.
Chhokar, J. S. and Wallin, J. A., A field study of the effects of feedback on frequency on perform-
ance, J. Appl. Psychol., 69, 524, 1984.
England, K. J., Porter, B. E., Geller, E. S., and DePasquale, J. P., Is television a health and safety haz-
ard? A longitudinal analysis of at-risk behavior on prime-time television, under editorial
review, 2001.
Geller, E. S., A training program in behavior modification: design, outcome, and implication, JSAS
Cat. Select. Doc. Psychol., 2, 29, 1972.
Geller, E. S., Applications of behavior analysis to litter control, in Behavioral Community Psychology:
Progress and Prospects, Glenwick, D. and Jason, L., Eds., Prager Press, New York, 1980.
Geller, E. S., Saving environmental resources through waste reduction and recycling: how the
behavioral community psychologist can help, in Helping in the Community: Behavior Applications,
Martin, G. L., and Osborne, J. G., Eds., Plenum, New York, 1980b.
Geller, E. S., Waste reduction and resource recovery: strategies for energy conservation, in Advances
in Environmental Psychology, Vol. III, Baum, A. and Singer, J., Eds., Lawrence Erlbaum
Associates, New Jersey, 1981.
Geller, E. S., Motivating safety belt use with incentives: a critical review of the past and a look to the
future, in Advances in Belt Restraint Systems: Design, Performance, and Usage, no. 141, Society of
Automotive Engineers, Inc., Warrendale, PA, 1984.
Geller, E. S., Environmental psychology and applied behavior analysis: from strange bedfellows to
a productive marriage, in Handbook of Environmental Psychology, Vol. I, Stokols, D. and Altman,
I., Eds., John Wiley & Sons, New York, 1987.
Geller, E. S., A behavioral science approach to transportation safety, Bull. NY Acad. Med., 64, 632,
1988.
Geller, E. S., Using television to promote safety belt use., in Public Communication Campaigns, 2nd
ed., Rice, R. E. and Atkin, C. K., Eds., SAGE Publications, Newberry Park, CA, 1989.
Geller, E. S., Preventing injuries and deaths from vehicle crashes: encouraging belts and discourag-
ing booze, in Social Influence Processes and Prevention, Edwards, J., Tindale, R. S., Heath, L., and
Posavac, E. J., Eds., Plenum, New York, 1990.
Geller, E. S., Applications of behavior analysis to prevent injury from vehicle crashes, monograph
published by the Cambridge Center for Behavioral Studies, Cambridge, MA, 1992a.
Geller, E. S., Solving environmental problems: a behavior change perspective, in In Our Hands:
Psychology, Peace, and Social Responsibility, Staub, S. and Green, P., Eds., New York University
Press, 1992b.
Geller, E. S., Ed., The educational crisis: issues, perspectives, solutions, monograph No. 7, Society
for the Experimental Analysis of Behavior, Inc., Lawrence, KS, 1992c.
Geller, E. S., Applications of behavioral science for road safety, in Promoting Health and Mental Health:
Behavioral Approaches to Prevention, Glenwick, D. and Jason, L., Eds., Springer, New York, 1993.
Geller, E. S. and Easley, A. T., Applied behavior analysis in the college classroom: Some ideas for
educators, in The Art of Teaching: Seven Perspectives, Bishop, L., Ed., The Academy of Teaching
Excellence, Virginia Polytechnic Institute and State University, Blacksburg, VA, 1986.
Geller, E. S. and Lehman, G. R., Drinking-driving intervention strategies: a person–situation-
behavior framework, in Snortum, J. R., Zimring, F. E., and Laurence, M. D., Eds., Social Control
of the Drinking Driver, The University of Chicago Press, Chicago and London, 1988.
Geller, E. S., Koltuniak, T. A., and Shilling, J. S., Response avoidance prompting: a cost-effective
strategy for theft deterrence, Behav. Counsel. Comm. Intervent., 3, 29, 1983.
Geller, E. S., Lehman, G. R., and Kalsher, M. J., Behavior Analysis Training for Occupational Safety,
Make-A-Difference, Inc., Newport, VA, 1989.
Geller, E. S., Winett, R. A., and Everett, P. B., Preserving the Environment: New Strategies for Behavior
Change. Pergamon Press, New York, 1982.
820049_CRC1_L1540_CH07 11/10/00 1:37 PM Page 128

128 Psychology of safety handbook

Geller, E. S., Eason, S., Phillips, J., and Pierson, M., Interventions to improve sanitation during food
preparation, J. Organ. Beh. Manage., 2, 229, 1980.
Geller, E. S., Johnson, D. F., Hamlin, P. H., and Kennedy, T. D., Behavior modification in a prison:
issues, problems, and compromises, Crim. Just. Beh., 4, 11, 1977.
Geller, E. S., Sleet, D., Elder, J., and Hovell, M., Behavior change approaches to deterring alcohol-
impaired driving, in Advances in Health Education and Promotion, Vol.III, Ward, W. B. and Lewis,
F. M., Eds., Jessica Kingsley Publishers, Philadelphia, PA, 1991.
Glenwick, D. and Jason, L., Eds., Behavioral Community Psychology, Praeger Publishers, New York,
1980.
Glenwick, D. and Jason, L., Eds., Promoting Health and Mental Health: Behavioral Approaches to
Prevention, Springer, New York, 1993.
Goldstein, A. P. and Krasner, L., Modern Applied Psychology, Pergamon Press, New York, 1987.
Greene, B. F., Winett, R. A., Van Houten, R., Geller, E. S., and Iwata, B. A., Eds., Behavior Analysis in
the Community: Readings from the Journal of Applied Behavior Analysis, University of Kansas,
Lawrence, KS, 1987.
Guastello, S. J., Do we really know how well our occupational accident prevention programs
work?, Saf. Sci., 16, 445, 1993.
Heinrich, H. W., Petersen, D., and Roos, N., Industrial Accident Prevention: A Safety Management
Approach, 5th ed., McGraw-Hill, New York, 1980.
Johnson, R. P. and Geller, E. S., Community mental health center programs, in Behavioral Community
Psychology: Progress and Prospects, Glenwick, D. and Jason, L., Eds., Praeger Press, New York, 1980.
Komaki, J., Heinzmann, A. T., and Lawson, L. Effect of training and feedback: component analysis
of a behavioral safety program, J. Appl. Psychol., 65(3), 261, 1980.
Kramer, K. D. and Geller, E. S., Community dental health promotion for children: integrating
applied behavior analysis and public health, Educ. Treat. Child., 10, 58, 1987.
Lehman, G.R. and Geller, E. S., Educational Technology for Health Workers in Nigeria, Academy for
Educational Development, Washington, D.C., 1987.
Ludwig, T. D. and Geller, E. S., Improving the driving practices of pizza deliverers: response
generalization and moderating effects of driving history, J. Appl. Behav. Anal., 24, 31, 1991.
Ludwig, T. D. and Geller, E. S., Managing injury control among professional pizza deliverers:
effects of goal setting and response generalization, J. Appl. Psychol., 82, 253, 1997.
Ludwig, T. D. and Geller, E. S., An organizational behavior management approach to safe driving
intervention for pizza deliverers, J. Organ. Behav. Manage., 19(4) monograph, 2000.
Mayer, J. and Geller, E. S., Motivating energy efficient travel: a cost-effective incentive strategy for
encouraging usage of a community bike path, J. Environ. Syst., 12, 99, 1982.
Pavlov, I. P., Conditional Reflexes. Anrep, G. V., Ed. and Transl., Oxford University Press, London, 1927.
Petersen, D., Safe Behavior Reinforcement, Aloray, Inc., New York, 1989.
Reichel, D. A. and Geller, E. S., Applications of behavioral analysis to conserve transportation
energy, in Advances in Environmental Psychology, Vol. III. Baum, A., and Singer, J., Eds.,
Lawrence Erlbaum Associates, New Jersey, 1981.
Roberts, D. S. and Geller, E. S., An “actively caring” model for occupational safety: a field test,
Appl. Prevent. Psychol., 4, 53, 1995.
Schnelle, J. F., Geller, E. S., and Davis, M. A., Law enforcement and crime prevention, in Behavioral
Approaches to Crime and Delinquency: Application, Research, and Theory, Morris, E. K., and
Braukmann, C. J., Eds., Plenum Press, New York, 1987.
Skinner, B. F., The Behavior of Organisms, Appleton-Century-Crofts, New York, 1938.
Snyder, S., Movies and juvenile delinquency: an overview, Adolescence, 26, 121, 1991.
Streff, F. M., Kalsher, M. J., and Geller, E. S., Developing efficient workplace safety programs:
observations of response covariation, J. Organ. Behav. Manage., 13(2), 3, 1993.
Sulzer-Azaroff, B. and De Santamaria, M. C., Industrial safety hazard reduction through perform-
ance feedback, J. Appl. Behav. Anal., 13, 287, 1980.
Wilde, G. J. S., Target Risk, PDE Publications, Toronto, Ontario, Canada, 1994.
Williams, J. H. and Geller, E. S., Behavior-based intervention for occupational safety: critical impact
of social comparison feedback, J. Saf. Res., 31(30), 135, 2000.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 129

chapter eight

Identifying critical behaviors

The practical “how to” aspects of this book begin with this chapter. The overall process is called DO
IT, each letter representing the four basic components of a behavior-based approach: Define target
behaviors to influence; Observe these behaviors; Intervene to increase or decrease target behaviors;
and Test the impact of your intervention process. This chapter focuses on developing a critical behav-
ior checklist for objective observing, intervening, and testing.

“As I grow older, I pay less attention to what men say, I just watch what they do.”
—Andrew Carnegie

Now the action begins. Up to this point, I have been laying the groundwork (rationale and
theory) for the intervention strategies described here and in the next three chapters. From
this information, you will learn how to develop action plans to increase safe behaviors,
decrease at-risk behaviors, and achieve a Total Safety Culture.
Why did it take me so long to get here—to the implementation stage? Indeed, if you
are looking for “quick-fix” tools to make a difference in safety you may have skipped or
skimmed the first two parts of this text and started your careful reading here. I certainly
appreciate that the pressures to get to the bottom line quickly are tremendous, but, remem-
ber, there is no quick fix for safety. The behavior-based approach that is the heart of this
book is the most efficient and effective route to achieving a Total Safety Culture. It is a never
ending continuous improvement process, one that requires ongoing and comprehensive
involvement from the people protected by the process. In industry, these are the operators
or line workers.
Long-term employee participation requires understanding and belief in the principles
behind the process. Employees must also perceive that they “own” the procedures that
make the process work. For this to happen it is necessary to teach the principles and ration-
ale first (as done in this Handbook), and then work with participants to develop specific
process procedures. This creates the perception of ownership and leads to long-term
involvement.
When people are educated about the principles and rationale behind a safety process,
they can customize specific procedures for their particular work areas. Then the relevance
of the training process is obvious, and participation is enhanced. People are more likely to
accept and follow procedures they helped to develop. They see such safe operating proce-
dures as “the best way to do it” rather than “a policy we must obey because management
says so.”
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 130

130 Psychology of safety handbook

Figure 8.1 Employee involvement is limited when a program is force fitted into a work
culture.

Obviously, the kind of safety consultant depicted in Figure 8.1 stifles employee
ownership and involvement. Yet, so many safety efforts start as off-the-shelf programs.
A videotape is shown and ready-made workbooks are followed to train step-by-step
procedures. Much more involvement occurs when consultants begin a new safety effort
by teaching rationale and principles and then guiding participants through the develop-
ment of specific procedures. Subsequently, people want to be trained on the implementa-
tion procedures.
When effective leaders or consultants guide the customization of a process, they state
expectations but they do not give mandates or directions. They show both confidence and
uncertainty (Geller, 2000; Langer, 1989, 1997). In other words, they are confident a set of
procedures will be developed but do not know the best way to do it. This allows employ-
ees room to be alert, innovative, and self-motivated. The result is that ownership and inter-
personal trust increase, which in turn leads to more involvement.
As we begin here to define principles and guidelines for action plans, it is important to
keep one thing in mind—you need to start with the conviction that there is rarely a generic
best way to implement a process involving human interaction. For a behavior-based safety
process to succeed in your setting, you will need to work out the procedural details
with the people whose involvement is necessary. The process needs to be customized to fit
your culture.

The DO IT process
For well over a decade, I have taught applications of the behavior-based approach to indus-
trial safety with the acronym depicted in Figure 8.2. The process is continuous and involves
the following four steps.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 131

Chapter eight: Identifying critical behaviors 131

Figure 8.2 Behavior-based safety is a continuous four-step process.

D: Define the critical target behavior(s) to increase or decrease.

O: Observe the target behavior(s) during a pre-intervention baseline period to set
behavior-change goals and, perhaps, to understand natural environmental or social
factors influencing the target behavior(s).
I: Intervene to change the target behavior(s) in desired directions.
T: Test the impact of the intervention procedure by continuing to observe and record
the target behavior(s) during the intervention program.

From this data obtained in the Test phase, you can evaluate the impact of your inter-
vention and make an informed decision whether to continue it, implement another strat-
egy, or define another behavior to target for the DO IT process.
This chapter focuses on the first two steps, defining and observing target behaviors.
Before we get into those specifics, however, I want to briefly outline the DO IT process to
make an important point: What I am explaining to you is all easier said than done.
Remember, there are no true quick fixes for safety.
To begin, just what are clear and concise definitions of target behaviors? This is the first
step in the DO IT process. There is so much to choose from: using equipment safely, lifting
correctly, locking out power appropriately, and looking out for the safety of others, to name
just a scant few. The outcome of behaviors, such as wearing personal protective equipment,
working in a clean and organized environment, and using a vehicle safety belt can also be
targeted.
If two or more people independently obtain the same frequency recordings when
observing the defined behavior or behavioral outcome during the same time period, you
have a definition sufficient for an effective DO IT process. Baseline observations of the
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 132

132 Psychology of safety handbook

target behavior should be made and recorded before implementing an intervention pro-
gram. More details on this aspect of the process are given later in this chapter.
What about the intervention step? This phase of DO IT involves one or more behavior-
change techniques, based on the simple ABC model depicted in Figure 8.3. As I discussed
in the preceding chapter, activators direct behavior and consequences motivate behavior.
For example, a ringing telephone or doorbell activates the need for certain behaviors from
residents, but residents answer or do not answer the telephone or door depending on cur-
rent motives or expectations developed from prior experiences. The activators listed in
Figure 8.3 are discussed in Chapter 10. Consequences are discussed in Chapter 11.
The DO IT process is based on operant learning, which we discussed in Chapter 7.
Figure 7.11, for example, shows the flashing blue light on the police car as a discriminative
stimulus (or activator) that signals (or directs) the motorist to perform certain driving
behaviors learned from past experience. We will respond to this activator if it is supported
by a consequence. For example, if we believe the police officer will give us a ticket if we do
not stop, we will pull over. Remember, the power of an activator to influence behavior is
determined by the type of consequence(s) it signals (Skinner, 1953, 1969).
Let us talk a little more about consequences. The strongest consequences are soon, siz-
able, and certain. In other words, we work diligently for immediate, probable, and large
positive reinforcers or rewards, and we work frantically to escape or avoid soon, certain,
and sizable negative reinforcers or punishers. This helps explain why safety is a struggle in
many workplaces. You see, safe behaviors are usually not reinforced by soon, sizable, and
certain consequences. In fact, safe behaviors are often punished by soon and certain

Figure 8.3 The ABC model is used to develop behavior change interventions.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 133

Chapter eight: Identifying critical behaviors 133

Figure 8.4 Compliance with some activators is supported by natural consequences.

negative consequences, including inconvenience, discomfort, and slower goal attainment.

Also, the consequences that motivate safety professionals to promote safe work practices—
reduced injuries and associated costs—are delayed, negative, and uncertain (actually
improbable) from an individual perspective.
Several illustrations make these points. The man in Figure 8.4 will surely comply with
the “Don’t Walk” activator and stay on the curb until the light changes. In this case, nega-
tive consequences from stepping off the curb at the wrong time would occur very soon.
Given the observable traffic flow, a negative consequence is almost certain to happen, and
the consequence could be quite sizable, perhaps fatal. However, most safety situations in
the workplace are not like this one.
Take, for example, the man in Figure 8.5. He is complying with the activator, but there
are no obvious (soon, certain, and sizable) consequences supporting this safe behavior. He
might not see a reason for the hard hat, and without supportive consequences from peers
his compliance could be temporary.
Research backs up these examples. When subjects in operant learning experiments no
longer receive a consequence for making the desired response, they eventually stop per-
forming the behavior. This is referred to as extinction, and it occurs in classical condition-
ing, too. When Pavlov stopped giving his dogs food (the unconditioned stimulus)
following the bell (the conditioned stimulus), the dogs eventually stopped salivating (con-
ditioned response) to the bell.
Competing with the lack of soon, certain, and sizable positive consequences for safe
behaviors are soon, certain (and sometimes sizable) positive consequences for at-risk
behaviors. Taking risks avoids the discomfort and inconvenience of most safe behaviors,
and it often allows people to achieve their production and quality goals faster and easier.
Supervisors sometimes activate and reward at-risk behaviors, unintentionally, of course, to
achieve more production. Because activators and consequences are naturally available
throughout our everyday existence to support at-risk behaviors in lieu of safe behaviors,
safety can be considered a continuous fight with human nature (as discussed earlier in
Section 2).
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 134

134 Psychology of safety handbook

Figure 8.5 Compliance with some activators is not supported by natural consequences.

Check out the two lawn-mower operators in Figure 8.6. Which one is having more fun?
Who is more uncomfortable? Who is safe? Chances are both men will complete mowing
their lawns without an injury. So which worker will have enjoyed the task more? Again,
this defines the fight against human nature. Safety typically means more discomfort,
inconvenience, and less fun than the more efficient at-risk alternative.
The DO IT process is a tool to use in this struggle with human nature. Developing and
maintaining safe work practices often requires intervention strategies to keep people
safe—strategies involving activators, consequences, or both, but we are getting ahead of
ourselves. First, we need to define critical behaviors to establish targets for our interven-
tion. Let us see how this is done.

Defining target behaviors

The DO IT process begins by defining critical behaviors to work on. These become the tar-
gets of our intervention strategies. Some target behaviors might be safe behaviors you want
to see happen more often, like lifting with knees bent, cleaning a work area, putting on per-
sonal protective equipment, or replacing safety guards on machinery. Other target behav-
iors may be at-risk behaviors that need to be decreased in frequency, such as misusing a
tool, overriding a safety switch, placing obstacles in an area designated for traffic flow,
stacking materials incorrectly, and so on.
A DO IT process can define desirable behaviors to be encouraged or undesirable
behaviors to be changed. What the process focuses on in your workplace depends on a
review of your safety records, job hazard analyses, near-hit reports, audit findings, inter-
views with employees, and other useful information.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 135

Chapter eight: Identifying critical behaviors 135

Figure 8.6 Compared to at-risk behavior, safe behavior is often uncomfortable, inconven-
ient, and less fun.

Critical behaviors to identify and target are

• At-risk behaviors that have led to a substantial number of near hits or injuries in the
past and safe behaviors that could have prevented these incidents.
• At-risk behaviors that could potentially contribute to an injury (or fatality) and safe
behaviors that could prevent such an incident.

Deciding which behaviors are critical is the first step of a DO IT process. A great deal
can be discovered by examining the workplace and discussing with people how they have
been performing their jobs. People already know a lot about the hazards of their work and
the safe behaviors needed to avoid injury. They even know which safety policies are some-
times ignored to get the job done on time. They often know when a near hit had occurred
because an at-risk behavior or environmental hazard had been overlooked. They also know
which at-risk behaviors could lead to a serious injury (or fatality) and which safe behaviors
could prevent a serious injury (or fatality).
In addition to employee discussions, injury records and near-hit reports can be con-
sulted to discover critical behaviors (both safe and at risk). Job hazard analyses or standard
operating procedures can also provide information relevant to selecting critical behaviors
to target in a DO IT process. Obviously, the plant safety director or the person responsible
for maintaining records for OSHA or MSHA (Mine Safety and Health Administration) can
provide valuable assistance in selecting critical behaviors.
After selecting target behaviors, it is critical to define them in a way that gets everyone
on the same page. All participants in the process need to understand exactly what behav-
iors you intend to support, increase, or decrease. Defining target behaviors results in an
objective standard for evaluating an intervention process.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 136

136 Psychology of safety handbook

There is another point to be made about the DO IT process. It involves translating

educational concepts into operational (behavioral) terms. From education, one learns
general principles, procedures, or policy. Training is the process of translating knowledge
from education into specific behaviors. Thus, when people describe critical behaviors
in objective and observable terms, they are transferring knowledge into meaningful
action plans. If done correctly, this can reduce biased subjectivity in various interpersonal
matters.

What is behavior?
The key is to define behaviors correctly. Let us begin by stepping back a minute to consider:
What is behavior?
Behavior refers to acts or actions by individuals that can be observed by others. In other
words, behavior is what a person does or says as opposed to what he or she thinks, feels, or
believes.
Yes, the act of saying words such as “I am tired,” is a behavior because it can be
observed or heard by others. However, this is not an observation of tired behavior. If the
person’s work activity slows down or amount of time on the job decreases, we might infer
that the person is actually tired. On the other hand, a behavioral “slow down” could result
from other internal causes, like worker apathy or lack of interest. The important point here
is that feelings, attitudes, or motives should not be confused with behavior. They are inter-
nal aspects of the person that cannot be directly observed by others. It is risky to infer inner
person characteristics from external behaviors.
The test of a good behavioral definition is whether other persons using the definition
can accurately observe if the target behavior is occurring. There are thousands of words in
the English language that can be used to describe a person. From all these possibilities, the
words used to describe behavior should be chosen for clarity to avoid being misinterpreted;
precision to fit the specific behavior observed; brevity to keep it simple; and their reference to
observable activity—they describe what was said or done. As shown in Figure 8.7, how-
ever, without a clear and precise definition, most action words can have more than one
interpretation.

Outcomes of behavior
Often it is easier to define and observe the outcomes of safe or at-risk behavior rather than
the behavior itself. These outcomes can be temporary or permanent, but they are always
observed after the behavior has occurred. For example, when observing a worker wearing
safety glasses, a hard hat, or a vehicle safety belt, you are not actually observing a behav-
ior, but rather you are observing the outcome of a pattern of safety behaviors (the behav-
iors required to put on the personal protective equipment). Likewise, a locked out machine
and a messy work area are both outcomes of behavior; one from safe behavior and one
from at-risk behavior.
This distinction between direct observations of behavior vs. behavioral outcomes is
important. You see, evaluating an outcome cannot always be directly attributed to a single
behavior or to any one individual, and the intervention to improve a behavioral outcome
might be different than an intervention to improve behaviors observed directly. For exam-
ple, direct guidance through instruction and demonstration (activators) might be the inter-
vention of choice to teach the correct use of a respirator; verbal recognition (a consequence)
would be more suitable to support the outcome of correctly wearing a respirator at the
appropriate time and place.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 137

Chapter eight: Identifying critical behaviors 137

Figure 8.7 Selective word definitions cause communication confusion.

Person–action–situation
Three elements comprise a complete description of behavior. The first is the person who is
behaving. Second, what the person says or does constitutes an action and last, but no less
important, is the situation in which the person acts (when or where). The importance of
safety-related behaviors depends on the situation in which the behavior occurs. Thus, tar-
get behaviors are often defined by environmental context. For example, safety glasses and
ear plugs are not needed in the personnel office but are needed in other work areas.
Here is another important point. You cannot study a behavior that does not happen.
The nonoccurrence of a recommended safety action in a given situation is often defined as
at-risk behavior. Although the lack of an appropriate safe action might put a person at risk,
the absence of a behavior is not a behavior.
For example, if a person is not wearing the required PPE, he or she is at risk. However,
this lack of doing something to protect oneself is not a behavior that can be observed in a
DO IT process. It is important to define what is happening—the at-risk behavior(s) occur-
ring in place of desired safe behavior. Such behavior can be observed and changed. Then,
it is possible to study the factors influencing this at-risk behavior and perhaps inhibiting
safe behavior.

Describing behaviors
A target behavior needs to be defined in observable terms so multiple observers can inde-
pendently watch one individual and obtain the same results regarding the occurrence or
nonoccurrence of the target behavior. There should be no room for interpretation. “Is not
paying attention,” “acting careless,” or “lifting safely,” for example, are not adequate
descriptions of behavior, because observers would not agree consistently about whether
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 138

138 Psychology of safety handbook

the behavior occurred. In contrast, descriptions like “keeping hand on handrail,” “moving
knife away from body when cutting,” and “using knees while lifting” are objective and
specific enough to obtain reliable information from trained observers. In other words, if
two observers watched for the occurrence of these behaviors, they would likely agree
whether or not the behavior occurred.

Interobserver reliability
The ultimate test for a behavioral description is to have two observers watch independ-
ently for the occurrence of the target behavior on a number of occasions, and then calculate
the percentage of agreement between observers. More specifically, agreement occurs
whenever the two observers report seeing or not seeing the target behavior at the same
time. Disagreement occurs whenever one person reports seeing the behavior when the
other person reports not seeing the behavior. Percentage of agreement is calculated by
adding the number of agreements and disagreements and dividing the total into the num-
ber of agreements. The quotient is then multiplied by 100 to give percentage of agreement.
If the result is 80 percent or higher, the behavioral definition is adequate and the observers
have been adequately trained to use the definition in a DO IT process (Kazdin, 1994).

Multiple behaviors
Let us look more closely at types of behaviors. Some workplace activities can be treated
effectively as a single behavior. Examples include “Looking left–right –left before crossing
the road,” “Walking within the yellow safety lines,” “Honking the fork lift horn at the inter-
section,” “Returning tools to their proper location,” “Bending knees while lifting,” and
“Keeping a hand on the handrail while climbing stairs.”
Some outcomes of behaviors also can be dealt with in singular terms, like “using ear
plugs,” “using a vehicle safety belt,” “climbing a ladder that is properly tied off,” “work-
ing on a scaffold with appropriate fall protection,” and “repairing equipment that had been
locked out correctly.” With a proper definition, an observer could readily count occurrences
of these safe behaviors (or outcomes) during a systematic audit.
Many safety activities are made up of more than one discrete behavior, however, and
it may be important to treat these behaviors independently in a definition and an audit.
“Bending knees while lifting,” for example, is only one aspect of a safe lift. Thus, if safe lift-
ing were the activity targeted in a DO IT process, it would be necessary to define the sepa-
rate behaviors (or procedural steps) of a safe lift. This would include, at least, checking the
load before lifting, asking for help in certain situations, lifting with the legs, holding the
load close to one’s body, lifting in a smooth motion, and moving feet when rotating (or not
twisting).
Each of the procedural steps in safe lifting requires a clear objective definition so two
observers could determine reliably whether the behavior in a lifting sequence had
occurred. Observing reliably whether the load was held close and knees were bent would
be relatively easy. However, defining a “smooth lift” so that observers could agree on 80
percent or more of the observations would be more difficult and, for observers to reliably
audit “asking for help,” the “certain situations” calling for this response would need to be
specified.
Role playing demonstrations are an important way to help define the behavioral steps
of a procedure. A volunteer acts out the behaviors while observers attempt to determine
whether each of the designated steps of the activity was “safe” or “at risk.” Suppose,
for example, your work group was interested in improving stair safety and decided the
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 139

Chapter eight: Identifying critical behaviors 139

safety-related behaviors in this activity include: keeping one hand on the handrail, taking
one step at a time, and walking rather than running. After defining these procedural steps
more completely in a group session, the participants should go to a setting with stairs and
observe people using the stairs. Observers should record independently whether each
behavior of the activity is “safe” or “at risk.” Then they should reconvene in a group meet-
ing and compare notes.
Group discussions about practice observations might very well lead to changed or
refined behavioral definitions. Additionally, it might be decided that some participants
need more education and training about the observation process. When observers can use
behavioral definitions and agree on the safe vs. at-risk occurrence of each behavior on 80
percent or more of the observation trials, you are ready for the next phase of DO IT—
Observation.

Observing behavior
The acronym “SOON” depicted in Figure 8.8 reviews the key aspects of developing ade-
quate definitions of critical behaviors to target for a DO IT process. You are ready for the
observation phase when you have a checklist of critical behaviors with definitions that are
Specific, Observable, Objective, and Naturalistic. We have already considered most of the
characteristics of behavioral definitions implied by these key words, and examples of
behavioral checklists are provided later in this chapter, as well as in Chapter 12 on “safety
coaching.”

S pecific
• Concise behavioral definition
• Unambiguous

O bservable
• Overt behaviors
• Countable and recordable

O bjective
• No interpretation
• nor attributions
• "What" not "W hy"

N aturalistic
• Normal interaction
• Real-world activity

Figure 8.8 Behavioral observations for the DO IT process should be “SOON.”

820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 140

140 Psychology of safety handbook

I have not yet explained one very important characteristic of “observable” behaviors.
They need to be quantifiable. In other words, observers of a target behavior should be able
to translate their experience into a form that can be counted and compared objectively with
other observations. Some meaningful aspect (or property) of the target behavior needs to
be recorded systematically so changes (or improvements) in the behavior can be monitored
over time. To do this, we have to consider various properties of behavior.

Properties of behavior
One property of behavior is intensity. When a person says something, the sound can range
from a low-intensity whisper to a high-intensity shout. Another property is speed. A fork
lift, for example, can be operated at fast or at slower speeds.
Still another property of behavior is duration. Some behaviors last only a few seconds,
like turning off the power to a machine, putting on safety glasses, or signing a “safety
pledge.” Others may continue for several hours, such as performing a series of responses
at a particular work station or discussing safe work practices during a group meeting.
For our purposes, the property of response frequency is usually most important. A partic-
ular response may occur once in a given period of time, or it may occur several times. The
rate of a behavior refers to its frequency of occurrence per unit of time. Most safety-related
behaviors can be considered in terms of frequency or rate. For example, the frequency or rate
of operating a fork lift at an at-risk speed is a meaningful target for a DO IT process.

Measuring behavior
Certain safe and at-risk behaviors start and stop often during a work period. Consider lift-
ing, smoking a cigarette, or praising a coworker. They are readily measurable in terms of
rate. On the other hand, the opportunities for some safety-related behaviors, like locking
out power to equipment, occur less often during the usual workday. Here, percentage of
occurrence per opportunity might be a more meaningful property to measure than fre-
quency or rate.
For example, situations that require locking out power, stacking racks less than three
high, and using cut-resistant gloves might vary considerably throughout the day. Thus, it
is most meaningful to consider the number of occasions the target behavior actually occurs
per total situations requiring that behavior. From these frequencies (total occurrences of
safe behavior and total opportunities for the safe behavior), a percentage of safe behavior
can be calculated and used to monitor safety performance.
Some behaviors should continue throughout lengthy work sessions. Protective apparel
such as safety glasses and ear plugs may need to be worn continuously. In these cases, it
might be most appropriate to observe and record the duration, or the total amount of time
the behavior occurs, rather than frequency. I shall return later to this issue of targeting and
measuring the most appropriate properties of safety-related behaviors. For now, it is
important to understand that the property targeted by a behavior-change intervention
depends on specific situational factors and program objectives. Generally, the goal of the
DO IT process is to increase the occurrence (frequency, rate, percentage, and/or duration)
of safe behaviors and decrease the occurrence of at-risk behaviors.

Recording observations
Accurate and permanent records of observed behavior are essential for a job safety analysis,
an injury investigation, and a successful DO IT process. Most existing records are in the form
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 141

Chapter eight: Identifying critical behaviors 141

of written comments, and often do not offer an objective behavioral metric, or a measure of
observable behavior. Before attempting to change a specific behavior, you should first
observe and record a certain property of that behavior. By measuring the frequency, dura-
tion, or percentage of occurrence per opportunity of a particular behavior over a sufficient
period of time, you determine the extent to which that behavior needs to be changed.
Careful observation of response frequency, for example, helps answer several impor-
tant questions.

• How does the frequency of the target behavior vary among different individuals?
• In what situations and at what times does the behavior occur most often?
• When and where does the behavior occur least often?
• How often does a person have an opportunity to make an appropriate safe behav-
ior but does not make it?
• What specific environmental changes occur before and after the target behavior
occurs?
• What environmental factors are supporting a particular at-risk behavior and/or
inhibiting, perhaps demotivating, a particular safe behavior?

Calculating a behavior rate. Two basic requirements are necessary to record the rate
of a behavior. First, a precise objective definition of the beginning and the end of the target
behavior is required. Any observer should be able to count the number of times the behav-
ior begins and ends within a given period. This results in a frequency measurement.
Second, it is important to record the time you begin observing the target behavior and the
time you stop. This gives you a record of the time interval during which the behaviors were
counted and enables you to convert response frequency into a response rate.
Response rate is calculated by dividing the frequency (for example, 45 occurrences of
a behavior) by the length of the time interval (for example, 15 minutes). The response rate
here is 3 responses per minute. Response rate is analogous to miles per hour. By translat-
ing frequencies into rates, comparisons between two measurements can be made even
when the lengths of observation periods are different. Independent frequency records can
only be compared if the lengths of the observation periods are the same; response rates are
comparable regardless of the lengths of the different recording periods.
Interval recording. Some safety-related behaviors begin and end relatively infre-
quently during a workday, but once they occur, they last for long durations. In this case, a
frequency or rate measure would not be as informative as a record of the length of time the
target behavior occurs. It might be more practical, though, to note periodically and sys-
tematically whether the target response is occurring in a particular situation. Instead of
watching an individual and counting the start and end of a particular behavior during a
given time period, an observer could intermittently look at the individual throughout the
work period in a given environmental setting and note whether or not the target behavior
is occurring. This measurement procedure is termed interval recording, in contrast to event
recording where the occurrence of a discrete behavior is counted during an observation
session and possibly converted to response rate.
At work, interval recording is often the most practical approach to observing and
recording critical behaviors. A checklist of critical behaviors is used and the observer
merely watches an individual work for a set period of time and checks off “safe” or “at-
risk” for each behavior on the list. The number of “safe” and “at-risk” checkmarks can be
totaled and used to calculate the percentage of “safe” behaviors recorded in a particular
interval. This is the approach recommended by Krause et al. (1996) and McSween (1995).
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 142

142 Psychology of safety handbook

Several methods are available for objectively observing and recording safe vs. at-risk
behaviors in real-world settings. Different situations call for different procedures. I give a
number of examples throughout this and subsequent chapters and hope at least one of
these methods will relate directly to the situation in which you want to apply the DO IT
process.

A personal example
My daughter Krista asked me to drive her to the local Virginia Department of Motor
Vehicles office to get her “learner’s permit.” She was 15 years old and thought she was
ready to drive. Of course, I knew better, but how do you fight a culture that puts teenagers
behind the wheel of motor vehicles before they are really ready for such a risky situation?
“Don’t worry, Dad,” my daughter said, “I’ve had driver’s education in high school.”
Actually, that was part of my worry. She was educated about the concepts and rules regard-
ing driving, but she had not been trained. She had not yet translated her education into
operations or action plans.
In order to obtain a license to operate a motor vehicle in Virginia before the age of 18,
teenagers with a learner’s permit are required to take seven two-hour instructional periods
of on-the-road experience with an approved driver-training school. For one-half of these
sessions they must be the driver; the rest of the time they sit in the back seat and perhaps
learn through observation. Thus, for seven one-hour sessions Krista drove around town
with an instructor in the front seat and one or more students in the back, waiting for their
turn at the wheel. This was an opportunity for my daughter to transfer her driver educa-
tion knowledge to actual performance.
Driving activators and consequences. On-the-job training obviously requires an
appropriate mix of observation and feedback from an instructor. Practice does not make
perfect. Only through appropriate feedback can people improve their performance. Some
tasks give natural feedback to shape our behavior. When we turn a steering wheel in a par-
ticular direction, we see immediately the consequence of our action, and our steering
behavior is naturally shaped. The same is true of several other behaviors involved in driv-
ing a motor vehicle, from turning on lights, windshield wipers, and cruise-control switches
to pushing gas and brake pedals.
However, many other aspects of driving are not followed naturally with feedback con-
sequences, particularly those that can prevent injury from vehicle crashes. Although we get
feedback to tell us our steering wheel, gas pedal, turn signal lever, and brakes work, we do
not get natural feedback regarding our safe vs. at-risk use of such control devices.
Therefore, as shown in Figure 8.9, feedback must be added to the driving situation if we
want behavior to improve.
Also, when we first learn to drive, we do not readily recognize the activators (or dis-
criminative stimuli) that should signal the use of various vehicle controls. This is com-
monly referred to as “judgment.” From a behavior-based perspective, “good driving
judgment” is recognizing environmental conditions (or activators) that signal certain vehi-
cle-control behaviors, and then implementing the controls appropriately.
I wondered whether my daughter’s driving instructor would give her appropriate and
systematic feedback regarding her driving “judgment.” Would he point out consistently
the activators that require safe vehicle-control behaviors? Would he put emphasis on the
positive, by supporting my daughter’s safe behaviors before criticizing her at-risk behav-
iors? Would he, or a student in the back seat, display negative emotional reactions in cer-
tain situations and teach Krista (through classical conditioning) to feel anxious or fearful in
particular driving situations? Would some at-risk driving behaviors by Krista or the other
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 143

Chapter eight: Identifying critical behaviors 143

Figure 8.9 Practice requires feedback to make perfect.

student drivers be overlooked by the instructor and lead to observational learning that
some at-risk driving behaviors are acceptable?
Developing a driving behavior checklist. Even if the driving instruction is optimal,
seven hours of such observation and feedback is certainly not sufficient to teach safe driv-
ing habits. I recognized a need for additional driving instruction for my daughter. We
needed a DO IT process for driving. The first step was to define critical behaviors to target
for observation and feedback. Through one-on-one discussion, my daughter and I derived
a list of critical driving behaviors and then agreed on specific definitions for each item. My
university students practiced using this critical behavior checklist (CBC) a few times with
various drivers and refined the list and definitions as a result. The CBC we eventually used
is depicted in Figure 8.10.

Using the critical behavior checklist

After refining the CBC and discussing the final behavioral definitions with Krista, I felt
ready to implement the second stage of DO IT—observation. I asked my daughter to drive
me to the university—about nine miles from home—to pick up some papers. I made it clear
I would be using the CBC on both parts of the roundtrip. When we returned home, I totaled
the safe and at-risk checkmarks and calculated the percentage of safe behaviors. Krista was
quite anxious to learn the results and I looked forward to giving her objective behavioral
feedback. I had good news. Her percentage of safe driving behaviors (percent safe) was 85
percent and I considered this quite good for our first time.
I told Krista her “percent safe” score and proceeded to show her the list of safe check-
marks, while covering the checks in the At-Risk column. Obviously, I wanted to make this
a positive experience, and to do this, it was necessary to emphasize the behaviors I saw her
do correctly. To my surprise, she did not seem impressed with her 85 percent safe score and
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 144

144 Psychology of safety handbook

Critical Behavior Checklist for Driving

Driver: Date: Day:
Observer 1 : Origin: Start Time:
Observer 2 : Destination: End Time:
Weather :
Road Conditions :
Behavior Safe At-Risk Comments
Safety Belt Use:
Turn Signal Use:
Left turn
Right turn
Lane change
Intersection Stop:
Stop sign
Red light
Yellow light
No activator
Speed Limits:
25 mph and under
25 mph- 35 mph
35 mph- 45 mph
45 mph- 55 mph
55 mph- 65 mph
Passing:
Lane Use:
Following Distance (2 sec):
Totals:
% Safe = Total Safe Observations = %
Total Safe + At-Risk Obs.

Figure 8.10 A critical behavior checklist (CBC) can be used to increase safe driving.

pushed me to tell her what she did wrong. “Get to the bottom line, Dad,” she asserted,
“Where did I screw up?” I continued an attempt to make the experience positive, by say-
ing, “You did great, honey, look at the high number of safe behaviors.” “But why wasn’t
my score 100 percent?” reacted Krista. “Where did I go wrong?”
This initial experience with the driving CBC was enlightening in two respects. It illus-
trated the unfortunate reality that the “bottom line” for many people is “where did I make
a mistake”? My daughter, at age 15, had already learned that people evaluating her per-
formance seem to be more interested in mistakes than successes. That obviously makes
performance evaluation (or appraisal) an unpleasant experience for many people.
A second important outcome from this initial CBC experience was the realization that
people can be unaware of their at-risk behaviors and only through objective feedback can
this be changed. My daughter did not readily accept my corrective feedback regarding her
four at-risk behaviors. In fact, she vehemently denied that she did not always come to a
complete stop. However, she was soon convinced of her error when I showed her my data
sheet and my comment regarding the particular intersection where there was no traffic and
she made only a rolling stop before turning right. I did remind her that she did use her turn
signal at this and every intersection and this was something to be proud of. She was devel-
oping an important safety habit, one often neglected by many drivers.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 145

Chapter eight: Identifying critical behaviors 145

I really did not appreciate the two lessons from this first application of the driving CBC
until my daughter monitored my driving. That is right, Krista used the CBC in Figure 8.10
to evaluate my driving on several occasions. I found this reciprocal application of a CBC to
be most useful in developing mutual trust and understanding between us. I found myself
asking my daughter to explain my lower than perfect score and arguing about one of the
recorded “at-risk” behaviors. I, too, was defensive about being 100 percent safe. After all,
I had been driving for 37 years and teaching and researching safety for more than 20 years.
How could I not get a perfect driving score when I knew I was being observed?
From our experience with the CBC, my daughter and I learned the true value of an
observation and feedback process. While using the checklist does translate education into
training through systematic observation and feedback, the real value of the process is the
interpersonal coaching that occurs. In other words, we learned not to get too hung up on
the actual numbers. After all, there is plenty of room for error in the numerical scores.
Rather, we learned to appreciate the fact that through this process people are actively car-
ing for the safety and health of each other in a way that can truly make a difference. We also
learned that even experienced people can perform at-risk behavior and not even realize it.
It is noteworthy that since these valuable feedback sessions with my daughter, the CBC
in Figure 8.10 had been refined for use in public transportation vehicles like buses and taxi
cabs (Geller, 1998). My students and I have evaluated a systematic application of the CBC
for driver training classes (DePasquale and Geller, 2001). Interestingly, we found it more
effective to present CBC feedback as an activator than a consequence. That is, students who
received their CBC results from a driving session immediately before their next session
showed significantly greater increases in safe driving behaviors than students who
received their CBC feedback immediately after a driving session.
With my daughter, I actually used the CBC as both an activator and a consequence,
which I recommend whenever possible. Specifically, I discussed the CBC results with my
daughter right after a driving session (as a consequence). Then prior to the next driving ses-
sion, I reviewed the CBC scores of her previous trip (as an activator). It seemed very useful
to remind my daughter of her prior success (to increase confidence and set high expecta-
tions for the current session) and to focus her attention on particular areas (i.e., behaviors)
for potential improvement. As mentioned previously, rigorous research has verified my
hypothesis (DePasquale and Geller, 2001).
My students and I have also produced an instructional videotape and workbook to
teach middle school children how to use a CBC to monitor the driving performance of their
parents (Geller et al., 1998). The process is called STAR for the critical components of an
effective observation and feedback process—“See”—“Think”—“Act”—“Reward.” The
psychology of setting children up as driving coaches for their parents is powerful if adults
can be open to such a process and show positive support. Perhaps your common sense tells
you such a process can have dramatic benefits for both parent and child. In fact, the prin-
ciples of psychology revealed in this text indicate the strongest long-term safety-related
benefits will occur for the child participant of a well received STAR process. The complete
rationale for this conclusion will be apparent by the time you finish reading this Handbook.
From unconscious to conscious Figure 8.11 depicts the process we often go through
when developing safe habits. Both Krista and I were unaware of some of our at-risk driv-
ing behaviors. For these behaviors, we were “unconsciously incompetent.” Through the
CBC feedback process, however, we became aware of our at-risk driving behaviors, but
awareness did not necessarily result in 100 percent safe behavior scores. Several feedback
sessions were needed before some safe driving behaviors occurred regularly and before
some at-risk behaviors decreased markedly or extinguished completely. In other
words, initial feedback made us “consciously incompetent” with regard to some driving
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 146

146 Psychology of safety handbook

Unconscious Conscious
Incompetence Incompetence
''bad habits'' ''learning''

Unconscious Conscious
Competence Competence
''safe habits'' ''rule governed''

UI: ''I didn't know there was a better way to do

it.''
CI: ''I know there's a better way; I need to learn
how to do it right.''
CC: ''I know I'm doing it right, because I'm
following the approved procedure.''
UC: '' I no longer think about it; I know it's right,
and now it's a safe habit.''

Figure 8.11 The DO IT process enables shifts from bad to good habits.

behaviors. Continuous feedback and mutual support resulted in beneficial learning, as

reflected in improved percent safe behavior scores on the CBC. Thus, for some driving
behaviors, we became “consciously competent.”
Feedback made us aware of certain driving rules or the driving situation (activator)
that calls for a particular safe behavior. Complying with these rules, developed with our
CBC feedback process, is referred to as “rule governed” behavior (Malott, 1988, 1992). This
stage involves thinking or talking to oneself to identify activators that require certain safe
behaviors and giving self-approval or self-feedback after performing the appropriate safe
behavior. With continuous observation and feedback from both others and ourselves, some
safe behaviors become automatic or habitual. They reach the “unconscious competence”
stage in Figure 8.11.
Some of my safe driving behaviors have progressed no further than the “conscious com-
petence” stage. The behavior has not become a habit. I need to remind myself on every occa-
sion to take the extra time or effort to set the safe example. These are the behaviors that
benefited most from the CBC feedback process, because over time I had gotten careless about
certain driving practices, especially stopping completely at intersections, maintaining a dis-
tance of two seconds behind vehicles in front, and staying in the right lane except to pass.
I am “unconsciously competent” about some safe driving practices, particularly safety-
belt and turn-signal use, but these behaviors were not always habitual. With safety-belt
use, I can recall going through each of the stages in Figure 8.11. When lap belts first
appeared in vehicles, I barely noticed them. I even remained “unconsciously incompetent”
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 147

Chapter eight: Identifying critical behaviors 147

for safety-belt use in 1974 when vehicles would not start unless the front-seat lap belts were
buckled. Like numerous other drivers (as observed by Geller et al., 1980), I merely buckled
my front-seat lap belt and sat on it.
In the mid-1970s, I learned well the statistics that justify the use of vehicle safety belts
on every trip. In fact, I actually taught the value of using safety belts in my safety work-
shops at the time. However, the popular quote “Do as I say, not as I do,” applied to me.
Even though I knew the value of safety-belt use, I still did not buckle up on every trip. I
was “consciously incompetent” with regard to this safe behavior.
From incompetence to competence. I started to buckle up consistently in the late
1970s only after my students made my belt use the target of an informal DO IT process.
My vehicle was visible from the large window in my research laboratory, and my stu-
dents began observing whether I was buckled up when entering and leaving the faculty
parking lot. After two weeks of collecting baseline data without my knowing it, they
informed me of their little “experiment” by displaying a graph of weekly percentages of
my safety-belt use.
I was appropriately embarrassed by the low percentages for the first two weeks of
the “project.” My students were holding me accountable for a behavior I should be
performing. That was sufficient to change my behavior. From that day on, I have always
buckled up. For about a year I had to think about it each time. I was “consciously compe-
tent.” Subsequently, safety-belt use in my vehicle became a habit, and I moved to the opti-
mal “unconscious competence” stage for this behavior. I bet many readers are now in this
stage for safety-belt use but can remember being at each of the earlier stages of habit
formation.
At what stage of habit formation are you when you get in the back seat of someone
else’s vehicle, like a taxi cab? It is possible to be “unconsciously competent” in some situa-
tions but be “consciously competent” or “consciously incompetent” in another situation
for the same behavior. For example, wearing safety glasses, ear plugs, and steel-toed shoes
might be a safe habit on the job, but which of these safe behaviors are followed when mow-
ing the lawn in your backyard?
A DO IT process could increase our use of personal protective equipment at home as
well as at work. First, we need to accept the fact that we can all be unconsciously or con-
sciously incompetent with regard to some behaviors. Next, we need to understand the
necessity of behavioral feedback to improve our performance. Then, we need others in our
family or work team to observe us with a CBC and then share their findings as actively car-
ing feedback.

Two basic approaches

The CBC examples described previously illustrate two basic ways of implementing the
Define and Observe stages of DO IT. The driving CBC I developed with my daughter illus-
trates the observation and feedback process recommended by a number of successful
behavior-based safety consultants (Krause, 1995; Krause et al., 1996; McSween, 1995). I
refer to this approach as one-to-one safety coaching because it involves an observer using
a CBC to provide instructive behavioral feedback to another person (Geller, 1995, 1998).
The second approach to the Define and Observe stages of DO IT involves a limited
CBC (perhaps targeting only one behavior) and does not necessarily involve one-to-one
coaching. This is the approach used in most of the published studies of the behavior-based
approach to safety (for example, see reviews by Petersen, 1989, and Sulzer-Azaroff, 1982,
1987). This was the approach used by my students years ago when they observed,
recorded, and graphed my safety-belt use as my vehicle entered and departed the
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 148

148 Psychology of safety handbook

faculty parking lot. In 1984, I taught this particular approach to plant safety leaders for 110
different Ford Motor Company plants (Geller, 1985, 1990).
Vehicle safety-belt use across all Ford plants increased from 8 percent to 54 percent, and
this behavior change in 1984 alone saved the lives of at least 8 employees and spared about
400 others from serious injury. Corporate cost savings were estimated at $10 million dur-
ing the first year and cumulated to $22 million by the end of 1985 (Gray, 1988).
Each of these approaches to the Define and Observe stages of DO IT are advantageous
for different applications within the same culture. Thus, it is important to understand the
basic procedures of each and to consider their advantages and disadvantages. For some
work settings, I have found it quite useful to start with the simpler approach of targeting
only a few CBC behaviors. With immediate success, behaviors are then added until even-
tually a comprehensive CBC is developed, accepted, and used willingly throughout a
worksite. For instance, after Ford Motor Company obtained remarkable success with appli-
cations of the DO IT process to increase vehicle safety-belt use, several Ford plants
expanded the process to target numerous on-the-job work practices (Geller, 1990).

Starting small
This approach targets a limited number of critical behaviors but does not require one-on-
one observation. A work group defines a critical behavior or behavioral outcome to
observe, as discussed earlier in this chapter. After defining their target so that two or more
observers can reliably observe and record a particular property of the behavior, usually fre-
quency of occurrence, the group members should give each other permission to observe
this work practice among themselves. If some group members do not give permission, it is
best not to argue with them. Simply exclude these individuals from the observations and
invite them to join the process whenever they feel ready.* They will likely participate even-
tually when they see that the DO IT process is not a “Gotcha Program” but an objective and
effective way to care actively for the safety of others and build a Total Safety Culture.
It often helps to develop a behavior checklist to use during observations. As discussed
earlier, target behaviors like “safe lifting” and “safe use of stairs” include a few specific
behaviors, either safe or at risk. Therefore, the CBC should list each behavior separately,
and include columns for checking “safe” or “at risk.” Figure 8.12 depicts a sample CBC for
safe lifting. Through use of this CBC, a work group might revise the definitions and possi-
bly add a lifting-related behavior relevant to their work area.
Participants willing to be observed anonymously for the target behavior(s) use the
CBC to maintain daily records of the safe and at-risk behaviors defined by the group. They
do not approach another individual specifically to observe him or her. Rather they look for
opportunities for the target behavior to occur. When they see a safe behavior opportunity (SBO),
they take out their checklist and complete it. If the target behavior is “safe lifting,” for
example, observers keep on the lookout for an SBO for lifting. They might observe such an
SBO from their work station or while walking through the plant. Of course, if they see an
at-risk lifting behavior and are close enough to reduce the risk, they should put their CBC
aside and intervene. Intervening to reduce risk must take precedence over recording an
observation of at-risk behavior.

* Some applications of the DO IT process have worked well without this permission phase, as in numerous safety-
belt promotion programs (Geller, 1985). However, obtaining permission first will help develop trust and increase
opportunities to expand the list of critical behaviors to target.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 149

Chapter eight: Identifying critical behaviors 149

Observer: Date:

Target Behavior Safe At-Risk

load appropriate
hold close
use legs
move feet - don't twist
smooth motion - no jerks

Comments (use back if necessary):

% Safe Observations:
Total Safe Observations X 100 =
Total Safe Observations + At-Risk Observations %

Figure 8.12 A critical behavior checklist (CBC) can be used to increase safe lifting.

This process could be used to hold people accountable for numerous behaviors or
behavioral outcomes. It is quite analogous to the standard environmental audit conducted
throughout industrial complexes to survey equipment conditions, environmental hazards,
and the availability of emergency supplies. Actually, most equipment and environmental
audits reflect behaviors. An equipment guard in place, a tool appropriately sharpened, a
work area neat and clean, and equipment power locked out properly are in that “safe” con-
dition because of employees’ behaviors. A behavior auditor might look for an SBO regard-
ing any number of safe environmental conditions. When they see an opportunity for the
safe target behavior to have occurred, they take out their CBC and record “safe” or “at risk”
to indicate objectively whether the desired safe behavior(s) had occurred to make the con-
dition safe.

Observing multiple behaviors

As the list of targets on a CBC increases, it becomes more and more difficult to complete a
checklist from a remote location. Auditing several critical behaviors usually puts observers
in close contact with another person (the performer), resulting in a one-on-one coaching sit-
uation (Geller, 1995). The observer should seek permission from the performer before
recording any observations, even though a work group might have agreed on the observa-
tion process in earlier education and training meetings. If the performer wishes not to be
observed, the observer should leave with no argument and a friendly smile. This helps to
build the trust needed to eventually reach 100 percent participation in the DO IT process.
Multiple behavior CBCs might be specific to a particular job or be generic in nature.
The driving CBC I used with Krista was a job-specific checklist, only relevant for operating
a motor vehicle. In contrast, a generic checklist is used to observe behaviors that may occur
at various job sites. The CBC depicted in Figure 8.13 is generic because it is applicable for
any job that requires the use of personal protective equipment (PPE). Because different PPE
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 150

150 Psychology of safety handbook

Critical Behavior Checklist for Personal Protective Equipment

Observation period (dates):

Observer:
TOTAL NUMBER NUMBER OF EMPLOYEES
OF EMPLOYEES OBSERVED USING
OBSERVED ALL REQUIRED PPE

PPE SAFE OBSERVATION AT-RISK OBSERVATION

(For Observed Area) (Proper Use of PPE) (Improper or No Use of PPE)

Gloves

Safety Glasses/Shield

Hearing Protection

Safety Shoes

Hard Hat

Lifting Belt

TOTAL

Figure 8.13 A critical behavior checklist (CBC) can be used to increase the use of personal
protective equipment (PPE).

might be required on different jobs, certain PPE categories on the CBC may be irrelevant
for some observations. For jobs requiring extra PPE, additional behaviors will be targeted
on the CBC. Obviously, the observer needs to know PPE requirements before attempting to
use a CBC like the one shown in Figure 8.13.
The CBC in Figure 8.13 includes a place for the observer’s name, but the performer’s
name is not recorded. Also, this CBC was designed to conduct several one-on-one behav-
ior audits over a period of time. Each time the observer performs an observation, he or she
places a checkmark in the left box (for total number of observations). If the performer was
using all PPE required in the work area, a check would be placed in the right-hand box.
From these entries, the overall percentage of safe employees can be monitored.
The checkmarks in the individual behavior categories of the CBC in Figure 8.13 are
totaled and, by dividing the total number of safe checks by the total safe and at-risk checks,
the percentage of safe behaviors for each PPE category can be assessed. See the formula at
the bottom of the CBC in Figure 8.12. This enables valuable feedback regarding the relative
use of various devices to protect employees. Such information might suggest a need to
make certain PPE more comfortable or convenient to use. It might also suggest the need for
special intervention as discussed in the next three chapters.
The formula at the bottom of the CBC in Figure 8.12 can be used to calculate an over-
all percent safe score. We have found it very effective to post this global score weekly for
different work teams. Such social comparison information presumably motivated per-
formance improvement through friendly intergroup competition (Williams and Geller,
2000). Chapter 12 also includes additional information on the design of CBCs for one-on-
one behavior observation.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 151

Chapter eight: Identifying critical behaviors 151

In conclusion
In this chapter we have gotten into the “nuts and bolts” of implementing a behavior-based
safety process to develop a Total Safety Culture. The overall process is referred to as DO IT,
each letter representing one of the four stages of behavior-based safety. This chapter
focused on the first two stages—Define and Observe.
Defining critical behaviors to target for observation and intervention is not easy. A
work team needs to consult a variety of sources, including the workers themselves, near-
hit reports, injury records, job hazard analyses, and the plant safety director. After select-
ing a list of behaviors critical to preventing injuries in their work area, the team needs to
struggle through defining these behaviors so precisely that all observers agree on a partic-
ular property of each behavior at least 80 percent of the time. The behavioral property most
often observed for industrial safety is frequency of occurrence per individual worker or per
group of employees.
A critical behavior checklist (CBC) is used to observe and record the relative frequency
(or percentage of opportunities) critical behaviors occur throughout a work setting. If the
CBC contains only a few behaviors or behavioral outcomes (conditions caused by behav-
ior), it is possible to conduct observations without engaging in a one-on-one coaching ses-
sion. This is often the best approach to use when first introducing behavior-based safety to
a work culture. It is not as overwhelming or time-consuming as one-on-one coaching with
a comprehensive CBC.
Over time and through building trust, a short CBC can be readily expanded and lead
to one-on-one safety coaching. Safety coaching is one very effective way to implement each
stage of the DO IT process and is detailed in Chapter 12. First, it is important to understand
how the first two stages of DO IT can facilitate a proper behavioral analysis of the situation.
This is the topic of the next chapter.

References
DePasquale, J. P. and Geller, E. S., Intervening to improve driving instruction: should behavioral
feedback be an antecedent or a consequence?, under editorial review, 2001.
Geller, E. S., Corporate Safety Belt Programs, Virginia Polytechnic Institute and State University,
Blacksburg, VA, 1985.
Geller, E. S., Performance management and occupational safety: start with a safety belt program, J.
Organ. Behav. Manage., 11(1), 149, 1990.
Geller, E. S., Safety coaching: key to achieving a Total Safety Culture, Prof. Saf., 40(7) 16, 1995.
Geller, E. S., Understanding Behavior-Based Safety: Step-by-Step Methods to Improve Your Workplace, 2nd
ed., J. J. Keller & Associates, Inc., Neenah, WI, 1998.
Geller, E. S., Ten leadership qualities for a Total Safety Culture: safety management is not enough,
Prof. Saf., 45(5), 38, 2000.
Geller, E. S., Casali, J. G., and Johnson, R. P., Seat-belt usage: a potential target for applied behavior
analysis, J. Appl. Behav. Anal., 13, 94, 1980.
Geller, E. S., Glaser, H. S., Chevaillier, C., McGorry, J., and Cronin, K., The Safety STAR Process:
Involving Young People in the Reduction of Highway Fatalities, Center for Applied Behavior
Systems, Virginia Tech, Blacksburg, VA, 1998.
Gray, D. A., Introduction to invited address by E. S. Geller at the annual National Safety Council
Congress and Exposition, Orlando, FL, October, 1988.
Kazdin, A. E., Behavior Modification in Applied Settings, 5th ed., Brooks/Cole Publishing Company,
Pacific Grove, CA, 1994.
Krause, T. R., Employee-Driven Systems for Safe Behavior: Integrating Behavioral and Statistical
Methodologies, Van Nostrand Reinhold, New York, 1995.
820049_CRC1_L1540_CH08 11/10/00 1:41 PM Page 152

152 Psychology of safety handbook

Krause, T. R., Hidley, J. H., and Hodson, S. J., The Behavior-Based Safety Process: Managing
Involvement for an Injury-Free Culture, 2nd ed., Van Nostrand Reinhold, New York, 1996.
Langer, E. S., Mindfulness, Perseus Books, Reading, MA, 1989.
Langer, E. S., Mindful Learning, Perseus Books, Reading, MA, 1997.
Malott, R. W., Rule-governed behavior and behavioral anthropology, Behav. Anal., 22, 181, 1988.
Malott, R. W., A theory of rule-governed behavior and organizational behavior management,
J. Organ. Behav. Manage., 12(2), 45, 1992.
McSween, T. E., The Values-Based Safety Process: Improving Your Safety Culture with a Behavioral
Approach, Van Nostrand Reinhold, New York, 1995.
Petersen, D., Safe Behavior Reinforcement, Aloray, Inc., Goshen, NY, 1989.
Skinner, B. F., Science and Human Behavior, Macmillan, New York, 1953.
Skinner, B. F., Contingencies of Reinforcement: A Theoretical Analysis, Appleton-Century-Crofts, New
York, 1969.
Sulzer-Azaroff, B., Behavioral approaches to occupational safety and health, in Handbook of
Organizational Behavior Management, Frederiksen, L., Ed., Wiley, New York, 1982.
Sulzer-Azaroff, B., The modification of occupational safety behavior, J. Occup. Accid., 9, 177, 1987.
Williams, J. H. and Geller, E. S., Behavior-based intervention for occupational safety: critical impact
of social comparison feedback J. Saf. Res., 31(3), 135, 2000.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 153

chapter nine

Behavioral safety analysis

The defining and observing processes of DO IT provide opportunities to evaluate the situational fac-
tors contributing to at-risk behavior and a possible injury. This chapter details the procedures of a
behavioral safety analysis, including a step-by-step examination of the situational, social, and per-
sonal factors influencing at-risk behavior in order to determine the most cost-effective corrective
action. Critical distinctions are made between four types of intervention—instruction, motivation,
support, and self-management—between training and education, and between accountability and
responsibility.

“A prescription without diagnosis is malpractice.”—Socrates

Chapter 8 introduced the DO IT process and provided some detail about the first two
steps—define target behavior(s) to improve and observe the target behavior occurring natu-
rally in the work environment. The CBC (critical behavior checklist) was introduced as a
way to look for the occurrence of critical behaviors during a work routine and then offer
workers one-on-one feedback about what was safe and what was at risk. This is behavioral
coaching and is explained in more detail in Chapter 12.
The checks in the safe and at-risk columns of a CBC can be readily summarized in a
“percent safe score.” As I discussed in the previous chapter, an overall global score can
be calculated by dividing the total number of behavioral observations (i.e., all checks on
all CBCs) into the total number of safe observations (i.e., all checks in the safe columns of
all CBCs). This provides an overall estimate of the safety of the workforce with regard to
the critical behaviors targeted in the observation step of DO IT.
The global “percent safe score” does not provide direction regarding which particular
behaviors need improvement, but it can provide motivation to a workforce that wants to
improve (Williams and Geller, 2000). It is an achievement-oriented index that holds
employees accountable for things they can control. This assumes, of course, that the work-
ers know the safe operating procedures for every work task.
If some employees are not sure of the safe way to perform a certain job, behavioral
direction is needed. A global “percent safe score” is not sufficient. When the CBC is
reviewed during a one-on-one coaching session, behavioral direction is provided. The
worker sees what critical behaviors were observed as “safe” and “at-risk.” A constructive
conversation with the coach provides support for safe behavior and corrective feedback for
behavior that could be safer. Often this includes suggestions for making the safe behavior
more convenient, comfortable, and easier to remember. It might also include the removal
of barriers (physical and social) that inhibit safe behavior.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 154

154 Psychology of safety handbook

Groups can receive support and direction for specific behavior if a “percent safe score”
is derived per behavior. That is, instead of adding safe and at-risk checks across behaviors,
calculate separate percentages for each behavior. Then the group can see which critical
behaviors are safe and which need improvement. The results can be posted in a prominent
location, as depicted in Figure 9.1, and discussed in team meetings.
It is important to have frank and open group conversations about the physical and
social barriers to safe behavior. The synergy from group discussion can result in creative
ways to make work settings more user friendly and conducive to safe behavior. Group con-
sensus-building (as detailed later in Chapter 18) can help establish new social norms
regarding the safe way to do something.
The key is constructive, behavior-focused conversation at both the group and individ-
ual level. One key to making a conversation constructive is to have useful information.
Where does this information come from? You guessed it; it comes from the behavioral
observations and it is more than the various percentages derived from tallying and divid-
ing certain column checks. Information is needed about environmental and social factors
that influence the occurrence or nonoccurrence of critical behaviors.
Often behaviors not targeted by the CBC become relevant, perhaps because they facil-
itate or hinder the performance of a critical behavior or because they should be included
on a revision of the CBC. Sometimes behavior reflects misunderstanding or a need for edu-
cation and training or certain behaviors could suggest fatigue, apathy, or a mismatch
between a person’s talents and the job.
Where on a CBC does an observer record these unanticipated conditions, events, and
behaviors so crucial to a constructive, behavior-based conversation? Every CBC should
have a place for comments (see, for example, Figures 8.10 and 8.12). Here is where the
observer records any observations that might be useful in a one-to-one feedback session or
a group discussion. The quality of a completed CBC usually increases with the number of
useful comments.
So while an observer is checking for safe or at-risk occurrences of critical behaviors, he
or she is watching for contributing factors to at-risk behavior, as well as for the occurrence
of behaviors that ought to be included on a revised CBC. This enables a second key to hav-
ing a constructive, behavior-based conversation—analysis.

Figure 9.1 The display of behavioral feedback provides direction and motivation for
improvement.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 155

Chapter nine: Behavioral safety analysis 155

The information in the “Comment” section of a CBC is invaluable in determining what

factors contribute to at-risk behavior and should be changed to reduce such behavior.
Before intervening to correct a problem, we need to conduct a proper behavior analysis.
This is the theme of this chapter. Without a careful behavioral analysis of the situation
requiring intervention or corrective action, we are indeed vulnerable to malpractice.
In addition to providing direction for recording observations in the comment section
of a CBC, this chapter offers basic guidelines for analyzing the behavioral aspects of a near
hit or injury. The purpose is to determine the most cost-effective corrective action. This is
obviously not fault finding nor victim blaming. Thus, we are not talking about “accident
investigation.” This is “incident analysis” with a focus on behavior.

Reducing behavioral discrepancy

It is important to consider human performance problems as a discrepancy rather than a
deficiency (Mager and Pipe, 1997). This places the focus on the behavior, not the individ-
ual. In other words, a difference exists between the behavior demonstrated and the behav-
ior desired. When evaluating safety problems, this discrepancy is between behavior
considered to be at risk vs. safe.
The behavioral discrepancy could be a “sin of omission” or a “sin of substitution.” The
worker might have failed to perform a particular safe behavior because he took a short cut
or the individual could have performed a certain behavior that puts someone at risk for
injury. After deciding what is safe and what is at risk for a particular individual and work
situation, an action plan can be designed to reduce the discrepancy between what is and
what should be.
Too often “retraining” or “discipline” (meaning punishment) are selected impulsively
as a corrective action for behavior change when another less costly and more effective
approach is called for. A proper behavioral safety analysis enables the selection of the most
cost-effective intervention. Let us consider the variety of situations or work contexts that
can influence a behavioral discrepancy.

Can the task be simplified?

Before designing an intervention to reduce a behavioral discrepancy, make sure all possi-
ble engineering “fixes” have been implemented. For example, consider the many ways the
environment could be changed to reduce physical effort, reach, and repetition. In other
words, entertain ways to make the job more user friendly before deciding what behaviors
are needed to prevent injury. This is, of course, the rationale behind ergonomics and the
search for engineering solutions to occupational safety and health.
As discussed earlier in Chapter 6, when people experience failure, as reflected by non-
compliance, property damage, or personal injury, they are more likely to place blame on
external than personal factors. In other words, as illustrated in Figure 9.2, people involved
in an injury feel more comfortable discussing environment-related causes than individual
factors. Given this self-serving bias (Schlenker, 1980), it makes sense to begin a behavior
analysis with a discussion of environmental or engineering factors. Afterward, the possi-
bility of a constructive discussion of personal factors potentially contributing to the inci-
dent increases markedly.
Sometimes behavior facilitators can be added, such as

1. Control designs with different shapes so they can be discriminated by touch as well
as sight.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 156

156 Psychology of safety handbook

Figure 9.2 We are reluctant to accept personal responsibility for our injuries.

2. Clear instructions placed at the point of application.

3. Color codes to aid memory and task differentiation (Norman, 1988).
4. Convenient machine lifts or conveyor rollers to help with physical jobs.

Plus, it is possible complex assignments can be redesigned to involve fewer steps or more
people. To reduce boredom or repetition, simple tasks might allow for job swapping.
Ask these questions at the start of a behavior analysis:

• Can an engineering intervention make the job more user friendly?

• Can the task be redesigned to reduce physical demands?
• Can a behavior facilitator be added to improve response differentiation, reduce
memory load, or increase reliability?
• Can the challenges of a complex task be shared?
• Can boring, repetitive jobs be swapped?

Is a quick fix available?

From their more than 60 combined years of analyzing and solving human performance
problems, Mager and Pipe (1997) conclude that many discrepancies between real and ideal
behavior can be eliminated with relatively little effort. More specifically, behavior might be
more at risk than desired because expectations are unclear, resources are inadequate, or
feedback is unavailable.
In these cases, solutions to reducing a behavioral discrepancy are obvious and rela-
tively inexpensive. Behavior-based instruction or demonstration can overcome invisible
expectations, and behavior-based feedback can enable continuous improvement.
Furthermore, a work team could decide what resources are needed to make a safe behav-
ior more convenient, comfortable, or efficient.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 157

Chapter nine: Behavioral safety analysis 157

When conducting this aspect of a behavioral analysis, ask these questions.

• Does the individual know what safety precautions are expected?

• Are there obvious barriers to safe work practices?
• Is the equipment as safe as possible under the circumstances?
• Is protective equipment readily available and as comfortable as possible?
• Do employees receive behavior-based feedback related to their safety?

Is safe behavior punished?

As I explained in Chapter 8, a key principle of applied behavior analysis is that behavior is
motivated by its consequences. In other words, our behavior results in favorable or unfa-
vorable consequences, and these consequences determine our future behavior. Sometimes
naturally occurring consequences work against us. This is especially true in safety because
safe behavior is usually less comfortable, convenient, or efficient than the at-risk alternative.
Those analyzing an incident need to try to see the situation through the eyes of the per-
former (Geller, 2000). This is called empathy and is illustrated in Figure 9.3. Some conse-
quences might actually seem positive to an observer but be viewed as negative by the
performer. For example, a safety manager might consider an individual’s public safety
award a positive consequence, but for the individual it could be a negative consequence
because of expected harassment from coworkers. How about the “rate busters” in indus-
try and school who excel in work quantity or quality? Praise from supervisors and teach-
ers is often overshadowed by punishing consequences from coworkers and classmates.
The result is often a reduction in individual output.
In some work cultures, the interpersonal consequences for reporting an environmental
hazard or near hit are more negative than positive. After all, these situations imply that

Figure 9.3 It is useful to see a situation through the eyes of the other person.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 158

158 Psychology of safety handbook

Figure 9.4 Sometimes exemplary performance is punished.

someone was irresponsible or careless. It is not unusual for people to be ridiculed for wear-
ing protective gear or using an equipment guard. It might even be considered “cool” or
“macho” to work unprotected and take risky short cuts. The hidden agenda might be that
“only a ‘chicken’ would wear that fall protection.”
Mager and Pipe (1997) refer to these situations as “upside-down consequences” and
suggest that whenever a behavioral discrepancy exists, part of the problem is owing to the
desired behavior being punished. Are people put down when they should be lifted up? Are
the consequences for performing well punishing rather than rewarding as illustrated in
Figure 9.4?
It is really not rare for the best performers to be “rewarded” with extra work. That
might seem like an efficient management decision, but the long-term results of this
“upside-down consequence” could be detrimental. Giving extra work for exemplary per-
formance can lead to avoidance behavior. In other words, a behavioral discrepancy might
occur to avoid extra work assignments.
Ask these questions during your behavior analysis.

• What are the consequences for desired behavior?

• Are there more negative than positive consequences for safe behavior?
• What negative consequences for safe behavior can be reduced or removed?

Is at-risk behavior rewarded?

As indicated previously, at-risk behavior is often followed by natural positive conse-
quences. Short cuts are usually taken to save time and can lead to a faster rate of output. So
taking on at-risk short cuts can be labeled “efficient” behavior. I have analyzed several
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 159

Chapter nine: Behavioral safety analysis 159

work environments, for example, where bypassing or overriding the power lockout
switches was acceptable because it benefited production—the bottom line. In these cul-
tures, the worker who could fix or adjust equipment without locking out the power was a
hero. He could handle equipment problems without slowing down production.
Behavior does not occur in a vacuum. Most people perform the way they do because
they expect to achieve soon, certain, and positive consequences or they expect to avoid
soon, certain, and negative consequences. People take calculated risks because they expect
to gain something positive and/or avoid something negative.
Ask these questions.

• What are the soon, certain, and positive consequences for at-risk behavior?
• Does a worker receive more attention, prestige, or status from coworkers for at-risk
than safe behavior?
• What rewarding consequences for at-risk behavior can be reduced or removed?

Are extra consequences used effectively?

Because the natural consequences of comfort, convenience, or efficiency usually support
at-risk over safe behavior, it is often necessary to add extra consequences. These usually
take the form of incentive–reward or disincentive –penalty programs. Unfortunately, many
of these programs do more harm than good because they are implemented ineffectively
(Geller, 1996). Disincentives are often ineffective because they are used inconsistently and
motivate avoidance behavior rather than achievement. Moreover, safety incentive pro-
grams based on outcomes stifle the development and administration of an effective safety
incentive program to improve behavior. Details about designing an effective safety incen-
tive–reward program are given in Chapter 11.
Ask these questions when analyzing the impact of extra consequences put in place to
motivate improved safety performance.

• Can the punishment consequences be implemented consistently and fairly?

• Can the safety incentives stifle the reporting of injuries and near misses?
• Do the safety incentives motivate the achievement of safety-process goals?
• Do monetary rewards foster participation only for a financial payoff and conceal the
real benefit of safety-related behavior—injury prevention?
• Are workers recognized individually and as teams for completing process activities
related to safety improvement?

Figure 9.5 depicts a variety of extra and extrinsic consequences that can influence
occurrences of safe or at-risk behavior. Three categories of behavior are shown (safe, at risk,
and production related) as potentially influenced by four different types of behavior-based
consequences. None of the consequence examples are natural or intrinsic to the task.
Rather they are added to the situation in an attempt to sustain desired behavior or change
undesired behavior. Therefore, each consequence manipulation in Figure 9.5 can be con-
sidered a corrective action.
The consequence examples in Figure 9.5 are not presented as recommended inter-
ventions. They are given only to illustrate the variety of potential individual and group
consequences that can change the context of a work situation and, thus, influence
occurrences of safety-related and production-related behaviors. Obviously, a careful
behavior analysis of the work situation is needed (as outlined previously) to determine
what kinds of motivating consequences should be removed from or added to the work
context.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 160

160 Psychology of safety handbook

Individual Consequence Group Consequence

Reward Penalty Reward Penalty
"Thank You "Sissy" or other Group Team ranked
Safe Card" for non-macho celebration after at bottom for
Behavior cleaning up comment for 100 coaching attendance at
spill using PPE sessions safety meetings
Praise for Verbal High-fives for Group
adjusting reprimand for team lifting reprimand for
At-Risk
equipment walking outside without a host unreported
Behavior
without locking yellow line property
out power damage
Praise for Written Group Work team
Production efficiency
working 12 warning for ranked last on
Behavior plaque for
hours overtime omitting a "Resource
quality check fastest work Management"

Figure 9.5 Various extrinsic consequences can influence safety-related behaviors.

Is there a skill discrepancy?

But what about those times when the individual really does not know how to do the
prescribed safe behavior? The person is “unconsciously incompetent.” This situation
might call for training which is a relatively expensive approach to corrective action. Mager
and Pipe (1997) claim that most of the time a behavioral discrepancy is not caused by a
genuine lack of skill. Usually people can perform the safe behavior if the conditions
and the consequences are right. So training should really be the least used approach for cor-
rective action.
Ask these questions to determine whether the behavioral discrepancy is caused by a
lack of skill.

• Could the person perform the task safely if his or her life depended on it?
• Are the person’s current skills adequate for the task?
• Did the person ever know how to perform the job safely?
• Has the person forgotten the safest way to perform the task?

What kind of training is needed?

Answers to the last two questions can help pinpoint the kind of intervention needed to
reduce a skill discrepancy. More specifically, a “yes” answer to these questions implies
the need for a skill maintenance program. Skill maintenance might be needed to help a
person stay skilled such as police officers practicing regularly on a pistol range to stay
ready to use their guns effectively in the rare situation when they need them. This is, of
course, the rationale behind periodic emergency training. People need to practice the
behaviors that could prevent injury or save a life during an emergency. Fortunately, emer-
gencies do not happen very often; but since they do not, people need to go through the
motions just to “stay in practice.” Then, if the infrequent event does occur, they will be
ready to do the right thing.
A very different kind of situation also calls for skill maintenance training. This is when
certain behaviors occur regularly, but discrepancies still exist. Contrary to circumstances
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 161

Chapter nine: Behavioral safety analysis 161

requiring emergency training, this problem is not lack of practice. Rather, the person gets
plenty of practice doing the behavior ineffectively or unsafely. In this case, practice does
not make perfect but rather serves to entrench a bad (or at-risk) habit.
Vehicle driving behavior is perhaps the most common and relevant example of this
second kind of situation in need of skill maintenance training. Most drivers know how to
drive a vehicle safely, and at one time they showed little discrepancy between safe and at-
risk driving. However, for many drivers, safe driving has deteriorated, with some safe
driving practices dropping out of some people’s driving repertoire completely.
Practice with appropriate behavior-based feedback is critical for solving both types
of skill discrepancies. However, if the skill is already used frequently but has deteriorated
(as in the driving example), it is often necessary to add an extra feedback intervention
to overpower the natural consequences that have caused the behavior to drift from the
ideal.
Whereas the police officer gets task-inherent feedback to improve performance on the
pistol range, at-risk drivers might need behavior-based coaching to improve. As described
in Chapter 8, I used this kind of behavior-based coaching to teach my daughter safe driv-
ing practices and she used the same technique to give me feedback about a few of my driv-
ing behaviors that had drifted from the ideal. More details on behavior-based coaching are
presented in Chapter 12.
Ask these questions to determine whether the cause of the apparent skill discrepancy
is owing to lack of practice or lack of feedback.

• How often is the desired skill performed?

• Does the performer receive regular feedback relevant to skill maintenance?
• How does the performer find out how well he or she is doing?

Is the person right for the job?

From this discussion it is clear a skill discrepancy can be handled in one of two ways.
Change the job or change the behavior. The first approach is exemplified by simplifying the
task, while the latter approach is reflected in practice and behavior-based feedback or
behavioral coaching, but what if a person’s interests, skills, or prior experiences are incom-
patible with the job?
The person might be like me, for example, and be “computer challenged.” (Yes,
I am writing the first draft of this Handbook by hand). Sure I could learn how to operate
and even fix a computer if my life depended on it, but the training process would take
relatively long and I would not like it. So before investing in skill training for a particular
individual, it is a good idea to assess whether the person is right for the task. If the
person does not have the motivation or the physical and mental capabilities for a par-
ticular assignment, the cost-effective solution is to replace the performer. If you do not,
you will not only suboptimize work output, you will increase the risk for personal
injury.
Ask these questions to determine whether the individual has the potential to handle
the job safely and effectively.

• Does the person have the physical capability to perform as desired?

• Does the person have the mental capability to handle the complexities of the task?
• Is the person overqualified for the job and, thus, prone to boredom or dissatis-
faction?
• Can the person learn how to do the job as desired?
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 162

162 Psychology of safety handbook

What is the Performance Discrepancy? Which Solution(s) yield the most for least effort?

Is Change Called For? What Kind of Training is Needed?

Can the Task be Simplified? Is the Person Right for the Task?

Are Expectations Clear? Is There a Skill Discrepancy?

Is Behavior-Based Feedback Available? What are the Natural Consequences?

Figure 9.6 Ask ten basic questions to conduct a behavior-based incident analysis.

In summary
Figure 9.6 summarizes the main steps of a behavior-based incident analysis with a flow
chart of ten basic questions to ask. Before an individual worker is targeted with a training
intervention, engineering strategies are considered for task simplification.
The bottomline. Before deciding on an intervention approach, conduct a careful
analysis of the situation, the behavior, and the individual(s) involved in an observed dis-
crepancy between desired and actual behavior. Do not impulsively assume corrective
action to improve behavior requires training or “discipline.” A behavioral incident analy-
sis will likely give priority to a number of alternative intervention approaches. I discuss
critical disadvantages of using “discipline” or a punishment approach to corrective action
in Chapter 11.

Behavior-based safety training

The principles and procedures of behavior-based safety, including behavioral observation
and interpersonal coaching, are new to most people. Therefore, to achieve a Total Safety
Culture, behavior-based safety training is needed throughout a work culture. Everyone in
a workforce needs to understand the basic rationale or theory behind the behavior-based
approach. Then work teams need to participate in exercises to customize observation,
analysis, and feedback procedures for their work areas. Finally, practice sessions are
needed in which individuals and teams receive supportive and corrective feedback regard-
ing their implementation of behavior-based safety—from designing a CBC and analyzing
CBC results to using a CBC for constructive intervention.
Why should employees want such training? Figure 9.7 illustrates one reason, but I
hope a more proactive rationale can motivate participation. First, as I have indicated in ear-
lier chapters, behavior-based safety works to reduce injuries. The principles and methods
of behavior-based safety are applicable in many situations—when and wherever human
performance is a factor and can be improved. Thus, training in behavior-based safety pro-
vides skills useful in numerous domains at work, at home, during recreational and sport
activities, and traveling in between.
In a comprehensive research project, my Ph.D. students and I systematically evaluated
20 different industrial sites where behavior-based safety had been in effect for at least one
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 163

Chapter nine: Behavioral safety analysis 163

Figure 9.7 Most employees in a work culture need basic behavior-based safety training.

year. After the implementation of the behavior-based approach, injuries were reduced sig-
nificantly at each of these sites. The purpose of this NIOSH-supported research was to
determine the critical success factors for behavior-based safety (DePasquale and Geller,
1999; Geller et al., 1998a,b). The factor which predicted the greatest amount of employee
participation in behavior-based safety was having effective and comprehensive training in
behavior-based safety.
Thus, the value of giving quality behavior-based safety training cannot be overem-
phasized. Obviously, people need to know how to carry out a process. They need sufficient
training to feel confident they can complete every procedural step effectively, but they also
need to believe the process is worthwhile. More specifically, they need to trust that imple-
menting the methods of behavior-based safety will work to prevent injuries. This requires
education, not training. There is a difference.

Safety training vs. safety education

Let us understand the difference between education and training. Actually, you already
know the distinction. Do you want your teenager to receive sex education or sex training?
In contrast, are you satisfied if your teenager receives only “driver education,” or do you
prefer some “training” with that education?
Because people know intuitively the difference between education and training, mis-
using these terms can lead to problems. We might perceive safety training as a step-by-step
procedure or program with no room for individual creativity, ownership, or empower-
ment. This is how safety can come to be viewed as a top-down “flavor of the month.” If we
do not educate people about the principles or rationale behind a particular safety policy,
program or process, they might participate only minimally. They will perceive the program
as a requirement rather than an opportunity to make a difference. They might even see
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 164

164 Psychology of safety handbook

Figure 9.8 Without education and training, we clown around with our biased common
sense.
themselves as animals in a “circus,” well trained to jump through hoops, rather than as
members of a safety community, empowered to go beyond the call of duty for safety.
By the same token, safety education without follow-up training will not reap optimal
benefits. Learning the theory or principles behind an intervention approach is crucial for
customizing intervention procedures for a particular work situation, but after the proce-
dures are developed—hopefully with input from an educated work team—training is nec-
essary. People need to know precisely what to do. With proper education, these
participants can refine or upgrade procedures when appropriate, and with a change in pro-
cedures, additional training is obviously needed.
Bottomline. People need both education and training to improve. As Deming is
known for reiterating at his quality and productivity workshops, “There’s no substitute for
knowledge” (Deming, 1991, 1992). Indeed, without gaining profound knowledge through
education and training, we are like the clown in Figure 9.8. We do our best with what we
now know. We use our biased and ineffective common sense.

Different teaching techniques

Teaching styles are not the same for education and training. When I lecture to large groups
of university students or to safety professionals and hourly workers, I use various tech-
niques to maintain attention and get participants involved in the learning process. I might
use brightly colored overheads, write statements on a blackboard or flipchart, make an
extreme statement to elicit contrary reaction, or ask pointed questions and solicit answers
from the audience. My purpose is to influence the participants’ cognitive or thinking
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 165

Chapter nine: Behavioral safety analysis 165

processes (Langer, 1997). This kind of presentation might increase profound knowledge or
critical thinking skills, and this could lead to behavior change. In this case, I try to “think a
person into behaving a certain way.” In other words, education targets thought processes
directly and might indirectly influence what people do.
Training targets behavior directly and might indirectly influence thought processes.
This typically calls for more than a lecture format. Training might start with a specification
of the steps needed to accomplish a particular task but more than this is needed to assure
certain skills or procedures are learned. Participants in a training course should practice the
desired behavior and receive pertinent feedback to support what is right and correct what
is wrong. If feedback is given genuinely in a trusting and caring atmosphere, behavior might
not only be directly improved, but one’s thinking or attitude associated with the behavior
might be positive. Then training would “act a person into a certain way of thinking.”

An illustrative example
My colleagues at Safety Performance Solutions use both education and training to teach
safety coaching skills. They start with education, teaching the basic principles behind a
behavior-based approach to coaching. Then they use group exercises to implement a train-
ing process.
In one small-group exercise, participants develop a brief skit to demonstrate the coach-
ing principles they have learned. For example, one person sets the stage, another person
demonstrates safe or at-risk behavior, and a third person gives rewarding and/or correc-
tive feedback. When the skit is performed in front of the group, everyone can give feedback
on how principles translate into practice. If done right, the feedback from the audience and
the educator/trainer improves the performance.
In one variation of this training process, we have asked groups to first show us the
wrong way to coach and then to demonstrate the right way. After the group acts out appro-
priate safety coaching, the audience can offer supportive and corrective feedback. Usually
the educator/trainer finds opportunities to add to an observer’s feedback and points out
how that feedback could have been more constructive. This teaching frequently includes
restating the underlying principle or rationale. In this way, education and training go hand-
in-hand to maximize real benefits from the learning process.

In summary
Although the title of this section is “Behavior-Based Safety Training,” I hope it is clear that
both training and education are needed. First, people need to understand and believe in the
theory and principles underlying the behavior-based approach to preventing injuries. This
is commonly referred to as education.
Understanding, belief, or awareness is not sufficient, however, to implement a partic-
ular behavior-based safety process. People need to learn the specific behaviors or activities
required for successful implementation. This requires training and should include behav-
ior-based observation and feedback. In other words, participants need to practice the
behaviors called for by the intervention process and then receive constructive behavior-
focused feedback from objective and vigilant observers.
Making this distinction between education and training in conversation and applica-
tion can help to straighten out the apparent confusion among safety professionals, con-
sultants, and employees regarding differences between attitudes and behaviors, and ways
to improve these critical human dimensions. Attitudes, beliefs, values, intentions, and per-
ceptions can be influenced directly through education; whereas behaviors are directly
influenced through training.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 166

166 Psychology of safety handbook

Education can influence behaviors indirectly if the education process changes an atti-
tude, intention, belief, or value which is perceived as linked to a certain behavior. Training
can also influence attitudes, intentions, beliefs, or values indirectly if the behavior change
is accepted by the participant and perceived as related to a particular attitude, intention,
belief, or value. The bottom line is that a strategic combination of both education and train-
ing is needed to improve both behavior and attitude.

Intervention and the flow of behavior change

Taken together, education and training are instruction, and instruction represents a type of
intervention. Under certain circumstances, instruction is sufficient to change behavior.
Sometimes instruction does not work, and another type of intervention is needed. Perhaps
a motivational intervention is called for or, maybe, only supportive intervention is needed.
A complete behavioral safety analysis should often include a recommendation for a certain
type of behavior-change intervention. This section provides information critical for mak-
ing such a recommendation. Then subsequent chapters provide guidance for designing a
certain type of behavior-change intervention.
We have already covered a variety of situational factors that influence the occurrence
of safe or at-risk behavior. This included a sequence of questions to ask in order to decide
whether instructional intervention is needed, whether another approach to corrective
action would be more cost-effective—from redesigning a task to clarifying expectations
and providing behavior-based feedback. Here we examine some basic principles about
behavior and behavior-change techniques that should influence your choice of an improve-
ment intervention. We begin with a distinction among other-directed, self-directed, and
automatic behavior (Watson and Tharp, 1997).

Three types of behavior

On-the-job behavior starts out as other-directed behavior, in the sense that we follow some-
one else’s instructions. Such direction can come from a training program, an operation’s
manual, or policy statement. After learning what to do, essentially by memorizing or inter-
nalizing the appropriate instructions, our behavior can become self-directed. We can talk
to ourselves or formulate an image before performing a behavior in order to activate the
right response.
Sometimes we talk to ourselves after performing a behavior in order to reassure our-
selves we performed it correctly or we figure out ways to do better next time. At this point,
we are usually open to corrective feedback that is delivered well.
After performing some behaviors frequently and consistently over a period of time
they become automatic. A habit is formed. Some habits are good and some are not good,
depending on their short- and long-term consequences. If implemented correctly, rewards,
recognition, and other positive consequences can facilitate the transfer of behavior from the
self-directed phase to the habit phase.
Of course, our self-directed behavior is not always desirable. When we take a calcu-
lated risk, for example, we are choosing intentionally to ignore a safety precaution or take
a short cut in order to perform more efficiently or with more comfort or convenience. In this
state, people are “consciously incompetent.” It is often difficult to change self-directed
behavior from incompetent to competent, because such a transition usually requires a rel-
evant change in personal motivation.
Before a bad habit can be changed to a good habit, the target behavior must become
self-directed. In other words, people need to become aware of their undesirable habit (as in
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 167

Chapter nine: Behavioral safety analysis 167

at-risk behavior) before adjustment is possible. Then, if the person is motivated to improve
(perhaps as a result of an incentive –reward program), his new self-directed behavior can
become automatic.
Let us see what kinds of behavior-based interventions are appropriate for the three
transitions referred to previously.

• Turning a risky habit (when the person is unconsciously incompetent) into self-
directed behavior.
• Changing risky self-directed behavior (when the person is consciously incompe-
tent) to safe self-directed behavior.
• Turning safe self-directed behavior (when the person is consciously competent) into
a safe habit (unconscious competence).

Three kinds of intervention strategies

In Chapter 8, I explained the ABC model as a framework to understand and analyze behav-
ior as well as to develop interventions for improving behavior. Recall that the “A” stands
for activators or antecedent events that precede behavior or “B”, and “C” refers to the con-
sequences following behavior and produced by it. Of course, you remember activators
direct behavior, while consequences motivate behavior.
Activators and consequences are external to the performer (as in the environment) or
they are internal (as in self-instructions or self-recognition). They can be intrinsic or extrin-
sic to a behavior, meaning they provide direction or motivation naturally as a task is per-
formed (as in a computer game) or they are added to the situation extrinsically in order to
improve performance. An incentive –reward program is external and extrinsic. It adds an
activator (an incentive) and a consequence (a reward) to the situation in order to direct and
motivate desirable behavior (Geller, 1996).
Instructional intervention. An instructional intervention is typically an activator or
antecedent event used to get new behavior started or to move behavior from the automatic
(habit) stage to the self-directed stage or it is used to improve behavior already in the self-
directed stage. The aim is to get the performer’s attention and instruct him or her to transi-
tion from unconscious incompetence to conscious competence. You assume the person wants
to improve, so external motivation is not needed—only external and extrinsic direction.
This type of intervention consists primarily of activators, as exemplified by education
sessions, training exercises, and directive feedback. Because your purpose is to instruct, the
intervention comes before the target behavior and focuses on helping the performer inter-
nalize your instructions. As we have all experienced, this type of intervention is more effec-
tive when the instructions are specific and given one-on-one. Role playing exercises
provide instructors opportunities to customize directions specific to individual attempts to
improve. They also allow participants the chance to receive rewarding feedback for their
improvement.
Supportive intervention. Once a person learns the right way to do something, prac-
tice is important so the behavior becomes part of a natural routine. Continued practice
leads to fluency and, in many cases, to automatic or habitual behavior. This is an especially
desirable state for safety-related behavior, but practice does not come easily and benefits
greatly from supportive intervention. We need support to reassure us we are doing the
right thing and to encourage us to keep going.
While instructional intervention consists primarily of activators, supportive inter-
vention focuses on the application of positive consequences. Thus, when we give people
rewarding feedback or recognition for particular safe behavior, we are showing our
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 168

168 Psychology of safety handbook

appreciation for their efforts and increasing the likelihood they will perform the behavior
again (Allen, 1990; Daniels, 1999; Geller, 1997). Each occurrence of the desired behavior
facilitates fluency and helps build a good habit.
Thus, after people know what to do, they need to perform the behavior correctly many
times before it can become a habit. Therefore, the positive regard we give people for their
safety-related behavior can go a long way toward facilitating fluency and a transition to the
automatic or habit stage. Such supportive intervention is often most powerful when it
comes from one’s peers—as in peer support.
Note that supportive intervention is typically not preceded by a specific activator. In
other words, when you support self-directed behavior you do not need to provide an
instructional antecedent. The person knows what to do. You do not need to activate desired
behavior with a promise (an incentive) or a threat (a disincentive). The person is already
motivated to do the right thing.
Motivational Intervention. When people know what to do and do not do it, a moti-
vational intervention is needed. In other words, when people are consciously incompetent
about safety-related behavior, they require some external encouragement or pressure to
change. Instruction alone is obviously insufficient because they are knowingly doing the
wrong thing. As I discussed earlier in Chapter 4, we refer to this as taking a calculated risk.
We usually perform calculated risks because we perceive the positive consequences of
the at-risk behavior to be more powerful than the negative consequences. This is because
the positive consequences of comfort, convenience, and efficiency are immediate and cer-
tain, while the negative consequence of at-risk behavior (such as an injury) is improbable
and seems remote. Furthermore, the safe alternative is relatively inconvenient, uncomfort-
able, or inefficient, and these negative consequences are immediate and certain. As a result,
we often need to add both activators and consequences to the situation in order to move
people from conscious incompetence to conscious competence.
This is when an incentive –reward program is useful. Such a program attempts to moti-
vate a certain target behavior by promising people a positive consequence if they perform
it. The promise is the incentive and the consequence is the reward. In safety, this kind of
motivational intervention is much less common than a disincentive–penalty program. This
is when a rule, policy, or law threatens to give people a negative consequence (a penalty) if
they fail to comply or take a calculated risk.
Often a disincentive –penalty intervention is ineffective because, like an injury, the neg-
ative consequence or penalty seems remote and improbable. The behavioral impact of
these enforcement programs is enhanced by increasing the severity of the penalty and
catching more people taking the calculated risk, but the large-scale implementation of this
kind of intervention can seem inconsistent and unfair. Because threats of punishment
appear to challenge individual freedom and choice (Skinner, 1971), this approach to behav-
ior change can backfire and activate more calculated risk taking, even sabotage, theft, or
interpersonal aggression (Sidman, 1989).
Motivational intervention is clearly the most challenging, requiring enough external
influence to get the target behavior started without triggering a desire to assert personal
freedom. Remember the objective is to motivate a transition from conscious incompetence
to a self-directed state of conscious competence.
Powerful external consequences might improve behavior only temporarily, as long as
the behavioral intervention is in place. Hence, the individual is consciously competent, but
the excessive outside control makes the behavior entirely other-directed. Excessive control
on the outside of people can limit the amount of control or self-direction they develop on
the inside.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 169

Chapter nine: Behavioral safety analysis 169

A long-term implementation of a motivational intervention, coupled with consistent

supportive intervention, can lead to a good habit. In other words, with substantial motiva-
tion and support, other-directed safe behavior can transition to unconscious competence
without first becoming self-directed. The design of effective motivational interventions is
covered in Chapter 11.

The flow of behavior change

Figure 9.9 reviews this intervention information by depicting relationships among four com-
petency states (unconscious incompetence, conscious incompetence, conscious competence,
and unconscious competence) and four intervention approaches (instructional intervention,
motivational intervention, supportive intervention, and self-management). When people
are unaware of the safe work practice (i.e., they are unconsciously incompetent), they need
repeated instructional intervention until they understand what to do. Then, as depicted at
the far left of Figure 9.9, the critical question is whether they perform the desired behavior.
If they do, the question of behavioral fluency is relevant. A fluent response becomes a habit
or part of a regular routine, and thus the individual is unconsciously competent.
When workers know how to perform a task safely but do not, they are considered con-
sciously incompetent or irresponsible. This is when an external motivational intervention
can be useful, as discussed previously. Then when the desired behavior occurs at least once,
supportive intervention is needed to get the behavior to a fluent state. Techniques for giv-
ing supportive recognition are described in Chapter 12.
Most people need supportive intervention for their safe behavior. In other words, most
experienced workers know what to do in order to prevent injury on their jobs, and they
have performed their jobs safely one or more times, but the safe way might not be habitual.

Conscious
Performer
Stage of

Competence
Other-Directed
Unconscious Conscious Unconscious
Accountable
Incompetence Incompetence Competence
Automatic Self-and Other- Automatic
Behavior Directed Conscious Behavior
At-Risk Habit Irresponsible Competence Safe Habit
Self-Directed
Responsible
Type of Intervention

Motivational Self-
Instructional Supportive
Intervention Management
Intervention Intervention
Activators and
Activators and Consequences
Activators Consequences
Consequences

NO Understand Perform Desired Behavior

Desired Desired Behavior Self-
Impact

Behavior? Behavior? Fluent? Directed?

YES YES

Figure 9.9 Awareness (conscious vs. unconscious) and safety-related behavior (compe-
tence vs. incompetence) determine which of four types of interventions is relevant.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 170

170 Psychology of safety handbook

The individual is consciously competent but needs supportive recognition or feedback for
response maintenance and increased fluency.
Figure 9.9 illustrates a distinction between conscious competence/other-directed and
conscious competence/self-directed. If a safe work practice is self-directed, the employee
is considered responsible and a self-management intervention is relevant. As detailed else-
where (Watson and Tharp, 1997), the methods and tools of effective self-management are
derived from behavioral science research and are perfectly consistent with the principles of
behavior-based safety.
In essence, self-management involves the application of the DO IT process introduced
in Chapter 8 to one’s own behavior. This means

1. Defining one or more target behavior(s) to improve.

2. Monitoring these behaviors.
3. Manipulating relevant activators and consequences to increase desired behavior
and decrease undesired behavior.
4. Tracking continual change in the target behavior(s) in order to determine the impact
of the self-management process.

See Geller (1998) and Geller and Clarke (1999) for more procedural details for safety self-
management

Accountability vs. responsibility

From the perspective of large-scale safety and health promotion, the distinction in
Figure 9.9 between accountable and responsible is critical. People often use the words
accountability and responsibility interchangeably. Whether you hold someone accountable
or responsible for getting something done, you mean the same thing. You want that person
to accomplish a certain task and you intend on making sure it happens. However, let us
consider the receiving end of this situation. How does a person feel about an assignment—
does he or she feel accountable or responsible? Here is where a difference is evident.
When you are held accountable, you are asked to reach a certain objective or goal, often
within a designated time period. However, you might not feel responsible to meet the dead-
line, or you might feel responsible enough to complete the assignment, but that is all. You do
only what is required and no more. In this case, accountability is the same as responsibility.
There are times, however, when you extend your responsibility beyond accountability.
You do more than what is required. You go beyond the call of duty as defined by a particular
accountability system. This is often essential when it comes to industrial safety and health. To
improve safety beyond the current performance plateau experienced by many companies,
workers need to extend their responsibility for safety beyond that for which they are held
accountable. They need to transition from an other-directed state to a self-directed state.
Many jobs are accomplished by a lone worker. There is no supervisor or coworker
around to hold the employee accountable for performing the job safely. The challenge for
safety professionals and corporate leaders is to build the kind of work culture that enables
or facilitates responsibility or self-accountability for safety.
An accountability system is needed that encourages personal involvement in and com-
mitment to safety. Then you will start a spiral of accountability feeding responsibility, feed-
ing more involvement and more responsibility, resulting in people becoming totally
committed to achieving an injury-free workplace. Psychological research on relationships
between environmental conditions or contingencies (as in an accountability system) and
people’s feeling states (like personal accountability or responsibility) suggests ways to
make this happen.
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 171

Chapter nine: Behavioral safety analysis 171

In conclusion
This chapter offered some basic guidelines for diagnosing the human behavior aspects of
a safety-related problem. Many situational, social, and personal factors contribute to a
behavioral discrepancy—a distinction between the behavior performed and the behavior
desired. In safety terms, this is the difference between at-risk and safe behavior.
Most of the factors contributing to a behavioral discrepancy are owing to the context
in which the task is performed or characteristics of the task itself. Common contextual
variables include

1. Unclear or misunderstood expectancies.

2. Upside-down contingencies that reward at-risk behavior or punish safe behavior.
3. The lack of behavior-based feedback to help people improve.

Often a job can be simplified or reengineered to reduce physical or mental effort, which
decreases the probability of personal injury.
Training should be considered only after critical contextual and task variables have
been analyzed and corrected. It is usually a good idea to include some education with the
training, meaning relevant theory, principles, and rationale are presented to justify the
step-by-step procedures taught and practiced. Adequate education also enables worker
customization of procedures to fit a particular work context. This, in turn, leads to
employee ownership of the process, feelings of responsibility, and increased involvement.
Some training is required to keep people in practice for handling a relatively rare event
(as in emergency training), while other training is needed to help people change frequently
occurring at-risk behavior to safe behavior. Then there is the training needed to introduce
a new procedure or process. Each of these training situations requires behavior-based feed-
back, but obviously the situation and the individuals involved determine the protocol for
delivering the feedback. This is one more analysis challenge.
Education and training reflect an instructional approach to corrective action. This type
of intervention is obviously most effective when the participants are willing to learn. They
are unaware of the correct procedures and are “unconsciously incompetent.” Instruction
will not help much for people who know what to do but do not do it. These individuals are
“consciously incompetent” and need a motivational intervention, as discussed later in
Chapter 12.
For most employees, the issue is not a matter of knowing what is safe. They periodi-
cally perform all of the safe operating procedures called for on the job. The problem is con-
sistency or fluency. They do not follow the safe protocol every time. These people need
supportive intervention to keep them safe.
When safe work practices are relatively convenient, like putting on PPE or buckling a
safety belt, the behavior can become habitual. When such behavior becomes a natural part
of the work routine, the participant is considered “unconsciously competent.” However,
some behaviors, like locking out a power source, are relatively complex and never reach
the automatic stage. Regular supportive intervention is often needed to keep these incon-
venient behaviors going, unless the individual is self-directed with regard to the particular
behavior.
Self-directed individuals hold themselves accountable for doing the right thing, even
when the behavior is relatively uncomfortable and inconvenient. These people certainly
appreciate supportive intervention from managers, friends, and coworkers, but they keep
performing the safe behavior when no one is around to support them. These self-directed
workers hold themselves accountable. They feel responsible and go beyond the call of duty
820049_CRC1_L1540_CH09 11/10/00 1:43 PM Page 172

172 Psychology of safety handbook

to prevent injuries to themselves and others. I call this “actively caring”—the focus
throughout Section 5 of this Handbook. Chapter 16, in particular, presents principles and
methods to help people transition from being held accountable to feeling responsible for
safety.
The bottom line Before selecting an intervention strategy, conduct a careful analysis
of the situation, the behavior, and the individuals involved in an observed discrepancy
between desired and actual performance. Do not impulsively assume corrective action
requires “training” or “discipline.” A behavioral safety analysis will likely give priority to
a number of alternative intervention approaches. Performing such an analysis before inter-
vening will help ensure your corrective action plan does not reflect malpractice.

References
Allen, J., I Saw What You Did and I Know Who You Are: Bloopers, Blunders and Success Stories on Giving
and Receiving Recognition, Performance Management Publications, Tucker, GA, 1990.
Daniels, A. C., Bringing out the Best in People: How to Apply the Astonishing Power of Positive
Reinforcement, 2nd ed., McGraw-Hill, New York, 1999.
Deming, W. E., Quality, productivity, and competitive position, four-day workshop presented by
Quality Enhancement Seminars, Inc., Cincinnati, OH, May 1991.
Deming, W. E., Quality concepts to solve societal crises: profound knowledge for psychologists,
invited address at the Centennial Convention of the American Psychological Association,
Washington, D.C., August 1992.
DePasquale, J. P. and Geller, E. S., Critical success factors for behavior-based safety: a study of
twenty industry-wide applications, J. Saf. Res., 30, 237, 1999.
Geller, E. S., The truth about safety incentives, Prof. Saf., 41(10), 34, 1996.
Geller, E. S., Key processes for continuous safety improvement: behavior-based recognition and cel-
ebration, Prof. Saf., 42(10), 40, 1997.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998.
Geller, E. S., Do you coach with feeling?, Ind. Saf. Hyg. News, 34(3), 16, 2000.
Geller, E. S. and Clarke, S. W., Safety self-management: a key behavior-based process for injury pre-
vention, Prof. Saf., 44(7), 29, 1999.
Geller, E. S., Boyce, T. E., Williams, J., Pettinger, C., DePasquale, J., and Clarke, S., Researching
behavior-based safety: a multi-method assessment and evaluation, in Proceedings of the 37th
Annual Professional Development Conference and Exposition, American Society of Safety Engineers,
Des Plaines, IL, 1998a.
Geller, E. S., DePasquale, J., Pettinger, C., and Williams, J., Critical success factors for behavior-
based safety, in Proceedings of Light Up Safety in the New Millennium: a Behavioral Safety
Symposium, American Society of Safety Engineers, Des Plaines, IL, 1998b.
Langer, E. J., The Power of Mindful Learning, Perseus Books, Reading, MA, 1997.
Mager, R. F. and Pipe, P., Analyzing Performance Problems or You Really Oughta Wanna, 3rd. ed., The
Center for Effective Performance, Inc., Atlanta, GA, 1997.
Norman, D. A., The Psychology of Everyday Things, Basic Books, New York, 1998.
Schlenker, B., Impression Management: the Self-Concept, Social Identity, and Interpersonal Relations,
Brooks/Cole, Monterey, CA, 1980.
Sidman, M., Coercion and Its Fallout, Authors Cooperative, Inc., Publishers, Boston, MA, 1989.
Skinner, B. F., Beyond Freedom and Dignity, Alfred A. Knopt, New York, 1971.
Watson, D. L. and Tharp, R. G., Self-directed Behavior: Self-Modification for Personal Adjustment, 7th
ed., Brooks/Cole Publishing, Pacific Grove, CA, 1977.
Williams, J. H. and Geller, E. S., Behavior-based intervention for occupational safety: critical impact
of social comparison feedback, J. of Saf. Res., 31(3), 135, 2000.
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 173

section four

Behavior-based intervention
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 174
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 175

chapter ten

Intervening with activators

Intervention techniques to increase safe behaviors or decrease at-risk behaviors are either activators or
consequences. This chapter explains activators, with real-world examples showing how to develop effec-
tive strategies. This discussion is framed by six principles for maximizing the impact of activators.

“Best efforts are not enough, you have to know what to do.”—W. Edwards Deming

In Chapter 9, I showed how the Activator–Behavior–Consequence (ABC) model can be

used to diagnose the contributing factors to an incident or at-risk behavior and to decide
on a plan for corrective action. With this chapter, we begin our discussion of interven-
tion design and implementation to improve safety-related behavior. As such, the ABC
model is used as introduced in Chapter 8—as a framework for designing behavior-change
interventions.
Psychologists who use the behavior-based approach to solve human problems design
activators (conditions or events preceding operant behavior) and consequences (conditions
or events following operant behavior) to increase the probability that desired behaviors
will occur and undesired behaviors will not. Activators precede and direct behavior.
Consequences follow and motivate behavior. This chapter explains basic principles about
activators to help you design interventions for increasing safe behavior and decreasing at-
risk behavior. The next chapter focuses on the use of consequences to motivate safety
achievement.
First, let me reiterate the need for safety interventions. As I have said before, main-
taining our own safe behavior is not easy. It is usually one long fight with human nature,
because in most situations activators and consequences naturally support risky behavior
in lieu of safe behavior. At-risk behavior often allows for more immediate fun, comfort,
and convenience than safe behavior, prompting the need for special intervention to
direct and motivate safe behavior. Activators are generally much easier and less expensive
to use than consequences, so it is not surprising that they are employed much more
often to promote safe behavior. Posters or signs are perhaps the most popular activators
for safety.

• Some bear only a general message—“Safety is a Condition of Employment”; others

refer to a specific behavior—”Hard Hat Required in this Area.”
• Some signs request the occurrence of a behavior—“Walk,” “Wear Ear Plugs in This
Area”; others want you to avoid a certain behavior—“Don’t Walk,” “No Smoking
Area.”
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 176

176 Psychology of safety handbook

• Sometimes a relatively convenient response is requested—“Buckle-Up,” while other

signs prompt relatively inconvenient behaviors—“Lock Out All Energy Sources
Before Repairing Equipment.”
• Some signs imply consequences—“Use Eye Protection: Don’t be blinded by the
light”; others do not—“Wear Safety Goggles.”
• We might be reminded of a general purpose—“Actively Care for a Total Safety
Culture,” or challenged—“100 Percent Safe Behavior is Our Goal This Year.”

I have visited a number of work environments where all of these types of safety signs
were displayed. In fact, I have seen situations that make the illustration in Figure 10.1 seem
not very far fetched. Does this sort of “over-kill” work to change behavior and reduce
injuries? If you answered “yes,” then this time your common sense was correct, because
you have been there and experienced the ineffectiveness of many safety signs.
Which signs would you eliminate from Figure 10.1? How would you change certain
signs to increase their impact? What activator strategies would you use instead of the
signs? This chapter will enable you to answer these questions—not on the basis of common
sense but from behavioral science research.
Let us consider six key principles for increasing the impact of activators. They are

• Specify behavior.
• Maintain salience with novelty.

Figure 10.1 Safety activators can be overwhelming and ineffective.

820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 177

Chapter ten: Intervening with activators 177

• Vary the message.

• Involve the target audience.
• Activate close to response opportunity.
• Implicate consequences.

Each of these principles is illustrated below with the help of some real-world examples.

Principle #1: Specify behavior

Behavioral research demonstrates that signs with general messages and no specification of
a desired behavior to perform (or an undesirable behavior to avoid) have very little impact
on actual behavior. However, signs that refer to a specific behavior can be beneficial.
For example, my students and I conducted several field experiments in the 1970s on
the behavioral effects of environmental protection messages. In one series of studies, we
gave incoming customers of grocery stores promotional flyers which included

1. A general antilitter message (“Please don’t litter. Please dispose of properly”).

2. No environmental protection message.
3. A specific behavioral request (“Please deposit in green trash can in rear of store”).

Later we searched the stores for our flyers and measured the impact of the different
instructions.
Our findings were consistent over several weeks, across different stores, and with dif-
ferent research designs (Geller et al., 1976, 1977). There were three useful conclusions. The
general antilitter message was no more effective than no message (the control condition) in
reducing littering or in getting flyers deposited in trash receptacles. In contrast, patrons
receiving the flyers with the specific behavioral request were significantly less likely to lit-
ter the store, and 20 to 30 percent of these flyers were deposited in the “green trash recep-
tacle.” In addition, a message that gave a rationale for the behavioral request “Please help
us recycle by depositing in green trash can in rear of store” was even more effective at
directing the desired behavior.
Our research on the importance of response specificity in activator interventions has
been replicated in other environmental protection research and in a few safety-belt pro-
motion studies. For example, specific response messages reduced littering in a movie the-
ater (Geller, 1975), increased the purchase of drinks in returnable bottles (Geller et al., 1973),
directed occupants in public buildings to turn off room lights when leaving the room
(Delprata, 1977; Winett, 1978), and reminded vehicle occupants to buckle up (Berry et al.,
1992; Geller et al., 1985; Thyer and Geller, 1987). As you will see, the activators in these
studies had characteristics besides response specificity to help make them effective.
Figure 10.2 illustrates “explosively” the need to include sufficient response informa-
tion with a behavioral request, but too much specificity can bury a message, as illustrated
in Figure 10.3. Activators ought to specify a desired response, but not overwhelm with
complexity, as I have seen in a number of industrial signs. Overly complex signs are easy
to overlook—with time they just blend into the woodwork. Keeping signs salient or notice-
able is clearly a challenge.

Principle #2: Maintain salience with novelty

All of the field research demonstrating the impact of response-specific signs was relatively
short term. None of the projects lasted more than a few months. The activators were salient
because they were different or novel. Customers rarely receive flyers when they enter
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 178

178 Psychology of safety handbook

Figure 10.2 Some activators are not specific enough.

grocery stores, so the flyers and their messages in our litter control and recycling research
were quite novel and salient. Plus, most customers did not see the messages every day,
because they rarely shopped more than once a week. Similarly, the subjects in the studies
that showed effects of safety belt messages on dashboard stickers (Thyer and Geller, 1987)
and on flash cards (Geller et al., 1985; Thyer et al., 1987) were exposed to the message only
once, or on average less than once a day.

Habituation
It is perfectly natural for activators like sign messages to lose their impact over time. This
process is called habituation, and it is considered by some psychologists to be the simplest
form of learning (Carlson, 1993). Through habituation we learn not to respond to an event
that occurs repeatedly.
Habituation happens even among organisms with primitive nervous systems. For
example, when you lightly tap the shell of a large snail it withdraws into its shell. After
about 30 seconds the snail will extend its body from the shell and continue on its way.
When you tap the shell again, the snail will withdraw again. However, this time the snail
will stay inside its shell for a shorter duration. Your third tap will cause withdrawal again,
but the withdrawal time will be even shorter. Each tap on the snail’s shell results in suc-
cessively shorter withdrawal time until eventually the snail will stop responding at all to
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 179

Chapter ten: Intervening with activators 179

Figure 10.3 Some signs are too complex to be effective.

your tap. The snail’s behavior of withdrawal to the activator—shell tapping—will have
habituated.
Habituation is perfectly consistent with an evolutionary perspective (Carlson, 1993). If
there is no obvious consequence (good or bad) from responding to a stimulus, the organ-
ism, be it an employee or a snail, stops reacting to it. It is a waste of time and energy to con-
tinue responding to an activator that seems to be insignificant. What would a snail do in a
rain storm if it did not learn to ignore shell taps that have no consequence?
Consider the distractions and distress you would experience daily if you could not
learn to ignore noises from voices, radios, traffic, and machinery. At first these environ-
mental sounds might be quite noticeable and perhaps distracting, but through habituation
they become insignificant background noise. They no longer divert attention nor interfere
with ongoing performance.
I have heard of a much more dramatic illustration of habituation that I want you to
only imagine. Please do not try this. If you were to take a frog and drop it in boiling water,
it would react immediately, leaping out to safety. However, if you put a frog in cold water
and slowly raise the heat over several hours, the frog will not jump out but eventually cook
in the boiling water. I confess I have not witnessed this myself nor read it in a scientific jour-
nal, but it does sound plausible, given the basic learning principle of habituation. Have you
not seen, for example, seemingly impossible situations of noise, heat, or squalor which peo-
ple seem to adjust to over time?
What is the relevance of habituation for safety? It is human nature to habituate to
everyday activators in our environment that are not supported by consequences. This is the
case with many safety activators. Staying attentive to safety activators is a continuous fight
with one aspect of human nature—habituation.
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 180

180 Psychology of safety handbook

Safety-belt reminders. Safety-belt reminders are a lesson in how easy it is to ignore

activators. How often do you notice the audible safety-belt reminder in your vehicle?
I have found many people unable to describe the sound. Is it a continuous tone, a beeping
signal, or a pleasant chime? How long does it last? Is it the same sound used for other warn-
ing signals in your vehicle?
Current safety-belt reminders in vehicles sold in the United States last from four to
eight seconds, as mandated by the National Highway Traffic Safety Administration. Do
they work? I suggest they are mostly ineffective (Geller, 1988), primarily owing to lack of
salience. The sound is usually the same for all warning signals in a vehicle. Not only do we
habituate to this sound, but it must compete for our attention with the background noise
initiating at the same time, from the roar of the engine, air conditioner, or heater to music
from a blaring radio. Even if safety-belt reminders were clearly audible, would they
increase safety-belt use? Why should a four- to eight-second audible reminder activate peo-
ple to buckle up?
Perhaps you never hear this reminder because you buckle up before turning on the
ignition. According to our field research (von Buseck and Geller, 1984), about half of the
drivers fasten their safety belt after turning on their ignition. Because the reminder starts
upon ignition, there is no opportunity for these drivers to buckle up and avoid the reminder.
When they buckle up, they are merely escaping the unpleasant activator. What if buckling
up enabled a driver to avoid a reminder that was salient and somewhat unpleasant?
In 1987, General Motors Research Laboratories loaned me a 1984 Cadillac Seville to
answer these important questions. This experimental vehicle was programmed to provide
any of the following reminder systems.

1. A standard six-second buzzer or chime triggered by engine ignition.

2. A six-second buzzer or chime that initiated five seconds after ignition.
3. A voice reminder (“Please fasten your safety belt”) that initiated five seconds after
engine ignition and was followed by a “Thank you” if the driver buckled up.
4. A second reminder option where the six-second buzzer, chime, or verbal prompt
kicked in if the driver was not buckled when the vehicle made its first stop after
exceeding ten miles per hour. This special vehicle had a portable computer in its
trunk to record each instance of belt use by the driver.

My students and I studied the impact of these different reminder systems by having
college students drive the experimental vehicle on a planned community course under the
auspices of an energy conservation study. We asked subjects to stop and park the vehicle at
six specific locations along this two-mile course and flip a toggle switch in the vehicle’s
trunk (presumably to record information on gasoline use). This gave the driver six oppor-
tunities to buckle up during a one-hour experimental session. Each subject returned peri-
odically to participate in this so-called “energy conservation study.” The number of days
between sessions varied from one to five. Results were examined on an individual basis in
order to study systematically the impact of a particular reminder system.
Our findings indicated that the more salient signals, especially the vocal reminder,
increased safety-belt use. A reminder signal had maximum impact if it could be avoided by
buckling up (Berry and Geller, 1991; Geller, 1988). General Motors applied some of our
findings in an innovative safety-belt reminder system for its line of Saturn vehicles. All
Saturns have an airbag, automatic shoulder belt, and a manual lap belt. The innovative
safety-belt reminder in the Saturn cannot be masked by other vehicle start-up noises
because it sounds six seconds after these noises have initiated. Most important, this
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 181

Chapter ten: Intervening with activators 181

activator will not occur if the driver buckles the lap belt within that six-second window.
Thus, Saturn drivers can buckle up after turning their ignition switch and avoid the safety-
belt reminder.
I do not believe anyone has systematically evaluated whether the Saturn activator for
lapbelt use is more effective than the standard system. There is plenty of evidence, how-
ever, that the lap belt offers optimal protection from vehicle ejection and fatalities (Evans,
1991). Moreover, field observations have revealed a decrease in lap-belt use in vehicles with
automatic shoulder belts (Williams et al., 1989). If any safety-belt reminder system can
increase the use of vehicle lap belts, it will be the Saturn activator because it is based on the-
ory and procedures developed from behavioral science research.
You can see how understanding some basic principles from behavioral science can
improve the design of simple activators like safety-belt reminders. Such knowledge could
also improve or alter public policy, as illustrated in the next example.
Radar detectors. A radar detector is a very effective activator to reduce speeding.
Why? Because it is consistent with the behavioral science principles discussed here. The
sound of this activator is distinct, so the user will not confuse this signal with other vehicle
sounds. More important, this activator is linked to a particular negative consequence. If a
driver is speeding and ignores this activator, a speeding ticket is possible, even likely.
Finally, this activator is voluntarily purchased by the drivers who tend to speed. The pur-
chaser is receptive to the information provided by this activator.
I find it quite disappointing that this activator is outlawed in my home state of Virginia.
If policy makers understood some basics of behavior-based safety and if they truly wanted
to reduce excessive speeding, they would not only allow radar detectors but they would
encourage their dissemination. Drivers who like to speed would purchase them, and read-
ily slow down following this activator’s distinct signal. Traffic enforcement agencies could
saturate risky areas (metropolitan loops or bypasses) with radar devices and monitor every
fifth or tenth device. Speeders would never know which signal was “real” and would thus
reduce their speeding to avoid a negative consequence.
A basic behavioral science principle, supported by substantial research (Chance, 1994;
Kimble, 1961), is that consequences occurring on an intermittent basis are much more effec-
tive at motivating long-term behavior change than consequences occurring on a continuous
basis or after every response. When consequences are improbable, as is currently the case
for receiving a speeding ticket, they can lose their influence entirely. The increased use of
radar detectors and the strategic placement of staffed and unstaffed radar devices would
make consequences for speeding more salient and immediate for those who most need con-
trol on their at-risk driving. I hope readers will teach policy makers and police officers these
basic principles about activators and consequences whenever they have the opportunity.

Warning beepers: a common work example

Figure 10.4 illustrates quite clearly the phenomenon of habituation and reduced activator
salience with experience. I bet you can reflect on personal experiences quite similar to the
one shown here. Not only has the brick mason habituated to the familiar “beep” of the
backing vehicle, but the driver is illustrating danger compensation (or risk homeostasis) as
I discussed in Chapter 6. He is not looking over his shoulder to check for a potential colli-
sion victim. He assumes the warning beeper is sufficient to activate coworkers’ avoidance
behavior and prevent injury. This particular activator has actually reduced the driver’s per-
ceived risk. This influences his at-risk behavior of looking forward instead of turning his
head to check his blind spots. A key point is that understanding the basic learning
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 182

182 Psychology of safety handbook

Figure 10.4 Some signals we rely on lose impact over time.

phenomenon of habituation can prevent overreliance on activators and support a need to

work more defensively.

Principle #3: Vary the message

What does habituation tell us about the design of safety activators? Essentially, we need to
vary the message. When an activator changes it can become more salient and noticeable.
The “safety share” discussed in Chapter 7 follows this principle. When participants in a
group meeting are asked to share something they have done for safety since the last meet-
ing, the examples will vary considerably. Similarly, group discussions of near hits and
potential corrective actions will also vary dramatically. The messages from safety shares
and near-hit discussions are also salient because they are personal, genuine, and distinct.

Changeable signs
Over the years I have noticed a variety of techniques for changing the message on safety
signs. There are removable slats to place different messages. I am sure most of you have
seen computer-generated signs with an infinite variety of safety messages. Some plants
even have video screens in break rooms, lunchrooms, visitor lounges, and hallways that
display many kinds of safety messages, conveniently controlled by user friendly computer
software.
Who determines the content of these messages? I know who should—the target audi-
ence for these signs. The same people expected to follow the specific behavioral advice
should have as much input as possible in defining message content. Many organizations
can get suggestions for safety messages just by asking. But if employees are not accustomed
to giving safety suggestions, they might need a positive consequence to motivate their
input.
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 183

Chapter ten: Intervening with activators 183

Figure 10.5 In 1986, Ford employees created buckle-up activators for display at Ford
World Headquarters.

Worker-designed safety slogans

This is what I am talking about. In 1985, employees and visitors driving into the main park-
ing lot for Ford World Headquarters in Dearborn, MI, passed a series of four signs arranged
with sequential messages, like the old Burma Shave signs. The messages were rotated peri-
odically from a pool of 55 employee entries in a limerick contest for safety-belt promotion.
My three favorites are illustrated in Figure 10.5.
Notice that the last sign in each series of safety-belt promotion messages at Ford World
Headquarters includes the name and department of the author. This public recognition,
with the author’s permission, of course, provides a positive consequence or reward to the
participant. It also reminds all sign viewers that many different people from various work
areas are actively involved in safety. Through positive recognition and observational learn-
ing, including vicarious reinforcement, this simple technique promotes ownership and
involvement in a safety process. This leads to the next principle. Involve the target audience.

Principle #4: Involve the target audience

This guideline is probably obvious by now. It is relevant for developing and implementing
any behavior-change intervention. When people contribute to a safety effort, their owner-
ship of and commitment to safety increase. Of course, this principle works both ways.
When individuals feel a greater sense of ownership and commitment, their involvement in
safety achievement is more likely to continue. Thus, involvement feeds ownership and
commitment, and vice versa.
The simple activator in Figure 10.6 illustrates an ownership-involvement connection
most of you can relate to, and it is a practical intervention strategy for many situations. The
name plate in Figure 10.6 would not have to be as obtrusive in a real-world application to
increase the perception of ownership. This would probably lead to a person taking greater
care of the equipment, including more attention to safety-related matters.
This ownership-involvement principle is supported by litter control research that
found much more littering and vandalism in public than private places (Ley and
Cybriwsky, 1974; Newman, 1972). When public trash receptacles include the logos of
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 184

184 Psychology of safety handbook

Figure 10.6 Some activators imply ownership and increase actively caring.

nearby businesses, the merchants whose logos are displayed typically take care of the
receptacle and keep the surrounding area clean. This principle is also supported by the suc-
cess of “Adopt-a-Highway” programs that have groups keep a certain roadway clear of lit-
ter and perhaps beautify with plants, scrubs, or flowers. Group ownership of a roadway
typically leads to actively caring for its appearance.
I would like to share three effective activator interventions I have used to involve a tar-
get audience. All three made use of hand-held cards with safety messages.

Safe behavior promise

First, in the mid-1980s, at a time when states did not have seat-belt laws and the use of vehi-
cle safety belts in the United States was below 15 percent, companies including General
Motors; Ford; Corning Glass in Blacksburg, VA; Burroughs Welcome in Greenville, NC;
and the Reeves Brothers Curon Plant in Cornelius, NC, more than doubled the use of safety
belts in company and private vehicles through “Buckle-Up Promise Cards” that employ-
ees were encouraged to sign (Geller and Lehman, 1991). Most of these cards were distrib-
uted after a lecture or group discussion about the value of using vehicle safety belts (Cope
et al., 1986; Geller and Bigelow, 1984; Kello et al., 1988).
My students and I have also distributed “Buckle-Up Promise Cards” during church
services (Talton, 1984), throughout a large university campus (Geller et al., 1989), and at the
Norfolk Naval Base (Kalsher et al., 1989). In every case, a significant number of pledge-card
signers increased their use of safety belts after their initial commitment behavior.
This simple activator approach also has had remarkable success in applications beyond
safety-belt promotion. Streff et al. (1993) found the technique successful at increasing
the use of safety glasses. More recently, Boyce and Geller (in press) used this promise card
technique to motivate university students to recognize the altruistic or helping behavior
of others.
Dr. Richard Katzev and his colleagues at Reed College in Portland, OR, have used this
activator to increase participation in community recycling programs (Katzev and Pardini,
1987 –1988; Wang and Katzev, 1990). Work teams at Logan Aluminum in Russellville, KY,
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 185

Chapter ten: Intervening with activators 185

instituted a “Public Safety Declaration” that had employees sign a poster at the plant
entrance specifying a safe-behavior commitment for the day—for example, “We wear hear-
ing protection in all designated areas.” Salience was maintained by changing the behav-
ioral target in the public commitment message weekly.
Figure 10.7 depicts a sample promise card for involving people in a commitment to
perform a particular behavior. The target behavior to increase in frequency could be
selected by a safety director, group leader, or through a group consensus discussion. This
behavior is written on the promise card, perhaps by each individual in a group. Group
members decide on the duration of the promise period and write the end date on the card.
Then each group member should be encouraged, not coerced, to sign and date a card. I
have found this group application of the safe behavior promise strengthens a sense of
group cohesion or belonging. Follow these procedural points for optimal results.

• Define the desired target behavior specifically.

• Involve the group in discussing the personal and group value of the target behavior.
• Make the commitment for a specified period of time that is challenging but not over-
whelming.
• Assure everyone that signing the card is only a personal commitment, not a com-
pany contract.
• There should be no penalties (not even criticism) for breaking a promise.
• Encourage everyone to sign the card, but do not use pressure tactics.
• Signers should keep their promise cards in their possession, or post them in their
work areas as reminders.

The more involvement and personal choice solicited during the completion of this acti-
vator strategy, the better each individual feels about the process. Personal commitment to
perform a specific behavior is activated as a result; those involved in the process should feel
obligated to fulfill the promise. Signing the card publicly in a group meeting also implicates
social consequences to motivate compliance. That is, many participants will be motivated
to keep their promise to avoid disapproval from a group member. When individuals keep
their promise, recognition and approval from the group reinforces and supports mainte-
nance of the targeted safe behavior.

e Card
vior Promis
Safe Beha

to
I promise

until
From
date

signature

Figure 10.7 A promise card activates a behavioral commitment.

820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 186

186 Psychology of safety handbook

The “Flash for Life”

The second activator intervention I want to relate dates back to 1984, when I developed the
“Flash for Life.” Here is how it worked. A person displayed to vehicle occupants the front
side of an 11- by 14-inch flash card that read, “Please Buckle-up—I Care.” If someone buck-
led up after viewing this message, the “flasher” flipped the card over to display the bold
words, “Thank You For Buckling Up.” For the first evaluation of this behavior change inter-
vention (Geller et al., 1985), the “flasher” was in the front seat of a stopped vehicle and the
“flashee” was the driver of an adjacent, stopped vehicle. The flash card was shown to 1087
unbuckled drivers, and of the 82 percent who looked at the card, 22 percent complied
immediately with the buckle-up request.
My youngest daughter, Karly, was the “flasher” for about 30 percent of the trials in the
study. As shown in Figure 10.8, Karly was only three and one-half years of age at the time.
On a few occasions we got a hand signal that was not used to indicate a right or left turn.
Once Karly asked: “Daddy, what does that mean?” and I answered, “It means you’re num-
ber one, honey, they are just using the wrong finger.”
When hearing about this “Flash for Life” project, many of my colleagues expressed
concern for my sanity. “Why do you waste your time?” some would say. “Getting 22 per-
cent to buckle up is not a big deal, and most of those who buckled up for your daughter
only did it the one time. They probably won’t buckle up the next day.”
I had two answers to this sort of pessimism. First, achievement is built on “small wins.”
People need to break up big problems or challenges into small, achievable steps, and then
work on each successive step, one at a time (Weick, 1984). We cannot expect to solve a major
safety problem like low use of personal protective equipment with one intervention tech-
nique, but we need to start somewhere. If everyone contributed a “small win” for safety,
the cumulative effects could be tremendous.
My second reply focused on the powerful influence of involvement. This intervention
procedure enabled my young daughter to get involved in a safety project, even though she
did not yet understand the concept of “safety.” Every time she “flashed” another person to
buckle up, her own commitment to practice the target behavior increased. I have never had
to remind her to use her safety belt. Actually, she has reminded me to buckle up, and
often monitors my driving speed. Now, I never taught her about speed limits, but her
early involvement in safety-belt promotion generalized to caring about other safety-
related behaviors. This was the real long-term benefit of involving Karly as a “Flash for
Life” activator.
You see, people who actively care for safety by encouraging—or activating—others to
practice safe behaviors strengthen their own personal safety commitment. When Karly was
in fourth grade she won a speech contest for a talk on her “flashing” experiences at age
three and one-half. Her early involvement in safety led to this later role as a safety teacher,
further strengthening her personal commitment to practice safe behaviors.
Thyer and colleagues (1987) demonstrated the benefits of another application of the
“Flash for Life” intervention by posting college students at campus parking lot
entrance/exit areas and asking them to “flash” vehicle occupants. Mean safety-belt use by
vehicle drivers increased from 19.5 percent (n  629) during an initial one-week baseline,
to 45.5 percent (n  635) during a subsequent week of daily flashing. The intervention was
withdrawn during the third week, and average belt use decreased to 28.5 percent (n  634).
When reinstating the intervention during the fourth week, the researchers observed a
prominent increase in mean belt use to 51.5 percent (n  625).
A follow-up study (Berry et al., 1992) showed that this activator had a substantially
greater impact when a person held the buckle-up sign, as opposed to the sign being
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 187

Chapter ten: Intervening with activators 187

Figure 10.8 Top: My daughter Karly “flashes” drivers to buckle up in 1984. Bottom: When
the driver buckles up, Karly flips over the flash card to give a positive consequence.

attached to the stop sign by the exit. At a few industrial sites, notably the Hanford Nuclear
site in Richland, WA, employees have implemented this activator intervention in their
parking lots. Vehicle occupants typically gave a smile or “thumbs-up” sign of approval
when they saw their coworkers “flashing for safety.” This rewarded the participants for
their involvement and increased the probability of their future participation in a safety
project.
In another variation, Roberts and his students (1990) disseminated vinyl folders with
the “Flash for Life” messages on front and back to 10,000 school children. They observed
children “flashing” throughout the community and found higher rates of safety-belt use among
children who received the flash card. Again, this points out the power of involvement.
820049_CRC1_L1540_CH10 11/10/00 1:44 PM Page 188

188 Psychology of safety handbook

I have personally distributed more than 3000 “Flash for Life” cards nationwide, usu-
ally upon request by an individual who heard about the intervention procedure. In addi-
tion, a number of safety-belt groups in Ohio, Tennessee, and Virginia have personalized the
flash card for distribution and use throughout their states. I have heard numerous “small
win” success stories from recipients of this “Flash for Life” activator.

The Airline Lifesaver

Now, my third personal experience with an activator intervention is one I have used since
November 1984, whenever boarding a commercial airplane (Geller, 1989a, b). I hand the
flight attendant a 3- by 5-inch “Airline Lifesaver” card like the one depicted in Figure 10.9.
The card indicates that airlines have been the most effective promoters of seat-belt use and
requests that someone in the flight crew make an announcement near the flight’s end to
activate safety-belt use in personal vehicles.
From November 1984 to January 1993, I distributed the “Airline Lifesaver” on 492
flights, and on 36 percent of these occasions the flight attendant gave a public buckle-up
reminder. In the period from March 1994 to February 1995, I gave the “Airline Lifesaver”
to 118 flight attendants and heard a buckle-up reminder on 54 percent of these flights.
Figure 10.10 depicts a graph of these data collected over a decade of field observations.
Perhaps you are wondering why I separated the two time periods when reporting
the preceding results and what could account for the significantly higher announcement
percentages during the second time period. Well, I used different Airline Lifesaver cards

The Airline Lifesaver

Airlines have been exemplary promoters of seat belt use.

Please, at the end of the trip would someone in
your flight crew announce the buckle-up reminder
below.
This announcement will show that your airline cares
about transportation safety. And who knows.. you might
save a life!
For important information, turn this card over.

Now that you have worn a

seat belt for the safest part of your trip...
...the flight crew would like to remind you to
buckle-up during your ground
transportation!!

00
Buckle Up PS

Figure 10.9 I use the Airline Lifesaver Card to activate a buckle-up reminder on airplanes.
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 189

Chapter ten: Intervening with activators 189

Figure 10.10 The percentage of compliance with the Airline Lifesaver request was higher
when a reward was offered.

during these periods. As shown in Figure 10.9, the cards distributed during the first phase
merely requested the buckle-up announcement; cards used during the second phase
offered prizes valued from $5 to $30 if the buckle-up reminder was given. See Figure 10.11
for an illustration of the back of this incentive card. If the announcement is made, I give the
attendant a postcard to mail to my office in order to redeem a reward. To date, of the atten-
dants who received this reward opportunity, 65 percent stamped and mailed the postcard
and they received a prize.
As with the “Flash for Life” activator, many friends have laughed at the “Airline
Lifesaver,” claiming I am wasting my time. A common comment was “No one listens to the
airline announcements anyway, and besides, do you really think an airline message could
be enough to motivate people to buckle up if they don’t already?”
Consider this personal experience from the mid-1980s. I observed a woman approach
the driver of an airport shuttle, asking her to “Please use your safety belt.” The driver
immediately buckled up. When I thanked the woman for making the buckle-up request,
she replied that she normally would not be so assertive but she had just heard a buckle-up
reminder on her flight, “and if a stewardess can request safety-belt use, so can I.”
Except for a few anecdotes like this one, it is impossible to assess the direct buckle-up
influence of the Airline Lifesaver. However, it is “safe” to assume that the beneficial, large-
scale impact of this activator is a direct function of the number of individuals who deliver
the reminder card to airline personnel. It is encouraging that several large corporations,
including Ford Motor Company; Tennessee Valley Authority; and Air Products and
Chemicals of Allentown, PA, have distributed Airline Lifesaver cards to their employees
for their own use during air travel. If the delivery of an Airline Lifesaver does not influence
a single airline passenger to use a safety belt during ground transportation, at least the act
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 190

190 Psychology of safety handbook

By reading the
"buckle-up reminder,"
you are actively caring
for the safety and
health of others.

For more information call (703) 231-8145

The Center for Applied Behavior Systems

Figure 10.11 The back of the new Airline Lifesaver card offers a reward for giving the
reminder.

of handing an Airline Lifesaver card to another person should increase the card deliverer’s
commitment to personal safety-belt use.
Of course, the primary purpose of getting involved in a safety intervention is to pre-
vent injury or improve a person’s quality of life. Unfortunately, we rarely see these most
important consequences. Thus, we need motivation, feedback, interpersonal approval, and
self-talk. We tell ourselves the safe behavior is “the right thing to do,” and that someday an
injury will be prevented. We cannot count the number of injuries we prevent; we just need
to “keep the faith.”
On December 28, 1994, I received a special letter from Steven Boydston, then assistant
vice president of Alexander & Alexander of Texas, Inc., which helps me “keep the faith”
that the “Airline Lifesaver” makes a difference. The encouraging words in this letter are
repeated in Figure 10.12. This success story is itself an activator for such proactive inter-
ventions as the Airline Lifesaver. It sure worked for me.

Principle #5: Activate close to response opportunity

Note that most of the effective activators discussed so far occurred at the time and place the
target behavior should happen. The litter-control messages were on the flyers that needed
disposal, the sign requesting lights be turned off was below the light switch, the “Flash for
Life” card was presented when people were in their vehicles and could readily buckle up,
and I give airline attendants the “Airline Lifesaver” card when boarding the plane.
Actually, I believe I would get more compliance with the request for a buckle-up announce-
ment if I handed an attendant the announcement card at the end of the flight—closer to the
opportunity to make the requested response. In fact, when I inquired about the lack of a
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 191

Chapter ten: Intervening with activators 191

Figure 10.12 I received these words of encouragement from Steven Boydston on

December 28, 1994.

buckle-up announcement while deplaning, some flight attendants tell me they forgot about
the card.

Point-of-purchase activators
In one study, my students and I systematically evaluated the impact of proximity between
activator and response opportunity. We distributed handbills prompting the purchase of
returnable drink containers at the entrance to a large grocery store or at the store location
where drinks can be picked up for purchase (Geller et al., 1971, 1973). As predicted, cus-
tomers purchased significantly more drinks in returnable than throwaway containers
when prompted at the point of purchase. This “point-of-purchase advertising” is pre-
sumed to be an optimal form of product marketing (Tilman and Kirpatrick, 1972).

Activating with television

You would think that product ad activators on television are less effective in directing
behavior than promotions at store locations. Similarly, it is reasonable to predict that pro-
moting vehicle safety-belt use on television would be less effective than presenting buckle-
up activators at road locations, as exemplified by the “Flash-for-Life” intervention. This
assumption is supported by the classic and rigorous evaluation of safety-belt promotion in
public service announcements on television by Robertson et al. (1974).
In this study, six different safety belt messages were shown during the day and during
prime time on one cable of a dual-cable television system. Residents in Cable System A
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 192

192 Psychology of safety handbook

(6400 homes) received the safety belt messages 943 times over a nine-month period. Each
viewer was exposed to the messages two to three times per week. The control residents in
Cable System B (7400 homes) did not receive any messages. In addition, for one month
before and the nine months during the television activators, the use of safety belts by vehi-
cle drivers was observed in a systematic rotating schedule from 14 different sites within the
community.
Vehicle license plate numbers were recorded and later matched with each owner’s
name and address from the files of the state Department of Motor Vehicles. The television
viewers did not know they were in an experiment, and the field observers could not know
the experimental condition of a particular vehicle observation.
Overall mean safety-belt use among drivers was 8.4 percent for males and 11.3 percent
for females for the intervention group, and 8.2 percent and 10.3 percent for the control
group. It is easy to conclude that television public service announcements have no effect on
whether a person buckles up (Robertson, 1976).
But consider that four of the six different television spots were based on a fear tactic,
highlighting the negative consequences of disfigurement and disability. Research suggests
that a fear-arousing approach is usually not desirable for safety messages (Leventhal et al.,
1983; Winett, 1986). Anxiety elicited by a vivid portrayal of the disfiguring consequence of
a vehicle crash can interfere with the viewer’s attention and retention (Lazarus, 1980). It
can cause viewers to “tune out” subsequent spots as soon as they appear (Geller, 1989).
Consequently, many viewers may have missed the end of these public service activators,
which demonstrated the problem’s solution—using safety belts.
At least part of the ineffectiveness of activating with television is owing to lack of prox-
imity between the specific response message and the later opportunity to perform the tar-
get behavior. However, this must be balanced with the great amount of exposure enabled
by television. A one percent effect of a television ad could translate to thousands using their
vehicle safety belt. It is also likely the naturalistic use of safety belts during actual televi-
sion episodes, as discussed in Chapter 7, would have greater impact than a commercial
activator or public service announcement (Geller, 1989; McGuire, 1984; Robertson, 1983).
Still, if communities and corporations activated safety at the time and place for the desired
behavior, the overall impact could be far greater than a television ad and the cost could be
minimal, as illustrated by the following field study.

Buckle-up road signs

Over a two-year period my students evaluated the behavioral impact of buckle-up activa-
tors located along the road in my hometown of Newport—a small rural community in
southwest Virginia. They started collecting baseline data in March 1993, by unobtrusively
observing and recording the safety-belt use of vehicle drivers and passengers from a parked
vehicle near the intersection of a four-lane highway (Highway 460) and the two-lane road
(Route 42) leading into Newport. Observations were taken of vehicles entering or leaving
Route 42 to Newport, as well as vehicles continuing on Highway 460, during most week-
days from approximately 4:00 to 6:30 p.m., when the Newport traffic was heaviest.
After 13 weeks of baseline observation, the sign shown in Figure 10.13 was positioned
approximately 7 feet from Route 42 and 300 feet from the intersection of Route 42 and
Highway 460. The sign was eight feet long by four feet high, and the buckle-up message
shown in Figure 10.13 was painted on both sides in black eight-inch high letters against a
white background. The sign could not be seen by occupants of vehicles continuing along
Highway 460.
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 193

Chapter ten: Intervening with activators 193

Figure 10.13 The feedback sign in Newport, VA, compared the safety-belt use of males
and females.

The identical message was posted on a three-foot by six-foot sign in front of the
Newport Community Center, located about one-half mile from the sign shown in Figure
10.13. The five-inch letters were black and removable. Every Monday the percentages were
changed to reflect mean safety-belt use for males and females during the prior week.
Vehicle observations continued for 24 weeks, then the feedback sign was removed.
After 21 weeks of observation during this withdrawal condition, the signs were reinstated,
but with a different message. We wanted to see if safety-belt use could be activated with a
sign that did not need to be changed weekly to reflect belt-use feedback. The new message
was, “WE BUCKLE UP IN NEWPORT TO SET AN EXAMPLE FOR OUR CHILDREN.”
The results of our long-term field observations are depicted in Figure 10.14. The weekly
percentages of drivers’ safety-belt use are graphed over the 77 weeks of the project.
Percentages were calculated separately for vehicles entering or exiting Newport (the inter-
vention group) vs. those continuing on Highway 460 (the control group).
Results show quite clearly that both signs increased safety-belt use substantially. While
mean safety-belt use in vehicles traveling on Highway 460 remained relatively stable, the
mean safety-belt use in vehicles entering or exiting the road on which the signs were placed
fluctuated systematically with placement and removal of the buckle-up activators. The
horizontal lines through the data points of the graph in Figure 10.14 depict mean driver
safety-belt use per phase and condition.
The overall impact of these activators was impressive and suggests that large-scale
increases in safety-belt use would occur if communities and companies nationwide imple-
mented this simple activator intervention. Other researchers have shown impressive
effects of feedback signs to increase safety-belt use at an industrial site (Grant, 1990) and to
reduce vehicle speeds at various community locations (Van Houten and Nau, 1983;
Ragnarsson and Björgvinsson, 1991). However, none of the other researchers evaluated
sign effects for as long a period as our study.
We showed relatively long-term benefits of the activator intervention with little habit-
uation effects. In addition, our findings suggest that an activator message referring to
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 194

194 Psychology of safety handbook

80

Percentage Using Available Shoulder Belt

70

60

50

Route 42
40
Highway 460

Baseline Feedback Sign Follow Up Second Sign

0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75
Y
Mar-93 May-93 Jul-93 Sep-93 Nov-93 Jan-94 Mar-94 May-94 Jul-94 Sep-94 Nov-94
Consecutive Weeks

Figure 10.14 The feedback sign on Route 42 increased the percentage of drivers buckled
up.
actively caring consequences can be as effective as a feedback sign that requires more effort
to implement owing to the need to collect behavioral data and post weekly feedback.
Finally, it is instructive to note that our second activator intervention ended abruptly
when vandals carried the 70-pound sign about 100 yards and threw it in my pond, as
shown in Figure 10.15. Consider this. We did not solicit community approval or involve-
ment when developing or implementing this intervention. We just built the signs and put
them in place. It is possible, even likely, that this apparent one-person decision to post a
community sign irritated some residents. A few were so outraged that they reacted with

Figure 10.15 Vandals threw the second sign in the pond.

820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 195

Chapter ten: Intervening with activators 195

countercontrol (Skinner, 1974), perhaps to gain a sense of freedom from our obvious
attempt to control their behavior (Brehm, 1966).
We will return to this issue again in Chapter 15 when I discuss ways to increase per-
ceptions of personal control and empowerment to boost involvement in efforts to achieve
a Total Safety Culture.

Principle #6: Implicate consequences

Field research has shown that activators which do not implicate consequences can influ-
ence behavior when they are salient and implemented in close proximity to an opportunity
to perform the specified target behavior. It is important to realize, however, that the target
behaviors were all relatively convenient to perform. We are talking about depositing hand-
bills in a particular receptacle, choosing certain products, using available safety glasses and
safety belts.
There is plenty of evidence that activators alone will not succeed when target behav-
iors require more than a little effort or inconvenience. My students and I, for example,
could not activate water conservation behaviors (Geller et al., 1983) or the collection and
delivery of recyclable newspapers (Geller et al., 1975; Witmer and Geller, 1976) without
adding rewarding consequences. The same was true for a number of attempts to promote
various energy conservation behaviors that involved more effort than flicking a light
switch (Hayes and Cone, 1977; Heberlein, 1975; Palmer et al., 1978).
Many of the activator strategies illustrated in this chapter were explicitly or implicitly
connected to consequences. Signing a promise card or public declaration, for example,
implicates social approval vs. disapproval for honoring vs. disavowing a commitment.
Consequences motivated employees to create safety slogans, and the most influential acti-
vators usually made reference to consequences. I have received more compliance with the
“Airline Lifesaver” since offering rewards for making the buckle-up announcement and
the “Flash for Life” included a “Thank You” consequence if the “flashee” buckled up.
Vehicle buzzers designed to promote safety-belt use were improved by implicating
consequences. When drivers of the Saturn buckle up within six seconds of turning the igni-
tion key, they avoid receiving the audible reminder. Avoiding an annoying stimulus is a
consequence that might motivate some people to buckle up. In a similar vein, the salient
beep of a radar detector effectively motivates reduced vehicle speeds because it enables
drivers to avoid a negative consequence—an encounter with a police officer.
Figure 10.16 illustrates the influence of negative consequences on activator impact. In
this case, however, the compliance will be reactive rather than proactive. That is, a negative
incident occurred because the specific behavior-focused instructions were not followed.
From now on, however, it is likely this activator will be effective for this person and if he
shares the negative incident and its messy consequences with other store personnel, this
activator will take on increased significance and behavioral impact.

Incentives vs. disincentives

Activators that signal the availability of a consequence are either incentives or disincen-
tives. An incentive announces to an individual or group, in written or oral form, the avail-
ability of a reward. This pleasant consequence follows the occurrence of a certain behavior
or an outcome of one or more behaviors. In contrast, a disincentive is an activator announc-
ing or signaling the possibility of receiving a penalty. This unpleasant consequence is con-
tingent on the occurrence of a particular undesirable behavior.
Research has shown quite convincingly that the impact of a legal mandate, for exam-
ple, drunk driving or safety-belt use laws, varies directly with the amount of media
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 196

196 Psychology of safety handbook

Figure 10.16 Negative consequences can increase the subsequent impact of an activator.

promotion or disincentive (Ross, 1982). Similarly, the success of an incentive program

depends on making the target population aware of the possible rewards. In other words,
marketing positive or negative consequences with activators (incentives or disincentives)
is critical for the motivating success of the consequence intervention.
The next chapter discusses how to design and apply consequences to motivate behav-
ior. At this point, it is important to understand that the power of an activator to motivate behav-
ior depends on the consequence it signals. Figure 10.17 illustrates this connection between
activator and consequence. If a sign like the one shown in Figure 10.17 motivated a driver
to attempt safer driving practices, it would work owing to the potential consequences
implied by the activator. Every time the driver got into the vehicle, she would be reminded

Figure 10.17 The most powerful activators imply immediate consequences.

820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 197

Chapter ten: Intervening with activators 197

of potential consequences for certain driving practices. Incidentally, do you perceive the
sign on the vehicle in Figure 10.17 as an incentive or disincentive?
My guess is you perceived it as a disincentive rather than an incentive. I bet you saw
the sign as a threat to reduce at-risk driving, rather than an incentive to encourage safe
driving. You would not expect dad to get a phone call commending his daughter’s driving.
If any phone calls were made, they would be to criticize at-risk or discourteous driving.
This is exactly how my daughter, Krista, perceived the activator in Figure 10.17. As the
illustration suggests, she flatly refused to drive with such a sign on our car.
Much to Krista’s chagrin, I actually painted and mounted the sign depicted in Figure
10.17, but the only benefit from that effort was my daughter’s increased motivation to work
with the critical behavior checklist described in Chapter 8. “Let’s lose this sign, Dad,” she
asserted, “and focus on giving each other feedback with the checklist.” I was happy to com-
ply. This exercise simply reminded me that we are socialized to expect more negative con-
sequences for our mistakes than positive consequences for our successes. The next chapter
explores this unfortunate reality in greater detail.

Setting goals for consequences

Let us talk about safety goals in the context of activators that imply consequences. Included
among Deming’s 14 points for quality transformation are “eliminate slogans, exhortations,
and targets for the work force . . . eliminate work standards . . . management by objec-
tives, and management by the numbers” (Deming, 1985). Does this mean we should stop
setting safety objectives and goals? Should we stop trying to activate safe behaviors with
signs, slogans, and goal statements? Does this mean we should stop counting OSHA
recordables and lost-time cases, and stop holding people accountable for their work
injuries?
Answers to all of these questions are “yes,” if you take Deming’s points literally.
However, my evaluation of Deming’s scholarship and workshop presentations, and my
personal communications with him in 1990 and 1991, have led me to believe that Deming
meant we should eliminate goal setting, slogans, and work targets as they are currently
implemented.
Deming was not criticizing appropriate use of goal setting, management by objectives,
and activators; rather he was lamenting the frequent incorrect use of these activator inter-
ventions. Substantial research evidence supports the use of objective goals and activators
to improve behaviors if these behavior-change interventions are applied correctly (Latham
and Yukl, 1975; Locke and Latham, 1990).
Incorrect goals. Setting zero injuries as a safety goal (as illustrated in Figure 10.1) is
a misuse of these principles and should in fact be eliminated. Holding people accountable
for numbers or outcomes they do not believe they can control is a sure way to produce neg-
ative stress or distress. Some people will not be distressed because they will not take these
outcome goals seriously. Experience has convinced them they cannot control the numbers,
so they simply ignore the goal-setting exhortations. These individuals overcome the dis-
tress of unrealistic management objectives or goals by developing a sense, a perspective or
attitude of helplessness.
What does the goal of zero injuries mean anyway? Is this goal reached when no work
injuries are recorded for a day, a month, six months, or a year? Does a work injury indicate
failure to reach the goal for a month, six months, or a year? Does the average worker believe
he or she can influence goal attainment, beyond simply avoiding personal injury?
Set SMART goals. I remember the techniques for setting effective goals with the
acronym SMART, as illustrated in Figure 10.18. SMART goal setting defines what will hap-
pen when the goal is reached (the consequences), and tracks progress toward achieving the
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 198

198 Psychology of safety handbook

Figure 10.18 SMART goals are effective activators.

goal. Rewarding feedback from completing intermediate steps toward the ultimate goal is
a consequence that motivates continued progress. Of course, it is critical that the people
asked to work toward the goal “buy in” or believe in the goal. They must believe the goal
is relevant to achieving a worthwhile consequence and that they have the skills and
resources to achieve it.
I once talked with a group of hourly workers about setting safety goals, and each mem-
ber of this safety steering committee was completely turned off to goal setting. Their past
experiences with corporate safety goals created a barrier to learning about SMART goals.
It was necessary for them to understand the difference between the right and wrong way
to set safety goals. I had them practice SMART goal setting for safety and then gave them
constructive feedback. However, they did not really believe in the power of goal setting
until they actually used SMART goals to facilitate their behavior-based safety process.
I have seen several corporate mission statements with the safety goal of zero injuries. As
I have indicated earlier, this is obviously an example of incorrect goal setting. It is easy enough
to track injuries, but employees’ daily experiences lead them to believe that many injuries
are beyond their direct control. For one thing, injuries usually happen to someone else—
what can they do about that? One injury in the workplace, perhaps resulting from another
person’s carelessness, ruins the goal of zero injuries. This leads to a perception of failure.
No one likes to feel like a failure. So people typically avoid situations where failure is
frequent or eminent, or they at least attempt to discount their own possible contribution to
the failure by blaming factors beyond their own control (as discussed in Chapter 6). This
fosters the belief that injuries are beyond personal control, and creates the sense that safety
goal setting is a waste of time.
So have I made my point? “Zero injuries” should not be specified as a safety goal.
Instead, zero injuries should be the aim or purpose of a safety mission—a mission that
depends on various safety processes motivated in part by SMART goals.
Focus on the process. Safety goals should focus on process activities that can con-
tribute to injury prevention. Workers need to discuss what they can do to reduce injuries,
from reporting and investigating near hits to conducting safety audits of environmental
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 199

Chapter ten: Intervening with activators 199

conditions and work practices. The safety steering committee I mentioned earlier wanted
to increase daily interpersonal communications regarding safety. They set a goal for their
group to achieve 500 safety communications within the following month. To do this, they
had to develop a system for tracking and recording “safety talk.” They designed a wallet-
sized “SMART Card” for recording their interactions with others about safety. One mem-
ber of the group volunteered to tally and graph the daily card totals.
Another work group I consulted with set a goal of 300 behavioral observations of lift-
ing. Employees had agreed to observe each other’s lifting behaviors according to a critical
behavior checklist they had developed. If each worker completed an average of one lifting
observation per day, the group would reach their goal within the month. Each of these
work groups reached their safety goals within the expected time period, and as a result
they celebrated their “small win” at a group meeting, one with pizza, and another with
jelly-filled donuts. Perhaps subsequent goal setting for these groups should target healthy
diet choices!
These two examples illustrate the use of SMART goals and depict safety as process-
focused and achievement-oriented, rather than the standard and less effective outcome-
focused and failure-oriented approach promoted by injury-based goals. More important,
these goals were employee driven. Workers were motivated to initiate the safety process
because it was their idea. They got involved in the process and owned it and they stayed
motivated because the SMART goals were like a roadmap telling them where they were
going, when they would get there, and how to follow their progress along the way.

In conclusion
In this chapter, I have presented examples of intervention techniques called activators.
They occur before desired or undesired behavior to direct potential performers. Based on
rigorous behavioral science research and backed by real-world examples, six principles for
maximizing effective activators were given.

• Specify behavior.
• Maintain salience with novelty.
• Vary the message.
• Involve the target audience.
• Activate close to response opportunity.
• Implicate consequences.

We are constantly bombarded with activators. At home we get telephone solicitations,

junk mail, television commercials, and verbal requests from family members. At work, it is
phone mail, e-mail, memos, policy pronouncements, and verbal directions from supervi-
sors and coworkers. On the road, there is no escape from billboards, traffic signals, vehicle
displays, radio ads, and verbal communication from people inside and outside our vehi-
cles. As discussed in Chapter 5, we selectively attend to some of these activators, ignoring
others. Only a portion of the activators we perceive actually influences our behavior.
Understanding the six principles discussed in this chapter can help you predict which ones
will influence behavior change.
Obviously, we do not need more activators in our lives. We certainly do need more effec-
tive activators to promote safety and health. It would be far better to make a few safety acti-
vators more powerful than to add more activators to a system already overloaded with
information. We need to plan our safety activators carefully so the right safety directives
receive the attention and ultimate action they deserve.
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 200

200 Psychology of safety handbook

If you want an activator to motivate action, you need to imply consequences. The most
powerful activators make the observer aware of consequences available following the
performance of a target behavior. Consequences can be positive or negative, intrinsic or
extrinsic to the task, and internal or external to the person. The next chapter will explain
the preceding sentence that is key to getting the most beneficial behavior from an inter-
vention process.

References
Berry, T. D. and Geller, E. S., A single-subject approach to evaluating vehicle safety belt reminders:
back to basics, J. Appl. Beh. Anal. 24, 13, 1991.
Berry, T. D., Geller, E. S., Calef, R. S., and Calef, R. A., Moderating effects of social assistance on verbal
interventions to promote safety belt-use: an analysis of weak plys, Environ. Behav., 24, 653, 1992.
Boyce, T. E. and Geller, E. S., Attempts to increase prosocial behavior: a comparison of reinforce-
ment and intrinsic motivation theory, Environ. Behav. in press.
Brehm, J. W., A Theory of Psychological Reactance, Academic Press, New York, 1966.
Carlson, N. R., Psychology: The Science of Behavior, 4th ed. Allyn & Bacon, Needham Heights, MA,
1993.
Chance, P., Learning and Behavior 3rd ed. Brooks/Cole Publishing Pacific Grove, CA, 1994.
Cope, J. G., Grossnickle, W. F., and Geller, E. S., An evaluation of three corporate strategies for
safety-belt use promotion, Accid. Anal. Prev., 18, 243, 1986.
Delprata, D. J., Prompting electrical energy conservation in commercial users, Environ. Behav., 9,
433, 1977.
Deming, W. E., Transformation of western style of management, Interfaces, 15(3), 6, 1985.
Evans, L., Traffic Safety and the Driver, Van Nostrand Reinhold, New York, 1991.
Geller, E. S., Increasing desired waste disposals with instructions, Man-Environ. Syst., 5, 125, 1975.
Geller, E. S., A behavioral science approach to transportation safety, Bull. NY Acad. Med., 65, 632,
1988.
Geller, E. S., The Airline Lifesaver: in pursuit of small wins, J. Appl. Behav. Anal., 22, 333, 1989a.
Geller, E. S., Using television to promote safety belt use, in Public Communication Campaigns, 2nd ed.
Rice, R. E. and Atkin, C. K., Eds., SAGE Publications, Inc., Newbury Park, CA, 1989b.
Geller, E. S. and Bigelow, B. E., Development of corporate incentive programs for motivating safety
belt use: a review, Traf. Saf. Eval. Res. Rev. 3, 21, 1984.
Geller, E. S. and Lehman, G. R., The buckle-up promise card: a versatile intervention for large-scale
behavior change, J. Appl. Behav. Anal. 24, 91, 1991.
Geller, E. S., Bruff, C. D., and Nimmer, J. G., “Flash for Life”: community-based prompting for
safety belt promotion, J. Appl. Behav. Anal. 18, 145, 1985.
Geller, E. S., Chaffee, J. L., and Ingram, R. E., Promoting paper-recycling on a university campus, J.
Environ. Syst., 5, 39, 1975.
Geller, E. S., Erickson, J. B., and Buttram, B. A., Attempts to promote residential water conservation
with educational, behavioral, and engineering strategies, Popul. Environ., 6, 96, 1983.
Geller, E. S., Farris, J. C., and Post, D. S., Promoting a consumer behavior for pollution control,
J. Appl. Behav. Anal., 6, 367, 1973.
Geller, E. S., Witmer, J. F., and Orebach, A. L., Institutions as a determinant of paper-disposal behav-
iors, Environ. Behav., 8, 417, 1976.
Geller, E. S., Witmer, J. F., and Tuso, M. E., Environmental interventions for litter control, J. Appl.
Psychol., 62, 344, 1977.
Geller, E. S., Wylie, R. C., and Farris, J. C., An attempt at applying prompting and reinforcement
toward pollution control, in Proceedings of the 79th Annual Convention of the American
Psychological Association, 6, 701, 1971.
Geller, E. S., Kalsher, M. S., Rudd, J. R., and Lehman, G. R., Promoting safety-belt use on a univer-
sity campus: an integration of commitment and incentive strategies, J. Appl. Soc. Psychol., 19, 3,
1989.
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 201

Chapter ten: Intervening with activators 201

Grant, B. A., Effectiveness of feedback and education in an employment-based seat belt program,
Health Educ. Res. 5(2), 197, 1990.
Hayes, S. C. and Cone, J. D., Reducing residential electrical use: payments, information, and feed-
back, J. Appl. Behav. Anal., 14, 81, 1977.
Heberlein, T. A., Conservation information: the energy crisis and electricity consumption in an
apartment complex, Energ. Syst. Pol., 1, 105, 1975.
Kalsher, M. J., Geller, E. S., Clarke, S. W., and Lehman, G. R., Safety-belt promotion on a naval base:
a comparison of incentives vs. disincentives, J. Saf. Res., 20, 103, 1989.
Katzev, R. and Pardini, A., The impact of commitment and token reinforcement procedures in pro-
moting and maintaining recycling behavior, J. Environ. Syst., 17, 93, 1987–1988.
Kello, J. E., Geller, E. S., Rice, J. C., and Bryant, S. L., Motivating auto safety-belt wearing in indus-
trial settings: from awareness to behavior change, J. Organ. Behav. Manage., 9, 7, 1988.
Kimble, G. A., Hilgard and Marquis’ Conditioning and Learning, Appleton-Century-Crofts, New York,
1961.
Latham, G. and Yukl, G., A review of research on the application of goal-setting in organizations,
Acad. Manage. J., 18, 824, 1975.
Lazarus, R., The stress and coping paradigms, in Theoretical Bases for Psychopathology, Eisdorfer, C.,
Cohen, D., Klienmen A., and Maxim, P., Eds., Spectum, New York, 1980.
Ley, D. and Cybriwsky, R., Urban graffiti as territorial markers, Ann. Assoc. Am. Geogr., 64, 491, 1974.
Leventhal, H., Shafer, M., and Panagis, D., The impact of communications on the self-regulation of
health beliefs, decision, and behavior, Health Educ. Q., 10, 3, 1983.
Locke, E. and Latham, G., A Theory of Goal Setting and Task Performance, Prentice-Hall, Princeton, NJ,
1990.
McGuire, W. J., Public communication as a strategy for inducing health promoting behavioral
change, Prev. Med., 13, 289, 1984.
Newman, O., Defensible Space, Macmillan, New York, 1972.
Palmer, M. H., Lloyd, M. E., and Lloyd, K. E., An experimental analysis of electricity conservation
procedures, J. Appl. Behav. Anal., 10, 665, 1978.
Ragnarsson, R. S. and Björgvinsson, T., Effects of public posting on driving speed in Icelandic traf-
fic, J. Appl. Behav. Anal., 24, 53, 1991.
Roberts, M. C., Alexander, K., and Knapp, L., Motivating children to use safety belts: a program
combining rewards and “Flash for Life,” J. Commun. Psychol., 18, 110, 1990.
Robertson, L., The great seat belt campaign flop, J. Commun., 26, 41, 1976.
Robertson, L., Injuries: Causes, Control Strategies, and Public Policy, Lexington Books, Lexington MA,
1983.
Robertson, L., Kelley, A., O’Neill, B., Wixom, C., Eisworth, R., and Haddon, W., Jr., A controlled
study of the effect of television messages on safety belt use, Am. J. Publ. Health, 64, 1071, 1974.
Ross, H. L., Deterring the Drinking Driver: Legal Policy and Social Control, Lexington Books,
Lexington, MA, 1982.
Streff, F. M., Kaisher, M. J., and Geller, E. S. Developing efficient workplace safety programs:
observations of response covariation, J. Organ. Behav. Manage., 13, 2, 3–15, 1993.
Skinner, B. F., About Behaviorism, Alfred A. Knoff, New York, 1974.
Talton, A., Increasing Safety-belt Usage through Personal Commitment: a Church-based Pledge
Card Program, Virginia Polytechnic Institute and State University, Blackburg, Masters’ thesis,
unpublished, 1984.
Thyer, B. A. and Geller, E. S., The “buckle-up” dashboard sticker: an effective environmental inter-
vention for safety belt promotion, Environ. Behav., 19, 484, 1987.
Thyer, B. A., Geller, E. S., Williams, M., and Purcell, S., Community-based “flashing” to increase
safety belt use, J. Exp. Educ., 53, 155, 1987.
Tillman, R. and Kirkpatrick, C. A., Promotion: Persuasive Communication in Marketing, Richard D.
Irwin, Homewood, IL, 1972.
Van Houten, R. and Nau, P. A., Feedback interventions and driving speed: a parametric and com-
parative analysis, J. Appl. Behav. Anal., 16, 253, 1983.
820049_CRC1_L1540_CH10 11/10/00 1:45 PM Page 202

202 Psychology of safety handbook

von Buseck, C. R. and Geller, E. S., The vehicle safety belt reminder: can refinements increase safety
belt use?, Technical Research Report for General Motors Research Laboratories, Warren, MI,
1984.
Wang, T. H. and Katzev, R., Group commitment and resource conservation: two field experiments
on promoting recycling, J. Appl. Soc. Psychol., 20, 265, 1990.
Weick, K. E., Small wins: redefining the scale of social problems, Am. Psychol., 39(1), 40, 1984.
Williams, A. F., Wells, J. D., Lund, A. K., and Teed, N., Observed use of seat belts in 1987 cars, Accid.
Anal. Prev., 19, 243, 1989.
Winett, R. A., Prompting turning-out lights in unoccupied rooms, J. Environ. Syst., 6, 237, 1978.
Winett, R. A., Information and Behavior: Systems of Influence, Erlbaum, Hillsdale, NJ, 1986.
Witmer, J. F. and Geller, E. S., Facilitating paper recycling: effects of prompts, raffles, and contests,
J. Appl. Behav. Anal., 9, 315, 1976.
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 203

chapter eleven

Intervening with consequences

Consequences motivate behavior and related attitudes. This happens in various ways. Consequences
can be positive or negative, intrinsic (natural) or extrinsic (extra) to a task, and internal or external
to a person. These characteristics need to be considered when designing and evaluating intervention
programs. This chapter explains why and provides principles and practical procedures for motivat-
ing people to work safely over the long term. In other words, I shall show you how to influence behav-
ior and attitudes so that both are consistent with a Total Safety Culture.

“Every act you have ever performed since the day you were born was performed
because you wanted something.”—Dale Carnegie

The introductory quotation from Carnegie’s classic book, How to Win Friends and Influence
People, first published in 1936, represents a key principle of human motivation and behav-
ior-based safety. Although supported by substantial research (Skinner, 1938), it actually
runs counter to common sense.
Think about it. When people ask us why we did something, we are apt to say, “I
wanted to do it,” or “I was told to do it,” or “I needed to do it.” These explanations sound
as if the cause of our behavior comes before we act. This perspective is supported by
numerous “pop psychology” self-help books and audiotapes that say people motivate
themselves with positive self-affirmations or optimistic thinking and enthusiastic expecta-
tions. In other words, behavior is caused by some external request, order, or signal or by an
internal force, drive, desire, or need.
Pop psychology often asserts that people cannot be motivated by others, only by them-
selves from within. This self-motivation is typically referred to as “intrinsic motivation”
and is a prominent theme in popular books by Deming (1993), Covey (1989), and Kohn
(1993). It is also the theme of the classic best seller by Peale (1952).
Indeed, Kohn reiterates throughout his book that any attempt to motivate people with
extrinsic procedures—incentives, praise, recognition, grades, and penalties—will detract
from intrinsic motivation and do more harm than good. Kohn concludes that interventions
set up to motivate others, even achievement-oriented reward and recognition programs,
are generally perceived as “controlling” and, thus, decrease “intrinsic” or self-motivation.
Fortunately, there is much solid research in behavioral science to discredit Kohn’s
assertions (see, for example, reviews by Cameron and Pierce, 1994; Carr et al., 1993;
Pearlstein, 1995; and Flora, 1990). I say “fortunately” because if all reward and recognition
programs detracted from our “intrinsic” (or internal) motivation to perform in certain
ways, many industry, school, and community motivational programs would be futile.
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 204

204 Psychology of safety handbook

This chapter will explain the fallacy in Kohn’s argument and show ways to maximize
the impact of an extrinsic reward process. Again, the research-supported principle is that
activators direct behavior and consequences motivate behavior, but the type of consequence
certainly influences the amount of motivation, as this chapter will explain.
Deming and Covey want people to act out of the knowledge that it is the right thing to
do. Covey (1989) refers to this motivation as “principle-centered.” I certainly agree with the
need for inner-directed, self-motivated behavior. When people consistently go out of their
way for the safety of themselves and others, they are principle centered. They have reached
the ultimate in safety. They hold safety as a value.
It seems Deming, Covey, Kohn, and others who have written about human motivation
presume people are already principle-centered for various activities, including safety. They
give advice from the perspective that people are “willing workers,” self-motivated to do
the right thing. Throughout this book, I have made the case that natural consequences
often motivate people to do the wrong thing when it comes to safety, like take risks. This is
the basic discrepancy between the person-based, or humanistic approach to safety, and
behavior-based safety.
Most people do not consistently avoid at-risk behavior. This calls for behavior-based
safety (including the use of consequences) to bring people to the principle-centered, self-
motivated stage. Recall the principle I have emphasized several times—people act them-
selves into new ways of thinking. In other words, people become principle-centered and
self-directed through their routine actions. As discussed in Chapter 9, behavior-based
intervention (instruction, support, or motivation) is needed to make safe behavior the rou-
tine. Then principle-centered or value-based safety eventually follows.

The power of consequences

Popular author and humorist Robert Fulghum (1988) wrote All I Really Need to Know I
Learned in Kindergarten, claiming he learned all the basic rules or norms for socially accept-
able adult behavior as a young child. The list of rules in Figure 11.1 was excerpted from
Fulghum’s famous book. Rules like share everything, play fair, don’t fight, and clean-up your
own mess were taught to most of us early on. These are clearly ideal edicts to live by. Perhaps
you still recall a teacher or parent using these rules to try to shape your behavior. Did it
work? Do you follow each of these basic norms regularly, for no other reason or conse-
quence except your realization that it is the right thing to do?
Imagine what a better world we would live in if everyone followed the simple rules
listed in Figure 11.1 from a self-directed, principle-focused perspective. Alas, there are
signs everywhere that this is not so. Take the automatic flushers in public facilities like air-
ports. They indicate that we have lost confidence in following the simplest of these rules—
“flush.” Flushing the toilet is followed by a natural consequence that should increase
future occurrences of this effortless response. Frankly, I like to control my own flush, thank
you, and I did not appreciate engineers taking that opportunity for personal control away
from me.
The last two kindergarten rules in Figure 11.1 are directly relevant to safety and, in fact,
reflect basic themes of this text. As discussed previously, especially in Chapter 10, safety
needs people to stick together in a spirit of shared belonging and interdependence.
However, sometimes we need activators to remind us of this critical rule and consequences
to keep us working together for safety.
“LOOK,” Fulghum’s last rule, is key to behavior-based safety and to achieving a Total
Safety Culture. This implies the “defensive working style” employees need to adopt. In a
Total Safety Culture, everyone looks for ways to improve safety by intervening to reduce
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 205

Chapter eleven: Intervening with consequences 205

❑ Share everything
❑ Play fair
❑ Don’t hit others
❑ Put things back where you found them
❑ Clean up your own mess
❑ Don’t take things that aren’t yours
❑ Say you’re sorry when you hurt someone
❑ Wash your hands before you eat
❑ Flush
❑ When you go out into the world, watch out
for traffic, hold hands, and stick together
❑ And remember the Dick-and-Jane books
and the first word you learned--the
biggest word of all: LOOK

Figure 11.1 Basic rules of social life we learned well as children we do not necessarily fol-
low as adults. Excerpted from Fulghum (1988). With permission.

at-risk behaviors and increase safe behaviors. In Chapter 10, we discussed ways to inter-
vene with activators. Here, we focus on the more powerful intervention approach—manip-
ulating consequences.

Consequences in school
Figure 11.2 reveals the power of consequences in school, the place where we heard most of
the rules listed in Figure 11.1. Many students have difficulty staying focused on their stud-
ies. Everyone tells them to stick with the program, put up with uninteresting teachers, and
do exactly as told. Why? Because if the student is diligent and patient, the hard work even-
tually pays off.
Some students are able to hang in there for the distant consequence of attaining a col-
lege degree and/or getting a good job. Of course, it is necessary to remind them of these
remote reasons. Sometimes, this is done by emphasizing grades, claiming that high grades
are necessary for a successful career. At any rate, academic behavior is typically motivated
by consequences, the most sizable being distant and remote.
Many students, though, are lured away from their studies by more immediate and cer-
tain consequences for distracting behaviors. As a result, the principles of behavioral science
discussed here for safety have been applied successfully to keep students on track. How? By
making classroom activities more rewarding (Sulzer-Azaroff and Mayer, 1972, 1986, 1991).
Do any students get soon, certain, and positive consequences for their school-based
behaviors? Who gets the letter sweaters, awards banquets, newspaper recognition, and
crowds of people cheering for their extra effort? Right—the athletes.
If my daughter, Karly, spent half the time working on academic-related tasks, even
reading for pleasure, as she does on sports, I would have no worry about her future, but
what should I expect? She has been playing baseball and basketball since the fourth grade,
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 206

206 Psychology of safety handbook

Figure 11.2 Students need consequences to keep them going.

and from the start she has received positive consequences for her performance, from tro-
phies and plaques to encouraging words from coaches, peers, and family members, and
guess who has been at almost every game, cheering her on. Dad, mom, grandma, and often
her older sister have.
What soon, certain, and valued consequences can keep Karly focused on improving
her academic performance? Letters on a report card every six weeks cannot compete with
the immediate ongoing rewards from her athletic performance. Academic activities are
boring, and sports activities are fun. Homework is work to be avoided, if possible. Sports
conditioning is work also, but necessary to achieve those rewards of successful athletic per-
formance. Actually, the soon, certain, and positive consequences available for any behav-
ior can determine whether it is boring or fun.
Peer influence. Obviously, consequences from peers are powerful motivators. We
work to achieve peer recognition or approval, and to avoid peer criticism or disapproval.
Think about this. Do students receive peer support when they demonstrate extra effort in
the classroom? In Karly’s high school, students who ask questions and show special inter-
est are often called “nerds.” In college, I was in the fraternity with most of the school’s
sports heroes and, in order to fit in with the group, I felt peer pressure to conceal my high
grades.
These days you are apt to see the names of honor students published in local newspa-
pers. You might see a bumper sticker proudly asserting that someone in a family made the
honor roll. Unfortunately, there has been a negative consequence to this sort of recognition,
reflected in the peer pressure bumper sticker depicted in Figure 11.3. When I saw this on a
pick-up truck in Blacksburg, VA, I was reminded once more of how hard it can be to see
rewarding consequences for academic success or improvement. We have no trouble in
the United States finding a “dream team” for athletic activities, but too often we fail to
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 207

Chapter eleven: Intervening with consequences 207

Figure 11.3 Peer pressure can inhibit academic performance.

motivate students to seek meaningful and admirable dreams through academic achieve-
ment. One root of the problem is misplaced consequences.

Intrinsic vs. extrinsic consequences

Most applied behavioral scientists view “intrinsic motivation” differently from the descrip-
tion used in pop psychology books (for example, Kohn, 1993). The behavior-based per-
spective is supported by research and our everyday experience. Plus, it is objective,
practical, and useful for developing situations and programs to motivate behavior change.
Simply put, “intrinsic” does not mean “inside” people, where it cannot be observed,
measured, and directly influenced (Horcones, 1987). Rather, intrinsic refers to the nature of
the task in which an individual is engaged. Intrinsically motivated tasks, or behaviors, lead
naturally to external consequences that support the behavior (rewarding feedback) or give
information useful for improving the behavior (corrective feedback).
Most athletic performance, for example, includes natural or intrinsic consequences that
give rewarding or correcting feedback. These consequences, intrinsic to the task, tell us
immediately how well we have performed at swinging a golf club, shooting a basketball,
or casting a fishing lure, for example. They motivate us to keep trying, sometimes
after adjusting our behavior as a result of the natural feedback directly related, or intrinsic,
to the task.
Take a look at the fisherman in Figure 11.4. Some psychologists would claim he is moti-
vated from within, or self-motivated. They use the term “intrinsic motivation” to refer to
this state (Deci, 1975; Deci and Ryan, 1985). In contrast, the behavioral scientist points to the
external consequences naturally motivating the fisherman’s behavior. These cause him to
focus so completely on the task at hand that he is not aware of his wife’s mounting anger—
or he is ignoring her. He may also be unaware that his supply of fish is creating a potential
hazard. In a similar way, safety can be compromised because of excessive motivation for
production. Rewards intrinsic to production can cause this motivation.
Notice that the “worker” in this picture does not receive a reward for every cast. In fact,
he is on an intermittent reinforcement schedule. He catches a fish once in a while. This kind
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 208

208 Psychology of safety handbook

Figure 11.4 Some tasks are naturally motivating because of intrinsic consequences.

of reward schedule is most powerful in maintaining continuous behavior. Anyone who has
gambled understands. Some say gambling is a disease, when in fact gambling is behavior
maintained by intermittent rewarding consequences.
Some tasks do not provide intrinsic or natural feedback. In this case, it is necessary to
add an extrinsic, or extra, consequence to support or redirect the behavior. Many, if not
most, safety behaviors fall in this category.
In fact, many safety practices have intrinsic negative consequences, such as discomfort,
inconvenience, and reduced pace, that naturally discourage their occurrence. Thus, there is
often a need for extrinsic supportive consequences, like intermittent praise, recognition,
novelties, and credits redeemable for prizes, to shape and maintain safe behaviors
(Skinner, 1982). The intent is not to control people, but to help people control their own
behavior by offering positive reasons for making the safe choice.
Now look at the student in Figure 11.5. He expects an extrinsic positive consequence
for completing an accurate calculation. Do you see a problem here? Sure, the pupil should
feel good about deriving the right answer. In other words, the intrinsic consequence of
completing a task correctly should be perceived as valuable and rewarding by the student.
The student should perceive the important payoff as getting the right answer.
Now we are talking about a person’s interpretation of the situation, which I refer to as
“internal” consequences in the next section. First, let us understand a very important point
reflected in Figure 11.5. Whenever there is an observable intrinsic consequence to a task, the
instructor, supervisor, or safety coach needs to help the performer see that consequence
and realize its importance. In other words, we need to help people perceive the intrinsic
consequences of their performance and show appreciation and pride in that outcome. This
helps to make the intrinsic consequence rewarding to the performer, thereby facilitating
ongoing motivation (Horcones, 1992). So, if the teacher in Figure 11.5 displayed genuine
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 209

Chapter eleven: Intervening with consequences 209

Figure 11.5 External rewards can reduce internal motivation.

approval and delight in the student’s achievement, an extra extrinsic reward might not be
needed to keep the performer motivated.

Internal vs. external consequences

The intrinsic and extrinsic consequences discussed so far are external to the individual. In
other words, they can be observed by another person. Behavioral scientists focus on these
types of consequences to develop and evaluate motivational interventions because they
can be objective and scientific when dealing with external, observable aspects of people.
Behavioral scientists, however, do not deny the existence of internal factors that moti-
vate action. There is no doubt that we talk to ourselves before and after our behaviors, and
this self-talk influences our performance. We often give ourselves internal verbal instruc-
tions, called intentions, before performing certain behaviors. After our activities, we often
evaluate our performance with internal consequences. In the process, we might motivate
ourselves to press on (with self-commendation) or to stop (with self-condemnation).
When it comes to safety and health, internal consequences to support the right behav-
ior are terribly important. Remember, external and intrinsic (natural) consequences for safe
behaviors are not readily available, and we cannot expect to receive sufficient support
(extra consequences) from others to sustain our proactive, safe, and healthy choices. So we
need to talk to ourselves with sincere conviction to boost our intentions. We also need to
give ourselves genuine self-reinforcement after we do the right thing to keep ourselves
going. When we receive special external consequences from others for our efforts, we need
to savor these and use them later to bolster our self-reinforcement.
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 210

210 Psychology of safety handbook

An illustrative story
A brief exchange I had with W. Edwards Deming at a seminar he conducted contrasts the
behavioral perspective on intrinsic consequences with the humanistic and more popular
view. Deming (1991) was describing how much he appreciated the special attention he
received from a flight attendant—Debbie. Debbie helped him into a wheelchair at the
arrival gate, pushed him a long distance across the airport to his ground transportation,
and then helped get him into the limousine. Pleased with Debbie’s actively caring behav-
ior, Deming pulled a five-dollar bill from his pocket to give her. She quickly refused his
offer, saying she was not allowed to accept gratuities from customers.
Deming told us he felt “so bad” about his attempt to reward Debbie. Later, he tried to
find Debbie’s last name so he could contact her and apologize for his “terrible mistake.” He
was so sure he had depreciated Debbie’s “intrinsic” motivation by his attempt to give her
an “extrinsic” reward. Deming used this story to explain the wide-spread pop psychology
notion that motivation only comes from within a person, and that any attempts to increase
it with extrinsic rewards will only decrease a person’s “intrinsic” motivation.
I was disappointed in Deming’s explanation of motivation and was distressed that an
audience of 600 or more might believe that any attempt to show appreciation for another
person’s performance with praise, some material reward, or award ceremony would be
done in vain, probably causing more harm than good. I ventured timidly to a microphone
to state a behavior-based perspective.*
I said that I was a behavioral scientist and a university professor and would like to offer
another perspective on his airline story. I began with the basic principle that behavior is
motivated by consequences. Some consequences are natural or intrinsic to the task and oth-
ers are sometimes added to the situation, like words of approval or money. Debbie’s behav-
iors were motivated by intrinsic consequences occurring while she wheeled him to his
destination, from observing sites along the way to enjoying conversation with a prominent
scholar, teacher, and consultant. The five dollars was an extrinsic consequence which could
add to or subtract from self-motivation depending upon personal interpretation.
If Debbie felt she deserved much more for her efforts, she might have been offended
and thought less of her customer, but it would not have detracted from the ongoing intrin-
sic (natural) consequences that make her job enjoyable to her. On the other hand, she might
interpret her job as quite boring or nonsatisfying, meaning the intrinsic consequences are
not enough to make her feel good about her work. In this case, any extrinsic consequence
could help justify her behavior and make her feel better about her job.
Deming nodded his head, saying, “Yes, thank you.” My experiences at Deming’s work-
shops led me to believe that such a reply from him represented sincere appreciation. Talk
about consequences. I interpreted his extrinsic response as a reward and I felt good about
my behavior—approaching the microphone. Plus, my self-motivation was increased fur-
ther by kind words and approval I received from other workshop participants as I returned
to my seat.
The bottom line is this. Our behavior is motivated by extrinsic or extra, as well as nat-
ural or intrinsic consequences; our self-motivation is influenced by how these external con-
sequences (intrinsic and extrinsic) are interpreted. Self-motivation can decrease if a
motivational program is seen as an attempt to control behavior. Thus, it is important that
praise, recognition, and other rewards are genuine expressions of appreciation. Individuals
or groups being recognized must believe they truly earned this consequence through their

* As anyone who has attended a Deming seminar will tell you, it was risky to voice a concern or question to Dr.
Deming. Thus, my nervousness was quite rational.
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 211

Chapter eleven: Intervening with consequences 211

own efforts. Rewards that we believe are genuine and earned by our own behaviors are
likely to increase our inner drive; consequences perceived as nongenuine, undeserved, or
administered only to control our behavior could be counterproductive.

Four types of consequences

Figure 11.6 summarizes the different types of consequences. Relative to a task or job assign-
ment, consequences can be natural (intrinsic) or extra (extrinsic). Natural consequences,
produced by the target behavior, are usually immediate and certain. In contrast, extra
consequences are added to the situation and are often delayed and may be uncertain.
Extra consequences are necessary when the natural consequences are insufficient to moti-
vate the desired behavior, as is often the case with safety-related activities (Geller, 1996;
Sulzer-Azaroff, 1992).
Relative to the person performing the task, consequences can be considered external or
internal. External consequences are observable by others and, thus, can be studied objec-
tively. Internal consequences are subjective and biased by the performer’s perceptions. It is
difficult to know objectively the exact nature of the internal consequences influencing an
individual’s performance. However, we know from personal experience that internal con-
sequences and evaluations accompany performance and dramatically influence motiva-
tion and subsequent performance.
I have eliminated the term “intrinsic” from this classification scheme because of the dif-
ferent uses of this word. Note, however, that “natural” is synonymous with “intrinsic”
from a behavior-based perspective (Skinner, 1957; Vaughan and Michael, 1982), while
“internal” is the same as “intrinsic” from a humanistic (or person-based) perspective (Deci,
1975; Kohn, 1993).
Figure 11.6 classifies various activities according to the type of consequence relative to
the task (natural vs. extra) and the task performer (internal vs. external). While these activ-
ities illustrate particular types of consequences available to motivate performance, the cat-
egorizations are neither mutually exclusive nor inclusive. Even the most straightforward
task classifications, for example, can overlap with other categories, according to percep-
tions of the performer.

Figure 11.6 Behavior is motivated by four different types of consequences.

820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 212

212 Psychology of safety handbook

Figure 11.7 Some tasks have natural rewarding consequences.

For example, if you play a musical instrument, complete a crossword puzzle (see
Figure 11.7), plant a garden, or participate in recreational sports, natural and external con-
sequences are immediately available. You have performed well, done a good job, or maybe
you are not pleased with the results. Add to this the fact that you might compare your
results to past results or the accomplishments of others. This is adding a personal evalua-
tion bias to the natural feedback—internalizing the external consequences. Now you have
created internal consequences to accompany your activity.
Let us take it a step further. Perhaps another person adds an extra consequence by
commending or condemning your performance. This could dramatically influence your
motivation. What if you got paid for gardening or playing the piano? Your motivation
could be further influenced.
As we have discussed, some activities or behaviors are not readily motivated by cer-
tain types of consequences, thus requiring extra support. Figure 11.6 can be used to iden-
tify these tasks and guide approaches for consequence intervention. Because safe behavior
competes with at-risk behavior that is supported by external and natural consequences, it
is usually necessary to support safe behavior with extra consequences. This leads us now
to a discussion of two very popular safety topics: rewards and penalties (actually referred
to as “discipline” in occupational settings).

Managing consequences for safety

At this point, I am sure you appreciate the special message reflected in Figure 11.8.
Submitting safety suggestions is an activity not typically followed by external motivating
consequences. In many work cultures, the idea of safety suggestions has long since passed.
The suggestion boxes are empty. Does this mean there are no more good suggestions? Is the
work force not creative enough? You know the answer to both of these questions is a
resounding “No.”
Let me give you an example. I once worked with safety leaders at a Toyota Motor man-
ufacturing plant in Georgetown, KY, whose 6,000 employees submitted more than 35,000
quality, production, or safety-related suggestions in 1994. A greater number of suggestions
were expected in 1995. Many employees in this culture are motivated internally to submit
suggestions, but external consequences are in place to keep the process going.
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 213

Chapter eleven: Intervening with consequences 213

Figure 11.8 Some tasks require supportive consequences.

Employees receive timely feedback regarding the utility and feasibility of every
suggestion, and if the suggestion is approved, they are empowered to implement it
themselves. Also, the individual or team responsible receives 10 percent of the savings
for the first year that result from the implemented suggestion. Such external, extra and
meaningful—in this case economic—consequences motivate a large work force to make a
difference.

Four behavior-consequence contingencies for motivational intervention

A behavior-consequence contingency is a relationship between a target behavior to be
influenced and a consequence that follows. Safety can be improved by managing—or
manipulating—four distinct behavior-consequence relationships. Specifically, the proba-
bility of injury can be reduced by

• Increasing positive consequences of safe behavior.

• Decreasing negative consequences of safe behavior.
• Decreasing positive consequences of at-risk behavior.
• Increasing negative consequences of at-risk behavior.

The contingencies can involve natural or extra consequences. When PPE is made avail-
able that is more comfortable or convenient to use, a natural behavior-consequence con-
tingency is put in place decreasing the previous negative consequences of safe
behavior—the possible feeling of discomfort and restricted movement that can come from
wearing PPE.
Still, this contingency may not be sufficient to overpower the natural convenience
and “get-the-job-done-quicker” contingency supporting the at-risk behavior of working
820049_CRC1_L1540_CH11 11/10/00 1:53 PM Page 214

214 Psychology of safety handbook

without PPE. It might be advisable to add an incentive/reward contingency to increase

PPE use or implement a disincentive/penalty contingency to increase negative conse-
quences of at-risk behavior.
An illustrative case study. Several years ago, I consulted with the managers and
safety leaders of a large work group who were genuinely concerned about the work pace
of their line employees. The probability of a cumulative trauma disorder, especially carpal
tunnel syndrome, was certainly a direct function of the work pace (Silverstein et al., 1987).
Their question was, “How can we reduce the work pace?” They essentially wanted my
advice on an education or incentive program that would decrease the work pace and lessen
the occurrence of cumulative trauma disorders (CTDs).
Before deriving a contingency to motivate behavior change, it is important to first
examine the existing contingencies that support undesirable behavior. In this case, the
behavior was a rapid work pace. The most obvious contingency supporting the at-risk
behavior was the relationship between work pace and the workers’ break time. As soon as
employees finished their assignment, they could go to the break area and remain until the
next work period. According to supervisors, this contingency was necessary for the partic-
ular work process and the union contract.
Do you think I recommended an education program—which would be an activator—
to reduce the work pace? Did I suggest positive consequences to motivate a slower pace?
Or did I advise negative consequences for a rapid pace?
I am sure you understand why the answer is “no” to each of these questions. I could
not recommend a feasible extra consequence powerful enough to overcome the current
negative consequence—less time in the break room—of a slower work pace and the ongo-
ing positive consequence—more time in the break room—contingent on a fast work pace.
I thought it necessary to alter the work-break reward to decrease positive support of the at-
risk behavior. This sort of systems change was not possible at the time, and the probability
of CTDs among these workers was not changed.
What is the lesson? Is there a lesson in my failure to make a difference? Perhaps the
most important lesson here is that some behaviors cannot be changed by merely adding a
consequence intervention to the situation. An existing behavior-consequence contingency
might overpower the impact of a feasible intervention program. Actually, this is a frequent
problem with efforts to improve safety. We should not expect activators or weak conse-
quences to improve safety over the long term if natural and powerful behavior-conse-
quence contingencies exist to support at-risk behavior. Sometimes it is necessary to change
the existing contingencies first.

The case against negative consequences

To subdue influences supporting at-risk behavior, it is often tempting to use a punishment
or penalty. All that is needed is a policy statement or some type of top-down mandate spec-
ifying a soon, certain, and sizable negative consequence following specific observable risky
behaviors. Could this contingency be powerful enough not to override the many natural
positive consequences for taking risks?
Yes, behavioral scientists have found negative consequences can permanently
suppress behavior if the punishment is severe, certain, and immediate (Azrin and
Holz, 1966). However, before using “the stick,” you should understand the practical
limitations and undesirable side effects of using negative consequences to influence
behavior.
Skinner (1953) deplored the fact that “the commonest technique of control in modern
life is punishment” (page 182). He protested against the human preoccupation with
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 215

Chapter eleven: Intervening with consequences 215

punishment until his death in 1990. Skinner’s animal research with relatively mild punish-
ment indicated that negative consequences merely suppress behaviors temporarily. Plus,
the use of negative consequences to control behavior has four undesirable side effects:
escape, aggression, apathy (Chance, 1999; Skinner, 1953; Sidman, 1989), and countercontrol
(Skinner, 1974).
Escape. Animals and people attempt to avoid situations with a predominance of
negative consequences. Sometimes, this means staying away from those who administer
the punishment. Humans will often attempt to escape from negative consequences by sim-
ply “tuning out” or perhaps cheating or lying. Sidman (1989) noted that the ultimate escape
from excessive negative consequences is suicide. Indeed, it is not uncommon for an indi-
vidual to commit suicide in order to escape control by aversive stimulation, which can
include the intractable pain of an incurable disease, physical or psychological abuse from
a family member, or perceived harassment by an employee or coworker.
Unpleasant attitudes or emotional feelings are produced when people work to escape
or avoid negative consequences. As shown in Figure 11.9, negative consequences can influ-
ence behavior dramatically, but such situations are usually unpleasant for the “victim.”
Under fear arousal conditions, people will be motivated to do the right thing, but only
when they have to. They feel controlled, and as discussed in Chapter 6, this can lead to dis-
tress and burnout. Obviously, this type of contingency and side effect is incompatible with
a Total Safety Culture where people feel “in control” and are ready and willing to go
beyond the call of duty for another person’s safety and health.
Aggression. Instead of escaping, people might choose to attack those perceived to be
in charge. For example, murder in the workplace is rapidly increasing in the United States,

Figure 11.9 Fear of negative consequences is motivating.

820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 216

216 Psychology of safety handbook

and the most frequent cause appears to be reaction to or frustration with control by nega-
tive means (Baron, 1993). Aggressive reaction to this kind of control, however, might not be
directed at the source (Oliver et al., 1974).
An employee frustrated by top-down aversive control at work might not assault his
boss directly, but rather slow down production, sabotage a safety program, steal supplies,
or vandalize industrial property, or the employee might react with spousal abuse. Then the
abused spouse might react by slapping a child. The child, in turn, might punch a younger
sibling and the younger sibling might punch a hole in a wall or kick the family pet—all as
a result of perceived control by negative consequences.
Apathy. Apathy is a generalized suppression of behavior. In other words, the nega-
tive consequences not only suppress the target behavior but might also inhibit the occur-
rence of desirable behaviors. Regarding safety, this could mean a decrease in employee
involvement. When people feel controlled by negative consequences, they are apt to sim-
ply resign themselves to doing only what is required. Going beyond the call of duty for a
coworker’s health or safety is out of the question.
Countercontrol. No one likes feeling controlled, and situations that influence these
feelings in people do not encourage buy-in, commitment, and involvement. In fact, some
people only follow top-down rules when they believe they can get caught, as typified by
drivers slowing down when noticing a police car. Some people look for ways to beat the
system they feel is controlling them, so you have vehicle drivers purchasing radar detec-
tors. This is an example of “countercontrol” (Skinner, 1974), the fourth undesirable side
effect of negative consequence contingencies.
I met an employee once who exerted countercontrol by wearing safety glasses without
lenses; when wearing his “safety frames,” he got attention and approval from certain
coworkers. Perhaps these coworkers were rewarded vicariously when seeing him beat the
system they perceived was controlling them also.
Figure 11.10 illustrates an example of countercontrol. Although the supervisors might
view the behavior as “feedback,” it is countercontrol if it occurred to regain control or assert
personal freedom. A perceived loss of control or freedom is most likely when a negative
consequence contingency is implemented. Also, countercontrol behavior is typically
directed at those in charge of the negative consequences.

Discipline and involvement

Let us specifically discuss traditional discipline for safety—a form of top-down control
with negative consequences. I have met many managers who include a “discipline session”
as part of the corrective action for an injury report. The injured employee gets a negative
lecture from a manager or supervisor whose safety record and personal performance
appraisal were tarnished by the injury.
These “discipline sessions” are unpleasant for both parties and, certainly, do not
encourage personal commitment or buy-in to the safety mission of the company. Instead,
the criticized and embarrassed employees are simply reminded of the top-down control
aspects of corporate safety, usually resulting in increased commitment not to volunteer for
safety programs nor to encourage others to participate. In this case, the culture loses the
involvement of invaluable safety participants.
Individuals who have been injured on the job have special insight into conditions and
behaviors that can lead to an injury. If persuaded to discuss their injuries with others, they
can be very influential in motivating safe work practices. Personal testimonies, especially
by people known to the audience, have much greater impact than statistics summarizing
the outcomes of a remote group (Sandman, 1991).
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 217

Chapter eleven: Intervening with consequences 217

Figure 11.10 Countercontrol is usually directed at those in charge of negative conse-

quences.

I propose you consider seven basic questions before applying a punishment contingency.
In most cases, you will conclude that a corrective action other than punishment is called for.
Some answers will offer direction as to what the alternative intervention should be.
1. Was a specific rule or regulation violated? If you answer “no,” punishment is
obviously unfair. Does this mean you need to write more rules or document more regula-
tions? I do not think so. You cannot write a rule for every possible at-risk behavior. Yet, a
proper injury analysis (as discussed in Chapter 9) will reveal some human errors more at-
risk at causing serious injury than behaviors already covered by a rule or regulation.
Because human errors are unintentional (as explained in Chapter 4), rules will not decrease
them. Plus, we need to allow for the possibility that noncompliance with an existing rule
or regulation can be unintended. This leads to the next question.
2. Was the behavior intentional? All human error is unintentional. We mean well
but have cognitive failures or “brain cramps.” Psychologists call these “slips” or “lapses”
(Norman, 1988) and they are typically owing to limitations of attention, memory, or infor-
mation processing. As covered in Chapter 4, these types of errors increase with experience
on the job. Skilled people often put their actions on “automatic mode” and perhaps add
other behaviors to the situation. How many of us fiddle with a cassette tape or juggle a cel-
lular phone while driving? You can see how an error can easily occur. This “unconscious
incompetence” needs to be corrected but certainly not with punishment.
As discussed in Chapter 9, there is also “conscious competence.” Sometimes poor judg-
ment is used to intentionally take a risk. I often make judgment errors and take calculated
risks on the tennis court. Sometimes, I rush the net when I should not or stroke the ball long
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 218

218 Psychology of safety handbook

when trying to hit a baseline corner. What I should do in these situations is quickly refocus
my attention or reconsider the risks I took. What I often do instead, however, is engage in
a self-defeating punishment strategy. I yell at myself internally and sometimes even talk
out loud. Occasionally, I will slam the ball against the fence or toss my racket. Does this
self-critical punishment ever help? Of course not. It only makes matters worse. The same
is true for your golf game and for meeting the continuous challenge of preventing injuries
in the workplace.
The deliberate or willful aspect of a calculated risk might seem to warrant punishment,
but this will not convince people their judgment was defective and that is what is needed
to change this kind of conscious incompetence to conscious competence. People need to
willingly talk about the factors contributing to their poor judgment and calculated risks.
Then, the kind of behavior analysis detailed in Chapter 9 can be conducted, enabling the
design and implementation of a corrective action plan that can truly decrease the undesired
at-risk behavior.
3. Was a rule knowingly violated? Researchers have proposed an inverse relation-
ship between one’s experience on the job and the probability of injury from a mistake. In
other words, the more knowledge or skill we have at doing something, the less likely we
are to demonstrate poor judgment. On the other hand, the tendency to take a calculated
risk increases with experience on the job. This is human nature, and it will not be changed
with punishment.
Some errors occur because the rule or proper safe behavior was not known and it is
possible for an experienced worker to forget or inadvertently overlook a rule. As discussed
in Chapter 9, training and behavior-based observation and feedback can reduce these types
of errors, but punishment certainly will not help.
4. To what degree were other employees endangered? This question reveals the
rationale I hear most often for punishing at-risk behavior. Many managers claim they only
use punishment at their workplace for the most serious matters. This is when a certain
behavior puts the individual at risk for a severe injury or fatality or places many individu-
als in danger. Failure to lock-out a power source during equipment adjustment or repair
work is the behavior most often targeted for punishment. Some are quick to add, however,
that for punishment to be warranted this behavior must be made knowingly and willfully.
“Knowingly” means the individual knew a rule was violated (Question 3), and “will-
fully” refers to the intentional issue (Question 2). Now, if an employee willfully and know-
ingly avoided a lock-out procedure to put himself and others at-risk for injury, then the
severest punishment is relevant. Actually, this person should be fired immediately, but this
rationale for at-risk behavior is very rare.
Many dangerous behaviors are mistakes resulting from poor judgment, not an uncon-
scious or conscious desire to circumvent safety policies and hurt someone. When a
calculated risk is taken, it is not performed with the idea that someone will get hurt. As dis-
cussed in Chapter 9, specific characteristics of the work environment or culture usually enable
or even encourage a calculated risk. Thus, the next question needs careful consideration.
5. What supports the at-risk behavior? This is the most important question of all.
People do not make errors or take calculated risks in a vacuum. Poor judgment occurs for
a reason and it is important to learn an employee’s rationale for taking a risk. This leads to
truly useful corrective action.
Did the individual lack knowledge or skills? Was a demanding supervisor or peer pres-
sure involved? Did equipment design invite error with poorly labeled controls? Was the
“safe way” inconvenient, uncomfortable, or cumbersome?
Let us look at the organizational culture. Is safety taken seriously only after an injury?
Is safety performance evaluated only in terms of injuries reported per month, instead of the
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 219

Chapter eleven: Intervening with consequences 219

number of proactive process activities performed to prevent injuries? These are only some
of the questions that need to be asked. See Chapter 9 for a more complete list of questions
needing answers for a comprehensive behavior analysis. My point here is that punishment
will make answers—and an effective corrective action plan—harder to come by.
6. How often has the individual performed the at-risk behavior? A particular error
or calculated risk is analyzed and punishment considered because something called atten-
tion to its occurrence. In other words, the analysis (commonly referred to as an “investiga-
tion”) is likely a reaction to an injury. But how many at-risk behaviors typically occur before
leading to an injury? As I reviewed in Chapter 7, Heinrich (1931) estimated 300 near hits
per one major injury. Bird observed this ratio to be 600 to one (as reported in Bird and
Germain, 1997). Both Henrich and Bird presumed numerous at-risk behaviors occur before
even a near hit is experienced, let alone an injury.
So what good is it to punish one of many at-risk behaviors? If the behavior is an error,
punishment will only stifle reporting, analysis, and the development of effective corrective
action. If the probability of getting caught while taking a calculated risk is low—and it is
miniscule if you wait until an injury occurs—any threat of punishment will have little
behavioral impact.
Remember, punishment does more to inhibit involvement in safety improvement
efforts than it does to reduce at-risk behavior. So consider your observation of an at-risk
behavior a mere sample of many similar at-risk behaviors and use the occasion to stimu-
late interpersonal dialogue about ways to reduce its occurrence.
7. How often have others escaped punishment? One sure way to lose credibility
and turn a person against your safety efforts is to punish an employee for behavior that
others have performed without receiving similar punishment. If the punished employee
has seen others perform similar behaviors without punishment, the situation is viewed as
unfair. Perceived inconsistency is the root of mistrust and lowered credibility (Geller, 1998).
So why risk such undersirable impact, especially when beneficial behavioral influence is
improbable anyway.
In summary. By posing these seven questions, I hoped to show the futility of using
punishment as a corrective measure in most situations. Errors (cognitive failures and mis-
takes) are unintentional and often caused by environmental factors. When errors are inten-
tional (as in calculated risks), the person did not intend to cause an injury. Rather, there
were factors in the situation that influenced the decision to take the risk. These factors need
to be discovered and addressed.
If the threat of personal injury is not sufficient to motivate consistent safe behavior (and
it often is not), it does not help to add one more threat (punishment) to the situation. We
need open and frank discussions with the people working at risk in order to analyze and
change management practices, equipment, or organizational systems that contribute to
much of the at-risk behavior we see in the workplace. This is only possible when the threat
of punishment is removed.
Figure 11.11 summarizes this discussion about various types of at-risk behavior and
the relevance of punishment for corrective action. Although I addressed this issue with
seven different questions, the two dimensions of Figure 11.11 are most critical. When the
behavior violates a designated rule or policy, the analysis boils down to considering
whether the act was intentional and whether the system or work culture influenced the
noncompliance.
What about progressive discipline? Whenever I teach behavior management prin-
ciples and procedures, the question of how to deal with the repeat offender fre-
quently comes up. Are there not times when punishment is necessary? Does not an indi-
vidual who “willfully” breaks the rules after repeated warnings or confrontations
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 220

220 Psychology of safety handbook

System Encouraged
Yes No

Calculated Risk Calculated Risk

Yes No Punishment May

Intentional
Punishment Be Warranted

Preventable Slip, Unpreventable Slip,

Lapse, or Mistake Lapse, or Mistake

No
No No
Punishment Punishment

Figure 11.11 Punishment is only warranted when the undesirable behavior is intentional
and not encouraged by the work culture.

deserve a penalty? Through progressive discipline these individuals receive top-down

penalties, starting with verbal warning, then written warnings, and eventually dismissal.
In some cases, dismissal is the best solution for noncooperative individuals who can be
a divisive and dangerous factor in the workforce. Fortunately, this worst case scenario
is rare.
The standard progressive discipline approach in safety enforcement includes three
steps. After the third infraction, it is common to send the employee home for a certain num-
ber of days without pay. In other words, “three strikes and you’re out.” But the wrongdoer
is not out for good. The individual is usually allowed back “in the game.” Here is the crit-
ical question. Is the person a better “player” upon his or her return?
When employees are punished by being temporarily dismissed, we expect them to per-
form better when they return to work. In other words, we hope they learn something from
this demeaning punishment. We also hope the learning is more than how to avoid getting
caught next time.
Actually, whether the right or wrong kind of learning occurs in this situation depends
on one key factor—attitude. If the employee is angry and does not own up to a calculated
risk, useful learning is unlikely. If negative or hostile emotions develop in an employee as
a result of the dismissal, do not expect the employee to return to work with a more pleas-
ant and co-operative demeanor. Instead, expect a more disgruntled worker, who might
give lipservice to following the safety rules to avoid another dismissal but will likely share
a negative attitude with anyone willing to listen. As we have all experienced, “returning a
rotten apple to a barrel makes other apples it contacts rotten.”
One way to avoid this problem is not to send an employee home without pay. Instead,
dismiss the employee with pay. Grote (1995) calls this “positive discipline.” This is not about
docking one’s wages for a safety infraction. It is about finding a meaningful way to reduce
a behavioral discrepancy. Now, this is not a free vacation day by any stretch of the imagi-
nation. The employee is required to think about the calculated risk and decide what can be
done to eliminate such at-risk behavior. By not withholding wages, this evaluative process
is not tainted by a negative or hostile attitude.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 221

Chapter eleven: Intervening with consequences 221

It should be clear that one option for the employee to consider is not to return to work.
The individual should seriously consider whether he or she can meet the safety standards
of the company. Is it too difficult to perform consistently with the paradigm of safety as a
core value?
It is unlikely a person will admit to not holding safety as a core value. Thus, it is
realistic and relevant for the employee to conduct a behavioral analysis (as outlined in
Chapter 9) and then develop a personal corrective action plan for reducing the behav-
ioral discrepancy implied by the rule infraction or calculated risk. Thus, at the end
of the dismissal day(s), the ultimate deliverable is a specific list of things the employee
will do to reduce the behavioral discrepancy and realign work practice with safety as a
value.
This corrective action plan should include a specification of environmental and inter-
personal supports the individual will summon in order to meet an improve-
ment objective. For example, the employee might recommend a modification of a worksta-
tion to make the desired behavior more convenient or add an activator to the area as a
behavioral reminder. The action plan also might include a solicitation of social support by
requesting certain coworkers to offer directive and/or supportive feedback (as detailed in
Chapter 12).
It is critical for a supervisor or safety leader to review this corrective action plan as soon
as the employee returns to work. Both parties must agree that the plan is reasonable, feasi-
ble, and cost effective. It is likely mutual agreement and commitment to a suitable action
plan will require significant discussion, consensus building, and refinement of the docu-
ment. The final document of the plan also should be signed by both parties. When a per-
son signs a commitment that took some effort to develop, the probability of compliance is
greatly enhanced (Cialdini, 1993; Geller and Lehman, 1991).
How about an employee discipline council? If a student at my university is caught
cheating by an instructor or another student, his or her name is submitted to the
“University Honor Council” along with details about the incident. Students volunteer to
serve on the honor council and a “Chief Justice” is elected by the entire student body.
University faculty or staff only get involved in this discipline system when making a refer-
ral or when presenting evidence during the honor council’s fact-finding and behavior
analysis mission.
After fact finding and deliberating, the honor council might dismiss the case, recom-
mend a penalty for the alleged cheater, and/or suggest changes in the instructor’s proce-
dures or policies. In one case, an instructor was given advice on the use of different test
forms and classroom seating arrangements. In another case, a professor was advised to
eliminate his “closed book, take home” exams. The rationale behind the university honor
council is that those most affected by cheating and those most capable of gathering and
understanding the facts about alleged cheating should run the system. This disciplinary
system is administered for and by the students it serves.
Given that employees typically have the most direct influence over their peers, and
that top-down discipline usually decreases voluntary involvement in desirable safety
processes, the idea of an Employee Discipline Council seems logical and intuitively appeal-
ing. If a council of people representative of the entire work force serve the fact-finding,
analysis, and corrective-action functions of safety discipline, employee involvement would
be enhanced rather than hurt by a discipline system. Such a council could offer the guid-
ance, leadership, and counseling implied by the Latin roots of discipline—disciplina mean-
ing instruction or training and discipulus referring to a learner. Disciple was also derived
from the same Latin roots.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 222

222 Psychology of safety handbook

“Dos” and “Don’ts” of safety rewards

Now let us look at the flip side of discipline—rewards. In one-on-one situations with chil-
dren at home or in school, using positive consequences to increase desirable behavior is
straightforward and easy. However, using rewarding consequences effectively with adults
in work settings is easier said than done, especially when it comes to safety. Throughout
my 35 years of professional experience in motivational psychology, I have seen more
inappropriate reward programs in occupational safety than in any other area. This is unfor-
tunate, because the effective use of extra positive consequences is often critically important
to overcome the readily available influences supporting risky behavior.
By this point, I am sure you understand the difference between an incentive and a
reward. An incentive is an activator that promises a particular positive consequence
(a reward) when a correct behavior occurs. Disincentives, on the other hand, are activators
such as rules and policies that announce penalties for certain undesired behavior.
Remember, the motivating power of incentives and disincentives depends on following
through. Rules or policies that are not consistently and justly enforced with penalties for
noncompliance are often disregarded. If promises of rewards are not fulfilled when the
behavioral criteria are reached, subsequent incentives might be ignored. This need for con-
sistent delivery of consequences—whether positive or negative—makes it quite challeng-
ing to develop and manage an effective motivational program for safety.

Doing it wrong
Most incentive/reward programs for occupational safety do not specify behavior.
Employees are rewarded for avoiding a work injury or for achieving a certain number of
“safe work days.” So, what behavior is motivated? Not to report injuries.
If having an injury loses one’s reward, or worse, the reward for an entire work group,
there is pressure to avoid reporting that injury, if possible. Many of these nonbehavioral,
outcome-based incentive programs involve substantial peer pressure because they use a
group-based contingency. That is, if anyone in the company or work group is injured,
everyone loses the reward. Not surprisingly, I have seen coworkers cover for an injured
employee in order to keep accumulating “safe days” and not lose their chance at a reward
possibility.
These incentive programs might decrease the numbers of reported injuries, at least
over the short term, but corporate safety is obviously not improved. Indeed, such programs
often create apathy or helplessness regarding safety achievement. Employees develop the
perspective that they cannot really control their injury record, but must cheat or beat the
system to celebrate the “achievement” of an injury reduction goal.
Figure 11.12 shows how the results of an outcome-based incentive program were dis-
played to the 1800 employees of a large industrial complex. The man on the ladder (twice
life size and named “I. M. Ready”) climbed one step higher every day there was no lost-
time injury. Whenever a lost-time injury occurred, I. M. Ready fell down the ladder and
started his climb again. In addition, a red light at the entrance–exit gate flashed for 12 hours
after one lost-time injury. Every employee was promised a reward as soon as I. M. Ready
reached the top of the ladder to signify 30 days without a lost-time injury.
At first this plan activated significant awareness, even enthusiasm, for safety. No spe-
cific tools or methods were added to reduce the injury rate, however. Safety did not
improve, and I. M. Ready did not reach the top of the ladder in 2 12 years. Initial zeal for the
program waned steadily. Eventually, people stopped looking at the display. The man on
the ladder and the flashing red light were reminders of failure. Most of the employees
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 223

Chapter eleven: Intervening with consequences 223

Figure 11.12 An outcome display of progress can activate feelings of helplessness and
demoralize a workforce.

were not personally responsible for the failure, yet they did not know what to do to stop
the injuries. Many workers became convinced they were not in direct control of safety at
their facility and developed a sense of learned helplessness (Seligman, 1975) about pre-
venting lost-time injuries.
There is a happy ending to this story. The outcome-based incentive program was
dropped, and a process-based approach was implemented. I taught an incentive steering
committee the guidelines presented below for doing it right, and the committee worked
out the details. After about six months, I. M. Ready reached the top of the ladder and a
plantwide celebration commemorated the achievement.

Doing it right
Here are seven basic guidelines for establishing an effective incentive/reward program
to motivate the occurrence of safety-related behaviors and improve industrial health and
safety.

1. The behaviors required to achieve a safety reward should be specified and per-
ceived as achievable by all participants.
2. Everyone who meets the behavioral criteria should be rewarded.
3. It is better for many participants to receive small rewards than for one person to
receive a big reward.
4. The rewards should be displayed and represent safety achievement. Coffee mugs,
hats, shirts, sweaters, blankets, or jackets with a safety message are preferable to
rewards that will be hidden, used, or spent.
5. Contests should not reward one group at the expense of another.
6. Groups should not be penalized or lose their rewards for failure by an individual.
7. Progress toward achieving a safety reward should be systematically monitored and
publicly posted for all participants.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 224

224 Psychology of safety handbook

Figure 11.13 Raffle drawings that result in few “lucky” winners and many “unlucky” los-
ers can do more harm than good.

Guideline 2 recommends against the popular lottery or raffle drawing. As illustrated

in Figure 11.13, a lottery results in one “lucky” winner being selected and a large num-
ber of “unlucky” losers. The announcement of a raffle drawing might get many people
excited, and if lottery tickets are dispensed for specific safe behaviors, there is some moti-
vational benefit. Eventually, however, the valuable reward is received by a lucky few. Also,
I perceive a disadvantage in linking chance with safety. It is bad enough we use the word
“accident” in the context of safety processes, as I pointed out earlier in Chapter 3.
I have worked with a number of safety directors who used a lottery incentive program
and vowed they would never do it again. Volk (1994) interviewed a number of safety direc-
tors who verified my observations. The big raffle prize, such as a snowmobile, pick-up
truck, or television set, was displayed in a prominent location. Everyone got excited—tem-
porarily—about the possibility of winning. Their attention was directed, however, at the
big prize instead of the real purpose of the program: to keep everyone safe. The material
reward in an incentive program should not be perceived as the major payoff. Incentives are only
reminders to do the right thing, and rewards serve as feedback and a statement of appre-
ciation for doing the right thing.
More important than external rewards is the way they are delivered. Rewards should
not be perceived as a means of controlling behavior but as a declaration of sincere gratitude
for making a contribution. If many people receive this recognition, you have many deposits
in the emotional bank accounts of potential actively caring participants in a Total Safety
Culture. That is why it is better to reward many than few (Guideline 3).
When rewards include a safety logo or message (Guideline 4), they become activators
for safety when displayed as illustrated in Figure 11.14. Also, if the safety message or logo
was designed by representatives from the target population, the reward takes on special
meaning (as discussed previously in Chapter 10). Special items like these cannot be pur-
chased anywhere and, from the perspective of internal consequences, they are more valu-
able than money.
As portrayed in Figure 11.15, contests that pit one group against another can lead to an
undesirable win –lose situation (Guideline 5). Safety needs to be perceived as win–win.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 225

Chapter eleven: Intervening with consequences 225

Figure 11.14 Rewards with safety messages are special to those who earn them.

Figure 11.15 Safety contests can motivate unhealthy competition.

This means developing a contract of sorts between each employee that makes everyone a
stockholder in achieving a Total Safety Culture. Everyone in the organization is on the
same team. Team performance within departments or work groups can be motivated by
providing team rewards or bonuses for team achievement. Every team that meets the
“bonus” criteria should be eligible for the reward. In other words, Guideline 2 should be
applied when developing incentive/reward programs to motivate team performance.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 226

226 Psychology of safety handbook

Penalizing groups for individual failure (Guideline 6) reflects a problem I have seen
with many outcome-based incentive/reward programs. The problem is typified in the I. M.
Ready program. It is certainly easy to administer a contingency that simply withdraws
reward potential from everyone whenever one person makes a mistake. This can do more
harm than good, however. As discussed earlier, it can promote unhealthy group pressure
and develop feelings of helplessness or lack of personal control. Displaying the results of
such a program only precipitates these undesirable perceptions and expectations.
On the other hand, when the incentive/reward program is behavior-based and per-
ceived as equitable and fair, it is advantageous to display progress toward reaching indi-
vidual, team, or company goals (Guideline 7). When people see their efforts transferred to
a feedback chart, their motivation and sense of personal control is increased, or at least
maintained.
Obviously, developing and administering an effective incentive/reward program for
safety requires a lot of dedicated effort. There is no quick fix, but it is worth doing, if you
take the time to do it right. As Daniels wisely stated, “If you think this is easy, you are doing
it wrong” (2000, page 179). Let us examine an exemplary case study.

An exemplary incentive/reward program

In 1992, I consulted with the safety steering committee of a Hoechst Celanese company of
about 2000 employees to develop a plant-wide incentive program that followed each of the
guidelines given previously. The steering committee, including four hourly and four salary
employees, met several times to identify specific behavior-consequence contingencies.
That is, they needed to decide what behaviors should earn what rewards. Their plan was
essentially a “credit economy” where certain safe behaviors, which could be achieved by
all employees, earned certain numbers of “credits.”
At the end of the year, participants exchanged their credits for a choice of different
prizes, all containing a special safety logo. The variety of behaviors earning credits
included attending monthly safety meetings; special participation in safety meetings; lead-
ing a safety meeting; writing, reviewing, and revising a job safety analysis; and conducting
periodic audits of environmental and equipment conditions, and certain work practices.
For a work group to receive credits for audit activities, the results of environmental and
PPE observations had to be posted in the relevant work areas. Only one behavior was
penalized by a loss of credits—the late reporting of an injury.
At the start of the new year, each participant received a “safety credit card” for tallying
ongoing credit earnings. Some individual behaviors earned credits for the person’s entire
work group, thus promoting group cohesion and teamwork. The audit aspects of this
incentive/reward program exemplify a basic behavior-based principle for health and
safety management—observation and feedback. Employees were systematically observed,
and they received soon, certain, and positive feedback (a reward) after performing a target
behavior. An incentive/reward program is only one of several methods to increase safe
work practices with observation and feedback. In the next chapter, I address feedback more
specifically as an external and extra consequence to prevent injuries.

Safety thank-you cards

I would be remiss if I did not describe “Safety Thank-You Cards” in a discussion of exem-
plary incentive/reward approaches. Figure 11.16 depicts a thank-you card that was avail-
able to all employees for distribution to coworkers whenever they observed them going
out of their way for another person’s safety (Roberts and Geller, 1995). The types of actively
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 227

Chapter eleven: Intervening with consequences 227

Front Back

C.C. Manufacturing
Examples of ACTIVELY CARING Behaviors
Thank You for ACTIVELY CARING

Please describe specifically the observed ACTIVELY

Date:
.. Recognizing and correcting an unsafe condition

.
CARING behavior: (see back for examples) Reminding a coworker not to perform an unsafe
act.

.
Removing or cleaning unsafe objects or debris
from a work area.

.
Giving positive feedback to a coworker for
working safely.

.
Reporting a near miss.

.
Making a task safer.
Observer's Code:
The first letter of the The first letter of your The number of the
Other
city where you were born mother's maiden name month you were born

Recipient's Code:
The first letter of the The first letter of your The number of the
city where you were born mother's maiden name month you were born
Hoechst Celanese

φ Thank You Limit: 55 φ

Elaine George Dave Salyer

Observer's Name
Tom Tillman Jim Woods
Recipient's Name
Department 1490

Figure 11.16 Employees can use a thank-you card to recognize each other’s safe behavior.

caring behaviors warranting recognition were listed on the back of the card, and involved
such things as suggesting a safer way to perform a task, pointing out a potential hazard
that might have been overlooked, or going beyond the call of duty to help another person
avoid an at-risk behavior.
Over the years, I have seen a wide variety of thank-you cards designed by work teams
and used successfully at a number of industrial sites, including Abbott Laboratories, Exxon
Chemical, Ford, General Motors, Hercules, Hoechst Celanese, Kal Kan, Logan Aluminum,
Phillip Morris, Westinghouse Hanford Company, and Weyerhaeuser,.
At some locations, thank-you cards were used in a raffle drawing, exchangeable for
food, drinks, or trinkets, or displayed on a plant bulletin board as a “safety honor roll.”
Sometimes the cards could be accumulated and exchanged for tee shirts, caps, or jackets
with messages or logos signifying safety achievement. At several plants, the person who
delivered a thank-you card returned a receipt naming the recognized employee and
describing the behavior earning the consequence, thus creating objective information to
define a “Safe Employee of the Month” (Geller, 1990).
At a few locations, the thank-you cards took on a special actively caring meaning.
Specifically, when deposited in a special collection container, each thank-you card was worth
25¢ toward corporate contributions to a local charity or to needy families in the community.
The actively caring card used at the Hoechst Celanese plant in Rock Hill, SC, is shown in
Figure 11.17. The back of the card included a colorful peel-off symbol which the recognized
employee could affix in any number of places as a personal reminder of the recognition. I
was surprised but pleased to see a large number of these thank-you stickers on employees’
hard hats. Obviously, this actively caring thank-you approach to safety recognition has great
potential as an inexpensive but powerful tool for motivating safe behavior.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 228

228 Psychology of safety handbook

I Thanks
for

From

front of card

Help this card

Makes A
Difference

Please deposit
this in the
collection box

"Hand-N-Hand
For Savety We
Stand"

back of card

Figure 11.17 This Actively Caring Thank-You Card offers reward leverage.

Motivational leverage with this simple actively caring thank-you card was illustrated
a few years ago at the Hercules chemical plant in Portland, OR. The actively caring cards
delivered for safety-related behaviors were similar to the one illustrated in Figure 11.17,
except the peel-off sticker depicted the company logo—a rhinoceros. The
incentive/reward contingency was simply stated. Give an Actively Caring Thank-You
Card and “Rhino Sticker” to anyone who goes beyond the call of duty for safety or health.
Here is special motivational leverage. Every actively caring card received and then
deposited in a designated “Actively Caring for Others” box was worth $1.00 to purchase
toys for disadvantaged children in and around Portland. With this program, the 64
line workers at this chemical plant contributed more than $1750 during the Christmas hol-
idays of 1996. Guess who picked out and delivered the toys? Children of the employees.
Now that is special actively caring leverage from a simple behavior-based incentive/
reward program.

The “Mystery Observee” program

The “Mystery Observee” incentive program developed and implemented at NORPAC
paper mill in Longview, WA, is exemplary for its ingenious way of targeting the right
behavior, its general, practical, and cost-effective applicability to almost any behavior-
based observation and feedback process, and its potential to add a fun and constructive
diversion to the standard work routine. The program started with 35 of 450 mill workers
agreeing to be a “mystery observee.” Each of these volunteers received a coupon
redeemable for a meal for two at a local restaurant.
The mystery observees were not to tell anyone they had the reward coupon. However,
when the mystery observees received coaching in a behavior-based observation and feed-
back process, they gave the reward coupon to their coach. More specifically, each week
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 229

Chapter eleven: Intervening with consequences 229

employees were asked to complete a critical behavioral checklist on a coworker (with per-
mission) and then present the results in a positive one-to-one feedback session. At the end
of the feedback conversation, the observer received a reward coupon if he or she happened
to select a mystery observee to coach. Then this coach became one of 35 mystery observees,
anticipating an opportunity to reward another coworker for completing a one-to-on behav-
ioral observation and feedback session.
You can see how this simple inexpensive incentive/reward program was both pleas-
ant and constructive. It got people talking about the behavior-based coaching process in
positive terms and it rewarded the most difficult aspect of a behavioral coaching process—
interpersonal feedback. It is relatively easy to complete a critical behavioral checklist (CBC)
compared with relaying the CBC results to an observee in a positive and constructive inter-
personal conversation. The potential reward for completing this last and most important
aspect of behavioral coaching added an element of fun to the whole process. It made it eas-
ier to transition from behavioral observation to interpersonal feedback.
This successful mystery observee program illustrates an important principle in incen-
tive/reward programs. You get what you reinforce. Programs that reward employees for
handing in a completed CBC will probably increase the number of checklists received, but
how about the quality of the CBC? Will the number of constructive comments on a CBC
decrease when a reward is given for quantity? You can count on this for employees who
view the reward as a “payoff” for their efforts. That is why it is important to educate peo-
ple about the rationale and true value of a particular safety effort. Then the big payoff is
injury prevention, and the extra reward can be perceived as a “token of appreciation” for
heartfelt participation.
It is possible, however, that some people will participate for the reward. Thus, it is cru-
cial to consider carefully what specific behavior is most desirable in a safety-related
process. The NORPAC employees believed the preeminent feature of interpersonal coach-
ing is the one-on-one feedback discussion. Thus, they linked the reward to this phase of
their behavioral safety coaching process.

In conclusion
Writing this book was challenging, tedious, overwhelming, tiresome, sacrificing, and
exhausting. Observers were apt to say I was self-directed and intrinsically motivated. Of
course, I know better, and you do too.
Incidentally, I literally wrote the various drafts of this text. I have never learned to type
and, therefore, have never benefited from the technological magic of computer word pro-
cessing. My colleagues explain that it is not necessary to be a skilled typist to reap the many
intrinsic benefits of preparing a manuscript on a computer. “I type slowly with only one
finger,” some say, “and still enjoy the wonderful benefits of high-tech computer word pro-
cessing. I could never go back to preparing a manuscript by hand. You don’t know what
you’re missing.”
I am sure you have noticed my disparate uses of “intrinsic” in the prior paragraphs and
you now understand the two meanings of this popular motivational term. Are my friends
and colleagues so enthusiastic about computer-based word processing because of intrinsic
(internal) motivation or because of intrinsic (natural) consequences linked to their com-
puter use? As a review of this chapter, I am sure you see my point.
Word processing on a computer allows for rapid “quick-fix” control of letters, words,
sentences, and paragraphs. Computer users also can walk to a printer and obtain a typed,
“hard copy” of their document for study, revision, or dissemination. All these soon, certain,
positive consequences are connected naturally to word-processing behavior. No wonder
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 230

230 Psychology of safety handbook

my friends and colleagues are motivated about computer word processing and urge me to
get on the high-tech “band wagon.”
While I did not reap the benefits of fast computer turnaround, there were plenty of
external and natural consequences to keep me going. I experienced a sense of rewarding
satisfaction (internal consequence) from seeing my thoughts and ideas take form. Almost
daily, I gave my writing to a secretary who processed my writing on computer disk (yes, I
do realize this is a step I would not need if I were computer literate). I got significant satis-
faction (or the internalization of external consequences) from reading and refining the
typed text. The next day my secretary delivered a refined version—another external con-
sequence from my work.
Throughout the writing process, I worked with a very talented illustrator who pro-
vided me with soon, certain, and positive consequences to feed my motivation. We talked
about concepts I wanted to portray, and in a few days I saw his artwork. Sometimes I
judged it ready. Other times we discussed revisions, and within a week or so I examined
the fruits of our discussions.
Continuous feedback from others was invaluable as a motivator and necessary mech-
anism for continuous improvement. As soon as a chapter appeared close to my internal
standard, I distributed copies to about ten colleagues and friends who had expressed inter-
est in reading early drafts and offering feedback. The feedback I received from these earlier
versions was valuable in refining this text and in motivating my progress. Feedback is obvi-
ously a powerful consequence intervention for improving behavior. In the next chapter,
I shall discuss ways to maximize the beneficial impact of feedback.
When we earn genuine appreciation and approval from others for what we do, we not
only become self-motivated, but we also maximize the chances that our activities will influ-
ence the behavior of others. In fact, this was my ultimate motivation for soliciting feedback
on earlier drafts of this text and for painstakingly refining the presentations. Practice can only
improve with feedback. We can only learn to communicate more effectively if we learn how
we are coming across to others. If I communicate effectively and earn the approval and appre-
ciation of readers for the principles and procedures presented in this text, injuries and fatali-
ties could be reduced on a large scale. This would be an external and natural consequence of
authoring this text—the remote but preeminent motivator for my writing behavior.

References
Azrin, N.H. and Holz, W.C., Punishment, in Operant Behavior: Areas of Research and Application,
Honig, W.K., Ed., Appleton-Century-Crofts, New York, 1966.
Baron, R. A., Violence in the Workplace: a Prevention and Management Guide for Businesses, Pathfinders
Publishing of California, Ventura, 1993.
Bird, F. E., Jr. and Germain, G. L., The Property Damage Accident, FEBCO, Inc., Loganville, GA, 1997.
Cameron, J. and Pierce, W. D., Reinforcement, reward, and intrinsic motivation: a meta-analysis,
Rev. Educ. Res., 64, 363, 1994.
Carnegie, D., How to Win Friends and Influence People, 1981 ed., Simon & Schuster, New York, 1936.
Carr, C., Mawhinney, T., Dickinson, A., and Pearlstein, R., Punished by rewards? A behavioral per-
spective, Perform. Improve Q., 8(2), 125, 1995.
Chance, P., Learning and Behavior, 4th ed., Brooks/Cole Publishing, Pacific Grove, CA, 1999.
Cialdini, R. B., Influence: Science and Practice, 3rd ed., Harper Collins College Publishers, New York,
1993.
Covey, S. R., The Seven Habits of Highly Effective People, Simon & Schuster, New York, 1989.
Daniels, A. C., Bringing out the Best in People, 2nd ed., McGraw-Hill, New York, 2000.
Deming, W. E., Quality, productivity, and competitive position, four-day workshop presented by
Quality Enhancement Seminars, Inc., Cincinnati, OH, May 1991.
820049_CRC1_L1540_CH11 11/10/00 1:54 PM Page 231

Chapter eleven: Intervening with consequences 231

Deming, W. E., The New Economics, MIT Press, Boston, MA, 1993.
Deci, E. L., Intrinsic Motivation, Plenum, New York, 1975.
Deci, E. L. and Ryan, R. M., Intrinsic Motivation and Self-Determination in Human Behavior, Plenum,
New York, 1985.
Fulghum, R., All I Really Need to Know I Learned in Kindergarten, Ivy Books, New York, 1988.
Flora, S. R., Undermining intrinsic interest from the standpoint of a behaviorist, Psychol. Rec., 40,
323, 1990.
Geller, E. S., Performance management and occupational safety: start with a safety belt program,
J. Organ. Behav. Manage., 11(1), 149, 1990.
Geller, E. S., The truth about safety incentives, Prof. Saf., 42(1), 40, 1996.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998.
Geller, E. S. and Lehman, G. R., The buckle-up promise card: a versatile intervention for large-scale
behavior change, J. Appl. Behav. Anal., 24, 91, 1991.
Grote, D., Discipline without Punishment, American Management Association, New York, 1995.
Heinrich, H. W., Industrial Accident Prevention, McGraw-Hill, New York, 1931.
Horcones, The concept of consequences in the analysis of behavior, Behav. Anal., 10, 291, 1987.
Horcones, Natural reinforcement: a way to improve education, in The education crisis: issues, per-
spectives, solutions, Geller, E. S., Ed., Monograph No. 7, J. Appl. Behav. Anal., Society for the
Experimental Analysis of Behavior, Inc., Lawrence, KS, 1992.
Kohn, A., Punished by Rewards: The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other
Bribes, Houghton Mifflin, Boston, MA, 1993.
Norman, D. A., The Psychology of Everyday Things, Basic Books, New York, 1988.
Oliver, S. D., West, R. C., and Sloane, H. N., Some effects on human behavior of aversive events,
Behav. Ther., 5, 481, 1974.
Peale, N. V., The Power of Positive Thinking, Prentice-Hall, New York, 1952.
Pearlstein, R. B., Rewarded by punishment, Performance Improvement Q., 8(2), 136, 1995.
Roberts, D. S. and Geller, E. S., An “actively caring” model for occupational safety: a field test, Appl.
Prev. Psychol., 4, 53, 1995.
Sandman, P. M., Risk  Hazard  Outrage: a formula for effective risk communication, Videotaped
presentation for the American Industrial Hygiene Association, Environmental Communication
Research Program, Cook College, Rutgers University, New Brunswick, NJ, 1991.
Seligman, M. E. P., Helplessness, W. H. Freeman, San Francisco, CA, 1975.
Sidman, M., Coercion and Its Fallout, Authors Cooperative, Boston, MA, 1989.
Silverstein, B. A., Fire, L. J., and Armstrong, T. J., Occupational factors and carpal tunnel syndrome,
Am. J. Ind. Med., 11, 343, 1987.
Skinner, B. F., The Behavior of Organisms, Copley Publishing, Acton, MA, 1938.
Skinner, B. F., Science and Human Behavior, Free Press, New York, 1953.
Skinner, B. F., Verbal Behavior, Appleton-Century-Crofts, New York, 1957.
Skinner, B. F., About Behaviorism, Alfred A. Knoff, New York, 1974.
Skinner, B. F., Contrived reinforcement, Behav. Anal., 5, 3, 1982.
Sulzer-Azaroff, B. and Mayer, G. R., Behavior Modification Procedures for School Personnel, Dryden
Press, New York, 1972.
Sulzer-Azaroff, B. and Mayer, G. R., Achieving Educational Excellence Using Behavioral Strategies, Holt,
Rinehart & Winston, New York, 1986.
Sulzer-Azaroff, B. and Mayer, G. R., Behavior Analysis for Lasting Change, Holt, Rinehart & Winston,
New York, 1991.
Sulzer-Azaroff, B., Is back to nature always best?, in The education crisis: issues, perspectives, solu-
tions, Geller, E. S., Ed., Monograph No. 7, Journal of Applied Behavior Analysis, Society for the
Experimental Analysis of Behavior, Lawrence, KS, 1992.
Vaughan, M. E. and Michael, J., Automatic reinforcement: an important but ignored concept,
Behaviorism, 10, 217, 1982.
Volk, D., Learn the do’s and don’ts of safety incentives, Saf. Health, xx, 54, 1994.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 233

chapter twelve

Intervening as a behavior-change
agent
This chapter presents the principles and procedure of safety coaching—a key behavior-change
process for safety improvement. The letters of COACH represent the critical sequential steps of
safety coaching: Care, Observe, Analyze, Communicate, Help. This coaching process is clearly rele-
vant for improving behaviors in areas other than safety and in settings other than the workplace.
Behavior-based feedback is critical for improvement in everything we do. This chapter shows you
how to give feedback effectively.

“There are risks and costs to a program of action. But they are far less than the long
range risks and costs of comfortable inaction.”—John F. Kennedy

The prior two chapters discussed guidelines for developing behavior change interventions.
Chapter 10 focused on the use of activators to direct behavior change. Chapter 11 detailed
the motivating role of consequences. Several examples employed both activators and con-
sequences. This is applying the three-term contingency (activator–behavior–consequence)
and is usually the most influential approach.
Large-scale behavior change is impossible without intervention agents—people will-
ing and able to step in on behalf of another person’s safety. In a Total Safety Culture, every-
one feels responsible for safety, pursuing it daily. They go beyond the call of duty to
identify at-risk conditions and behaviors and intervene to correct them (Geller, 1994). This
chapter is about becoming a behavior-focused change agent.
In simplest terms, this means observing and supporting safe behaviors or observing
and correcting at-risk behaviors. It might involve designing and implementing a particu-
lar intervention process for a work team, organizational culture, or an entire community. It
might mean merely engaging in behavior-focused communication between an observer
(the intervention agent) and the person observed. This is safety coaching (Geller, 1995) and,
to be effective, certain principles and guidelines need to be followed.

Selecting an intervention approach

The number of ways to intervene on behalf of safety by using activators, consequences, and
their combination is almost limitless. A steering committee for an organization or commu-
nity needs to design specific procedures for each aspect of the three-term contingency.
What are the desired (or undesired) target behaviors? How will the target behavior(s) be
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 234

234 Psychology of safety handbook

activated? What consequences can be employed to motivate behavior change or support

the occurrence of desired behavior?

Various intervention approaches

Figure 12.1 gives brief definitions of 23 different ways to use activators and consequences
for improving safety-related behaviors. These come from the research literature on tech-
niques to change behaviors at individual and group levels. Representative sources include

ACTIVATORS
1. Lecture: Unidirectional oral communication concerning the rationale for specific
behavior change.
2. Demonstration: Illustrating the desired behavior for target subject(s).
3. Policy: A written document communicating the standards, norms, or rules for desired
behavior in a given context.
4. Written Activator: A written communication that attempts to prompt desired behavior.
5. Commitment: A written or oral pledge to perform a desired behavior.
6. Discussion/Consensus: Bidirectional oral communication between the deliverers and
receivers of an intervention program.
7. Oral Activator: An oral communication that urges desired behavior.
8. Assigned Individual Goal: One person decides for another person the level of desired
behavior he or she should accomplish by a certain time.
9. Individual Goal: An individual decides the level of desired behavior (the goal) that
should be accomplished by a specific time.
10. Individual Competition: An intervention promotes competition between individuals
to see which person will accomplish the desired behavior first (or best).
11. Individual Incentive: An announcement to an individual in written or oral form of the
availability of a reward following one or more designated behaviors.
12. Individual Disincentive: An announcement to an individual specifying the possibility
of receiving a penalty following one or more undesired behaviors.
13. Assigned Group Goal: A group leader decides the level of desired behavior a group
should accomplish by a certain time.
14. Group Goal: Group members decide for themselves a level of group behavior they
should accomplish by a certain time.
15. Group Competition: An intervention promotes competition between specific groups to
see which group will accomplish the desired behavior first (or best).
16. Group Incentive: An announcement specifying the availability of a group reward
following the occurrence of desired group behavior.
17. Group Disincentive: An announcement specifying the possibility of receiving a group
penalty following the occurrence of undesired group behavior.
CONSEQUENCES
18. Individual Feedback: Presentation of either oral or written information concerning an
individual’s desired or undesired behavior.
19. Individual Reward: Presentation of a pleasant item to an individual, or the withdrawal
of an unpleasant item from an individual for performing desired behavior.
20. Individual Penalty: Presentation of an unpleasant item to an individual, or the
withdrawal of a pleasant item from an individual following undesired behavior.
21. Group Feedback: Presentation of either oral or written information concerning a
group’s desired or undesired behavior.
22. Group Reward: Presentation of a pleasant item to a group, or the withdrawal of an
unpleasant item from a group or team following desired group behavior.
23. Group Penalty: Presentation of an unpleasant item to a group or the withdrawal of a
pleasant item from a group or team following undesired group behavior.

Figure 12.1 Various intervention techniques are available to influence behavior. Adapted
from Geller (1998a). With permission.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 235

Chapter twelve: Intervening as a behavior-change agent 235

Cone and Hayes (1980), Glenwick and Jason (1980, 1993), Geller et al. (1982), and most
research articles published in the Journal of Applied Behavior Analysis from 1968 until the
present (for example, Greene et al., 1987).
The first 17 approaches are activators, occurring before the target behavior is per-
formed. They attempt to persuade or direct people, can be classified as passive or active,
and target individuals or groups. The three basic types of consequence approaches—
reward, penalty, and feedback—can be given to an individual or to a group. Therefore,
Figure 12.1 defines six different consequence procedures.
As discussed earlier, most interventions consist of various techniques listed in
Figure 12.1. Education and training programs to promote safety and health, for example,
often include discussion/consensus building; demonstrations; lectures; and written
activators, including signs, newsletters, memos, or verbal reminders. Role-playing
exercises employ antecedent instructions to direct desired behaviors and consequen-
tial feedback to support what is right and correct what can be improved. An incentive/
reward program requires a variety of activators (incentives) to announce the availability of
a reward for certain behaviors, and the consequence (reward) can be given to individuals
or groups.
Also, as depicted in Figure 12.2, various items or events can be used for rewards, from
special privileges and promotional items to special individual or team recognition. Notice
that receiving a reward for particular behavior is a form of feedback—information regard-
ing the occurrence of desired behavior. A person can receive feedback, however, without
acquiring any of the rewards listed in Figure 12.2. Safety coaching, for instance, always
involves feedback. When the coaching process recognizes safety behavior, it is usually per-
ceived as a social attention reward.

Industry Privileges Exchangeable Tokens

Time off Cash
Extra Break Food coupon
Refreshments Ticker to an event
Preferred parking Rebate coupon
Special assignment Gift certificate

Useful Items Chance to Win a Contest

Coffee mug Lottery ticket
Litter bag, Car wax Bingo number
Tire gauge Poker card, Game symbol
Umbrella, Pocket knife Raffle coupon
Flashlight, Pen

Promotional Items Social Attention

Safety button Name in newspaper
Bumper sticker Posted picture
Key chain Letter of commendation
Hardhat sticker T. V. interview
T-shirt Handshake, Thank-you card

Figure 12.2 A variety of possible rewards are available to motivate safe behaviors in orga-
nizational settings.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 236

236 Psychology of safety handbook

Multiple intervention levels

Interventions fluctuate widely in terms of cost, administrative effort, and participant
involvement. Written activators like signs, memos, and newsletter messages are relatively
effortless. Other activators like demonstrations, commitment techniques, and consensus-
building discussions take considerable time and effort to design and deliver. As discussed
in Chapter 11, implementing an incentive/reward process correctly requires continuous
attention and periodic refinement from a team of intervention agents.
Some people benefit from the simplest and least expensive interventions, such as signs
or policy statements specifying the correct behavior for a certain situation. However, for a
variety of reasons, many of which were considered in Section 2, some people do not alter
their at-risk behavior after reading a simple safety message. They need to be motivated
and, therefore, require a more intensive, intrusive, and expensive intervention. Exposing
them to more signs, posters, and memos is generally a waste of time.
By the same token, when people do the right thing following the least intrusive inter-
vention technique, such as passive activators, it is a waste of time and effort to target them
with a motivational intervention. As discussed in Chapter 9, these “consciously competent”
individuals may need supportive intervention (such as interpersonal recognition) to become
more fluent. You can help them become more responsible and self-directed (Geller, 1998b;
Watson and Tharp, 1997) if you enroll these folks as agents of change (Katz and Lazarfeld,
1955). In other words, do not “preach to the choir”—enlist the “choir” to preach to others.
Figure 12.3 depicts a multiple intervention level (MIL) hierarchy (adapted from Geller,
1998a). It summarizes the important points about different levels of intervention impact
and intrusiveness and illustrates the need to multiply the number of intervention agents as
the level of intensity and intrusiveness increases.
At the top of the hierarchy (Level 1), the interventions are least intense and intrusive.
They are instructive and target the maximum number of people for the least cost per per-
son. At this level, an intervention is designed to have maximum large-scale appeal, while
minimizing contact between target individuals and intervention agents. Those showing
the desired target behavior at a particular intervention level need to become more fluent.
In this regard, supportive intervention can be useful, as discussed in Chapter 9. However,
individuals unaffected by initial exposures to a particular intervention level will “fall
through the cracks.” Repeated exposure to interventions at the same level will have no
effect. These individuals require a more influential and costly intervention approach.
This hierarchy lists four intervention levels, but to date this number has not been
empirically verified. Each level has height, length, and width, representing different char-
acteristics of the intervention. The height of each intervention box represents the “financial
investment per person” to participate in or experience that particular intervention. Notice
how the investment per person increases as the levels increase. The length of each box rep-
resents the “probability” that an individual will be influenced to change his or her behav-
ior as a result of experiencing that intervention. This probability increases as the level
increases.
The width of each intervention level is marked with the letters “A,” “B,” “C,” etc. This
indicates repeated applications of the same intervention approach. The hierarchy predicts
that repeating the same intervention over and over typically will not influence additional
people. This is because those who were susceptible to changing their behavior at that level
have already done so. The others have fallen through the cracks. Let us look at some
examples.
A Level 1 intervention might include relatively inexpensive signs, posters, or bill-
boards with safety messages or slogans. Placed around the plant, people typically notice
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 237

Chapter twelve: Intervening as a behavior-change agent 237

Figure 12.3 When people are not influenced by interventions at one impact level, they fall
through the cracks and need a more intrusive intervention. Adapted from Geller (1998a).
With permission.

them when they are first put in place or when the message changes, but soon the activators
are forgotten or ignored. A certain percentage of the plant population might change their
behavior and perform more safely as a result, but others are not influenced by this level of
intervention intrusiveness.
For these people, a Level 2 intervention is in order. This might include weekly safety
meetings for each work group, where employees talk about safety issues for an allotted
time. Meetings require more participation and involvement from the employees. They give
participants a greater sense of personal control over safety and offer more opportunities for
social support. This instructional intervention should have a greater impact. Also, owing
to the more intense nature of Level 2 programs, the personal investment per participant has
increased. This will likely add more converts to our safety task force, but for nonpartici-
pants it will not be enough.
So these employees “fall through the cracks” again, to be faced with a Level 3 inter-
vention. This might be an incentive/reward program or a goal-setting and feedback
process. Of course, these motivational techniques require more time and effort from both
intervention agents and participants.
Finally, notice the large arrow on the left of the diagram in Figure 12.3. It indicates
that once individuals are influenced at any given level, they can participate in the next
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 238

238 Psychology of safety handbook

intervention level as a behavior-change agent, helping others improve their safety per-
formance. Higher level interventions require more change agents. The highest level inter-
vention for safety is one-on-one, the level for safety coaching.

Increasing intervention impact

Based on an extensive literature review and our own studies in safety-belt promotion, my
students and I (Geller et al., 1990) proposed that the success of any intervention program is
a direct function of

1. The amount of specific response information transmitted by the intervention.

2. The degree of external consequence control.
3. The target individual’s perception of personal control or personal choice regarding the
behavior change procedures.
4. The degree of group cohesion or social support promoted.
5. The amount of participant involvement facilitated by the intervention.

Response information varies according to the amount of new behavioral knowledge

transmitted by the intervention. As discussed in Chapter 10, this can be facilitated by
increasing the salience of the information presented and the proximity between the time a
behavioral request is made and the opportunity or ability to perform the desired response.
External control is determined by the type of behavior-consequence contingency used; as
covered in Chapter 11, the nature of the consequence will influence the target individual’s
perception of personal control over the behavior change procedures.
Negative consequences and nongenuine or insincere positive consequences decrease
personal control—and the long-term benefits of an intervention process. When people
get to choose aspects of an intervention, such as which behaviors to focus on and what
rewards to offer, their perceptions of personal control increase, and the intervention’s
impact is enhanced. The concept of personal control is discussed more completely in
Chapters 14 and 15.
Social support is shaped by the amount of peer, family, or friend encouragement result-
ing from the intervention process. Person factors, such as an individual’s natural tendency
to interact in group settings and various group dynamics, such as degree of group cohesion
or sense of belonging, also affect the amount of perceived social support associated with an
intervention. The person-based aspects of perceived social support are discussed in more
detail in Chapter 14.
Participant involvement in an intervention process also depends on certain person fac-
tors, including an individual’s degree of introversion vs. extroversion (Eysenck, 1976;
Eysenck and Eysenck, 1985) and perceived locus of control (Strickland, 1989). Extroverts,
for example, usually participate more than introverts in interventions involving a high
level of activity and social interaction. In addition, people with an internal locus of control
typically prefer situations that allow them greater personal control rather than being at the
mercy of others or chance factors. The reverse is true for individuals with an external locus
of control (Phares, 1973).
More important than person factors in determining intervention involvement is the
ratio of number of intervention agents to individuals in the target population. More inter-
vention agents per target population usually promote greater participation. A one-to-one
agent-to-target ratio is the highest level of intervention intensity and effectiveness, and
occurs in safety coaching. This is the ultimate intervention for safety, and the remainder of
this chapter details the ingredients of an effective safety coaching intervention process.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 239

Chapter twelve: Intervening as a behavior-change agent 239

Intervening as a safety coach

Coaching is essentially a process of one-on-one observation and feedback. The coach sys-
tematically observes the behaviors of another person and provides behavioral feedback
on the basis of the observations. Safety coaches recognize and support the safe behaviors
they observe and offer constructive feedback to reduce the occurrence of any at-risk behav-
iors. This chapter specifies the steps of safety coaching, points out trainable skills needed
to accomplish the process, and illustrates tools and support mechanisms for increasing
effectiveness.

Athletic coaching vs. safety coaching

The term “coach” is very familiar to us in an athletic context. In fact, winning coaches prac-
tice the basic observation and feedback processes needed for effective safety coaching.
They follow most of the guidelines reviewed here. As illustrated in Figure 12.4, the most
effective team coaches observe the ongoing behaviors of individual players and record
their observations in systematic fashion, using a team roster, behavioral checklist, or
videotape.
Football coaches, for instance, spend hours and hours analyzing film. Then they
deliver specific and constructive feedback to team members to instruct, support, or moti-
vate desirable behavior and/or to decrease undesirable behavior. Sometimes the feedback
is given in a group session, perhaps by critiquing videotapes of team competition. At other
times, the feedback is given individually in a personal one-on-one conversation. Usually,
the one-to-one format has greater impact on individual performance.
Differential Acceptance. The most effective athletic coaches communicate feedback
so team members learn from the exchange and increase their motivation to continuously

Figure 12.4 Systematic observation and feedback are key to effective coaching.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 240

240 Psychology of safety handbook

improve. The same is true for safety coaches. The basic principles and procedures for effec-
tive coaching are the same whether communicating behavioral feedback to improve ath-
letic or safety performance. It is usually more challenging, however, to coach for safety than
for sports.
People are generally more accepting of information to improve their performance in
sports. In fact, they usually offer sincere words of appreciation for feedback to improve an
athletic behavior, such as a golf swing, a tennis stroke, a batter’s stance, a basketball
maneuver, or a football blocking technique. But, how often do you hear an individual offer
genuine thanks for being corrected on a safety-related behavior?
Safety coaching is often viewed as a personal confrontation. Actually, we are usually
more willing to accept and appreciate advice regarding work production and quality than
work safety.
Consequences of coaching. When we adjust our behavior following constructive
athletic coaching, it does not take very long to notice an improvement in our performance.
We see a direct connection between the improvement and the coach’s feedback. Sometimes,
we even see an increase in the individual or team scores as a result of individual or group
feedback. Thus, the process of athletic coaching is often supported by consequences occur-
ring naturally and soon after the behaviors targeted by the coach. The value of athletic
coaching becomes obvious.
Usually, the value of safety coaching is not obvious, because on a day-to-day basis
there is no clear connection between safety coaching and the ultimate purpose of
coaching—reduced injuries. When people follow the advice of a safety coach they usu-
ally do not perceive an immediate difference, either in their own safety or the company’s
injury rate. People do not expect injuries to happen to them and, because their everyday
experience supports this belief, they do not perceive a personal need for advice from a
safety coach.
Changing the way we keep score for safety can increase acceptance and appreciation
for safety coaching. While injury reduction is the ultimate purpose of coaching, the imme-
diate goal is behavioral improvement. Because at-risk behavior is involved in almost every
workplace injury, noting an increase in safe work behavior or a decrease in at-risk behav-
ior owing to safety coaching should result in coaching being appreciated as a proactive,
upstream approach to reducing injuries. This requires a behavior-based evaluation process,
and tools for accomplishing this are covered here.
As I have emphasized throughout this text and have written in other articles (Geller,
1994, 1995, 1999, 2000), a behavior-based approach to safety treats safety as an achievement-
oriented (rather than failure-oriented) process (not outcome) that is fact finding (not fault
finding) and proactive (rather than reactive). This chapter illustrates coaching techniques
that meet these criteria and demonstrates the critical value of safety coaching for achieving
a Total Safety Culture.

The safety coaching process

As shown in Figure 12.5 the five letters of the word COACH can be used to remember
the basic ingredients of the most effective coaching—whether coaching the members
of a winning athletic team or the individuals in a work group striving for safe beha-
viors. This is my favorite instructional acronym, because it not only contains the com-
ponents of an effective coaching process, but it also lists them in the sequence in which they
should occur.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 241

Chapter twelve: Intervening as a behavior-change agent 241

C are
• Show that you care
• Set caring examples

O bserve
• Define target behavior
• Record behavioral occurrences

A nalyze
• Identify existing contingencies
• Identify potential contingencies

C ommunicate
• Listen actively
• Speak persuasively

H elp
• Recognize continuous
improvement
• Teach and encourage the
process

Figure 12.5 The five letters of COACH represent the basic ingredients of effective safety
coaching.

“C” for care

Caring is the basic underlying motivation for coaching. Safety coaches truly care about the
health and safety of their coworkers and they act on such caring. In other words, they
“actively care” (Geller, 1991, 1994, 1996). When people realize by a safety coach’s words
and body language that he or she cares, they are more apt to listen to and accept the coach’s
advice. When people know you care, they care what you know.
Our emotional bank accounts. Covey (1989) explained the value of interdependence
among people—exemplified by appropriate safety coaching—with the metaphor of an
“emotional bank account.” People develop an emotional bank account with others through
personal interaction. Deposits are made when the holder of the account views a particular
interaction to be positive, as when he or she feels recognized, appreciated, or listened to.
Withdrawals from a person’s emotional bank account occur whenever that individual feels
criticized, humiliated, or less appreciated, usually as a result of personal interaction.
Sometimes, it is necessary to offer constructive criticism or even state extreme displeasure
with another person’s behavior. However, if such negative discourse occurs on an “overdrawn
or bankrupt account,” this corrective feedback will have limited impact. In fact, continued
withdrawals from an overdrawn account can lead to defensive or countercontrol reactions
(Skinner, 1974). The person will simply ignore the communication or actually do things to
discredit the source or undermine the process or system implicated in the communication.
Thus, safety coaches need to demonstrate a caring attitude through their personal
interactions with others. This maintains healthy emotional bank accounts—operating in
the “black.” The woman in Figure 12.6 is requesting a deposit along with the withdrawal.
Our emotional reaction to police officers depends on the proportion of deposits vs. with-
drawals we have experienced with them.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 242

242 Psychology of safety handbook

Figure 12.6 Our attitude toward police officers would be more positive if we received
deposits along with withdrawals.

A shared responsibility. People are often unwilling to coach or to be coached for

safety because they view safety from an individualistic perspective. To them, it is a matter
of individual or personal responsibility. This is illustrated by the verbal expression or
internal script, “If Molly and Mike want to put themselves at risk, that’s their problem, not
mine.” For some people a change in personal attitude or perspective is needed in order to
motivate coaching. People need to consider safety coaching a shared responsibility to pre-
vent injuries throughout the entire work culture. This requires a shift from an individual to
a collective perspective (Triandis, 1977, 1985), but this is not easy, as reflected in the insight-
ful poem “The Cookie Thief” by Valerie Cox, reproduced in Figure 12.7.
Many people accept a collective or team attitude when it comes to work productivity and
quality. Coaching for production or quality is part of the job, but coaching for personal safety
is often perceived as meddling. People need to understand that safety-related behaviors
require as much, if not more, interpersonal observation and feedback as any other job activity.
One way to convince people to accept and support safety coaching as a shared respon-
sibility is to point out their plant’s injury record for a certain period of time. While an injury
did not happen to them, it did happen to someone, and everyone certainly cares about that.
Given this underlying caring attitude, the challenge is to convince others that effec-
tive safety coaching by them will reduce injuries to their coworkers. This is enabled by a
behavior-based measurement system, as discussed next.

“O” for observe

Safety coaches observe the behavior of others objectively and systematically, with an eye
for supporting safe behavior and correcting at-risk behavior. Behavior that illustrates going
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 243

Chapter twelve: Intervening as a behavior-change agent 243

The Cookie Thief

by Valerie Cox
A woman was waiting at an airport one night,
With several long hours before her flight.
She hunted for a book in the airport shop,
Bought a bag of cookies and found a place to drop.
She was engrossed in her book, but happened to see,
That a man beside her, as bold as could be
Grabbed a cookie or two from the bag between,
Which she tried to ignore, to avoid a scene.
She read, munched cookies, and watched the clock,
As the gutsy "cookie thief" diminished her stock.
She was getting more irritated as the minutes ticked by,
Thinking, "If I wasn’t so nice, I’d blacken his eye!"
With each cookie she took, he took one, too.
When only one was left, she wondered what he’d do.
With a smile on his face and a nervous laugh,
He took the last cookie and broke it in half.
He offered her half, as he ate the other.
She snatched it from him and thought, "Oh, brother,
This guy has some nerve, and he’s also rude,
Why, he didn’t even show any gratitude!"
She had never known when she had been so galled
And sighed with relief when her flight was called.
She gathered her belongings and headed for the gate,
Refusing to look back at the "thieving ingrate."
She boarded the plane and sank in her seat,
Then sought her book, which was almost complete.
As she reached in her baggage, she gasped with surprise.
There was her bag of cookies in front of her eyes!
"If mine are here," she moaned with despair,
"Then the others were his and he tried to share!"
Too late to apologize, she realized with grief,
That she was the rude one, the ingrate, the thief!

Figure 12.7 Independence from one person can stifle interdependence from another.

beyond “the call of duty” for the safety of another person should be especially supported.
This is the sort of behavior that contributes significantly to safety improvement and can be
increased through rewarding feedback. As illustrated in Figure 12.8, a safety observer does
not hide or spy, and always asks permission first. Only with permission should an obser-
vation process proceed.
Observing behavior for supportive and constructive feedback is easy if the coach

1. Knows exactly what behaviors are desired and undesired (an obvious requirement
for athletic coaching).
2. Takes the time to observe occurrences of these behaviors in the work setting.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 244

244 Psychology of safety handbook

Figure 12.8 Safety coaches are up-front about their intentions and ask permission before
observing.

It is often advantageous—and usually essential—to develop a checklist of safe and at-risk

behaviors and to rank them in terms of risk. Ownership and commitment is increased
when workers develop their own behavioral checklists.
Developing a critical behavioral checklist. Observation checklists can be generic or
job-specific. A generic checklist is used to observe behaviors that may occur during several
jobs. A job-specific checklist is designed for one particular job. Deciding which items
to include on a critical behavior checklist (CBC) is a very important part of the coach-
ing process. A CBC enables coaches to look for critical behaviors. A critical behavior is a
behavior that

1. Has led to a large number of injuries or near hits in the past.

2. Could potentially contribute to a large number of injuries or near hits because many
people perform the behavior.
3. Has previously led to a serious injury or a fatality.
4. Could lead to a serious injury or fatality.

If only a few behaviors are observed in the beginning, which is often a good way to
start a large-scale coaching process, a CBC should be designed for only the most critical
behaviors.
Several sources can be consulted to obtain behaviors for a CBC, including injury
records, near-hit reports, job hazard analyses, standard operating procedures, rules and
procedural manuals, and the workers themselves. People already know a lot about their
own safety performance. They know which safety rules they sometimes ignore, and they
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 245

Chapter twelve: Intervening as a behavior-change agent 245

know when a near hit has occurred to themselves or to others because of at-risk behavior.
In addition, it is often useful to obtain advice from the plant doctor, nurse, safety director,
or anyone else who maintains injury statistics for the plant.
When starting out, do not develop an exhaustive checklist of critical behaviors. A list
can get quite long in a hurry. A long list for one-on-one observations can appear over-
whelming, and can inhibit the process. As with anything that is new and needs voluntary
support, it is useful to start small and build. With practice, people find a CBC easy to use,
and they accept additions to the list. They will also contribute in valuable ways to refine
the CBC, from clarifying behavioral definitions to recommending behavioral additions
and substitutions. The development and use of a CBC is really a continuous improve-
ment process. Further development and refinement benefits coaching observations, and
vice versa.
A work group on a mission to develop a CBC needs to meet periodically to select crit-
ical behaviors to observe. I have found the worksheet depicted in Figure 12.9 useful in
beginning the development of a CBC. Through interactive discussions, work groups define
safe and at-risk behaviors in their own work areas relevant to each category. The category
on body positioning and protecting, for example, includes specific ways workers should
protect themselves from environment or equipment hazards. This can range from using
certain personal protective equipment to positioning their body parts in certain ways to
avoid possible injury.
Some categories in Figure 12.9 may be irrelevant for certain work groups, like locking
or tagging out equipment or complying with certain permit policies. A work group might
add another general procedural category to cover particular work behaviors. Notice that

Operating Procedures Safe Observation At - Risk Observation

BODY POSITIONING/PROTECTING
Positioning/protecting body parts
(e.g., avoiding line of fire, using PPE,
equipment guards, barricades, etc.).
VISUAL FOCUSING
Eyes and attention devoted to ongoing
task(s).
COMMUNICATING
Verbal or nonverbal interaction that
affect safety.
PACING OF WORK
Rate of ongoing work (e.g., spacing
breaks appropriately, rushing).
MOVING OBJECTS
Body mechanics while lifting,
pushing/pulling.
COMPLYING WITH LOCKOUT/TAGOUT
Following procedures for
lockout/tagout
COMPLYING WITH PERMITS
Obtaining, then complying with
permit(s). (e.g., Confined space entry,
hot work, excavation, open line, hot
tap, etc.).

Figure 12.9 Use this worksheet to develop a generic critical behavior checklist (CBC).
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 246

246 Psychology of safety handbook

defining safe and at-risk behaviors results in safety training in the best sense of the word.
Participants learn exactly what safe behaviors are needed for a particular work process.
A list of work behaviors covering all the generic categories in Figure 12.9 can be exten-
sive and overwhelming. This gives numerous opportunities for coaching feedback, but
remember, it takes time and practice to observe behaviors reliably—and to get used to
being observed while working. I have found it useful to start the observation procedure
with a brief CBC of four or five behaviors, and then build on the list with practice, group
discussion, and more practice.
Each of the generic categories in Figure 12.9 could be used as a separate checklist at the
initiation of a coaching process. The first category, for example, could lead to the develop-
ment of a CBC for observation of personal protective equipment. Specific PPE behaviors
for the work area could be listed in a left-hand column, with space on the right to check safe
and at-risk observations. This kind of CBC could be used to record the observations of sev-
eral individuals, by simply adding checks in the safe or at-risk columns for each observa-
tion of an individual’s use or nonuse of a particular PPE item.
Sample critical behavior checklist. Figure 12.10 depicts a comprehensive CBC for
recording the results of a coaching observation. This kind of CBC recording sheet should
be used after the participants (optimally, everyone in a particular work area) have derived
precise behavioral definitions for each category and have practiced rather extensively with
shorter CBCs. Such practice enables careful refinement of behavioral definitions and builds
confidence and trust in the process.
The CBC in Figure 12.10 allows for recording two or more one-on-one coaching obser-
vations. A “1” would be placed in either the “safe” or “at-risk” column for each behavioral
observation of Person 1, and a “2” would be used to indicate specific safe and at-risk behav-
iors for Person 2. Note that only the name of the observer is included on the data sheet.
When people realize that safety coaching is only to increase safe behavior and decrease at-
risk behavior, not to identify unsafe workers, voluntary participation will increase, along
with trust.
Scheduling observation sessions. There is no best way to arrange for coaching
observations. The process needs to fit the setting and work process. This can only happen
if workers themselves decide on the frequency and duration of the observations and derive
a method for scheduling the coaching sessions. I have seen the protocol for effective coach-
ing observations vary widely across plants, and across departments within the same plant.
The success of those process has not varied predictably as a function of protocol.
The 350 employees at one Exxon Chemical plant, for example, designed a process call-
ing for people to schedule their own coaching sessions with any two other employees. On
days and at times selected by the person to be observed, two observers show up at the indi-
vidual’s worksite and use a CBC to conduct a systematic, 30-minute observation session.
This plant started with only one scheduled observation per month, and observers were
selected from a list of volunteers who had received special coaching training. One year
later, employees scheduled two observations per month, and any plant employee could be
called on to coach. With slight periodic revisions, this interpersonal coaching process has
been in place for eight years (at the time of this writing) and it has enabled this plant to
reach and sustain a record-low injury rate.
The Exxon procedure is markedly different from the “planned 60-second actively car-
ing review” implemented at a Hoechst Celanese plant. For this one-on-one coaching
process, all employees attempt to complete a one-minute observation of another employee’s
work practices in five general categories: body position, personal apparel, housekeeping,
tools/equipment, and operating procedures. The initial plant goal was for each of the 800
employees to complete one 60-second behavioral observation every day. Results were
entered into a computer file for a behavioral safety analysis of the work culture.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 247

Chapter twelve: Intervening as a behavior-change agent 247

Observer:
Person 1 Person 2
Date: Date:
Time: Time:
Department: Department:
Building: Building:
Floor: Area: Floor: Area:

Operating Procedures Safe At-Risk

Observation Observation
BODY POSITIONING / PROTECTING

VISUAL FOCUSING

COMMUNICATING

PACING OF WORK

MOVING OBJECTS

COMPLYING WITH
LOCKOUT / TAGOUT

COMPLYING WITH PERMITS

1 = observations for first person; 2 = observations for second

person

Total Safe Observations

% Safe Behaviors: = %
Total Safe and
At-Risk Observations

Figure 12.10 A comprehensive critical behavior checklist (CBC) enhances the learning
from one-on-one observations.

The CBC used for the one-minute coaching observations is shown in Figure 12.11. The
front of each card includes the five behavioral categories, a column to check “safe” or “at
risk” per category, and columns (“Feedback Targets”) to write comments about the obser-
vations. These comments facilitate a feedback session following the observation session, if
it is convenient. The back of this CBC includes examples (“memory joggers”) related to
each behavioral category on the front of the card. These examples summarize the category
definitions developed by the CBC steering committee and determine whether “safe” or
“at-risk” should be checked on the front of the card.
Critical features of the observation process. Duration, frequency, and scheduling
procedures of CBC observations vary widely. Still, there are a few common features. First
and foremost, the observer must ask permission before beginning an observation process.
The name of the person observed must never be recorded. To build trust and increase par-
ticipation, a “no” to a request to observe must be honored.
Asking permission to observe serves notice to work safely and, thus, biases the obser-
vation data, right? In other words, when workers give permission to be coached, their
attention to safety will likely increase and they will try to follow all safety procedures. It is
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 248

248 Psychology of safety handbook

Observer: Location: Date:

Feedback Targets:
Audit
Safe At Risk Safe At Risk
Category

Position

Safe Apparel

Housekeeeping

Tools/Equip.

Procedures

Total

Front of One-Minute Audit Card

Observation Observation
Targets Safe At Risk Targets Safe At -Risk

Position Tools/Equip.
* Line of Fire * Condition
* Falling * Use
* Pinch Points * Guards
* Lifting
Safe Apparel Procedures
* Hair * SOP's
* Clothes * JSA's
* Jewelry * Permits
* PPE * Lockout
Housekeeping * Barricade
* Equipment
* Floor Release
* Equipment
* Storage of
Materials

Back of One-Minute Audit Card

Figure 12.11 Employees used this critical behavior checklist for one-minute observations.
possible, though, for people to overlook safety precautions, even when trying their best.
They could be unconsciously at risk. When people give permission to be coached, their
willingness to accept and appreciate feedback is maximized, even when it is corrective.
What if people sneak around and conduct behavioral observations with no warning?
This is in fact an unbiased plant-wide audit of work practices. It might even be accepted if
those observed were not identified. However, if one-on-one coaching is added to this pro-
cedure, an atmosphere of mistrust can develop.
Safety coaching should not be a way to enforce rules or play “gotcha.” It needs to be
seen as a process to help people develop safe work habits through supportive and con-
structive feedback. Giving corrective feedback after “catching” an individual off-guard
performing an at-risk behavior will likely lead to defensiveness and lack of appreciation,
even for a well-intentioned effort. It can also reduce interpersonal trust and alienate a per-
son toward the entire safety coaching process.
Feedback is essential. Each observation process with a CBC provides for tallying
and graphing results as “percent safe behavior” on a group feedback chart. The CBC shown
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 249

Chapter twelve: Intervening as a behavior-change agent 249

in Figure 12.10 includes a formula at the bottom for calculating percent safe behavior per
coaching session. In this case, all checks for safe observations are added and divided by the
total number of checks (safe plus at-risk behaviors). The result is multiplied by 100 to yield
percent safe behavior.
Applying the formula in Figure 12.10 to checks written on the front of the CBC shown
in Figure 12.11 results in a conservative estimate of percent safe behavior. That is because a
safe check mark on the CBC in this application meant that each separate behavior of a cer-
tain category was marked safe on the back. Thus, this calculation required all behaviors
relating to a particular observation category to be safe for a “safe” designation. The CBC
shown in Figure 12.10 does not use this all-or-nothing calculation and generally results in
higher percentages of safe behaviors.
There is no best way to do these calculations. What is important is for participants to
understand the meaning of the feedback percentages. As shown in the lower half of Figure
12.12, these percentages can be readily displayed on a group feedback chart or graph.
While feedback percentages are valuable, it is vital to realize that the process is more impor-
tant than the numbers.
The true value of the coaching process is not in the behavioral data, which are no doubt
biased by uncontrollable factors, but in the behavior-based interaction between employees.
I have actually seen observers get so caught up in recording the numbers, such as fre-
quency of safe and at-risk behaviors, that they let coworkers continue to perform an at-risk
behavior while they observe and check columns on a CBC.
An individual’s safety must come before the numbers in any observation process.
When observers see an at-risk behavior that immediately threatens a person’s health or
safety, they should intervene at once. They can usually pick up the observation process

Figure 12.12 Feedback from a critical behavior checklist can be given one-on-one and in
groups.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 250

250 Psychology of safety handbook

afterwards. On the other hand, if the CBC was partially completed before they stepped in,
it might be most convenient to communicate other observations, especially if there are
some safe behaviors to report. This “deposit” will help compensate for the “withdrawal”
that was probably implicated by the need to stop a risky behavior.
Observation procedures always include a provision for one-on-one feedback, although
some processes have made immediate feedback optional. On the one-minute CBC (Figure
12.11), there is a place to check whether feedback was given or not. The employees who
designed this protocol decided to make feedback optional in order to increase participation
in the observation process, and at least obtain group feedback from all departments. The
number of CBC cards collected per department was exhibited in a large display case at
the plant entrance, along with daily percentages of observations resulting in one-on-one
feedback.

“A” for analyze

When interpreting observations, safety coaches draw on their understanding of the ABC
contingency (for activator–behavior–consequence) introduced in Chapter 8 and the behav-
ior analysis principles discussed in Chapter 9. They realize observable reasons usually exist
for why safe or risky behaviors occur They know certain dangerous behaviors are triggered
by activators such as work demands, risky example setting by peers, and inconsistent mes-
sages from management. They also appreciate the fact that at-risk behaviors are often moti-
vated by one or more natural consequences, including comfort, convenience, work breaks,
and approval from peers or work supervisors.
This understanding is critical if safety coaching is to be a “fact-finding” rather than
“fault-finding” process. It also leads to an objective and constructive analysis of the situa-
tions observed. This is how people discover the reasons behind at-risk behaviors and
design interventions to decrease them.
An ABC analysis can be done before giving feedback to the person observed or during
the one-on-one feedback process. Discussing the activators and consequences that possibly
influenced certain work practices can lead to environment or system improvements for
decreasing at-risk behavior.

• Was the behavior observed activated by a work demand or a desire to go on break

or leave work early?
• Does the design of the equipment or environment, or the ergonomic design of the
task, influence at-risk behavior?
• Is certain personal protective equipment uncomfortable or difficult to use?
• Are fellow workers or supervisors activating dangerous behavior by requesting or
demanding an excessive work pace?
• Are certain people motivating at-risk behavior from others by giving rewarding
consequences, like words of appreciation, for work done quickly at the expense
of safety?

Answers to these and other questions are explored with the observee in the next phase
of safety coaching—the heart of the process.

“C” for communicate

A good coach is a good communicator. This means being an active listener and persuasive
speaker. Because none of us are born with these skills, communication training sessions
that incorporate role-playing exercises can be invaluable in developing the confidence
and competence needed to send and receive behavioral feedback. Such training should
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 251

Chapter twelve: Intervening as a behavior-change agent 251

emphasize the need to separate behavior (actions) from person factors (attitudes and feel-
ings). This enables corrective feedback without stepping on feelings.
People need to understand that anyone can be at risk without even realizing it, as in
“unconscious incompetence,” and performance can only improve with behavior-specific
feedback. Once this fact is established, corrective feedback that is appropriately given will
be appreciated, regardless of who is giving the feedback. Work status is not a factor.
The right delivery I remember key aspects of effective verbal presentation with the
“SOFTEN” acronym listed in Figure 12.13. First, it is important for the observer to initiate
communication with a friendly smile and an open (flexible) perspective. “Territory”
reflects the need to respect the fact that you are encroaching upon another person’s work
area. You should ask the person where it would be appropriate and safe to talk. It is also
important to maintain a proper physical distance during this interaction.
Standing too close or too far from another person can cause interference and discom-
fort. Hall (1959, 1966) coined the term “proxemics” to refer to how we manage space, and
he researched the distances people keep from each other in various situations. There are
prominent cultural differences in interpersonal distance norms. In the United States, com-
municating closer than 18 inches with another person—measured nose to nose—is con-
sidered an intimate distance (Hall, 1966), with 0 to 6 inches reserved for comforting,
protesting, lovemaking, wrestling, and other full-contact behaviors. The far phase of the
intimate zone (6 to 18 inches) is used by individuals who are on very close terms.
Safety coaches in the United States should most likely communicate at a “personal dis-
tance” (18 inches to 4 feet). According to Hall (1966), the near phase of the personal distance
(18 to 30 inches) is reserved for those who are familiar with one another and on good terms.
The far phase of the personal zone (2.5 to 4 feet) is typically used for social interactions
between friends and acquaintances. This is likely to be the most common interaction zone
for a workplace coaching session. Some coaching communication might occur at the near
phase (4 to 7 feet) of Hall’s social distance, which is typical for unacquainted individuals
interacting informally. These distance recommendations are not hard and fast rules of con-
duct but rather personal territory norms we need to consider.

Figure 12.13 Principles of effective sending can be remembered with SOFTEN.

820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 252

252 Psychology of safety handbook

Figure 12.14 Body language communicates more than words.

The “E” of our acronym represents three important directives to remember when
coaching—energy, enthusiasm, and eye contact. Your energy and enthusiasm can activate
concern and caring on the part of the person you are communicating with. We all know that
excited, committed coaches can make “true believers” out of the troops—and that indif-
ferent or distracted coaches can have the opposite effect. Actually, as depicted in Figure
12.14, our body language speaks louder than words.
Proper eye contact is body language critically important to maintain throughout a
coaching session. Everyone has experienced the uncomfortable feeling of talking or listen-
ing to someone who does not look at them. In contrast, there is the piercing stare we some-
times perceive with too much eye-to-eye contact. Thus, we learn through natural feedback,
often unconsciously, the definition of proper eye contact.
Finally, we need to remember that the dearest word to anyone’s ears is his or her own
name (Carnegie, 1936). Refer to the other person by name, but make it clear the behavioral
observations you have recorded will remain anonymous.
The power of feedback. You know the old saying, “Practice makes perfect”? I bet
you have heard it a hundred times. Well, I hate to tell you, it is wrong. Practice does not
make perfect. Practice only makes permanence. Feedback makes perfect. Without the right
feedback, we cannot improve our performance. We need to know how we are doing so we
can make adjustments if they are called for. When we are doing great, we need to know
that, too, so we will be reinforced and keep on doing things right.
Some behaviors are followed by consequences that provide natural feedback. Take
sports, for example. When you hit a tennis ball or throw a football, the path of the ball is
your feedback. You can see how close it came to where you aimed it, and you can adjust
your technique the next time, based on the information you received. The same thing hap-
pens when we hammer a nail, type a word, or organize our work area—we observe natu-
ral consequences that give us feedback about our performance.
We hardly ever get natural feedback about our safety-related behavior. When we do
get natural feedback, what form does it usually take? If we are lucky, it is only a near hit. If
we are unlucky, it is an injury. So where do we get the feedback to improve our safety-
related behaviors? It must come from people. Giving feedback to others, and receiving
feedback from others about safety, is vital to improving safety-related behavior. In fact, it is
the most cost-effective intervention technique a safety coach can use.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 253

Chapter twelve: Intervening as a behavior-change agent 253

Research evidence. The power of feedback is evident in the famous Hawthorne

studies. Ask any safety manager, industrial consultant, or applied psychologist whether
they have heard of the “Hawthorne Effect,” and they probably will say, “yes.” They might
not be able to describe any details of the studies that occurred between 1927 and 1932 at the
Western Electric plant in the Hawthorne community near Chicago that led to the classic
Hawthorne Effect. Most, however, will be able to paraphrase the infamous finding from
these studies that the hourly output rates of the employees studied increased whenever an
obvious environmental change occurred in the work setting.
The explanation of the Hawthorne results is also well-known and recited as a potential
confounding factor in numerous field studies of human behavior. Specifically, it is com-
monly believed the Hawthorne studies showed that people will change their behavior in
desired directions when they know their behavior is being observed. The primary
Hawthorne sources (Mayo, 1933; Roethisberger and Dickson, 1939; Whitehead, 1938) leave
us with this impression and, in fact, this interpretation seems intuitive. The fact is, however,
this interpretation of the Hawthorne studies is not accurate—it is nothing but a widely dis-
seminated myth.
Parsons conducted a careful re-examination of the Hawthorne data and interviewed
eyewitness observers, including one of the five female relay assemblers who were the pri-
mary targets of the studies. Parsons’ findings were published in a seminal Science article
entitled “What happened at Hawthorne?” (Parsons, 1974). What happened was the five
women observed systematically in the Relay Assembly Test Room received regular feed-
back about the number of relays each had assembled. “They were told daily about their
output, and they found out during the working day how they were doing simply by get-
ting up and walking a few steps to where a record of each output was being accumulated”
(Parsons, 1980, page 58).
From his scientific detective work, Parsons concluded that performance feedback was
the principal extraneous, confounding variable that accounted for the Hawthorne Effect.
The performance feedback was important to the workers (so they were apt to respond to
it) because their salaries were influenced by an individual piecework schedule—the more
relays each employee assembled, the more money each earned.
There is one other point I would like to make about the Hawthorne studies. The five
test subjects preferred working in the test room rather than in the regular department, but
when asked why, they did not mention anything about receiving feedback. Instead, their
reasons included “smaller groups,” “no bosses,” “less supervision,” “freedom,” and “the
way we are treated” (Roethlisberger and Dickson, 1939, pages 66–67). This is worth noting,
because it suggests you should not rely only on verbal report to discover factors influenc-
ing work performance. Sometimes, people are not aware of the basic contingencies con-
trolling their behavior. Through systematic and objective observation these factors can be
uncovered—and instructive feedback given.
Feedback for safety. For anyone who has studied the behavior-based approach to
performance management, the only surprise in Parsons’ research is that the critical role of
performance feedback was not documented in the original reports of the Hawthorne stud-
ies. Numerous research studies have shown that posting results of behavioral observations
related to safety, production, or quality has a positive impact on targeted work behaviors.
If desired work behaviors are targeted, they increase in frequency; when undesired behav-
iors are targeted for observation and feedback, they decrease in probability (e.g., Austin
et al., 1996; Geller et al., 1980; Kim and Hammer, 1982; Komaki et al., 1980; Sulzer-Azaroff,
1982; Williams and Geller, 2000).
Individual feedback. Whether the aim is to support or correct, feedback should be
specific and timely. It should specify a particular behavior and occur soon after the target
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 254

254 Psychology of safety handbook

behavior is performed. Also, it should be private, given one-on-one to avoid any interfer-
ence or embarrassment from others. Corrective feedback is most effective if the alternative
safe behavior is specified and potential solutions for eliminating the at-risk behavior are
discussed.
Anyone giving feedback must actively listen to reactions. This is how a safety coach
shows sincere concern for the feelings and self-esteem of the person on the receiving end
of feedback. The best listeners give empathic attention with facial cues and posture, para-
phrase to check understanding, prompt for more details, accept stated feelings without
interpretation, and avoid arrogance such as, “When I worked in your department, I always
worked safely.”
Figure 12.15 reviews the critical characteristics of effective rewarding and correcting
feedback. This figure can be used as a guide for group practice sessions. Because it is not
easy to give safety feedback properly and because many people feel awkward or uncom-
fortable doing it, practice is important.
In training sessions, I ask groups of three to seven individuals to develop a brief skit
that illustrates rewarding or correcting feedback. Each skit involves at least three partici-
pants: a safety coach, a worker receiving feedback, and a narrator who sets the scene. The
exercise can be more fun if groups first demonstrate the wrong way to give feedback and
then show the correct procedure.
Afterward, the audience should provide feedback. The presenters should hear about
particular strengths of their demonstration and places where it could be improved. A good
facilitator can draw out important lessons from this communication exercise, while

Figure 12.15 Maximize the beneficial impact of rewarding and correcting feedback with
these key points.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 255

Chapter twelve: Intervening as a behavior-change agent 255

keeping the atmosphere congenial and enjoyable. It has to be informative and rewarding
for all involved, because everyone will need to participate in several demonstrations. It
takes practice and peer support for participants to feel comfortable and effective at giving
rewarding and correcting feedback. Fortunately, we learn much from watching others per-
form (Bandura, 1969), and so the audience learns by vicariously observing demonstrations
by peers.
Group feedback. As we have discussed, observations recorded on a CBC can be
summarized as a calculation of “percent safe behavior.” These percentages can be calcu-
lated per day or per week or per month and displayed on a feedback graph (see Figure
12.12). When this graph is posted in a prominent place, perhaps next to the plant’s statisti-
cal process control charts, employees can monitor their progress and be naturally rewarded
for their efforts. This gives safety the same status as quality and promotes group achieve-
ment. People can monitor the progress of their work team as its percentage of safe behav-
iors increases.
Percentages of safe observations can be calculated separately for various workplace
activities (see Figures 12.10 and 12.11). Graphs can readily show the percentage of employ-
ees wearing safe apparel, using appropriate personal protective equipment, avoiding the
line of fire, lifting or moving objects safely, keeping work areas neat and free of trip haz-
ards, using tools and equipment safely, and complying with lockout/tagout procedures or
work-permit requirements.
Monitoring behavioral categories separately lets you see what needs special attention.
A variety of interventions may be called for, from implementing special training sessions
to ergonomically rearranging a particular work area. Observation and feedback provide
invaluable diagnostic information. The graphs hold people accountable for process num-
bers they can control on a daily basis—in contrast to outcome numbers like total recordable
injury rate or workers’ compensation costs. Improving these upstream numbers (percent
safe behavior) will eventually reduce the outcome number (work injuries).
The power of social comparison feedback. In a recent industrial safety study,
Williams and Geller (2000) systematically compared the impact of weekly posting specific
vs. general behavioral feedback. Specifically, the percent safe scores from daily CBCs were
posted as separate percentages per each of the nine CBC behaviors or as an overall global
percent safe score which was calculated across all nine behaviors on the CBC. The percent-
age of safe behavior increased with both group feedback methods. However, as expected,
specific feedback resulted in significantly greater improvement.
When we added social comparison percentages to group feedback charts, we were
pleasantly surprised. A group receiving two global percent safe scores, one for their own
group and another for employees performing the same tasks on another shift, showed the
same amount of behavioral improvement as the group who received specific percentages
for each CBC behavior and a group who received specific percentages for both their own
and a comparable group.
Because global feedback was just as effective as specific feedback when social compar-
ison feedback was included, we presumed most of the 97 employees of the soft-drink bot-
tling facility did not need an instructional intervention. They knew how to perform their
jobs safely but needed some extrinsic motivation to follow the nine safety policies implied
by the nine target behaviors. This was provided by a global percent safe score from a sim-
ilar work group.
The finding that a global percent safe score from a comparable work group led to as
much improvement as providing separate percentages per each CBC behavior has practi-
cal significance. That is, it took 5 to 6 hours to prepare the weekly graphs of specific feed-
back, whereas it only took about 30 minutes to calculate the percentages for the global
feedback displays.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 256

256 Psychology of safety handbook

Note that global percentages from CBC records can only be effective when workers
know the safe operating procedures and need a motivational intervention. In situations
where employees are inexperienced and unaware of the safest work practices, specific
feedback is needed. It is best, however, to provide these employees with one-on-one coach-
ing and specific behavioral feedback. Thus, a combination of individual and group feed-
back is usually most cost effective, whereby specific behavioral feedback is provided
during one-on-one coaching and global percent safe percentages are posted on a group
feedback chart. However, our research suggests you will increase the impact of the global
feedback if you add a percent safe score from another similar workgroup.

“H” for help

The word “help” summarizes what safety coaching is all about. The purpose is to help an
individual prevent injury by supporting safe work practices and correcting at-risk prac-
tices. It is critical, of course, that a coach’s help is accepted. The four letters of HELP outline
strategies to increase the probability that a coach’s advice, directions, or feedback will be
appreciated.

“H” for humor

Safety is certainly a serious matter, but sometimes a little humor can add spice to our com-
munications, increasing interest and acceptance. It can take the sting out what some find to
be an awkward situation. In fact, researchers have shown that laughter can reduce distress
and even benefit our immune system (Goodman, 1995; National Safety News, 1985).

“E” for esteem

People who feel inadequate, unappreciated, or unimportant are not as likely to go beyond
“the call of duty” to benefit the safety of themselves or others as people who feel capable
and valuable (see Chapter 15 for support of this argument). The most effective coaches
choose their words carefully, emphasizing the positive over the negative, to build or avoid
lessening another person’s self-esteem. Although Figure 12.16 is humorous, it is unfortu-
nately an accurate portrayal of the atmosphere in many organizational cultures, including
the university environment in which I have worked for more than 30 years.

“L” for listen

One of the most powerful and convenient ways to build self-esteem is to listen attentively
to another person. This sends the signal that the listener cares about the person and his or
her situation. And it builds self-esteem—”I must be valuable to the organization because
my opinion is appreciated.” After a safety coach listens actively, his or her message is more
likely to be heard and accepted. As Covey (1989) put it, “seek first to understand, then to
be understood” (page 235). The next chapter covers this principle more completely.

“P” for praise

Praising others for their specific accomplishments is another powerful way to build self-
esteem. If the praise targets a particular behavior, the probability of the behavior reoccur-
ring increases. This reflects the basic principle of positive reinforcement and motivates
people to continue their safe work practices and look out for the safety of coworkers.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 257

Chapter twelve: Intervening as a behavior-change agent 257

Figure 12.16 Standard feedback more often depreciates than appreciates a person’s self-
esteem.
Behavior-focused praising is a powerful rewarding consequence which not only increases
the behavior it follows, but also increases a person’s self-esteem. This, in turn, increases the
individual’s willingness to actively care for the safety of others, as I discuss more com-
pletely in Section 5 of this Handbook.
Human nature directs more attention to mistakes than successes. Errors stick out and
disrupt the flow, so they attract reaction and attempts to correct them. As illustrated in
Figure 12.17, however, when things are going smoothly—and safely—there is usually no
stimulus to signal success. A person’s good performance is typically taken for granted. We
need to resist the tendency to go with the flow, and sometimes express sincere appreciation
for ongoing safe behavior. I give specifics on how to do this in the next chapter.

What can a safety coach achieve?

The safety coaching process described here is founded on the basic premise of behavior-
based safety. Injuries are a direct function of at-risk behaviors, and if these behaviors can
be decreased and safe behaviors increased, injuries will be prevented. Indeed, the well-
known Heinrich Law of Safety implicates unsafe acts as the root cause of most near hits and
injuries (Heinrich et al., 1980).
Over the past 20 years, a variety of behavior-based research studies have verified this
aspect of Heinrich’s Law by systematically evaluating the impact of interventions designed
to increase workers’ safe behaviors and decrease their at-risk behaviors. Feedback from
behavioral observations was a common ingredient in most of the successful interventions,
whether it was delivered through tables, charts, interpersonal communication, congratula-
tory notes, or a reward following a particular behavior (see, for example, comprehensive
reviews by Balcazar et al., 1986; and Petersen, 1989; or individual research articles by
Austin et al., 1996; Cooper et al., 1994; Chhokar and Wallin, 1984; Geller et al., 1980; Komaki
et al., 1980; Sulzer-Azaroff and De Santamaria, 1980).
The behavior-based feedback and coaching process described here is analogous to the
behavior-based safety process detailed by Krause et al. (1996) and McSween (1995) and
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 258

258 Psychology of safety handbook

Figure 12.17 People need frequent rewarding feedback.

taught in training videotapes and workbooks developed by Tel-A-Train, Inc. (1995) and
J. J. Keller and Associates (Geller, 1998c). In addition to the research referred to previously,
there are hundreds of real-world case studies which provide evidence of the the injury pre-
vention impact of behavior-based coaching. I have personally been teaching variations of
this approach to industry for more than 25 years and have never seen the process fail to
work when implemented properly. I have also witnessed numerous cases of companies
receiving less than desired benefits owing to incomplete or inadequate implementation.
There is no quick-fix substitute for this process, no effective step-by-step cookbook.
Achievements from safety coaching are a direct function of the effort put into it. The guide-
lines presented in this chapter need to be customized. Who knows best what step-by-step
coaching procedures will succeed in a given work area? The people employed there know
best and they need to be empowered to develop their own safety coaching process.
An ExxonMobil Chemical facility in Texas has demonstrated exemplary success with a
coaching process based on the principles and procedures described in this chapter. By the
end of two years, they had almost 100 percent participation and have reaped extraordinary
benefits. From an outcome perspective, they started with a baseline of 13 OSHA recordable
injuries in 1992 (TRIR  4.11), and progressed to 5 OSHA recordables in 1993 (TRIR  1.70).
They sustained only one OSHA recordable in 1994 (TRIR  0.30) and reached their target
of zero OSHA recordables in 1997 and 1999. At the time of this writing (mid-2000), they are
still injury free for the year. Figure 12.18 depicts the total recordable injury rate (TRIR) for
this plant from 1991 to mid-2000. They had received behavior-based coaching training in
the latter half of 1992, implemented their observations and feedback process plantwide in
1993, and by 1994 everyone was on board as a behavior-focused coach. They have contin-
ued this process ever since and have had numerous occasions to celebrate their phenome-
nal safety success.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 259

Chapter twelve: Intervening as a behavior-change agent 259

Figure 12.18 The ten-year TRIR record of one ExxonMobil facility indicates powerful
influence of interpersonal behavior-based coaching.

Obviously, many factors contributed to this extraordinary performance, but there is lit-
tle doubt their safety coaching process played a critical role. At the end of 1994, for exam-
ple, 98 percent of the workforce had participated as observers to complete a total of 3350
documented safety coaching sessions. They identified 51,048 behaviors, of which 46,659
were safe and 4,389 were at risk.
In 1992, a safety culture survey was administered before the safety coaching process
was initiated. It was repeated again in 1994. Results revealed statistically significant
improvement in perceptions and attitudes toward industrial safety, intentions to actively
care for other workers’ safety, and feelings of belonging and group cohesion throughout
the work culture. It is important to realize that these dramatic improvements in safety per-
ceptions, attitudes, and intentions occurred while Exxon and the petrochemical industry
experienced significant downsizing.

Self appraisal of coaching skills

The self-survey in Figure 12.19 reflects attributes of ideal safety coaches. Several of the con-
cepts—particularly self-esteem, self-confidence, optimism, and teamwork—are discussed
in more detail in Section 5 of this Handbook. By rating how often you accomplish the ideal
coaching characteristic implied by each item in this questionnaire you will review key
points of this chapter. If you are honest and frank, you will gain important insight from this
exercise. Define your strengths and weaknesses, then apply what you have learned here to
improve your competence as an actively caring safety coach.

In conclusion
Safety coaching is a key intervention process for developing and maintaining a Total Safety
Culture. In fact, the more employees effectively apply the principles of safety coaching dis-
cussed here, the closer an organization will come toward achieving a Total Safety Culture.
The same is true for preventing injury in the community and among our immediate
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 260

260 Psychology of safety handbook

Answer each of these questions honestly to determine Highly Agree

your current level of readiness and competence to be an Agree
effective safety coach. Read each statement, then circle Not Sure
the number that best describes your current feelings.
Disagree
Highly Disagree
1. I take a balanced approach to long-term goals and short-term results. 1 2 3 4 5
2. I give credit for a job well done. 1 2 3 4 5
3. I refer to specific observable actions when discussing a worker's
performance. 1 2 3 4 5
4. I explain the rationale for policies and procedures. 1 2 3 4 5
5. I provide both direction and encouragement when requesting behavior
change. 1 2 3 4 5
6. I avoid talking down to other workers I supervise. 1 2 3 4 5
7. I practise active listening. 1 2 3 4 5
8. I display a sense of humor. 1 2 3 4 5
9. I only make promises I can keep. 1 2 3 4 5
10. I work with others to set performance standards. 1 2 3 4 5
11. I treat others fairly. 1 2 3 4 5
12. I express interest in the career growth of workers I supervise. 1 2 3 4 5
13. I ask others for ideas and opinions. 1 2 3 4 5
14. I promote feelings of ownership among team members. 1 2 3 4 5
15. I demonstrate personal integrity when dealing with others. 1 2 3 4 5
16. I practise principles of appropriate rewarding feedback. 1 2 3 4 5
17. I practise principles of appropriate correcting feedback. 1 2 3 4 5
18. I take ownership and responsibility for personal decisions. 1 2 3 4 5
19. I treat others with dignity and respect. 1 2 3 4 5
20. I encourage and accept performance feedback from others. 1 2 3 4 5
21. I find ways to celebrate others accomplishments. 1 2 3 4 5
22. I accept others' failures as opportunities to learn. 1 2 3 4 5
23. When appropriate I challenge higher level management. 1 2 3 4 5
24. I create an atmosphere of interpersonal trust. 1 2 3 4 5
25. I evaluate others' performance as objectively as possible. 1 2 3 4 5
26. I help team members solve problems constructively. 1 2 3 4 5
27. I act to support the value that "people are our most important asset." 1 2 3 4 5
28. I encourage others to participate actively in conversations, discussions,
and meetings. 1 2 3 4 5
29. I show sensitivity to the feelings of others. 1 2 3 4 5
30. I promote synergy among team members. 1 2 3 4 5
31. I promote a win/win approach to problem solving. 1 2 3 4 5
32. I promote others' self-esteem. 1 2 3 4 5
33. I promote others' sense of personal control. 1 2 3 4 5
34. I promote others' perceptions of self-confidence. 1 2 3 4 5
35. I am optimistic. 1 2 3 4 5
36. I encourage teamwork. 1 2 3 4 5

Figure 12.19 Use this survey to evaluate your coaching skills.

family members at home. Indeed, we need to practice the principles of safety coaching in
every situation where an injury could occur following at-risk behavior.
Systematic safety coaching throughout a work culture is certainly feasible in most set-
tings. Large-scale success requires time and resources to develop materials, train necessary
personnel, establish support mechanisms, monitor progress, and continually improve the
process and support mechanisms whenever possible. For example, the following questions
need to be answered at the start of developing an initial action plan.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 261

Chapter twelve: Intervening as a behavior-change agent 261

• Who will develop the critical behavior checklist (CBC)?

• How extensive will the first CBC be?
• What information will be used to define critical behaviors?
• How will safety coaches be trained and receive practice and feedback?
• How many coaches will be trained initially and how can additional people volun-
teer to participate as a safety coach?
• How will the coaching sessions be scheduled, how often will people be coached, and
how long will the coaching sessions last?
• Where will the group feedback graphs be posted, and who will be responsible for
preparing the displays of safe behavior percentages?
• Who will be on the steering committee to oversee the safety coaching process,
answer these and other questions about process implementation, maintain records,
monitor progress, and refine procedural components whenever necessary?

This does not cover all the issues, yet the list might appear overwhelming at first. There
is no formula for a quick-fix solution. Organizational cultures vary widely according to
personnel, history, policy, the work process, environmental factors, and current contingen-
cies. So implementation procedures need to be customized. There must be significant input
from the people protected by the coaching process and from whom long-term participation
is needed. It is likely to take significant time, effort, and resources to achieve a plant-wide
safety coaching process. With this end in mind, I recommend starting small to build confi-
dence and optimism on small-win accomplishments. Then with patience and diligence, set
long-term goals for continuous improvement. Remember to celebrate achievements that
reflect successive approximations of your vision—an organization of people who consis-
tently coach each other effectively to increase safe work practices and decrease at-risk
behaviors.
Safety coaching is a critically important intervention approach, but keep in mind the
many other ways you can contribute to the health and safety of a work culture. In other
words, safety coaching is one type of intervention for the “I” stage of the DO IT process.
Any variety of activator and consequence strategies explained in Chapters 10 and 11,
respectively, can be used as a behavior-based intervention. These steps require people to go
beyond their normal routine to help another person. The next chapter shows how we can
support and, thereby, improve safety with everyday interpersonal conversations and infor-
mal coaching.

References
Austin, J., Kessler, M. L., Riccobono, J. E., and Bailey, J. S., Using feedback and reinforcement to
improve the performance and safety of a roofing crew. J. Organ. Behav. Manage., 16, 49, 1996.
Balcazar, F., Hopkins, B. L., and Suarez, Y., A critical, objective review of performance feedback,
J. Organ. Behav. Manage., 7(3/4), 65, 1986.
Bandura, A., Principles of Behavior Modification, Holt, Rinehart & Winston, New York, 1969.
Carnegie, D., How to Win Friends and Influence People, Simon & Schuster, New York, 1936.
Chhokar, J. S. and Wallin, J. A., A field study of the effects of feedback frequency on performance,
J. Appl. Psychol., 69, 524, 1984.
Cone, J. D. and Hayes, S. C., Environmental Problems: Behavioral Solutions, Brooks/Cole, Monterey,
CA, 1980.
Cooper, M. D., Phillips, R. A., Sutherland, V. J., and Makin, P. J., Reducing accidents using goal set-
ting and feedback: a field study, J. Organ. Psychol., 67, 219, 1994.
Covey, S. R., The Seven Habits of Highly Effective People, Simon & Schuster, New York, 1989.
Eysenck, H. J., The Structure of Human Personality, 3rd ed., Methuen, London, 1976.
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 262

262 Psychology of safety handbook

Eysenck, H. J. and Eysenck, M. W., Personality and Individual Differences: a Natural Science Approach,
Plenum Press, New York, 1985.
Geller, E. S., If only more would actively care, J. Appl. Behav. Anal., 24, 607, 1991.
Geller, E. S., Ten principles for achieving a Total Safety Culture, Prof. Saf., 39(9) 18, 1994.
Geller, E. S., Safety coaching: key to achieving a total safety culture, Prof. Saf., 40(7), 16, 1995.
Geller, E. S., Managing the human element of occupational health and safety, in Essentials of
Occupational Safety and Health Management, Lack, R. W., Ed., CRC Press/Lewis Publishers, Boca
Raton, FL, 1996.
Geller, E. S., Applications of behavior analysis to prevent injuries from vehicle crashes, 2nd ed.,
Cambridge Center for Behavioral Studies Monograph Series: Progress in Behavioral Studies,
Monograph No. 2. Cambridge Center for Behavioral Studies, Cambridge, MA, 1998a.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998b.
Geller, E. S., Practical Behavior-Based Safety: Step-by-Step Methods to Improve Your Workplace, 2nd ed.,
J. J. Keller & Associates, Inc., Neenah, WI, 1998c.
Geller, E. S., Behavior-based safety: confusion, controversy and clarification, Occup. Health Saf.,
68(1), 40, 1999.
Geller, E. S., Maintaining involvement in occupational safety: fourteen key points, Occup. Health
Saf., 69(1), 72, 2000.
Geller, E. S., Berry, T. D., Ludwig, T. D., Evans, R. E., Gilmore, M. R., and Clarke, S. W., A conceptual
framework for developing and evaluating behavior change interventions for injury control,
Health Educ. Res. Theor. Pract., 5(2), 125, 1990.
Geller, E. S., Eason, S. L., Phillips, J. A., and Pierson, M. D., Intervention to improve sanitation dur-
ing food preparation, J. Organ. Behav. Manage., 2(3), 229, 1980.
Geller, E. S., Winett, R. A., and Everett, P. B., Preserving the Environment: New Strategies for Behavior
Change, Pergamon Press, New York, 1982.
Glenwick, D. S. and Jason, L. A., Behavioral Community Psychology: Progress and Prospects, Praeger,
New York, 1980.
Glenwick, D. S. and Jason, L. A., Eds., Promoting Health and Mental Health in Children, Youth, and
Families, Springer, New York, 1993.
Goodman, J. B., Laughing matters: taking your job seriously and yourself lightly, J. Am. Med. Assoc.,
267(13), 1858, 1995.
Greene, B. F., Winett, R. A., Van Houten, R., Geller, E. S., and Iwata, B. A., Eds., Behavior Analysis in
the Community: Readings from the Journal of Applied Behavior Analysis, University of Kansas Press,
Lawrence, KS, 1987.
Hall, E. T., The Silent Language, Doubleday, Garden City, NY, 1959.
Hall, E. T., The Hidden Dimension, Doubleday, Garden City, NY, 1966.
Heinrich, H. W., Petersen, D., and Roos, N., Industrial Accident Prevention: a Safety Management
Approach, 5th ed., McGraw-Hill, New York, 1980.
Katz, E. and Lazarfeld, P. E., Personal Influence: The Part Played by People in the Flow of Mass
Communication, Free Press, Glencoe, Il, 1955.
Kim, J. and Hammer, C., Effect of performance feedback and goal setting on productivity and satis-
faction in an organizational setting, J. Appl. Psycho., 61, 48, 1982.
Komaki, J., Heinzmann, A. T., and Lawson, L., Effect training and feedback: component analysis of
a behavioral safety program, J. App. Psychol., 65(3), 261, 1980.
Krause, T. R., Hidley, J. H., and Hodson, S. J., The Behavior-Based Safety Process: Managing
Involvement for an Injury-Free Culture, 2nd ed., Van Nostrand Reinhold, New York, 1996.
Mayo, E., The Human Problems of an Industrialized Civilization, Harvard University Graduate School
of Business Administration, Boston, MA, 1933.
McSween, T. E,. The Values-Based Safety Process: Improving Your Safety Culture with a Behavioral
Approach, Van Nostrand/Reinhold, New York, 1995.
National Safety News, Laughter could really be the best medicine, p. 15, 1985.
Parsons, H. M., What happened at Hawthorne?, Science, 183, 922, 1974.
Parsons, H. M., Lessons for productivity from the Hawthorne studies, in Proceedings of Human
820049_CRC1_L1540_CH12 11/10/00 1:55 PM Page 263

Chapter twelve: Intervening as a behavior-change agent 263

Factors Symposium sponsored by the Metropolitan Chapter of the Human Factors Society, Columbia
University, New York, 1980.
Petersen, D., Safe Behavior Reinforcement, Aloray, Inc., New York, 1989.
Phares, E. J., Locus of Control: a Personality Determinant of Behavior, General Learning Press,
Morristown, NJ, 1973.
Rothlisberger, F. J. and Dickson, W. J., Management and the Worker, Harvard University Press,
Cambridge, MA, 1939.
Skinner, B. F., About Behaviorism, Alfred A. Knoff, New York, 1974.
Sulzer-Azaroff, B., Behavioral approaches to occupational health and safety, in Handbook of
Organizational Behavior Management, Frederiksen, L. W., Ed. John Wiley and Sons, New York,
1982.
Sulzer-Azaroff, B. and De Santamaria, M. C., Industrial safety hazard reduction through perform-
ance feedback, J. Appl. Behav. Anal., 13, 287, 1980.
Tel-A-Train, Inc., Actively Caring for Safety, Wescott Communications, Dallas, TX, 1995.
Triandis, H. C., Interpersonal Behavior, Brooks/Cole, Monterey, CA, 1977.
Triandis, H. C., The self and social behavior in differing cultural contexts, J. Personal. Soc. Psychol.,
96, 506, 1985.
Watson, D. L. and Tharp, R. G., Self-Directed Behavior: Self-Modification for Personal Adjustment, 7th
ed., Brooks/Cole, Pacific Grove, CA, 1997.
Whitehead, T. N., The Industrial Worker, Harvard University Press, Cambridge, MA, 1938.
Williams, J. H. and Geller, E. S., Behavior-based intervention for occupational safety: critical impact
820049_CRC1_L1540_CH13 11/11/00 2:38 PM Page 265

chapter thirteen

Intervening with supportive

conversation
Interpersonal conversation defines the culture in which we work. It can create conflict and build bar-
riers to safety improvement or it can cultivate the kind of work culture needed to make a major break-
through in injury prevention. Interpersonal conversation also affects our intrapersonal
conversations or self-talk, which in turn influences our willingness to get involved in safety-
improvement efforts. This chapter explains the reciprocal impact of inter- and intrapersonal conver-
sation and offers guidelines for aligning both toward the achievement of a Total Safety Culture.

“Leadership is the ability to persuade others to do what you want them to do because
they want to do it.”—Dwight Eisenhower

Up to this point, the intervention procedures in Section 4 have been relatively formal. In
other words, I recommended a set of guidelines for developing and implementing activa-
tor strategies (Chapter 10) to direct behavior, for combining activator and consequence
strategies to motivate behavior (Chapter 11), and for using interpersonal coaching to both
direct and motivate behavioral improvement (Chapter 12).
Coaching was presented as a rather formal step-by-step process whereby a critical
behavior checklist (CBC) is developed and used to observe and analyze the safe vs. at-risk
behaviors occurring in a particular work procedure. Then, as illustrated in Figure 13.1, the
CBC is used to present directive and/or motivational feedback in a one-to-one interpersonal
conversation. Also, percent safe scores are derived from a variety of CBCs and presented on
a group feedback chart. Comments written on the CBCs are discussed in group meetings to
analyze areas of concern and to find ways to make safe behavior more likely to occur.
This chapter is also about interpersonal conversation and coaching, but the emphasis
is on brief informal communication to support safe behavior and help it become more flu-
ent. How we talk with others (interpersonal communication) influences their attitude and
ongoing behavior, and how we talk to ourselves (intrapersonal communication) influences
our own behavior and attitude. Therefore, this chapter also addresses self-talk—the men-
tal scripts we carry around in our heads before, during, and after our behaviors.
It is fair to say that the nature of our safety-related conversations with others influences
their degree of involvement in safety. The variety of the safety-related conversations we
have with ourselves influences whether we feel accountable to someone else for our safe
behaviors or whether we feel self-accountable for our safety-related behaviors. This is the
distinction I introduced in Chapter 9 between feeling accountable or other-directed vs.
responsible or self-directed.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 266

266 Psychology of safety handbook

Figure 13.1 Formal coaching includes the use of a checklist to give one-on-one behavioral
feedback.

The bottom line is that I believe the long-term success of any effort to prevent injuries
in the workplace, in the home, and on the road is determined by conversation.

The power of conversation

I am convinced the dramatic success companies experience with behavior-based safety is
essentially owing to an increase in the quality and quantity of safety correspondence—not
the high-tech communication referred to in Figure 13.2, but one-to-one interpersonal

Figure 13.2 The power of conversation comes from face-to-face communication.

820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 267

Chapter thirteen: Intervening with supportive conversation 267

conversation about safety. Such improvement, in turn, benefits people’s self-talk about
safety, increasing their sense of personal control and optimism regarding their ability to
prevent occupational injuries.
This chapter offers guidelines and techniques for getting more beneficial impact from our
conversations with others and with ourselves. Then four types of safety management are pre-
sented, each defined by the nature of interpersonal conversation. I cannot expect you to spend
the time and effort needed to improve communications until you truly appreciate the power
of conversation. So let us consider the impact of effective conversation or the lack thereof.

Building barriers
Almost everyone has seen how lack of communication can escalate a minor incident into
major conflict. Here is an example: You see a coworker and say, “Hello,” yet she passes by
without reacting. Maybe she did not see you, or had other thoughts on her mind. Still, it is
easy for you to assume the person is unfriendly or does not like you. So the next time you
see this person, you avoid being friendly. You might even talk about that person’s
“unfriendliness” to others. I think you can see how the barrier starts to build.
This is only one of many situations that can stifle interpersonal communication and
lead to negative feelings and judgments. The result is perhaps the perception of interper-
sonal conflict, an unpleasant relationship, lowered work output, and reduced willingness
to actively care for another person’s safety.

Resolving conflict
If the lack of conversation can initiate or fuel conflict, it is not surprising that the occurrence
of conversation can prevent or eliminate conflict. “Let’s talk it out,” as the saying goes. Of
course, it is the quality of that conversation that will determine whether any perceived con-
flict is heightened or lessened. This issue of conversation quality is covered later in this
chapter. Here I only want you to consider the power of interpersonal talk. It can make or
break interpersonal conflict which, in turn, enables constructive or destructive relation-
ships. The nature of relationships determines whether individuals are willing to actively
care for another person’s safety and health (Geller, 1994, 2000a,b).

Bringing intangibles to life

What are love, friendship, courage, loyalty, happiness, and forgiveness? Sure, you can
describe behaviors that reflect these concepts, but where is the true meaning? I think we
derive the meaning of these common words from our conversations. Think about how we
“fall in or out of love” depending on how we talk to ourselves and others. Likewise, we can
convince ourselves we are happy through our self-talk, and this inner conversation is obvi-
ously influenced by what we hear others say about us.
We define another person’s friendship, courage, or loyalty by talking about that indi-
vidual in certain ways, both to ourselves and to others. Our mental scripts and verbal
behavior are powerful—giving useful meaning to concepts that define the very essence of
human existence. When groups, organizations, or communities communicate to define
these concepts, we get a “culture.” Perhaps, it is fair to say that culture is conversation—
both spoken and unspoken.

Defining culture
Does the term “unspoken conversation” make sense? I am referring to customs or unwrit-
ten rules we heed without mention. For example, we might realize the “teacher’s pets” sit
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 268

268 Psychology of safety handbook

in the front row or the boss does not want to hear about a “near miss” or the real purpose
of the safety incentive program is to stifle the reporting of OSHA recordables. We might
also know characteristics that bias certain managers’ performance appraisals, from gender
and seniority to ability on the golf course.
It might be understood that a male with high seniority and a low golf handicap is more
likely to get the special training assignment, but such prejudice is certainly not expressed.
If it were, a productive conversation would be possible—one that could reduce the conflict
and bias that hinder the optimization of a work system and the achievement of a Total
Safety Culture.
The bottom line is that spoken and unspoken words define cultures and subcultures,
and then cultures can change, for better or worse, through interpersonal and intrapersonal
conversation.

Defining public image and self-esteem

Public conversation defines public image, whether referring to an individual or an organ-
ization. Sometimes, though, different groups talk about an individual or organization in
different ways. As a result, we have mixed messages and an inconsistent public image. The
image of our president changes, depending on whether you are listening to Democrats or
Republicans. However, as shown in Figure 13.3, public conversation has generally
demeaned the image of most politicians.
The safety image of an organization can vary, depending on who is doing the talking.
Safety professionals might question a company’s touting of “zero injuries,” while the pub-
lic would not doubt this company’s elite ranking.
How we talk to ourselves both influences and reflects our self-esteem. In fact, it is prob-
ably fair to say our mental script about ourselves is our self-esteem. We can focus our self-
talk on the good things people say about us or on other people’s critical statements about

Figure 13.3 Conversation can build or demean public image.

820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 269

Chapter thirteen: Intervening with supportive conversation 269

Figure 13.4 Negative self-talk can ruin a good thing.

us. The result is a certain kind of self-talk we call “interpretation.” Such intrapersonal
communication can increase or decrease how we feel about ourselves. In other words, our
self-esteem can go up or down according to how we talk to ourselves about the way others
talk about us. As illustrated in Figure 13.4, negative self-talk can interfere with the kind of
positive experiences that should lead to positive intrapersonal communication and an
increase in self-esteem.

Making breakthroughs
In his provocative book Leadership and the Art of Conversation, Krisco (1997) defines a
breakthough as going beyond business as usual and getting more than expected. This
requires people to realize new possibilities, commit to going for more, and then make a
concerted effort to overcome barriers. So how can we visualize possibilities, show com-
mitment to go for a breakthrough, and identify barriers to overcome? You guessed it—
through conversation.
Expect barriers and resistance to change warns Krisco. The greater the change, the
greater the resistance, but remember, most barriers to change are interpretations or peo-
ple’s self-talk about perceived reality. Conversation, both interpersonal and intrapersonal,
enables us to overcome the barriers that hold back the accomplishment of breakthroughs.

In summary
Given the power of conversation to resolve interpersonal conflict, achieve breakthroughs,
and define public image, self-esteem, and culture, we need to direct this powerful tool to
support safety. How do we maximize the impact of our interpersonal and intrapersonal
conversations? What kinds of conversations are more likely to provoke and maintain ben-
eficial improvement in occupational safety? This is the theme for the remainder of this
chapter.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 270

270 Psychology of safety handbook

The art of improving conversation

The focus here is on improving safety-related conversations between people. Conversa-
tion, as discussed previously, is a powerful tool that shapes personal and team attitudes
about loyalty, commitment, social support, and safety. Each of the techniques presented
here can get employees more involved in safety, and improve the overall level of workplace
safety performance.
Applying these techniques can also improve how you talk to yourself—your
“self-talk.” The payoff is increased self-esteem and perceptions of empowerment,
which are essential for increasing our willingness to actively care for the safety and health
of others.

Do not look back

Has this ever happened to you? You ask for more safety involvement from a particular indi-
vidual and you get a reaction like, “I offered a safety suggestion three years ago and no one
responded.” You may have attended a safety meeting where people spent more time going
over past accomplishments or failures than discussing future possibilities and deriving
action plans.
These are examples of conversations stuck in the past. The discussion might be enjoy-
able but little or no progress is made. Conversations about past events help us connect with
others and recognize similar experiences, opinions, and motives, but such communication
does not enable progress toward problem solving or continuous improvement. For this to
happen, the conversation must leave the past and move on.
Krisco (1997) maintains that leaders need to help people move their conversations from
the past to the future and then back to the present. If you want conversation to lead to
improvement-focused behavior, possibilities need to be entertained (future talk) and then
practical action plans need to be developed (present talk). This is the case for group
conversation at a team meeting, as well as for one-on-one advising, counseling, or safety
coaching.
The power of future talk is illustrated by President John F. Kennedy’s vision, stated on
May 25, 1961, that by the end of the decade the United States would put people on the
moon and return them safely to earth. Many thought this prediction was absurd. While the
Russians had completed several successful space missions at the time, America lagged
behind in the “space race.” Renowned U.S. scientists warned that a moon landing was
impossible because of insufficient fuel and computer technology. Yet, on July 20, 1969, the
world watched in awe when astronaut Neil Armstrong took that “giant leap for mankind”
and planted an American flag on the moon.
It obviously took a lot of science and technology, and monumental team effort to pull
off that historic lunar landing. One must wonder, however, if that mission would ever have
been accomplished if the leader of our country had not communicated his vision—his
future talk. Some say President Kennedy actually spoke a manned lunar landing into exis-
tence (Blair, 2000; Krisco, 1997). Progress begins with transitioning conversation to the
future (the vision) and then returning to the present for the development of goal-directed
action plans.
To direct the flow of a conversation from past to future and then to the present, you first
must recognize and appreciate what the other person has to say. Then shift the focus
toward the future. Remember, you are approaching this person to discuss possibilities for
safety improvement and specific ways to get started now.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 271

Chapter thirteen: Intervening with supportive conversation 271

Seek commitment
You know your interpersonal conversation is especially productive when someone makes
a commitment to improve in a certain way. This reflects success in moving conversation
from the past to the future and then to a specific action plan. A verbal commitment also tells
you that something is happening on an intrapersonal level within that other person. The
person is becoming self-motivated, increasing the probability the target behavior will
improve (Cialdini, 1993).
Now you can proceed to talk about how that commitment can be supported or how to
hold the individual accountable. For example, one person might offer to help a coworker
meet an obligation through verbal reminders or an individual might agree to honor a com-
mitment by showing a coach behavioral records that indicate improvement. This is, of
course, the kind of follow-up conversation that facilitates personal achievement.

Stop and listen

In their eagerness to prevent injury, safety advocates often give corrective feedback in a
top-down, controlling manner. In other words, passion for safety sometimes leads to an
overly directive approach to get others to change their behavior. You know from personal
experience, and clinical psychologists have shown through research, that a nondirective
approach to giving advice is often more effective, especially over the long term (Bandura,
1982, 1997; Deci and Ryan, 1985; Ryan and Deci, 2000).
Think about it. How do you respond when someone overtly tells you what to do? Now
it certainly depends on who is giving the instruction, but I bet your reaction is not entirely
positive. You might follow the instruction, especially if it comes from someone with the
power to control consequences, but how will you feel? Will you be motivated to make a
permanent change? You might if you asked for the direction, but if you did not request
feedback, you could feel insulted or embarrassed.
Corrective feedback that can be interpreted as an “adult-child” confrontation will
probably not work. The supervisor in Figure 13.5 means well, but the worker does not see
it that way. When a directive conversation is interpreted as controlling or demeaning, it is
essentially ineffective, so play it safe. Try to be more nondirective when using interpersonal
conversation to affect behavior change.
Let me explain what I mean here. The theme of nondirective psychotherapy is active
listening (Rogers, 1951). The objective is to get clients to reveal their concerns, problems,
and solutions on their own terms. The therapist’s role is to be a passive catalyst, enabling
and facilitating a conversation that is directed and owned by the client. I am not suggest-
ing safety leaders become therapists, but we can take some useful lessons from this nondi-
rective approach.
The mother in Figure 13.6 certainly means well, but her directive stance causes mis-
communication. A nondirective perspective would allow mom to understand where the
child is coming from. As Covey (1984) recommends with his fifth habit for highly effective
people, “Seek first to understand before being understood.”

Ask questions first

Instead of telling people what to do, try this. Get them to tell you, in their own words, what
they ought to be doing in order to be safe. You can do this by asking questions with a sin-
cere and caring demeanor. Avoid at all costs a sarcastic or demeaning tone, but, first, point
out certain safe behaviors you noticed—it is important to emphasize positives.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 272

272 Psychology of safety handbook

Figure 13.5 Corrective feedback can feel demeaning.

Figure 13.6 A directive paradigm can stifle understanding.

Then move on to the seemingly at-risk behavior by asking, “Is there a safer way to per-
form that task?” Of course, you hope for more than a “yes” or “no” response to a question
like this. However, if that is all you get, you need to be more precise in follow-up ques-
tioning. You might, for example, point out a particular work routine that seems risky and
ask whether there is a safer way.
I recommend approaching a corrective feedback conversation as if you do not know
the safest cooperating procedures, even though you think you do. You might, in fact, find
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 273

Chapter thirteen: Intervening with supportive conversation 273

Figure 13.7 Diagnosis requires questions and answers.

your presumptions to be imperfect. The “expert” on the job might know something you do
not know. If you approach the situation with this mindset, you will not get the kind of reac-
tion given by the woman in Figure 13.7.
By asking questions, you are always going to learn something. If nothing else you will
hear the rationale behind taking a risk over choosing the safer alternative. You might
uncover a barrier to safety that you can help the person overcome. A conversation that
entertains ways to remove obstacles that hinder safe behavior is especially valuable if it
translates possibilities into feasible and relevant action plans.
You will know your nondirective approach to correction worked if your colleague
owns up to his or her mistake even under a cloud of excuses. Remember, it is only natural
to offer a rationale for taking a risk. It is a way of protecting self-esteem. Let it pass, and
remind yourself that when someone admits a mistake before you point it out, there is a
greater chance for both acceptance of responsibility and behavior to change.

Transition from nondirective to directive

What if the person does not give a satisfactory answer to your questions about safer alter-
natives? What if the individual does not seem to know the safest operating procedure?
Now you need to shift the conversation from nondirective to directive. You need to give
behavior-focused advice.
In this case, start with the phrase, “As you know,” as my friend Drebinger (2000)
advises. Open the conversation with a phrase that implies the person really does know the
safe way to perform, but for some reason just overlooked it (or forgot) this time. This could
happen to anyone. Such an opening can help prevent others from feeling their intelligence
or safety knowledge has been insulted.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 274

274 Psychology of safety handbook

Beware of bias
Every conversation you have with someone is biased by prejudice or prejudgment filters—
in yourself and within the other person. You cannot get around it. From personal experi-
ence, people develop opinions and attitudes and these, in turn, influence subsequent
experience. With regard to interpersonal conversation, we have subjective prejudgment fil-
ters that influence what words we hear, how we interpret those words, and what we say in
response to those words. In Chapter 5, I referred to this bias as premature cognitive com-
mitment (Langer, 1989). Every conversation influences how we process and interpret the
next conversation.
Figure 13.8 illustrates what I mean. The female driver is merely trying to inform the
other driver of an obstacle in the road, but that is not what the driver of the pick-up hears.
This driver’s prior driving experiences lead to a biased interpretation of the warning. You
could call such selective listening an “autobiographical bias” (Covey, 1989). Of course, fac-
tors besides prior experience can bias interpersonal communication, including personality,
mood state, physiological needs, and future expectations.
It is probably impossible to escape completely the impact of this premature bias in our
conversations, but we can exert some control. Actually, each of the conversation strategies
discussed here is helpful. For example, the nondirective approach attempts to overcome
this bias by listening actively and asking questions before giving instruction. With this
approach, a person’s biasing filters can be identified and considered in the customization
of a plan for corrective action.
Certain words or phrases in a conversation can be helpful in diminishing the impact of
prejudice filter. For example, when you say “as you know” before giving behavior-based
advice, you are limiting the perception of a personal insult and the possibility of a “tune-
out” filter. By asking people for their input up front, you reduce the likelihood they will
later tune you out. It is the principle of reciprocity (Cialdini, 1993). By listening first, you
increase the odds the other person will listen to you without a tune-out filter.

Figure 13.8 Selective listening can be hazardous to your health.

820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 275

Chapter thirteen: Intervening with supportive conversation 275

If you think a person might tune you out because he or she heard your message before,
you could use opening words to limit the power of tune-out filters. Specifically, you might
start the conversation with something like, “Now I realize you might have heard some-
thing like this before, but. . . . “ In this way, you are anticipating the kind of intrapersonal
conversation (or mental script) that triggers a tune-out reaction and, therefore, you reduce
such filtering.
In the same vein, do not let your prejudices about a speaker limit what you hear. Do you
ever listen less closely to certain individuals, perhaps because the person seldom has any-
thing useful to say or because you think you can predict what he will say? If so, you have
let your past conversations with this individual bias future conversation. Becoming aware
of this “stuck-in-the-past” prejudice can enable more active, even proactive, listening.
Do not let the speaker limit what you hear. Tell yourself you are not listening to some-
one, rather you are listening for something (Krisco, 1997). You are not listening reactively to
confirm a prejudice—you are listening proactively for possibilities.
Pay close attention to the body language and tone in conversations. I am sure you have
heard many times that the method of delivery can hold as much or more information as the
words themselves. Listen for passion, commitment, or caring. If nothing else, you could
learn whether the messenger understands and believes the message and, perhaps, you will
learn a new way to deliver a message yourself. The bottom line is our intrapersonal con-
versations can either facilitate or hinder what we learn from interpersonal conversation.

Plant words to improve self-image

Earlier in this chapter, I discussed how conversation influences both public and self-image.
How we talk about others influences interpersonal perceptions. How we hear others talk
about us shapes our own self-image and how we talk to ourselves about these viewpoints
can make them a permanent feature of our self-concept or self-esteem.
Do you want to change how others perceive you? Change the conversations people are
having about you. Through proactive listening, you can become aware of negative inter-
personal conversations about you and then you can interject new statements about your-
self into conversations, especially with people who have numerous contacts with others.
If you suspect, for example, that colleagues consider you to be forgetful and disorgan-
ized, you could mention certain self-management strategies you have been using lately to
improve memory and organization. Of course, you need to actually practice these tech-
niques so you will also change your self-dialogue. If you focus on new positive qualities
rather than past inadequacies in your conversations with others and with yourself, you will
surely improve your self-image and self-esteem. Plant key messages about your commit-
ment to become a more effective safety leader and you will eventually see yourself that
way and behave accordingly.

In summary
The strategies covered here for getting the most from interpersonal conversation are
reviewed in Figure 13.9. Each technique is relevant for getting more safety-related
involvement from others. Applying these strategies effectively can improve one’s self-talk
or intrapersonal conversation. This leads to increased self-esteem and perceptions of
empowerment—person states which enhance an individual’s willingness to actively
care for the safety and health of others. Evidence for this is detailed in Section 5 of this
Handbook.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 276

276 Psychology of safety handbook

Safety Conversation Checklist

 Listen attentively and proactively.
 Focus on the positive actions observed.
 Draw out responses from the other person.
 Influence others to tell you what they should do to be
safe.
 Ask questions with a sincere and caring demeanor.
 Act as if you don’t know the answer, even though you
think you do.
 Shift the focus to future possibilities for improvement.
 Bring the conversation back to the present and develop
an action plan.
 Seek a verbal commitment to follow the action plan.
 Plant words to improve public and self-image.

Figure 13.9 Follow these strategies to get the most from interpersonal conversation.

First, consider the tendency to focus interpersonal and intrapersonal conversation on

the past. This helps us connect with others, but it also feeds our prejudice filters and limits
the potential for conversation to facilitate beneficial change. We enable progress when we
move conversations with ourselves and others from past to future possibilities and then to
the development of an action plan.
Expect people to protect their self-esteem with excuses for their past mistakes. Listen
proactively for barriers to safe behavior reflected in these excuses. Then help the conversa-
tion shift to a discussion of possibilities for improvement and personal commitment to
apply a practical action plan. This is more likely to occur with a nondirective than a direc-
tive approach, in which more questions are asked than directives given, and when open-
ing words are used to protect self-esteem and limit the impact of reactive bias.
Remember that planting certain words in self-talk and conversations with others can
improve your self-image and confidence as a facilitator of beneficial change. Tell others of
your increased commitment to facilitate more effective safety conversations. Then, tell
yourself the strategies you will use to improve interpersonal conversation and commend
yourself when you do. In this way, intra- and interpersonal conversations work together to
help achieve a Total Safety Culture.

Conversation for safety management

Safety is managed through conversation, and the success of safety management is deter-
mined, in large part, by the effectiveness of interpersonal communication. This starts with
listening proactively to understand the other person’s situation before giving direction,
advice, or support. Then one of four types of interpersonal conversation should occur,
depending on what kind of management is called for. As depicted in Figure 13.10 the con-
versation can reflect coaching, supporting, instructing, or delegating (Blanchard et al.,
1985), depending on the amount of direction and motivation given.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 277

Chapter thirteen: Intervening with supportive conversation 277

Figure 13.10 Management conversation is determined by amount of direction and moti-

vation needed.

Coaching conversation
As detailed in Chapter 12, coaches give direction and provide feedback. They present a
plan, perhaps specific behaviors needed for a certain task, and then follow up with support
and empathic correction to pinpoint what worked and what did not. Periodic reminders
keep people on the right track, while intermittent recognition provides support to keep
people going.

Delegating conversation
Sometimes it is best to give an assignment in general terms (without specific direction) and
to limit interpersonal behavior-focused feedback. This is when team members are already
motivated to do their best and will give each other direction, support, and feedback when
needed. These individuals should be self-accountable (or responsible) and expected to use
self-management techniques (activators and consequences) to keep themselves motivated
and on the right track (Geller, 1998a,b; Geller and Clarke, 1999).

Instructive conversation
Some people are already highly motivated to perform well, but do not know exactly what
to do. This is often the case with new hires. They want to make a good first impression, and
the newness of the job is naturally motivating. They are nervous, however, because of
response uncertainty. They are not sure what to do in the relatively novel situation. In this
case, managers need to focus on giving behavior-focused instruction.
This type of conversation should also be the approach at most athletic events.
Individuals and teams in a sports contest do not typically need motivation. The situation
itself, from fan support to peer pressure, often provides plenty of extrinsic motivation. Such
competitors need directional focus for their motivation. They need to know what specific
behaviors are needed to win in various situations. This said, my personal experience with
athletic coaches is not consistent with this analysis. For example, are the half-time speeches
of team coaches more likely to be directional or motivational?
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 278

278 Psychology of safety handbook

Supportive conversation
What about the experienced worker who does the same tasks day after day? This individ-
ual does not need direction, but could benefit from periodic expressions of sincere thanks
for a job well done.
There are times when experienced workers know what to do but do not consistently
perform up to par. This is not a training problem, but rather one of execution (Geller,
2000b). Through proactive listening, a manager can recognize this and provide the kind of
support that increases motivation. This could involve broadening a job assignment, vary-
ing the task components, or assigning leadership responsibilities. At least, it includes the
delivery of one-to-one recognition in ways that increase a person’s sense of importance
and self-worth.

Recognizing safety achievement

Each of the four management styles reviewed here includes supportive conversation. So,
let us discuss effective ways to do this. First, let us realize we are much more inclined to
notice the mistakes people make. In fact, we are more inclined to beat ourselves up for our
own mistakes, instead of celebrating our personal successes. Now, how can this be
explained? Why do we pay more attention to the negative things in our lives?
One reason is the mistakes stick out. They upset the flow and are readily noticed. When
people are doing the right thing, the process runs smoothly, and we keep on going. We
go with the flow. We hardly notice the variety of good behaviors occurring at the time.
Instead of being a “good finder,” we wait for the obvious mistake and then make our move.
The women in Figure 13.11 have a knack for finding good in a situation when the bad
is obvious.
Another reason for our focus on the negative is we have come to believe people learn
best by making mistakes. We think paying attention to errors is the best way to improve
performance. Perhaps, you have heard a pop psychologist or motivational speaker assert

Figure 13.11 Even when the context is negative, “good-finders” struggle to find a silver
lining.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 279

Chapter thirteen: Intervening with supportive conversation 279

we need to fail in order to learn. I heard one attempt to make his point by asking, “Who
was the greatest home-run hitter in the history of baseball?” The audience shouted, “Babe
Ruth,” and the presenter agreed. Then, the motivational speaker asked, “Do you know
who struck out more times than any professional baseball player?” The audience was
primed to respond, “Babe Ruth,” and again the “expert” on stage supported this answer.
The implication is that “The Babe” learned his fantastic skill by making errors. Now,
that might make us feel better about our own mistakes, but it is wrong. Hank Aaron hit
more home runs than Babe Ruth. Reggie Jackson hit fewer home runs than either of them,
but he is the player who struck out the most. If we believe people need to make mistakes
in order to learn, and we act on that belief, we can do more harm than good. It can be an
excuse for focusing more on failures than on successes.
Behavioral scientists have shown quite convincingly that success—not failure—pro-
duces the most learning (Chance, 1999). Thorndike (1911, 1931) did critical research on this
at the beginning of the century. He placed chickens, cats, dogs, fish, monkeys, and humans
in various problem-solving situations. As a result, he observed they solved the problems
through a process he called “trial and accidental success.”
At first, Thorndike’s subjects tried various random behaviors. When a behavior
resulted in no gain, the behavior was less likely to occur again. However, when a behavior
was followed by success, the behavior was much more likely to be repeated. Thorndike’s
subjects learned to solve the problems with greater and greater ease by discovering which
behavior worked and then repeating that behavior. He coined the “Law of Effect” to refer
to the fact that learning depends upon the consequences of behavior.
Now, it is ironic that when people began talking and writing about Thorndike’s work,
many referred to this type of learning as “trial and error learning.” That is where the com-
mon phrase “learning by trial and error” came from, but Thorndike knew better. We do
learn something from our mistakes. They teach us what not to do, but the positive conse-
quences—the successes, not the failures—produce the most learning.
Thus, it is easy to understand why we criticize more than we praise. It is clear, how-
ever, that we need to turn that around. We learn best when we get positive reinforcement
for doing the right thing. As discussed in Chapter 11, positive consequences are good for
our attitude. You know how you feel when you get recognition—when it is genuine. You
feel good, and that is what we need. We need people feeling good about themselves when
they go out of their way for safety.
We need to have the same mindset about safety that the gold prospectors had about
their challenge. Their focus was in finding gold. They sifted to find the good, not the bad.
Likewise, we need to prospect for the good behaviors, even when the bad might be more
obvious. Mom has the right idea in Figure 13.12.
After finding good behavior, it is important to recognize the right way. Most of us have
not been taught how to give recognition effectively. Our common sense is not sufficient.
Behavioral research, however, has revealed strategies for making interpersonal recognition
most rewarding. When you know how to maximize the impact of your recognition, you
might use this powerful supportive intervention more often. Listed in Figure 13.13 are
seven guidelines for giving quality recognition. Let us consider each one in order.

Recognize during or immediately after safe behavior

In order for recognition to provide optimal direction and support, it needs to be associated
directly with the desired behavior. People need to know what they did to earn the appre-
ciation. If it is necessary to delay the recognition, then the conversation should relive the
activity that deserves recognition. Reliving the behavior means talking specifically about
what warrants the attention. You could ask the person you are recognizing to describe
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 280

280 Psychology of safety handbook

Figure 13.12 Prospect for the good in others.

aspects of the situation and the desirable behavior. This facilitates direction and motivation
to continue the behavior. When you connect a person’s behavior with recognition you also
make the supportive conversation special and personal.

Make recognition personal for both parties

A supportive conversation is most meaningful when it is personal. The recognition should
not be general appreciation that could fit anyone in any situation. Rather, it needs to be cus-
tomized to fit a particular individual and circumstance. This happens, naturally, when the
recognition is linked to specific behavior.
When you recognize someone you are expressing personal thanks. It is tempting to say
“we appreciate” rather than “I appreciate” and to refer to company gratitude instead of

How to Give Supportive Recognition

❑ Recognize during or immediately after
safe behavior.
❑ Make it personal for both parties.
❑ Connect specific behavior with general
higher-level praise.
❑ Deliver it privately and one-on-one.
❑ Let it stand alone and soak in.
❑ Use tangibles for symbolic value only.
❑ Second-hand recognition has special
advantages.

Figure 13.13 Follow seven conversation guidelines when giving recognition to support
safety achievement.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 281

Chapter thirteen: Intervening with supportive conversation 281

personal gratitude. Speaking for the company can come across as impersonal and insin-
cere. Of course, it is appropriate to reflect value to the organization when giving praise, but
the focus should be personal. “I saw what you did to support our safety process and I really
appreciate it. Your example illustrates actively caring and demonstrates the kind of lead-
ership we need around here to achieve a Total Safety Culture.” This second statement illus-
trates the next guideline for giving quality recognition.

Connect specific behavior with general higher-level praise

A supportive conversation is most memorable when it reflects a higher-order characteris-
tic. Adding a universal attitude like leadership, integrity, trustworthiness, or actively car-
ing to the recognition statement makes the recognition more rewarding and most likely to
increase the kind of intrapersonal communication that boosts self-esteem. It is important to
state the specific behavior first and then make a clear connection between the behavior and
the positive attribute it reflects.

Deliver recognition privately and one-on-one

Because quality recognition is personal and indicative of higher-order attributes, it needs to
be delivered in private. After all, the recognition is special and only relevant to one person.
So, it will mean more and seem more genuine if it is given from one individual to another.
It seems conventional to recognize individuals in front of a group. This approach is
typified in athletic contests and reflected in the pop psychology slogan, “Praise publicly
and reprimand privately.” Many managers take the lead from this common-sense state-
ment and give their individual recognition at group meetings. Is it not maximally reward-
ing to be held up as an exemplar in front of one’s peers? Not necessarily, as I mentioned
earlier in Chapter 1; many people feel embarrassed when receiving special attention in a
group. Part of this embarrassment is owing to fear of subsequent harassment by peers.
Some peers might call the recognized individual an “apple-polisher” or “brown-noser,” or
accuse him or her of “kissing up to management.”
In sports, individual performance is measured objectively and the winner is deter-
mined fairly. While behavior-based safety recognition is also objective, it is usually impos-
sible to assess everyone’s safety-related behaviors and obtain a fair ranking for individual
recognition. However, such ranking sets up a win–lose atmosphere. This may be appro-
priate for sporting events, but it is certainly inappropriate in a work setting where the elim-
ination of injuries is dependent upon everyone looking out for the safety of everyone else.
It is useful, of course, to recognize teams of workers for their accomplishments, and
this can be done in a group setting. Usually, group accomplishment worthy of recognition
can be documented for public review. Because individual responsibility is diffused or dis-
persed across the group, there is minimal risk of individual embarrassment or later peer
harassment. However, it is important to realize that group achievement is rarely the result
of equal input from all team members. Some take the lead and work harder, while others
do less and count on the group effort to make them look good. Thus, it is important to
deliver personal and private recognition to those individuals who went beyond the call of
duty for the sake of their team.

Let recognition stand alone and soak in

I have heard pop psychologists recommend a “sandwich method” for enhancing the
impact of interpersonal communication. “First say something nice, then give corrective
feedback, and then say something nice again.” This approach might sound good, but it is
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 282

282 Psychology of safety handbook

not supported by communication research. In fact, this mixed-message approach can cause
confusion and actually reduce credibility. The impact of initial recognition is canceled
by the subsequent correction. Then the corrective feedback is neutralized by the closing
recognition.
You need to keep a supportive conversation simple and to the point. Give your behav-
ior-based praise a chance to soak in. In this fast-paced age of trying to do more with less,
we try to communicate as much as possible when we finally get in touch with a busy per-
son. After recognizing a person’s special safety effort, we are tempted to tag on a bunch of
unrelated statements, even a request for additional behavior. This comes across as, “I
appreciate what you’ve done for safety, but I need more.” Resist the temptation to do more
than praise the good behavior you saw. If you have additional points to discuss, it is better
to reconnect later, after your praise has had a chance to sink in and become a part of the
person’s self-talk.
By giving quality interpersonal support, we give people a script they can use to reward
their own behavior. In other words, our quality recognition improves the other person’s
interpersonal conversation. Positive self-talk is crucial for long-term maintenance of safe
behavior. In other words, when we allow our recognition to stand alone and soak in, we
give people words they can use later for self-motivation.

Use tangibles for symbolic value only

Tangibles can detract from the self-motivation aspect of quality recognition. If the focus of
a recognition process is placed on a material reward, the words of appreciation can seem
less significant. In turn, the impact on one’s intrapersonal conversation system is lessened.
On the other hand, tangibles can add to the quality of interpersonal recognition if they
are delivered as tokens of appreciation. As discussed in Chapter 11, if tangibles include a
safety slogan, they can help to promote safety, but how you deliver a trinket will determine
whether it adds to or subtracts from the value of your supportive conversation. The bene-
fit of your praise is weakened if the tangible is viewed as a payoff for the safety-related
behavior. On the other hand, if the tangible is seen as symbolic of going beyond the call of
duty for safety, it strengthens the praise.

Secondhand recognition has special advantages

Up to this point, I have been discussing one-on-one verbal conversation in which one per-
son recognizes another person directly for a particular safety-related behavior. It is also
possible to recognize a person’s outstanding efforts indirectly, and such an approach can
have special benefits. Suppose, for example, you overhear me talk to another person about
your outstanding safety presentation. How will this secondhand recognition affect you?
Will you believe my words of praise were genuine?
Sometimes people are suspicious of the genuineness of praise when it is delivered face-
to-face. The receiver of praise might feel, for example, there is an ulterior motive to the
recognition. The deliverer of praise might be expecting a favor in return for the special
recognition. Perhaps both individuals had recently attended the same behavior-based
safety course, and the verbal exchange is viewed as only an extension of a communication
exercise. It, thus, will be devalued as sincere appreciation. Secondhand recognition, how-
ever, is not as easily tainted with these potential biases. Therefore, its genuineness is less
suspect.
Suppose I tell you that someone else in your workgroup told me about the superb
job you did leading a certain safety meeting. What will be the impact of this type of
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 283

Chapter thirteen: Intervening with supportive conversation 283

secondhand recognition? Chances are you will consider the recognition genuine because I
was only reporting what someone else said; because that person reported your success to
me rather than you, there was no ulterior motive for the indirect praise. Such secondhand
recognition can build a sense of belonging or win–win teamwork among people. When
you learn that someone was bragging about your behavior, your sense of friendship with
that person will probably increase.
My main point here is that gossip can be good—if it is positive. When we talk about the
success of others in behavior-specific terms, we begin a cycle of positive communication
that can support desired behavior. It also helps to build an internal script for self-motiva-
tion. We also set an example for the kind of inter- and intrapersonal conversations that
increase self-esteem, empowerment, and group cohesion. As explained in Section 5 of this
Handbook, these are the very person states that increase actively caring behaviors and cul-
tivate the achievement of a Total Safety Culture.

Receiving recognition well

The list of guidelines for giving quality recognition is not exhaustive, but it does cover the
basics. Following these guidelines will increase the benefit of a conversation to support
desirable performance. The most important point is that more recognition for safe behav-
ior is needed in every organization, whether given firsthand or indirectly through positive
gossip. It only takes a few seconds to deliver quality recognition.
Perhaps, realizing the beneficial consequences we can have on people’s behaviors and
attitudes with relatively little effort will be self-motivating enough for us to do more rec-
ognizing. Even more important, however, are the social consequences we receive when
attempting to give quality recognition. In other words, the reaction of the people who are
recognized can have dramatic impact on whether supportive conversations increase or
decrease throughout a work culture. We need to know how to respond to recognition in
order to assure quality recognition continues.
Most of us get so little recognition from others we are caught completely off guard
when acknowledged for our actions. We do not know how to accept appreciation when it
finally comes. Some claim they do not deserve the special recognition. Others actually
accuse the person giving recognition of being insincere or wanting something from them.
This can be quite embarrassing to the person doing the recognizing. It could certainly dis-
courage that person from giving more recognition.
Remember the basic motivational principle that consequences influence the behaviors
they follow. Well, this is true for both the person giving recognition and the person receiv-
ing recognition. Quality recognition increases the behavior being recognized and one’s
reaction to the recognition influences whether the recognizing behavior is likely to occur
again. Thus, it is crucial to react appropriately when we receive recognition from others.
Seven basic guidelines for receiving recognition are listed in Figure 13.14. Here is an expla-
nation for each.

Avoid denial and disclaimer statements

Whenever I attempt to give quality recognition, whether to a colleague, student, waitress,
hotel clerk, or a member of the baseball team I coach, the most common reaction I get is
awkward denial. Some act as if they did not hear me and keep doing whatever they are
doing, or they offer a disclaimer like, “It really was nothing special,” “Just doing my job,”
“I really could not have done it without your support,” or “Other members of our team
deserve more credit than I.”
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 284

284 Psychology of safety handbook

How to Receive Supportive Recognition

❑ Avoid denial and disclaimer statements.
❑ Listen actively with genuine appreciation.
❑ Relive recognition later for self-motivation.
❑ Show sincere appreciation.
❑ Recognize the person for recognizing you.
❑ Embrace the reciprocity principle.
❑ Ask for recognition when it is deserved but
not forthcoming.

Figure 13.14 Follow seven conversation guidelines when receiving recognition in order to
increase the occurrence of interpersonal support.
We need to accept recognition without denial and disclaimer statements and we
should not deflect the credit to others. It is okay to show pride in our small-win accom-
plishments, even if others contributed to the successful outcome. After all, the vision of a
Total Safety Culture includes everyone going beyond the call of duty for their own safety
and that of others. In this context, most people deserve recognition on a daily basis. It is not
“employee of the month,” it is “employee of the moment.”
Accept the fact that recognition will be periodic and inconsistent. When your turn
comes, accept the recognition for your current behavior and for the many safety-related
behaviors you performed in the past that went unnoticed. Remind yourself that your
genuine appreciation of the recognition will increase the chance more recognition will be
given to others.

Listen attentively with genuine appreciation

Listen proactively to the person giving you recognition. You want to know what you did
right. Plus, you can evaluate whether the recognition is given well. If the recognition does
not pinpoint a particular behavior, you might ask the person “What did I do to deserve
this?” This will help to improve that person’s method of giving recognition.
Of course, it is important not to seem critical but rather to show genuine appreciation
for the special attention. Consider how difficult it is for most people to go out of their way
to recognize others. Then revel in the fact you are receiving some recognition, even if its
quality could be improved. Remember that a person who recognizes you is showing grat-
itude for what you do and will come to like you more if you accept the recognition well.

Relive recognition later for self-motivation

Obviously, most of your safety-related behaviors go unnoticed. You perform many of these
when no one else is around to observe you. Even when other people are available, they will
likely be so preoccupied with their own routines they will not notice your extra effort. So
when you finally do receive some recognition, take it in as well-deserved. Remember the
many times you have gone the extra mile for safety but did not get noticed.
You need to listen intently to every word of praise, not only to show you care but also
because you want to remember this special occasion. Do not hesitate to relive this moment
later by talking to yourself. Such self-recognition can motivate you to continue going
beyond the call of duty for safety.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 285

Chapter thirteen: Intervening with supportive conversation 285

Show sincere appreciation

After listening actively with humble acceptance, you need to show sincere gratitude with a
smile, a “Thank you,” and maybe special words like, “You’ve made my day.” As I have
already emphasized, your reaction to being recognized can determine whether similar
recognition will occur again. So be prepared to offer a sincere “Thank you,” and words to
reflect pleasure in the special conversation. I find it natural to add “You’ve made my day,”
to the thank you because it is the truth. When people go out of their way to offer me quality
recognition, they have made my day and I often relive such situations to improve a later day.

Recognize the person for recognizing you

When you accept recognition well, you reward the person giving support for their extra
effort. This can motivate that individual to do more recognizing. Sometimes, you can do
even more to increase quality recognition. Specifically, you can recognize the person for
recognizing you. In this case, you apply quality recognition principles to reward certain
aspects of the supportive conversation. You might say, for example, you really appreciate
the pinpointing of a certain behavior and the reference to higher-order praise. Such reward-
ing feedback provides direction and motivation for those aspects of the recognition process
that are especially worthwhile and need to become habitual.

Embrace the reciprocity principle

Some people resist receiving recognition because they do not want to feel obligated to give
recognition to others. This is the reciprocity norm at work. If we want to cultivate a Total
Safety Culture, we need to embrace this norm. Research has shown that when you are nice
to others, as when providing them with special praise, you increase the likelihood they will
reciprocate by showing similar behavior (Cialdini, 1993; Geller, 1997). You might not
receive the returned favor, but someone will.
The bottom line is to realize your genuine acceptance of quality recognition will acti-
vate the reciprocity norm, and the more this norm is activated from positive interpersonal
conversation, the greater the frequency of interpersonal recognition. So accept recognition
well and embrace the reciprocity norm. The result will be more interpersonal involvement
consistent with the vision of a Total Safety Culture.

Ask for recognition when deserved but not forthcoming

There is one final strategy I would like to recommend for increasing recognition conversa-
tion throughout a culture. If you feel you deserve recognition, why not ask for it? This
might result in recognition viewed as less genuine than if it were spontaneous, but the out-
come from such a request can be quite beneficial. You might receive some words worth
reliving later for self-motivation. Most important, you will remind the other individual in
a nice way that he or she missed a prime opportunity to offer quality recognition. This
could be a valuable learning experience for that person.
Consider the possible benefit from your statement to another person that you are
pleased with a certain result of your extra effort. With the right tone and effect, such verbal
behavior will not seem like bragging but rather a declaration of personal pride in a small-
win accomplishment. The other person will probably support your personal praise with
individual testimony, and this will bolster your self-motivation. Plus, you will teach the
other person how to support the safety-related behaviors of others.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 286

286 Psychology of safety handbook

Many years ago, I started a self-recognition process among my research students that
increased our awareness of the value of receiving praise, even when we ask for it. I told my
students that during class or group meetings they could request a standing ovation at any
time. All they had to do was specify the behavior they felt deserved recognition and then
ask for a standing ovation. Obviously, such recognition is not private, personal, and one-
on-one, and therefore it is not optimal. Plus, the public aspects of the process inhibited
many personal requests for a standing ovation.
Over the years, however, a number of my students have made a request for a standing
ovation, and the experience has always been positive for everyone. Each request has
included a solid rationale. Some students express pride in an exemplary grade on a proj-
ect. Others acknowledge an acceptance letter from a graduate school, internship, or jour-
nal editor. The actual ovation is fun and feels good, whether on the giving or receiving end.
Plus, we all learn the motivating process of behavior-based recognition, even when it does
not follow all of the quality principles.

Quality safety celebrations

So far I have been talking about individuals recognizing individuals, but what about group
recognition? What about those celebrations where people are being recognized as a group
for reaching some kind of safety milestone? I have been at many of these, and almost all of
them could have been a lot more effective. Let us review the seven principles in Figure
13.15. When group safety celebrations follow these guidelines, they support teamwork and
build a sense of belonging and interdependency. When this happens, people are more will-
ing to actively care for the safety and health of their coworkers.

Do not announce celebrations for injury reduction

Many organizations celebrate when their injury rate reaches a record low. Organizations
often give groups of employees a celebration dinner after a certain number of weeks or
months pass with no recordable injury. Going several months without an injury is certainly
worth celebrating, but you had better be sure “injury-free” was reached fairly. If people
cheat to win by not reporting their injuries, the celebration will not mean much.
If a celebration for record-low injuries is announced as an incentive, motivation to
cheat increases. In other words, when managers promise employees a reward for working

Figure 13.15 Follow seven guidelines to celebrate group achievement of safety mile-
stones.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 287

Chapter thirteen: Intervening with supportive conversation 287

a certain number of days without an injury, they make it tempting to hide a personal injury.
Some workers will feel pressure from peers to avoid reporting an injury if they can get
away with it. This is peer pressure to cheat—a situation that reduces interpersonal trust
and a sense of personal control over workplace injuries.

Celebrate the outcome but focus on the journey

Most of the safety celebrations I have seen give far too little attention to the journey or
processes that enabled the record reduction in injuries. Too often the focus is on the end
result, like having no injuries for a certain period of time, with little discussion about how
this outcome was achieved. It is natural to toast the bottom line, but there is more to be
gained from taking the opportunity to recognize the process.
Valuable direction and motivation can be obtained from pointing out aspects of the
journey which made it possible to reach a safety milestone. Participants learn what they
need to do in order to continue a successful process. Those who performed the behaviors
identified as contributing to the injury prevention receive a boost in self-effectiveness, per-
sonal control, and optimism. They also add information to their intrapersonal dialogues for
later self-motivation. The most important reason for pinpointing journey activities that
lead to injury prevention is it gives credit where credit is owed. Focusing on the process
credits the people and their behaviors that made the difference.

Show top-down support but facilitate bottom-up involvement

Safety celebrations typically start off with speeches by representatives from top manage-
ment. They state their extreme pleasure in the lowered injury rate. Sometimes they display
charts to compare the past with the improved present. Often a manager points out the
amount of money the company saved with the reduction in injury rate. A sincere request
for continuous improvement is made, as well as a promise for a bigger celebration if injury
rates continue to decrease.
Occasionally, a motivational speaker gives an uplifting and entertaining talk. Special
reward placards are often given to individuals or team captains, along with a firm hand-
shake from a top-management official. Along with the steak dinner, participants some-
times receive a certificate and a trinket with a safety slogan. Rarely, however, do the
participants discuss the processes they implemented in order to reach the celebrated
milestone.
In the typical safety celebration, managers give and the operators receive. This cer-
tainly shows impressive top-down support. The ceremony would be more memorable and
supportive, however, if the employees were more participants than recipients. In other
words, the effects of a safety celebration are more beneficial and last longer if line workers
do more talking about their experiences along the journey than listening to managers’
pleasure with the bottom line.

Relive the journey toward injury reduction

By doing more listening than speaking, managers and supervisors enable discussions of
activities that led to the celebrated accomplishment. By reliving the activities that made the
journey successful, people strengthen the internal scripts that support a successful process.
When managers listen to such discussions with genuine interest and gratitude, they
acknowledge the behaviors that led to success. Plus, they empower the employees to con-
tinue their journey toward higher levels of achievement.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 288

288 Psychology of safety handbook

The most effective safety celebration I ever observed featured a series of brief presen-
tations by teams of hourly workers. These employees shared numerous safety ideas they
had put in place to prevent workplace injuries. Some showed off new personal protective
equipment, some discussed their procedures for encouraging near-hit analyses and cor-
rective actions, and others displayed graphs of data obtained from audits of behaviors and
environmental conditions. One team presented an ergonomic analysis and redesign of a
work station.
The after-dinner entertainment was also employee-driven. Representatives from both
management and the workers’ union participated in safety-related skits and in a talent
show. There was no need to hire a band for live music. The workforce of 1200 included a
number of talented musicians. The program was planned and presented by the people
whose daily involvement in various safety processes enabled a celebration of a record-low
injury rate.

Facilitate discussion of successes and failures

The discussions of safety projects at the celebration mentioned previously included both
successes and failures. Work teams not only presented the positive consequences of their
special efforts, they also relived their disappointments, their frustrations, and their dead
ends. They featured the highs and the lows. This made their presentations realistic. It also
made clear the great amount of dedicated work needed to carry out their action plans and
contribute to the celebrated reduction in injury rate.
Pointing out hardships along a journey to success justifies the celebration. It shows that
the celebrated bottom line was not luck. It was accomplished by hard work, interdepend-
ence, win–win collaboration, and synergy. Many people had to go beyond the call of duty
to make a small-win contribution.

Use tangible rewards to establish a memory

When people discuss the difficulties in reaching a milestone, the accomplishment is mean-
ingful. When managers listen to these discussions with sincere interest and appreciation,
the incident becomes even more significant. When a tangible reward is distributed appro-
priately at this occasion, a mechanism is established to support the memory of this experi-
ence and promote its value.
As discussed and illustrated in Chapter 11, the best tangibles include words, perhaps
a worker-designed safety slogan, symbolizing the safety. The tangible should also be some-
thing readily displayable or used in the workplace, from coffee mugs, placards, and pencil
holders to caps, shirts, and umbrellas. Of course, these rewards need to be delivered as only
tokens of appreciation. They were selected “to remind you how you achieved our real
reward—fewer people getting hurt.”
A week after the employee-driven safety celebration I just described, I received a
framed group photograph of everyone who attended the celebration. That picture hangs in
my office today. Every time I look at it, I am reminded of a special time many years ago
when management did more listening than talking in a most memorable safety celebration.

Solicit employee input

When I told my colleagues I was writing about how to celebrate, Josh Williams promptly
responded, “That’s easy, a $100 bottle of cognac, a $6 cigar, and a special friend.” I informed
Josh I was not talking about that kind of celebration. It did occur to me, however, that
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 289

Chapter thirteen: Intervening with supportive conversation 289

everyone has his or her own way of celebrating success. When it comes to group celebra-
tion, we often impose our prejudices on others. We often do not take the time to ask the rel-
evant persons what kind of celebration party they would like.
When asking people how they would like to celebrate, challenge them to think beyond
tangible rewards. A discussion about material rewards puts the celebration in a payoff-for-
behavior mode. This is not the purpose of a safety celebration. How you celebrate a safety
milestone determines whether the occasion is meaningful and memorable or just another
misguided but well-intentioned attempt to show management support.
A safety celebration with top-down support and bottom-up involvement encourages
teamwork and builds a sense of belonging among participants. Therefore, the most effec-
tive safety celebrations are planned by representatives from the work group whose efforts
warrant the celebration. At these celebrations, managers do more listening than talking.
They show genuine approval and appreciation of the challenges addressed and the diffi-
culties handled in achieving the bottom line. The employee-focused discussions of the jour-
ney help write internal scripts for continued self-talk and self-accountability to achieve
more for the next safety celebration.

Choosing the best management conversation

So how can we know what type of safety management conversation to use? This is where
empathy comes into play. Your assessment of situations and people—through observing,
listening, and questioning—will determine which approach to use. Given the dynamic
characteristics of most work settings and the changing nature of people, you need to make
this assessment periodically per situation and worker.
Consider, for example, the new employee who needs specific direction at first. Then, as
he or she becomes familiar with the routine, more support than instruction is called for.
Later, you decide to expand this individual’s work assignment with no increase in finan-
cial compensation. This situation will likely benefit most with a coaching conversation
whereby both direction and support are given, at least at first.
Eventually, a delegating approach might be most appropriate, whereby varying
assignments are given with only outcome expectations. These workers are able to manage
themselves with self-direction and self-motivation, but, as discussed previously, these
individuals still benefit from genuine words of appreciation and gratitude when expecta-
tions are met.
The proactive managers of work teams change their interpersonal conversations quite
dramatically as groups get more familiar with team members and their mission. In the
beginning, during the forming and storming stages of team progress (Tuckman, 1965),
work groups need structure, including specific direction and support. This implies a coach-
ing or directing format. Later, when the group members become familiar with each other’s
interests and talents, and progress to the norming and performing stages of team develop-
ment (Tuckman, 1965), supporting and delegating conversations are needed.

The role of competence and commitment

Figure 13.16 illustrates how two critical characteristics—competence and commitment—
should influence a manager’s conversation approach (adapted from Blanchard, 1999).
When competence is high, people know what to do and, therefore, do not need a directive
conversation. However, they need supportive conversations when their motivation or
commitment is low. This is particularly evident when employees perform irregularly or
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 290

290 Psychology of safety handbook

Figure 13.16 Management conversation is determined by the recipient’s level of compe-

tence and commitment.
inconsistently. Their good days indicate they know what to do, while the occurrence of bad
days suggests a motivational problem.
Causes of low commitment vary dramatically, from interpersonal conflict on the job to
emotional upheaval at home. Such causes can only be discovered through proactive lis-
tening. At times, the diagnosis and subsequent treatment of a motivational problem require
special assistance. In this case, the best a manager can do is recognize a need for profes-
sional help and offer advice and support.
Coaching conversations are needed when a person’s competence and commitment
regarding safety are relatively low. You can improve competence through specific direction
and feedback, and increase commitment by sincerely giving appreciation and support.
Anything that increases a person’s perception of importance or self-worth on the job can
enhance commitment. What makes that happen? It is not always obvious, but if you listen,
observe, and ask questions, you will find out.
Delegating is relevant when people know what to do (competence) and are motivated
to do it. You can often know when an individual or work team advances to this level by
observing successive progress. However, it is often useful to ask people whether they are
ready for this level of conversation. If they say “no,” then ask them what they need to reach
this stage. Do they need more competence through direction or more commitment through
some kind of support the organization could make available?

In conclusion
I hope I have convinced you that the status of safety in your organization is greatly deter-
mined by how safety is talked about—from the managers’ board room to the workers’
break room. Whether we feel responsible for safety and are committed to go for a break-
through depends on our interpretation or mental script about safety conversation.
We often focus our interpersonal and intrapersonal conversations on the past. This
helps us connect with others, but it also feeds our prejudice filters and limits the potential
for conversation to facilitate beneficial change. We enable progress when we move conver-
sations with ourselves and others from past to future possibilities and then to the develop-
ment of an action plan.
Expect people to protect their self-esteem with excuses for their past mistakes.
Listen proactively for barriers to safe behavior reflected in these excuses. Then help the
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 291

Chapter thirteen: Intervening with supportive conversation 291

conversation shift to a discussion of possibilities for improvement and personal commit-

ment to apply a feasible action plan. This is often more likely to occur with a nondirective
than directive approach in which more questions are asked than directives given. It is also
useful to use opening words to protect the listener’s self-esteem and limit the impact of
reactive bias.
Proactive listening enables one to determine whether a coaching, instructing, support-
ing, or delegating conversation is most appropriate. Coaching involves both direction and
support and is needed when a person’s competence and commitment in a particular set-
ting are relatively low. In contrast, delegating is relevant when people know what to do and
are motivated to do it. In this case, they are both competent and committed and can direct
and motivate themselves. Then, delegating conversations provide clear expectations and
show sincere appreciation for worthwhile work.
When people are internally motivated to perform well but do not know how to maxi-
mize their efforts for optimal performance, an instructive conversation is called for. In other
situations, people know what is needed for optimal performance but do not always work
at optimal levels. This reflects an execution problem which cannot be solved with directive
conversation; rather, supportive conversation is needed.
Actually, everyone can benefit from supportive recognition. William James, the first
renowned American psychologist, wrote “the deepest principle in human nature is the
craving to be appreciated” (from Carnegie, 1936, page 19). In Chapter 9, I introduced the
flow of intervention and behavior change model which includes supportive intervention
or behavior-based recognition as critical in helping safety-related behavior become fluent.
Thus, it is extremely important to improve the quality of our interpersonal recognition
conversations and our group celebrations. This chapter presented guidelines for making
this happen.
We also need to increase the quantity of interpersonal support given for safety.
Education and training on how to give behavior-based recognition can certainly help, but
how we receive recognition from others is also critical. Following the guidelines given here
for responding to a supportive conversation can provide both direction to improve the
quality of subsequent recognition and motivation to increase the quantity of supportive
conversations. Receiving recognition well can also activate the reciprocity norm which, in
turn, helps to cultivate a culture of actively caring people working interdependently to
keep each other injury-free.
The next three chapters (Section 5) further address the challenge of increasing actively
caring behavior throughout a work culture. In Chapters 14 and 15, you will learn what psy-
chological research has revealed regarding conditions and individual characteristics that
influence people’s willingness to actively care for the safety and health of others. Then, in
Chapter 16, I discuss ways of integrating behavior-based and person-based psychology to
increase actively caring throughout an organization, community, neighborhood, and fam-
ily. When we teach people the appropriate tools for improving behavior, as presented in
Section 3 of this text; and show them how to increase their willingness to use these tools as
interdependent actively caring intervention agents, we are en route to achieving a Total
Safety Culture.

References
Bandura, A., Self-efficacy mechanism in human agency, Am. Psychol., 37, 747, 1982.
Bandura, A., Self-Efficacy: The Exercise of Control, W. H. Freeman, New York, 1997.
Blair, E. H., How conversation influences safety performance, Update: Newsl. Colon. VA Chap. ASSE,
41(8), 1, 2000.
820049_CRC1_L1540_CH13 11/11/00 2:39 PM Page 292

292 Psychology of safety handbook

Blanchard, K., Building gung ho teams: how to turn people power into profits, workshop presented
at the Hotel Roanoke, Roanoke, VA, November 1999.
Blanchard, K., Zigarmi, P., and Zigarmi, D., Leadership and the One Minute Manager, William
Morrow, New York, 1985.
Carnegie, D., How to Win Friends and Influence People, 1981 ed., Simon & Schuster, New York, 1936.
Chance, P., Learning and Behavior, 4th ed., Wadsworth, Belmont, CA, 1999.
Cialdini, R. B., Influence: Science and Practice, 3rd ed., Harper Collins, New York, 1993.
Covey, S. R., The Seven Habits of Highly Effective People: Restoring the Character Ethic, Simon &
Schuster, New York, 1989.
Deci, E. L. and Ryan, R. M., Intrinsic Motivation and Self-Determination in Human Behavior, Plenum,
New York, 1985.
Drebinger, J. W., Jr., Mastering Safety Communication: Communication Skills for a Safe, Productive and
Profitable Workplace, 2nd ed., Wulamoc Publishing, Galt, CA, 2000.
Geller, E. S., Ten principles for achieving a Total Safety Culture, Prof. Saf., 39(9), 18, 1994.
Geller, E. S., The social dynamics of occupational safety, in Proceedings of the 36th Annual ASSE
Professional Development Conference, New Orleans, American Society of Safety Engineers, Des
Plaines, IL, 1997.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Neenah, WI, 1998a.
Geller, E. S., Building Successful Safety Teams: Together Everyone Achieves More, J. J. Keller &
Associates, Neenah, WI, 1998b.
Geller, E. S., Behavioral safety analysis: a necessary precursor to corrective action, Prof. Saf., 45(3),
29, 2000a.
Geller, E. S., How to sustain involvement in occupational safety: from research-based theory to
real-world practice, in Proceedings of the 2000 ASSE Professional Development Conference and
Exposition, American Society of Safety Engineers, Des Plaines, IL, 2000b.
Geller, E. S. and Clarke, W. S., Safety self-management: a key behavior-based process for injury pre-
vention, Prof. Saf., 44(7), 29, 1999.
Krisco, K. H., Leadership and the Art of Conversation, Prima Publishing, Rocklin, CA, 1997.
Langer, E., Mindfulness, Addison-Wesley, Reading, MA, 1989.
Rogers, C. R., Client-Centered Therapy, Houghton-Mifflin, Boston, 1951.
Ryan, R. M. and Dici, E. L., Self-determination theory and the facilitation of intrinsic motivation,
social development, and well-being, Am. Psych., 55, 68, 2000.
Thorndike, E. L., Animal Intelligence: Experimental Studies, Hafner, New York, 1911.
Thorndike, E. L., Human Learning, MIT Press, Cambridge, MA, 1931.
Tuckman, B. W., Developmental sequence in small groups, Psychol. Bull., 63, 384, 1965.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 293

section five

Actively caring for safety

820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 294
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 295

chapter fourteen

Understanding actively caring

Actively caring is planned and purposeful behavior, directed at environment, person, or behavior
factors. It is reactive or proactive and direct or indirect. Direct, proactive, and behavior-focused
active caring is most challenging, but it is usually most important for large-scale injury prevention.
This chapter discusses conditions and situations that inhibit actively caring behavior. We need to
understand why people resist opportunities to actively care for safety. Then, we can develop inter-
ventions to increase this desired behavior which is critical for achieving a Total Safety Culture.

“We cannot live only for ourselves. A thousand fibers connect us with our fellow men;
and among those fibers, as sympathetic threads, our actions run as causes; and they
come back to us as effects.”—Herman Melville

This quotation from Herman Melville appeared in a popular paperback entitled Random
Acts of Kindness (page 31). Here, the editors of Conari Press (1993) introduced the idea of
randomly showing kindness or generosity toward others for no ulterior motive except to
benefit humanity. This notion seems quite analogous to the actively caring concept I have
discussed earlier in various contexts. Indeed, a recurring theme in this book is that a Total
Safety Culture can only be achieved if people intervene regularly to protect and promote
the safety and health of others. Actively caring, however, is not usually random. It is
planned and purposeful; plus, as implied in Melville’s quote, actively caring behaviors
(actions) are supported by positive consequences (effects). Sometimes, the consequences
are immediate, as when someone expresses their appreciation for an act of caring or they
are delayed, but powerful, as in working with care to develop a safer work setting and
prevent injuries.
Section 4 of this Handbook addresses the need to increase actively caring behavior
throughout a culture and to get the maximum safety and health benefits from this type of
behavioral intervention. Psychologists have identified conditions and individual charac-
teristics (or person states) that influence people’s willingness to actively care for the safety
or health of others. I shall present these and link them to practical things we can do to
increase the occurrence of active caring.
While the concept of “random acts of kindness” is thoughtful, benevolent, and clearly
related to actively caring behavior, I propose a more systematic goal-directed approach
with this concept. I suggest we define actively caring behaviors that give us the “biggest
bang for our buck” in particular situations and, then, manage situations and response–con-
sequence contingencies to increase the frequency of such behaviors.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 296

296 Psychology of safety handbook

In Section 4, I presented techniques that actively caring intervention agents could use
to increase safe behaviors and reduce at-risk behaviors. I propose we practice systematic
and purposeful acts of kindness to keep other people safe and healthy. We clearly need
more of this in our society. Before examining ways to increase actively caring behaviors,
though, it is necessary to define the concept more precisely and objectively.

What is actively caring?

Figure 14.1 presents a simple flow chart summarizing the basic approach to culture change
presented up to this point. We start a culture-change mission with a vision or ultimate pur-
pose—for example, to achieve a Total Safety Culture. With group consensus supporting the
vision, we develop procedures or action plans to accomplish our mission. These are
reflected in process-oriented goals which hopefully activate goal-related behaviors.
It is revealing that many consultants and pop psychologists stop here. As I have indi-
cated earlier, the popular writings of Covey (1989, 1990), Peale (1952), Kohn (1993), and
Deming (1986, 1993) suggest that behavior is activated and maintained by self-affirma-
tions, internal motivation, and personal principles or values. For example, I heard Barker
(1993), the futurist who convinced us to change the dictionary meaning of “paradigm,”
proclaim that “vision alone is only dreaming and behavior alone is only marking time.”
Barker explained, however, that turning vision into goals that specify behaviors will lead
to positive organization change.
Appropriate goal setting, as I described in Chapter 10, self-affirmations, and a positive
attitude can activate behaviors to achieve goals and visions, but we must not forget one of
Skinner’s most important legacies—”selection by consequences” (Skinner, 1981). As
depicted in Figure 14.1, consequences are needed to support the right behaviors and cor-
rect wrong ones. Without support for the “right stuff,” good intentions and initial efforts
fade away. Sometimes, natural consequences are available to motivate desired behaviors,
but often—especially in safety—consequence-contingencies need to be managed to moti-
vate the behavior needed to achieve our goals. As discussed in Chapter 11, we might be

Vision

Goal Setting

Behavior

Consequences

Figure 14.1 A Total Safety Culture requires vision and behavior management.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 297

Chapter fourteen: Understanding actively caring 297

Vision

Goal Setting

Actively Caring

Behavior

Consequences

Figure 14.2 Continuous improvement requires actively caring.

able to eliminate positive consequences that motivate undesired behavior. More often,
however, it is necessary to add some positive consequences for desired behavior.
In Figure 14.2, a new box is added to the basic flow diagram in Figure 14.1. My point
is simple but extremely important. Vision, goals, and consequence-contingencies are not
sufficient for culture change. People need to actively care about goals, action plans, and
consequences. They need to believe in and own the vision. They need to feel obligated to
work toward attaining goals that support the vision, and they need to give rewarding or
corrective feedback to increase behaviors consistent with vision-relevant goals. This is the
key to continuous improvement and to achieving a Total Safety Culture.

Three ways to actively care

The “Safety Triad” (Geller, 1989) introduced in Chapter 3 is useful to categorize actively
caring behaviors. These behaviors can address environment factors, person factors, or
behaviors. When people alter environmental conditions or reorganize or redistribute
resources in an attempt to benefit others, they are actively caring from an environment per-
spective. Actively caring safety behaviors in this category include attending to housekeep-
ing details, posting a warning sign near an environmental hazard, designing a guard for a
machine, locking out the energy source to production equipment, and cleaning up a spill.
Person-based actively caring occurs when we attempt to make other people feel better.
We address their emotions, attitudes, or mood states. Proactively listening to others, inquir-
ing with concern about another person’s difficulties, complimenting an individual’s per-
sonal appearance, and sending a get-well card are examples. This type of active caring is
likely to boost a person’s self-esteem, optimism, or sense of belonging—which, in turn,
increases his propensity to actively care. I discuss this in detail in Chapter 15; also included
are reactive behaviors performed in crisis situations. For example, if you pull someone out
of an equipment pinch point or administer cardiopulmonary resuscitation, you are actively
caring from a person-based perspective.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 298

298 Psychology of safety handbook

Person
Attempting to make
another person feel better.
(e.g., intervening in a crisis,
actively listening, sending
a get-well card)
Environment
Reorganizing or redistributing
resources in an attempt to
benefit others
Safety (e.g., cleaning another's work
area, putting money in
another's parking meter,
donating blood)
Culture

Behavior
Attempting to influence another person's behavior in desired directions.
(e.g., giving rewarding or correcting feedback, demonstrating or
teaching desirable behavior, actively caring coaching

Figure 14.3 Actively caring can target three factors.

From a proactive perspective, behavior-based actively caring is most constructive and

most challenging. This happens when you apply an instructive, supportive, or motiva-
tional intervention to improve another person’s safe behavior. When we teach others
about safe work practices or provide rewarding or corrective feedback regarding observed
behavior, we are actively caring from a behavior focus. Obviously, the one-on-one
coaching process described in Chapter 12 represents behavior-based actively caring.
Giving someone behavior-based recognition in a supportive conversation, as discussed in
Chapter 13, is also actively caring with a behavior focus.
However, when we give feedback on the results of a critical behavior checklist, we
need to consider the feelings of the recipient. We should make more deposits than with-
drawals and actively listen to reactions and suggestions. This is actively caring from a
person perspective. Thus, a good safety coach practices both behavior-focused and person-
focused actively caring.
Figure 14.3 categorizes actively caring behaviors. Obviously, this concept applies to
behaviors outside the safety field. In the fall of 1991, Brown (1991) gave his son, who was
leaving home to begin his freshman year at college, a list of 511 principles to live by. Later
that year, these principles were published in a best seller, entitled Life’s Little Instruction
Book. Then, two years later, Brown (1993) included 517 more tips in a sequel. Figure 14.4
lists several of the life tips Brown gave his son, which I consider actively caring behaviors.
Can you categorize them according to the schema in Figure 14.3? In other words, is each
item environment-focused, person-focused, or behavior-focused?
This might not seem like a straightforward exercise. It might not be clear, for example,
whether an actively caring behavior focuses on a person’s feeling states or behaviors or,
perhaps, both. In some cases, it would be necessary to assess the intentions of the actively
caring agent. The categorizations I recommend for this list are given in Figure 14.5.

Why categorize actively caring behaviors?

So why go to the trouble of categorizing actively caring behaviors? Good question! I think
it is useful to consider what these behaviors are trying to accomplish and to realize the rel-
ative difficulty in performing each of them. Environment-focused active caring might be
easiest for some people because it usually does not involve interpersonal interaction. When
people contribute to a charity, donate blood, or complete an organ donor card, they do not
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 299

Chapter fourteen: Understanding actively caring 299

1. Compliment three people every day.

36. Donate two pints of blood every year.
72. Give to charity in your community and support it
generously with your time and money.
149. Skip one meal a week and give what you would
have spent to a street person.
336. Get your next pet from the animal shelter.
386. Turn off the tap when brushing your teeth.
424. Sign and carry your organ donor card.
475. Don't expect others to listen to your advice and
ignore your example.
511. Call your mother.
561. Say bless you" when you hear someone sneeze.
611. When boarding a bus, say "hello" to the driver. Say
"thank you" when you get off.
612. Write a short note inside the front cover when
giving a book as a gift.
667. Everyone loves praise. Look hard for ways to give
it to them.
769. Don't accept unacceptable behavior.
770. Never put the car in "drive" until all passengers
have buckled up.
802. Leave a quarter where a child can find it.
804. Place a note reading "Your license number has been
reported to the police" on the windshield of a
car illegally parked in a handicapped space.
831. Don't allow children to ride in the back of a pickup
truck.
876. Get your name off mailing lists by writing to: Mail
Preference Service, 11 W. 42nd ST., P.O. Box
3861, New York, NY 10163-3861.
919. Put love notes in your child's lunch box.

Figure 14.4 These items from Brown (1991, 1993) typify actively caring. With permission.

Item Focus Interaction

1 P D
36 E I
72 E I
149 E D
336 E I
386 E I
424 E I
475 B I
511 P D
561 P D
611 P D
612 P I
667 P D
769 B D
770 B D
802 E I
804 B I
831 B D
876 E I
919 P I

Figure 14.5 These items from Figure 14.4 can be categorized with regard to the focus—
environment (E), person (P), or behavior (B), and whether interaction was direct (D) or
indirect (I).

interact personally with the recipient of the contribution. These behaviors are certainly
commendable and may represent significant commitment and effort, but the absence of
personal encounters between giver and receiver warrants consideration separate from
other types of actively caring behavior.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 300

300 Psychology of safety handbook

Certain conditions and personality traits might facilitate or inhibit one type of actively
caring behavior and not the other. For example, communication skills are needed to
actively care on the personal or behavioral level and different skills usually come into play.
Behavior-focused active caring is more direct and usually more intrusive than person-
focused active caring. It is riskier and potentially confrontational to attempt to direct or
motivate another person’s behavior than it is to demonstrate concern, respect, or empathy
for someone.
Helping someone in a crisis situation certainly takes effort and requires special skills, but
there is rarely a possibility of rejection. On the other hand, attempting to correct someone’s
behavior could lead to negative, even hostile, reactions. Actually, effective behavior-based
active caring, as in safety coaching, usually requires both person skills to gain the person’s
trust and behavior-based skills to support desired behavior or correct undesired behavior.
Classifying actively caring behaviors also provides insight into their benefits and lia-
bilities. Both Brown (1991) and the editors of Conari Press (1993) recommend we feed
expired parking meters to keep people from paying excessive fines. Let us consider the
behavioral impact of this environment-focused “random act of kindness” (Conari Press,
1993). What will the vehicle owner think when finding an unexpired parking meter? Could
this lead to a belief that parking meters are unreliable—and further mismanagement
of time? Is there a price to pay in people becoming less responsible about sharing public
parking spaces?
When considering the long-term and large-scale impact of some actively caring strate-
gies, other approaches might come to mind. In the parking meter situation, for example,
the potential impact would be improved by adding some behavior-focused active caring.
Along with feeding the expired parking meter, why not place a note under the vehicle’s
windshield wiper explaining the act? The note might also include some time management
hints. This additional step might not only improve behavior, but set an example. The recip-
ient of the note is probably more inclined to actively care for someone else.
Each type of actively caring behavior can be direct or indirect, with direct behavior
requiring effective communication strategies. For instance, leaving a note to explain an
actively caring act does not involve interpersonal conversation. Similarly, you can report
an individual’s safe or at-risk behavior to a supervisor and eliminate the need for one-to-
one communication skills.
In the same vein, person-focused actively caring does not always involve interpersonal
dialogue. You can send someone a get-well card or leave a friendly uplifting statement on
an answering machine or by e-mail. It is also possible that environment-focused acts will
include personal confrontation, say if you deliver a contribution to a needy individual. This
additional category for actively caring behavior is illustrated in Figure 14.6. You can assess
your understanding by assigning a D (for direct) or I (for indirect) to each item in Figure
14.4. Then compare your answers with those in Figure 14.5.

An illustrative anecdote
Several years ago, I was driving on a toll road in Norfolk, VA, en route to the Fort Eustis
Army Base, where a transportation safety conference was being held. Of my students, three
were with me. Each was scheduled to give a 15-minute talk at the conference. This was to
be their first professional presentation and they appeared quite distressed. Each was pag-
ing frantically through his or her notes making last-minute adjustments.
“Were you this nervous Doc, when you gave your first professional address?” one stu-
dent asked.
“No, I don’t think so,” I replied in jest, “ I was obviously better prepared.”
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 301

Chapter fourteen: Understanding actively caring 301

Figure 14.6 Actively caring is usually most challenging and useful when direct and
behavior-focused.

“Can we just read our paper?” asked another.

“Absolutely not,” I retorted. “Anyone could read your paper. It’s much more profes-
sional and instructive to just talk about your paper informally with the audience.”
Naturally, this conversation just caused more anxiety and distress for my students.
Something had to be done to distract them—to break the tension. As we approached the
first of several tollbooths along the highway, I thought of an actively caring solution. After
paying a quarter for my vehicle, I handed the attendant another quarter and said, “This is
for the vehicle behind us; the driver is using a safety belt and deserves the recognition.” My
students put down their papers and watched the attendant explain to the driver that we
paid her toll because she was buckled up. Because we slowed down to observe this, the
driver caught up with us, pulled next to us in the right lane, and acknowledged our actively
caring behavior with a “shoulder belt salute”—a smile and tug on her shoulder strap.
At the next toll booth, the driver of the vehicle behind us was not buckled up, but that
did not stop me. I gave the attendant an extra quarter and said, “This is for the vehicle
behind us; please ask the driver to buckle up.” Again, we slowed down to watch and, to
our delight, the driver buckled up on the spot. When the vehicle passed us, the driver gave
us a smile and a “thumbs up” sign.
I kept doing this at every tollbooth until exiting the highway, by which time my stu-
dents had almost forgotten about their papers. They seemed relaxed and at ease when
entering the conference room and each gave an excellent presentation. Later, we discussed
how the toll booth intervention actively took their minds off their papers and their anxiety.
Brown (1991) recommended that his son occasionally pay the toll for the vehicle
behind him. This is redistribution of resources. It is also actively caring with an environ-
ment focus. By adding a safety-belt message, I was able to accomplish more than the “ran-
dom act of kindness” suggested by Brown (1991) and the editors of Conari Press (1993).
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 302

302 Psychology of safety handbook

I was able to support and to reward those who were already buckled up and to influence
some drivers to buckle up. In other words, realizing the special value of behavior-based
actively caring enabled me to get more benefit from an environment-focused strategy with
very little extra effort. This behavior-based effort was particularly convenient and effortless
because it was indirect. You can see how the system for categorizing actively caring behav-
ior allows us to compare real and potential acts of kindness and then to consider ways to
increase their impact.

A hierarchy of needs
The hierarchy of needs proposed by the humanist Abraham Maslow (1943, 1954) is proba-
bly the most popular theory of human motivation. It is taught in a variety of college
courses, including introductory classes in psychology, sociology, economics, marketing,
human factors, and systems management. It is considered a stage theory. Categories of
needs are arranged hierarchically and we do not attempt to satisfy needs at one stage or
level until the needs at the lower stages are satisfied.
First, we are motivated to fulfill our physiological needs, which include basic survival
requirements for food, water, shelter, and sleep. After these needs are under control, we are
motivated by safety and security needs—the desire to feel secure and protected from future
dangers. When we prepare for future physiological needs, we are proactively working to
satisfy our need for safety and security.
The next motivational stage includes our social acceptance needs—the need to have
friends and to feel like we belong. When these needs are gratified, our concern focuses on
self-esteem, the desire to develop self-respect, gain the approval of others, and achieve per-
sonal success.
When I ask audiences to tell me the highest level of Maslow’s Hierarchy of Needs, sev-
eral people usually shout “self-actualization.” When I ask for the meaning of “self-actual-
ization,” however, I receive limited or no reaction. This is probably because the concept of
being self-actualized is rather vague and ambiguous. In general terms, we reach a level of
self-actualization when we believe we have become the best we can be, taking the fullest
advantage of our potential as human beings. We are working to reach this level when we
strive to be as productive and creative as possible. Once accomplished, we possess a feel-
ing of brotherhood and affection for all human beings and a desire to help humanity as
members of a single family—the human race (Schultz, 1977). Perhaps, it is fair to say that
these individuals are ready to actively care.
Maslow’s Hierarchy of Needs is illustrated in Figure 14.7, but self-actualization is not
at the top. Maslow (1971) revised his renowned hierarchy shortly before his death in 1970
to put self-transcendence above self-actualization. Transcending the self means going
beyond self-interest and is quite analogous to the actively caring concept. According to
Frankl (1962), for example, self-transcendence includes giving ourselves to a cause or
another person and is the ultimate state of existence for the healthy person. Thus, after sat-
isfying needs for self-preservation, safety and security, acceptance, self-esteem, and self-
actualization, people can be motivated to reach the ultimate state of self-transcendence by
reaching out to help others—to actively care.
It seems intuitive that various self-needs require satisfaction before self-transcendent
or actively caring behavior is likely to occur. Actually, there is little research support for
ranking needs in a hierarchy. In fact, it is possible to think of a number of examples where
individuals have actively cared for others before satisfying all of their own needs.
Mahatma Gandhi is a prime example of a leader who put the concerns of others before his
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 303

Chapter fourteen: Understanding actively caring 303

Self-Transcendency

Self-Actualization

Self-Esteem Needs

Acceptance Needs

Safety and Security Needs

Physiological Needs

Figure 14.7 The highest need in Maslow’s revised hierarchy reflects actively caring.

own. He suffered imprisonment, extensive fasts, and eventually assassination in his 50-
year struggle to help his poor and downtrodden compatriots.
Figure 14.8 includes a story about one of Gandhi’s actively caring behaviors. Notice
how quickly Gandhi reacted in order to leave his second shoe next to the one he acciden-
tally lost from the train. Actively caring was obviously habitual for Gandhi, developed over
a lifetime of active public service. Gandhi focused on the most fundamental of human
responsibilities—our responsibility to treat others as ourselves (Nair, 1995).
I am sure you can think of individuals in your life, including perhaps yourself, who
reached the level of self-transcendence before satisfying needs in the lower stages. I shall
demonstrate in Chapter 15, however, that while satisfying lower level needs might not be

As Ghandi stepped aboard a train one day,

one of his shoes slipped off and landed on
the track. To the amazement of his
companions, Ghandi calmly took off his other
shoe and threw it back along the track to land
close to the first. Asked by a fellow
passenger why he did so, Ghandi smiled.
“The poor man who finds the shoe lying on the
track will now have a pair he can use.”

Figure 14.8 Actively caring was a mindful habit for Mohandas Karamchand Ghandi.
Adapted from Fadiman (1985). With permission.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 304

304 Psychology of safety handbook

necessary for actively caring behavior, people are generally more willing to actively care
after satisfying the lower level needs in Maslow’s hierarchy.
From a behavior-based perspective, you can see that these different need levels simply
define the kinds of consequences that motivate our behavior. When we are at the first stage
of the hierarchy, for example, we are working to achieve consequences—or avoid losing
consequences—necessary to sustain life. We need money to buy food and pay the rent or
mortgage. Then, consequences that imply safety and security are reinforcing. Money is
needed to buy insurance or feed a savings account, for example. At the social acceptance
level, we perform to receive peer support or to avoid negative peer pressure.
Consequences (rewards) that recognize our efforts build our self-esteem and eventu-
ally enable us to be self-actualized. At the highest stages of Maslow’s Hierarchy of Needs—
self-actualization and self-transcendence—we are presumably rewarded by the realization
that we have helped another person. At these levels, we truly believe it is better to give than
to receive. How can we help people get to this motivation level? Let us see how psycholo-
gists have attempted to answer this important question.

The psychology of actively caring

Walking home on March 13, 1964, Catherine (Kitty) Genovese reached her apartment in
Queens, NY, at 3:30 a.m. Suddenly, a man approached with a knife, stabbed her repeatedly
and then raped her. When Kitty screamed, “Oh my God, he stabbed me! Please help me!”
into the early morning stillness, lights went on and windows opened in nearby buildings.
Seeing the lights, the attacker fled, but when he saw no one come to the victim’s aid, he
returned to stab her eight more times and rape her again. The murder and rape lasted more
than 30 minutes and was witnessed by 38 neighbors. One couple pulled up chairs to their
window and turned off the lights so they could get a better view. Only after the murderer
and rapist departed for good did anyone phone the police. When the neighbors were ques-
tioned about their lack of intervention, they could not explain it.
The reporter who first publicized the Kitty Genovese story, and later made it the sub-
ject of a book (Rosenthal, 1964), assumed this bystander apathy was caused by big city life.
He presumed that people’s indifference to their neighbors’ troubles was a conditioned
reflex in crowded cities like New York. After this incident, hundreds of experiments were
conducted by social psychologists in an attempt to determine causes of this so called
“bystander apathy” (Latané and Darley, 1968). This research has actually discredited the
reporter’s common-sense conclusion. Several factors other than big city life contribute to
bystander apathy. Actually, common sense suggests that if more people are present during
a crisis, there is a greater chance that a victim will receive help.
Several years ago, a tragic incident occurred in Detroit, MI, that paralleled the Kitty
Genovese incident and, unfortunately, many others just like it. On Saturday morning,
August 20, 1995, Deletha Word (age 33) leaped off the Detroit River Bridge to escape
Martell Welch (age 19) who had smashed her car with a tire iron after a fender bender.
Dozens of people just watched as Ms. Word was attacked. There were reports that some
spectators actually cheered, presumably encouraging Ms. Word to jump. In an attempt to
save Ms. Word, two men did dive into the river but the victim reportedly resisted their
efforts. City Council President Maryann Mahaffey interpreted this resistance as indicating,
“She was apparently so frightened that she couldn’t trust anyone” (Curley, 1995, page 3A).
An editorial appearing in USA Today (1995) reflects concern for the bystander apathy
in this incident. This editorial is given verbatim in Figure 14.9. It refers to psychological
research to interpret the tragedy. Should the editorial have said more about the relevance
of psychological research? Was the reference to psychological research accurate and
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 305

Chapter fourteen: Understanding actively caring 305

Lack of Heroes
(from USA Today, 1995, p. 10A)
We all complain about crime. Tragically, though, when some of us have a
chance to do something about it, we fail miserably.
Weekend revelers in Detroit had a chance to stop a crime and save a life.
Instead, they apparently gawked. And Deletha Word, 33, working mother, is
dead.
She was pulled from her car by a teenager who tore off most of her
clothes, hit and chased her until she jumped off a bridge to her death in the
Detroit River.
Unfortunately, there were no heroes in that part of Detroit on that tragic
night. No one answered Word’s pleas for help.
Missing were the kinds of bystanders who tackled Francisco Duran after
he shot at the White House last fall. Or Don Lanini, 40, who, in June,
jumped before a Manhattan subway train to rescue a woman.
Two men did dive into the river in a futile attempt to save Word. But
apparently no one acted sooner.
Defenders of the Detroit crowd say maybe the bystanders weren’t sure
what was happening. Maybe they thought someone else would assume
responsibility. Or maybe they were just afraid to interfere with the young
toughs, one weighing about 200 pounds and brandishing a crowbar.
In fact, psychologists say individuals are more likely than crowds to risk
helping in an emergency. Individuals alone tend to act, then think; they
can’t wait for someone else. Crowds tend to inhibit their members.
But that’s no excuse to tolerate violence and inhumanity. Indifference
encourages evildoers. Someone could have rallied the crowd to rush the
assailant. Or yelled and stopped more motorists to help.
Some of the Detroit spectators undoubtedly are tormented by guilt. They
deserve it. Others doubtless distanced themselves from the tragedy and feel
nothing.
Ask yourself. What would you do?

Figure 14.9 Is a sole observer more likely to intervene? Excerpted from USA Today, (1995,
page 10A). With permission.

complete? Could this editorial reduce future bystander apathy? Use your common sense to
answer these questions, then read on for research-based answers.

Lessons from research

Latané, Darley, and their colleagues studied bystander apathy by staging emergency
events observed by varying numbers of individuals. Then, they systematically recorded
the speed at which one or more persons came to the victim’s rescue. In the most controlled
experiments, the observers sat in separate cubicles (as depicted in Figure 14.10) and could
not be influenced by the body language of other subjects. In the first study of this type, the
subjects introduced themselves and discussed problems associated with living in an urban
environment. In each condition, the first individual introduced himself and then casually
mentioned he had epilepsy and that the pressures of city life made him prone to seizures.
During the course of the discussion over the intercom, he became increasingly loud and
incoherent, choking, gasping, and crying out before lapsing into silence. The experimenters
measured how quickly the subjects left their cubicles to help him.
When subjects believed they were the only witness, 85 percent left their cubicles within
three minutes to intervene. However, only 62 percent of the subjects who believed 1 other
witness was present left their cubicle to intervene, and only 31 percent of those who thought
5 other witnesses were available attempted to intervene. Within 3 to 6 minutes after the
seizure began, 100 percent of the lone subjects, 81 percent of the subjects with 1 presumed
witness, and 62 percent of the subjects with 5 other bystanders left their cubicles to intervene.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 306

306 Psychology of safety handbook

Figure 14.10 Subjects in the Latané and Darley experiment could not see each other and
thought they were conversing with one, two, or five other individuals.

The reduced tendency of observers of an emergency to help a victim when they believe
other potential helpers are available has been termed the bystander effect and has been
replicated in several situations (Latané and Nida, 1981). Researchers have systematically
explored reasons for the bystander effect and have identified conditions influencing this
phenomenon. The results most relevant to safety management are reviewed here. Some
suggest ways to prevent bystander apathy—a critical barrier to achieving a Total Safety
Culture. Keep in mind this research only studied reactions in crisis situations, what we
would categorize as reactive, person-focused actively caring. It seems intuitive, though,
that the findings are relevant for both environment-focused and behavior-focused actively
caring in proactive situations.
Many years ago, my students and I (Jenkins et al., 1978) studied the bystander effect in
a situation requiring environment-focused, actively caring behavior. We planted litter (a
small paper bag and sandwich wrappings from a fast-food restaurant) next to a 50-gallon
trash barrel located along a busy sidewalk of our university campus. Then, we watched
people walk by to see if anyone would pick up the litter. Several people used the trash bar-
rel, but only 1 person (a female) of 598 people who walked past the trash barrel, alone or in
groups, stopped to pick up and deposit the litter. The fact that this actively caring person
was alone lends some minuscule support to the bystander effect, but the more remarkable
finding was that almost everyone walked around or over the litter without stopping to per-
form a relatively convenient act of caring. Those who noticed the litter, and several did look
down as they walked, probably assumed someone else would take care of the problem.
They presumed it was someone else’s responsibility. Let us consider this and other factors
affecting our inclination to actively care.
Diffusion of responsibility. Similar to our litter example, a key contributor to the
bystander effect is a presumption that someone else should assume the responsibility. It is
likely, for example, many observers of the Kitty Genovese rape and murder assumed that
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 307

Chapter fourteen: Understanding actively caring 307

another witness would call the police or attempt to scare away the assailant. Perhaps, some
observers waited for a witness more capable than they to rescue Kitty.
Does this factor contribute to lack of intervention for occupational safety? Do people
overlook environmental hazards or at-risk behaviors in the workplace because they pre-
sume someone else will make the correction? Perhaps some people assume, “If the employ-
ees who work in the work area don’t care enough to remove the hazard or correct the risk,
why should I?”
Social psychology research suggests that teaching people about the bystander effect
can make them less likely to fall prey to it themselves (Beaman et al., 1978). Also, elim-
inating a “we–they” attitude or a territorial perspective (“I’m responsible for this area;
you’re responsible for that area”) will increase willingness to look out for others
(Hornstein, 1976).
A helping norm. Many if not most, U.S. citizens are raised to be independent rather
than interdependent. However, intervening for the benefit of others, whether reactively in
a crisis situation or proactively to prevent a crisis, requires sincere belief and commitment
toward interdependence. Social psychologists refer to a “social responsibility norm” as the
belief that people should help those who need help. Subjects who scored high on a meas-
ure of this norm, as a result of upbringing during childhood or special training sessions,
were more likely to intervene in a bystander intervention study, regardless of the number
of other witnesses (Bierhoff et al., 1991).
Some cultures are more interdependent, or collectivistic, than others and promote
social responsibility and group welfare. Chinese and Japanese children, for example, learn
collectivism early on. American and British children are raised to be more individualistic.
Figure 14.11 contrasts the slogans or common phrases repeated in our culture with those
found in the Japanese culture. The difference between an individualistic and collectivistic

Figure 14.11 Expressions reflect socialization and cultural norms.

820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 308

308 Psychology of safety handbook

perspective (Triandis et al., 1990) is clearly shown here and suggests that an interdepend-
ence or helping norm is stronger in Japan than the United States. A survey by Chinese
psychologist Hing-Keung Ma (1985) supported this prediction by showing greater concern
and responsiveness for other people’s problems among residents in Hong Kong vs.
London.
Knowing what to do. When people know what to do in a crisis, they do not fear mak-
ing a fool of themselves and do not wait for another, more appropriate person to intervene.
The bystander effect was eliminated when observers had certain competencies, such as
training in first-aid treatment, which enabled them to take charge of the situation (Shotland
and Heinold, 1985). In other words, when observers believed they had the appropriate
tools to help, bystander apathy was decreased or eliminated.
This conclusion is also relevant for proactive or preventive action, as in safety inter-
vention. When people receive tools to improve safety, and believe the tools will be accepted
and effective to prevent injuries, bystander apathy for safety will decrease. This implies, of
course, the need to promote a social responsibility or interdependence norm throughout
the culture and teach and support specific intervention strategies or tools to prevent work-
place injuries.
It is important to belong. Researchers demonstrated reduced bystander apathy
when observers knew one another and had developed a sense of belonging or mutual
respect from prior interactions (Rutkowski et al., 1983). Most, if not all of the witnesses to
Kitty Genovese’s murder did not know her personally and it is likely the neighbors did not
feel a sense of comradeship or community with one another. Situations and interactions
that reduce a we–they, or territorial perspective, and increase feelings of togetherness or
community will increase the likelihood of people looking out for each other.
Mood states. Several social psychology studies have found that people are more
likely to offer help when they are in a good mood (Carlson et al., 1988). The mood states
that facilitated helping behavior were created very easily—by arranging for potential
helpers to find a dime in a phone booth, giving them a cookie, showing them a comedy
film, or providing pleasant aromas. Are these findings relevant for occupational safety?
Daily events can elevate or depress our moods. Some events are controllable, some
are not. Clearly, the nature of our interactions with others can have a dramatic impact
on the mood of everyone involved. As depicted in Figure 14.12, even a telephone
conversation can lift a person’s spirits and increase his or her propensity to actively care.
Perhaps, remembering the research on mood and its effect will motivate us to adjust our
interpersonal conversations with coworkers (see Chapter 13). We should also interact in a
way that could influence a person’s beliefs or expectations in certain directions, as
explained next.
Beliefs and expectancies. Social psychologists have shown that certain personal
characteristics or beliefs influence one’s inclination to help a person in an emergency.
Specifically, individuals who believe the world is fair and predictable, a world in which
good behavior is rewarded and bad behavior is punished, are more likely to help others in
a crisis (Bierhoff et al., 1991). Also, people with a higher sense of social responsibility and
the general expectancy that people control their own destiny showed greater willingness
to actively care (Schwartz and Clausen, 1970; Staub, 1974).
The beliefs and expectancies that influence helping behaviors are not developed
overnight and obviously cannot be changed overnight. A work culture, however, (includ-
ing its policies, appraisal and recognition procedures, educational opportunities, and
approaches to discipline) can certainly increase or decrease perceptions or beliefs in a just
world, social responsibility, and personal control, and, in turn, influence people’s willing-
ness to actively care for the safety of others (Geller, 1998).
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 309

Chapter fourteen: Understanding actively caring 309

Figure 14.12 Telephone conversations can lift moods and increase one’s propensity to
actively care.

Deciding to actively care

As a result of their seminal research, Latané and Darley (1970) proposed that an observer
makes five sequential decisions before helping a victim. The five decisions (depicted in
Figure 14.13) are influenced by the situation or environmental context in which the emer-
gency occurs, the nature of the emergency, the presence of other bystanders and their reac-
tions, and relevant social norms and rules.
While the model was developed to evaluate intervention in emergency situations—
where there is need for direct, reactive, person-focused actively caring—it is quite relevant
for the other types of actively caring. Actually, the model has been used effectively in a
variety of intervention situations, ranging from preventing a person from driving drunk
(behavior-focused actively caring) to making an environment-focused decision to donate a
kidney to a relative (Borgida et al., 1992; Rabow et al., 1990).

Step 1. Is something wrong?

The first step in deciding whether to intervene is simply noticing that something is wrong.
Some situations or events naturally attract more attention than others. This point relates to
the discussion in Chapter 10 about relative attention and habituation to various activators.
Most emergencies are novel and upset the normal flow of events. However, as shown by
Piliavin et al. (1976), the onset of an emergency such as a person slipping on a spill or falling
down a flight of stairs will attract more attention and helping behavior than the aftermath
of an incident, as when a victim is regaining consciousness or rubbing an ankle after a fall.
Of course, we should expect much less attention to a nonemergency situation.
Context also plays a role here. A significant amount of research has shown, for exam-
ple, that people are more helpful in rural than urban settings (Steblay, 1987), and this dif-
ference may be owing partly to context (Schroeder et al., 1995). The stimulus overload of
the city might lead to people not noticing a need to intervene. Indeed, in active and noisy
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 310

310 Psychology of safety handbook

NOTICE a Need
Is something wrong?

Yes No

INTERPRET as Requiring
Intervention
No
Am I needed?

No
Yes Intervention
ASSUME Personal
Responsibility No
Should I intervene?

Yes No

CHOOSE an Intervention
What should I do?

Yes

PERFORM Actively
Caring Behavior

Figure 14.13 Actively caring requires five sequential decisions.

environments, like various work settings, many people narrow their focus to what is per-
sonally relevant. They learn to tune out irrelevant stimuli. Bickman et al. (1973) used this
stimulus overload theory to explain their finding that university students living in high-
density, high-rise dormitories were less likely to return a lost letter than were students
residing in less densely populated buildings.
Matthews and Canon (1975) tested the stimulus overload theory directly in a real-
world field study. On several trials, a research accomplice wearing a wrist-to-shoulder cast,
dropped several boxes of books a few feet in front of a potential helper. Researchers
observed systematically whether the potential helper intervened. Environmental stimula-
tion was manipulated by running a power lawn mower nearby on half of the trials. In the
noisy condition, only 15 percent of the potential helpers showed actively caring behavior;
however, without the excessive noise, 80 percent of the subjects stopped to help pick up the
dropped boxes.
What is going on here? It is possible the loud noise may have had a negative effect on
the mood of the potential helpers. In fact, mood state may be a critical factor in stimulus
overload studies. Environmental stressors like noise, pollution, and crowding usually have
a negative impact on mood states (Bell et al., 1990), with depressed moods leading to self-
centeredness and lower awareness of another person’s needs.
If stimulus overload can affect people’s attention to an emergency, it can certainly
reduce attention to common everyday situations that are not very obtrusive, but neverthe-
less require actively caring behavior. Consider, for example, the various needs for proactive
behavior that can prevent an injury. Environmental hazards are easy to overlook, especially
in a busy and noisy workplace requiring focused attention on a demanding task. Even less
noticeable and attention-getting are the ongoing safe and at-risk behaviors of people
around us. Yet, these behaviors need proactive support or correction as in the safety coach-
ing approach described in Chapter 12.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 311

Chapter fourteen: Understanding actively caring 311

Now, even if the need for proactive intervention is noticed, actively caring behavior will
not necessarily occur. The observer must interpret the situation as requiring intervention.
Which leads us to the next question that must be answered before deciding to intervene.

Step 2. Am I needed?
As shown in Figure 14.14, people can come up with a variety of excuses for not helping.
Distress cues, such as cries for help, and the actions of other observers can clarify an event
as an emergency. When people are confused, they look to other people for information and
guidance. In other words, through observational learning (Chapter 7), people figure out
how to interpret an ambiguous event and how to react to it. Thus, the behavior of others is
especially important when stimulus cues are not present to clarify a situation as requiring
intervention (Clark and Word, 1972). This was illustrated in one of the early seminal exper-
iments by Latané and Darley (1968).
Professors Latané and Darley invited male students to discuss problems they experi-
enced at a large urban university. While the students were completing a questionnaire,
pungent smoke began puffing through a vent into the testing room. Smoke quickly filled
the room. The danger of the situation was rather ambiguous, however, because real smoke
was not used. The experimenters expected the subjects to rely on others when deciding
what to do.
The social context of the situation varied. Some subjects were alone in the room. Others
filled out the questionnaire with two other subjects who were strangers. Some subjects
were with two accomplices of the researchers who shrugged their shoulders and acted as
if nothing were wrong. Social context had a dramatic impact on whether the subjects left
the room, presumably to save their lives. Of the students who were alone, 75 percent left
the room to report the smoke, but only 10 percent of the subjects with two passive strangers
left the room. In fact, many of the subjects in this circumstance later reported that they

Figure 14.14 People give a variety of excuses for not helping.

820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 312

312 Psychology of safety handbook

believed nothing was wrong. Some concluded the smoke was “truth gas.” Thus, the pas-
sive behavior of others led most subjects to interpret the situation as safe and requiring no
intervention.
Is this relevant to many work situations? How often are environmental hazards or at-
risk behavior overlooked or ignored because the social context—other people—gives the
signal that nothing is wrong?
What about the situation with three naive subjects? Does your common sense tell you
that at least one of these subjects left the room to inquire about the smoke? With three peo-
ple uninformed about the risk, the probability that someone will take action should be
high, right? Wrong! Only 38 percent of the time did anyone leave the room to inquire about
the smoke. Each subject tried to “stay cool.” Thus, when looking around for social cues
each subject saw two other individuals remaining calm, cool, and collected. The group
developed a shared illusion of safety. The investigators labeled this phenomenon pluralis-
tic ignorance.
Follow-up research to pluralistic ignorance caused by mutual passive reaction to
potential dangers has demonstrated the critical value of people’s reactive words in the sit-
uation (Wilson, 1976). We know these words to be verbal activators. Staub (1974), for exam-
ple, varied systematically what his accomplice said after pairs of bystanders, a subject and
the accomplice, heard a crash in an adjoining room and a female’s cry for help. When the
accomplice said, “That sounds like a tape-recording. Maybe they are trying to test us,” only
25 percent of the subjects left the room to help. On the other hand, when the accomplice
reacted with, “That sounds bad. Maybe we should do something,” 100 percent of the sub-
jects intervened.
Thus, in situations where the need for intervention or corrective action is not obvious,
people will seek information from others to understand what is going on and to receive
direction. This is the typical state of affairs when it comes to safety in the workplace. In fact,
the need for proactive actively caring behavior is rarely as obvious as smoke entering a
room or the sound of a crash. If activators like these occurred in the workplace, many peo-
ple would likely react in a hurry. Such events would be noticeable and likely would be
interpreted as needing attention.
Would you assume personal responsibility and respond? Surely the bystanders in the
Kitty Genovese and Deletha Word incidents described earlier noticed the event and inter-
preted it as requiring assistance. Steps 1 and 2 of Latané and Darley’s decision model
were likely satisfied. The breakdown probably occurred at Step 3—perceiving personal
responsibility.

Step 3. Should I intervene?

In this stage you ask yourself, “Is it my responsibility to intervene?” The answer would be
obvious if you were the only witness to a situation you perceive as an emergency. However,
you might not answer “yes” to this question when you know that other people are also
observing the emergency or the safety hazard. In this case, you have reason to believe
someone else will intervene, perhaps a person more capable than you. This perception
relieves you of personal responsibility, but what happens when everyone believes the other
guy will take care of it? This is likely what happened in the Kitty Genovese and Deletha
Word incidents and many other tragedies just like these.
A breakdown at this stage of the decision model does not mean the observers do not
care about the welfare of the victim. Actually, it is probably incorrect to call lack of inter-
vention “bystander apathy” (Schroeder et al., 1995). The bystanders might care very much
about the victim but defer responsibility to others because they believe other observers are
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 313

Chapter fourteen: Understanding actively caring 313

more likely or better qualified to intervene. Similarly, people might care a great deal about
the safety and health of their coworkers, but they might not feel capable of acting on their
caring. People might resist taking personal responsibility to actively care because they do
not believe they have the most effective tools to make a difference. This can be remedied by
teaching employees the principles and procedures presented in Section 4 of this Handbook.
In addition to having a “can do” belief, people need to believe it is their personal respon-
sibility to intervene. In many work situations, it is easy to assume that safety is someone
else’s responsibility—the safety director or a team safety captain. After all, these individu-
als have “safety” in their job titles, and they meet regularly to discuss safety issues. They get
to go off site now and then to attend a safety conference—where they learn the techniques
that make them the most capable to intervene. Therefore, it is their responsibility.
Psychologists have shown that people will take responsibility, even among strangers,
if their responsibility is clearly specified (Baumeister et al., 1988). In an interesting field
study, for example, researchers staged a theft on a public beach and then observed whether
assigning responsibility to some individuals increased their frequency of intervention
(Moriarty, 1975). Researchers posing as vacationers randomly asked individual sunbathers
to watch their possessions, including a radio, while they went for a walk on the beach. In
the control condition, the researchers only asked sunbathers for a match and then left for a
walk. A short time later, a second researcher approached the unoccupied towel, snatched
the radio, and ran down the beach.
How often did the individual sunbather intervene? Surprisingly, 94 percent of the sun-
bathers assigned the “watchdog” responsibility intervened, often with dramatic and phys-
ical displays of aggressive protection. In contrast, only 20 percent of the bystanders in the
control condition reacted in an attempt to retrieve the radio. Perhaps, your common sense
predicted the correct answer this time, and that is why you have asked strangers in public
places to watch your possessions for a short period of time.
The challenge in achieving a Total Safety Culture is to convince everyone they have a
responsibility to intervene for safety. Indeed, a social norm or expectancy must be estab-
lished that everyone shares equally in the responsibility to keep everyone safe and healthy.
Furthermore, safety leaders or captains need to accept the special responsibility of teach-
ing others any techniques they learn at conferences or group meetings that could increase
a person’s perceived competence to intervene effectively. All this is easier said than
done, of course. Unfortunately, if we do not meet this challenge, many people are apt to
decide that actively caring safety intervention is not for them. They could feel this way
even after viewing an obvious at-risk behavior or condition that would benefit from their
immediate action.

Steps 4 and 5. What should I do?

These last two steps of Latané and Darley’s decision model point out the importance
of education and training. Education gives people the rationale and principles behind
a particular intervention approach. It gives people information to design or refine inter-
vention strategies, leading to a sense of ownership for the particular tools they help to
develop. Through training, people learn how to translate principles and rules into specific
behaviors or intervention strategies. As I discussed in Chapter 12, for example, safety
coaching training should include role-playing exercises so people practice certain commu-
nication techniques and receive specific feedback regarding their strengths and weak-
nesses. The bottom line here is that people who have learned how to intervene effectively
through relevant education and training are likely to be successful agents of actively-
caring intervention.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 314

314 Psychology of safety handbook

Research by Shotland and Heinold (1985) showed that bystanders without first-aid
training were just as likely to intervene for a victim with obvious arterial bleeding as were
bystanders who previously received first-aid training. The choice and execution of an inter-
vention, however, varied significantly depending on prior training. Those with training
intervened with much greater competence, with some untrained helpers doing more harm
than good. Similarly, Clark and Word (1974) demonstrated that people without proper
information regarding electricity would sometimes impulsively touch a victim who was
holding a “live” electrical wire, jeopardizing their own lives.

Summary of the decision framework

In this section, I have reviewed the decision process model Latané and Darley proposed as
a sequence of choices people make before actively caring on behalf of another person.
Although developed to understand the bystander effect in emergency situations, this deci-
sion framework is certainly relevant for proactive situations and for each type of actively
caring behavior defined in this chapter (direct vs. indirect, and environment-focused,
person-focused or behavior-focused).
The model can help us understand why an individual might not actively care for
another person’s health and safety, and it can be used to guide the development of strate-
gies to increase the frequency of actively caring behaviors. For example, conditions (acti-
vators) will increase the probability of this behavior if they increase the likelihood a person
will notice and perceive a need for intervention and assume personal responsibility for
helping. Moreover, education and training sessions that increase skills and self-confidence
to actively care effectively will increase the amount of constructive actively caring behav-
ior occurring throughout a culture.
More strategies for increasing these behaviors are entertained in Chapter 16. Before
turning to a discussion of ways to increase actively caring, however, we need to consider
another approach to interpreting bystander intervention, or the lack of it. Because behav-
ior is motivated by consequences, a person’s decision to actively care can be analyzed
according to the perceived positive vs. negative consequences one expects to receive. If
people are motivated to maximize positive consequences and minimize negative conse-
quences, actively caring behavior will only occur if perceived rewards outweigh perceived
costs. This framework suggests strategies for increasing actively caring not prompted by
Latané and Darley’s sequential decision model.

A consequence analysis of actively caring

When I related the Kitty Genovese and Deletha Word incidents to my family and asked
their opinions, I received a unanimous reaction that I could not readily explain with the
decision model discussed previously. My wife and two daughters proclaimed that most
observers did not help these women because they feared for their own safety. The perpe-
trator was armed with a knife in the 1964 incident and a tire iron in the 1995 tragedy. Each
was obviously dangerous. The onlookers could certainly see there was an emergency
requiring specific assistance from anyone who would take responsibility.
According to an interpretation based on our understanding of the power of conse-
quences, people resisted taking responsibility because they perceived that it could mean
more trouble—or potential harm—than it was worth. It was safer to assume that someone
else more capable would intervene. According to this consequence model, people hesitated
to intervene because they perceived more potential costs than benefits, not because they
were apathetic or failed to interpret a need to take personal responsibility.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 315

Chapter fourteen: Understanding actively caring 315

Piliavin and colleagues (1969, 1981) have developed a cost-reward model to interpret
people’s propensity to help others in various emergency situations. There are two basic cat-
egories of potential negative consequences for helping that include personal costs, includ-
ing effort, inconvenience, potential injury, and embarrassment; and costs to the victim if no
intervention occurs. This latter category includes two subcategories: the personal costs of
not helping, including criticism, guilt, or shame; and empathic costs from internalizing the
victim’s distress and physical needs. The authors combine these negative consequences for
direct intervention and for not intervening in order to predict whether actively caring
behavior will occur under certain circumstances.
The matrix in Figure 14.15 combines two levels of cost (low vs. high) to the potential
intervention agent and the victim in order to predict when actively caring behavior will
occur. It is most likely (lower left cell of Figure 14.15) when costs for helping are low, for
example, convenient and not dangerous; and costs to the victim for not helping are high,
as when the victim is seriously injured. On the other hand, intervention is least likely when
the perceived personal costs for intervening are high, for example, effortful and risky; and
the apparent costs to the victim for no intervention are low, as when an experienced worker
is performing at-risk behavior with no negative consequences.
The Genovese and Word incidents fit the lower right cell of Figure 14.15—high per-
ceived cost for both helper and victim. Although the costs for not helping these individu-
als were extremely high, resulting ultimately in their deaths, the costs for helping were also
high, in fact, potentially fatal. This means significant conflict for the person deciding what
to do. The conflict can be resolved by helping indirectly, say by telephoning police or an
ambulance; or by reinterpreting the situation (Schroeder et al., 1995). This can be done by
presuming someone else will intervene—diffusion of responsibility—or perhaps by ration-
alizing that the person does not deserve help.
A bystander might rationalize, for example, that Genovese should not have been walk-
ing the streets in that neighborhood at 3:30 in the morning and Word brought on the attack
by crashing into the assailant’s vehicle. Rationalization reduces the perceived costs for not
intervening and enables the bystander to ignore the situation without excessive shame or
guilt. According to this cost-reward interpretation, when bystanders perceive high costs
both for intervening and for not intervening in a crisis, they recognize the need for action,
hesitate because of perceived personal costs, and then search for an excuse to do nothing
(Schroeder et al., 1995; Schwartz and Howard, 1981).
The upper left quadrant of Figure 14.15 represents situations most analogous to
actively caring for injury prevention. Although a simple low-cost intervention might be

Figure 14.15 Costs to bystanders for intervening and costs to a victim for not intervening
determine the probability of intervention. Adapted from Piliavin et al. (1981). With per-
mission.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 316

316 Psychology of safety handbook

called for to correct an environmental hazard or an at-risk behavior, there is no immediate

emergency and, thus, no need for immediate action. There is low perceived cost if no action
is taken: “We’ve been working under these conditions for months and no one has been
hurt.” Piliavin et al. (1981) presume that intervention in situations represented by this cost
quadrant is most difficult to predict. Many factors can influence perceived consequences
that are positive and negative, and small changes in these factors can tilt the cost-reward
balance in favor of stepping in or standing back.
Through testimonials and constructive discussions, employees can be convinced that
the potential cost of not intervening is higher than they initially thought. This can occur, for
example, by considering the large degree of plant-wide exposure to a certain uncorrected
hazard. Also, it might be worthwhile to remind people of the large-scale detrimental learn-
ing that could occur from the continuous performance of risky behavior. Furthermore, edu-
cation and role-playing exercises can reduce the perceived personal costs of actively caring.
It is also true that personal factors, such as mood states discussed earlier, determine
whether intervention occurs.
Figure 14.16 illustrates the cost-reward approach of a rational potential helper, as
described by Wrightsman and Deaux (1981, page 261). Notice that the potential helper in
the story is considering the rewards as well as the costs for intervening and for not inter-
vening. Although the matrix in Figure 14.15 (from Piliavin et al., 1981) focused entirely on
negative consequences, it is important to consider that positive consequences can also play
a prominent role in determining one’s decision to get involved. In occupational safety, for
example, proactive actively caring behavior can not only prevent a serious injury, but it also
can set the right example for others to follow. It can also increase certain positive personal
states in both the doer and beneficiary of the act which, in turn, increases the probability of
actively caring behavior by both in the future. These positive outcomes from giving and
receiving are detailed in the next chapter.

Figure 14.16 Does actively caring depend on a rational cost-reward analysis? Adapted
from Wrightsman and Deaux (1981, page 261). With permission.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 317

Chapter fourteen: Understanding actively caring 317

The power of context

The influence of context in determining whether we actually care for another person’s
safety cannot be overemphasized. Context actually can influence each step of the Latané
and Darley (1970) decision model described previously and summarized in Figure 14.13.
The context in which behavior occurs can affect one’s evaluation of the costs and benefits
of helping vs. not helping a victim. In other words, the perceived consequences of actively
caring depend to a significant extent on the environmental and social context in which the
relevant behaviors occur. Let us look more closely at this context variable, and consider its
impact on safety-related behavior.
According to my copy of The American Heritage Dictionary (1991), context refers to “the
circumstances in which a particular event occurs” (page 316). It includes both the outside
and inside stuff surrounding people when they are performing. This refers to what we see
others doing on the outside and how we feel on the inside— from feelings of competence,
confidence, and commitment to perceptions of insecurity, uncertainty, and risk.
Figure 14.17 is worth more than one thousand words to describe context. Have you
seen a mild mannered and polite person turn into an impatient and hostile creature after
getting behind the wheel of an automobile? The environmental and competitive context of
driving interacts with certain personality characteristics to produce “Mr. Hyde” on the
road. Then, we have a nationwide epidemic of “road rage.” Incidentally, our research
attempts to identify those individuals most prone to demonstrate road rage have shown
that almost anyone can experience the negative emotions reflected in road rage, given the
“right” context (DePasquale et al., in press).

Experiencing context
I use a simple demonstration to teach the influence of context in my course in introductory
psychology. I ask volunteers to simultaneously stick one hand in a bucket of ice water and
the other in a bucket of hot water (around 100°F). After about 10 seconds, I ask the volun-
teers to remove their hands from the two buckets and put both hands in a third bucket

Figure 14.17 In the context of driving, many individuals transition from mild-mannered
to rude, hostile, and impolite.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 318

318 Psychology of safety handbook

filled with water at room temperature (about 70°F). However, the volunteers do not expe-
rience room temperature. In fact, one hand feels quite warm, while the other hand per-
ceives a rather cool temperature.
You do not have to be there to appreciate how the prior brief temperature exposure
influenced subsequent perception. In fact, you have probably already guessed which hand
experienced warm water and which hand experienced cold water. We live this simple con-
text effect everyday. Coming indoors from the cold gives the impression of warmth but, in
contrast to a hot summer day, the same indoor temperature can appear quite cool. Yet expe-
riencing “warm” in one hand and “cool” in the other while soaking in the same bucket of
water brings expressions of surprise to my students.

An illustrative anecdote
On a ski weekend in Snowshoe, WV, a few years ago, I was reminded of the dramatic influ-
ence context has on human behavior. First, I need to explain that this was only the third
time in my life I had ever tried to ski, and the first time was in 1974. Furthermore, the hills
were quite icy, and the so-called beginner hills at Snowshoe appeared quite steep to me. I
did not see a “bunny hill” anywhere, but my daughter urged me on. So even with low com-
petence and confidence and perceptions of uncertainly and high risk, I took to the crowded
slopes. One near hit after another did not stop me, nor did one “wipe out” after another.
My numerous bruises qualified for several OSHA recordables. My only consolation
was that I was not the only one in pain. The next day, many guests at the Silver Creek Lodge
were limping around; some were sitting with legs wrapped and elevated—more OSHA
recordables. Most other skiers in my age range were much more experienced than I, and
several told me they were having a difficult time because of the icy conditions. Their admo-
nitions were not sufficient for me to ignore my daughter’s urgings, “Come on, dad, just one
more hill; you can do it.” I was also influenced by the “big bucks” I had paid for this ski
weekend. I wanted us to get my money’s worth.
Taking risks on the lifts. The risky behavior of the slopes generalized to the ski lifts.
Herein lies the real context lesson of my story. The lift chairs had protection bars that could
be pulled down conveniently. The signs requesting the use of these restraining bars hardly
seemed necessary; the need for this protective device was obvious. The lifts rose to heights
over 200 feet above the ground. It would not take much for someone to slip off the seat,
especially given the slick material of most ski pants. When the lift stopped, the chairs
rocked forward and backward slightly, making the need for this protective device even
more evident. Here is the kicker— the bottom line. More often than not, I observed the bars
in the upright position. Most skiers were not using this protective device. Did the risky con-
text of the skiing experience influence decreased use of this protective device?
At every lift, a “courtesy patrol” person guided lines of people to the entrance, and
another individual helped people take their seats. There was ample opportunity for these
“professionals” to remind skiers to use the protection device, but I never heard such a
reminder. The many long lines I stood in that weekend gave me numerous opportunities
to hear such a safety message. In fact, I learned later that my daughter’s friends rode the
lift several times at first without pulling down the protection bar because they did not real-
ize it was there.
I noticed the protective device and used it everytime— well almost everytime. I must
confess that once my daughter and I rode a lift with two young men who appeared to be
expert skiers. This time I did not pull down the bar, at least not at first. Instead, I waited for
one of “the experts” to take control. Within the context of my insecurity and reduced self-
confidence, I waited for someone else to intervene. Only when our chair stopped and
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 319

Chapter fourteen: Understanding actively caring 319

rocked a bit, about 100 feet above the ground, did I reach up to pull down the protection
bar. There I was, a researcher and educator who has studied and lectured about safety for
over 25 years and I hesitated to protect my daughter, myself, and two strangers.
Context can inhibit actively caring. Context is my only excuse for my lack of actively
caring behavior. Not only did the use of this protective bar seem insignificant within the
context of the greater perceived risk of skiing, but I hesitated to take control within the con-
text of two experienced skiers. I might add that the two experts seemed quite perturbed at
my protective behavior. They both grimaced slightly, with one having to move his ski poles
to make room for the protective bar. Long before we reached the end of our ride, one expert
raised the protective bar, presumably preparing to dismount.
There were other examples that weekend of how my behavior was shaped by the con-
text of what was going on around me. I think you can see how this story relates to safety
and actively caring in the workplace.
A ski resort is a mini-culture, with its own set of rules, norms, behavioral patterns,
and attitudes. The environmental and social context at this busy ski resort was not
conducive to actively caring for safety. The overriding purpose or mission of the resort is
to give people the exhilarating experience of gliding down snow-covered hills of varying
steepness. Nowhere in the resort’s mission statement was there a message about safety.
Actually, for some people, an attempt to link safety with skiing would seem inconsistent.
After all, skiers pay big bucks to take extraordinary risks. Why should we look out for their
personal safety?

Context at work
Does the mission statement of your industry reflect an overarching concern for production
and quality? Is safety considered a priority (instead of a value) that gets shifted when pro-
duction quotas are emphasized? Is safety viewed as a top-down condition of employment
rather than an employee-driven process supported by management? Are safety programs
handed down to employees with directives to “implement per instructions” rather than
“customize for your work area”?
Are safety initiatives discussed as short-term “flavor-of-the-month” programs rather
than an ongoing process that needs to be continuously improved to remain evergreen? Are
near-hit and injury “investigations” perceived as fault-finding searches for a single cause
rather than fact-finding opportunities to learn what else can be done to reduce the proba-
bility of personal injury? Are the elements of a safety initiative considered piecemeal fac-
tors independent of other organizational functions rather than aspects of an organizational
system of interdependent functions?
Are employees held accountable for outcome numbers that hold little direction for
proactive change and personal control rather than process numbers that are diagnostic
regarding achieving an injury-free workplace? Do employees take a dependency stance
toward industrial safety whereby they depend on the organization to protect them with
rules, regulations, engineering safeguards, and personal protective equipment?
A “yes” answer to any of these questions implies contextual barriers that need to be
overcome in order to achieve the ultimate injury-free workplace. A “no” answer to all of
these questions is symptomatic of a work context that encourages people to actively care
for the health and safety of others. In this kind of work culture, it is not sufficient to rely on
the organization’s safe operating procedures or even on personal responsibility and self-
discipline but on interpersonal teamwork and a shared interdependent responsibility to
protect each other. In this work context, actively caring can be cultivated and a Total Safety
Culture achieved.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 320

320 Psychology of safety handbook

Summary of contextual influence

Here, I reflected on a personal experience at a ski resort to illustrate the critical impact of
environmental and social context (or culture) on individual health and safety. I hope it is
clear that the context in which we perform can have a dramatic effect on our behavior and
attitude. A key part of this influential context is the behavior and attitude of other people.
Think back to my daughter urging me on or the savvy skiers I shared a lift with who dis-
dained using the restraining bar.
Some organizational cultures inhibit the kinds of behavior needed to reduce industrial
injuries. Getting employees involved in safety is difficult within the context of top-down
rules, regulations, and programs supported almost exclusively with the threat of negative
consequences. In contrast, employee involvement is much more likely with top-down sup-
port of safety processes developed, owned, and continuously improved upon by work
teams educated to understand relevant rationale and principles.
Metrics used to evaluate the safety performance of individuals, teams, and the organ-
ization as a whole have a powerful influence on context. Employee commitment, owner-
ship, and involvement can increase or decrease depending on the evaluations employed.
Injury statistics provide an overall estimate of the distance from a vision of “injury free,”
but they are not a diagnostic tool for proactive planning. If used as the only index of safety
achievement (or failure), injury-related outcome numbers can do more harm than good,
alienating people rather than empowering them to actively care for safety.
On the other hand, numbers that measure the quantity and quality of process activities
related to safety performance provide the context needed to motivate individual and team
responsibility. They direct continuous improvement of the process. Chapter 19 in Section 6
of this Handbook presents more details on developing a process-based evaluation system for
continuously improving safety. The following chapters in Section 5 recommend a
variety of additional strategies for cultivating a work culture that promotes actively caring
behavior.

In conclusion
Actively caring behavior is planned and purposeful. It can be direct or indirect and its focus
is environment, person, or behavior. Actively caring that addresses the environment is usu-
ally easiest to perform because it does not involve interpersonal confrontation. Behavior-
focused actively caring is often most proactive but is most difficult to carry out effectively
because it attempts to influence another person’s behavior in a nonemergency situation.
Practically all of the research related to this concept has studied crisis situations in which a
victim needs immediate assistance. This is essentially person-focused and reactive caring.
Psychologists have determined factors that influence the probability of actively caring
behavior in emergencies, and the results are relevant for both environment-focused and
behavior-focused actively caring. Understanding the conditions that lead to an increase
or decrease in reactive caring behavior can help us find ways to facilitate proactive caring
for safety.
The finding that people often refuse to act in a crisis, especially when they can share
the responsibility of intervening with others, is quite analogous to most work settings.
Hence, it is important to understand the factors that can influence this resistance, referred
to as “bystander apathy.” For example, people with a sense of social responsibility and
comradeship for others at work, and who believe they have personal control in a just
world, are more apt to intervene for the safety of others. It is possible to increase these per-
sonal characteristics among people through policy, procedures, and personal interaction.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 321

Chapter fourteen: Understanding actively caring 321

Increasing these states, and thus the willingness to actively care for safety, is key to achiev-
ing a Total Safety Culture and is addressed in the next two chapters.
A decision model developed by Latané and Darley helps us understand why we do not
see more actively caring behavior. Before we step in, either reactively or proactively, we
presumably make five sequential decisions.

1. Is something wrong?
2. Is my help needed?
3. Is it my responsibility to intervene?
4. What kind of intervention strategy should I use?
5. Exactly when and how should I intervene?

This decision logic suggests certain methods for increasing the likelihood that people
will get involved. For example, the model shows the importance of teaching employees
how to recognize and correct environmental hazards and at-risk behaviors. It is also criti-
cal to promote the ultimate aim or corporate mission to make safety a value. For this to hap-
pen, everyone must assume responsibility for safety and never wait for someone else to act.
A consequence, or cost –benefit model, offers more guidance for increasing actively car-
ing behavior. It enables us to analyze motivational factors that shape decisions to actively
care. Conditions and situations that increase perceptions of costs to victims (for not inter-
vening) and reduce perceptions of personal costs to the intervention agent (for intervening)
increase the probability of action being taken. In addition, it is important to help people
realize the potential positive consequences or rewards available to both the giver and
receiver of an actively caring intervention. When these perceived internal and external
rewards outweigh the rewards for doing nothing, people will probably actively care.
Most safety situations involve relatively low costs and rewards to both the recipient
and deliverer of the intervention. Although the relative costs to an individual for interven-
ing may be low, the recipient of a proactive safety intervention is only a potential victim, so
the perceived cost for not intervening is also low. Education and training can reduce these
perceived costs to the intervention agent and increase the perceived costs to potential
victims. The result is more frequent actively caring behavior for safety. However, educa-
tion and training are not sufficient to achieve the amount of actively caring needed for
a Total Safety Culture. The next two chapters deal more specifically with how to develop
and implement strategies to increase actively caring behavior for occupational and com-
munity safety.

References
American Heritage Dictionary, 2nd College ed., Houghton Mifflin, New York, 1991.
Barker, J. A., The Power of Vision videotape, Chart House International Learning Corp., New York,
1993.
Baumeister, R. F., Chesner, S. P., Sanders, P. S., and Tice, D. M., Who’s in charge here? Group leaders
do lend help in emergencies, Personal. Soc. Psychol. Bull., 14, 17, 1998.
Beaman, A. I., Barnes, P. J., Klentz, B., and McQuirk, B., Increasing helping rates through informa-
tional dissemination: teaching pays, Personal. Soc. Psychol., 37, 1835, 1978.
Bell, P. A., Fisher, J. D., Baum, A., and Greene, T., Environmental Psychology, 3rd ed., Holt, Rinehart,
& Winston, Fort Worth, TX, 1990.
Bickman, L., Teger, A., Gabriele, T., McLaughlin, C., Berger, M., and Sunaday, E., Dormitory density
and helping behavior, Environ. Behav., 5, 465, 1973.
Bierhoff, H. W., Klein, R., and Kramp, P., Evidence for the altruistic personality from data on acci-
dent research, J. Personal., 59, 263, 1991.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 322

322 Psychology of safety handbook

Borgida, E., Conner, C., and Manteufel, L., Understanding living kidney donation: a behavioral
decision-making perspective, in Helping and Being Helped, Spacapan, S. and Oskamp, S., Eds.,
Sage, Newbury Park, CA, 1992.
Brown, H. J., Jr., Life’s Little Instruction Book, Rutledge Hill Press, Nashville, TN, 1991.
Brown, H. J., Jr., Life’s Little Instruction Book, Vol. II, Rutledge Hill Press, Nashville, TN, 1993.
Carlson, M., Charlin, V., and Miller, N., Positive mood and helping behavior: a test of six hypothe-
ses, J. Personal. Soc. Psychol., 55, 211, 1988.
Clark, R. D., III and Word, L. E., Why don’t bystanders help? Because of ambiguity?, J. Personal. Soc.
Psychol., 24, 392, 1972.
Clark, R. D., III and Word, L. E., Where is the apathetic bystander? Situational characteristics of the
emergency, J. Personal. Soc. Psychol., 29, 279, 1974.
Covey, S. R., The Seven Habits of Highly Effective People, Simon & Schuster, New York, 1989.
Covey, S. R., Principle-Centered Leadership, Simon & Schuster, New York, 1990.
Curley, T., Anger, disbelief swirling after Detroit attack, USA Today, Wednesday, August 23, 1989, 3A.
Deming, W. E., Out of the Crisis, Massachusetts Institute of Technology, Center for Advanced
Engineering Study, Cambridge, MA, 1986.
Deming, W. E., The New Economics for Industry, Government, Education, Massachusetts Institute of
Technology, Center for Advanced Engineering Study, Cambridge, MA, 1993.
DePasquale, J. P., Geller, E. S., Clarke, S. W., and Littleton, L. C., Measuring road rage: development
of the propensity for angry driving scale, J. Saf. Res., in press.
Editors of Conari Press, Random Acts of Kindness, Conari Press, Emeryville, CA, 1993.
Fadiman, C., Ed., The Little Brown Book of Anecdotes, Little, Brown, Boston, 1985.
Frankl, V., Man’s Search for Meaning: an Introduction to Logotherapy, Beacon Press, Boston, 1962.
Geller, E. S., Managing occupational safety in the auto industry, J. Organ. Beh. Manage., 10(1), 181,
1989.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998.
Hornstein, H. A., Cruelty and Kindness: a New Look at Aggression and Altruism, Prentice-Hall,
Englewood Cliffs, NJ, 1976.
Jenkins, E., Cuddiky, K., Hearn, K., and Geller, E. S., When will people pick up and pitch in?, paper
presented at the Virginia Academy of Science meeting, Blacksburg, VA, April 1978.
Kohn, A., Punished by Rewards: the Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes,
Houghton Mifflin, Boston, 1993.
Latané, B. and Darley, J. M., Group inhibition of bystander intervention, J. Personal. Soc. Psychol., 10,
215, 1968.
Latané, B. and Darley, J. M., The Unresponsible Bystander: Why Doesn’t He Help?, Appleton-Century-
Crofts, New York, 1970.
Latané, B. and Nida, S., Ten years of research on group size and helping, Psychol. Bull., 89, 308, 1981.
Ma, H., Cross-cultural study of altruism, Psychol. Rep., 57, 337, 1985.
Maslow, A. H., A theory of human motivation, Psychol. Rev., 50, 370, 1943.
Maslow, A. H., Motivation and Personality, Harper, New York, 1954.
Maslow, A. H., The Farther Reaches of Human Nature, Viking, New York, 1971.
Mathews, K. E. and Canon, L. K., Environmental noise level as a determinant of helping behavior,
J. Personal. Soc. Psychol., 32, 571, 1975.
Moriarty, T., Crime, commitment, and the responsive bystander: two field experiments, J. Personal.
Soc. Psychol., 31, 370, 1975.
Nair, K., A clue from Gandhi, Sky, May, 26–31, 1995.
Peale, N. V., The Power of Positive Thinking, Prentice-Hall, New York, 1952.
Piliavin, I. M., Rodin, J., and Piliavin, J. A., Good samaritanism: an underground phenomenon?,
J. Personal. Soc. Psychol., 13, 289, 1969.
Piliavin, J. A., Piliavin, I. M., and Broll, L., Time of arousal at an emergency and likelihood of help-
ing, Personal. Soc. Psychol. Bull., 2, 273, 1976.
Piliavin, J. A., Dovidio, J. F., Gaertner, S. L., and Clark, R. D., III, Emergency Intervention, Academic
Press, New York, 1981.
820049_CRC1_L1540_CH14 11/10/00 2:03 PM Page 323

Chapter fourteen: Understanding actively caring 323

Rabow, J., Newcomb, M. D., and Hernandez, A. C. R., Altruism in drunk driving situations:
personal and situational factors in helping, Soc. Psychol. Q., 53, 199, 1990.
Rosenthal, A. M., Thirty-Eight Witnesses, McGraw-Hill, New York, 1964.
Rutkowski, G. K., Gruder, C. L., and Romer, D., Group cohesiveness, social norms, and bystander
intervention, J. Personal. Soc. Psychol., 44, 545, 1983.
Schroeder, D. A., Penner, L. A., Dovidio, J. F., and Piliavin, J. A., The Psychology of Helping and
Altruism, McGraw-Hill, New York, 1995.
Schwartz, S. H. and Clausen, G. T., Responsibility, norms, and helping in an emergency, J. Personal.
Soc. Psychol., 16, 299, 1970.
Schwartz, S. H. and Howard, J. A., A normative decision-making model of altruism, in Altruism and
Helping Behavior: Social, Personality, and Developmental Perspectives, Rushton, J. P. and Sorrentino,
R. M., Eds., Erlbaum, Hillsdale, NJ, 1981.
Schultz, D., Growth Psychology: Models of the Healthy Personality, Van Nostrand, New York, 1977.
Shotland, R. L. and Heinold, W. D., Bystander response to arterial bleeding: helping skills, the
decision-making process, and differentiating the helping response, J. Personal. Soc. Psychol., 49,
347, 1985.
Skinner, B. F., Selection by consequences, Science, 213, 502, 1981.
Staub, E., Helping a distressed person: social, personality, and stimulus determinants, in Advances
in Experimental Social Psychology, Vol. 7, Berkowitz, L., Ed., Academic Press, New York, 1974.
Steblay, N. M., Helping behavior in rural and urban environments: a meta-analysis, Psychol. Bull.,
102, 346, 1987.
Triandis, H. C., McCusker, C., and Hui, C. H., Multimethod probes of individualism and collec-
tivism, J. Personal. Soc. Psychol., 59, 1006, 1990.
USA Today, Lack of heroes, Wednesday, August 23, 1995, 10A.
Wilson, J. P., Motivation, modeling, and altruism: a person X situation analysis, J. Personal. Soc.
Psychol., 34, 1078, 1976.
Wrightsman, L. S. and Deaux, K., Social Psychology in the 1980’s, 3rd ed., Brooks/Cole Publishing,
Monterey, CA, 1981.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 325

chapter fifteen

The person-based approach to

actively caring
Our willingness to actively care for others is affected by certain feelings and states of mind. If we
have a strong sense of self-esteem, self-efficacy, personal control, optimism, and belonging, there is a
greater chance we will go beyond the call of duty. Each of these person states is explained in this
chapter and the research supporting direct relationships between these states and actively caring
behavior is reviewed. Understanding these connections enables us to design conditions and inter-
ventions to increase actively caring behavior throughout an organization or community.

“Our deeds determine us, as much as we determine our deeds.”—George Eliot

This quotation by George Eliot (Editors of Conari Press, 1993, page 83) indicates that our
behaviors influence something about us and implies that good deeds or actively caring
behaviors are good for us. They change something about us and this, in turn, affects sub-
sequent behavior. Does this mean our actively caring behaviors influence us to actively care
even more? It is a nice thought and seems intuitive, but what does it really mean?
This chapter explores a host of questions arising from the concept reflected in Eliot’s
words. What is it about us that changes as a result of our good deeds, and will this change
lead to more good deeds? Can making people more willing to actively care be influenced
in ways other than managing activators and consequences to directly change behavior? In
other words, can we change something about people that will make them more willing to
actively care for the safety and health of others? If answers to these questions can be turned
into practical procedures, we will know how to increase actively caring behaviors through-
out a culture.
Several years ago, a heart-warming story appeared in our local newspaper, The Roanoke
Times and World Report. The newspaper report of the incident is reprinted in Figure 15.1 and
it is clearly opposite to the Kitty Genovese and Deletha Word tragedies reviewed in
Chapter 14. In this case, two individuals, Tywanii Hairston and John McKee, went out of
their way to save the life of a truck driver named Don Arthur, whose truck was blocking
traffic because he had blacked out. Several individuals had already driven around his truck
without intervening—for a variety of possible reasons discussed in Chapter 14, but two
individuals did interrupt their routine to actively care, and the result was a life saved.
Hairston and McKee went into action, rather than succumb to bystander apathy. Why?
Were there special characteristics of the two heroes? Both individuals did have some life-
saving training in the past, so as discussed in Chapter 14, they might have felt more respon-
sible than others because they knew what to do.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 326

326 Psychology of safety handbook

Pair Breathes Life into Driver

Tywanii Hairston was on the way to pay her water bill
Tuesday when she pulled up behind a Roanoke city truck at
a red light on Campbell Avenue.
The light turned green, but the truck didn’t budge. After
a moment the driver of another car honked the horn, and
still the truck didn’t move. Hairston thought it had stalled.
As she slowly drove around the truck, she saw a man
slumped over the wheel. A former nursing assistant,
Hairston parked her car and jumped out to see what was
wrong.
Meanwhile, John McKee, a driver for Alert Towing on
his way to a job, also had seen the man and stopped. He
radioed his dispatcher to call 911, and with the help of
another passer-by, pulled the man from the truck and laid
him on the street.
Together, Hairston and McKee went into action. Neither
had ever administered cardiopulmonary resuscitation, or
CPR, before, although both knew the procedure. Hairston
started giving mouth-to- mouth, and McKee pumped the
man’s chest until paramedics arrived. Their swift action
quite possibly saved Don Arthur’s life. On Thursday,
Arthur, 60, was in stable condition at Community Hospital,
one of only a few who make it back after venturing so close
to death.

Figure 15.1 Actively caring behavior saved a life. (from The Roanoke Times, Friday, Sept. 8,
1995)
This chapter examines additional person-based reasons (or internal characteristics of
people) that may have contributed to the success story summarized in Figure 15.1. Did the
two intervening agents have personality traits conducive to helping? Did recent experi-
ences influence their personality state in some beneficial way? Obviously, we will never
know the answers to these questions but exploring the possibilities can help us better grasp
the person factors contributing to actively caring behavior. If these factors can be altered
systematically, then it is possible to increase people’s willingness to actively care for the
safety and health of others.

Actively caring from the inside

Perhaps you recall earlier discussions in this text about “outside” vs. “inside” aspects of
people. In Chapter 3, for example, I distinguished between behaviors (outside) vs. inten-
tions, attitudes, and values (inside), and emphasized that we should start with behaviors.
A prime principle of behavior-based psychology is that it is easier, especially for large-scale
culture change, to “act a person into safe thinking” than it is to address attitudes and val-
ues directly in an attempt to “think a person into safe acting.” Another key principle of
behavior-based psychology is that the consequences of our behavior influence how we feel
about the behavior. Generally, positive consequences lead to good feelings or attitudes;
negative consequences lead to bad feelings or attitudes.
Long-term behavior change requires people to change “inside” as well as outside. The
promise of a positive consequence or the threat of a negative one can maintain the desired
behavior while the response–consequence contingencies are in place. What happens when
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 327

Chapter fifteen: The person-based approach to actively caring 327

they are withdrawn? What happens when people are in situations, like at home, when no
one is holding them accountable for their behavior? If people do not believe in the safe way
of doing something and do not accept safety as a value or a personal mission, do not count
on them to choose the safe way when they have the choice. In addition, if people are
not self-motivated to keep themselves safe, do not expect them to actively care for the
safety of others.
Figure 15.2 illustrates how person factors interact with the basic activator-behavior-
consequence model of behavior-based psychology (adapted from Kreitner, 1982). As detailed
earlier, activators direct behavior (Chapter 10) and consequences motivate behavior (Chapter
11). However, as shown in Figure 15.2, these events are first filtered through the person.
As discussed in Chapter 5, numerous internal and situational factors influence how we
perceive activators and consequences. For example, if we see activators and consequences
as nongenuine ploys to control us, our attitude about the situation will be negative. If we
believe the external contingencies are genuine attempts to help us do the right thing, our
attitude will be more positive. Thus, personal, or internal dynamics determine how we
receive activator and consequence information. This can influence whether environmental
events enhance or diminish what we do.
Let us keep in mind that people operate within a context of environmental factors that
have complex and often unmeasurable effects on perceptions, intentions, beliefs, attitudes,
values, and behaviors. This is represented by the environment side of The Safety Triad
(Geller, 1989) and the “environment” designation in Figure 15.2. Such complex interactions
among person, behavior, and environment dimensions of everyday existence often make it
extremely difficult—sometimes impossible—to predict or influence what people will do.
However, certain changes in external and internal conditions can influence people’s behav-
iors consistently and substantially.

Figure 15.2 Activators and consequences to change behavior are filtered through the per-
son. Adapted from Kreitner (1982). With permission.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 328

328 Psychology of safety handbook

In Chapters 10 and 11, I showed how direct manipulations of activators and conse-
quences can influence behavior on a large scale. Now, let us see if changes in internal per-
son factors can benefit behavior change. In particular, how do “inside” factors affect
actively caring for safety?

Person traits vs. states

Some person factors are presumed to be traits, while others are states. Theoretically, traits
are relatively permanent characteristics of people and do not vary appreciably over time or
across situations. The popular Myers-Briggs Type Indicator (Myers and McCaulley, 1985),
for example, was designed to measure where individuals fall along four dichotomous per-
sonality dimensions: extroversion vs. introversion, sensing vs. intuition, thinking vs. feel-
ing, and judgment vs. perception. The various combinations of these dimensions allow for
16 different personality types, each with its special personality characteristics. These traits
are presumed permanent and unchangeable, as determined largely by physiological or bio-
logical factors.
In contrast, states are characteristics that can change moment-to-moment depending
on circumstances and personal interactions. When our goals are thwarted, for example, we
can be in a state of frustration. When experience leads us to believe we have little control
over events around us, we can be in a state of apathy or helplessness. These states can influ-
ence certain behaviors.
Frustration often provokes aggressive behavior; perceptions of helplessness can inhibit
constructive behavior or facilitate inactivity. In contrast, certain life experiences can affect
positive person states, such as optimism, personal control, self-confidence, and belonging-
ness. This, in turn, increases constructive behavior, including actively caring.

Searching for the actively caring personality

Beginning as early as 1928, psychologists have attempted to identify stable personality
traits of helpful people. Although some psychologists might disagree (Huston and Korte,
1976; Rushton, 1984; Staub, 1974), the search has not been particularly successful. For
example, Hartshorne and May (1928) developed 33 different tests to measure positive
social behaviors, including helping, honesty, and resistance to temptation; then they gave
these tests to hundreds of children. Although some children reported a greater willingness
to actively care than others, inconsistencies were more common than consistencies. For
instance, children who stated they would work as a Red Cross volunteer or send get-well
cards to hospitalized peers were not necessarily more willing to share money with class-
mates, and vice versa. The authors concluded that helping behavior is determined more by
situational factors than personality differences. Follow-up research has generally sup-
ported this conclusion (Bar-Tal, 1976; Schwartz, 1977).
In one follow-up study, Gergen et al. (1972) administered a battery of personality tests
to 72 college students and, subsequently, recorded the students’ responses to five different
requests for help from the psychology department. Results were neither simple nor
straightforward. A personality trait that correlated significantly with one actively caring
behavior did not relate to another type of helping behavior. Moreover, relationships
between traits and behaviors were not consistent for males and females. Again, the
researchers had to conclude that personality traits interact in complex ways with both sit-
uational factors and the nature of the helping behavior.
There is some evidence that certain people are consistently more generous, kind,
and helping than others (Rushton, 1984), and that these people score higher on certain
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 329

Chapter fifteen: The person-based approach to actively caring 329

personality scales. Specifically, individuals with a high propensity to actively care tend to
demonstrate empathy or emotional concern or compassion for the welfare of others
(Batson, 1991; Oliner and Oliner, 1988). They score relatively high on measures of moral
development and social responsibility (Staub, 1978), and tend to be self-confident (Aronoff
and Wilson, 1985), idealistic rather than pragmatic (Gilligan, 1982; Waterman, 1988), and
possess a sense of self-control, self-directedness, flexibility, self-acceptance, and independ-
ence (Oliner and Oliner, 1988; Rosenhan, 1970).
It is possible, though, these characteristics are states rather than traits. They might vary
within people according to situations. If so, these qualities of people can be changed with
planned intervention. We can influence them through the use of response –consequence
contingencies and by changing the culture of interpersonal relationships. People can even
be taught to be more empathic toward others, and interventions can be set up to increase
and support empathic or altruistic behavior. It makes sense to treat most person factors
related to actively caring as changeable “states” rather than permanent “traits.” Now, we
can consider ways to change these states to facilitate active caring.

Actively caring states

Using animals, psychologists have influenced marked changes in performance by altering
certain physiological states of their subjects through food, sleep, or activity deprivation.
Similarly, behavioral scientists have demonstrated significant behavior change in both nor-
mal and developmentally disabled children by altering aspects of the social context
(Gewirtz and Baer, 1958a,b) or the temporal proximity of lunch and response–consequence
contingencies (Vollmer and Iwata, 1991). Behavioral scientists typically refer to these
manipulations of physiological conditions or psychological states as “establishing opera-
tions” (Michael, 1982). They set the stage or establish circumstances to facilitate the impact
of an intervention program.
Likewise, certain past or present situations or environmental conditions can influence
or establish physiological or psychological states within individuals. This, in turn, can
affect their behavior. From the behavioral science perspective, a basic mechanism for doing
this is to use the power of positive consequences.
I contend that actively caring characteristics internal to people are states, not traits.
Plus, certain conditions—including activators and consequences—can influence these psy-
chological states (Geller, 1991, 1995, 1998). These states are illustrated in Figure 15.3, a
model my associates and I have used many times to stimulate one-to-one and group con-
versations among employees. We talk of specific situations, operations, or incidents that
influence their willingness to actively care for the achievement of a Total Safety Culture. Let
us examine these influential states in more detail.
Self-esteem (“I am valuable”). How do you feel about yourself? Generally good or
generally bad? Your level of self-esteem is determined by the extent to which you generally
feel good about yourself. If we do not feel good about ourselves, it is unlikely we will care
about making a difference in the lives of others. As illustrated in Figure 15.4, a person’s self-
esteem can get pretty low. The better we feel about ourselves, the more willing we are to
actively care for the safety and health of other people.
One’s self-concept, or feeling of worth, is the central theme of most humanistic thera-
pies (Rogers, 1957, 1977). According to Rogers and his followers, we have a real and ideal
self-concept. That is, we have notions or dreams of what we would like to be (our ideal self)
and what we think we are (our real self). The greater the gap between our real and ideal
self-concepts, the lower our self-esteem. Thus, a prime goal of many humanistic therapies
is to help a person reduce this gap. This can be done by raising a person’s perceptions of
his real self-concept—“Count your strengths and blessings and you will see that you are
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 330

330 Psychology of safety handbook

Personal Control
"I'm in control
Self-Efficacy Optimism
"I can do it "I expect the best

Empowerment
"I can make a difference"

1 2
4

Self-Esteem 3 Belonging
I'm valuable" I belong to a team"

1. I can make valuable differences.

2. We can make a difference.
3. I'm a valuable team member.
4. We can make valuable differences.

Figure 15.3 Certain person states influence an individual’s willingness to actively care for
the safety and health of others.

Figure 15.4 A person’s self-esteem can get pretty low.

much better than that.” The alternative is to lower one’s aspirations or ideal self-concept—
“You expect too much; no one is perfect; take life one step at a time and you will eventu-
ally get there.”
It is important to maintain a healthy level of self-esteem and to help others raise their
self-esteem. Research shows that people with high self-esteem report fewer negative
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 331

Chapter fifteen: The person-based approach to actively caring 331

emotions and less depression than people with low self-esteem (Straumann and Higgins,
1988). Those with higher self-esteem also handle life’s stresses better (Brown and McGill,
1989). Recall the discussion of stress vs. distress in Chapter 7. Higher self-esteem turns
stress into something positive, rather than negative distress.
Researchers have also found that individuals who score higher on measures of self-
esteem are less susceptible to outside influences (Wylie, 1974), more confident of achieving
personal goals (Wells and Marwell, 1976), and make more favorable impressions on others
in social situations (Baron and Byrne, 1994). Supporting the actively caring model depicted
in Figure 15.3, people with higher self-esteem help others more frequently than those scor-
ing lower on a self-esteem scale (Batson et al., 1986).
Here is something to keep in mind, though. It has also been found that people with
high self-esteem are less willing to ask for help than people with low self-esteem (Nadler
and Fisher, 1986; Weiss and Knight, 1980). Later in this chapter, I discuss in more detail
research that shows a direct relationship between self-esteem and actively caring behavior.
Empowerment (“I can make a difference”). In the management literature, empow-
erment typically refers to delegating authority or responsibility, or sharing decision
making (Conger and Kanungo, 1988). In contrast, the person-based perspective of empow-
erment focuses on how the person who receives more power or influence reacts.
From a psychological perspective, empowerment is a matter of personal perception.
Do you feel empowered or more responsible? Can you handle the additional assign-
ment? This view of empowerment requires the personal belief that “I can make a
difference.”
Perceptions of personal control (Rotter, 1966), self-efficacy (Bandura, 1977), and opti-
mism (Scheier and Carver, 1985, 1993; Seligman, 1991) strengthen the perception of
empowerment. An empowered state is presumed to increase motivation to “make a differ-
ence,” perhaps by going beyond the call of duty. As I discuss later in this chapter, there is
empirical support for this intuitive hypothesis (Bandura, 1986; Barling and Beattie, 1983;
Ozer and Bandura, 1990; Phares, 1976).
Figure 15.5 includes an instructive and provocative story about the loss of empower-
ment in a simple but typical school situation. Many readers will relate empathetically with
the young boy, having been in similar situations themselves. They know what it is like to
have their empowerment sapped (Byham, 1988). The first teacher takes too much personal
control over the situation. As a result, the student loses a sense of self-efficacy (“I can do it
myself”), personal control (“I am in control”), and even optimism (“I expect the best”). All
of this diminishes the sense of being able to contribute. Instead, the individual learns to
wait for top-down instructions and is motivated to do only what is required. This is sad and
all too common in home, school, and occupational settings. What is even more dishearten-
ing is that many people, like the first teacher in this story, sap empowerment from others
and do not even realize it.
Let us look more closely at these three factors affecting our sense of worth and ability—
and our propensity to actively care.
Self-efficacy is the idea that “I can do it.” This is a key factor in social learning theory,
determining whether a therapeutic intervention will succeed over the long term (Bandura,
1990, 1994, 1997). We are talking about your self-confidence. Dozens of studies have found
that subjects who score relatively high on a measure of self-efficacy perform better at a
wide range of tasks. They show more commitment to a goal and work harder to pursue
it. They demonstrate greater ability and motivation to solve complex problems at
work. They have better health and safety habits and they are more apt to handle stressors
positively, rather than with negative distress (Bandura, 1982; Betz and Hackett, 1986;
Hackett et al., 1992).
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 332

332 Psychology of safety handbook

Figure 15.5 A simple story illustrates common sapping of empowerment and unfortunate
consequences. Adapted from Canfield and Hansen (1993). With permission.

Self-efficacy contributes to self-esteem, and vice versa; but these constructs are differ-
ent. Simply put, self-esteem refers to a general sense of self-worth; self-efficacy refers to
feeling successful or effective at a particular task. Self-efficacy is more focused and can vary
markedly from one task to another. One’s level of self-esteem remains rather constant
across situations.
When I am losing to an opponent on the tennis court, my self-efficacy usually drops
considerably. However, my self-esteem might not change at all. I might protect my self-
esteem by rationalizing that my opponent is younger and more experienced or that I am
more physically tired and mentally preoccupied than usual. My damaged self-efficacy will
undoubtedly lead to reduced optimism about winning the match. If I continue to lose at
tennis and run out of excuses, my self-esteem could suffer if I think it is important for me
to play tennis well. In this case, there would be a prominent gap between my real self, a
loser at tennis, and ideal self, a winner on the court.
Personal control is the feeling that “I am in control.” Rotter (1966) used the term locus of
control to refer to a general outlook regarding the location of forces controlling a person’s
life—internal or external. Those with an internal locus of control believe they usually have
direct personal control over significant life events as a result of their knowledge, skill, and
abilities. They believe they are captain of their life’s ship.
In contrast, persons with an external locus of control believe factors like chance, luck,
or fate play important roles in their lives. In a sense, externals believe they are victims, or
sometimes beneficiaries, of circumstances beyond their direct personal control (Rotter,
1966; Rushton, 1984). As depicted in Figure 15.6, however, there are times when everyone
likes to feel that their successes resulted from their own efforts.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 333

Chapter fifteen: The person-based approach to actively caring 333

Figure 15.6 At times, we all want credit for our personal control.

Personal control has been one of the most researched individual difference dimensions
in psychology. Since Rotter developed the first measure of this construct in 1966, more than
2000 studies have investigated the relationship between perceptions of personal control
and other variables (Hunt, 1993). Internals are more achievement-oriented and health con-
scious than externals. They are less prone to distress and more likely to seek medical treat-
ment when they need it (Nowicki and Strickland, 1973; Strickland, 1989). In addition,
having an internal locus of control helps reduce chronic pain, facilitates psychological and
physical adjustment to illness and surgery, and hastens recovery from some diseases
(Taylor, 1991). Internals perform better at jobs that allow them to set their own pace,
whereas externals work better when a machine controls the pace (Eskew and Riche, 1982;
Phares, 1991).
Optimism is reflected in the statement, “I expect the best.” It is the learned expectation
that life events, including personal actions, will turn out well (Peterson, 2000; Scheier and
Carver, 1985, 1993; Seligman, 1991). Optimism relates positively to achievement. Seligman
(1991) reported, for example, that world-class swimmers who scored high on a measure of
optimism recovered from defeat and swam even faster compared to those swimmers scor-
ing low. Following defeat, the pessimistic swimmers swam slower.
Compared to pessimists, optimists maintain a sense of humor, perceive problems or
challenges in a positive light, and plan for a successful future. They focus on what they can do
rather than on how they feel (Carver et al., 1989; Sherer et al., 1982; Peterson and Barrett, 1987).
As a result, optimists handle stressors constructively and experience positive stress rather
than negative distress (Scheier et al., 1986). Optimists essentially expect to be successful at
whatever they do, and so they work harder than pessimists to reach their goals. As a result,
optimists are beneficiaries of the self-fulfilling prophecy (Tavris and Wade, 1995). Figure 15.7
shows how an optimistic perspective can influence one’s attempt to achieve more.
The self-fulfilling prophecy (Merton, 1948) starts with a personal expectation about
one’s future performance and ends with that expectation coming true because the individ-
ual performs in such a way to make it happen. An experiment by Feather (1966) demon-
strated how quickly the self-fulfilling prophecy can take effect. For 15 trials, he asked
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 334

334 Psychology of safety handbook

Figure 15.7 Optimists expect more from their efforts.

female college students to unscramble letters to make a word. Prior to each trial, the sub-
jects predicted their chances of solving the puzzle or anagram. For the first five trials, half
of the women received easy anagrams, while the other subjects received five anagrams
with no solution.
As you might expect, the group that started with easy anagrams increased their esti-
mates of success on subsequent trials; those who received the five insoluable puzzles
became pessimistic about their future success. The optimistic subjects performed markedly
better on the last ten anagrams which were soluble and the same for both groups. The
higher a person’s expectation for success, the more anagrams he solved. When people are
optimistic and expect the best, they work hard to make their prediction come true. As a
result, they often achieve the best.
What do you expect when your boss or supervisor asks to see you? Do you expect the
best? Our past experiences with top-down control and the use of negative consequences to
influence our behavior often results in pessimistic rather than optimistic expectations.
Moreover, our approach to this situation, illustrated in Figure 15.8, can support our nega-
tive expectations.
If you expect to be punished or reprimanded every time your boss or supervisor calls
you into the office, then your body language and demeanor will subtly reflect that expec-
tation. You will “telegraph” these signals to your boss, who might think, “Scott sure looks
guilty, I wonder what he’s done that needs to be punished?”
However, if you approach the interaction with an optimistic attitude, reflected in your
body language and verbal behavior, the results could be more positive. You could, for
example, write a different internal script. “No one is perfect, and I might have missed
something. Everyone can improve with specific behavioral feedback. If I help to make the
interaction constructive, the outcome can only be positive.”
It is important to understand that fulfilling a pessimistic prophecy can depreciate our
perceptions of personal control, self-efficacy, and even self-esteem. Realizing this should
motivate us to do whatever we can to make interpersonal conversations positive and
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 335

Chapter fifteen: The person-based approach to actively caring 335

Figure 15.8 When our boss asks to see us, we expect the worst.

constructive. This will not only increase optimism in a work culture but also promote a
sense of group cohesiveness or belonging—another person state that facilitates actively
caring behavior.
Belonging. In his best seller, The Different Drum: Community Making and Peace, Peck
(1979) challenges us to experience a sense of true community with others. We need to
develop feelings of belonging with one another regardless of our political preferences, cul-
tural backgrounds, and religious doctrine. We need to transcend our differences, overcome
our defenses and prejudices, and develop a deep respect for diversity. Peck claims we must
develop a sense of community or interconnectedness with one another if we are to accom-
plish our best and ensure our survival as human beings. As illustrated in Figure 15.9, the
opposite of this perspective or win–lose independence is often experienced on the road,
fueling “road rage,” and contributing to numerous vehicle crashes and fatalities.
It seems intuitive that building a sense of community or belonging among our co-
workers will improve organizational safety. Safety improvement requires interpersonal
observation and feedback and, for this to happen, people need to adopt a collective
win–win perspective instead of the individualistic win–lose orientation common in many
work settings. A sense of belonging and interdependency leads to interpersonal trust and
caring—essential features of a Total Safety Culture.
In my numerous group discussions with employees on the belonging concept, some-
one inevitably raises the point that a sense of belonging or community at their plant has
decreased over recent years. “We used to be more like family around here” is a common
theme. For many companies, growth spurts, continuous turnover—particularly among
managers—or “lean and mean” cutbacks have left many employees feeling less connected
and trusting. It seems, in some cases, people’s need level on Maslow’s hierarchy (see Figure
14.7) has regressed from satisfying social acceptance and belonging needs to concentrating
on maintaining job security in order to keep food on the table.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 336

336 Psychology of safety handbook

Figure 15.9 A win–lose independent perspective makes vehicle travel more risky.

Figure 15.10 lists a number of special attributes prevalent in most families where inter-
personal trust and belonging are often optimal. We are willing to actively care in special
ways for the members of our immediate family. The result is optimal trust, belonging, and
actively caring behavior for the safety and health of our family members. To the extent we
follow the guidelines in Figure 15.10 among members of our “corporate family,” we will
achieve a Total Safety Culture. Following the principles in Figure 15.10 will develop trust
and belonging among people and lead to the quantity and quality of actively caring behav-
ior expected among family members—at home and at work.
The psychological construct most analogous to the actively caring concept of belong-
ing is group cohesion—the sum of positive and negative forces attracting group members
to each other (Wheeless et al., 1982). Satisfaction is considered a key determinant of group
cohesiveness. The more cohesive a group, the more satisfied are members with belonging
to the group. Also, the greater the member satisfaction with the group, the greater the
group cohesiveness.
Wheeless et al. (1982) identified two beneficial levels of satisfaction in interpersonal
relationships: independence and involvement. Independence refers to an internal locus of
control in group decision making and group involvement reflects the level of interpersonal
concern, respect, and warmth present in the group.
Ridgeway (1983) defined five benefits of group cohesiveness, including increased
quantity and quality of communication, individual participation, group loyalty and satis-
faction, ability to enforce group norms and focus energy toward goal attainment, and elab-
oration of group culture, typified by special behavioral routines that increase the group’s
sense of togetherness. From this conceptualization, it follows that members of a cohesive
group should demonstrate actively caring behavior for each other. The actively caring
model also predicts that group cohesiveness will increase this behavior for targets—
persons, behaviors, and environments—outside the group.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 337

Chapter fifteen: The person-based approach to actively caring 337

• We use more rewards than penalties with

• We use more rewards than penalties with
family members.
family members.
• We don’t pick on the mistakes of family
• We don’t pick on the mistakes of family mem-
members.
bers.
• We don’t rank one family member against
• We don’t rank one family member against
another.
another.
• We brag about the accomplishments of family
• We brag about the accomplishments of family
members.
members.
• We respect the property and personal space
• We respect the property and personal space of
of family members.
family members.
• We pick up after other family members.
• We pick up after other family members.
• We correct the at-risk behavior of family
• We correct the at-risk behavior of family
members.
members.
• We accept the corrective feedback of family
• We accept the corrective feedback of family
members.
members.
• We are our brothers/sisters keepers of family
• We are our brothers/sisters keepers of family
members.
members.
• We actively care because they’re family.
• We actively care because they’re family.

Figure 15.10 Incorporating an actively caring family perspective in an organization will

help to cultivate a Total Safety Culture.

Measuring actively caring states

Surveys that measure workplace safety cultures are quite popular these days (Geller, 1992;
Geller and Roberts, 1993; Simon and Simon, 1992). Some proponents recommend their use
to discriminate between “safe” and “unsafe” employees (Krause, 1992, 1995). To justify
these surveys, consultants often teach that individuals have stable personality traits deter-
mining both their motivation level for particular tasks and their propensity to have an
injury. This perspective seems to be on the rise today, as safety consultants peddle their
“quick-fix” measurement devices.
I believe the idea of a stable personality bias can interfere with the more practical and
cost-effective behavior-based approach to managing human resources, which is what this
Handbook is all about. If you recall in Chapter 1, the comprehensive research comparisons
by Guastello (1993) revealed this personnel selection approach to be the most popular—but
quite ineffective at reducing industrial injuries.
At my industrial safety workshops, I explain that valid individual difference scales
do not exist to reliably predict an individual’s propensity to get hurt on the job
(Geller, 1994). Even if they were available, you still must account for the influence of
such contextual factors as environmental conditions, management systems, response–
consequence contingencies, and peer interactions—in addition to personality factors.
I have found, though, that assessing individual differences—including different life-
styles, personality factors, perceptions, and cognitive strategies—can be useful in an
employee education and training program to teach the concept of individual diversity and
to increase employees’ awareness of their own idiosyncrasies that relate potentially to
injury proneness.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 338

338 Psychology of safety handbook

A safety culture survey

The Safety Culture Survey which Steve Roberts, Mike Gilmore, and I developed for culture
assessment and corporate training programs (Geller and Roberts, 1993; Geller et al., 1992,
1996) includes subscales to measure safety-related perceptions and risk propensity, includ-
ing cognitive failures (Broadbent et al., 1982), sensation seeking (Zuckerman, 1979), psycho-
logical reactance (Tucker and Byers, 1987), and extroversion (Eysenck and Eysenck, 1985)
The most useful subscale of our Safety Culture Survey, from both a training and cul-
ture-change perspective, is the actively caring scale, which includes adaptations from
standard measures of self-esteem (Rosenberg, 1965), self-efficacy (Sherer et al., 1982), per-
sonal control (Nowicki and Duke, 1974), optimism (Scheier and Carver, 1985), and group
cohesion (Wheeless et al., 1982). The survey also includes direct measures of willingness to
actively care from an environment focus (“I am willing to pick up after another employee
in order to maintain good housekeeping”), a person focus (“If an employee needs assis-
tance with a task, I am willing to help even if it causes me inconvenience”), and a behav-
ior-change focus (“I am willing to observe the work practices of another employee in order
to provide direct feedback to him/her”). Respondents’ reactions to each of the 154 items of
the survey are given on a 5-point Likert-type scale ranging from “Highly Disagree” to
“Highly Agree.”

Support for the actively caring model

Analyzing Safety Culture Survey results from three large industrial complexes shows
remarkable support for the actively caring model (Geller et al., 1996; Roberts and Geller,
1995). The personal control factor was consistently most influential in predicting willing-
ness to actively care. Belonging scores predicted significant differences in actively caring
propensity at two of three plants. Self-esteem and optimism always correlated highly with
each other, and with willingness to actively care, but only one or the other predicted inde-
pendent variance in actively caring propensity. For these tests, our survey did not include
a measure of self-efficacy. The multiple regression coefficients and sample sizes were
0.54 (n  262), 0.57 (n  307), and 0.71 (n  207) at the three plants, respectively (see Geller
et al., 1996 for details).
These regression results were not of much interest to the plant managers, supervisors,
and trainers at the three facilities, but there is a practical value to classifying actively car-
ing attributes according to various work groups, including managers, operators, secre-
taries, contractors, and laboratory personnel. At another plant, for example, relatively high
levels of willingness to actively care convinced the plant manager to support an actively
caring training and intervention process. In another case, extreme differences in the incli-
nation to help across work areas prompted the development of special intervention pro-
grams for certain work groups.

Check your understanding

The 20 questions included in Figure 15.11 were selected from the actively caring person
scale of our Safety Culture Survey. Each of the five actively caring states discussed in this
chapter is assessed. There are only four questions per state, so this should not be consid-
ered a reliable nor a valid measure of these factors. In other words, do not read too much
into this survey. Just respond to each query according to the instructions and then check
the answer key in Figure 15.12 to increase your understanding of the five actively caring
person states.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 339

Chapter fifteen: The person-based approach to actively caring 339

This is a questionnaire about your beliefs and Highly Agree

feelings. Read each statement, then circle the Agree
number that best describes your current feelings. Not Sure
There are no "right" or "wrong" answers. This Disagree
questionnaire only asks about your personal opinion. Highly Disagree

1. I feel I have a number of good qualities. 1 2 3 4 5

2. Most people I know can do better job than I can. 1 2 3 4 5
3. On the whole, I am satisfied with myself. 1 2 3 4 5
4. I feel I don't have much to be proud of. 1 2 3 4 5
5. When I make plans, I am certain I can make them work. 1 2 3 4 5
6. I give up on things before completing them. 1 2 3 4 5
7. I avoid challenges. 1 2 3 4 5
8. Failure just makes me try harder. 1 2 3 4 5
9. People who never get injured are just plain lucky. 1 2 3 4 5
10. People's injuries result from their own carelessness. 1 2 3 4 5
11. I am directly responsible for my own safety. 1 2 3 4 5
12. Wishing can make good things happen. 1 2 3 4 5
13. I hardly ever expect things to go my way. 1 2 3 4 5
14. If anything can go wrong for me, it probably will. 1 2 3 4 5
15. I always look on the bright side of things. 1 2 3 4 5
16. I firmly believe that every cloud has a silver lining. 1 2 3 4 5
17. My work group is very close. 1 2 3 4 5
18. I distrust the other workers in my department. 1 2 3 4 5
19. I feel like I really belong to my work group 1 2 3 4 5
20. I don't understand my coworkers. 1 2 3 4 5

Figure 15.11 These 20 survey items assess the 5 actively caring person states.

Comparing the items that measure self-esteem with those that assess self-efficacy, for
example, helps you understand the distinct difference between these two constructs.
Remember, these person factors are presumed to be states that fluctuate day-to-day and
from situation-to-situation. The score you get today might be quite different than the one
you would obtain on another day under a different set of circumstances.

Theoretical support for the actively caring model

The actively caring model depicted in Figure 15.3 certainly makes intuitive sense. Does
your willingness to help others often change, depending on the person states given in the
model? When we feel better about ourselves, we are less preoccupied with personal prob-
lems and more likely to do something nice for someone else. Of course, we feel better about
ourselves when reaching our aspirations through self-efficacy and personal control. In
turn, this satisfaction can lead to optimistic expectations and heightened self-esteem.
Moreover, when it comes to helping others, we are more apt to help those we like and to
whom we feel close. Theorizing a direct relationship between the probability of actively
caring behavior and the degree of belonging between the helper and the victim should not
surprise anyone.
The common-sense appeal of the actively caring model is also supported by compar-
ing it with Maslow’s popular and intuitive motivation theory. Maslow’s hierarchy of needs
was discussed in Chapter 14 (see Figure 14.7), with particular reference to Maslow’s (1971)
later addition of “self-transcendency” to the top of his hierarchy. I bet you see the similar-
ity between self-transcendency and actively caring.
Figure 15.13 depicts a hierarchy of the concepts in the actively caring model, along with
the need levels of Maslow’s hierarchy. The similarities are noteworthy. Indeed, one-half of
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 340

340 Psychology of safety handbook

Self-Esteem (items 1-4) = feelings of self-worth and value.

Actual scale = 16 items
Total 1 =
(a) Add numbers for items 1 & 3
Total 2 =
(b) Add numbers for items 2 & 4 and subtract from 12.

Self-Efficacy (items 5-8) = general level of belief in one's competence.

Actual scale = 23 items Total 1 =

(a) Add numbers for items 5 & 8 Total 2 =
(b) Add numbers for items 6 & 7 and subtract from 12.

Personal Control (items 9-12) = the extent a person believes he or

she is personally responsible for his/her life situation.
Actual scale = 25 items
(a) Add numbers for items 10 & 11 Total 1 =
(b) Add numbers for items 9 & 12 and subtract from 12. Total 2 =

Optimism (items 13-16) = the extent to which a person expects

the best will happen for him/her.
Actual scale = 8 items
(a) Add numbers for items 15 & 16 Total 1 =
(b) Add numbers for items 13 & 14 and subtract from 12.
Total 2 =

Belonging (items 17-20) = the perception of group cohesiveness

or feelings of togetherness.
Actual scale = 20 items
(a) Add numbers for items 17 & 19 Total 1 =
(b) Add numbers for items 16 & 20 and subtract from 12. Total 2 =

ACTIVELY CARING SCORE = Sum of Self-Esteem, Self-Efficacy.

Optimism, Personal Control, and Belonging Totals. Total Score =

Figure 15.12 Scoring your answers to the 20 person-state items will increase your under-
standing of the actively caring model.

the actively caring concepts—belonging, self-esteem, and actively caring—are exactly the
same as three of Maslow’s need levels. The other three person states—self-efficacy, per-
sonal control, and optimism—can be readily linked to the remaining three need levels. It
makes sense to relate self-efficacy or self-confidence to an individual’s drive to satisfy basic
physiological needs. Increasing one’s sense of personal control is basic to feeling safe and
secure. Indeed, increasing employees’ personal control of safety is fundamental to achiev-
ing a Total Safety Culture.
Linking optimism with self-actualization might be a bit of a stretch, but does it not
seem that optimism reflects self-actualization, or vice versa? When people believe they
are the best they can be, when they are self-actualized, they are happiest and most opti-
mistic about the future. Their self-fulfilling prophecy comes true, and they continue to be
self-actualized.
All of this sounds good, but we cannot allow common sense to determine the value of
a theory. That would be inconsistent with a primary theme of this Handbook. Principles and
procedures must be based on valid results from research, not on common sense.
There is, however, a practical benefit to connecting a theory or model to common
sense—it scores points for what you are trying to prove, but the accuracy or applicability
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 341

Chapter fifteen: The person-based approach to actively caring 341

Actively
Caring
(Self-Transcendency)

Optimism
(Self-Actualization)

Self-Esteem
(Self-Esteem Needs)

Belonging
(Acceptance Needs)

Personal Control
(Safety and Security Needs)

Self-Efficacy
(Physiological Needs)

Figure 15.13 The actively caring person states are reflected in Maslow’s hierarchy of
needs.
of a theory or concept cannot be based on whether it sounds good or seems acceptable. As
discussed earlier, Maslow’s hierarchy of needs has intuitive charm, but limited empirical
research has been conducted to support the concept of a motivational hierarchy. On the
other hand, there has been substantial research on helping behavior that can be related to
the actively caring model. Let us see if research supports the theory. Perhaps, some of it will
suggest practical applications for injury prevention.

Research support for the actively caring model

I have found a number of empirical studies, mostly in the social psychology literature, sup-
porting individual components of the actively caring model. Although these studies did
not address more than one factor at a time, the combined evidence gives substantial empir-
ical support for the model.
The bystander intervention paradigm (as described in Chapter 14) has been the most
common and rigorous laboratory method used to study person factors related to actively
caring behaviors. This research measures or manipulates self-esteem, empowerment, and
belonging among a group of subjects. Then, these individuals are placed in a situation
where they have an opportunity to help another person presumably encountering some
kind of a personal crisis, like falling off a ladder, dropping belongings, or feigning a heart
attack or illness. The delay in coming to the rescue is studied as it relates to a subject’s social
situation or personality state. As pointed out in Chapter 14, the actively caring behaviors
studied in these experiments were reactive and person-focused. They were never proactive
and behavior-focused.
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 342

342 Psychology of safety handbook

Self-esteem
According to Coopersmith (1967), self-esteem can be relatively general and enduring—and
also situational and transitory. Often considered a general trait, self-esteem can be affected
markedly by situations, response–consequence contingencies, and personal interactions.
When circumstances return to “normal,” self-esteem usually returns to its chronic level, but
permanent changes in circumstances or perceived personal competencies can have a last-
ing impact. Maturity, for example, can shape or reshape personal aspirations, and so
change the gap between a person’s perceived real and ideal self.
Michelini et al. (1975) and Wilson (1976) measured subjects’ self-esteem with a sentence
completion test (described in Aronoff, 1967). Then, they observed whether these subjects
helped out in a bystander intervention test. High self-esteem subjects were significantly
more likely than those with low self-esteem to help another person pick up dropped books
(Michelini et al., 1975) and to leave an experimental room to assist a person in another room
who screamed he had broken his foot following a mock “explosion” (Wilson, 1976).
Similarly, subjects with higher self-esteem scores were more likely to help a stranger (the
experimenter’s accomplice) by taking his place in an experiment where they would pre-
sumably receive electric shocks (Batson et al., 1986).
In a naturalistic field study of actively caring behavior, Bierhoff et al. (1991) compared
individual differences among people who helped or only observed at vehicle crashes.
People who stopped at the road scene were identified by ambulance workers, and were
later given a questionnaire measuring certain personality constructs. Those who helped
scored significantly higher on self-esteem, personal control, and social responsibility.

Personal control
Some studies have measured subjects’ locus of control and then observed the probability
of actively caring behavior in a bystander intervention trial. Another study manipulated
subjects’ perceptions of personal control prior to observing their actively caring behaviors.
The field study discussed in the previous section by Bierhoff et al. (1991) found more active
caring at vehicle-crash scenes by bystanders with an internal locus of control.
Also, those high self-esteem subjects who showed more active caring than low self-
esteem subjects in Wilson’s (1976) bystander intervention study (discussed previously)
were also characterized as internals, in contrast to the lower self-esteem externals who were
less apt to actively care. In addition, Midlarsky (1971) found more internals than externals
willing to help a confederate perform a motor coordination task that involved the recep-
tion of electric shocks.
Sherrod and Downs (1974) asked subjects to perform a task in the presence of a loud,
distracting noise. They manipulated subjects’ perception of personal control by telling one-
half the subjects they could terminate the noise, if necessary, by notifying them through an
intercom. The subjects who could have terminated the noise but did not were significantly
more likely to comply with a later request by a accomplice to help solve math problems
requiring extra time and resulting in no extrinsic benefits.

Optimism
As mentioned in Chapter 14, researchers have manipulated optimistic states or moods by
giving test subjects unexpected rewards or positive feedback and, then, observing the fre-
quency of actively caring behaviors. Isen and Levin (1972), for example, observed that 84
percent of those individuals who found a dime in the coin-return slot of a public phone
(placed there by researchers) helped an accomplice pick up papers he dropped in the sub-
ject’s vicinity. In contrast, only four percent of those who did not find a dime helped the
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 343

Chapter fifteen: The person-based approach to actively caring 343

man pick up his papers. Similarly, students given a cookie while studying at a university
library were more likely than those not given a cookie to agree to help another person by
participating in a psychology experiment (Isen and Levin, 1972).
Isen et al. (1976) delivered free samples of stationery to people’s homes and then called
them later to request an act of caring. Specifically, the caller said he had dialed the wrong
number but since he had used his last dime, he needed the subject to call a garage to tow
his car. Subjects who had received the stationery were more likely to make the call than
subjects who had not received this gift.
Carlson et al. (1988) reviewed these and other studies that showed direct relationships
between mood—or optimism—and actively caring behavior. They reported that these
pleasant experiences increased active caring, purportedly by inducing a positive mood or
optimistic outlook: finding a dime, receiving a packet of stationery, listening to soothing
music, being on a winning football team, imagining a vacation in Hawaii, and being
labeled a charitable person. The authors suggested that the state or mood caused by the
pleasant experiences may have increased the perceived rewarding value of helping others.
Berkowitz and Connor (1966) found a direct relationship between perceived success
and actively caring behavior. Their subjects were instructed to complete certain puzzles in
less than two minutes. The task was manipulated to allow one-half of the subjects to suc-
ceed and one-half to fail. Afterward, successful subjects made more boxes for the
researcher’s accomplice than did the unsuccessful subjects. In a series of analogous labo-
ratory studies, Isen (1970) manipulated performance feedback on a perceptual-motor task.
Subjects told they had performed extremely well were more likely to donate money to
charity, pick up a dropped book, and hold a door open for a confederate than those who
were told that they had performed very poorly.
These later studies that manipulated the outcome of a task illustrated a potential over-
lap between optimism, self-efficacy, and personal control. It is reasonable to assume per-
formance feedback increases one’s perception of self-efficacy and personal control, as well
as one’s optimism. Indeed, Scheier and Carver’s (1985) measure of optimism correlated
significantly with locus of control. Optimism, self-efficacy, and personal control determine
feelings of empowerment, according to the actively caring model. Thus, these perform-
ance-feedback studies support the general hypothesis that we can increase the chances for
active caring by boosting individual perceptions of empowerment.

Belonging
Staub (1978) reviewed studies showing that people are more likely to help victims who
belong to their own group, with “group” determined by race, nationality, or an arbitrary
distinction defined by preference of a particular artist’s paintings. Similarly, Batson et al.
(1986) found subjects more likely to help a confederate if they rated her as similar to them.
In a bystander intervention experiment, pairs of friends intervened faster to help a female
experimenter who had fallen from a chair than did pairs of strangers. Thus, the bystander
intervention effect, which holds that victims are less likely to be helped as the number of
observers increases, may not occur when friends are involved. Group cohesiveness or a
sense of belonging counteracts the diffusion of responsibility that presumably accounts for
bystander inaction.
By experimentally manipulating group cohesion in groups of two and four, Rutkowski
et al. (1983) tested whether group cohesion can reverse the usual bystander intervention
effect. Cohesiveness was created by having the groups discuss topics and feelings they had
in common related to college life. Then, the researchers studied the number of subjects who
left the experimental room to assist a “victim” who had presumably fallen off a ladder and
measured how long it took to respond. Findings indicated that group cohesiveness
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 344

344 Psychology of safety handbook

increased actively caring behaviors, perhaps because of reduced diffusion of responsibility.

Both frequency and speed of helping were greater for the cohesive groups. Subjects in the
high cohesion/four-person group were most likely to respond quickly.
Blake (1978) studied real-world relationships between group cohesion and the ultimate
act of caring—altruistic suicide. He gathered his data from official records of Medal of Honor
awards given during World War II and Vietnam. The independent variable was the cohe-
siveness of combat units, estimated by group training and size; the dependent variable was
percentage of “grenade acts”—voluntarily using one’s body to shield others from explod-
ing devices. Results revealed that the smaller, more elite, specially-trained combat units—
the Marine Corps and Army airborne units—accounted for a substantially larger percentage
of “grenade acts” than larger, less specialized units—Army nonairborne units. These find-
ings also supported the hypothesis that group cohesion increases actively caring behavior.

Direct test of the actively caring model

As I indicated, none of these empirical studies were designed to test the actively caring
model, but our research with the Safety Culture Survey has shown a direct relationship
between employees’ scores on the five person states and their self-reported actively caring
behavior (Geller and Roberts, 1993; Geller et al., 1996). People who scored high on measures
of self-esteem, self-efficacy, personal control, optimism, and belonging reported that they had
performed more acts of caring in the past, and they reported a significantly greater willing-
ness to actively care in the future. A major weakness of this research was its complete reliance
on verbal report. Actual behavior was not observed. Two field tests overcame this weakness.
Roberts and Geller (1995) studied relationships between on-the-job actively caring
behaviors of 65 employees and prior measures of their self-esteem, optimism, and group
cohesion. Self-efficacy and personal control were not assessed. Employees were told to give
co-workers “Actively Caring Thank-You Cards” (Figure 11.16) redeemable for a beverage
in the cafeteria whenever they saw a co-worker going beyond the call of duty for safety.
Employees were trained to look for proactive actively caring behavior that removed a haz-
ard, supported safe behavior, or corrected at-risk behavior. Employees who gave or
received a thank-you card scored significantly higher on measures of self-esteem and
group cohesion than those who did not give nor receive a card.
Of my students (Buermeyer et al., 1994), five tested the entire actively caring model by
asking 156 of their peers (75 males and 86 females) who had just donated blood at a cam-
pus location to complete a 60-item survey that measured each of the five person factors in
the model. The high return rate of 92 percent was consistent with an actively caring profile.
Most remarkable, though, was that the blood donors scored significantly higher on each of
the five subscales than did a group of 292 randomly selected students from the same uni-
versity population. The blood donors also scored significantly higher than the others on the
self-report measures of willingness to actively care.
The prominently higher survey scores from blood donors could have resulted from the
immediate effects of donating blood. As reflected in the opening quote of this chapter,
actively caring might very well increase a person’s sense of self-esteem, self-efficacy, per-
sonal control, optimism, and belonging. Whether the differences between blood donors
and the control group were due to pre-existing states of those who gave blood or to the
impact of giving blood, the entire actively caring model was supported by this research.
Actively caring person states are probably present before acts of caring, serving as
establishing conditions that activate the caring behavior, and these states are likely affected
in positive directions after performing an act of caring. Obviously, more research is needed
to study this model as a predictor of when people will actively care and as a predictor of
changes in person states following actively caring behavior. In the next chapter, I use the
820049_CRC1_L1540_CH15 11/10/00 2:04 PM Page 345

Chapter fifteen: The person-based approach to actively caring 345

model as a framework for exploring ways to increase actively caring behavior throughout
an organization or culture.

Actively caring and emotional intelligence

The same person states I have described here as influencing people’s willingness to actively
care for the safety and health of others also reflect a most important kind of human wis-
dom—emotional intelligence (Goleman, 1995, 1998). How important is emotional intelli-
gence? Well, it is probably much more responsible for our successes and failures in life than
mental capacity, or the “intelligence quotient” (IQ) measured by standard IQ tests. From
his comprehensive review of the research, Goleman (1995) concludes that, “At best, IQ con-
tributes about 20 percent to factors that determine life success, which leaves 80 percent to
other factors” (page 34).
Goleman shows convincing evidence that a majority of the other factors contributing
to personal achievement can be associated with “emotional intelligence” or one’s ability to

• Remain in control and optimistic following personal failure and frustration.

• Understand and empathize with other people and work with them cooperatively.

In his influential book, Gardner (1993) refers to the first ability as “intrapersonal intel-
ligence” and the second as “interpersonal intelligence.” We show intrapersonal intelligence
when we keep our negative emotions (including frustration, anger, sadness, fear, disgust,
and shame) in check and use our positive emotions or moods (such as joy, passion,
love, optimism, and surprise) to motivate constructive action. The driver in Figure 15.14
is attempting to control his negative emotions elicited by an unfriendly interpersonal
communication.

Figure 15.14 Interpersonal communication can affect intrapersonal communication, and

vice versa.
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 346

346 Psychology of safety handbook

We demonstrate interpersonal intelligence when we correctly recognize the moods,

emotions, motives, or feeling states of other people and react appropriately. As I discussed
earlier in Chapters 12 and 13, this kind of emotional intelligence requires proactive listen-
ing, behavior-based feedback, and actively caring conversation. Thus, people with high
interpersonal intelligence communicate with others to increase their self-confidence, per-
sonal control, optimism, and self-esteem.
When we communicate with interpersonal intelligence, we facilitate the cultivation of
intrapersonal intelligence in others. As illustrated in Figure 15.15, we often have a choice in
communicating our dissatisfaction. We can impulsively scream and shout our disappoint-
ment or we can attempt a more positive and productive approach. Sometimes, just paus-
ing to think first before reacting can make a big difference. We need to envision how the
process of interpersonal communication is reciprocal and mutually supportive of con-
structive or destructive emotional states.
I am sure you can see the strong connection between the emotional intelligence concept
and the actively caring model discussed previously. Each of the actively caring person
states—self-esteem, self-efficacy, personal control, optimism, and belonging—reflect an
aspect of emotional intelligence, as conceptualized by Goleman (1995, 1998).
Consider also that actively caring for safety increases emotional intelligence in our-
selves and in others. In other words, when we help people avoid taking a calculated risk in
order to achieve a delayed and remote positive consequence (avoiding an injury), we
increase this special intelligence in ourselves. When these people willingly follow our
safety advice and give up the efficiency, comfort, or convenience of an at-risk short cut, they
are enhancing this sort of intelligence in themselves.

Safety, emotions, and impulse control

Safety leaders need to develop emotional intelligence in themselves and others. Think
about the range of emotions that come into play as we struggle to improve work-
place safety and health. We need the curiosity to assess objectively the impact of our safety

Figure 15.15 Interpersonal conversation affects the actively caring person states.
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 347

Chapter fifteen: The person-based approach to actively caring 347

interventions, persistence to continue successful programs in the face of active resistance,

flexibility to try new approaches, resilience to bounce back after failure, and passion to
try again.
Achieving the vision of an injury-free workplace requires awareness and control of our
own emotions, as well as the ability to assess, understand, and draw on the influence of
other people’s emotions. This requires empathic and persuasive communication skills
(interpersonal intelligence), as well as self-confidence, personal control, self-esteem, and
optimism (intrapersonal intelligence) to develop and implement new tools for safety
management.
Perhaps, we can increase personal responsibility for safety by helping people under-
stand the fundamental emotional problem at the root of all safety intervention. Safety
requires impulse control under the most difficult circumstances. We ask employees to do
things that are uncomfortable or inconvenient in order to avoid a negative consequence
that seems remote and improbable. This takes a special kind of emotional intelligence, both
from us as safety leaders and from the employees with whom we are working.
Goleman (1995) considers impulse control “the root of all emotional self-control” (page
81) and he demonstrates its power in the classic research of Walter Mischel, a renowned
psychology professor at Stanford University. In the 1960s, Mischel gave four-year olds a
“Marshmallow Test” to measure impulse control. Here is how the test worked. Children
were given a marshmallow and told they could eat it now or wait until later and receive
two marshmallows. Some children ate the single marshmallow within a few seconds after
the researcher left the room; others were able to wait the 15 to 20 minutes for the researcher
to return.
The diagnostic power of this simple test was shown when these preschoolers were fol-
lowed up as adolescents (Shoda et al., 1990). Those who put off immediate gratification for
a bigger but delayed reward demonstrated greater intrapersonal and interpersonal intelli-
gence. They handled stressors and frustration with more confidence, personal control, and
optimism. They were more self-reliant, trustworthy, and dependable, and less likely to shy
away from social contacts than the children who had not waited for two marshmallows at
age four.
In comparison, the adolescents who had devoured the single marshmallow 12 to 14
years earlier were now more stubborn and indecisive, more prone to jealousy and envy,
and more readily upset by stress or frustration than the adolescents who had waited for the
extra marshmallow.
When evaluated again during their last year of high school, those who had waited
patiently at age four were far superior as students than those who failed the marshmallow
test. They were clearly more academically competent; they had better study habits and
appeared more eager to learn. They were better able to concentrate, to express their ideas,
and to set goals and achieve them. Most astonishingly, these higher achievers scored signif-
icantly higher on both the math and verbal portions of the SAT (by an average of 210 total
points) than the students who had not delayed gratification at age four (Shoda et al., 1990).

Nurturing emotional intelligence

Although Mischel’s research and Goleman’s conclusion suggest that some degree of emo-
tional intelligence begins early in life, there is plenty of evidence that emotional intelligence
(both intra- and interpersonal) can be learned. Goleman describes a number of educa-
tional/training programs that have demonstrated success at increasing the emotional
intelligence of children. In fact, this Handbook, especially the next chapter, reviews the
actively caring feeling states among adults (including self-confidence, personal control,
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 348

348 Psychology of safety handbook

optimism, belonging, and self esteem) which imply an increase in intrapersonal or inter-
personal intelligence.
I am sure you see the relevance of emotional intelligence to improving occupational
health and safety. Obviously, safety leaders need to remain self-confident and optimistic
(intrapersonal intelligence) in their attempts to prevent injuries, and much of their success
depends upon their ability to facilitate involvement, empowerment, and win–win cooper-
ation among those who can be injured (interpersonal intelligence). However, it is easy for
safety leaders to get discouraged and frustrated, because so often safety seems to take a
back seat to seemingly more immediate demands like meeting production quotas and
quality standards. Controlling these negative emotions is reminiscent of Mischel’s
“Marshmallow Test.”
Doing things for safety (from using protective equipment to completing behavioral
and environmental audits) is equivalent to asking someone to delay immediate gratifica-
tion for the possibility of receiving a larger reward (preventing a serious injury). In other
words, safety often (if not always) requires people to control their impulse to procure an
immediate consequence (if only to be more comfortable or to complete a task faster). Note,
however, that in the “Marshmallow Test” the delayed and larger consequence of two
marshmallows was certain and positive.
The children saw one marshmallow right there in front of them and, if there was one
marshmallow, there could be two. So getting an extra reward for waiting was credible.
When it comes to safety, however, the consequences for impulse control or delaying imme-
diate rewards are usually uncertain and actually improbable.
When we ask people to actively care for health and safety, we are asking them to give
up a powerful immediate reward—the ease, speed, or comfort they get from at-risk behav-
ior. In return for extra effort, we promise a bigger reward—they will prevent personal
injury or perhaps reduce the possibility a coworker will be injured. Unfortunately, this
delayed reward might not seem credible. People have learned they can get away with at-
risk behavior, and many people have not made the connection between their own behav-
ior and the reduction of injuries among others.
Safety requires more emotional intelligence than that shown by the children who
waited for two marshmallows. Believing in the availability of an extra marshmallow was
easy compared to believing participation in a safety process will have a beneficial conse-
quence. Even though we will likely never see a direct connection between a particular
safety practice or process and a reduction in our plant’s safety record, we maintain our faith
(or intrapersonal communication) that certain work practices and interpersonal communi-
cations need to continue. This is yet another reason why actively caring for safety is so chal-
lenging and why it is so important for safety leaders to nurture intra- and interpersonal
intelligence in themselves and among others.

In conclusion
In this chapter, we continued to develop an understanding of actively caring behavior as it
relates to injury prevention. A person-based approach was emphasized; we considered
subjective factors inside people as potential determinants of active caring including their
emotional intelligence. The notion of a general actively caring personality was entertained
at first, but discarded because of relatively limited empirical support. Even if some people
do have permanent traits that make them more or less prone to actively care, it is unlikely
we could use this information to improve safety or prevent injuries. However, we could
benefit safety if certain person states influence people’s willingness to actively care and if
these states can be manipulated by controllable outside factors.
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 349

Chapter fifteen: The person-based approach to actively caring 349

A total of five person states were proposed as influencing people’s willingness to

actively care—self-esteem, self-efficacy, personal control, optimism, and belonging. Each of
these person variables has a prosperous research history in psychology and some of
this research relates directly to the actively caring model. Research that tested relation-
ships between person states and actual behavior was reviewed. Although the results
strongly supported the model, none of the actively caring behavior studied was proactive
or behavior-based.
A few direct tests support the model but suggest the need for further research. A par-
ticularly important question is whether actively caring states are both antecedents and con-
sequences of a caring act. It seems intuitive that performing an act of kindness that is
effective, accepted, and appreciated could increase a helper’s self-esteem, self-efficacy, per-
sonal control, optimism, and sense of belonging. This, in turn, should increase the proba-
bility of more actively caring behavior. In other words, one act of caring, properly
appreciated, should lead to another . . . and another. A self-supporting actively caring
cycle is likely to occur.
The increasingly popular and research-supported concept of emotional intelligence
relates directly to the actively caring model and to improving safety at work, at home, and
on the road. Each of the person states in the Actively Caring Model reflects emotional intel-
ligence, and when people go beyond the call of duty to actively care for the safety or health
of others, they build emotional intelligence in themselves and in the people they help.
Thus, you can see how important it is to get actively caring behavior started and accepted
among a large group of individuals. This challenge is addressed in the next chapter.

References
Aronoff, J., Psychological Needs and Cultural Systems, Van Nostrand, New York, 1967.
Aronoff, J. and Wilson, J. P., Personality in the Social Process, Erlbaum, Hillsdale, NJ, 1985.
Bandura, A., Self-efficacy: toward a unifying theory of behavioral change, Psychol. Rev., 84, 191,
1977.
Bandura, A., Self-efficacy mechanism in human agency, Am. Psychol., 37, 122, 1982.
Bandura, A., Social Foundations of Thought and Action, Prentice-Hall Englewood Cliffs, NJ, 1986.
Bandura, A., Self-regulation of motivation through goal systems, in Nebraska Symposium on
Motivation, Vol. 38, Dienstbier, R. A., Ed., University of Nebraska Press, Lincoln, 1990.
Bandura, A., Self-efficacy, in Encyclopedia of Human Behavior, Vol. 4, Academic Press, Orlando, FL,
1994.
Bandura, A., Self Efficacy: The Exercise of Control, W.H. Freeman, New York, 1997.
Bar-Tal, D., Prosocial Behavior: Theory and Research, Hemisphere, Washington, 1976.
Barling, J. and Beattie, R., Self-efficacy beliefs and sales performance, J. Organ. Behav. Manage., 5, 41,
1983.
Baron, R. A. and Byrne, D., Social Psychology: Understanding Human Interaction, 7th ed., Allyn &
Bacon, Boston, 1994.
Batson, C. D., The Altruism Question: Toward a Social-Psychological Answer, Erlbaum, Hillsdale, NJ,
1991.
Batson, C. D., Bolen, M. H., Cross, J. A., and Neuringer-Benefiel, H. E., Where is altruism in the
altruistic personality?, J. Personal. Soc. Psychol., 1, 212, 1986.
Berkowitz, L. and Connor, W. H., Success, failure, and social responsibility, J. Personal. Soc. Psychol.,
4, 664, 1966.
Betz, N. E. and Hackett, G., Applications of self-efficacy theory to understanding career choice
behavior, J. Soc. Clin. Psychol., 4, 279, 1986.
Bierhoff, H. W., Klein, R., and Kramp, P., Evidence for altruistic personality from data on accident
research, J. Personal., 59(2), 263, 1991.
Blake, J. A., Death by hand grenade: altruistic suicide in combat, Suicide Life-Threat. Behav., 8, 46,
1978.
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 350

350 Psychology of safety handbook

Broadbent, D., Cooper, P. F., Fitzgerald, P., and Parker, K., The cognitive failures questionnaire
(CFQ) and its correlates, Br. J. Clin. Psychol., 21, 1, 1982.
Brown, J. D. and McGill, K. L., The cost of good fortune: when positive life events produce negative
health consequences, J. Personal. Soc. Psychol., 57, 1103, 1989.
Buermeyer, C. M., Rasmussen, D., Roberts, D. S., Martin, C., and Gershenoff, A. G., Red Cross
blood donors vs. a sample of students: an assessment of differences between groups on
“actively caring” person factors, paper presented at the Virginia Academy of Science,
Harrisonburg, VA, May 1994.
Byham, W. C., Zapp! The Lightening of Empowerment, Fawcett Columbine, New York, 1988.
Canfield, J. and Hansen, M. V., Chicken Soup for the Soul, Health Communications, Inc., Deerfield
Beach, FL, 1993.
Carlson, M., Charlin, V., and Miller, N., Positive mood and helping behavior: a test of six hypothe-
ses, J. Personal. Soc. Psychol., 55, 211, 1988.
Carver, C. S., Scheier, M. F., and Weintraub, J. K., Assessing coping strategies: a theoretically based
approach, J. Personal. Soc. Psychol., 56, 267, 1989.
Conari Press, Random Acts of Kindness, Emeryville, CA, 1993.
Conger, J. A. and Kanungo, R. N., The empowerment process: Integrating theory and practice,
Acad. Manage. Rev., 13, 471, 1988.
Coopersmith, S., The Antecedents of Self-Esteem, W. H. Freeman, San Francisco, CA, 1967.
Eskew, R. T. and Riche, C. V., Pacing and locus of control in quality control inspection, Hum. Fact.,
24, 411, 1982.
Eysenck, H. J. and Eysenck, M. W., Personality and Individual Differences: a Natural Science Approach,
Plenum, New York, 1985.
Feather, N. T., Effects of prior success and failure on expectations of success and subsequent per-
formance, J. Personal. Soc. Psychol., 3, 287, 1966.
Gardner, H., Multiple Intelligences, Basic Books, New York, 1993.
Geller, E. S., Managing occupational safety in the auto industry, J. Organ. Behav. Manage., 101, 181,
1989.
Geller, E. S., If only more would actively care. J. Appl. Behav. Anal., 24, 601, 1991.
Geller, E. S., A critical review of human dimension sessions at the 1992 National Safety Council
Congress and Exposition, technical report for Exxon Chemical Company, Virginia Polytechnic
Institute and State University, Blacksburg, VA, 1992.
Geller, E. S., Survey reliability vs. validity: digging into important technical issues, Ind. Saf. Hyg.
News, 12, 12, 1994.
Geller, E. S., Actively caring for the environment: an interaction of behaviorism and humanism,
Environ. Behav., 27, 184, 1995.
Geller, E. S., Understanding Behavior-Based Safety: Step-by-Step Methods to Improve Your Workplace, 7th
ed., J. J. Keller & Associates, Neenah, WI, 1998.
Geller, E. S. and Roberts, D. S., Beyond behavior modification for continuous improvement in occu-
pational safety, paper presented at the FABA/OBM Network Conference, St. Petersburg, FL,
January 1993.
Geller, E. S., Roberts, D. S., and Gilmore, M. R., Achieving a Total Safety Culture through Employee
Involvement, Make-A-Difference, Inc., Newport, VA, 1992.
Geller, E. S., Roberts, D. S., and Gilmore, M. R., Predicting propensity to actively care for occupa-
tional safety, J. Saf. Res., 27, 1, 1996.
Gergen, K. J., Gergen, M. M., and Meter, K., Individual orientation to prosocial behavior, J. Soc. Iss.,
28(3), 105, 1972.
Gewirtz, J. L. and Baer, D. M., Deprivation and satiation of social reinforcers as drive conditions,
J. Abnorm. Soc. Psychol., 57, 165, 1958a.
Gewirtz, J. L. and Baer, D. M., The effect of brief social deprivation on behaviors for a social rein-
forcer, J. Abnorm. Soc. Psychol., 56, 49, 1958b.
Gilligan, C., In a Different Voice: Psychological Theory and Women’s Development, Harvard University
Press, Cambridge, MA, 1982.
Guastello, S. S., Do we really know how well our occupational accident prevention programs
work?, Saf. Sci., 16, 445, 1993.
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 351

Chapter fifteen: The person-based approach to actively caring 351

Goleman, D., Emotional Intelligence, Bantam Books, New York, 1993.

Goleman, D., Working with Emotional Intelligence, Bantam Books, New York, 1998.
Hackett, G., Betz, N. E., Casas, J. M., and Rocha-Singh, I. A., Gender, ethnicity, and social cognitive
factors predicting the academic achievement of students in engineering, J. Counsel. Psychol., 39,
527, 1992.
Hartshorne, H. and May, M. A., Studies in the Nature of Character, Vol. 1, Studies in Deceit, Macmillan,
New York, 1928.
Hunt, M. M., The Story of Psychology, Doubleday, New York, 1993.
Huston, T. L. and Korte, C., The responsive bystander: why he helps, in Moral Development and
Behavior: Theory, Research, and Social Issues, Lickona, T., Ed., Holt, Rhinehart & Winston, New
York, 1976.
Isen, A. M., Success, failure, attention, and reaction to others: the warm glow of success, J. Personal.
Soc. Psychol., 15, 294, 1970.
Isen, A. M. and Levin, P. F., Effect of feeling good on helping: cookies and kindness, J. Personal. Soc.
Psychol., 21, 384, 1972.
Isen, A. M., Clark, M., and Schwartz, M., Duration of the effect of good mood on helping:
“Footprints in the sands of time,” J. Personal. Soc. Psychol., 34, 385, 1976.
Krause, T. R., Behavior science methods for accident prevention, presented at the National Safety
Council Congress and Exposition, Orlando, FL, October 1992.
Krause, T. R., Employee-Driven Systems for Safe Behavior: Integrating Behavioral and Statistical
Methodologies, Van Nostrand Reinhold, New York, 1995.
Kreitner, R., The feedforeward and feedback control of job performance through organizational
behavior management (OBM), J. Organ. Behav. Manage., 4(2), 3, 1982.
Maslow, A. H., The Farther Reaches of Human Nature, Viking, New York, 1971.
Merton, R., The self-fulfilling prophecy, Antioch Rev., 8, 193, 1948.
Michael, J., Distinguishing between discriminative and motivational functions of stimuli, J. Exp.
Anal. Behav., 37, 149, 1982.
Michelini, R. L., Wilson, J. P., and Messe, L. A., The influence of psychological needs on helping
behavior, J. Psychol., 91, 253, 1975.
Midlarsky, E., Aiding under stress: the effects of competence, dependency, visibility, and fatalism,
J. Personal., 39, 132, 1971.
Myers, I. B. and McCaulley, M. H., Manual: A Guide to the Development and Use of the Myers-Briggs
Type Indicator, Consulting Psychologists Press, Palo Alto, CA, 1985.
Nadler, A. and Fisher, J. D., The role of threat to self-esteem and perceived control in recipient reac-
tion to help: theory development and empirical validation, in Advances in Experimental Social
Psychology, Vol. 19, Berkowitz, L., Ed., Academic Press, San Diego, CA, 1986.
Nowicki, S. and Duke, M. P., A locus of control scale for college as well as noncollege adults,
J. Personal. Assess., 38, 136, 1974.
Nowicki, S. and Strickland, B. R., A locus of control scale for children, J. Consult. Psychol., 40, 148, 1973.
Oliner, S. P. and Oliner, P. M., The Altruistic Personality: Rescuers of Jews in Nazi Europe, Free Press,
New York, 1988.
Ozer, E. M. and Bandura, A., Mechanisms governing empowerment effects: a self-efficacy analysis,
J. Personal. Soc. Psychol., 58, 472, 1990.
Peck, M. S., The Different Drum: Community Making and Peace, Simon & Schuster, New York, 1979.
Peterson, C., The future of optimisim, Am. Psychol., 55(1), 44, 2000.
Peterson, C. and Barrett, L. C., Explanatory style and academic performance among university
freshmen, J. Personal. Soc. Psychol., 53, 603, 1987.
Phares, E. J., Locus of Control in Personality, General Learning Press, Morristown, NJ, 1976.
Phares, E. S., Introduction to Personality, 3rd ed., Harper Collins, New York, 1991.
Ridgeway, C. L., The Dynamics of Small Groups, St. Martin’s Press, New York, 1983.
The Roanoke Times, Pair breathes life into driver, September 8, 1995, A1, A2.
Roberts, D. S. and Geller, E. S., An “actively caring” model for occupational safety: a field test, App.
Preven. Psychol., 4, 53, 1995.
Rogers, C., The necessary and sufficient conditions of therapeutic personality change, J. Consult.
Psychol., 21, 95, 1957.
820049_CRC1_L1540_CH15 11/10/00 2:05 PM Page 352

352 Psychology of safety handbook

Rogers, C., Carl Rogers on Personal Power: Inner Strength and Its Revolutionary Impact, Delacorte, New
York, 1977.
Rosenhan, D. L., The natural socialization of altruistic autonomy, in Altruism and Helping Behavior,
Macaulay, J. and Berkowitz L., Eds., Academic Press, New York, 1970.
Rosenberg, M., Society and the Adolescent Self-Image, Princeton University Press, Princeton, NJ, 1965.
Rotter, J. B., Generalized expectancies for internal versus external control of reinforcement, Psychol.
Monog., 80, No. 1, 1966.
Rushton, J. P., The altruistic personality: evidence from laboratory, naturalistic and self-report
perspectives, in Development and Maintenance of Prosocial Behavior, Staub, E., Bar-Tal, D.
Karylowski, J., and Reykowski, J., Eds., Plenum, New York, 1984.
Rutkowski, G. K., Gruder, C. L., and Romer, D., Group cohesiveness, social norms, and bystander
intervention, J. Personal. Soc. Psychol., 44, 545, 1983.
Scheier, M. F. and Carver, C. S., Optimism, coping and health: assessment and implications of
generalized outcome expectancies, Health Psychol., 4, 219, 1985.
Scheier, M. F. and Carver, C. S., On the power of positive thinking: the benefits of being optimistic,
Cur. Direct. Psychol. Sci., 2, 26, 1993.
Scheier, M. F., Weintraub, J. K., and Carver, C. S., Coping with stress: divergent strategies of
optimists and pessimists, J. Personal. Soc. Psychol., 51, 1257, 1986.
Schwartz, S. H., Normative influences on altruism, in Advances in Experimental Social Psychology,
Vol. 10, Berkowitz L., Ed., Academic Press, New York, 1977.
Seligman, M. E. P., Learned Optimism, Alfred A. Knoff, New York, 1991.
Sherer, M., Maddox, J. E., Mercandante, B., Prentice-Dunn, S., Jacobs, B., and Rogers, R. W., The
self-efficacy scale: construction and validation, Psychol. Rep., 51, 663, 1982.
Sherrod, D. R. and Downs, R., Environmental determinants of altruism: the effects of stimulus
overload and perceived control on helping, J. Exp. Soc. Psychol., 10, 468, 1974.
Shoda, Y., Mischel, W., and Peake, P. K., Predicting adolescent cognitive and self-regulatory compe-
tencies from preschool delay of gratification, Dev. Psychol., 26, 978, 1990.
Simon, S. and Simon, R., Improving safety through innovative behavioral and cultural approaches,
presented at the National Safety Council Congress and Exposition, Orlando, FL, October 1992.
Staub, E., Helping a distressed person: social, personality, and stimulus determinants, in Advances
in Experimental Social Psychology, Vol. 7, Berkowitz, L., Ed., Academic Press, New York, 1974.
Staub, E., Positive Social Behaviors and Morality: Social and Personal Influence, Academic Press, New
York, 1978.
Strickland, B. R., Internal–external control expectancies: from contingency to creativity, Am.
Psychol., 44, 1, 1989.
Straumann, T. J. and Higgins, E. G., Self-discrepancies as predictors of vulnerability to distinct syn-
dromes of chronic emotional distress, J. Personal., 56, 685, 1988.
Tavris, C. and Wade, C., Psychology in Perspective, Harper Collins, New York, 1995.
Taylor, S. E., Health Psychology, 2nd ed., McGraw-Hill, New York, 1991.
Tucker, R. K. and Byers, P. Y., Factorial validity of Mertz’s psychological reactance scale, Psychol.
Rep., 61, 811, 1987.
Vollmer, T. R. and Iwata, B. A., Establishing operations and reinforcement effects, J. Appl. Behav.
Anal., 24, 279, 1991.
Waterman, A. S., On the uses of psychological theory and research in the process of ethical inquiry,
Psychol. Bull., 103, 283, 1988.
Weiss, H. M. and Knight, P. A., The utility of humility: self-esteem, information search and problem
solving efficiency, Organ. Behav. Hum. Perform., 25, 216, 1980.
Wells, L. E. and Marwell, G., Self-esteem, Sage, Beverly Hills, CA, 1976.
Wheeless, L. R., Wheeless, V. E., and Dickson-Markman, F., The relations among social and task
perceptions in small groups, Small Group Behav., 13, 373, 1982.
Wilson, J. P., Motivation, modeling, and altruism: a person x situation analysis, J. Personal. Soc.
Psychol., 34, 1078, 1976.
Wylie, R., The Self-Concept Vol. 1, University of Nebraska Press, Lincoln, 1974.
Zuckerman, M., Sensation Seeking: Beyond the Optimal Level of Arousal, Lawrence Erlbaum, Hillsdale,
NJ, 1979.
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 353

chapter sixteen

Increasing actively caring

behaviors
This chapter integrates principles and procedures from previous chapters to address the most criti-
cal question regarding the achievement of a Total Safety Culture. Namely, how can we increase
actively caring behavior throughout a work culture? Some conditions and interpersonal techniques
facilitate this behavior indirectly by benefiting self-esteem, empowerment, or a sense of belonging.
Other procedures can boost actively caring directly with certain activators and consequences. In
addition, the social influence principles of reciprocity and consistency can be applied to enhance the
caring behavior we desire. In sum, this chapter shows you how to increase the likelihood that people
will go beyond their normal routines to help keep people safe.

“In a Total Safety Culture . . . people ‘actively care’ on a continuous basis for safety.”

This quote from my 1994 article in Professional Safety (Geller, 1994, page 18) reflects the
ultimate vision of a Total Safety Culture. Everyone periodically goes beyond his or her per-
sonal routine for the safety and health of others. To meet this challenge, we need to find
ways to increase actively caring behaviors.
So how do we do this at work, in our homes, and the community at large? I hope you
can already provide some answers after reading Chapters 14 and 15, where I alluded to
some ways to increase actively caring behaviors. Here, I want to expand on earlier sugges-
tions and add more. We can classify these approaches as indirect or direct.
Indirect strategies for facilitating actively caring behavior follow from the theory and
principles discussed in Chapter 15. The actively caring model supported by research pro-
poses that certain person states inside people increase their willingness to look beyond self-
interests and consider the safety of others. Thus, conditions and procedures that increase
these states will indirectly increase the amount of actively caring among people. More
direct ways to increase these behaviors can be derived from behavior-based principles of
learning and social influence. These are discussed in the latter part of this chapter.

Enhancing the actively caring person states

Sometimes at seminars and workshops, I hear participants express concern that the actively
caring person-state model might not be practical. “The concepts are too soft or subjective,”
is a typical reaction. Employees accept the behavior-based approach because it is straight-
forward, objective, and clearly applicable to the workplace. However, person-based
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 354

354 Psychology of safety handbook

concepts like self-esteem, empowerment, and belonging appear ambiguous, “touchy-feely,”

and difficult to deal with. “The concepts sound good and certainly seem important, but
how can we get our arms around these ‘warm fuzzies’ and use them to promote safety?”
These person states are more difficult to define, measure, and manage than behaviors.
That is why I said early on in this text that it is more cost effective to work on behaviors first.
Whenever the ultimate outcome is behavior change, it is usually most efficient to deal
directly with behaviors. However, it is always important to consider people’s feelings when
designing behavior-change interventions. That is why I recommend against using negative
consequences whenever possible and why I offer ways to design intervention strategies that
account for subjective person-states like commitment, ownership, and involvement.
You see, we just cannot ignore the importance of how people feel about a behavior-
change intervention. For people to accept a behavior-change process and sustain the target
behaviors over the long term, we must consider internal person states while designing and
implementing an intervention. We can measure and evaluate person states through one-
on-one interviews, group discussions, or questionnaires. The 20-item questionnaire in the
previous chapter, for example, illustrates how the actively caring person states can be
measured systematically. Using more extensive surveys than this, researchers have
demonstrated significant positive relationships between these person states and actively
caring behavior. So, while measures of what occurs inside people—those person states—
are less objective and precise than systematic observations of behavior, such measurement
is possible and it can help gauge the impact of strategies to increase these states and
actively caring behavior.
After introducing the actively caring person-state model at my workshops, I often
divide participants into discussion groups. I ask group members to define events, situa-
tions, or contingencies that decrease and increase the person state assigned to their group.
Then, I ask the groups to derive simple and feasible action plans to increase their assigned
state. This promotes personal and practical understanding of the concept.
Feedback from these workshops tells me the concept may be soft, but it is not too hard
to grasp. Action plans have been practical and quite consistent with techniques used by
researchers. Also, there has been substantial overlap of practical recommendations—work-
shop groups dealing with different person states have come up with similar contributory
factors and action plans.
Some techniques suggested to increase self-esteem, for example, have been offered by
groups assigned to increase self-efficacy, personal control, and optimism. Such intercon-
nectedness of the person states is consistent with our research (Geller and Roberts, 1993;
Geller et al., 1996; Roberts and Geller, 1995). Distinct action plans, however, have emerged
from discussions of each particular person state, verifying the utility of teaching a five-state
actively caring model.
Let us take a look at what workshop participants have come up with for factors and
strategies regarding each of these person states.

Self-esteem
Factors consistently mentioned as shaping self-esteem include communication techniques,
reinforcement and punishment contingencies, and leadership styles. Participants suggest a
number of ways to build self-esteem, including

1. Provide opportunities for personal learning and peer mentoring.

2. Increase recognition for desirable behaviors and individual accomplishments.
3. Solicit and follow up on a person’s suggestions.
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 355

Chapter sixteen: Increasing actively caring behavior 355

• Accept - appreciate diversity

• Actively Listen - with verbal and nonverbal behavior
• Agree - with verbal and nonverbal behavior
• Admire - “attractive dress,” “nice tie”
• Appreciate - “please” (activator) and “thank you”
(consequence)
• Acknowledge - the achievements of others
• Approve - praise for good behavior
• Ask - for feedback, advice, opinions, etc.
• Attend - lend a helping hand
• Avoid Criticizing - it won’t be accepted anyway
• Argue Less - arguments are win/lose situations

Figure 16.1 Apply “A” strategies to increase others’ self-esteem.

Communication strategies. Figure 16.1 lists 11 words beginning with the letter “A”
that imply a specific verbal technique for increasing a person’s self-esteem. Each “A” word
suggests a slightly different communication approach, from stating simple words of agree-
ment, admiration, appreciation, and approval to acknowledging the achievement and indi-
vidual creativity of others through active listening and praise. It is also a good idea to argue
less and avoid criticizing. Arguments waste time and usually promote a win–lose per-
spective, and criticism always does more harm than good.
No one likes to be criticized, but some people are more resilient to negative feed-
back than others. People with a high and stable self-esteem have developed mechanisms
to protect themselves. Hence, these people take criticism in stride, protecting their
self-esteem for example, by just denying a mistake, rationalizing that the flack was
biased and unwarranted, or optimistically focusing on positive aspects of the verbal
exchange. On the other hand, some people are overly sensitive to critical comments. These
folks usually have a relatively low or unstable self-esteem and perceived daggers can
devastate them.
Correct with care. Figure 16.2 depicts a supervisor giving feedback to a worker quite
appropriately. He notes achievements before pointing out a flaw. This represents a com-
mendable strategy of making several deposits before making a withdrawal from that emo-
tional bank account (Covey, 1989). Still, this person is overwhelmed by the corrective
feedback and perceives it as an attack, possibly because his current self-esteem is low. He
could be “emotionally bankrupt.” Perhaps prior exchanges did not focus on the positive.
So despite an effective feedback technique, the net effect in Figure 16.2 is still negative.
What can you do to avoid or fix these situations?
First, recognize that the situation depicted in Figure 16.2 is possible. Observe body lan-
guage carefully to assess the impact of your words. If you note potential blows to self-
esteem, then do some damage control. You might re-emphasize the positive—”I see much
more good stuff here than bad.” You could indicate that such errors are not uncommon
and, in fact, you have made them yourself.
Above all, focus on the act, not the actor. Stress that the error only reflects behavior
that can be corrected, not some deeper character flaw. The worst thing you can do is be
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 356

356 Psychology of safety handbook

Figure 16.2 We selectively hear more negatives than positives.

judgmental. Do not come off as a judge of character, implying that a mistake suggests some
subjective personal attribute like “carelessness,” “apathy,” “bad attitude,” or “poor moti-
vation.” Figure 16.3 illustrates this nonbehavioral and destructive approach to giving feed-
back, which we find too often among family members.

Figure 16.3 Nonbehavioral feedback can be detrimental to self-esteem.

820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 357

Chapter sixteen: Increasing actively caring behavior 357

When offering corrective feedback, it is critical to be a patient, active listener. Allow the
person to make excuses, and do not argue about these. Resist the temptation. Giving
excuses is just a way to protect self-esteem, and it is generally a healthy response.
Remember, you already made your point by showing the error and suggesting ways to
avoid the mistake in the future (as discussed in Chapter 12 on coaching). Leave it at that.
If a person does not react to corrective feedback, it might help to explore feelings.
“How does this make you feel?” you might ask. Then listen empathetically to assess
whether self-esteem has taken a hit. You will learn whether some additional communica-
tion is needed to place the focus squarely on what is external and objective, rather than on
very subjective, internal states.

Self-efficacy
The feedback situation depicted in Figure 16.2 is clearly job related, so it is likely to have
greater impact on self-efficacy than self-esteem. As I discussed in Chapter 15, self-efficacy
is more situation specific than self-esteem, and so it fluctuates more readily. Job-specific
feedback should actually affect only one’s perception of what is needed to complete a par-
ticular task successfully. It should not influence feelings of general self-worth. Keep in
mind, though, that repeated negative feedback can have a cumulative effect, chipping
away at an individual’s self-worth. Then, it takes only one remark, perhaps what you
would think is innocuous and job-specific, to trigger what seems like an overreaction. The
internal reaction “He hates me” in Figure 16.2 suggests the negative statement influenced
more than self-efficacy—probably a fragile self-esteem.
Hence, it is important to recognize that our communication may not be received as
intended. We might do our best to come across positively and constructively, but because
of factors beyond our control, the communication might be misperceived. One’s inner state
can dramatically bias the impact of feedback. I am referring, of course, to variation in per-
sonal perception, or selective sensation, as discussed earlier in Chapter 5.
The communication process. To try to avoid these problems, let us deconstruct the
process of communication. As shown in Figure 16.4, a one-to-one exchange consists

Figure 16.4 The communication process consists of six key stages.

820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 358

358 Psychology of safety handbook

essentially of six components or steps—three per individual. There is plenty of room for
bias or misperception. First, the sender’s idea (or intention) is filtered through psycholog-
ical mechanisms influenced by autobiographical biases and person states. The message is
then affected by the sender’s verbal and nonverbal behavior during the exchange. I am
sure you have heard more than once about the critical importance of voice tone and body
language. These are affected by the sender’s prior experience and current person states.
The sender’s imperfect message, biased by perceptions and presentation mannerisms,
is transmitted to the receiver’s imperfect system. (Let us face it, none of us are perfect.) As
discussed in Chapter 5, we selectively receive information, sometimes filtering out mes-
sages we do not want to hear, and then we exert personal bias when comprehending the
distorted message. Thus, the last stage of the communication cycle depicted in Figure 16.4
is critical. Listeners need to check their understanding of the message, and senders need to
encourage clarification. In this way, constructive feedback allows for continuous improve-
ment of speaking and listening. Again, it is often important to check for changes in feeling
states and to make repairs when needed.
Achievable tasks. What makes for a “can do” attitude? Personal perception is the
key. A supervisor, parent, or teacher might believe he or she has provided everything
needed to complete a task successfully. However, the employee, child, or student might not
think so. Hence, the importance of asking, “Do you have what you need?” We are check-
ing for feelings of self-efficacy. This is easier said than done because people often hesitate
to admit they are incompetent. Really, who likes to say, “I can’t do it?” Instead, we try to
maintain the appearance of self-efficacy.
I have often found it necessary to ask open-ended questions of students to whom I give
assignments, in order to assess whether they are prepared to get the job done. In large
classes, however, such probing for feelings of self-efficacy is impossible. As a result, many
students get left behind in the learning process (frequently because they skipped classes or
an important reading assignment). As they get farther and farther behind in my class, their
low self-efficacy is supported by the self-fulfilling prophecy and diminished optimism.
Sometimes, this leads to “give-up behavior” and feelings of helplessness (Peterson et al.,
1993; Seligman, 1975). All too often, these students withdraw from my class or resign them-
selves to receiving a low grade.
Figure 16.5 reflects the need to focus on “small wins” (Weick, 1984) when assigning
tasks and communicating performance feedback. Of course, the kind of situation depicted
in Figure 16.5 requires one-on-one observation and feedback in which the individual’s ini-
tial competencies can be assessed. Then, successively more difficult performance steps can
be designed for the learner. The key is to reduce the probability that the learner will make
an error and feel lowered effectiveness or self-efficacy. Celebrating small-win accomplish-
ments builds self-efficacy and enables support from the self-fulfilling prophecy.
Suppose you were teaching a young child to put together a puzzle. How would you
apply the principle of reducing errors and celebrating small wins to build self-efficacy?
Would you lay the puzzle pieces on a table and then encourage each attempt to find the
right piece and put it in the right location? This might work if the child were experienced
with the puzzle, but this approach could be perceived as overwhelming by the child, and
a lack of initial success and perception of effectiveness could lead to frustration and give-
up behavior. In other words, the first performance step could be too large.
Analogous to the scenario in Figure 16.5, a puzzle-learner could experience initial suc-
cess and self-efficacy if he or she watched the teacher put all of the pieces of the puzzle
together except for the last piece. Then the child has a relatively easy task to do, and the
teacher and child can celebrate the completed picture. Notice the role of observation learn-
ing when the child watches the teacher pick and place the various puzzle pieces. A good
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 359

Chapter sixteen: Increasing actively caring behavior 359

Figure 16.5 Small, successive steps to success build self-efficacy.

teacher would include appropriate verbal description with each selection and placement of
the puzzle pieces, and ask the child for suggestions along the way.
On the next trial, the teacher would complete the puzzle again with the exception of
two or three puzzle pieces. The child would handle the next step successfully with no frus-
tration and, thus, would feel more effective at the task. Eventually, with a patient and
actively caring teacher, the child would put the entire puzzle together without experienc-
ing significant errors or debilitation to self-efficacy. The result is a “can do” attitude in this
situation, as well as perceptions of personal control and optimism.
I hope you can relate this story to situations in your workplace and consider ways to
apply the “reduced errors and small win” approach with your colleagues. How about at
home? Figure 16.6 illustrates this “small win” principle with a couch potato. Hopefully, the
“stretch” goals set for yourself and those you coach are greater than that shown here.
Personal strategies. Watson and Tharp (1997) suggest the following five steps to
increase perceptions of self-efficacy. First, select a task at which you expect to succeed, not
one you expect to fail. Then, as your feelings of self-efficacy increase, you can tackle more
challenging projects. A cigarette smoker who wants to stop smoking, for example, might
focus on smoking 50 percent fewer cigarettes per week rather than attempting to quit “cold
turkey.” With early success at reducing the number of cigarettes smoked, the individual
could make the criterion more stringent (like smoking no cigarettes on alternate days).
Continued success would lead to more self-efficacy.
Second, it is important to distinguish between the past and the present. Do not dwell
on past failures (recall the key point in Chapter 13 about shifting the focus of inter- and
intrapersonal conversation from past to future to present). Instead, focus on a renewed
sense of self-confidence and self-efficacy. Past failures are history—today is the first day of
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 360

360 Psychology of safety handbook

Figure 16.6 Improvement and self-efficacy develop from successive approximations and
small wins.

the rest of your life. The cigarette smoker might review past attempts to quit smoking, for
example, in order to decide on initial goals. (Recall the discussion of SMART goals in
Chapter 10). Choosing an attainable goal gives self-efficacy a chance to build when the goal
is reached.
Third, it is important to keep good records of your progress toward reaching your goal.
Our cigarette smoker should record the number of cigarettes smoked each day, and note
when the rate of smoking is 50 percent less for a week. This should be noted as an achieve-
ment, and then a new goal should be set. Focusing on your successes (rather than failures)
represents the fourth step in building self-efficacy.
The fifth step is to develop a list of tasks or projects you would like to accomplish and
rank them from easiest to most difficult to accomplish. Then, whenever possible, start with
the easier tasks. The self-efficacy and self-confidence developed from accomplishing the
less demanding tasks will help you tackle the more challenging situations on your list.
Focus on the positive. Many of the strategies I have presented for improving behav-
iors and person-states include a basic principle—focus on the positive. Whether attempt-
ing to build our own self-efficacy or that of others, success needs to be emphasized over
failure. Thus, whenever we have the opportunity to teach others or give them feedback, we
must look for small-win accomplishments and give genuine approval before commenting
on ways to improve. This approach is easier said than done.
Failures are easier to spot than successes. They stick out and disrupt the flow. That is
why most teachers give rather consistent negative attention to students who disrupt the
classroom, while giving only limited positive attention to students who remain on task and
go with the flow. Furthermore, many of us have been conditioned (unknowingly) to believe
negative consequences (penalties) work better than positive consequences (rewards) to
influence behavior change (Notz et al, 1987).
Figure 16.7 illustrates how natural variation in behavior can lead to a belief that penal-
ties have more impact on behavior than rewards. If people receive rewards for their
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 361

Chapter sixteen: Increasing actively caring behavior 361

Reward Reward
Above

Avg.
Perf.

Below
Penalty Penalty

Time

Figure 16.7 Natural variation of performance makes it easier to see a presumed behav-
ioral impact of negative rather than positive consequences.

superior performance and penalties for their inferior performance, natural variation in per-
formance (or regression to average performance) can give the impression that penalties are
more powerful than rewards. In other words, the natural tendency to do less than superior
work over time can mask the behavioral impact of a reward given for superior performance.
On the other hand, the natural tendency to improve following inferior performance can give
the impression that prior punishment improved the performance.
Imagine you see an employee demonstrating peak performance at a particular task.
This behavior deserves special commendation, and so you offer genuine words of appreci-
ation. The next time you see the individual you rewarded, you note less than optimal per-
formance. There are many reasons for the noticeable drop in performance, including the
fact that few people can perform at peak levels all of the time. There is natural regression
to average performance. You recall giving this person special recognition (a reward) earlier
for superior performance, and now you notice a decrement. Might you conclude that
rewards do not work to improve performance?
Now, imagine you observe an employee doing below-average work, and you decide to
intervene to improve performance. You issue a “progressive discipline” warning citation
which goes in the employee’s file. Sometime later, you observe this person performing
notably better. Although there are many reasons for the improvement, you naturally pre-
sume your punishment procedure the other day was responsible for this beneficial change.
Consequently, you develop the inaccurate belief that negative consequences work better
than positive consequences to change behavior.
In summary, normal circumstances make it relatively difficult to focus on the positive.
Mistakes are more noticeable than “go-with-the-flow” successes, and natural regression to
average performance can develop a faulty belief that negative consequences have more
behavioral impact than positive consequences. However, considering the impact conse-
quences have on internal person states (especially self-efficacy and self-esteem), a positive
consequence (like praise and social approval) is always preferable to a negative conse-
quences (like criticism or ridicule).

Personal control
Most people find a healthy imbalance between internal (personal) and external control.
They essentially believe, “I’m responsible for the good things that happen to me, but bad
luck or uncontrollable factors are responsible for the bad things” (Beck, 1991; Taylor, 1989).
Thus, people are apt to attribute injuries to rotten luck and beyond their control, especially
when they happen to them. (Recall our discussion of attributional bias in Chapter 6). This
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 362

362 Psychology of safety handbook

is the prime reason industrial safety must focus on process achievements rather than fail-
ure (injury) outcomes. To achieve a Total Safety Culture, people must believe they have per-
sonal control over the safety of their organization.
Employees at my seminars on actively caring have listed a number of ways to increase
perceptions of personal control, including

1. Setting short-term goals and tracking progress toward long-term accomplishment.

2. Offering frequent rewarding and correcting feedback for process activities rather
than only for outcomes.
3. Providing opportunities to set personal goals, teach others, and chart “small wins”
(Weick, 1984).
4. Teaching employees basic behavior-change intervention strategies (especially feed-
back and recognition procedures).
5. Providing people time and resources to develop, implement, and evaluate inter-
vention programs.
6. Showing employees how to graph daily records of baseline, intervention, and fol-
low-up data.
7. Posting response feedback graphs of group performance.

Figure 16.8 illustrates humorously a personal control perspective. Obviously, this is an

extreme and unrealistic scenario, but would it not be nice if people would attempt to take
personal control of safety issues at their industrial sites with the same passion and com-
mitment some individuals have for their golf game? I believe differences in perceived per-
sonal control for safety vs. golf are largely owing to contrasting scoring procedures.
Suppose you could not receive direct and immediate feedback about your golf game.
That is, each time you hit a golf ball you wore a blindfold and could not see where the ball
landed. Even when putting on the greens, you are blindfolded and cannot tell whether your
ball goes into the cup. Imagine also that you do not receive a score per hole or per game.
However, you do receive negative feedback whenever your ball lands in a sandtrap. Under
these circumstances, would you feel “in control” of your golf game? Would you attribute

Figure 16.8 Sometimes we try extra hard to exert personal control.

820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 363

Chapter sixteen: Increasing actively caring behavior 363

balls hit into sandtraps to personal control or just bad luck? Would you continue playing
golf or give it up for an activity in which you can experience greater personal control?
Of course, the golf scenario I have asked you to imagine is far-fetched, but is this not
the way it is for safety at many industrial sites? The primary evaluation tool used to rank
companies and determine performance appraisals and bonuses is an outcome number
(such as total recordable injury rate) which is quite remote from the daily plant processes
people have control over. Without a scoring system that focuses on controllable processes
(as discussed earlier in this Handbook), safety will be viewed as beyond personal control. An
injury is just “bad luck,” analogous to hitting a golf ball in a sandtrap while blindfolded.
Now, imagine you are playing golf without a blindfold and you are really playing well.
In fact, your scores indicate peak performance. Suddenly, it begins to storm. If you (unwisely)
continue to play, your golf game would deteriorate, and your score would well exceed par.
How would you evaluate this situation? Would you continue to feel a high degree of personal
control and self-efficacy or would you give up some control to uncontrollable factors?
My point here is that people need to distinguish between factors they can control on a
personal level and factors beyond their domain of influence. Similarly, Covey (1989) rec-
ommends we distinguish between our “Circle of Concern” and “Circle of Influence,” and
focus our efforts in the Circle of Influence. Thus, it is healthy to admit there are things we
are concerned about but have little influence over. Then, when negative consequences
occur outside our domain of personal influence, we will not attribute personal blame and
reduce our sense of self-efficacy, personal control, or optimism.
Obviously, we cannot have complete control over all factors contributing to an injury.
That is why I think it is wrong to say “all injuries are preventable.” However, there is much
we can do within our own domain of influence, and we can prepare for factors outside our
personal control. Thus, we take an umbrella to the golf course in case it rains, and we wear
personal protective equipment in case we are exposed to risks beyond our domain of per-
sonal control. Likewise, we protect our children from events beyond their control, as illus-
trated in Figure 16.9.

Figure 16.9 We cannot control everything!

820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 364

364 Psychology of safety handbook

The power of choice

About 30 years ago when I was conducting research in cognitive science, I carried out a
very simple experiment and obtained very simple results. The implications of the findings,
however, were immense and relevant to this discussion. Of the 40 subjects in this experi-
ment, one-half were shown a list of five three-letter words (cat, hat, mat, rat, bat) and asked
to select one. Then, after a warning tone, the selected word was presented on a screen in
front of the subject, and he or she pressed a lever as fast as possible after seeing the word.
The delay in milliseconds between the presentation of the word and the subject’s response
was a measure of simple reaction time.
This sequence of warning signal, word presentation, and subject reaction occurred for
25 trials. If a subject reacted before the stimulus word was presented, the reaction time was
not counted, and the trial was repeated. The experimental session took less than 15 min-
utes per subject.
The word selected by a particular subject was used as the presentation stimulus for the
next subject. Thus, this subject did not have the opportunity to choose the stimulus word.
As a result, the word choices of 20 subjects were assigned (without choice) to 20 other sub-
jects. Therefore, this simple experiment had two conditions—a “choice” condition (in
which subjects chose a three-letter word for their stimulus) and an “assigned” condition (in
which subjects were assigned the stimulus word selected by the previous subject). The
mean reactions of subjects in the “choice” group were significantly faster than those of sub-
jects in the “assigned” group.
I explained these results by presuming the opportunity to choose a stimulus word
increased motivation to perform in the reaction time experiment. I must admit, however, I
did not expect the differences to be as large as they were. How could the simple choice of
a three-letter word motivate faster responding in a simple reaction time experiment? I
frankly did not feel confident in my explanation for these results, and thus I did not pur-
sue publication of these data in a professional research journal. However, subsequent lab-
oratory studies verified these findings and were published (Monty et al., 1979; Monty and
Perlmuter 1975; Perlmuter et al., 1971).
From laboratory to classroom. About a year after the simple reaction time experi-
ment described previously, I tested the theory of “choice” as a motivator in the college
classroom. I was teaching two sections of social psychology; one at 8:00 a.m. Monday,
Wednesday, and Friday, and the other at 11:00 a.m. on the same days. There were about 75
students in each class. Instead of distributing a preprepared syllabus with weekly assign-
ments on the first day of classes, I distributed only a general outline of the course. This out-
line introduced the textbook, the course objectives, and the basic criteria for assigning
grades (a quiz on each textbook chapter and a comprehensive final exam on classroom lec-
tures, discussions, and demonstrations).
I told the 8:00 class they could choose the order in which the ten textbook chapters
would be presented, they could submit multiple choice questions for me to consider using
for the chapter quizzes, and they could hand in short answer and discussion questions for
possible use on the final exam. The 11:00 class received the order of textbook chapters
selected previously by the 8:00 class in an open discussion and voting process. Also, the
11:00 class was not given an opportunity to submit quiz or exam questions.
Thus, I derived “choice” and “assigned” classroom conditions analogous to the two
reaction-time groups I had studied earlier. Two of my undergraduate research assistants
attended each of these classes, posing as regular students, and systematically counted
amount of class participation. These observers did not know about my intentional choice-
assigned manipulations.
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 365

Chapter sixteen: Increasing actively caring behavior 365

From the day the students in my 8:00 class voted on the textbook assignments, this
class seemed more lively than the later 11:00 class. My perception was verified by the par-
ticipation records of the two classroom observers. Furthermore, the quiz grades, final exam
scores, and my teaching evaluation scores from standard forms distributed during the last
class period were significantly higher in the “choice” class than the “assigned” class.
Although several students from the 8:00 class submitted potential quiz and final exam
questions, I did not use any of these questions per se. Each class received the same quizzes
and final exam, and the final grades were significantly higher for the 8:00 class than the
11:00 class.
There are several possible reasons for the group differences, but I am convinced the
“choice” vs. “assigned” manipulation was a critical factor. I believe the initial opportunity
to choose reading assignments increased class participation. This increased involvement
fed on itself and led to more involvement, choice, and learning. The students’ attitudes
toward the class improved as a result of feeling more “in control” of the situation rather
than “controlled.” It is likely the “choice” opportunities in the 8:00 class were especially
powerful because they were so different from the traditional top-down classroom atmos-
phere experienced in other classes (and typified by my 11:00 class). In other words, the con-
trast with the students’ other courses made the choice opportunities especially salient and
powerful.
The implications of these “choice” vs. “assigned” findings in laboratory and classroom
investigations are far reaching. Indeed, the notion that “choice increases involvement”
relates to a number of motivational theories supported by psychological research (Steiner,
1970; White, 1959). Essentially, when people believe they have personal control over a sit-
uation, they are generally more motivated to achieve and get more involved. How can this
sense of personal control be increased? You guessed it. We can increase a belief in personal
control by increasing the number of “choice” opportunities in a situation.
Opportunities to choose lead to involvement, and more involvement leads to increased
perceptions of personal control. More personal control leads to more choice and more
involvement—and this continuous involvement cycle continues. This is a primary route to
feeling empowered.
Some people’s past experiences have made them less likely to develop a sense of per-
sonal control when “choice” is offered. They may mistrust the “choice” situation or lack
confidence in their ability to make something positive come out of a “choice” opportunity.
They might not feel comfortable with the added responsibility of “choice” and, thus, resist
the change implied by new “choice” potentials.
Usually the best way to deal with this resistance is to not confront it directly. I hope
enough other people will take personal control of their choice opportunities and eventu-
ally convince the resisters to get involved. (I address the topic of resistance more com-
pletely in Chapter 18.) It is important, however, that people with “choice” feel competent
to make appropriate decisions and, therefore, education and training might be needed.
The consequences of choice are critically important. If the subjects in my simple reac-
tion time experiment or my social psychology class did not believe their choices made a dif-
ference in the situation, the choice opportunities would not have made a difference in their
motivation. The reaction-time subjects saw me use their choice in the stimulus presenta-
tions, and the psychology students observed me change the class structure and process as
a function of their choices. When we see a consequence consistent with decisions from our
choice opportunities, we increase our trust of the people who gave us the power to choose.
We gain confidence in our abilities to take personal control of the situation.
From research to real world. An Exxon Chemical facility with 350 employees exem-
plifies the power of choice to make a difference for safety. The employees initiated an
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 366

366 Psychology of safety handbook

actively caring observation, feedback, and coaching process in 1992, reaping amazing ben-
efits for their efforts. In 1994, for example, 98 percent of the workforce had participated in
behavioral observation and feedback sessions, documenting a total of 3350 coaching ses-
sions for the year. A total of 51,408 behaviors were systematically documented on critical
behavior checklists, of which 46,659 were safe and 4,389 were at risk. Such comprehensive
employee involvement led to remarkable outcomes. At the start of their process in late
1992, the plant safety record was quite good (13 OSHA recordables for a Total Recordable
Incidence Rate of 4.11). This record improved to 5 OSHA recordables in 1993 (TRIR  1.70),
and in 1994 they had the best safety performance in Exxon Chemical with only one OSHA
recordable (TRIR  0.30). This enviable safety performance continued for the next seven
years, as illustrated earlier in Figure 12.18.
I have seen many companies subtantially improve safety performance with processes
based on the principles of behavior-based safety, but this plant holds the record for effi-
ciency in getting everyone involved and in obtaining exceptional results. I am convinced a
key factor was the employees’ “choice” in developing, implementing, and maintaining the
process. Choice has led to ownership. Here is what I mean.
Each month, employees schedule a behavioral observation and feedback session with
two other employees, who are safety observers. They select the task, day, and time for the
coaching session, as well as two individuals to observe them. Employees choose their
observers—and coaches—from anyone in the plant. At the start of their process, the num-
ber of volunteer safety coaches was limited to about 30 percent of the workforce, but today
everyone is a potential coach.
At first, some employees did not completely trust the process and resisted active
participation. Some tried to beat the system by scheduling observation and feedback ses-
sions at slow times when the chance of an at-risk behavior was minimal—such as when
watching a monitor or completing paperwork. Today, most employees choose to schedule
their coaching sessions during active times when the probability of an at-risk behavior is
highest.
Frequently, the observed individual uses the opportunity to point out an at-risk behav-
ior necessitated by the work environment or procedure, such as a difficult-to-reach valve,
a hose-checking procedure too cumbersome for one auditor, a walking surface made slip-
pery by an equipment leak, or a difficult-to-adjust machine guard. This often leads to
improved environmental conditions or operating procedures. Another benefit is an
increased perception of personal control for safety.

Optimism
As discussed in Chapter 15, optimism results from thinking positively, avoiding negative
thoughts, and expecting the best to happen. Anything that increases our self-efficacy
should increase optimism. Also, if our personal control is strengthened, we perceive more
influence over our consequences. This gives us more reason to expect the best. Again, we
see how the person states of self-efficacy, personal control, and optimism are clearly inter-
twined. A change in one will likely influence the other two.
Recall from Chapter 15 that simple events like finding a dime in a coin return, receiv-
ing a cookie, listening to soothing music, and being on a winning football team are suffi-
cient to boost optimism and willingness to actively care. It is not necessary for a person to
perceive personal control over a pleasant consequence for that consequence to build opti-
mism and possible actively caring behavior. I can think of no better reason for offering
words of appreciation and approval to others. As told by Art Buchwald in Figure 16.10, a
kind word can go a long way.
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 367

Chapter sixteen: Increasing actively caring behavior 367

I was in New York the other day and rode with a friend in a taxi. When we got out my friend said to the driver,
“Thank you for the ride. You did a superb job of driving.”
The taxi driver was stunned of a second. Then he said:
“Are you a wise guy or something?”
“No my dear man, and I’m not putting you on. I admire the way you keep cool in heavy traffic.”
Yeh,” the driver said and drove off.
“What was that all about?” I asked.
“I am trying to bring love back to New York,” he said. “I believe it’s the only thing that can save the city.”
“How can one man save New York?”
“It’s not one man. I believe I have made the taxi driver’s day. Suppose he has 20 fares — he’s going to be nice to
those 20 fares because someone was nice to him. Those fares in turn will be kinder to their employees or
shopkeepers or waiters or even their own families. Eventually the goodwill could spread to al least 1000 people.
Now that isn’t bad, is it?”
“But you’re depending on that taxi driver to pass your goodwill to others.”
I’m not depending on it, my friend said. “I’m aware that the system isn’t foolproof so I might deal with 10
different people today. If out of 10, I can make three happy, then eventually I can indirectly influence the attitudes
of 3000 more.”
“It sounds good on paper,” I admitted, “but I’m not sure it works in practice.”
“Nothing is lost if it doesn’t. I didn’t take any of my time to tell that man he was doing a good job. He neither
received a larger tip nor a smaller tip. If it fell on deaf ears, so what? Tomorrow there will be another taxi driver
whom I can try to make happy.”
“You’re some kind of nut,” I said.
“That shows you how cynical you have become. I have made a study of this. The thing that seems to be lacking,
besides money of course, for our postal employees, is that no one tells people who work for the post office what a
good job they’re doing.
“But they’re not doing a good job.”
“They’re not doing a good job because they feel no one cares if they do or not. Why shouldn’t someone say a kind
word to them?”
We were walking past a structure in the process of being built and passed five workmen eating their lunch. My
friend stopped. “ That’s a magnificent job you men have done. It must be difficult and dangerous work.”
The five men eyed my friend suspiciously.
“When will it be finished?”
“June,” a man grunted.
“Ah. That really is impressive. You must all be very proud.”
We walked away. I said to him, “I haven’t seen anyone like you since “The Man from La Mancha.”
“The most important thing is not to get discouraged. Making people in the city become kind again is not an easy
job, but if I can enlist other people in my campaign…”
“You just winked at a very plain looking woman, I said.
“Yes, I know,” he replied. “And if she’s a schoolteacher, her class will be in for a fantastic day.

Figure 16.10 A kind word can go a long way. Adapted from Adler and Towne (1990). With
permission.

Belonging
Here are some common proposals given by my seminar discussion groups for creating and
sustaining an atmosphere of belonging among employees.
• Decrease the frequency of top-down directives and “quick-fix” programs.
• Increase team-building discussions, group goal-setting and feedback, as well as
group celebrations for both process and outcome achievements.
• Use self-managed or self-directed work teams.
When groups are given control over important matters like developing a safety obser-
vation and feedback process or a behavior-based incentive program, feelings of both
empowerment and belonging can be enhanced. When resources, opportunities, and talents
enable team members to assert, “We can make a difference,” feelings of belonging occur
naturally. This leads to synergy, with the group achieving more than could be possible from
members working independently (see Figure 16.11).
Social psychologists have studied conditions that influence the productivity of people
working alone vs. in groups. Results of this research are relevant to increasing actively caring
behavior, because conditions that spur on group productivity and synergy also boost feelings
of empowerment and belonging. When group members experience synergy, they especially
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 368

368 Psychology of safety handbook

Figure 16.11 Higher goals can be reached through synergy.

appreciate the group process and project pride in their output. This feeds a win–win per-
spective and more perceptions of group empowerment and belonging. Obviously, a Total
Safety Culture requires teamwork, belonging, and synergy. Yet, as you will see below, tradi-
tional approaches to safety management can throw a wet blanket on this process.
Social loafing: the whole can be less than the sum of its parts. First, I need to digress
for a moment to describe a phenomenon known as social loafing. In the 1930s, social psy-
chologists measured the effort people exerted when pulling a rope in a simulated tug-of-
war contest (Dashiell, 1935). Researchers measured the force exerted by each of eight
subjects when pulling on the rope alone and when pulling on the rope as a team. If the
group effort was greater than the sum of the individual efforts, synergy would have been
shown. What do you think happened?
The eight subjects worked harder alone than as a team. In fact, the total pulling force
of the group was only about one-half the total of the eight individual efforts. This finding
was not a fluke. This phenomenon, termed social loafing, has been demonstrated in more
than 50 experiments conducted in the United States, India, Thailand, Japan, and Taiwan
(Gabrenya et al., 1983). For example, blindfolded subjects were asked to shout or clap their
hands as loud as they could while listening through headphones to the sound of loud
shouting or hand clapping. Subjects told they were doing the clapping or shouting with
others produced about one-third less noise than when they thought they were performing
alone (Latané et al., 1979). So what contributed to the social loafing?
Identifiability: Can I be evaluated? When subjects were told the sound equipment
could measure their individual clapping as well as the team effort, social loafing disap-
peared (Williams et al., 1981). Thus, when subjects believed their personal effort could be
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 369

Chapter sixteen: Increasing actively caring behavior 369

identified, they worked just as hard in the group as alone. When are individual efforts iden-
tifiable in the workplace? Are there more opportunities to recognize individual effort in
production and quality aspects of the job than for safety?
One thing is certain. The typical approach to measuring safety achievement—by
counting number of hours or days without an OSHA recordable or lost-time injury—does
not provide much opportunity for individual recognition, unless a worker is injured. In
this case, the recognition is failure-focused, which motivates behavior through fear rather
than desire. So what happens? The injury is often covered up to avoid individual embar-
rassment or punishment. That is not the behavior we want to see.
Responsibility: Am I needed on this job? Social loafing has been shown to increase
when a person’s efforts are duplicated by another person. In one experiment, for example,
subjects worked in groups of four to report whenever a dot appeared in a particular quad-
rant of a video screen (Harkins and Petty, 1982). Social loafing did not occur when subjects
were assigned their own quadrant to watch. However, social loafing was found when all
four subjects were asked to observe all four quadrants. Social loafing can be expected to
increase as people’s sense of personal responsibility decreases. When people believe their
individual contribution is important, perhaps indispensable, for the team effort, synergy is
more likely than social loafing.
Do your safety processes allow individuals to feel a sense of personal responsibility for
safety achievements? A safety system that is evaluated only by numbers of injuries or
workers’ compensation costs does not provide individuals with feedback regarding their
contributions. A safety monitoring process that tallies the safe versus at-risk observations
of team members, however, can build a sense of individual responsibility for a group’s
safety record. Likewise, when teams develop interventions to reduce hazards and increase
safe behaviors, personal responsibility for team progress is realized and social loafing is
reduced. Tracking the success of intervention efforts verifies perceptions of empowerment
and promotes feelings of group belonging.
Interdependence: We need each other. Covey (1989) reminds us that we all come
into this world completely dependent on others to survive. As we mature and learn, we
reach a level of independence—we strive to achieve success or avoid failure on our own,
but higher levels of achievement and quality of life are usually reached after we develop a
perspective of interdependence and act accordingly. According to Covey, acting to achieve
interdependence means we actively listen (Habit 5: “Seek first to understand, then to be
understood”) and develop relationships or contingencies with people that reflect positive
outcomes for everyone involved (Habit 4: “Think win–win”). Consistently practicing these
habits leads to synergy (Habit 6).
Research on social loafing supports Covey’s idea, and suggests approaches for facili-
tating synergistic teamwork. Social loafing is reduced when group members know each
other well and agree on common goals (Williams et al., 1981). Assessing personal perform-
ance against an objective standard or the performance of other team members also dimin-
ishes social loafing. This happens even when evaluations are not publicized and there are
no external consequences (Harkins and Szymanski, 1984; Szymanski and Harkins, 1987).
In safety management, this can be achieved by tracking individual contributions to
safety observations and intervention processes. For example, team members can conduct
regular audits of safe vs. at-risk conditions and behaviors, providing invaluable data for
calculating group percentages of safe conditions and behaviors. Work teams can also
develop intervention strategies to correct environmental hazards or motivate safe work
practice and, then, devise a process to monitor individual contributions to the process.
How large is your group? Most of us feel a sense of belonging with our immediate
family and, as discussed in Chapter 15, we usually do not hesitate to actively care for the
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 370

370 Psychology of safety handbook

safety and health of our immediate family (see Figure 15.7). In the same vein, the more
interdependence and belonging we experience at work, the more likely we are to look out
for coworkers’ safety. How large is your work group?
Obviously, a large “family” means more people come under the protective wing of
actively caring behavior. So it makes sense to have an extended family at work. In a Total
Safety Culture, everyone actively cares for everyone else’s safety. In effect, everyone
belongs to one group. Now what are the barriers to an extended “work family”?
Figure 16.12 depicts an obstacle, or win–lose perspective, that I come across all too
often. The “we–they” attitude spun off by traditional management-labor differences often
makes for a dysfunctional “family.” It seems some unions attempt to justify their existence
by focusing on disagreement, conflict, and mistrust between management and labor. For
its part, management supports this “we–they” split with an alienating communications
style that asserts its ultimate power and control.
I have seen management memos, for example, that might have been well-intentioned,
but were written in top-down, control language that sounded like an adult talking to a
child. See, for example, the management memos listed in Figure 16.13 which were pur-
ported to have actually been distributed in an occupational setting. They were the finalists
for the “Dilbert Award” in 1999.
Through actively caring, and enhancing self-esteem, empowerment, and belonging,
we can bring down the “us vs. them” walls that entrap a work culture. Active caring
spreads mutual trust and interdependence throughout the culture. In a Total Safety
Culture, everyone benefits from each individual’s efforts.
Does actively caring imply the elimination of unions? No, but it might suggest altered
visions and mission statements for organized labor groups. Labor unions can certainly help
enhance the five person states that facilitate actively caring behavior. To do this, they need
to work with management from a win–win perspective that appreciates interdependence
and the power of synergy.
At the same time, managers need to relinquish their hold on the “control buttons” of
operations and processes that workers can manage themselves, perhaps through self-
directed work teams. A truly “empowered work force” is one trusted by managers and
supervisors to get the job done without direct supervision. Obviously, this cannot happen
overnight, but a solid foundation is cemented when the five actively caring person states
are strengthened.

Figure 16.12 Win–lose competition inhibits teamwork and synergy.

820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 371

Chapter sixteen: Increasing actively caring behavior 371

• “… What I need is a list of specific unknown

problems we will encounter.”
• “… This project is so important, we can’t let
things that are more important interfere with it.”
• “… Teamwork is a lot of people doing what I say.”
• “… We know that communication is a problem,
but the company is not going to discuss it with
the employees.”
• My boss asked me to submit a status report … I
asked him if tomorrow would be soon enough.
He said, “If I wanted it tomorrow, I would have
waited until tomorrow to ask for it.”

Figure 16.13 Some memos promote a “we–they” split between management and line
workers.

Directly increasing actively caring behaviors

You can treat actively caring behavior just like any other target behavior. Many interven-
tions that increase the occurrence of safe work behaviors can be used to boost the frequency
of actively caring behaviors. The four chapters in Section 4 covered principles and proce-
dures for directly influencing behavior. You will recall that the techniques were classified
as activators and consequences, with activators considered directive or instructional, and
consequences being motivational. Let us take up that discussion again because it applies to
actively caring behavior.

Education and training

There is some evidence that educating people about the barriers inhibiting actively caring
behavior, as detailed in Chapter 14, will increase acts of caring. Beaman and colleagues
(1978) randomly assigned students to listen to either a lecture about the bystander inter-
vention research conducted by Latané and Darley (1970) or a lecture on unrelated topics.
Two weeks later, the students participated in a presumed unrelated sociology experiment.
They encountered a student lying on the floor. He could have been hurt, of course, but then
he could have only been resting after studying all night for an exam. An accomplice of the
researchers posed as another participant and acted unconcerned.
Most of the students followed the lead of the accomplice and did not stop to inquire
whether the student needed help, but the lecture about bystander intervention appeared to
make a difference. Of those who heard this lecture, 43 percent stopped to help the victim,
compared to only 25 percent of the students who heard an unrelated lecture.
Higher education. The impact of education should be even more dramatic if some of
the concepts discussed in Chapter 14 are taught. It seems particularly useful to explain that
actively caring behavior can be proactive or reactive, direct or indirect, and focused on the
environment, internal person-states, and behavior. Also, it should be taught that direct,
proactive, and behavior-focused active caring is most challenging—and most useful to pre-
vent injuries.
Ferrari and I have advocated getting college students actively involved in research
“projects focusing on improving the quality of life of others . . . (thereby) nurturing a sense
of personal growth and belonging to a larger community” (Ferrari and Geller, 1994, page
12). At junior and community colleges and a large university, Ferrari has involved under-
graduate students in evaluating interventions to increase blood donations, promote the use
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 372

372 Psychology of safety handbook

of child safety-belt use in shopping carts, encourage young adults to lessen their risk of
hearing problems from high-volume headsets, and to assess the rewarding consequences
from helping AIDS sufferers.
As a faculty member at Virginia Tech for more than 30 years, I have involved numer-
ous graduate and undergraduate students in community and organizational projects to
increase recycling, vehicle safety-belt and child safety-seat use, and safe work practices;
and to decrease littering, shoplifting, energy and water consumption, alcohol and other
drug abuse, and alcohol-impaired driving.
Ferrari and I are certainly not unique in involving university students in field studies
designed to help others. Many of our colleagues have conducted applied research that
involves students as intervention agents to promote proactive actively caring behavior
(see, for example, numerous studies published in Greene et al., 1993; Seligman, 1987). My
point is we need more of this kind of instruction, not only at colleges and universities, but
also at elementary and high schools and within civic and social organizations such as Girl
Scouts, Boy Scouts, 4-H clubs, and church groups. Of course, we need similar education
and training in industry.
Promoting actively caring education and participation among children and adults can
dramatically increase the number of caring people in our society. Participative education is
particularly powerful because it follows the classic Confucian principle.

Tell them and they’ll forget . . . .

Show them and they’ll remember . . . .
Involve them and they’ll understand . . . .

Education at home. Parents have profound impact on their children’s current

and future acts of caring. As discussed in Chapter 8, children learn continuously and indi-
rectly by watching their parents (observational learning), and their behavior is directly
influenced by the consequences they receive (operant learning). Children are more
apt to actively care for others when their mothers are warm, sympathetic, and empathic,
and when their fathers are perceived as generous and compassionate (Rutherford
and Mussen, 1968; Zahn-Waxler et al., 1979). Other studies find that people are
more actively caring as adults when their parents have been open and tolerant of other
people, teaching them they are part of humanity in general (collectivism) rather than
some elite or special group (Staub, 1990, 1992). Moreover, children were observed
to perform kind acts when they saw an adult set an example two months earlier
(Rushton, 1975).
Parents should promote active caring among their children. Society will reap rewards
for the seeds you plant now. Staub (1975, 1979) found that children who made toys for poor
hospitalized children or taught skills to younger children were later more likely to exhibit
actively caring behavior. Keep in mind, though, it is important to provide a rationale for
actively caring. Stress the importance of empathy and caring (Eisenberg, 1992). In other
words, give children internal attributions for their behavior. When you see them perform-
ing acts of caring, praise them as helpful and generous individuals.
Grusec and Redler (1980) conducted an interesting and instructive experiment to com-
pare the impact of internal attribution and external praise on children’s short-term and
long-term actively caring behavior. First, children were persuaded to share some tokens,
which could be exchanged for prizes, they had won previously in a game with poor chil-
dren. Following this induced actively caring behavior, the experimenter either verbally
praised the child—“It was good that you gave your tokens to those poor children . . . that
was a nice and helpful thing to do”—or offered an internal attribution about the child’s
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 373

Chapter sixteen: Increasing actively caring behavior 373

behavior—“I guess you’re the kind of person who likes to help others whenever you
can . . . you are a . . . helpful person.”
The researchers assessed the amount of actively caring behavior from the two groups
of children immediately after these two verbal consequences and then after one week and
three weeks. Initially, there were no differences between groups in amount of actively car-
ing behavior, measured by the number of additional tokens given to poor children, but one
week later, the group given an internal attribution for their earlier behavior made more
playhouse roofs for the teacher than the group given only praise for their token contribu-
tions. Three weeks later, subjects in the internal attribution group collected significantly
more craft material for sick children than did the children in the praise only group. Helping
children write an internal script that they are an actively caring person will likely increase
their future behavior in the absence of external prompts, models, or rewards.
Researchers have also shown that actively caring behavior can be inhibited by negative
consequences (Hoffman, 1975; Zahn-Waxler et al., 1983). When children act in noncaring or
selfish ways, it is important to explain why this behavior is antisocial and inappropriate.
Do not spank a selfish child for not sharing, but urge him or her on with words like “Please
share your toys with Sherry. You made her cry and now she is unhappy. You can be a helper
and make her feel better.”

Consequences for actively caring

Rewards should increase actively caring behavior. This comes from the basic operant learn-
ing principle that behavior increases following positive consequences. Indeed, researchers
have demonstrated increases in active caring following monetary rewards (Wilson and
Kahn, 1975) and social approval (Deutsch and Lamberti, 1986). Even a simple “thank you”
can be effective.
In a study by McGovern et al., (1975), a female research accomplice asked male subjects
to endure a brief electric shock for her. For one group, she responded with a “thank you” if
the student agreed. For a control group, she said nothing if they agreed. After the initial
shock trial, subjects who received a thank you showed more actively caring behavior
throughout the experiment than did subjects in the control group.
Positive vs. negative consequences. In a field study by Moss and Page (1972), indi-
viduals on a busy street were approached by a researcher and asked for directions to a local
department store. When the individual agreed to give directions, the researcher varied the
consequences for this act of caring. For one group of subjects, the researcher smiled and
said, “Thank you very much, I really appreciate this.” For another group of subjects,
the researcher gave a negative consequence by rudely interrupting and remarking, “I
can’t understand what you’re saying. Never mind, I’ll ask someone else.” For a control
condition, the researcher listened to the directions and gave neither a positive nor a nega-
tive consequence.
A short time after giving directions, the subjects encountered another person, a
research accomplice, who “accidentally” dropped a small bag. Of those subjects rewarded
with a “thank you” for their previous actively caring behavior, 93 percent stopped to help
this person. In contrast, only 40 percent of those subjects who received a negative conse-
quence for giving directions earlier offered to help. Of those in the control group, 85 per-
cent stopped to help.
You can see the importance of responding positively to an individual displaying
actively caring behavior, but it is perhaps even more important not to respond negatively
when observing a caring act. A negative reaction could make that person avoid a subse-
quent opportunity to actively care for safety. I made a similar point when discussing safety
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 374

374 Psychology of safety handbook

coaching in Chapter 12. A person’s reaction to a safety coach could determine whether or
not that person goes out of their way to coach again for safety.
The “Actively Caring Thank-You Cards” described in Chapter 11 (see Figures 11.16 and
11.17) are a rather generic technique to boost caring in the workplace. You will recall that
employees used these cards to thank their colleagues for going beyond the call of duty for
the safety of others (Roberts and Geller, 1995). Some cards included a space for the inter-
vention agent to define the act of caring. Some cards could be exchanged for inexpensive
rewards, like a beverage in the company cafeteria. Other cards had peel-off stickers for
public display. In some applications, each card was worth some amount of money—
25 cents or 50 cents—toward corporate contributions to local charities.
The contributions to charity idea promotes actively caring in more than one way. First,
those more ready to actively care, perhaps because of high levels of the internal person
states, will find these “charity thank-you cards” more rewarding (Roberts and Geller, 1995).
These people not only set admirable examples for others, they are also likely to become
champions of the process and the actively caring concept. Also, public displays of the char-
itable contributions can help get others involved in the process.
A child researcher. Besides our own research, I could find only one other study that
used the opportunity to help other people as the reinforcing consequence for an actively
caring behavior. The research was conducted by the 10-year-old grandson of the eminent
behavioral scientist and teacher, Dr. Fred S. Keller. Jacob Keller (1991) wanted to increase
participation in the curbside recycling program in his neighborhood. After counting the
percentage of homes putting out recyclables along two streets, Jacob intervened on a street
by delivering weekly handwritten notes (activators) to each of the 44 homes.
Jacob’s first note notified residents that he was monitoring recycling participation and,
if recycling improved, a local grocery store “has offered to give two $10 gift certificates
to a homeless shelter” (Keller, 1991, page 618). The next three weekly notes specified
the amount of improvement and thanked the residents for their behavior. The improve-
ment was significant, increasing from 34 percent participation before the intervention to
53 percent during the last two intervention weeks. Participation on the street that did not
receive the intervention (40 homes) remained quite stable at about 35 percent throughout
the study.
This community research project is noteworthy on two accounts. Sure, it illustrates the
impact of rewards on actively caring behavior with an environment focus but, more impor-
tantly, it was conducted competently by a 10-year-old, who also documented it for profes-
sional publication. Children throughout communities should be empowered to conduct
such projects. Their “small wins” will add up to substantial benefits to our environment
and its inhabitants. Even more significantly, these kids will be introduced to the actively
caring concept and, perhaps, experience the self-rewarding power of this behavior. This
will have a multiplying effect, leading to involvement in other actively caring projects.

The reciprocity principle

Reciprocity: “Do for me and I’ll do for you”

Some sociologists, anthropologists, and moral philosophers consider reciprocity a univer-
sal norm that motivates a good deal of interpersonal behavior (Cialdini, 1993; Gouldner,
1960). Simply put, people are expected to help those who have helped them. You can expect
people to comply with your request if you have done a favor for them. This is the principle
behind Covey’s (1989) claim that we need to make deposits in another person’s emotional
bank account before making a withdrawal.
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 375

Chapter sixteen: Increasing actively caring behavior 375

I witnessed the reciprocity norm rather dramatically when I worked in the Virginia
prison system in the mid-1970s. Several inmates used gifts and other forms of personal
assistance to influence other inmates. I knew an inmate, for example, who went to great
lengths to place a gift in the cell of a new inmate he wanted to influence. Accepting such a
gift meant that the target inmate, actually a victim, was now obliged to return a favor in
order to “save face.”
In a similar vein, my father-in-law had a favorite expression whenever someone
offered to do him a favor. He would say, “I don’t want to be obligated.” Without knowing
the principle, he was saying that he did not want the favor because he would feel a duty to
reciprocate.
Have you ever felt a little uncomfortable after someone did you a favor? I certainly
have. I interpret my discomfort as the reciprocity principle in action. Another person’s kind
act makes me feel pressured to respond in kind. What does this mean for safety? I think it
means we should look for opportunities to go out of our way for another person’s safety.
Doing this, we increase the likelihood they will help when we need them.
The workplace setting might enhance chances for reciprocity. Research shows that
people are most likely to pay back individuals they expect to see again (Carnevale et al.,
1982). Also, the more actively caring behavior one receives, the more such behavior they
feel obligated to return (Kahn and Tice, 1973). So, when we actively care frequently for the
safety of coworkers we see often, we will obligate them to return in kind. This actually has
far-reaching benefits.
Suppose you “do the right thing” for a coworker’s safety, but you are not available to
receive his or her reciprocal act of caring. Will that person be more likely to act on behalf of
another individual? Yes, according to informative research by Berkowitz and Daniels
(1964). Subjects in one experimental group received favors, while other subjects did not.
Later, the person who did the favor was unavailable, but another individual (a research
accomplice) was in apparent need of help. Subjects who had received the earlier favor were
significantly more likely to assist than those who had not.
The reciprocity principle also works among strangers or when there is no expectation
of future interaction (Goranson and Berkowitz, 1966). Two researchers (Kunz and
Woolcott, 1976) demonstrated this by mailing Christmas cards to a sample of total
strangers. To the researchers’ surprise, many of these strangers (about 20 percent)
responded by sending holiday greeting cards to the return address. They went beyond the
call of duty to return a favor, even though they did not know who they were sending the
card to.
Gifts aren’t free. Has someone snared your attention to hear a sales pitch after giv-
ing you a free gift? Have you ever felt obligated to contribute to a charity after receiving
gummed individualized address labels and a stamped envelope for your check? Ever pur-
chase food in a supermarket after eating a free sample? Do you feel obliged to buy some-
thing after using it for a 10-day “free” trial period? As illustrated in Figure 16.14, how about
feeling obligated to do the dishes after someone else fixes the meal?
If you answer “yes” to any of these questions, it is likely you have been influenced by
the reciprocity principle. Many marketing or sales-promotion efforts count on this “free
sample” gimmick to influence purchasing behavior. A classic experiment by Regan (1971)
gives credibility to this ploy. During a break in an “art appreciation” experiment in which
pairs of individuals rated paintings, one subject (actually the experimenter’s accomplice)
performed an unexpected favor—returning from the break with a Coke for the other sub-
ject. For a control condition, the accomplice returned from the break empty handed.
At the end of the experiment, the accomplice asked the subject for a favor. He was sell-
ing raffle tickets for a new car and would win $50 if he sold the most tickets. The subjects
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 376

376 Psychology of safety handbook

Figure 16.14 The dating game is frequently influenced by reciprocity.

who had received the unexpected favor purchased twice as many raffle tickets as those in
the control group. This sounds fair because of the reciprocity norm, but soft drinks cost
only 10 cents at the time of this study, and the raffle tickets were 25 cents. Attempting to
reciprocate, the students purchased an average of two tickets each. The free drink was not
“free.” It cost them five times as much than if they had paid for it themselves.
Does this justify distributing free safety gifts, such as pens, tee-shirts, caps, cups, and
other trinkets? Yes, to some extent, but you have to take into account perceptions. How spe-
cial is the gift? Was the gift given to a select group of people, or was it distributed to every-
one? Does the gift or its delivery represent significant sacrifice in money, time, or effort?
Can it be purchased elsewhere, or does its safety slogan make it special?
A “special” safety gift—as perceived by the recipient—will trigger more acts of caring
in response. Remember, too, that the way a gift is bestowed can make all the difference in
the world. The labels and slogans linked with it can influence the amount and kind of
responsive action. If the gift is presented to represent the actively caring safety leadership
expected from a “special” group of workers, a certain type of reciprocity is activated.
People will tell themselves they are considered leaders, and they need to justify this label
by going beyond the call of duty for others. If they have learned about the various cate-
gories of actively caring behavior, they know the best way to lead is by taking action that
is direct, proactive, and focused on supporting safe behavior or correcting at-risk behavior.
Door-in-the-face: start big and retreat. Suppose the plant safety director pulls you
aside and asks you to chair the safety steering committee for the next two years. This
request seems outrageous, given your other commitments and the fact you never even
served on the committee. You say, “Thanks, but no thanks!” The safety director says he
understands, and then asks if you would be willing to serve on the committee. Research
shows that because the safety director “backed down” from his first request, you will feel
subtle pressure to make a similar concession—to reciprocate—and agree to the second, less
demanding assignment.
Cialdini and his associates (1975) posed as representatives of the “County Youth
820049_CRC1_L1540_CH16 11/10/00 2:05 PM Page 377

Chapter sixteen: Increasing actively caring behavior 377

Counseling Program” and asked students if they would be willing to chaperone juveniles
on a trip to the zoo. Most of the students (16.7 percent) refused, but the researchers got
50 percent compliance when they preceded this minor request with a major one—to serve
as counselors for juvenile delinquents for two hours a week over a two-year period. When
the researchers accepted the subjects’ refusal of their extreme request and retreated to a
more reasonable one, one-half of the subjects apparently felt obligated to reciprocate and
agree to the less demanding favor.

Commitment and consistency

Cialdini (1993) refers to commitment and consistency as an influence mechanism lying
deep within us, directing many of our actions. It reflects our motivation to be, and appear
to be, consistent. “Once we make a choice or take a stand, we will encounter personal and
interpersonal pressures to behave consistently with that commitment” (Cialdini, 1993,
page 51). Cialdini suggests the pressure comes from three basic sources.

• Society values consistency within people.

• Consistent conduct benefits daily existence.
• A consistent orientation allows us to take shortcuts when processing information
and making decisions. We do not have to stop to consider everything involved;
instead we fall back on our prior commitment or decision and act accordingly to
remain true to ourselves.

Public and voluntary commitment. The “Safe Behavior Promise Card” described in
Chapter 10 derives its power to influence from the commitment and consistency principle.
When people sign their name to a promise card they commit to behaving in a certain way.
Then, they act in a way that is consistent with their commitment.
Commitments are most effective, or influential, when they are visible, require some
amount of effort, and are perceived to be voluntary, not coerced (Cialdini, 1993). It makes
sense, then, to have employees state a public rather than private commitment to actively
care for safety and to have them sign their name to a promise card rather than merely raise
their hand. It is critically important for those making a pledge to believe they did it volun-
tarily. In reality, decisions to make a public commitment are dramatically influenced by
external activators and consequences, including peer pressure. However, if people sell
themselves on the idea that they made a personal choice, consistency is likely to follow the
commitment.
Figures 16.15 and 16.16 illustrate a public commitment intervention implemented at a
safety seminar for supervisors, safety leaders, and maintenance personnel of Delta
Airlines. After giving a keynote address on the concept of actively caring for the safety and
health of others, I signed my name to a “Declaration of Interdependance” as a symbol of
commitment to look out for the safety of others (Figure 16.15). Then, I urged the audience
to follow suit.
The social context was partly responsible for the great number of individuals signing
the declaration (Figure 16.16) which is now prominently displayed at the employees’ work-
site. The public and voluntary nature of the commitment request contributed to the effec-
tiveness of this exercise to activate awareness and the development of relevant action
plans.
Foot-in-the-door: start small and build. This strategy follows directly from the com-
mitment and consistency principle. To be consistent, a person who follows a small request
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 378

378 Psychology of safety handbook

Figure 16.15 The author signs a “Declaration of Interdependence.”

Figure 16.16 Delta Airlines employees sign a “Declaration of Interdependence.”

will likely comply with a larger request later. During the Korean War, the Chinese commu-
nists used this technique on American prisoners by gradually escalating their demands,
which started with a few harmless requests (Schein, 1956). First, prisoners were persuaded
to speak or write trivial statements. Then they were urged to copy or create statements that
criticized American capitalism. Eventually, the prisoners participated in group discussions
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 379

Chapter sixteen: Increasing actively caring behavior 379

of the advantages of communism, wrote self-criticisms, and gave public confessions of

their wrong-doing.
Research has found this “start small and build” strategy succeeds in boosting product
sales, monetary contributions to charities, and blood donations. In a classic study,
researchers posed as volunteers in a local traffic safety campaign and went door-to-door to
ask residents permission to install a large ugly sign in their front yards with the message
“Drive Carefully.” Only about 17 percent consented. However, of those residents who two
weeks earlier had signed a safety legislation petition or had agreed to display a three-inch
square “Be a Safe Driver” sign in their home, 76 percent allowed the large sign to be
installed (Freedman and Fraser, 1966).
The “Safe Behavior Promise Card” uses this principle. After people sign on to perform
a certain behavior for a specified period of time, such as “Buckle vehicle safety belts for one
month,” “Use particular personal protective equipment for two months,” or “Walk behind
yellow lines for the rest of the year,” they are more likely to actually perform the target
behavior.
This “foot-in-the-door” technique only works when people go along with the first
small request. If a person says “No” right away, he or she might find it even easier to resist
subsequent, more important requests. So, if your first call for actively caring behavior is
shot down, you did not start small enough. Be prepared to retreat to something less
demanding and build reciprocity from there.
Throwing a curve ball: raising the stakes later. This technique of “throwing a curve”
occurs when you persuade someone to make a decision or commitment because of the rel-
atively low stakes involved. Then you raise the level of involvement required.
For example, being a safety steering committee member is not asking too much if meet-
ings are held only once a month. However, after attending the first two safety meetings, the
stakes are raised. More meetings are requested for a special project. Because of the com-
mitment and consistency principle, the individual will likely stick with the original deci-
sion and remain an active member of the committee.
Cialdini and colleagues (1978) used this technique to get college students to sign-up for
an early morning experiment on “thinking processes.” During solicitation phone calls, the
7:00 a.m. start time was mentioned up-front for one-half of the subjects. Only 24 percent
agreed to participate. For the others, the caller first asked if they wanted to participate in
the study. Then, after 56 percent agreed, the caller threw them the “curve” and said the
experiment started at 7:00 a.m. The subjects had the chance to change their minds, but none
did. Plus, 95 percent showed up at the 7:00 a.m. appointment time. Practically every one of
them showed consistency and kept their commitment—in spite of being thrown a curve.
The effectiveness of this technique has been shown in several other studies (see, for exam-
ple, Brownstein and Katzev, 1985; Burger and Petty, 1981; Joule, 1987).
This procedure is similar to the foot-in-the-door technique: a larger request occurs after
you get agreement with a smaller one. A key difference, though, is that only one basic deci-
sion is made in the curve ball procedure, with costs or stakes raised after that initial
commitment.
This compliance tactic is almost legendary among car dealers. A customer agrees to a
special purchase price, say $800 below all other competitors. Then, the price is raised. A
number of reasons—we should say excuses—are given. The sales manager will not
approve the deal. Certain options were not included in the special offer. The manager
decreased the value of the customer’s trade-in. The ploy usually works. Customers agree-
ing to the special price usually do not flinch when thrown the curve because reneging
might suggest a lack of consistency or failure to fulfill an obligation, even though the obli-
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 380

380 Psychology of safety handbook

gation is only imaginary. Often customers develop their own reasons—or excuses—to jus-
tify their initial choice and the additional costs (Teger, 1980).

Some influence techniques can stifle trust

Throwing the curve ball raises a critical issue when trying to increase actively caring behav-
ior. How do you feel when someone throws you a curve? Let us talk about trust for a
moment.
Do you trust the waiter who brings you an expensive wine list—only after you have
been seated and made selections from the food menu? It probably depends on whether you
believe this sequence of events was done intentionally to get you to buy more. Similarly,
you might not dislike or mistrust the car salesman who jacks up the price unless you sus-
pect that his advertised price was just a lure to get you in the showroom.
In other words, our trust, appreciation, or respect for people might fall off considerably
if we believe they intentionally tricked or deceived us into modifying our attitude or
behavior. Of course, there may be no harm done if the result is clearly for our own good, as
for our health or safety, and we realize this.
Some of the influence strategies reviewed here are more likely to raise suspicion and
reduce interpersonal trust. The curve ball and door-in-the-face techniques are usually the
most difficult to pull off. Reciprocity might do the most harm, however, if the recipient
believed the kindness or favor was a self-serving manipulation done only to force feelings
of indebtedness, as is often the case in prison cultures. Trying to increase actively caring
behavior is not self-serving. We are trying to build a safety culture that benefits everyone.
This must be made clear. Once the purpose of using the influence techniques described
here is understood, interpersonal trust or mutual respect between agents and recipients of
interventions will not diminish but may actually grow.

Reinforcers vs. rewards

Technically, a behavioral consequence is a positive reinforcer only if it increases the fre-
quency, intensity, or duration of the targeted behavior. In reality, it is usually impossible to
know if this happens. You could give a person recognition, positive feedback, or even a
financial bonus, and not influence the behavior you are intending to reward.
Giving special recognition to someone for following all safety procedures, for example,
cannot improve the targeted behaviors because they are already at 100 percent. The finan-
cial bonus an individual receives for demonstrating better safety performance last month
does not necessarily influence any behavior. The behavior most often reinforced by recog-
nition ceremonies like a steak dinner celebrating a safety milestone is not safety-related
behavior but attendance at the dinner.
Because it is usually impossible to know whether a positive consequence has the
intended beneficial impact on behavior in real-world situations, I rarely use the term “pos-
itive reinforcer.” Instead, I use the more common term “reward.” A reward is a positive
consequence given to an individual or a group with the intention of improving, supporting,
or maintaining desired behavior. The reward can be one-on-one recognition, a group cele-
bration dinner, a positive feedback conversation, credits toward the purchase of a cata-
logue gift item, a financial bonus, or a small trinket with a safety logo.
A reward might be given long after occurrence of the desired behavior and, therefore,
it is unlikely to have a direct effect on that behavior. Some rewards are not even associated
with specific behaviors. For example, the behavior most likely reinforced by awarding
companies with a safety improvement placard is someone walking to the stage to receive
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 381

Chapter sixteen: Increasing actively caring behavior 381

the prize and public recognition.

The strict behavioral approach to safety does not recognize much, if any, value in
rewards (Daniels, 2000; Malott et al., 1997). If a behavior analyst observes no change in a
target behavior when a particular consequence follows it, that consequence is considered
useless in that situation and no longer applied. Here is my point. Even if a reward does not
improve behavior directly, it has other special benefits.
From the person-based perspective presented in Section 5 of this Handbook, a positive
consequence or reward has value beyond the behavior change. If given genuinely, inter-
personal recognition, group celebration, and positive feedback directly improve internal
unobservable aspects of people. In a word, they make you feel better. This is a worthwhile
outcome by itself. Moreover, it is likely that safety-related behavior will be indirectly
improved.
Let me explain by asking how you feel after being rewarded for exemplary perform-
ance. Do you get a boost in self-esteem? Feel better about yourself? Do you feel more com-
petent at the task singled out for the reward? Do you sense a greater degree of personal
control over the targeted activity? Are you more optimistic that you will be successful
in the future? Do you feel more connected with other team members who attended the
celebration event?
Answering “yes” to any of these questions testifies to the value of rewards. Not only
because they make you feel good, but because they can improve the person states implied
in these questions. As I discussed in Chapter 15, when any of these internal feeling states
is increased, a person’s willingness to look out for the welfare of others is also increased. So
whether or not a reward increases the behavior it follows, it is apt to improve one or more
of the feeling states that make people more likely to actively care for the safety of others.
This justifies using rewards, even when behavior is not directly influenced. Delivered
appropriately, rewards always bring out the best in people because they improve those
feeling states—self-esteem, self-efficacy, personal control, optimism, or belonging—that
make it more likely an individual will go beyond the call of duty to help others.

In conclusion
The information reviewed in this chapter is critical in terms of practical application.
Integrating principles and procedures from the prior chapters, I have tried to address this
crucial challenge. Continuous safety improvement leading to a Total Safety Culture
requires people to actively care—for others as well as themselves. Research-derived pro-
cedures to increase the frequency of actively caring behavior throughout a culture was dis-
cussed. Some of these influence techniques indirectly increase actively caring behavior by
benefiting the person states that facilitate one’s willingness to care. Other influence strate-
gies target behaviors directly.
Indirect strategies are deduced from the actively caring model explained in Chapter 15.
Any procedure that increases a person’s self-esteem, perception of empowerment—includ-
ing self-efficacy, personal control, and optimism—or sense of belonging or group cohesion
will indirectly benefit active caring. A number of communication techniques enhance more
than one of these states simultaneously, particularly actively listening to others for feelings
and giving genuine praise for accomplishments. There are barriers to focusing on the pos-
itive, and we discussed these in the hope that awareness will help overcome the obstacles.
We need only reflect on our own lives to appreciate the power of choice and how the
perception of choice and personal control makes us more motivated, involved, and com-
mitted. Choice activates and sustains actively caring behavior.
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 382

382 Psychology of safety handbook

Perceptions of belonging are important, too. They increase when groups are given con-
trol over important decisions and receive genuine recognition for accomplishments.
Synergy is the ultimate outcome of belonging and win–win group involvement. It occurs
when group interdependence produces more than what is possible from going it alone. We
reviewed barriers to belonging and synergy, specifically, social loafing. More on this topic
is presented in Chapter 17 when I discuss the challenges of facilitating teamwork.
The behavior-change techniques detailed in Section 4 of this text—from setting
SMART goals and signing promise cards to offering soon, certain, and positive conse-
quences for the right behavior—can be used to enhance those actively caring person states.
They can also directly increase actively caring behavior. Education and direct participation
in actively caring projects foster the behavior we are seeking as well.
The interpersonal influence principles of reciprocity and commitment –consistency
were introduced as they apply to our everyday decisions and behaviors. The commitment
and consistency principle is behind the success of safe behavior promise cards, the foot-in-
the-door technique (“start small and build”), and throwing a curve ball (“raise the stakes
later”). These principles and specific strategies can be applied to directly boost actively
caring behaviors.
It is critical to realize that social influence strategies can reduce respect or trust between
people if they are applied within a context of win–lose or top-down control. On the other
hand, when the purpose of the influence technique is clearly to increase actively caring
behavior and improve the safety and health of everyone involved, a win–win climate is
evident and mutual respect and trust is nurtured.
Positive reinforcers, by definition, directly increase the actively caring behaviors they
follow. However, since we usually do not know the behavioral impact of a positive conse-
quence, I suggest using the term “reward” when referring to the application of a positive
consequence to motivate or support desired behavior. Even if the reward does not increase
the target behavior, it will likely have an indirect effect on actively caring behavior because
it can strengthen one or more of the five person states that influence one’s willingness to
actively care for a person’s safety or health. So look for opportunities to reward quality per-
formance, and deliver the reward well.

References
Adler, R. B. and Towne, N., Looking out/Looking in, 6th ed., Holt, Rinehart, & Winston, Philadelphia,
PA, 1990.
Beaman, A. L., Barnes, P. J., Klentz, B., and McQuirk, B., Increasing helping rates through informa-
tion dissemination: teaching pays, Personal. Soc. Psychol. Bull., 4, 406, 1978.
Beck, A. T., Cognitive therapy: a 30-year retrospective, Am. Psychol., 46, 368, 1991.
Berkowitz, L. and Daniels, L. R., Affecting the salience of the social responsibility norm: effect of
past help on the responses to dependency relationships, J. Abnorm. Soc. Psychol., 68, 275, 1964.
Brownstein, R. and Katzev, R., The relative effectiveness of three compliance techniques in eliciting
donations to a cultural organization, J. Appl. Soc. Psychol., 15, 564, 1985.
Burger, J. M. and Petty, R. E., The low-ball compliance technique: task or person commitment?,
J. Personal. Soc. Psychol., 40, 492, 1981.
Carnevale, P. J., Pruitt, D. G., and Carrington, P. I., Effects of future dependence, liking, and
repeated requests for help on helping behaviors, Soc. Psychol. Q., 45, 9, 1982.
Cialdini, R. B., Influence: Science and Practice, 3rd ed., Harper Collins College, New York, 1993.
Cialdini, R. B., Cacioppo, J. T., Bassett, R., and Miller, J. A., Low-ball procedure for producing com-
pliance: commitment then cost, J. Appl. Soc. Psychol., 15, 492, 1978.
Cialdini, R. B., Vincent, J. E., Lewis, S. K., Catalan, J., Wheeler, D., and Darby, B. L., Reciprocal
concessions procedure for inducing compliance: the door-in-the-face technique, J. Personal. Soc.
Psychol., 1, 206, 1975.
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 383

Chapter sixteen: Increasing actively caring behavior 383

Covey, S. R., The Seven Habits of Highly Effective People, Simon & Schuster, New York, 1989.
Daniels, A. C., Bringing Out the Best in People: How to Apply the Astonishing Power of Positive
Reinforcement, 2nd ed., McGraw-Hill, New York, 2000.
Dashiell, J. F., Experimental studies of the influence of social situations on the behavior of individ-
ual human adults, in A Handbook of Social Psychology, Murcheson, C., Ed., Clark University
Press, Worchester, MA, 1935.
Deutsch, F. M. and Lamberti, D. M., Does social approval increase helping?, Personal. Soc. Psychol.
Bull., 12, 148, 1986.
Eisenberg, N., The Caring Child, Harvard University Press, Cambridge, MA, 1992.
Ferrari, J. R. and Geller, E. S., Developing future caregivers by integrating research and community
service, Commun. Psychol., 27(2), 12, 1994.
Freedman, J. L. and Fraser, S. C., Compliance without pressure: the foot-in-the-door technique,
J. Personal. Soc. Psychol., 4, 195, 1966.
Gabrenya, W. K., Jr., Latané, B., and Wang, Y. E., Social loafing in cross cultural perspective, J. Cross-
Cult. Psychol., 14, 368, 1983.
Geller, E. S., Ten principles for achieving a Total Safety Culture, Prof. Saf., 39(9), 18, 1994.
Geller, E. S. and Roberts, D. S., Beyond behavior modification for continuous improvement in occu-
pational safety, paper presented at the FABA/OBM Network Conference, St. Petersburg, FL,
January 1993.
Geller, E. S., Roberts, D. S., and Gilmore, M. R., Predicting propensity to actively care for occupa-
tional safety, J. Saf. Res., 27, 1, 1996.
Goranson, R. and Berkowitz, L., Reciprocity and responsibility reactions to prior help, J. Personal.
Soc. Psychol., 18, 227, 1966.
Gouldner, A. W., The norm of reciprocity: a preliminary statement, Am. Soc. Rev., 25, 161, 1960.
Greene, B. F., Winett, R. A., Van Houten, R., Geller, E. S., and Iwata, B. A., Eds., Behavior Analysis in
the Community 1968 –1986 from the Journal of Applied Behavior Analysis, reprint series, Vol. 2,
Society for the Experimental Analysis of Behavior, Inc., Lawrence, KS, 1987.
Grusec, J. E. and Redler, E., Attribution, reinforcement, and altruism: a developmental analysis,
Dev. Psychol., 16, 525, 1980.
Harkins, S. G. and Szymanski, K., Social loafing and group evaluation, J. Personal. Soc. Psychol., 56,
934, 1984.
Harkins, S. G. and Petty, R. E., Effects of task difficulty and task uniqueness on social loafing,
J. Personal. Soc. Psychol., 43, 1214, 1982.
Hoffman, M. L., Altruistic behavior and the parent child relationship, J. Personal. Soc. Psychol., 31,
937, 1975.
Joule, R. V., Tobacco deprivation: the foot-in-the-door technique versus the low-ball technique, Eur.
J. Soc. Psychol., 17, 361, 1987.
Kahn, A. and Tice, T., Returning a favor and retaliating harm: the effects of stated initiation and
actual behavior, J. Exp. Soc. Psychol., 9, 43, 1973.
Keller, J. J., The recycling solution: how I increased recycling on Dilworth Road, J. Appl. Behav.
Anal., 24, 617, 1991.
Kunz, P. R. and Woolcott, M., Season’s greetings: from my status to yours, Soc. Sci. Res., 5, 269, 1976.
Latané, B. and Darley, J. M., The Unresponsive Bystander: Why Doesn’t He Help?, Appleton-Century-
Crofts, New York, 1970.
Latané, B., Williams, K., and Harkins, S., Many heads make light the work: the causes and conse-
quences of social loafing, J. Personal. Soc. Psychol., 37, 823, 1979.
Malott, R. W., Whaley, D. L., and Malott, M. E., Elementary Principles of Behavior, 3rd ed., Prentice-
Hall, Upper Saddle River, NJ, 1997.
McGovern, L. P., Ditzian, J. L., and Taylor, S. P., The effect of one positive reinforcement on helping
behavior, Bull. Psychon. Soc., 5, 421, 1975.
Monty, R. A. and Perlmuter, L. C., Persistence of the effect of choice on paired-associate learning,
Mem. Cognit., 3, 183, 1975.
Monty, R. A., Geller, E. S., Savage, R. E., and Perlmuter, L. C., The freedom to choose is not always
so choice, J. Exp. Psychol.: Hum. Learn. Mem., 37, 170, 1979.
820049_CRC1_L1540_CH16 11/10/00 2:06 PM Page 384

384 Psychology of safety handbook

Moss, M. K. and Page, R. S., Reinforcement and helping behavior, J. Appl. Soc. Psychol., 2, 360, 1972.
Notz, W. W., Boschman, I., and Tax, S. S., Reinforcing punishment and extinguishing reward: on the
folly of OBM with SPC, J. Organ. Behav. Manage., 9(1), 33, 1987.
Perlmuter, L. C., Monty, R. A., and Kimble, G. A., Effect of choice on paired-associate learning,
J. Exp. Psychol., 91, 47, 1971.
Peterson, C., Maier, S. F., and Seligman, M. E. P., Learned Helplessness: a Theory for the Age of Personal
Control, Oxford University Press, New York, 1993.
Regan, D. T., Effect of a favor on liking and compliance, J. Exp. Soc. Psychol., 7, 627, 1971.
Roberts, D. S. and Geller, E. S., An “actively caring” model for occupational safety: a field test, Appl.
Prevent. Psychol., 4, 53, 1995.
Rushton, J. P., Generosity in children: immediate and long term effects of modeling, preaching, and
moral judgment, J. Personal. Soc. Psychol., 31, 459, 1975.
Rutherford, E. and Mussen, P., Generosity in nursery school boys, Child Dev., 39, 755, 1968.
Schein, E., The Chinese indoctrination program for prisoners of war: a study of “brainwashing,”
Psychiatry, 19, 149, 1956.
Seligman, M. E. P., Helplessness: On Depression Development and Death, Freeman San Francisco, 1975.
Staub, E., To rear a prosocial child: reasoning, learning by doing, and learning by teaching others, in
Moral Development: Current Theory and Research, DePalma, D. and Folley, J., Eds., Lawrence
Erlbaum, Hillsdale, NJ, 1975.
Staub, E., Positive Social Behavior and Morality: Socialization and Development, Vol. 2, Academic Press,
New York, 1979.
Staub, E., Moral exclusion, personal goal theory and extreme destructiveness, J. Soc. Iss., 46, 47,
1990.
Staub, E., The origins of aggression and the creation of positive relations among groups, in
Psychology and Social Responsibility: Facing Global Challenges, Staub, S. and Green, P., Eds., New
York University Press, 1992.
Steiner, I. D., Perceived freedom, in Advances in Experimental Social Psychology, Vol. 5, Berkowitz, L.,
Ed., Academic Press, New York, 1970.
Szymanski, K. and Harkins, S. G., Social loafing and self-evaluation with a social standard, J.
Personal. Soc. Psychol., 53, 891, 1987.
Taylor, S. E., Positive Illusions: Creative Self-Deception and the Healthy Mind, Basic Books, New York,
1989.
Teger, A. I., Too Much Invested to Quit, Pergamon Press, Elmsford, NJ, 1980.
Watson, D. C. and Tharp, R. G., Self-Directed Behavior: Self-Modification for Personal Adjustment, 7th
ed., Brooks/Cole Publishing, Pacific Grove, CA, 1987.
Weick, K. E., Small wins: redefining the scale of social problems, Am. Psychol., 39, 40, 1984.
White, R. W., Motivation reconsidered: the concept of competence, Psychol. Rev., 66, 297, 1959.
Williams, K., Harkins, S., and Latané, B., Identifiability as a deterrent to social loafing: two cheering
experiments, J. Personal. Soc. Psychol., 40, 303, 1981.
Wilson, D. W. and Kahn, A., Rewards, costs, and sex differences in helping behavior, Psychol. Rep.,
36, 31, 1975.
Zahn-Waxler, C., Radke-Yarrow, M., and King, R., Child rearing and children’s prosocial initiations
toward victims of distress, Child Dev., 50, 319, 1979.
Zahn-Waxler, C., Radke-Yarrow, M., and Chapman, M., Prosocial dispositions and behavior, in
Manual of Child Psychology: Socialization, Personality, and Social Development, Vol. 4, Hetherington,
E. M., Ed., Wiley, New York, 1983.
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 385

section six

Putting it all together

820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 386
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 387

chapter seventeen

Promoting high-performance
teamwork

Everyone talks about teamwork, but not everyone gets the best from their teams. Teamwork just does
not come naturally. This chapter explains why and what we can do about it. Principles and practi-
cal procedures are offered for initiating and sustaining productive teamwork. The functions of seven
different safety teams are described. Each of these teams contributes to improving the human dynam-
ics of occupational safety, as taught in this Handbook. Each team depends on the output of other
teams to optimize the system and cultivate a Total Safety Culture.

“My responsibility is to get my twenty-five guys playing for the name on the front of
their uniform and not the name on the back.”—Tommy Lasorda

Imagine a workplace where everyone coaches each other about the safest way to perform
a job. A workplace in which actively caring for other people’s safety is a natural part of the
everyday routine. When people depend on each other in this way to improve safety, they
understand teamwork. They have an interdependent mindset and realize the true meaning
of synergy. For these individuals TEAM means Together Everyone Achieves More.
Reaching this level of teamwork does not come easy. After all, look at how we have
been raised. “Be independent,” we are told. We compete with other individuals to get
ahead, whether at work or at play. Remember, “Nice guys finish last.” A win–lose, me-first
mindset is promoted by almost everything in our culture, from the grades we get in school
to salary promotions at work.
This chapter is about win–win teamwork. It builds on the principles of behavior-based
safety presented in Section 3, the intervention tools from behavior-based safety detailed
in Section 4, and the concepts of group belonging and interdependence discussed in
Section 5.
First, I review some of the ways we must change our perspective in order to achieve
high-performance team results. Next, I introduce social psychology principles that explain
why some teams are not productive. Then, I cover seven procedural steps for getting the
best from teamwork, from selecting the right team members to evaluating team perform-
ance. Finally, I discuss the different stages teams go through on the way to producing syn-
ergistic results.
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 388

388 Psychology of safety handbook

Paradigm shifts for teamwork

Learning to work effectively as a team takes patience. Teamwork calls for different
approaches to work from those most of us are used to. By understanding these differences,
we can appreciate the various barriers to high-performance teamwork. Sometimes, policies
or procedures get in the way, and we can modify them to remove a teamwork barrier.
Often, however, the best we can do is change our perspective or expectancy regarding
teamwork and, then, adjust our behavior accordingly.
We are talking about changing paradigms here. As discussed in Chapter 3, a paradigm
is a powerful personal perception or a premature cognitive commitment (Langer, 1989). In
other words, our paradigm represents our attitude or expectancy in a particular situation
and biases the way we view that situation. It also influences what we take from a situation.
As a result, we often experience what we expect to happen, and we learn what we expect
to learn. It is a self-fulfilling prophecy. We act a certain way to be consistent with our para-
digm and so increase the likelihood that what we believe will actually occur.
Also, as I discussed earlier in Chapter 2, what we do—our method—influences our
perspective or how we view the situation. In other words, our behavior influences our per-
ceptions. We act ourselves into new ways of thinking. Consider this principle of human
nature while reviewing the following five paradigm shifts needed for high-performance
teamwork. We can literally act ourselves into becoming better team players—if we follow
these interconnected paradigm shifts.

From individual to team performance

Traditional work holds people accountable for their own behavior. Effective teamwork,
though, requires mutual accountability. It is not “What you do is what you get,” but
rather “How you collaborate with others is what the group gets.” Effective collaboration
nets performance results greater than what individuals can do by themselves. This is a
synergistic outcome.

From individual jobs to team tasks

Synergy occurs when each team member contributes individual talent and effort to improve
team performance. Team members receive task assignments from each other and carry out
their responsibilities to support the rest of the team. This is much different from traditional
work, which has us completing individual job assignments to please a supervisor.
Teamwork requires a shift from working exclusively to achieve personal goals to work-
ing to achieve shared team goals. This requires belief in the power of teamwork, commit-
ment toward the team’s mission, and trust that every team member will do his or her part
to meet team objectives.

From competitive rewards to rewards for cooperation

It takes a special mindset to revel in the accomplishments of a team effort. When team mem-
bers value their mutual purpose and believe their teammates will cooperate to achieve their
shared goals, they put forth their best efforts. When they see cooperation pay off, they
develop a unique appreciation for teamwork. They feel personally recognized when the
team is rewarded. Then, they cooperate more to fulfill their team’s next objectives.
Thus, when people see the synergistic results of team members working cooperatively,
their own team behaviors are reinforced and they contribute more for the team. They
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 389

Chapter seventeen: Promoting high-performance teamwork 389

change their mindset from “How can I do better for myself?” to “How can I do more
for the team?” Experiencing the rewards of cooperation is benefited by the next para-
digm shift.

From self-dependence to team-dependence

We come into this world dependent on others to take care of us. As children we depended
on our family for all of our basic life needs. As adolescents, however, we looked for oppor-
tunities to be on our own. It seems, in fact, a primary mission of most teenagers is to resist
dependency and assert independence. This reliance on self rather than others is promoted
and reinforced throughout our culture, from high-school and college classrooms to the
corporate boardroom. High-performance teamwork requires a dependency perspective,
however.
This might seem like regression, but it is really progression. In fact, it is more appro-
priate to consider this a paradigm shift to interdependence rather than dependence. That
is because the dependency between team members is reciprocal. While you depend on
team members to complete their task assignments, others depend on you to do your part.
We are moving from independence to interdependence here, as discussed earlier in
Chapter 16. The more you trust the ability and intentions of the other individuals on your
team, the more you will depend on your teammates for their contributions—and the more
you will feel obligated to complete your own task assignments.

From one-to-one communication to group interaction

Trust and interdependency are developed and supported through interpersonal conversa-
tion—the theme of Chapter 13. Some of this certainly happens through one-to-one interac-
tion. However, the synergistic power of teamwork is more readily realized through
effective team meetings. For example, I can assess the individual talents and motives of my
students and research associates during one-to-one conversation, but I learn much more
about our team’s potential to meet a challenge during group meetings.
Through group interaction, individuals see how their diverse talents can combine to
produce synergistic results. This leads to greater feelings of self-esteem and self-efficacy
and increases personal commitment to meet team objectives. It also cultivates group cohe-
sion and feelings of belonging which, in turn, motivates high-performance teamwork.
Of course, the amount of beneficial impact from group interaction depends on how
team meetings are run. Actually, a productive group communication can facilitate each of
the paradigm shifts reviewed here. Effective group discussions build trust in the abilities
and intentions of team members. Trust leads to cooperation. This strengthens a paradigm
of interdependency and increases personal motivation to work harder for the team. Later
in this chapter, I offer strategies for getting the most out of a team meeting. First, let us con-
sider specific barriers to synergistic teamwork.

When teams do not work well

Research psychologists have demonstrated certain drawbacks of teamwork. Specifically,
there are occasions when teams do not work well—when teamwork results in inappro-
priate or ill-advised decisions and actions. Understanding the circumstances surround-
ing poor team decisions can suggest direction for preventive action. In other words,
knowing when teamwork goes bad is useful in planning how to maintain productive
high-performance teams.
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 390

390 Psychology of safety handbook

On January 28, 1986, at 11:30 a.m., many Americans were “glued to the tube” watching
the takeoff of the Challenger on another NASA space shuttle launch. Elementary schools
interrupted lessons and rolled TV sets into classrooms so their students could watch this
event. NASA had completed a string of successful shuttle launches, but this one was spe-
cial. One of the crew members was Christa McCauliffe, a high-school teacher from
Concord, NH, who won a nationwide competition to join the shuttle team. Viewers all
over the world saw the cheerful space crew enter the Challenger and then witnessed a suc-
cessful launch.
Suddenly, the unthinkable happened. Only 73 seconds into the flight, the spacecraft
exploded and burst into flames. Television cameras captured this disaster as a bright flash
in the sky followed by a white trail of smoke. The tragedy was replayed numerous times
for viewers worldwide.
From one perspective, the root cause of this catastrophe was an engineering flaw. The
rocket seals did not hold up under freezing temperatures. However, the engineers of the
rocket booster had anticipated a potential seal problem under the weather conditions and
warned against the launch. Unfortunately, the NASA executive who made the final “go”
decision was shielded from the engineer’s warnings. Frustrated by several launch delays,
but confident from prior successes, a NASA management team decided to silence the engi-
neer’s warnings.
This group decision was perhaps the real cause of the Challenger explosion. This and
other historical fiascoes, including failure to anticipate the Japanese attack on Pearl Harbor,
the Bay of Pigs invasion of Cuba, the escalation of the Vietnam war, the Watergate cover-
up, the Iran-Contra affair, and the Chernobyl reactor tragedy, resulted from teams of well-
intentioned professionals making unwise and at-risk decisions.
As illustrated in Figure 17.1, this problem has been generally labeled “groupthink”
(Janis, 1972, 1983). The result is a deterioration of mental effectiveness, practical consider-
ations, and moral judgment as a result of various group process factors. The group process
factors that contribute to groupthink have been studied extensively by social psycholo-
gists. Let us review these and explore their relationship to industrial safety and the opti-
mization of team performance.

Group gambles
A number of decision-making studies in the 1960s showed that groups generally make
riskier decisions than individuals (Stoner, 1961; Wallach et al., 1962). This is contrary to
the common-sense belief that groups tend to be more conservative than their individual
members.
To compare the risk-taking of groups vs. individuals, researchers asked subjects to

1. Make a series of decisions individually on a questionnaire.

2. Form a group to discuss each choice on the questionnaire and arrive at a consensus.
3. Complete the questionnaire again individually.

Not only were the group consensus decisions more risky than the average of the initial
decisions; but after the group discussions, the individual choices became more risky. This
group process phenomenon is referred to as “the risky shift” (Kogan and Wallach, 1964).
Group polarization. In some cases, risky means more extreme or opinionated. When
team members discuss an issue the majority favors or opposes, they become more cer-
tain and confident in their viewpoint. This exaggeration of individual opinion as a result
of group discussion is referred to as “group polarization” (Isenberg, 1986; Levine and
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 391

Chapter seventeen: Promoting high-performance teamwork 391

Figure 17.1 Groupthink occurs when group members acquiesce to a leader without con-
sidering alternative options or proposals.

Moreland, 1998). When the group starts with the right values, principles, and methodol-
ogy, group polarization will increase commitment and facilitate beneficial behavior
change. However, the polarization that frequently occurs when like-minded people discuss
their concerns or attitudes can distort reality and lead to groupthink and unwise decisions.
Diffused responsibility. It is natural for team members to feel they cannot be held
accountable for the failure of a group decision. The risk of failure is spread around. Think
about it. How often does a work group or team decide to bypass or overlook a safety pro-
cedure, perhaps in the name of productivity? If someone gets injured as a result, no one
individual can be held responsible. The risk was a team decision.
Deindividuation. Often diffusion of responsibility is accompanied by deindividua-
tion (Postmes and Spears, 1998)—people lose their sense of self-awareness and individu-
ality within the team context. As with group polarization, this can be positive or negative,
depending on the values and principles of the group. When team members compromise to
achieve goals consistent with the organization’s purpose and mission statement, the effects
of deindividuation are likely to be beneficial. On the other hand, deindividuation can lead
to abandoning fundamental individual constraints and to less careful or less safe decisions
and behaviors.
Deindividuation is facilitated when group members wear uniforms and cut their hair
in similar ways, as in prisons, cults, monasteries, and the military (Zimbardo, 1969). Work
teams certainly do not experience the degree of deindividuation found in prisons, monas-
teries, and military boot camps. Group meetings of all sorts, however, often stifle individ-
ual contributions; and for some people, this can lead to a perceived loss of personal control.
Individuals may fail to speak up when they disagree. Silence is interpreted as consent and
leads to an illusion of group unanimity. With the whole group on the same wavelength
(even if risky), individuals feel protected and develop an illusion of invulnerability (Taylor
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 392

392 Psychology of safety handbook

Figure 17.2 Important group decisions require diversity.

and Brown, 1988; Weinstein, 1980). The liability of uniformity and deindividuation and the
need for diversity in group decision making is illustrated in Figure 17.2.

Overcoming groupthink
It is useful to realize that team meetings can lead to groupthink through the social dynam-
ics of diffusion of responsibility, group polarization, and deindividuation. To prevent this
problem, it is important to recognize a few telltale signs. When team meetings do not
encourage a variety of viewpoints nor promote diversity of knowledge and experience,
groupthink is probable. Groupthink is even more likely when members of a team actively
stifle disagreement and seek harmony and unanimity at all costs. So teams that attempt to
reach a quick decision about a serious issue without substantial discussion are on the verge
of groupthink.
A group facilitator who embraces diverse opinions, invites input and critique of a
developing solution, or assigns task forces to explore alternative action plans will decrease
the probability of groupthink and increase the likelihood of synergy. Also, when a team
has a well-defined purpose and mission statement (as discussed later) that is principle-
centered and compatible with the organization’s vision, the group processes of polariza-
tion and deindividuation will lead to group commitment, loyalty, trust, and a win–win
perspective. This also increases synergistic decision making and action plans to optimize
the system for the advantage of everyone.

Cultivating high-performance teamwork

Now, let us consider the main phases of teamwork, from start to finish, and see how
each relates to the development of optimum group performance for safety improvement.
In other words, let us move our discussion of paradigms and principles to a real-world
application. Here, I provide real-world answers to questions like, “How can we establish a
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 393

Chapter seventeen: Promoting high-performance teamwork 393

1. Select the Right Team Members

• Understand and appreciate behavior-based safety.
• Commitment, interpersonal interest,
communication, and caring.
2. Clarify the Assignment
• State general mission or purpose.
• Specify resources, authority, and accountability.
• Get acquainted.
• Develop understanding of TEAM, interdependency,
and synergy.
3. Establish a Team Charter
• Write a mission statement.
• Set ground rules.
• Define deliverables and accountability.
• Specify budget details and direct reports.
• Assign standard team roles.
4. Develop Action Plan
• Set goals with SMARTS.
• Assign task responsibilities.
• Develop time lines.
5. Engage in the Process
• Conduct productive team meetings.
• Use brainstorming and consensus-building.
• Give each other supportive and corrective
behavior-based feedback.
6. Evaluate Team Performance
• Recognize process results.
• Document product results.
• Celebrate accomplishments.
7. Disband, Restructure, or Renew the Team

Figure 17.3 Follow seven steps for team success.

successful safety team?” and “Once we have a safety team, what can we do to make it more
effective?”
Figure 17.3 outlines seven consecutive phases of teamwork, from selecting team mem-
bers to disbanding or renewing the team. These are the basic steps of teamwork as dis-
cussed by leading team-building trainers and consultants (Cadwell, 1997; Katzenbach and
Smith, 1994; Lloyd, 1996; Rees, 1997; Torres and Fairbanks, 1996; Wellins et al., 1991). Let us
examine each of these steps in more detail as they relate specifically to industrial safety.

Selecting team members

Obviously, the first crucial step in successful safety teamwork is to select the right people
for your team. Someone is ultimately responsible for choosing team members. In safety,
this is often the safety director or the person responsible for maintaining injury reports and
lost-time records. In some cases, however, it is advantageous for a small committee of
safety champions representing a cross section of the workforce to select potential members
of a safety team. I say “potential” because it is important for membership to be voluntary.
So, a safety champion or selection committee should come up with a list of people to
approach one-on-one and ask if they would be willing to serve on a particular safety team.
So, what kinds of people should you look for as potential members of a safety team?
Perhaps, first and foremost, the candidate should be committed to safety. Has the individ-
ual done something recently to indicate personal concern for the safety or health of a
coworker? Perhaps, she turned in a comprehensive near-hit report in a work culture where
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 394

394 Psychology of safety handbook

Figure 17.4 This survey can be used to select potential members of a behavior-based
safety team.
such reports are rare or, maybe, the employee was injured recently and has given testimony
to a renewed regard for safety initiatives.
Besides demonstrating special commitment to occupational safety, the best team
members also have other qualities. They have interpersonal skills (they like to work with
other people), they communicate well (they actively listen and speak with passion), and
they are willing to actively care for the safety and health of others.
The brief survey in Figure 17.4 addresses the desirable characteristics of participants on
a behavior-based safety team. You could readily customize these questions to assess peo-
ple’s desire and readiness to serve on any safety team. Ask people to be frank and honest
in their answers and do not imply any value to a high score. Then, those who score high on
this survey are likely candidates for a particular safety team.
The survey items in Figure 17.4 are classified into the constructs they target. If you use
this device for only assessment and not instruction, you should administer the 15 items
without the category labels. However, the labels are useful for instructional purposes. Plus,
you will find these labels useful when attempting to match an individual profile with a par-
ticular team assignment.

Clarify the assignment

From the start, it is important for the team members to understand their assignment. They
need to know the overall mission of the team and the resources available to accomplish it.
They also need to understand their authority with regard to the mission. For example, they
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 395

Chapter seventeen: Promoting high-performance teamwork 395

need to realize the degree of control the team has over the consequences of their decisions.
Will any other authority influence the outcome of the team’s decisions? In other words, to
what extent is the team truly empowered to carry out the processes needed to accomplish
its mission?
This is when the true purpose of a team is discussed. Why is the assignment given to a
team instead of individuals working alone? What are the advantages of using a team? Own
up to the fact that the start-up process will take some time, and some might think one
dedicated person would be more efficient. In the long run, however, teamwork will be
more effective. Now it is appropriate to explain the TEAM acronym—Together Everyone
Achieves More.
Besides explaining the general mission and the TEAM concept, the team facilitator
should also provide opportunities for the participants to get acquainted. Introductions
could be initiated with a statement like, “Let’s get to know each other better by each per-
son stating your name, your job, and your general expectations, if any, about our team
assignment.” Later, you might also ask each team member to say something about safety.
“Tell us what you know about behavior-based safety or about your personal interest or
commitment to occupational safety.”
During the first introductions, the team facilitator should take notes on each indi-
vidual. I have found it most useful to draw a seating chart of the room and write each
person’s name at his or her seating location on the chart. Then after the introductions, I can
call on each participant by name, usually just the first name. In this way, I quickly learn
everyone’s name.
It is amazing how much more comfortable I feel as a facilitator when I know everyone’s
name. I realize I am setting the right example by addressing each person by name when
answering their questions or calling on them to react to a comment. This is one sure way to
facilitate broad participation—a critical aspect of the initial stages of teamwork.

Establish a team charter

During the prior “getting acquainted” phase, the mission was given as a general descrip-
tion of the team’s assignment. Now, it is time to write a formal mission statement that artic-
ulates the overall purpose of the team, define the ground rules for team meetings, address
budget issues, specify what the team will produce (its deliverables), and assign various
team roles. Some standard team roles are as follows, although I have seen many situations
where one person assumes more than one role.

• Team leader—provides direction and obtains outside resources.

• Team facilitator—keeps meetings focused and prompts total participation.
• Team administrator—handles various administrative duties like distributing
reports, networking with outside individuals and groups, and reminding members
of team meetings.
• Treasurer—tracks input and output of finances.
• Reporter—documents and distributes meeting agenda and minutes.

Developing an accountability system. The team might also give itself a name to
reflect its basic function and provide a sense of identity for its members. It is also impor-
tant to specify the accountability system for the team. For example, answers to the follow-
ing questions are usually needed.

• What kind of progress reports are desired? When and how many?
• Who receives the progress reports and evaluates them?
820049_CRC1_L1540_CH17 11/10/00 2:06 PM Page 396

396 Psychology of safety handbook

Figure 17.5 Body language speaks louder than words.

• What individuals or groups outside the team will review or use its deliverables?
• Who will give the team performance feedback? When and how often?
• Who will hold the team accountable and by what standard(s)?
• What information will be used to assess team success?
• What could be the individual consequences for team success or failure?

It is vital that every team member understands and affirms the mission statement,
deliverables, ground rules, accountability system, and team role assignments. Therefore, a
“team charter” is developed through consensus-building, which is the opposite of top-
down decision making. It is not the same as negotiating, calling for a vote and letting the
majority win, or working out a compromise between two different sets of opinions. The
team facilitator in Figure 17.5 is not likely to cultivate consensus.
Building consensus. Negotiating, voting, or compromising comes across as
win–lose and decreases the interpersonal trust needed for high performance teamwork. A
majority of the team might be pleased, but others will be discontented and might actively
or passively resist involvement. Even the “winners” could feel lowered interpersonal trust.
“We won this decision, but what about next time?” Without everyone’s buy-in for a group
decision, the teamwork will be less synergistic than possible.
So how can group consensus be developed? How can the outcome of a heated debate
on ways to solve a problem be perceived as a win–win solution everyone supports, instead
of a win –lose compromise or negotiation? Practical answers to these questions are easier
said than done. Consensus-building takes time, energy, and patience. It requires open and
frank conversation among all team members.
The scenario depicted in Figure 17.6 is unacceptable. All participants must be willing
to state their honest opinion without fear of ridicule or reprisal. It takes a good team facil-
itator to make this happen. He or she needs to solicit the opinions of everyone throughout
the team meeting.
Rees (1997) describes six basic steps to reaching a consensus decision. Briefly, these are

• Set the decision goal. What is the aim or purpose of the consensus-building exer-
cise? What will be the end result, one way or another, of the group’s decision
process?
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 397

Chapter seventeen: Promoting high-performance teamwork 397

Figure 17.6 Consensus-building requires frank and open communication.

• Spell out the criteria needed to make the group decision worthwhile. What
qualities or characteristics of the decision are needed to reach a particular goal?
Which criteria are consistent with the team’s ultimate purpose or mission? What are
the budget constraints? What principles or quality standards are relevant? What will
a decision provide for the team? What criteria or restrictions are essential and what
criteria are desirable but not absolutely necessary?
• Gather information. What information is useful for making the decision? Where
is this information and who can provide it? How should the information be sum-
marized for optimal understanding and consideration?
• Brainstorm possible options. What are the variety of possible solutions? Does
everyone understand each option and its ramifications? How does everyone feel
about the possibilities? Has everyone had a chance to voice a personal opinion?
• Evaluate the brainstormed options against the group’s criteria. Which solutions
appear to meet the “must have” criteria? Which options seem to meet the “nice
but not necessary” criteria? To what degree will each option meet both the “neces-
sary” and “desirable” criteria? Can certain solution options be combined to meet
more criteria?
• Make the final decision as a team. Which option or combination best meets all of
the “necessary” criteria and most of the “desirable” criteria? Who has reservations
and why? How can we resolve individual skepticism? Can everyone support the
most popular option? What can be altered in the most popular action plan to attract
unanimous support and ownership?

I am sure you see that building consensus around a group process or action plan is not
easy. There is no quick fix to do this. It requires plenty of interpersonal communication, includ-
ing straightforward opinion sharing, intense discussion, emotional debate, proactive listen-
ing, careful evaluation, methodical organization, and systematic prioritizing. On important
matters, however, the outcome is well worth the investment. When you develop a solution
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 398

398 Psychology of safety handbook

or process every potential participant can get behind and champion, you have cultivated the
degree of interpersonal trust and ownership needed for total involvement. Involvement, in
turn, builds personal commitment, more ownership, and then more involvement.
Developing ground rules. Total participation of every team member during impor-
tant discussions is vital for consensus-building. Thus, open and frank discussion should be
a team-meeting ground rule accepted by everyone. Figure 17.7 lists this and other poten-
tial ground rules to consider adopting for increasing the effectiveness of your team meet-
ings. These are general guidelines or rules of conduct the team members need to agree on
and hold each other accountable to follow.
You cannot just post the contents of Figure 17.7 and call them “our team ground rules.”
Rather, you need to discuss the issue of ground rules with team members and get every-
one’s opinion. Then, use a consensus-building approach to get everyone’s acceptance of the
final list. You can use the suggestions in Figure 17.7 to “prime the pump” or stimulate dis-
cussion or just keep the seven items in mind as you facilitate group discussion and look for
opportunities to direct comments toward these topics. Your list of ground rules might be
similar to the following items which were given by Lloyd (1996, page 26) and reflect the
same meaning as those in Figure 17.7.
We agree to

• Speak respectfully to one another and about one another.

• Listen without interrupting.
• Express opinions and feelings openly and honestly.
• Ask for help when needed and offer help when possible.
• Make commitments seriously—and keep them.

1. Everyone participates
Active participation comes with the territory. This means
always being prepared for team meetings and problem-
solving discussions.
2. No barbs or put-downs
Team members show respect for each other. They don’t
say anything that could hurt someone’s feelings or limit the
involvement of others.
3. Every idea counts
Team members listen with respect to everyone’s opinion
regardless of how silly it might seem at first. The strangest
sounding idea can be the seed for creative invention.
4. Strive to be completely informed
Team members actively listen during team meeting to know
exactly what’s going on. They openly ask questions about
anything they don’t understand.
5. Follow through on commitments and meet deadlines
Keeping promises builds interpersonal trust and assures team
progress. This includes showing up for all meetings
on time.
6. Support team decisions
Team members voice their concerns during decision-making
discussions because in the end they realize they must support
a team decision.
7. Think win/win interdependency
Team members want everyone to win. Synergy depends on
everyone contributing individual talents for the good of all.

Figure 17.7 Seven ground rules can promote effective team meetings.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 399

Chapter seventeen: Promoting high-performance teamwork 399

• Support the team and each other.

• Focus on problems and solutions, not blame and accusations.

Writing a mission statement. You can use this basic consensus-building approach to
arrive at a mission statement. In other words, you should have a basic idea of the team’s
overall mission or purpose. Then, use consensus building to write a mission statement
everyone can support. Be sure to explain that the team’s mission statement must be a
clearly stated purpose that serves to direct and motivate team members. It answers three
basic questions.

• What does the team do?

• How does the team do its work?
• Who are the team’s customers?

Develop an action plan

Keeping in mind the team mission and ground rules, the team now plans how it will pro-
ceed. This planning process consists of three primary steps.

• Define specific goals needed to accomplish the mission.

• Decide on a time line for completing each goal.
• Assign goal-relevant tasks to each team member.

Let us consider some general approaches to taking these steps. The specifics will vary
depending upon the safety mission, the team charter, and the talents, skills, and opinions
of the team members. Remember to follow the basic consensus-building process when
arriving at goals and assigning tasks.
Follow-up work plans to accomplish the team’s mission can usually follow the basic
format of the first work plan. Refinements and additions will be needed to account for new
team members, rotation of team member roles or assignments, additional mission-related
challenges, and organizational changes. However, the essence of the work plan will be con-
stant for a given safety team, so there is some consolation in knowing that only the initial
planning process will be both time consuming and challenging.
Defining tasks. Process goals imply specific tasks, while outcome goals do not. For
example, the process goal, “complete 100 behavior-based observation sessions by the end
of the month” is quite task specific and stipulates what needs to be done. In contrast, the
outcome goal to “reduce the total recordable injury rate by 50 percent this year” does not
suggest any actions or behaviors. Assuming this outcome goal is judged achievable by the
team, it is necessary to decide on specific tasks needed to achieve the goal. One of those
tasks might be, in fact, to conduct periodic behavior-based observations throughout a work
area. In this case, the process goal to achieve 100 observations would be needed to attain
the outcome of reduced injuries.
It would likely take several process goals to reach a certain injury-reduction goal.
Process goals could be derived for attendance of safety meeting, for reporting of near hits
and property damage incidents, for removing environmental hazards, and for engaging in
one-on-one safety-related conversations with coworkers. All of these are tasks that could
contribute to reaching an injury-reduction milestone. So, the achievement of an injury-
reduction outcome goal is contingent on reaching certain process goals. All of these goals
are consistent with the mission or purpose of obtaining an injury-free workplace.
Assign task responsibilities The key to successful teamwork is to develop a list of
specific tasks needed to achieve team goals and, then, to assign the right persons to take on
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 400

400 Psychology of safety handbook

the various task responsibilities. Also critical, of course, is the setting of appropriate dead-
lines for each task to be completed. Adding a deadline or completion date to specific
assignments results in a SMARTS goal. These goals are SMART, as discussed earlier in
Chapter 10, with “S” for specific, “M” for motivational, “A” for achievable, “R” for rele-
vant, and “T” for trackable. The added “S” for SMARTS team goals stands for shared.
SMARTS goals are then organized into a time line for scheduling teamwork.
The time line reflects interdependency, because accomplishing certain goals are con-
tingent on reaching other goals. For example, the percentage of safe behaviors for a work
team cannot be posted until team members submit their data, and this cannot happen with-
out the development of a critical behavior checklist and the training of all observers on the
proper use of the checklist.
A time line of team goals is not carved in stone. It needs to be updated as team mem-
bers complete their tasks. Thus, a prime purpose for periodic team meetings is to review
progress on tasks, acknowledge goal accomplishments, and determine when additional
tasks need to be initiated. Sometimes, the results of task assignments will not be as desir-
able as expected. When this happens, team meetings are needed to redesign tasks and/or
make adjustments in task assignments.

Make it happen
After setting SMARTS goals, assigning task responsibilities, and developing a time line, the
real work begins. Process goals have been set and team members are motivated to fulfill
their interdependent roles for the sake of their team mission. This is the performing stage
of teamwork. Regular team meetings are still needed to keep the process going and to pro-
mote continuous improvement. More specifically, team meetings provide opportunities for
team members to connect with one another and

• Hold each other accountable for achieving specific tasks.

• Review project progress and acknowledge achievements of individuals, subgroups,
and the team as a whole.
• Discuss problems and entertain corrective action plans.
• Check the time line and make refinements and additions.
• Plan for next steps and assign new task responsibilities.

As discussed earlier, team meetings are run by a designated leader or facilitator, and
notes are taken by the recorder. For some teams the same person serves as both leader and
facilitator, setting the tone, prompting discussion, and encouraging total participation.
Often, however, it is a good idea to have a leader and a facilitator. The facilitator steps the
group through the agenda and calls for reports, comments, and suggestions when appro-
priate, while the team leader or champion offers insightful commentary and challenging
observations as a regular member of the discussion group.
It is also a good idea to rotate the role of facilitator among team members. Giving team
members opportunities to direct the flow of a team meeting increases both personal confi-
dence and team commitment. In this case, involvement precedes increases in commitment
and the perception of individual competence and confidence. These perceptions, in turn,
lead to more involvement and commitment. Let us consider some key aspects of running
a productive team meeting.
Prepare an agenda. A well-planned and well-run meeting starts with an agenda. This
keeps the meeting on track and focused and prevents off-track or time-wasting behavior. A
good agenda is neither wordy nor complex. It is simply an outline of items to be covered.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 401

Chapter seventeen: Promoting high-performance teamwork 401

At the start of the meeting a copy of the agenda should be distributed to all partici-
pants. When participants keep an agenda in front of them throughout the meeting, they are
apt to keep their comments on track. This also prompts team members to offer their per-
spectives and recommendations at the appropriate times.
The basic structure of the agenda will not vary much from one meeting to the next. The
following components make up most meeting agendas and occur in the order given here.

• Review the purpose of the team meeting.

• Make any organizational announcements relevant to the team’s mission.
• Call for progress reports from team members, including project objectives, accom-
plishments since the last meeting, and special assistance or resources needed for the
next steps.
• Discuss special issues, difficulties, and solutions with a focus on the positive or on
examining ways to overcome problems raised.
• Identify what needs to happen next per project or task assignment in order to
progress and continuously improve.
• Set the time and date for the next meeting and offer a preview of critical topics or
project reports to be covered.

The Safety Share. I introduced this process earlier in Chapter 7. Every team meeting
should start with this simple and brief technique. It gives safety special status and inte-
grates it into regular team business. The meeting facilitator merely asks the team members
to report something they have done for safety since the last team meeting. Team members
are encouraged to mention anything they did for safety, regardless of how insignificant it
may seem. For example, buckling a safety belt or reminding a coworker to use certain per-
sonal protective equipment qualifies as a “safety share.” Eventually, team members expect
to be asked about their safety achievements and, therefore, they prepare for the request.
This motivates some people to go out of their way for safety in order to have an impressive
safety behavior to share.
Notice that this simple technique puts an achievement perspective on safety. The focus
is not on failure—like how many injuries we have had—but on success—what we have
done for safety. For some meetings, you might vary this process slightly by asking team-
mates to share what they intend to do for safety before the next team meeting. This encour-
ages people to think proactively about safety.
When you state your intentions out loud, especially publicly, you will make a strong
mental note. There is a good chance you will remember your promise and follow through.
Plus, when you make a public commitment to do something, as discussed in Chapter 16,
peer support is activated to help influence behavior. Team members are motivated to honor
their commitment because they do not want disapproval from the team. When a person fol-
lows through with a commitment, approval from a team member rewards and supports
the person’s commitment and personal involvement for safety.
Keep discussion on track. Even with a clear agenda, meetings can get bogged down
with digressions or distracted with side conversations. The meeting facilitator needs to be
assertive at the right times to keep the discussion on track and productive. For a statement
that is off track, the facilitator might say, “Your point is interesting, Mike, but let’s table it
for now.” She might also ask, “How does your point relate to the topic of our discussion?”
At times, certain individuals will dominate a meeting with their verbal behavior. This
not only results in a lopsided discussion, it also inhibits others from stating their view-
points. Under these circumstances, some team members will consider the session a waste
of time and consensus will not be developed. So, it is important for the meeting facilitator
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 402

402 Psychology of safety handbook

to take charge with a comment like, “Pardon me, Joe, can we get some opinions from other
team members?” “Excuse me, but we need to wrap up this discussion in a few minutes
so we can move on to the next agenda item. Does anyone else want to offer their perspec-
tive briefly?”
Manage time well. How many times have you heard someone say, “I could get so
much more done in a day if I didn’t have so many meetings to go to?” Meetings have the
reputation of robbing people of valuable time. So, if time is managed well at your safety
team meetings, everyone will be appreciative. More will get accomplished, and people will
feel good about the time they gave up. I suggest the following for managing your meeting
time effectively.

• Designate a start and stop time, and make sure everyone knows what these are.
• Start the meeting on time, even if everyone has not shown up. This sets the stage for
on-time arrivals.
• Stop the meeting on time, even if every agenda item was not covered completely.
This sets the stage for efficient use of meeting time.
• Allot specific time periods to each agenda item and remind participants of these
throughout the meeting.
• If breaks are given, state a precise time for participants to return. Start the meeting
again at this time, even if everyone has not returned.
• If discussions get long, remind participants of the time remaining and the number
of agenda items left.
• Hold meetings prior to lunch time or at the end of the workday in order to provide
an incentive to get things done on time.
• Do not allow cellular phones or pagers in the meetings or you will set the stage for
distraction.

Record minutes. Decisions and assignments made during a team meeting are usu-
ally critical for team success, yet they can be easily forgotten. Therefore, it is important for
someone to document key events of the team meeting. When team members are confident
the designated “recorder” will take good notes, they will not be distracted by their own
note-taking behavior. They can listen attentively and participate actively throughout the
meeting. Later the meeting notes are made available as a permanent record of team
progress. This reminds teammates of their accomplishments and their obligations for con-
tinual success. The recorder should consider the following points when preparing the min-
utes of a team meeting.

• Include location, date, time, and names of those in attendance.

• Document the flow of the meeting in chronological order.
• Record key points and who made them.
• Summarize accomplishments from project reports and suggestions for the next
steps of the follow-up.
• Specify task assignments and who is responsible for each.
• Include the location, date, and time for next meeting.
• Distribute the minutes as soon as possible after the meeting.

Communicate between meetings. Team members need to support each other when
completing their assignments. Do not wait until the next team meeting to inquire about a
teammate’s progress on a project. Asking others how their specific assignments are going
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 403

Chapter seventeen: Promoting high-performance teamwork 403

sends the message you care about their contributions to your team. Taking the time to lis-
ten to a report of progress or actually reviewing results from a particular assignment does
more to show your concern and can be a powerful motivator.
It is awfully important to give supportive and corrective feedback for specific behav-
iors related to team assignments. As I discussed earlier in Chapter 16, behavior-based feed-
back is extremely powerful in directing and motivating desirable behavior. Team members
need to be alert to the kinds of behaviors needed from each other in order to have a suc-
cessful team. They also need to follow certain guidelines for giving effective feedback to
their teammates, as I detailed earlier in Chapter 12.
I think it is important to reiterate that there is special value in coworkers giving each
other behavior-based feedback. Coworkers’ corrective comments are less likely to come
across as a “gotcha” indictment of performance. They are more likely to support the spirit
of interdependent teamwork. Plus, fellow employees are more likely to be present when
immediate feedback is necessary. They also can best shape a message—probably without
even giving it much thought—to the expectations, abilities, and experience level of the
recipient.
Finally, encouragement or especially corrections from a coworker are more apt to be
taken as a sign of true caring for team success. Interpersonal behavior-based feedback is the
most effective way to direct and motivate teammates to do what needs to be done to make
a team successful. In other words, interpersonal feedback is crucial for reaping optimal
synergy from teamwork.

Evaluate team performance

I am sure you realize the value of performance evaluation when it is done well. In fact, the
discussion in the previous section about feedback says it all. Willing workers cannot
improve without receiving feedback directly related to their performance, and such feed-
back is only available through an objective and periodic evaluation process. Evaluation
is the key to accountability and responsibility as I detail in the next chapter. Here, I offer a
few guidelines regarding the evaluation of safety teams so they might improve their
performance.
My guess is most readers will find Figure 17.8 humorous. Why? What is the problem
with most performance appraisals? If performance appraisals were objective, fair, and
based on behavior, you would not see any humor in this illustration. Right! So, a useful
evaluation of team performance needs to be objective, fair, and related to changeable
behaviors and conditions.
Your team needs to evaluate its performance periodically in order to assess successive
improvements made possible by prior evaluations. Then, it has reason to celebrate its
accomplishments. Quality team celebrations (as discussed in Chapter 13) are the key to
enhancing team cohesiveness and mutual responsibility toward the accomplishment of
more shared goals.
It is important to realize that although performance evaluation is listed sixth in this list
of successive teamwork steps, this topic is inherent in every step. Whether selecting team
members, establishing a team charter, or setting goals and assigning task responsibilities,
evaluation plays an integral role.
Team members continually evaluate each other’s opinions and reactions throughout
group discussions in order to arrive at decisions everyone can support. The presentation of
project reports at team meeting is essentially an evaluation process. Team members
appraise whether a project is progressing as planned and decide whether the time line
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 404

404 Psychology of safety handbook

Figure 17.8 Subjective, nonbehavioral, and infrequent evaluations are not taken seriously.

needs adjusting when the results call for refinements or additions to the list of task assign-
ments. All of this involves the ongoing study and interpretation of information in order to
make the best decisions. This is evaluation in the truest sense of the word.
Overall evaluation. Figure 17.9 contains a general team evaluation form that can be
used by any safety team to assess and improve their process. The form is easy to fill out and
score, and the results can be quite diagnostic regarding which aspects of teamwork are
going well and which components need refinement or enhancement. Reviewing the results
of this evaluation survey with team members can prompt constructive discussion about
action plans needed to improve target areas pinpointed by the survey.
You will note nothing special about the various items in this survey. You will realize
you can readily refine the items or add your own in order to make the evaluation most per-
tinent to your team. Ideas for such customization will come forth naturally during team
discussion of your initial application of the survey in Figure 17.9.
Evaluating team meetings. Each of the various items on the survey in Figure 17.9
relates to a particular characteristic of the teamwork process—from deriving a mission
statement, performance goals, and task assignments to working on specific team projects.
In contrast, the survey in Figure 17.10 targets only the team meeting. Because team meet-
ings define the work of the team and can inspire or discourage members to work interde-
pendently toward synergy, it is essential to find out how the team meetings are perceived
and how they can be improved.
I suggest you periodically administer a survey to evaluate team meetings. The sur-
vey given in Figure 17.10 will work fine the first time, but your results and post-survey
discussions will probably suggest survey items to refine or eliminate and others to add. By
examining the mean score for each item of the survey, you can determine particular aspects
of your team meetings that need improvement. Through brainstorming and consensus-
building, you can define task assignment and responsibilities aimed at benefiting future
team meetings.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 405

Chapter seventeen: Promoting high-performance teamwork 405

Each item of the survey refers to a particular aspect Highly Agree

teamwork - from developing a mission statement and Agree
performance goals to completing assignments with Not Sure
adequate leadership, resources, direction, inter- Disagree
dependency, and interpersonal trust. Circle the Highly Disagree
number next to each statement to indicate how much
you agree with it.

1. Our team mission is clear and directional. 1 2 3 4 5

2. Our task assignments and team goals are specific, motivational,
achievable, and shared. 1 2 3 4 5
3. Team members are highly committed to the team's mission. 1 2 3 4 5
4. Consensus is reached about important matters without sacrificing
quality. 1 2 3 4 5
5. Our teamwork is planned organized, and completed effectively. 1 2 3 4 5
6. Team meetings are productive. 1 2 3 4 5
7. Team members are kept well-informed about team-relevant data,
events, policies, and organizational changes. 1 2 3 4 5
8. Team members know their individual roles on team assignments. 1 2 3 4 5
9. The team has authority and control over its decisions and their
applications. 1 2 3 4 5
10. Team members have adequate resources, knowledge, and skills to
accomplish their assignments. 1 2 3 4 5
11. Our team's leadership is effective and supportive. 1 2 3 4 5
12. The team assignments are fair in light of individual workloads. 1 2 3 4 5
13. I feel completely accepted as a productive member of the team. 1 2 3 4 5
14. Team members listen to each other proactively in order to
completely understand diverse perspectives. 1 2 3 4 5
15. The team capitalizes on each member's unique talents and
capabilities. 1 2 3 4 5
16. Team members give and receive behavior-based feedback in a
caring and constructive way. 1 2 3 4 5
17. Working on this team has been challenging and satisfying. 1 2 3 4 5
18. Working on this team has been a valuable learning experience. 1 2 3 4 5
19. Our team periodically reviews its progress toward accomplishing
team goals. 1 2 3 4 5
20. Our team periodically recognizes and celebrates the achievements
of process and outcome goals. 1 2 3 4 5

Figure 17.9 This generic survey can be used to assess overall team performance.

It should be team policy (or a ground rule) for members to submit potential survey
items for any team evaluation form, whether for a general evaluation or for an evaluation
of a particular aspect of teamwork.

Disband, restructure, or renew the team

Many books and manuals on teamwork discuss this final stage as the time when members
of a work team realize their work is done and adjourn or disband. This is rarely the case,
however, for safety teams. The work of these teams is never done. Consider, for example,
the seven teams defined in Figure 17.11 which I propose are needed to address compre-
hensively the human dynamics of industrial safety. For more details on these different
safety teams see Geller (1998b,c).
Specific projects or assignments may come and go, but safety teams need to work per-
sistently on their general missions in order to achieve continuous safety improvement
throughout a work culture. The membership of these teams will change periodically and
team goals will vary, but the challenges of behavioral observation and feedback, incident
analysis and corrective action, ergonomics analysis and intervention, and behavior-based
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 406

406 Psychology of safety handbook

Highly Agree
The statements below reflect ideal characteristics of Agree
team meetings. Circle the number next to each Not Sure
statement to indicate how much you agree with it. Disagree
Highly Disagree

1. The meeting room and its provisions are adequate. 1 2 3 4 5

2. Team meetings start and end on time. 1 2 3 4 5
3. The agenda for our team meetings is clear and available for all to
see. 1 2 3 4 5
4. Team members show up to meetings on time and leave on time. 1 2 3 4 5
5. We have the right number of team meetings. 1 2 3 4 5
6. Team meetings are facilitated well. 1 2 3 4 5
7. Team members listen well to each other and acknowledge good
ideas. 1 2 3 4 5
8. All team members actively contribute to consensus-building
discussions. 1 2 3 4 5
9. Discussions are well balanced across team members. 1 2 3 4 5
10. Decisions and assignments are posted for review and comment. 1 2 3 4 5
11. When team members disagree they seek consensus or win/win
compromise. 1 2 3 4 5
12. Adequate time is allowed to complete the meeting agenda. 1 2 3 4 5
13. Working safely is the number one priority in my plant. 1 2 3 4 5
14. Ideas and decision alternatives are accurately recorded on a flip
chart, overhead, or white board. 1 2 3 4 5
15. Team members periodically evaluate the success of meetings. 1 2 3 4 5

Figure 17.10 This survey can be used to evaluate team meetings.

recognition and celebration will remain. Of course, the methods and procedures used to
meet these team functions will change and, in fact, they will successfully improve if appro-
priate evaluation processes are implemented.
Safety teams will learn to work more effectively and efficiently over time, but they will
need to keep working. Even when your workplace becomes injury free, the safety teams
listed in Figure 17.11 are needed to maintain this enviable situation. Thus, this final step for
safety team success should be considered restructuring or renewing, not disbanding.
Restructuring. Restructuring could mean a change in focus, in team membership, or
in the methods and procedures the team uses to accomplish its mission. For example, after
observation and feedback teams are in operation throughout your work culture, the Safety

Safety Steering Team - oversees the effort of all other teams listed here.
Observation and Feedback Team - develops, implements, evaluates, and
refines behaviour-based observation and feedback procedures.
Ergonomics Team - conducts periodic audits of workplace settings,
evaluates employee suggestions regarding ergonomic issues, and
recommends corrective action for environment, behaviour, or both.
Incident Analysis Team - conducts fact-finding evaluations of near-hit
reports and injuries, including behavioral, environment, and person-
based factors; and recommends corrective action.
Celebration Team - plans and manages celebration events to recognize
process activities and reward achievements of milestones.
Incentives/Rewards Team - oversees the design, implementation,
evaluation, and refinement of behaviour-based incentive/reward programs
to motivate participation in designated safety-improvement activities.
Preventive Action Team - evaluates reports of rule/policy violations,
decides whether the violator should be punished, and chooses the penalty.

Figure 17.11 Various types of employee teams are needed to improve occupational safety.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 407

Chapter seventeen: Promoting high-performance teamwork 407

Steering Team changes its focus from promoting and training to advising and maintaining.
In other words, after employee teams get involved in behavior-based coaching, the chal-
lenge becomes one of sustaining the process. The prime issue changes from “How can we
teach coworkers behavior-based coaching procedures and convince work teams to use crit-
ical behavior checklists on a regular basis,” to “How can we keep work teams motivated to
keep their behavior-based observation and feedback process going.”
In the beginning, the efforts of the Incentive/Reward Team might focus on convincing
management and coworkers to substitute behavior-based safety incentives for their tradi-
tional safety incentive program that offers rewards for reductions in injury rate. If success-
ful at this, the team’s challenge changes to developing an acceptable and effective
behavior-based safety incentive program. Then, the team needs to evaluate the impact of
this incentive program and refine it for another application. This plan-implement-evaluate-
refine process needs to be repeated over the long term to maximize the beneficial impact of
behavior-based safety incentives.
Earlier in Chapter 11, I presented details on the design, administration, and evaluation
of safety incentive programs. My point here is that each of the four key phases—planning,
implementing, evaluating, and refining—implies a different team focus, along with unique
goals and task assignments. Special training, resources, and individual talents are needed
for each phase, requiring appropriate adjustment in team leadership, meeting agenda, and
task assignments. These four phases are not peculiar to a Safety Incentive/Reward Team.
They are relevant for each of the seven safety teams listed in Figure 17.11.
The best advice I can give any team for maintaining interest in their mission statement
and its relevant applications is to follow the four stages illustrated previously. Regardless
of your team mission, you need to plan or develop an action plan, implement your plan
throughout the workplace, evaluate the effects of your efforts, and then use information
from the evaluation to refine an improved application. This is, in fact, a basic principle of
behavior-based safety, incorporated in the DO IT process introduced in Chapter 8. It is key
to learning from research.
Nothing helps a team more to stay motivated and aligned with its mission than an
objective presentation of the good they have done and an opportunity to learn how to
improve their intervention and do more good. This is the essence of an objective and equi-
table accountability system, which leads to people’s responsibility for safety extending
beyond the numbers. As I detail elsewhere (Geller, 1998a), this is fundamental to attaining
and maintaining an injury-free workplace.
Renewing. When team members observe the “fruits” of their labor, their motivation
to continue their efforts is bolstered. In other words, observation of success breeds more
success. So, an optimal way to renew the confidence and purpose of a team is to display
clear and objective evidence that their efforts make a difference. More can and should be
done, however, to move teams forward with renewed concern and commitment.
Team-building sessions can be conducted with the sole purpose of restoring team
members’ motivation toward teamwork. Often, it is beneficial to hire an outside consultant
or facilitator to conduct such a session. However, it is possible your company training
department employs a person who could facilitate an effective team-building session. The
aim of these sessions is to get a team back on track and increase effectiveness by

1. Diagnosing and solving a particular problem.

2. Improving the management of a team.
3. Restoring interpersonal trust among team members.
4. Clarifying role expectations and obligations of team members.
5. Negotiating the modification or reassignment of certain team-member responsi-
bilities.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 408

408 Psychology of safety handbook

The following questions might be addressed, for example, at a team-building session

aimed at diagnosing potential problems and finding ways to improve team success.

• How are we doing?

• What problems need to be worked on?
• What opportunities should we be taking advantage of?
• What are our unique strengths and how can we better capitalize on these?
• Based on our experiences at teamwork, what should we do differently?
• What parts of our team charter need elimination or refinement?
• What needs to be added to our team charter?
• What organizational factors or barriers reduce team effectiveness?
• What changes in team leadership or membership are called for?
• What additional outside support/resources could benefit team output?

These ten questions certainly do not represent an exhaustive list. They only provide an
example of the kinds of issues that might be addressed at a team-building session. You
could break the participants into small discussion groups with different issues or questions
to address. Even pairs of team members could work as small discussion groups. After suf-
ficient discussion time (perhaps 30 minutes), call the subgroups back and have a represen-
tative report their findings or suggestions to the entire team. Then, a consensus-building
session would be useful, whereby issues and suggestions are classified into categories or
themes and specific action plans identified and agreed upon.
At follow-up team meetings, SMARTS goals can be set and tasks assigned according
to the action plans derived for increasing team effectiveness. Of course, an evaluation plan
is needed to assess whether the intervention actually increases the success of the safety
team. To the extent behaviors can be identified as barriers to team success, a practical action
plan identifies dysfunctional behaviors to eliminate and functional behaviors to support.
As such, an evaluation plan would assess whether the target behaviors change in the
desired directions.

In summary
This is, obviously, only a brief overview of basic steps involved in developing and sustain-
ing high-performance safety teams. Additional details related to each of these procedural
steps are available in other texts (e.g., Geller, 1998b; Lloyd, 1996; Parker, 1996; Rees, 1997).
You realize, of course, that effective safety teams do not develop overnight. Each process
reviewed here takes time and patient application of the various interpersonal and group
process strategies described in Sections 3 and 4 of this text, including behavior-based obser-
vation and feedback, proactive listening, directive and supportive coaching, individual
recognition, group celebrations, and actively caring performance evaluations.
The benefits of implementing these teamwork strategies will not be immediate. The
“sell” for teamwork is analogous to the “sell” for safety. Safety leaders are well aware of the
need to perform certain inconvenient, inefficient, and even uncomfortable safety-related
behaviors in order to reap the potential long-term benefits of injury prevention. Likewise,
the rewards of teamwork require substantial up-front investment in resources, time, and
collective effort.
The journey will not be easy, and interpersonal conflict and frustration are inevitable.
In fact, the early stages of team development can seem quite chaotic and nonproductive,
but this is a result of a natural process of team development. Let me explain, because know-
ing how teams mature over their existence builds understanding and patience among team
members and directs effective team leadership.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 409

Chapter seventeen: Promoting high-performance teamwork 409

The developmental stages of teamwork

Although researchers, scholars, and organizational consultants have used different terms
to refer to the four developmental stages of teamwork, there is no disagreement that these
four stages exist and call for special kinds of leadership. The most popular labels for these
progressive phases are forming, storming, norming, and performing (Tuckman, 1965).
In general, during the early stages of group interaction (forming and storming), work
groups need structure and a clear vision and mission statement. At this time, an autocratic
or directive leadership style is often most appropriate, although it is still best to get input
from team members before the first group meeting. After the group members become
familiar with each other and start implementing their assignments (the norming and per-
forming stages), a democratic leadership style is usually most effective.
Interpersonal trust is lowest during the first stage (forming) and highest during the
fourth stage (performing). Understanding the stages of team-building enables leaders to
set realistic expectations (teams do not really perform efficiently until the fourth stage) and
allows team members to feel better about their meetings because they recognize the need
to assume certain roles at each stage.

Forming
Structure and assertive leadership are important in the beginning. Group members are get-
ting to know each other, including sizing up each member’s role and potential influence in
achieving the team’s mission. Members begin to evaluate how much they can trust each other.
As illustrated in Figure 17.12, during the forming stage the leader of team meetings should

• Provide vision, structure, and clear direction.

• Allow participants to get to know one another.
• Demonstrate proactive listening skills.
• Promote active and total involvement.
• Establish ground rules for team meetings (as discussed previously).
• Create a climate of interdependency and optimism.
• Provide education and training when needed to accomplish team goals.

Figure 17.12 During the FORMING stage, the team gets its assignment and members get
acquainted.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 410

410 Psychology of safety handbook

Figure 17.13 During the STORMING stage, the team struggles over its purpose and mem-
bers compete for influence.

Storming
After a cordial beginning, the storming stage takes over (as depicted in Figure 17.13). Team
members now engage in debate, argument, conflict, and basic power struggles. At this
stage interpersonal trust is questioned because some members may attempt to assert per-
sonal control or individual superiority. Some participants get frustrated and consider the
meetings a waste of time. Strong leadership is needed at this stage to keep the group on
task. As a facilitator and teacher, the effective leader helps team members weather the
storm by

• Expressing positive expectations and optimism.

• Acknowledging that conflict is normal.
• Reminding members of the team mission and goals.
• Pointing out the value of diversity.
• Guiding discussion toward consensus.
• Getting members to assume mission-related responsibilities.
• Providing mission-related education and training.

Norming
The group members begin productive teamwork during the norming stage (as depicted in
Figure 17.14). They develop roles for working together, realize each other’s talents, and
develop mutual trust and respect. Group cohesion grows as members begin to understand
their own roles in the group process and witness interdependency and synergy. The effec-
tive leader in this stage gives up control and serves as a cheerleader and coach by

• Providing behavior-based feedback and support.

• Encouraging total participation.
• Recognizing individual and group accomplishments.
• Being flexible and allowing for less structure.
• Working to prevent groupthink (as discussed previously).
• Asking for suggestions on how to improve.
• Promoting interpersonal trust and group cohesion.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 411

Chapter seventeen: Promoting high-performance teamwork 411

Figure 17.14 During the NORMING stage, team members settle down to productive and
interdependent work.

Performing
Here a team realizes its synergistic benefits. Team thinking, team behavior, and team loy-
alty are the norm. Individuals identify with the team and take pride in team accomplish-
ments. Interdependency and interpersonal trust peak during this stage, so much so that
team members cover for one another even without request. Social loafing (Latané et al.,
1979) is not a problem, but “burnout” is possible. As illustrated in Figure 17.15, the team
leader is inconspicuous, serving as sponsor and consultant by

• Providing advice and new information as needed.

• Giving supportive and corrective feedback at the team level.
• Planning and sponsoring group celebrations of milestones.
• Allowing the team to manage itself.
• Providing relevant training and education.
• Watching for signs of burnout (e.g., negative attitudes, cynicism, pessimism).
• Keeping communication open, honest, and candid.

Figure 17.15 During the PERFORMING stage, momentum peaks and teamwork leads to
synergy.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 412

412 Psychology of safety handbook

Adjourning
I need to explain one final stage of group dynamics—adjourning. This occurs when the
group achieves its mission, celebrates its achievements, and possibly disbands. It is impor-
tant for groups to realize when it is time to change its operation or its membership. Group
members might enjoy their fellowship so much they will resist changing membership. At
this point, the team leader needs to facilitate a healthy transition to a new team mission, to
a new team with the same mission, or to complete termination.
As indicated previously, however, when referring to safety teams this fifth stage is
really about transforming rather than adjourning. Circumstances might allow you to com-
bine teams or team missions. With less need for a formal safety incentive program, for
example, an incentive/reward team could combine with the celebration team and/or the
preventive active team, as discussed previously, or an ergonomics team could be combined
with an incident analysis team.
Changing company policies, priorities, or personnel might also require teams to be
reorganized, but there will always be a need for teamwork around three critical functions

1. Overseeing and reviewing safety programs and processes (a safety steering team).
2. Observing work practices and providing feedback and coaching (an observation
and feedback team).
3. Holding people accountable to substitute safe for at-risk behavior (an accountabil-
ity/motivation team).

When it comes to industrial safety, the fifth stage of team development is much more likely
to be renewal, reorganization, or transformation than adjournment.

In conclusion
In this chapter, I have presented barriers to productive teamwork, as well as step-by-step
methods for initiating and conducting productive teamwork. Learning to work effectively
as a team takes patience. Membership and leadership on a team call for a different
approach to work than most people are used to. As summarized in Figure 17.16, the stan-
dard work of most contemporary organizations is a lot different than teamwork. Indeed,
the teamwork perspective reflected in Figure 17.16 represents a paradigm shift for many
people.
Many of us are not used to the collaborative and cooperative interdependency of high-
performance teamwork. This includes the people who hold us accountable for our work
output. Therefore, the teamwork perspective of mutual accountability for shared goals
needs to be appreciated by managers and supervisors as well as by team members. The
people in organizations who provide the resources and opportunities for teamwork need
to understand what it takes to reap the benefits of synergy, even it they are not members of
a team themselves.
Figure 17.17 depicts a typical scenario in the hustle and bustle of our everyday lives.
Everyone is doing his or her own thing from a win–lose, individualistic framework. The
outcome seems like utter chaos and leaves individuals with the impression that their per-
sonal goals are temporarily thwarted. This can lead to frustration and a bad attitude
toward the whole situation. A bad attitude can influence risky or win–lose behavior, which
in turn adds fuel to a bad attitude. This spiral of frustrated behavior feeding bad attitude,
feeding more frustration, and so on, can lead to the kind of aggressive behavior we see in
“road rage.”
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 413

Chapter seventeen: Promoting high-performance teamwork 413

Figure 17.16 High-performance teamwork requires these paradigm shifts.

Our hero in Figure 17.17 has a different perspective on the whole situation. He is able
to take a broader view of the situation and appreciate the marvelous interdependent trans-
portation system. This viewpoint, or systems-thinking paradigm, toward everyday cir-
cumstances can be greatly beneficial to the safety and health of individuals and groups.
Systems thinking benefits teamwork and it is a consequence of teamwork. Therefore,
systems thinking feeds the interdependency and collectiveness needed for high-perform-
ance teams, and productive teamwork feeds more systems thinking. This attitude-behav-
ior spiral is constructive and motivates the special commitment and dedication needed to
build and maintain successful safety teams.

Figure 17.17 Systems thinking reflects the interdependency paradigm needed for high-
performance teamwork.
820049_CRC1_L1540_CH17 11/10/00 2:07 PM Page 414

414 Psychology of safety handbook

References
Cadwell, C. M., Team Up for Success: Building Teams in the Workplace, American Media Inc., West Des
Moines, IA, 1997.
Geller, E. S., Beyond Safety Accountability: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998a.
Geller, E. S., Building Successful Safety Teams: Together Everyone Achieves More, J. J. Keller &
Associates, Inc., Neenah, WI, 1998b.
Geller, E. S., Seven teams for highly efficient organizations, Ind. Saf. Hyg. News, 32(11), 10, 1998c.
Isenberg, D. J., Group polarization: a critical review and meta-analysis, J. Personal. Soc. Psychol., 50,
1141, 1986.
Janis, I. L., Victims of Groupthink, Houghton-Mifflin, Boston, 1972.
Janis, I. L., Groupthink: Psychological Studies of Policy Decisions and Fiascoes, 2nd ed., Houghton
Mifflin, Boston, 1983.
Katzenbach, J. R. and Smith, D. K., The Wisdom of Teams: Creating High-Performance Organizations,
Harper Business, New York, 1994.
Kogen, N. and Wallach, M. A., Risk-Taking: a Study in Cognition and Personality, Holt, Rinehart, &
Winston, New York, 1964.
Langer, E. J., Mindfulness, Perseus Books, Reading, MA, 1989.
Latané, B., Williams, K., and Harkins, S., Many heads make light the work: the causes and conse-
quences of social loafing, J. Personal. Soc. Psychol., 37, 822, 1979.
Levine, J. M. and Moreland, R. L., Small groups, in Handbook of Social Psychology, Vol. 2, Gilbert, D.
T., Fiske S. T., and Lindzey, G., Eds., McGraw-Hill, Boston, MA, 1998.
Lloyd, S. R., Leading Teams: The Skills for Success, American Media Inc., West Des Moines, IA, 1996.
Parker, G. M., ed., The Handbook of Best Practices for Teams, Vol. 1, HRD Press, Amherst, MA, 1996.
Postmes, T. and Spears, R., Deindividuation and antinormative behavior: a meta-analysis, Psychol.
Bull., 123, 238, 1998.
Rees, F., Teamwork from Start to Finish, Jossey-Bass, San Francisco, CA, 1997.
Scholtes, P. R., Teams in the age of systems, in The Handbook of Best Practices from Teams, Vol. 1,
Parker G. M., Ed., HRD Press, Amherst, MA, 1996.
Stoner, J. A. F., A Comparison of Individual and Group Decisions Involving Risk, Master’s thesis,
School of Industrial Management, Massachusetts Institute of Technology, Cambridge, MA,
Unpublished, 1961.
Taylor, S. E. and Brown, J. D., Illusions and well-being: a social psychology perspective on mental
health, Psychol. Bull., 103, 193, 1988.
Torres, C. and Fairbanks, D. F., Teambuilding: The ASTD Trainer’s Sourcebook, McGraw-Hill, New
York, 1996.
Tuckman, B. W., Developmental sequence in small groups, Psychol. Bull., 63, 384, 1965.
Wallach, M. A., Kogan, N., and Bem, D. J., Group influence on individual risk taking, J. Abnorm.
Soc. Psychol., 65, 75, 1962.
Weinstein, N. D., Unrealistic optimism about future life events, J. Personal. Soc. Psychol. 39, 806,
1980.
Wellins, R. S., Byham, W. C., and Wilson, J. M., Empowered Teams: Creating Self-Directed Work Groups
That Improve Quality, Productivity, and Participation, Jossey-Bass, San Francisco, CA, 1991.
Zimbardo, P. G., The human choice: individuation, reason, and order versus deindividuation,
impulse, and chaos, in Nebraska Symposium on Motivation, Arnold, W. J. and Levine, D., Eds.,
University of Nebraska, Lincoln, 1969.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 415

chapter eighteen

Evaluating for continuous

improvement
Continuous improvement demands proper evaluation. This chapter explains how to evaluate the
impact of safety interventions from an environment, behavior, and person perspective. More employ-
ees need to contribute information pertinent to intervention evaluation. This chapter shows you how
to make this happen. The principles described here will make you a smarter consumer of marketed
safety programs and help you evaluate your own customized intervention process.

“What gets measured gets done; what gets measured and rewarded gets done well.”
—Larry Hansen

Hansen (1994) used these words in his Professional Safety article on managing occupational
safety (page 41). You have probably heard words to this effect. Indeed, they are key to any
continuous improvement effort, but there is a problem with how workplace safety is tra-
ditionally measured. As I indicated earlier in Chapter 3, too much weight is given to out-
come numbers that people cannot control directly. People must be held accountable for
results they can control. Yet, corporations, divisions, plants, and departments are often
ranked according to abstract outcome numbers like the total recordable injury rate. These
rankings often determine bonus rewards or penalties.
What behavior improves when safety awards are based only on an injury rate? If
employees can link their daily activities to safety results, then celebrating reduced injury
rates can be useful, even motivating. It is critical, however, to recognize the behaviors, pro-
cedures, and processes that led to fewer injuries or lower workers’ compensation costs.
If you do not focus on the real causes of improvement, you run the risk of actually demo-
tivating the folks deserving recognition. Employees might think continuous improvement
is caused by luck or chance—events beyond personal control. This can lead to feelings of
apathy or learned helplessness (Seligman, 1975), as I discussed earlier (e.g., see Chapters 6,
15, and 16). If we want employees to work for continuous improvement, we need to recog-
nize and reward the “right stuff.” This requires the right kind of measurement procedures.

Measuring the right stuff

Deming (1991) admonished his audiences for ranking people, departments, and organiza-
tions. In fact, he recommended that grades and performance appraisals be abolished
completely from education and business. In his words, “The fact is that performance
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 416

416 Psychology of safety handbook

appraisal, management by the numbers, M.B.O., and work standards have already devas-
tated Western industry . . . the annual rating of performance has devastated Western indus-
try . . . Western management has for too long focused on the end product” (Deming, 1986,
page 1).
Part of Deming’s rationale comes from the fact that standard approaches to measuring
academic and work performance are often subjective, relative, and not clearly related to
individual behavior. Teachers and professors, for example, use contrived distribution
curves and cut-offs to assure only a designated percentage of students can attain certain
grades. Knowledge tests are necessarily biased and imperfect assessment devices, and are
often only remotely linked to specific behaviors within a student’s control—including
attending class, taking notes, reading the textbook, and studying the material on a regular
basis. I have known many demotivated students who felt their daily efforts were over-
shadowed by the emphasis on exams.

Limitations of performance appraisals

I have met very few employees who respect and appreciate annual performance
appraisals. They do not see the evaluations as being fair, objective, and motivational. I am
sure you know department heads and supervisors who complete performance appraisals
only because it is mandatory. They are not interested in providing constructive, perform-
ance-focused feedback for continuous improvement. They do not maintain ongoing
records of employees’ accomplishments and less-than-adequate performance. Instead,
they typically wait until a few days before appraisals are due and then make their best-
guess estimate of an individual’s ranking, using nebulous performance dimensions like
“competent,” “enthusiastic,” “self-motivated,” “cooperative,” “responsive to feedback,”
and “willing to improve.”
Employees give these appraisals the attention they deserve. They simply ignore them
and try not to get too stressed out when appraisal time comes around. Some employees try
to gain control of the situation—and turn distress into stress—by “performing” for the
boss. They put on last minute “exhibitions” in order to improve their “scores.” All too often
this strategy works. As a result, these short-term exhibitions are the only behavior rein-
forced as a result of the evaluation.
The manager in Figure 18.1. is obviously soliciting biased performance feedback.
Given the principle of reciprocity discussed in Chapter 16, the suggestion boxes might con-
tain a number of signed comments near the time he must write performance appraisals.
Actually, Figure 18.1 does depict a way for employees to deliver and receive timely per-
formance-appraisal feedback whenever it is warranted. If the suggestions are based on the
principles for giving rewarding and correcting feedback reviewed in Chapter 12, then per-
formance appraisals would indeed facilitate continuous improvement.
Some managers get proactive and behavior-based with their appraisal responsibilities
and have measured and rewarded the right stuff. The process is not easy and takes extra
time, but it is well worth the effort. Here is how it works. The manager meets with employ-
ees individually at the start of the evaluation period. The pair set operational (behavioral)
and customized definitions for performance criteria. The process is very much interactive
and includes setting SMART goals (see Chapter 10), as well as developing a way to report
progress toward goal attainment and to receive feedback. The manager then keeps contin-
uous records of the employee’s behaviors related to those specific goals. The evaluation
process is, indeed, an ongoing process—not some last-minute rush to judgment.
Performance appraisals should be reciprocal. Effective managers ask employees for
input on their own management-related goals and encourage feedback on how well they
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 417

Chapter eighteen: Evaluating for continuous improvement 417

Figure 18.1 For continuous improvement, performance appraisals need to be frequent,

frank, and followed.

are progressing toward these goals. Employees should be required to complete periodic
evaluations of supervisors, reflecting expected and desired performance. This increases
their perceptions of empowerment and belonging and facilitates continuous performance
improvement on the part of the supervisors.

What is performance improvement?

If “what gets measured gets done,” what are we measuring when we measure perform-
ance? From a systems standpoint, performance means output. In the context of psychology,
performance means behavior or output from a human system. We cannot discuss measure-
ment and evaluation without an operational definition of performance. Let us examine dif-
ferent definitions more closely.
Individual performance. The first fact I learned in my introductory psychology
course was that psychology is the study of the individual. Later, each psychology course I
took in graduate school examined some aspect of individual performance (the dependent
measure) as it was influenced by particular environmental or experimenter manipulations
(the independent variables). For example, in my courses on learning, I studied the effects
of prior experiences and reinforcement history on individual performance. In social psy-
chology, I learned how other people influence an individual’s performance; and in physi-
ological psychology, I learned how specific changes to one’s nervous system through
drugs, electrical stimulation, or surgery influenced individual performance. Today, in my
management and organizational psychology courses, I teach students how an individual’s
work performance can be affected by training techniques, the three-term contingency (acti-
vator-behavior-consequence), and differences in a person’s knowledge, skills, ability, atti-
tude, and personality.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 418

418 Psychology of safety handbook

System performance. At a four-day Deming seminar, I learned that many corporate

leaders and experts in the field of organizational management—including safety manage-
ment—define performance from an industrial systems perspective. The second afternoon
was devoted to the famous red/white bead game, a demonstration to convince the audi-
ence that individual performance is a relatively insignificant determinant of organizational
or system performance.
Deming asked for 10 volunteers from the more than 600 attendees. He assigned six
volunteers the job of “willing worker,” two other volunteers became “inspectors” of
each worker’s output, another volunteer played the role of “quality control supervisor” or
inspector of the inspectors’ work, and the tenth volunteer was the “recorder.”
After assigning work duties, Deming mixed 800 small red beads with 3200 white
beads—all the same size—in a box. He showed the “willing workers” how to scoop out
beads from this box using a beveled paddle that had indentations for 50 beads. Each scoop
of the paddle was considered a day’s production. The inspectors counted the numbers of
red and white beads on each worker’s paddle, and after a reliability check by the bead
inspector, the number of red beads (considered mistakes or defects) was recorded and dis-
played publicly on an overhead projector by the recorder.
Deming played the role of the corporate executive officer and he urged the workers to
produce no defects (red beads). When workers scooped fewer than ten red beads, they
were praised by the CEO. Scooping more than 15 red beads, however, resulted in severe
criticism and an exhortation to do better, otherwise “we will go out of business.” Deming
reacted to each scoop of beads as a product of individual performance. If performance out-
come was poor, he again demonstrated the correct bead-sampling procedure or gave cor-
rective feedback to a worker during the sampling process. It is noteworthy that regression
to the mean resulted in good individual performance getting worse and poor individual
performance getting better. As discussed in Chapter 16, this can give the impression that
correcting feedback works better than rewarding feedback.
The sampling, inspecting, reliability checking, recording, public graphing of results,
and corrective action for defects continued for ten samples (representing ten days) per
worker. This portion of the demonstration lasted more than an hour and led logically to
Deming’s concluding statements that “the actual number of red beads scooped by each
worker was out of that worker’s control. The workers only delivered defects. Management,
which controls the system, caused the defects through system design.”
This demonstration has profound lessons for achieving a Total Safety Culture. As I
have discussed earlier, the typical evaluation procedure used by both the government and
private sector to judge the safety record of companies is based on organizational or system
performance—“the total recordable injury rate” or number of “OSHA recordables”—
uncontrollable by most individual workers.
Individuals are not only held accountable, perhaps even “disciplined,” for their own
injuries but also for the injuries of others, even employees outside their immediate work
group. They are often blamed for injuries caused by a number of factors outside their con-
trol, such as at-risk environmental conditions or equipment, excessive workloads, system
contingencies causing a fast work pace, and a culture that supports “rugged individual-
ism” and thwarts group cohesion and a “brothers/sisters keepers” perspective.
In July 1992, Deming wrote me this explanation to clarify his teaching intentions with
his red/white bead demonstration.

Currently, management works under the assumption that people

and not the systems they work in are responsible for performance.
We therefore reward and punish people but the system they work in
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 419

Chapter eighteen: Evaluating for continuous improvement 419

remains unchanged. The point is not that differences between indi-

viduals are unimportant in and of themselves (but) that focus on
individual differences alone yields possibilities for improvement
that are trivial compared with transformation of the entire system
that they work in. In fact, it is only individuals that can change a
given system to improve performance. (W. E. Deming, personal
communication, July 11, 1992)

System vs. individual performance. In the fall of 1991, a demonstration in my

graduate course in behavior and systems management profoundly illustrated the role
of individual performance in organizational output. It revealed the need to consider both
individual and system-level factors when evaluating safety achievement. One of my
students, Mario Beruvides, directed a probability “game” that, to his surprise and my
delight, illustrated the significant and unexpected impact of individual performance fac-
tors. Playing cards were used instead of beads to illustrate the importance of system-level
performance.
Mario asked four volunteers and myself to play the role of workers for a hypothetical
company, “Geller Inc., Non-Face Card Manufacturers.” He gave us each a sealed deck of
cards, and instructed us to open them, remove the two jokers and instruction cards and
shuffle them thoroughly. Then, we placed our decks face down on the seminar table, and
per Mario’s instructions, drew five cards from the top of our decks to represent a day’s
work. Face cards were counted as defects or errors, and the number of face cards per “will-
ing worker” was recorded and displayed to the class on a “production log.”
As in Deming’s bead game, the CEO, played by Mario, urged the workers to care for
the welfare of the company and produce as few defects as possible. Defects from each
worker were recorded, summarized, and displayed by a student assigned as “supervisor.”
The five-card drawing and recording continued for five trials, representing a work week.
After each trial, the CEO reprimanded the workers whose rate of defects exceeded the aver-
age and praised those who had just one or zero defects. To the delight of the class and sur-
prise of the CEO, my drawing always included the most face cards, or defects, and I always
received the most criticism. Before the sixth round of drawings, marking the second work
week, the CEO unveiled a company-wide incentive plan. Whenever the face cards from a
group of five “hands” are combined within 15 seconds to form a pair or three or more of a
kind, as in a poker hand, the face cards will not be counted as defects. On each trial, I con-
tinued to draw the most face cards, usually three or more, and I immediately took charge
of organizing the face cards from the other “hands” into poker combinations. This leader-
ship and teamwork decreased the number of defects per individual dramatically. Now, my
defect rate was at least as low as any other worker’s.
During the lively classroom discussion that ensued, Mario admitted that our results
were remarkably different from those obtained by the management training and develop-
ment specialist who invented the card-game analog of Deming’s bead demonstration
(Storey, 1989). In a completely random system, one worker is not supposed to always pro-
duce the most defects. Moreover, Storey reported that the incentive condition should not
reduce defects, even over five draws, presumably because some workers are reluctant to
trade cards owing to lack of experience, knowledge, or leadership.
While discussing reasons for the unexpected results from “Geller, Inc.,” I made a
provocative discovery—I had been drawing cards from a pinochle deck containing 50 per-
cent face cards. Mario had inadvertently purchased one pinochle deck and happened to
give me that deck by chance. This serendipitous innovation in Mario Beruvides’ demon-
stration resulted in profound awareness of the power in individual differences.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 420

420 Psychology of safety handbook

Mario admitted his preset expectation that the game would show random, uncontrol-
lable variation. His anticipation to find only “common cause” prevented him from recog-
nizing “special cause” contributions from individuals. He was caught completely off guard
by the success of the incentive program.
The students identified individual performance factors that made the incentive pro-
gram effective, particularly the special effects of the pinochle deck and its chance assign-
ment to the assertive class instructor. This led us to discuss how companies can benefit
from appropriate employee selection and placement, education and training, interpersonal
communication, and individual and group recognition processes. We also saw the need to
shift from our usual preoccupation with individual performance and attend to group and
system-wide performance as well. This requires an ecological analysis (Willems, 1974) of
multiple factors that can affect individual, group, and organizational performance.
A Total Safety Culture, then, requires us to balance how we measure and manage both
organizational performance and individual behavior. The challenge is to accurately attrib-
ute change in safety performance. Is it due to individual behavior, groups, or the system?
Individuals should only be held directly responsible for their own safety performance;
teams should be held accountable for outcomes directly related to their team performance.
Continuous improvement in organizational safety performance, using the yardstick of
OSHA recordables, requires improvements in the system as a whole, of which individual
employees play an integral part both as individuals and members of work teams. As I have
discussed earlier, holding individuals responsible for safety performance outside their per-
ceived control develops attitudes and perceptions, such as apathy, helplessness, and pes-
simism, that interfere with both personal and organizational safety improvement.

Developing a comprehensive evaluation process

In April and August of 1995, I had the pleasure of working with a panel of evaluation
experts to develop a set of measurement guidelines for the National Safety Council and the
Centers for Disease Control.* Our mission was to develop a handbook of practical guide-
lines that field personnel can use to evaluate the impact of an intervention to improve
safety on family and commercial farms. The results of our four days of deliberations are
documented in Steel (1996), which includes useful insight and direction for the design of
evaluation procedures.
We agreed that evaluation was essential to hold people accountable for achieving pro-
gram objectives. This particularly pertains to those developing and implementing the
intervention. The evaluation process should measure whether intervention procedures are
consistent with relevant principles and mission statements of the organization, reach the
desired audience and are implemented as planned, and are efficient and effective (Vojtecky
and Schmitz, 1986). In sum, a complete evaluation process should assess how an interven-
tion was conceived, designed, and implemented, and how efficient and effective it was.
Our panel discussed a number of approaches to assess these three basic areas—con-
ceptualization, implementation, and impact. Each suggests that certain records be kept and
examined systematically. For example, the rationale, goals, action plan, and techniques of

* These four days of meetings were organized by Dr. Sam Steel of the National Safety Council and facilitated by
Dr. Jerry Burk of the BURK group (Bakersville, CA). Other panelists included Professor Joe Miller (Penn State),
Dr. Doreen Greenstein (Cornell University), Professor Stephen J. Guastello (Marquette University), Professor
Midge Smith (University of Maryland), Jim Williams (Country Companies Bureau Insurance), and David Hard
(National Institute of Occupational Safety and Health).
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 421

Chapter eighteen: Evaluating for continuous improvement 421

the intervention must be documented in order to check if the intervention design is con-
sistent with relevant corporate principles and aims of a safety effort. When a safety steer-
ing committee deliberates to decide whether to implement a specific training process or to
hire a particular team of consultants, they essentially make this kind of evaluation. They
decide whether the goals, objectives, and procedures of the training or intervention pro-
gram fit their philosophy, purpose, and mission. Obviously, this evaluation is typically
made before training or intervention begins.
The actual intervention process must be documented to decide if the procedures are
being implemented as planned and whether the intervention is reaching appropriate num-
bers of people. Attendance records, for example, provide an efficient means of measuring
the coverage of training programs. Participation in an intervention can be measured read-
ily if the procedures include completion or delivery of materials with the participant’s
name, such as recognition thank-you cards, critical behavior checklists, or safe behavior
promise cards.
Often, it is useful to record reactions from participants and nonparticipants. What did
they think of the intervention principles and procedures? This feedback can help find ways
to increase program involvement. Thus, it is useful to conduct this kind of evaluation dur-
ing the early stages of an intervention process. Personal interviews or questionnaires can
rate levels of satisfaction or dissatisfaction and lead to “course corrections” if needed.
Assessing intervention effectiveness is most difficult but most critical. Our lengthy
panel discussions on this issue revealed the different levels of performance discussed pre-
viously—individual, group, and system. We also debated process vs. outcome issues and
the most objective and efficient ways to measure safety processes and outcomes. We con-
cluded that behavior was the optimal process measure and injury reduction the ultimate
measure of intervention success.
Attitudes, perceptions, and beliefs were presumed to influence whether the interven-
tion is accepted and has potential for long-term success. We all agreed, though, that survey
techniques to estimate these subjective person states are relatively difficult to develop and
evaluate. Issues of questionnaire reliability and validity need to be addressed as I discuss
later in this chapter.
Figure 18.2 summarizes our panel’s deliberations on how to evaluate intervention
impact. It integrates the primary issues discussed so far in this chapter. Lower levels of the
hierarchy represent process activities needed to improve the higher-level outcomes of a
safer environment and ultimate injury reduction. Immediate causes of injury reduction are
changes in environment or behavior—or both.
Let us pause for a moment to consider the cause-and-effect connection between
process and outcome. Behavior can be viewed as an outcome, shaped by a process that pur-
sues changes in employee knowledge, perceptions, or attitudes. In this case, the behavior-
change goal depends on employee participation in an intervention aimed at influencing a
person state. Of course, intervention processes can be designed to circumvent person states
and directly change behavior to reduce injuries, as discussed in Section 3 of this text.
Figure 18.2 points out the relativity of process and outcome. An education process, for
example, can lead to behavior change (an outcome), while a process to change behaviors
might result in outcome changes to the environment, say improved housekeeping.
Completing a work process in a safer environment can affect the ultimate outcome—fewer
injuries.
Figure 18.2 also reflects the three basic areas requiring attention for injury prevention—
environment, behavior, and person. This is, of course, the Safety Triad (Geller, 1989b) intro-
duced in Chapter 2 to categorize intervention strategies and referred to later in Chapter 14
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 422

422 Psychology of safety handbook

Injury
Reduction

Environment
Change

Behavior
Change

Perception, Belief,
Attitude, Intention

Knowledge Gain
(Education)

Participation
(Information Received)

Figure 18.2 Measures of intervention impact vary according to remoteness from immedi-
ate injury causation.

to classify different types of actively caring behaviors. The hierarchical levels of inter-
vention impact in Figure 18.2 reflect one or more of these three domains and suggest a
particular approach to measurement. Changes in environments and behaviors can be
assessed directly through systematic observation, but changes in knowledge, perceptions,
beliefs, attitudes, and intentions are only accessible indirectly through survey techniques,
usually questionnaires. Let us review these three basic areas of evaluation.

What to measure?
Most safety interventions focus on either environmental conditions—including engineer-
ing controls—or human conditions, as reflected in employees’ perceptions, attitudes, or
behaviors. It might seem reasonable to us to evaluate change only in the area we have tar-
geted—environment, behavior, or person state. When the target is corporate culture,
employee perceptions or attitudes are typically evaluated. If behavior change is the focus,
then behaviors are observed and analyzed in terms of their frequency, rate, duration, or
percentage of occurrence as reviewed in Chapter 8. When environments or engineering
technologies are evaluated, mechanical, electrical, chemical, or structural measurements
are taken (Geller, 1992).
I have heard culture-change consultants advocate perception surveys in place of envi-
ronmental audits. Presentations on behavior-based safety emphasize direct observations of
work practices, often in lieu of the subjective evaluation of personal perceptions and atti-
tudes. Given the need, however, for employees to “feel good” about a behavior-based
safety process and their need to participate continuously in managing and monitoring this
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 423

Chapter eighteen: Evaluating for continuous improvement 423

process, I think it is obvious that we need to check perceptions and attitudes and ongoing
behaviors. For example, a comprehensive evaluation of a simple change in equipment
design should probably include an assessment of relevant human factors like employees’
work behaviors around the new equipment and their attitudes and perceptions regarding
the equipment change.
I hope you can see that a comprehensive evaluation for safety requires a three-way
audit process covering environmental conditions, safety-related behaviors, and person
states such as perceptions, attitudes, beliefs, and intentions.

Evaluating environmental conditions

In many ways, environmental audits are the easiest and most acceptable type of evaluation.
In fact, regular environmental or housekeeping audits are already standard practice at
most companies. These evaluations can often be improved by involving more employees
in designing audit forms, conducting systematic and regular assessments, and posting the
results in relevant work areas.
A safety incentive program established in 1992 at a Hoechst Celanese plant in Narrows,
VA, awarded employees weekly “credits” for accomplishing these components of envi-
ronmental evaluation. At the end of the year, the credits could be exchanged for various
commodities containing a special safety logo. This company recognized the need to
involve as many employees as possible in the regular auditing of environmental condi-
tions, including tools, equipment, and operating conditions. Its incentive/reward process
motivated involvement.
Figure 18.3 depicts a generic environmental checklist for safety that can be used to
graph for public display the percentage of safe conditions and the percentage of potential
corrective actions taken for at-risk tools, equipment, or operating conditions. My associates
at Safety Performance Solutions typically teach work teams the rationale behind the envi-
ronmental checklist and then assist them in applying the checklist in their plant. Employees
then customize the checklist and graphing procedures for their particular work areas.
Regular audits and feedback sessions increase accountability for environmental factors
that can be changed to prevent an injury.
The property damage incident. A provocative book by Bird and Germain (1997)
focuses on environmental assessment, in particular, the need to investigate thoroughly the
“property damage accident.” Then, the property damage needs to be fixed in order to pre-
vent workplace injury. The authors claim the investigation and correction of property dam-
age or property in need of repair are key to improving workplace safety. Yet, the property
damage incident is sorely overlooked.
If you are underwhelmed by this so called “missing link,” I understand. I felt the same
when Bird first related his passionate thoughts about the property damage accident. I did
not understand the profound implications of this concept until reading his book and dis-
cussing it with line workers. For example, when I introduce the property damage incident
at my workshops and seminars, line workers in attendance show special interest. They
often testify to the extreme amount of property damage at their work sites, including stock-
piles of broken ladders, tools in disrepair, machine guards that do not work properly, and
dents in equipment, walls, and vehicles. Each dent signals an incident, perhaps a near hit,
that was not investigated.
Front-line workers also verify the dramatic impact of property damage on their work
demeanor which, in turn, influences their attitude about safety. To them, unrepaired prop-
erty damage signifies that “management doesn’t care about our work situation,” or “it’s
okay to damage property as long as we meet production demands.”
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 424

424 Psychology of safety handbook

Observer: Date: Time:

Department: Building: Floor: Area:

*Corrective
Operating Conditions/Tools & Equip. Safe At-Risk Actions Taken
Electrical wiring (properly enclosed)
Air nozzles (limited to 30 P.S.I.)
Chemicals (exposure concern)
Eyewash station
Emergency shower
Barricades (in place where necessary)
Storage of materials (neat/safe)
Hazard Communication labels
(appropriate)
Floors (dry)
Exits, aisles, sidewalks and
walkways (clear of debris)
Lighting (adequate)
Housekeeping (satisfactory)
Tools (safe operating condition)
Guards (adequate and in place)
Fire extinguisher (monthly inspection)
Fire extinguisher (in appropriate location)
GoJo (safe operation)
GoJo driver (license in possession)
Tow truck (safe operation)
Tow truck driver (license in possession)
Chairs in (safe condition)

Totals

Total Safe Observations

Percent Safe Conditions: x 100= %
Total Safe Observations + At-Risk Observations
Total Corrective Actions Taken:
Percent Corrective Actions Taken: x 100= %
Total At-Risk Observations

* Please list and define the corrective actions taken on back of this sheet.

Figure 18.3 An environmental checklist can be used to evaluate the safety of tools, equip-
ment, and operating conditions.

During the breaks at my seminars, participants tell me many personal stories about
property needing repair and how it adversely affects their safety-related behavior and
attitude. They disclose incidences in which equipment that did not work properly was a
root cause of a serious injury. In one case, a drawer that got stuck frequently was responsi-
ble for a serious injury to a worker’s elbow and costly surgery. From the time the cabinet
was new, the drawer had never worked properly, but the damage was never reported.
Sometimes the drawer moved in and out smoothly, but most of the time it got stuck and
needed a severe pull. The injury occurred because the maintenance worker pulled on the
drawer with extreme force. The drawer slid smoothly out of the cabinet and the employee’s
elbow smashed against a sharp object. The safety team leader who told me this story
declared that our discussion about the property damage incident made his participation at
the daylong seminar very worthwhile, even if he learned nothing else.
Years ago in a university research project, my students and I validated the popular slo-
gan, “litter begets litter.” In other words, we showed empirically that planting litter in com-
mercial settings led to more littering behavior (Geller et al., 1977). These research findings
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 425

Chapter eighteen: Evaluating for continuous improvement 425

can be generalized to the workplace with the message, “property damage begets more
property damage.” Comments from workshop participants clearly support this slogan.
The Heinrich Triangle. Most safety professionals are familiar with “Heinrich’s Law.”
As illustrated in Figure 18.4, Heinrich proposed over 60 years ago a 300 : 29 : 1 ratio
between “near-miss” incidents, minor injuries, and major injuries (Heinrich, 1931; Heinrich
et al., 1980). Ever since, safety professionals have been encouraged to investigate near hits
in order to reduce minor and major injuries. Heinrich also estimated that 88 percent of all
near hits and workplace injuries resulted from unsafe acts. As a result, some presentations
of “Heinrich’s Law” add a wider base to the triangle with the label “unsafe acts.”
It is interesting that the 300 : 30 ratio of near hits to injuries is referred to as a “law,”
when, in fact, it was only an estimate. It was not until more than 30 years later that this
“law” was actually tested empirically. As Director of Engineering Services for the
Insurance Company of America, Frank E. Bird, Jr., analyzed 1,753,498 “accidents” reported
by 297 companies. These companies employed a total of 1,750,000 employees who worked
more than three billion hours during the exposure period analyzed.
The result was a new ratio. For every 600 near hits, there will be 30 property damage
incidents, 10 minor injuries, and 1 major injury. Now, we see the critical link of property-
damage incidents which were unidentified in Heinrich’s estimates.
Bird and Germain (1997) are quick to point out that the 600 : 30 : 10 : 1 ratio depicted in
Figure 18.5 was obtained from incidents reported and discussed, some during 4000 hours
of confidential interviews by trained supervisors. It is likely the base of the Bird Triangle is
much larger than 600. Notice the dramatic difference between the two ratios. Suppose the
number of minor injuries in Bird’s ratio were multiplied by a factor of 3 to make it compa-
rable to the 29 minor injuries in the Heinrich ratio. Then you would have 1800 near hits and
90 property damage incidents per 30 minor injuries. Do you see how linking property
damage to workplace injuries can encourage more incident reporting, analysis, and cor-
rective action?

1
Major Injury

29
Minor Injuries

300
Near-Miss Incidents

Figure 18.4 Heinrich (1931) estimated the ratio between “near-miss” incidents, minor, and
major injuries.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 426

426 Psychology of safety handbook

1
Major
Injury

10
Minor Injuries

30
Property Damage Incidents

600
Near-Miss Incidents

Figure 18.5 In 1969, Bird investigated the ratio between near hits, property damage,
minor, and major injuries.

Property damage is a physical trace of an incident and the precursor of an injury. It

can also be diagnostic in a comprehensive analysis as I detailed earlier in Chapter 9. The
behavior that contributed to a property damage incident was likely unintentional human
error. Punishment is probably not warranted as discussed in Chapter 11. However, failure
to report such property damage and assure correction is intentional and shows disregard
for safety. Punishment might be warranted. That is the paradigm shift enabled by a focus
on the property damage incident.
An industrial example. Let me tell you a true story to illustrate how a focus on prop-
erty damage can make a difference. Walking along a scaffold, a worker slipped on a metal
plate and almost fell several stories to his death. Fortunately, he was able to catch himself
with his arms and pull himself back onto the walkway. Members of the safety committee
decided to do more than the typical “reactive investigation” of this incident. They did not
simply blame the welder responsible for securing the plate. Instead, they looked for other
contributing factors to prevent similar mishaps.
Guess what they discovered? At least a dozen people had slipped on that same loose
plate and said nothing about it. No one reported a near hit. They did not want to report a
“near miss,” implying careless or thoughtless behavior.
However, if the loose plate had been reported as property damage that needed imme-
diate repair, the idea of individual blame would have been removed. Thus, not to report
such property damage and assure correction should be considered careless and thoughtless.
Removing personal blame from incidents that set the stage for personal injury will
enable more proactive reporting, evaluating, and correcting. When your periodic environ-
mental audits show less and less property damage, you can be assured you are preventing
injuries. In fact, I am convinced this is actually a more reliable and valid metric for safety
improvement than the standard injury and illness rates derived from employees’ self
reports and visits to the plant infirmary. So a comprehensive safety measurement system
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 427

Chapter eighteen: Evaluating for continuous improvement 427

should include systematic audits of damage to the work environment. The repair of
environmental damage should be continuously tracked as an ongoing measure of safety
improvement.

Evaluating work practices

The systematic auditing of work practices was the theme of Chapters 8, 9, and 12. In
Chapter 8, I introduced the overall DO IT process—“D” for define target behaviors, “O” for
observe target behaviors, “I” for intervene to increase safe behavior or decrease at-risk
behavior, and “T” for test (or evaluate) the impact of your intervention. How to develop
two types of observation checklists was discussed in Chapter 8—a generic version for basic
work practices applicable anywhere, such as prescribed lifting techniques and the use of
certain personal protective equipment (see Figures 8.12 and 8.13); and a job-specific check-
list for particular tasks, like the safe driving checklist my daughter and I developed (see
Figure 8.10).
The coaching process detailed in Chapter 12 also discussed how to develop and apply
both generic and job-specific checklists for one-on-one observation and feedback sessions
with coworkers (see Figures 12.9 and 12.10). As mentioned previously, using a behavioral
checklist to observe and evaluate ongoing work practices is the type of performance
appraisal that can lead to continuous improvement.
Chapter 9 on “Behavioral Safety Analysis” was all about evaluation from the perspec-
tive of work practices. A series of ten questions was proposed for conducting a step-by-step
examination of the situational, social, and personal factors influencing at-risk behavior.
Answers to these questions (see Figure 9.6) provide direction for deriving the most cost-
effective corrective action plan.
My experience has been that group auditing and analysis—of people using vs. not
using personal protective equipment, for example—is readily accepted by most employees
and relatively easy to implement. However the one-on-one audits and analysis in coaching
(see Chapter 12) with employees observing other employees who volunteer to be moni-
tored, are not readily accepted in some corporate cultures. A plant-wide education and
training intervention is often necessary to teach the rationale and procedures for this eval-
uation process and to develop the necessary interpersonal understanding, empathy, and
trust. My associates and I at Safety Performance Solutions have developed a Safety Culture
Survey to assess if a corporate culture is ready for one-on-one coaching. This involves eval-
uating person factors related to safety.

Evaluating person factors

As discussed earlier in this text, person factors refer to subjective or internal aspects of
people. They are reflected in commonly used terms like attitude, perception, feeling,
intention, value, intelligence, cognitive style, and personality trait. You can find many
surveys that measure specific person factors of target populations ranging from children to
adults. Some of these factors are presumed to be traits, others are considered states. It is
important to understand the difference when you consider the evaluation potential of a
particular survey.
Person traits. Theoretically, traits are relatively permanent characteristics of people;
they do not vary much over time or across situations. The popular Myers-Briggs Type
Indicator, for example, was designed to measure where individuals fall along four dichoto-
mous personality dimensions: extroversion vs. introversion, sensing vs. intuition, thinking
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 428

428 Psychology of safety handbook

vs. feeling, and judgment vs. perception (Myers and McCaulley, 1985). The various combi-
nations of these attributes allow for 16 different personality types, each with special char-
acteristics. These traits are presumed permanent and unchangeable, determined largely by
physiological or biological factors.
Because traits are relatively permanent, questionnaires that measure them cannot
gauge the impact or progress of a culture-change intervention. Trait measures serve as a
tool to teach individual differences but, in safety management, their application is limited
to selecting people for certain job assignments (Geller, 1994). This is very risky, though, and
should not be done without understanding the validity limitations described later in this
chapter.
Person states. Person states are characteristics that can change from moment to
moment, depending on situations and personal interactions (as discussed in Chapter 15).
When our goals are thwarted, for example, we can be in a state of frustration. When expe-
riences lead us to believe we have little control over events around us, we can be in a state
of apathy or helplessness. Person states can influence behaviors. Frustration, for example,
often provokes aggressive behavior (Dollard et al., 1939); and perceptions of helplessness
inhibit constructive behavior or facilitate inactivity (Abramson et al., 1989).
In contrast, certain life experiences can affect positive person states, such as optimism,
personal control, self-confidence, and belonging. These, in turn, boost constructive behavior.
This was the indirect approach to increasing actively caring behavior discussed in Chapter
16. The woman in Figure 18.6 is in a positive person state referred to as optimism. She might
drive her friends crazy, but research has shown that healthier and happier people are more
often in this state (Peterson, 2000; Seligman, 1990). Plus, as I discussed earlier in Chapter 15,
optimistic people are more likely to actively care for the health or safety of others.
Measures of person states can be used to evaluate perceptions of culture change and to
pinpoint areas of a culture that need special intervention attention. Like most culture sur-
veys, our Safety Culture Survey asks participants to answer questions on a five-point con-
tinuum (from highly disagree to highly agree) about their perceptions of the safety culture.
Issues include the perceived amount of management support for safety, the willingness of

Figure 18.6 An optimistic person state facilitates happiness, perseverance, achievement,

health, and actively caring.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 429

Chapter eighteen: Evaluating for continuous improvement 429

employees to correct at-risk situations and look out for the safety of coworkers, the per-
ceived risk level of the participant’s job, and the nature of interpersonal consequences fol-
lowing an injury.
Our survey also measures factors that increase one’s willingness to actively care for
another person’s safety. These include self-esteem, belonging, and empowerment as
detailed in Chapter 15. Sample items from our survey that measure the actively caring per-
son states are given in Chapter 15. They were adapted from professional measures of these
characteristics, and have been evaluated for reliability and validity, as discussed later.
Figure 18.7 contains 20 items from the safety perception and attitude portion of our survey.
You will note nothing very special about the items in this scale. They ask employees to react
to straightforward statements about safety management and improvement.

Highly Agree
Agree
Not Sure
Disagree
Highly Disagree

1. The risk level of my job concerns me quite a bit. 1 2 3 4 5

2. When told about safety hazards, supervisors are
appreciative and try to correct them quickly. 1 2 3 4 5
3. My immediate supervisor is well informed about
relevant safety issues. 1 2 3 4 5
4. It is the responsibility of each employee to seek out
opportunities to prevent injury. 1 2 3 4 5
5. At my plant, work productivity and quality usually
have a higher priority than work safety. 1 2 3 4 5
6. The managers in my plant really care about safety
and try to reduce risk levels as much as possible. 1 2 3 4 5
7. When I see a potential safety hazard (e.g., oil spill),
I am willing to correct it myself if possible. 1 2 3 4 5
8. Management places most of the blame for an accident
on the injured employee. 1 2 3 4 5
9. "Near misses" are consistently reported and investigated
at our plant. 1 2 3 4 5
10. I am willing to warn my coworkers about working
unsafely. 1 2 3 4 5
11. Employees seen behaving unsafely in my department
are usually given corrective feedback by their
coworkers. 1 2 3 4 5
12. Compared to other plants, I think mine is rather risky. 1 2 3 4 5
13. Working safely is the number one priority in my plant. 1 2 3 4 5
14. I have received adequate job safety training. 1 2 3 4 5
15. Many first aid cases in my plant go unreported. 1 2 3 4 5
16. Information needed to work safely is made available to
all employees. 1 2 3 4 5
17. Management here seems genuinely interested in
reducing injury rates. 1 2 3 4 5
18. Safety audits are conducted regularly in my department
to check the use of personal protective equipment. 1 2 3 4 5
19. I know how to do my job safely. 1 2 3 4 5
20. Most employees in my group would not feel comfortable
if their work practises were observed and recorded
by a coworker. 1 2 3 4 5

Figure 18.7 These questionnaire items measure personal perception regarding the safety
of an organization and were selected from the Safety Culture Survey developed by Safety
Performance Solutions, Inc. With permission.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 430

430 Psychology of safety handbook

You could compare employees’ reactions to the items in Figure 18.7 before and after
implementing a safety improvement process. Studying reactions prior to an intervention
helps identify issues or work areas needing special attention. This information can lead you
to choose a particular intervention approach or to customize one. Data from a baseline per-
ception survey might even indicate that a culture is not ready for a given intervention
process, suggesting the need for more education and discussion to get employees to “buy
in,” for example.
The impact of an intervention can be measured by comparing perception surveys
given before and after implementation. At one plant, our baseline Safety Culture Survey
indicated that secretaries had below-average levels of perceived empowerment, as
assessed by the measures of self-efficacy, personal control, and learned optimism described
earlier in Chapter 15. A special recognition intervention was devised and later the survey
was administered again to measure changes in the five actively caring person states as well
as safety perceptions and attitudes.
Reviewing the results of this survey helped employees understand the relationship
between work practices, perceptions, and attitudes. It also revealed that the recognition pro-
gram improved some person states both for employees who received and administered it.
Limitations of questionnaires. Although measures of person states are more useful
for safety management than measures of person traits, there are critical limitations to both.
Given the increasing popularity of these evaluation tools among safety professionals, I
urge you to give these limitations serious thought. First, surveys of person factors, traits, or
states are neither as objective nor as reliable as audits of behaviors. Second, results are not
as straightforward and easy to analyze and interpret as information from behavioral obser-
vations. Third, developing, administering, and interpreting surveys designed to evaluate
person factors relevant for safety requires a basic understanding of reliability and validity.

Reliability and validity

What is the practical value of a questionnaire or survey? This can be assessed with a vari-
ety of research methods and statistical tools. Many are beyond the scope of this text, but a
few basic concepts are pertinent. First, questionnaires to measure person factors can be reli-
able, though not valid. To be valid, they must be reliable. A reliable survey gives consistent
results. You assess this by comparing answers across different survey items that suppos-
edly measure the same factor or by comparing two different administrations of the same
survey.
If a scale indicated that I weighed 250 pounds on Monday, 249 pounds on Wednesday,
and 251 pounds on Saturday of a given week, the scale would get a high reliability rating,
even though I really weigh in at about 180 pounds. This scale gave consistent results; it is
reliable, but the numbers are invalid.
Validity refers to whether the survey instrument measures what it claims to measure.
There are three basic types of validity for a measurement scale, each with particular exper-
imental and statistical methodologies for evaluation. These are content validity (do relevant
experts agree that the survey appears to measure what it is supposed to measure?), crite-
rion validity (can scores from the survey be used to predict individual behavior or per-
formance?), and construct validity (are the relationships found with the survey consistent
with relevant theory and research?).
In the weight example, the scale looks like it measures the correct weight of a person
standing on it (content validity), but if results were compared with readings of other scales
or with results of another estimate of weight, the numbers would not correspond.
Construct validity would be questionable. Plus, this weight scale could not predict other
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 431

Chapter eighteen: Evaluating for continuous improvement 431

variations in individual performance, such as running speed or calories consumed per day,
presumed to be influenced by weight. Criterion validity could not be demonstrated.
Criterion validity can be evaluated with two different validity-testing techniques: con-
current and predictive validity. Concurrent validity is most frequently used and refers sim-
ply to the relationship between the scale results—in this case my weight in pounds—and
another simultaneous assessment of the factor the scale is supposed to measure. This
assessment could result from measuring my weight with another scale or visually estimat-
ing my weight.
Predictive validity is much more difficult to assess. It refers to the ability of an evalua-
tion tool to predict future behavior. In our example, testing predictive validity requires that
the scale results (how much someone weighs) be compared with a future outcome that the
scale is purported to predict, such as a person’s quickness, general health, or diet.
Determining if the results of a perception survey predict the degree of employee involve-
ment in a safety-improvement effort is another example of testing for predictive validity.
The construct validity of a scale is usually evaluated with tests of convergent and
divergent validity. Convergent validity refers to the extent that other measures of the same
construct (for example, a visual estimate of weight) relate to each other. Divergent validity
indicates the extent that scores from surveys unrelated to the construct do not correlate with
survey scores related to the construct. In other words, divergent validity implies the extent
a particular questionnaire measures special characteristics not measured by other scales.
Regardless of how a person scale is used, whether for teaching, pinpointing problems,
or measuring trends or change, it is important to use measurement tools with acceptable
levels of reliability and validity. However, if results of a person scale are only used to teach
diversity or to measure group change, statistically unacceptable levels of reliability or
validity will not cause harm or injustice to someone. This is obviously not the case, how-
ever, when a person scale is used to select individuals for a particular job as some evalua-
tion tools on the safety market purport to do (Burke, 1994; Job Safety Consultant, 1995;
Krause and Kamp, 1994). When using a questionnaire to identify individuals, it is critical
that prior research with the scale has demonstrated acceptable levels of criterion and con-
struct validity.
What is acceptable? The basic statistic used to measure reliability and validity is a
correlation coefficient, which describes the relationship between two sets of survey scores
with a number ranging from 1.0 to 1.0. The greater a positive number (between 0 and
1), the greater the direct relationship between measures (a high score on the predictor scale
indicates a high score on the criterion). Negative correlations (between 0 and 1.0) indicate
an indirect (or inverse) relationship (a high score on the predictor scale indicates a low
score on the criterion), and the closer the correlation to 1.0, the greater the inverse
relationship.
The closer the correlation to  1 or 1 and the larger the sample, the more confidence
one can have that the relationship is true. However, it is important to realize the difference
between statistical significance and practical significance. A correlation of 0.30, for example,
would be statistically significant for most statistical tests and sample sizes. In some cases,
however, this number might not represent practical significance, given that the square of the
correlation coefficient indicates the degree of variance overlap between the two measures.
For example, a correlation of 0.30 between the results of a safety perception survey and
other measures of safety, such as employees’ frequency of coaching sessions completed or
percentages of at-risk behavior per observation period, sounds good until you realize that
only 9 percent of the variance in one measurement device could be accounted for by the
other (0.302  0.09 or 9 percent). In this case, 91 percent of the variance in people’s safety
perception scores could not be explained by the other estimate of a person’s safety.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 432

432 Psychology of safety handbook

The importance of construct validity. It is possible that a direct relationship between

a predictor (such as the measure of accident proneness) and a criterion (such as the num-
ber of at-risk behaviors or recordable injuries) can be found (predictive validity) without
supporting the underlying principle(s) or theory. This would indicate the absence of con-
struct validity. Suppose, for example, an individual could figure out how to answer the sur-
vey questions in order to receive a favorable score. Then, construct validity would be
questionable, even if criterion validity were high.
Every survey I have seen that attempts to assess injury proneness has items which are
transparent and enable a respondent to “fake good.” Faking good is called impression
management in the research literature (Schlenker, 1980; Umstot, 1984) and leads to signifi-
cant bias in many survey administrations. If a scale to measure injury proneness is used to
select individuals for a job, and if the respondents know this, impression management
could easily bias the test results. For example, honest risk seekers who confine their thrills
to off-the-job free time may be rejected, while deceptive individuals covering up their risk-
seeking tendencies may be selected.
The ethics of survey administration require that respondents give their informed con-
sent to be tested and that they know how their answers will be used. This second ethic,
termed “demand characteristics” in the research literature (Orne, 1969), is problematic for
an employee selection device with transparent items. A significant relationship between
scores on such a survey and other indicators (convergent validity) could reflect principles
and theory other than those presumed.
For example, the significant correlation could reflect motivation or intelligence factors
rather than actual injury propensity, thus concurrent validity would be shown without con-
struct validity. This kind of hiring survey might select individuals who are most skilled at
impression management rather than less injury prone.
Here is another important point to consider. Research accomplished to test the validity
of a survey must occur under the same demand characteristics as the proposed use of the
survey. For example, if a safety survey is to be used to select individuals for a particular job,
then tests of validity should occur with respondents knowing the survey will be used to
select them for a job. This particular demand characteristic is often difficult to pull off in a
testing situation; you should check for it when reading the technical manual accompany-
ing a survey used to screen individuals.
I hope you can see that determining an acceptable level of validity is not a straightfor-
ward process. It requires you to distinguish between statistical and practical significance
and to carefully evaluate the experimental methodology used to assess validity.
Unfortunately, this kind of evaluation requires special training and experience beyond the
purview and expertise of most safety professionals. I think it is wise to seek advice from an
appropriate consultant—one who has nothing to gain if the target survey is used or not
used. A more cost-effective approach is to study the research literature associated with the
survey. If there is research published in a peer-reviewed scientific journal, it is likely the
survey has passed at least one rigorous test of validity.
If the survey has not been reviewed and accepted by the scientific community, it should
not be used to select out individuals for any purpose. There are too many potential biases
in using any survey of subjective human factors, and to use a survey without sufficient
validity to identify individuals is just too risky, especially for a safety professional. It is far
safer to use surveys of person factors to identify person characteristics contributing to the
increase or decrease of injuries, and to monitor the impact of interventions designed to
change attitudes, perceptions, intentions, or mood states. Although it might sound good to
use a person scale to select safe employees, I urge extreme caution.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 433

Chapter eighteen: Evaluating for continuous improvement 433

Social validity. We must consider one final type of validity when evaluating a safety
intervention. It comes from researchers and practitioners in behavior-based psychology
(Baer et al., 1968) and refers essentially to practical significance. It includes using rating
scales, interviews, or focus-group discussions to assess

1. The social significance of goals. . . .

2. The social appropriateness of the procedures. . . .
3. The social importance of the effects . . . (Wolf, 1978, page 207).

It is critical to obtain social validity evaluations from the actual recipients of the program
or intervention, as illustrated in Figure 18.8.
A comprehensive evaluation of an intervention’s social validity is more complex than
it seems, as I reviewed in a lengthy monograph on social validity (Geller, 1991). There are
many perspectives on what makes intervention goals socially significant, procedures
socially appropriate, and results socially important. Plus, there are various ways to assess
the social validity of an intervention, from unobtrusive behavioral observation to surveys
of reactions from those involved in the process.
To understand various perspectives on social validity, I have found it useful to consider
the four basic components of an intervention process: selection, implementation, evalua-
tion, and dissemination (Geller, 1989b). Selection refers to the importance or priority of
the target problem and the population addressed. The social validity of selecting work-
place and community safety as an intervention target is obvious, given that unintentional
injury is responsible for the greatest percentage of years of potential life lost before age
65 (Sleet, 1987).

Figure 18.8 Social validity assessment should target the recipients of an intervention.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 434

434 Psychology of safety handbook

Assessing the social validity of the implementation stage includes evaluating the goals
and procedures of the program plan—how acceptable are they to potential participants
and other parties, even those tangentially associated with the intervention (Schwartz and
Baer, 1991)? In the case of a corporate safety program, this means obtaining acceptability
ratings not only from employees, but also employees’ family members and customers of
the company. This assessment clearly relates to one of the evaluation recommendations
from the panel of experts mentioned earlier in this chapter. That is, are intervention proce-
dures consistent with an organization’s values, and do they reach the appropriate audi-
ence? One difference is that Schwartz and Baer recommend a broader assessment of
acceptability from both direct and indirect consumers of the intervention process.
The social validity of the evaluation stage refers, of course, to the impact of the inter-
vention process. This includes estimates of the costs and benefits of an intervention (dis-
cussed later in this chapter) as well as measures of participant or consumer satisfaction.
Figure 18.9 depicts the various ways to evaluate program impact. The far-left column of
Figure 18.9 lists aspects of a work setting that can be measured before, during, and after
implementation of a safety intervention. The order of these characteristics reflects the eval-
uation hierarchy presented earlier in Figure 18.2. The top items are directly measurable and
relate most immediately to the ultimate purpose of a safety intervention—injury preven-
tion. Therefore, improvements in injury-related incidents, behaviors, and environmental
conditions would indicate more social validity for the evaluation phase than would benefi-
cial changes in attitudes, perceptions, knowledge, opinions, or program participation.
The center column of Figure 18.9 includes examples of the type of measurement tool or
index that can measure the dimensions in the left column. Each of these measurement
devices can be classified according to three basic sources of data: direct observation,
archival data (obtained from examining plant documents, memos, and government
reports), and self report (such as verbal answers to interview questions or written reactions

What is Measured? How is it Measured? What is the Score?

Near hit reports Frequency and type of near hits
Injury-related Number and type of injury-
Injury reports
Incidents producing incidents
Worker compensation costs Monetary expenditures
Percentage of safe conditions
Observation of worksite
per opportunity
Environment
Percentage of items in proper
Housekeeping audit
location
Percentage of safe behaviors per
Direct observation
opportunity
Behavior
Number of items corrected for
Corrective action survey
safety
"Safety score" reflecting overall
Attitudes Questionnaire safety attitude, perception, or
Perceptions person state
Person States Statements of specific and
Interview
general attitudes about safety
Questionnaire Percentage correct
Statements indicating
Knowledge
Interview awareness of a hazard or
a safety procedure
Opinions Questionnaire Opinion score
Direct observation Number of participants
Participation
attendance records per opportunity

Figure 18.9 A variety of measurement devices can be used to evaluate intervention

impact.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 435

Chapter eighteen: Evaluating for continuous improvement 435

on questionnaires). If the direct observations and archival data are reliable, these measures
have greater social validity than self-report measures (Hawkins, 1991).
The third column in Figure 18.9 reflects the scores or numbers obtained from various
measurement devices. What are meaningful and useful numbers? Of course, they need to
be reliable and valid, but they also need to be understood by the people who use them. If
they are not, the evaluation scheme cannot lead to continuous improvement.
Meaningless numbers also limit the dissemination potential and large-scale applica-
bility of an intervention. This is the social validity of the dissemination stage of the process.
Next, I want to address confusing characteristics of statistical analysis that reduce social
validity as it pertains to large-scale acceptance and application of an intervention process.

Cooking numbers for evaluation

The issue of using socially valid numbers reminds me of insightful lessons from a good
friend and eminent behavioral scientist, Ogden R. Lindsley (Professor Emeritus, University
of Kansas). Lindsley completed his graduate studies at Harvard University with B. F.
Skinner and has dedicated most of his creative and prolific research and scholarship to
applying the principles and procedures of behavior-based psychology to improving edu-
cation (Lindsley, 1992). Now he spends considerable time and effort sharing his profound
knowledge with corporations, especially regarding the use of behavior-based principles to
evaluate organizational change at individual and group levels.
Figure 18.10 illustrates wisdom from Lindsley that is relevant to our discussion of eval-
uation. Program evaluators often lose important information from their observations and
reduce social validity by “cooking” their “raw data” with complex statistical tests. That is,
they transform the numbers from the field into composite scores and test results in order
to determine whether the differences (or similarities) they found are statistically signifi-
cant. Often, the outcome of these statistical tests are meaningless to those responsible
for improving the process. This is similar to using injury-rate data to try to change safety
behaviors.

Figure 18.10 Raw data are cooked before an evaluation.

820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 436

436 Psychology of safety handbook

Computers and software programs make it easy to crunch numbers that might be sta-
tistically significant, but all too often they have limited practical utility—or social validity.
Many of my graduate students, for example, become remarkably skilled at running
impressive statistical tests on raw data. They speak eloquently and persuasively about the
step-by-step procedures required for a particular statistical technique but many are
stumped by my simple question, “What does it mean? How does your interesting and com-
petent statistical evaluation apply in the real world?” Often, the theory or rationale behind
statistical results is lost when software programs mix numbers with a formula and churn
out pages of computer output.

What do the numbers mean?

Years ago, my students and I posted our evaluation data daily on large graphs. Posted
numbers represented the frequency or percentage of certain target behaviors observed. The
baseline data, obtained from observations prior to implementing an intervention, told us
the amount and variability of the desired or undesired behaviors we were targeting. This
feedback was invaluable when deciding whether to intervene. If intervention was called
for, the graphs helped us decide when to act. For example, it is easier to show the clear
impact of an intervention if the baseline data is relatively stable at the start of the process.
Posting numbers from daily observations allowed us to monitor the progress of our
attempts to improve performance. Sometimes we modified intervention procedures as a
result. We all paid attention to these daily numbers, getting a surge of motivation when-
ever they improved.
When the numbers stabilized, we typically withdrew the behavior-change procedures
and noted whether the target behavior(s) reverted to baseline levels. Obviously, many real-
world applications of an intervention process does not include a Withdrawal Phase. We
implemented this evaluation approach for research purposes. If target behaviors returned
to near-baseline levels after the intervention process was removed—which was indeed the
case for most of the behavior-change techniques described in Chapters 10 and 11—we had
the most convincing demonstration of the intervention’s impact (Hersen and Barlow, 1976;
Kazdin, 1994).
I am sure you have noted the similarity between this scheme for intervention evalu-
ation and the DO IT process introduced in Chapter 8. The “T,” or test phase, of DO IT
implies a comparison of the target behaviors before and after the “I,” or intervention phase,
is initiated.
Note also that I have used the past-tense to describe this evaluation process. Now, my
students punch numbers into high-tech computer programs to allow for multiple statisti-
cal transformations. Still, I recommend the earlier, straightforward, and low-tech approach
over our current protocol.
We still record behavioral frequencies or percentages on a session-by-session basis, but
these graphs are not processed daily for ongoing feedback and evaluation. Instead, weeks
after a study has been completed—and all data has been entered into the computer—we
get a final printout. The figures are quite attractive and ready for publication, but I miss the
frequent posting of daily observations and the personal reaction and interaction it stimu-
lated. Obviously, we could still post the results of our daily observations and reap the ben-
efits of both evaluation approaches. Several times my students and I have discussed the
need to revive the old low-tech approach, but other priorities take control over the people
empowered to make this happen.
My students feel pressure, actually mandates, from other faculty to conduct sophisti-
cated statistical conversions of research data, so computer processing takes priority. I hope
most of you feel no need to “cook” raw data and will reap the benefits of the low-tech way.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 437

Chapter eighteen: Evaluating for continuous improvement 437

Figure 18.11 Cooked data are often confusing and require special consultation.

You will be spared the interpretative expertise of a statistical consultant, as depicted in

Figure 18.11, plus the raw numbers will be most useful for directing action and motivating
continued involvement.
The problem with percentages. Lindsley (1997) even advises against one of the sim-
plest calculations—percentages. Percentages are not problematic when comparing static
levels of performance across different groups or conditions. However, they can be very
misleading when monitoring fluctuations in the performance of a single group over sev-
eral observation sessions. This is because percent change is not symmetrical. If you add 20
percent to a number (like 100) and then subtract 20 percent from the result, you are not back
from where you started. You will be below the start point (at 96). In fact, if you start at 100
and then add and subtract 20 percent on ten trials, you will end up at 66.6—a result well
below the starting point.
Another problem with using percentages, of course, is the disregard for sample size.
The basketball player who goes to the foul line once and makes one of two free throws is
at the same 50-percent effectiveness level as the player who makes 10 of 20 free-throw
attempts. However, it is likely these players are not equivalent performers. The player
going to the line more often might be more aggressive and valuable to the team effort. The
same holds true in safety. Performing one at-risk behavior out of two opportunities seems
less problematic in terms of exposure and potential injury than performing 50 at-risk
behaviors on 100 trials.
Changes in percentages. Another problem is that people can be confused by changes
in percentages. Suppose during a coaching observation session you record 60-percent safe
behavior on your critical behavior checklist. Then, during subsequent sessions, you
observe 70-percent safe behavior. How do you report this increase in safe behavior per-
centage? Is this a 10 percent increase? I have seen many people report changes in percent
this way, because it seems logical. By this logic, however, the percentage increase from
20 percent safe to 30 percent safe is the same as the percentage increase from 60 to 70 per-
cent safe. There is a flaw in such logic, at least as conceptualized by a percent-change
transformation.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 438

438 Psychology of safety handbook

A percentage increase from one number to another implies reference to the starting
number. Thus, the percentage increase from 20 to 30 is 50 percent, calculated by subtract-
ing 20 from 30 to determine amount of increase and then dividing 20 (the beginning
amount or reference number) into 10 (the amount of increase). By similar logic and calcu-
lations, the percentage increase from 60 to 70 is 16.7 percent (10 divided by 60).
Now, watch what happens when calculating the percentage decrease from the second to
first numbers given above. It can get quite confusing when realizing that the decrease per-
centages are not the same as the increase percentages. For example, the percentage decrease
from 30 to 20 is 33.3 percent (10 divided by 30), and the percentage decrease from 70 to 60
is 14.3 percent (10 divided by 70). Of course, the root problem of this confusion is the
change in the reference number when going up vs. going down. That is why starting at 100
and adding and subtracting 20 percent results in 66.6 after 10 trials (Lindsley, 1997).
Calculations of percent change are logical and understandable when we keep the ref-
erence number (or baseline) in mind. Thus, a 50 percent increase from 100 is 150, and a 50
percent decrease from 100 is 50. Problems can occur if we lose sight of the starting point or
reference number. Some people find it disconcerting, however, that adding 10 to 10 (as in
10 percent safe) comes across as 100 percent improvement, while adding 10 from a starting
point of 80 percent safe might get reported as only 12.5 percent improvement. With this
percent-change logic, safety excellence rewards are more easily won by organizations that
start with the worst safety record.
One approach to handling the confusion of shifting the point of reference when calcu-
lating change percentages is to report change in percentage points. With this logic, the dif-
ference between percentages are reported as an increase or decrease in percentage points
with no reference to starting (or baseline) levels. For example, increases in percentage safe
behavior from 10 to 20, 50 to 60, and 80 to 90 would all be reported as increases of 10 per-
centage points. Likewise, changes from the second to first number in each pair would be
referred to as decreases of 10 percentage points.
Is it fair to determine safety improvement awards on the basis of change in percentage
points? Do you believe it is critical to consider an organization’s baseline level at the begin-
ning of the evaluation period? Should the company with the lower baseline, and thus
greater opportunity to improve, have an advantage? Actually, the better one’s safety
record, the more difficult it is to improve. Not only are the percent-change calculations
biased against organizations with enviable baseline records, the reality of making a notice-
able improvement stacks the deck for companies with the most improvement needed.
Now, I hear Deming (1991, 1992) warning us again to “abolish the ranking of people,
departments, and organizations.”

An exemplar
I think you can see how important it is to realize even the simplest transformation of raw
numbers can add confusion, eliminate instructive information, and detract from construc-
tive feedback. I received the display depicted in Figure 18.12 five years ago from Kitty
Morgan of the ExxonMobil Polyolefins Plant in Baton Rouge, LA. She was rightfully proud
of the safety-belt use at her plant and the benefits of unannounced buckle-up checks con-
ducted by the plant’s “Family Safety Council.” Indeed, this is an example of the DO IT
process and, in this case, it might have saved a life. Shortly after leaving the plant follow-
ing his night shift, an employee was involved in a crash with an oncoming vehicle travel-
ing without headlights. He sustained only minor injuries because he was buckled up,
something he claimed always to do “strictly because of the seat belt inspections” (personal
communication from Kitty Morgan, November 3, 1995).
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 439

Chapter eighteen: Evaluating for continuous improvement 439

Figure 18.12 Posting frequencies of drivers buckled vs. not buckled is more informative
than percentages. Data adapted from the Baton Rouge Polyolifins Plant, 1995. With per-
mission.

The numbers in Figure 18.12 provide more information than percentages. The plant’s
employees look at such a display and know exactly where they stand regarding safety-belt
use. They see how many coworkers are like them, buckled or unbuckled; and they clearly
see a dwindling of peer conformity in the at-risk category over four gate checks, from 120
to only 13. It is noteworthy that the first three data points were obtained before the recent
safety-belt use law went into effect in Louisiana. Thus, the high voluntary buckling up at
Paxon is something to be proud of.
When I received this data, I immediately calculated the buckle-up percentages of 79,
91, 95, and 97 across the four check periods, respectively. My extensive personal history of
examining safety-belt use percentages had conditioned me to do this. After completing
these calculations and obtaining results I could compare with prior results, I realized the
display of raw numbers Morgan sent me was, in fact, the most complete and clear way to
share the results. Often the simplest approach to an evaluation process has the most social
validity, from a dissemination perspective.

Evaluating costs and benefits

The ExxonMobil Family Safety Council can readily justify its safety-belt promotion efforts.
The benefits from protecting one employee from serious injury, perhaps a fatality, are
clearly greater than the effort and financial costs of the program. For example, it has been
conservatively estimated by the National Highway Traffic Safety Administration (1984)
that every employee fatality costs industry more than $120,000 in direct payments, prop-
erty damage, and medical care; and it would take $2,400,000 in sales at a 5 percent margin
to offset such a loss. Plus, the costs can be much greater for a nonfatality if, for example, the
injury causes permanent disability and requires lengthy rehabilitation. Also, estimating
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 440

440 Psychology of safety handbook

direct costs does not include the expense of hiring and training replacements—and associ-
ated productivity losses.
The benefits of an occupational safety program are illustrated dramatically by the
cost–benefit evaluation of the company-wide safety-belt program initiated at Ford Motor
Company. This corporate-wide program was initiated in the spring of 1984, before any
state passed a mandatory seat-belt use law. Safety committees at the Ford plants developed
their own safety-belt programs, based on the behavior-change principles described in
Section 3 of this text. The combination of activator interventions like awareness sessions,
incentives, and commitment pledge cards with rewarding consequences more than tripled
safety-belt use among Ford employees. After only one year, this increase saved the lives of
at least eight employees, spared about 400 others from serious injury, and reduced corpo-
rate costs by 10 million dollars. After the second year of increased belt use, the benefits
more than doubled, amounting to direct monetary savings of approximately $22 million
(Geller, 1985; Gray, 1988).
This cost –benefit analysis was invaluable in sustaining top-management support for
the campaign at Ford. It also provided motivating feedback to the many employees who
developed, managed, or evaluated the plant-based interventions. It is important to main-
tain records of direct and indirect costs associated with injuries—and with injury preven-
tion—even if these calculations are only estimates. Comparing estimates with the costs of
implementing and maintaining particular safety programs illustrates specific benefits and
justifies continued program support, especially if you can show that the program has sub-
stantially reduced injury frequency and costs.
When calculating program costs, you should document every expense, including pro-
motional materials, teaching aids, evaluation supplies, rewards, media expenditures, and
wages paid for program assistance. Employees’ time away from the job to plan, present,
evaluate, or participate in the program should be estimated, even if the time is voluntary—
for example, on evenings or weekends. If you are comprehensive when calculating pro-
gram costs, then you are justified in estimating the numerous direct and indirect costs
resulting from a job-related injury.
Injury records should be consulted before and after an intervention process has been
started in order to show the savings from fewer work-related injuries. Direct costs that
should be calculated per injury include

• Wages paid to absent employees (workers’ compensation).

• Property damage.
• Medical expenses.
• Physical and vocational rehabilitation costs.
• Survivor benefits.

These direct costs may be the proverbial tip of the iceberg when considering the indi-
rect or hidden costs of business disruptions caused by a loss-time injury. However, indirect
costs can be difficult or impossible to calculate. You should, however, try to estimate such
costs in these categories

• Overtime pay used to cover the work of an injured employee.

• Scheduling work tasks to cover for an injured employee.
• Additional administrative hassles, extra wages, training time, and inefficient work
associated with temporary replacements.
• Special costs of losing a skilled employee.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 441

Chapter eighteen: Evaluating for continuous improvement 441

• Extra time from work supervisors to schedule shift changes, temporary replace-
ments, or employee training necessitated by the absence of an injured worker.
• Retraining and readjusting for employees returning to work after an extended
absence. (A permanent disability from the injury might call for a new job assign-
ment.)
• Special costs for extensive recruitment procedures and on-the-job training for per-
manent replacements for injured employees who do not return to work or who
return with a permanent disability.
• Special administrative costs to investigate and document the incident and medical
treatments for compliance with state workers’ compensation laws and with other
state and federal regulations, such as OSHA standards.

Considering the direct and indirect costs to a company from work-related injury is
overwhelming in at least two respects. It is certainly an intimidating chore to estimate these
costs, and it is stunning to think that the corporate losses from one employee’s injury can
be so great. These dollars clearly justify considerable intervention activity. Just anticipating
the negative consequences from a work-related injury should motivate support and par-
ticipation in proactive efforts. This is the first critical step in “selling” safety—the theme of
the next chapter.

You cannot measure everything

Deming (1991, 1992) condemned grades and performance appraisals because they provide
a limited picture of an individual’s contributions and potential. They might also constrain
the number and type of interventions used to improve the quality of a work culture. If,
for example, the only procedures implemented to improve safety are those that allow
for objective measurement, the number and quality of safety interventions is severely
restricted.
In Chapter 16, for example, I discussed a number of ways to increase actively caring
behaviors directly, through applications of learning and social influence principles, and
indirectly, through improving the five personal states that increase willingness to actively
care. It is impractical and impossible to measure the impact of many of these interventions.
Should we avoid doing so just because we cannot measure their occurrence and impact?
Deming (1991, 1992) explained there are many things we should do for continuous
improvement without attempting to measure their impact. We should not do these things
only to influence performance indicators, but because they are the right things to do for
people. You might never be able to measure the impact of treating an employee with spe-
cial respect and dignity, but you do it anyway. Such treatment may, in fact, contribute to
achieving a Total Safety Culture but you will never know it. Likewise, you will never know
how many injuries you prevent with proactive actively caring behaviors and you will
never know how much actively caring behavior you will promote by taking even small
steps to increase coworkers’ self-esteem, empowerment, and sense of belonging. You need
to continue doing these things anyway. Many things that cannot be measured and
rewarded still need to get done.
I first realized the fallacy of the common management dictum, “You can’t manage
what you can’t measure” in May 1991, when attending my first Deming workshop. Each
afternoon during the 3-day seminar, the 600 participants split into small work groups to
discuss various topics. On the second day, the groups were asked to “explain why
it is wrong to suppose that if you cannot measure something, you cannot manage it
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 442

442 Psychology of safety handbook

effectively—a costly myth.” Except for our facilitator, this statement caused mental conflict
or cognitive dissonance (Festinger, 1957) for all of us. Each of us had been living by the
manage-by-measurement philosophy and believed in the opening quote of this chapter.
We entertained the points reviewed earlier in this chapter, that

• Outcome measures are usually imperfect and deceptive estimates of critical process
activities.
• Outcome measures remotely connected to process activities provide minimal if any
useful direction and motivation.
• There are numerous immeasurable things we need to do on a regular basis to help
people optimize their system.
• We cannot expect external rewards and recognition for most of the important man-
agement-related behaviors we need to do. In other words, we cannot measure
everything for which we need to be accountable and feel responsible (Geller, 1998).
We need to start with the right vision, theory, and principles, and hold ourselves
accountable with internal consequences.

Our discussion group did, however, agree that we should try to develop objective,
process-based measures for our quality—or safety—objectives. Although we cannot meas-
ure every important process directly, defining and tracking desired actions or behaviors
guide proper procedures and motivate continuous improvement. In other words, the quote
from Hansen (1994) at the start of this chapter is accurate, but it does not say it all. Many
factors affect performance. Not only is it impossible to monitor all of them, it is often
impossible to identify the specific change in performance that led to an improved system.

In conclusion
At the start of this Handbook, I explained the fallacy of basing decisions on common sense.
Rather than adopt intervention programs that sound good, we need to use procedures that
work, but how do we know what works? Of course, you know the answer to this question.
Only through rigorous program evaluation can we know whether an intervention is worth
pursuing. Now comes the more difficult question. What kind of program evaluation is
most appropriate for a particular situation?
Actually, every chapter of this text has addressed program evaluation in one way or
another. Early on, I explained the need for achievement-oriented methods to keep score of
your safety efforts. This enables people to consider safety in the same work-to-achieve con-
text as production and quality. This implies, of course, the need for program evaluation
numbers people can understand and learn from. This is how evaluation leads to continu-
ous improvement.
Throughout this text, I have referred to published research in order to justify psycho-
logical principles or recommendations for intervention procedures. Information presented
in this text is founded on rigorous evaluation, not common sense. Evaluation techniques
used in published research is, indeed, more rigorous and complex in terms of reliability,
validity, and statistical analysis than those needed for continuous improvement of real-
world safety programs. The basic principles and issues presented in this chapter, however,
are relevant to both researchers (seeking to contribute to professional scholarship) and
practitioners (seeking continuous improvement of an intervention process).
To publish their findings, researchers need to demonstrate reliability and validity of
their measures and find statistical significance. However, they can and do ignore several
evaluation principles presented in this chapter. For example, their measures typically
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 443

Chapter eighteen: Evaluating for continuous improvement 443

target only one dimension (environment, behavior, or person factors), are short-term
(applied for a limited number of observation sessions), are subjected to statistical transfor-
mations and analyses that take substantial time to complete and are not readily understood
by the average person, and often do not include a cost–benefit analysis. You see, reports of
their procedures and results only need to be understood and appreciated by a select, often
esoteric, group of professionals who specialize in the particular issue or problem addressed
by the research.
However, you cannot overlook the basic principles presented in this chapter when
evaluating practical interventions to achieve continuous improvement. Data collection
procedures and statistical analyses often can be less rigorous, but safety practitioners need
to address several important issues often bypassed by professional researchers.
Specifically, they need to

1. Define the level of performance targeted by the intervention, while appreciating

limitations in attacking individual vs. organizational performance.
2. Use measures for the three dimensions of safety improvement—environment,
behavior, and person factors.
3. Apply process measures periodically over the long term, especially checks on envi-
ronmental conditions and work practices.
4. Include a cost–benefit analysis to justify continued intervention and evaluation
efforts.
5. Keep score with numbers that are both meaningful to all program participants and
provide direction for intervention refinement.

These last two principles are critical to meeting the challenge addressed in the next chap-
ter—obtaining and maintaining support for an effective intervention process.

References
Abramson, L. V., Metalsky, G. I., and Alloy, L. B., Hopelessness depression: a theory-based subtype,
Psychol. Rev., 96, 358, 1989.
Baer, D. M., Wolf, M. M., and Risley, T., Some current dimensions of applied behavior analysis,
J. Appl. Behav. Anal., 1, 91, 1968.
Bird, F. E., Jr., and Germain, G. L., The Property Damage Accident: the Neglected Part of Safety, Institute
Publishing Inc., Loganville, GA, 1997.
Burke, A., Putting job candidates to the safety test, Ind. Saf. Hyg. News, 28(4), 19, 1994.
Deming, W. E., Drastic changes for western management, abstract for the meeting of TIMS/ORS at
Gold Coast City, Australia, July 1986.
Deming, W. E., Quality, productivity, and competitive position, four-day workshop presented in
Cincinnati, OH, by Quality Enhancement Seminars, Inc., Los Angeles, CA, May 1991.
Deming, W. E., Instituting Dr. Deming’s methods for management of productivity and quality, two-
day workshop presented in Washington, D.C. by Quality Enhancement Seminars, Inc., January
1992.
Dollard, J., Doob, L., Miller, N., Mowrer, O. H., and Sears, R. R., Frustration and Aggression, Yale
University Press, New Haven, 1939.
Festinger, L., A Theory of Cognitive Dissonance, Row, Peterson, Evanston, IL, 1957.
Geller, E. S., Corporate Safety Belt Programs, Virginia Polytechnic Institute and State University,
Blacksburg, VA, 1985.
Geller, E. S., Applied behavior analysis and social marketing: an integration for environmental
preservation, J. Soc. Iss., 45, 17, 1989a.
Geller, E. S., Managing occupational safety in the auto industry, J. Organ., Behav. Manage., 10(1), 181,
1989b.
820049_CRC1_L1540_CH18 11/10/00 2:08 PM Page 444

444 Psychology of safety handbook

Geller, E. S., Ed., Social Validity: Multiple Perspectives, Monograph Number 5, Society for the
Experimental Analysis of Behavior, Inc, Lawrence, KS, 1991.
Geller, E. S., A critical review of human dimension presentations given at the 1992 National Safety
Council Congress and Exposition, technical report for Exxon Chemical Americas, Virginia
Polytechnic Institute and State University, Blacksburg, VA, 1992.
Geller, E. S., What’s in a perception survey?, Ind. Saf. Hyg. News, 28(11), 11, 1994.
Geller, E. S., Beyond Safety Accountable: How to Increase Personal Responsibility, J. J. Keller &
Associates, Inc., Neenah, WI, 1998.
Geller, E. S., Witmer, J. F., and Tuso, M. E., Environmental interventions for litter control, J. Appl
Psychol., 62, 344, 1977.
Gray, D. A., Introduction to invited address by E. S. Geller at the annual National Safety Council
Congress and Exposition, Orlando, FL, October 1988.
Hansen, L., Rate your B.O.S.S.—benchmarking organizational safety strategy, Prof. Saf., 39(6), 37,
1994.
Hawkins, R. P., Is social validity what we are interested in? Argument for a functional approach,
J. Appl. Behav. Anal., 24, 240, 1991.
Heinrich, H. W., Industrial Accident Prevention: a Scientific Approach, McGraw-Hill, New York, 1931.
Heinrich, H. W., Petersen, D., and Nestor, R., Industrial Accident Prevention—a Safety Management
Approach, 5th ed., McGraw-Hill, New York, 1980.
Hersen, M. and Barlow, D. H., Single Case Experimental Designs: Strategies for Studying Behavior
Change, Pergamon Press, New York, 1976.
Job Safety Consultant, Hiring safety-conscious employees—Can it be done? Is it legal?, August, Issue
267, 1, 6, 7, 1995.
Kazdin, A. E., Behavior Modification in Applied Settings, 5th ed., Brooks/Cole, Pacific Grove, CA, 1994.
Krause, T. R. and Kamp, J. Viewpoint: selecting safe employees, Ind. Saf. Hyg. News, 28(10), p. 25,
1994.
Lindsley, O. R., Precision teaching: discoveries and effects, J. Appl. Behav. Anal., 25, 51, 1992.
Lindsley, O. R., Performance is easy to monitor and hard to measure, in Handbook of Human
Performance Systems, Kaufman, R. Thiagarajan, S., and MacGillis, P., Eds., University Associates,
San Diego, CA, 1997.
Myers, I. B. and McCaulley, M. H., Manual: a Guide to the Development and Use of the Myers-Briggs
Type Indicator, Consulting Psychologists Press, Palo Alto, CA, 1985.
National Highway Traffic Safety Administration, The Profit in Safety Belts: a Handbook for Employees,
DOT HS 806 443, Washington, D.C., U.S. Department of Transportation, 1984.
Orne, M. T., Demand characteristics and the concept of quasi-controls, in Artifact in Behavior
Research, Rosenthal, R. and Rosnow, R., Eds. Academic Press, New York, 1969.
Peterson, C., The future of optimism, Am. Psychol., 55, 44, 2000.
Schlenker, B. R., Impression Management, Brooks/Cole, Monterey, CA, 1980.
Schwartz, I. S. and Baer, D. M., Social validity assessments: is current practice state of the art?,
J. Appl. Behav. Anal., 24, 189, 1991.
Seligman, M. E. P., Helplessness: On Depression, Development, and Death, W. H. Freeman, San
Francisco, 1975.
Seligman, M. E. P., Learned Optimism, Alfred A. Knopf, New York, 1990.
Sleet, D. A., Motor vehicle trauma and safety belt use in the context of public health priorities,
J. Trauma, 27, 695, 1987.
Steel, S., Ed., Evaluating the Effectiveness of Agricultural Safety and Health Initiatives, The National
Safety Council, Itasca, IL, 1996.
Storey, C., Excellence: it’s in the cards, Train. Dev. J., September, 46, 1989.
Umstot, D. D., Understanding Organizational Behavior, West Publishing, St. Paul, MN, 1984.
Vojtecky, M. A. and Schmitz, M. F., Program evaluation and health and safety training, J. Saf. Res.,
17, 57, 1986.
Willems, E. P., Behavioral technology and behavioral ecology, J. Appl. Behav. Anal., 7, 151, 1974.
Wolf, M. M., Social validity: the case for subjective measurement or how behavior analysis is find-
ing its heart, J. Appl. Behav. Anal., 11, 203, 1978.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 445

chapter nineteen

Obtaining and maintaining

involvement
You cannot effectively put to use the principles in this book without ongoing support from both man-
agers and employees. This chapter focuses on ways to initiate and maintain that support, including
ways to promote leadership, build commitment and involvement, expand the scope of interventions,
reduce active resistance, and sustain momentum.

“Character consists of what you do on the third and fourth tries.”—James Michener

Culture change is never quick, never easy. The “quick-fix” illustrated in Figure 19.1 is
clearly ridiculous. As absurd as this notion is, it comes to us naturally. We want speedy
solutions to difficult challenges. It is easy to lose patience, enthusiasm, and optimism along
the way. After all, our society demands immediate gratification—just look at all the movies
and television shows that begin with dramatic problems and come to happy endings
within 30 to 90 minutes. Plus, the faster we solve any problem, the sooner we experience
rewarding positive consequences.

Figure 19.1 There is no quick-fix solution to culture change.

820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 446

446 Psychology of safety handbook

Of course, we know the insinuation depicted in Figure 19.1 is preposterous. Common

sense tells us, correctly this time, that fairy-tale solutions are utopian and unrealistic. It
makes much more sense to plan and work for the small wins that come from SMART goals
(Chapter 11). We need to remember the well-known answer to the silly question, “How do
you eat an elephant?” One bite at a time!
This chapter brings us to the point where we start pulling things together. I want to dis-
cuss the broad challenge of initiating a culture-change process aimed at achieving a Total
Safety Culture. First, come general guidelines for starting a process and maintaining sup-
port. Then, I address concepts of leadership, communication, and resistance. You will not
find step-by-step cookbook procedures here; a generic recipe is just not available. Instead,
take the principles and procedures presented in prior chapters, add the information found
here, and you will be well on your way to an innovative experience in safety improvement.

Starting the process

Management support
You cannot do without it. How many times have you heard, “Whatever management really
pushes and supports will happen.” Implicitly, “Whatever upper management does not
push and does not support will fail.” If managers emphasize housekeeping, quality con-
trol, or cost-reduction, improvements in these areas are likely to follow. Strong top-down
support, involvement, and commitment alone will not make a campaign succeed, but
they are essential ingredients. Plus, management and labor must collaborate to make the
process work.

Creating a safety steering team

A Safety Steering Team plays a critical role in developing a Total Safety Culture, providing
policy-making, oversight, and general support. All this is simply more than any one per-
son can handle. At the start, there are at least two functions this team serves.

1. Content function groups typically produce better ideas, and more creative problem
solutions than individuals working alone.
2. Process function-group-based solutions or decisions lead to more commitment and
enthusiastic involvement than individual solutions or decisions handed down.

Before creating a Safety Steering Team, it is important to look at the existing commit-
tee structure in your organization. You do not want to duplicate the efforts of the safety
department or some other relevant standing committee. For example, a current employee
team might be able to take on the responsibility of coordinating efforts to achieve a Total
Safety Culture.
Careful planning is needed to determine

• What is the mission of the Safety Steering Team?

• What are the ground rules for how it operates? (See Chapter 17.)
• What are the group’s limitations or restrictions?
• What are the priorities?
• Who should be on this team?
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 447

Chapter nineteen: Obtaining and maintaining involvement 447

Developing evaluation procedures

“Did it do any good?” This is the central question the Safety Steering Team must be pre-
pared to answer about any intervention. Chapter 18 offered guidelines for evaluating
impact. At the start of an evaluation process, these questions need to be answered.

• What indicators should we look at—behaviors, attitudes, opinions, or outcomes?

• When should we measure?
• What types of data should we collect and analyze?
• What is the cost of this evaluation process?
• How should we summarize and display results?

Setting up an education and training process*

Ensuring that employees learn key principles and procedures to improve safety is a major
responsibility for the committee. At the minimum, the following elements should be incor-
porated into planning an effective education and training program**:

• Develop education content and procedures.

• Plan the education and training process.
• Plan for follow-up sessions.
• Identify and prepare instructors.
• Measure the impact of the program.

Let us discuss each of these elements in a bit more detail, keeping in mind that
my suggestions need to be customized at the plant level to get the most “bang for your
buck.”
Identifying and preparing instructors. Selecting the “right” instructors is critical
because teaching is at the heart of effective education and training. If the teachers do a poor
job, they undermine other training tools such as videos and booklets. You should consider
these factors when choosing instructors.

• Prior experience in educating or training.

• Current level of teaching ability (aptitude and achievement).
• Credibility with employees to be educated.
• Level of motivation and interest in doing the instruction.
• Prior familiarity with psychology, especially behavior-based principles.
• Belief that the principles and procedures can help achieve a Total Safety Culture.

To help selected teachers prepare for their task, they will need to understand the prin-
ciples and relevant procedures in this text so they can represent them accurately to the

*As discussed in Chapter 9, I distinguish between educating and training, with educating referring to teaching the
principles or theory behind a process or set of procedures and training referring to teaching techniques or proce-
dures with hands-on experience and behavioral feedback.
** A process is a long-term continuous effort, mission, or set of procedures which might include specific programs
with a beginning and end. Thus, a training process might include a particular training program that is refined or
updated on a regular basis.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 448

448 Psychology of safety handbook

participants; feel comfortable with the specific process of the plant-wide education and
training which, ideally, they will help to develop; practice basic communication skills,
demonstrate leadership at meetings; and learn how to facilitate discussions—particularly
if they have little experience at stand-up training and small-group meetings.
In-house training staff can help volunteers build their teaching skills. These topics are
often covered in various corporate training programs. Plus, instructional programs are
readily available from outside vendors or consultants.*
Developing course content. When considering the specific topics to include in each
instructional session, you need to address these points.

• What are the specific goals of a particular session–for example, to teach principles,
train procedures, or build commitment and motivation?
• In addition to this text, what sources are available for relevant content and support,
such as local case studies of behavior-based safety?
• What relevant films, videotapes, and other instructional materials are available?**
• To meet session goals, what key content points should be covered, and in what
order?
• What specific plant facts, statistics, and case studies can be incorporated into the
session?
• How much information can be covered effectively in one session?

Involve your selected instructors as much as possible in developing the specific edu-
cation and training plan.
Planning the instructional process. “Everything was covered but nobody paid
much attention.” This is a common complaint about education or training. The transla-
tion is good content is important, but not sufficient. You have got to “package” your
content and present it in a way that hits home with participants. Obviously, you want
them to practice what is preached. Most instructors know their material; the challenge
lies in conveying that knowledge. How do you get your message across? Here are some
points to consider (for additional practical information on effective group presentation, see
Drebinger, 2000).

• An interactive/participative approach is typically more effective than a “top down”

lecture coming from the podium.
• Like a good pitcher, change speeds. Do not rely on one pitch, one way of presenting
information.
• It is easier to involve small groups of participants than large ones.
• Regardless of the main objective of a session, some initial awareness raising makes
participants more receptive to the content.
• Integrate demonstrations into the program.
• If possible, have participants practice the skill taught with appropriate feedback in
the classroom or on the job.

* My associates at Safety Performance Solutions (SPS) provide education and training and guide organizations
through the entire process reviewed here. For more information, write Safety Performance Solutions, 1007 N.
Main St., Blacksburg, VA 24060, call (540) 951-SAFE (7233), e-mail safety@safetyperformance.com, or visit our
website at safetyperformance.com.
** A variety of instructional materials on behavior-based safety and actively caring, including videotapes, audio-
tapes, and facilitator guides, is available through Safety Performance Solutions, Inc.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 449

Chapter nineteen: Obtaining and maintaining involvement 449

• Resolve those administrative questions, including where and when the instructional
sessions should be held, and how long they should take.

Evaluating effectiveness. As covered in Chapter 18, you cannot overlook measure-

ment. At this point in the overall intervention process, you need to determine impact of
education and training on participants’ knowledge, skill, and attitudes.
Knowledge of content can be assessed the old-fashioned way, through written tests
given at the end of a session. Skills can be evaluated by systematic behavioral observation
in the classroom or out in the workplace. Participants’ reactions to the session can be meas-
ured with brief questionnaires. For this you should keep in mind

• Brevity.
• Choice and complexity of wording.
• Combining objective ratings and written comments.

Figure 19.2 depicts a simple evaluation form to assess reactions to an instructional ses-
sion. Distributing a form like this shows that teachers care about improving the sessions.
Often, you will find that participants give useful information to develop action plans for
implementing principles. The form also solicits suggestions for improving subsequent ses-
sions. Consider using the form shown in Figure 19.2, or a refinement, to continuously
improve your education and training process.

Training Evaluation

Please evaluate this training session along the characteristics listed below (circle the
number corresponding to your answer).

Effectiveness of the presentations:

1 2 3 4
Ineffective Somewhat Definitely Highly
Effective Effective Effective

Satisfaction with what you learned:

1 2 3 4
Not Somewhat Definitely Highly
Satisfied Satisfied Satisfied Satisfied

Usefulness of material presented:

1 2 3 4
Not Useful Somewhat Definitely Highly
Useful Useful Useful

The training session was:

1 2 3
Too Long About Too Short
Right
If too long or short, what would you add, delete, or change?

• From your viewpoint, what are the most significant principles or procedures
you learned from this session?
• What needs to happen over the course of the next six months to implement
these principles or procedures?
• What would you like to learn next about the "Psychology of Safety?"
• Is there anything else you would like to tell us?

Figure 19.2 Use this questionnaire to evaluate a safety training session.

820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 450

450 Psychology of safety handbook

Sustaining the process

Continued upper-management support. Many safety programs get a big send-off, only
to drop off the radar screen. Then, it is “out of sight, out of mind” as some new program is
pushed. This is why safety is often derided for its “flavor of the month” approach. So how
do we sustain a safety process? (Note that I prefer the concept of “process” to “program”
here because processes flow on while programs begin and end.) First and foremost, we
need continued and visible support from top management. If management endorses the
process on an ongoing basis, it can become integrated into normal plant operations.
The Safety Steering Team needs to work hard to convince managers that their com-
mitment is fundamental to the process—not only to get it going, but to keep it going. Here
are some thoughts on maintaining that all-important backing of top managers.

• First, you need to gain access to upper management. Identify a manager to cham-
pion your cause in the executive offices. Bring him or her into the loop, ask him or
her to attend all team meetings.
• Keep managers informed. Submit team reports to them on a regular basis but do not
overwhelm busy managers with minutiae.
• Ask for their input, keep managers involved. Solicit their comments and “concerns”
about the process you have underway. Of course, you should be doing this with all
levels of the organization to create a top-to-bottom sense of ownership.
• Promote and market your efforts. Publish articles and announcements about the
safety process in the employee newsletter on a regular basis.
• Keep at it. Identify the benefits of your process and continue to “sell” them to upper
management.

Awareness support—activators
Reminders of various kinds can be extremely helpful to keep managers, supervisors, and
employees aware and involved in your safety process. Vary these support devices and, by
all means, keep them coming. In addition to management commitment and activity, other
sustaining process drivers are

• Environmental supports, such as signs, slogans, newsletter, articles.

• Incentive supports, especially publicity about reward strategies.
• Promise cards to pledge behavioral commitment.
• Social support at group meetings and social gatherings.

Performance feedback—consequences
As you know, a cornerstone of the DO IT approach to motivating behavior change is the
principle that behavior is influenced by its consequences. In most plants, however, there
are no clear-cut consequences for performing safe acts. Sometimes, there are actually sub-
tle rewards for failing to follow proper safety procedures (Chapter 9). Employees might
save time and avoid discomfort by not wearing appropriate protection, for instance. As
detailed in Chapter 12, rewarding feedback should be given following safe behaviors, and
correcting feedback should be given when at-risk behaviors are observed.
Effective performance feedback is, of course, closely linked to measurement and
evaluation of the test phase of DO IT. As with measurement, performance feedback
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 451

Chapter nineteen: Obtaining and maintaining involvement 451

should be given systematically and consistently. Ideally, feedback regarding safe behavior
should occur at several levels, including plant-wide (for overall organizational perform-
ance), across shifts or work groups (for group performance), and individually (for personal
performance).
Finally, it is important to acknowledge supervisors and others for supporting a safety
achievement effort. As discussed in Chapter 12, safety coaches must feel appreciated for
their actively caring efforts. They need to be recognized for their competence at giving sup-
portive and corrective feedback to others.
Without clear behavioral consequences, performance tends to drift (Hayes et al., 1980),
usually in the direction of minimal effort. If you want to sustain the energy it takes to be a
safety coach, you need to recognize their work. You have to convince top managers that
coaches need their support, as well.

Tangible consequences
Performance-contingent consequences can go beyond just saying “good job” or “you need
to do better.” The use of tangible rewards to initiate desirable behaviors is fundamental.
The advantages of incentive strategies over the disincentive approach are detailed in
Chapter 11. That chapter also describes procedures for implementing an incentive/reward
program. At this point, it is important to consider these general questions about tangible
reward interventions.

• Should rewards, punishers, or both be used?

• What specific events or items should be used as rewards or punishers?
• Should the rewards be individual or group-based, or both?
• Who should administer tangible consequences?
• How often should these consequences be available?

Ongoing measurement and evaluation

You cannot maintain and improve a safety process unless you regularly measure its impact.
As discussed in Chapters 11 and 12, even the mere act of tracking a given set of behaviors
usually improves the performance of those behaviors. Performance tracking should be
done continuously, and questions to be resolved include

• Who conducts ongoing measurement?

• What will be measured?
• How frequently should the program’s impact be measured?
• What types of evaluation data should be collected?
• How will data be summarized and tabulated?
• What are the costs of the evaluation?
• What are the benefits? Are the benefits real or potential?

Follow-up instruction/booster sessions

Even with ongoing support, a comprehensive safety improvement process cannot succeed
without carefully planned follow-up instruction. From time to time, education and train-
ing content must be updated to reflect changes in plant conditions, the use of new machines
or protective equipment, and the like. Do not delay in keeping pace with change; what is
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 452

452 Psychology of safety handbook

being taught should correspond exactly to current plant conditions. Follow-up education
and training involve these issues

• When and how should basic instruction be repeated for new employees?*
• What are the objectives of follow-up instruction?
• How should new material be integrated?
• How often should follow-up sessions occur?
• What should be the content?
• How should the material be presented—film, lecture, on-the-job training, discus-
sion groups, audio tapes, interactive computer program?
• How can monthly safety talks on team meetings be used as boosters or activators?

Involvement of contractors
Let us pause to consider the influence of those outside our process, specifically contractors.
Noncompliance by outside contractors can undermine our efforts by demonstrating dan-
gerous behavior to plant personnel—in addition to possibly compromising the safety of
everyone on site.
Are outside personnel exempt from wearing safety glasses? Are they exempt from
wearing hard hats and steel-toed shoes in construction areas? From following speed limits
posted on plant property? If so, you have probably heard one of your employees complain,
“If they don’t follow safe work practices, why should we?”
Here are some strategies for getting outside personnel to understand and comply with
your safety procedures.

• Request information pertinent to safety-related issues in the bidding process.

• Conduct safety meetings and behavior-based instruction for contractors.
• Obtain full cooperation/commitment of the contractor to follow local safety practices.
• Provide verbal instructions and feedback (rewarding and correcting) to contractors.
• Enable and support safety coaching of contractors by regular plant employees and
vice versa.
• Include contractors in DO IT and reward/recognition programs.

Troubleshooting and fine-tuning

Once a safety achievement process is up and running, the Safety Steering Team con-
fronts the responsibility of fine-tuning the procedures. This is based on ongoing evalua-
tions. If the process is going to be sustained, employees and managers must perceive
it as current—state of the company—in terms of content, adaptable, and responsive. It sim-
ply cannot be “frozen” nor left unattended. You cannot “wind up” a process at the start and
expect it to run forever like that battery-powered bunny. Some keys to fine-tuning include

• Discuss the impact—is it working? What do the data from participants’ reaction
sheets, as well as other measures, tell us?
• Identify strengths and weaknesses—based on the data, which elements should be
kept, changed (and how), or replaced?

* For information on an interactive computer instructional program with on-line access, contact Safety
Performance Solution, Inc., 1007 N. Main St., Blacksburg, VA 24060, call (540) 951-7233, or e-mail safety@
safetyperformance.com.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 453

Chapter nineteen: Obtaining and maintaining involvement 453

• Cope with change—be sure all those affected by the process are fully informed
about changes when they occur. Ideally, all participants should be actively involved
in trouble-shooting and fine-tuning interventions.

Cultivating continuous support

Starting a safety improvement process and maintaining it over the long term requires the
three essential support processes depicted in Figure 19.3. Leaders are needed to champion
new principles and procedures. In fact, leadership makes the difference between a “flavor
of the month” safety initiative and a long-term continuous improvement process.
My colleagues and I at Safety Performance Solutions have seen the principles and pro-
cedures presented in this book lead to remarkable success and, eventually, a Total Safety
Culture. All too often, however, we have seen good intentions and superb introductory
instruction fizzle out and go nowhere. Why? It is a matter of leadership. You can launch a
process with excellent education and training, but you cannot keep the momentum going
without individuals who provide energy, enthusiasm, and the right example. This section
covers some essentials of effective leadership.

Where are the safety leaders?

First, we have to find the leaders. Who are they? The traditional definition of one person
exerting influence over a group does not quite work for safety. Ask any safety manager
who has been expected to do it all. To achieve a Total Safety Culture, everyone needs to

2
Leadership Recognition

4
1 3

Communication

1. Leaders communicate effectively.

2. Leaders recognize desired performance.
3. Recognition is communicated effectively.
4. Leaders recognize desired performance
effectively through a variety of
communication channels.

Figure 19.3 Continuous improvement depends on three support systems.

820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 454

454 Psychology of safety handbook

accept a leadership role in reducing injuries. Everyone needs to feel responsible for safety
and go beyond the call of duty to protect others. This requires leadership skills: giving
supportive feedback for another person’s safe behavior and constructive feedback for at-
risk behavior.
Psychologists have studied leadership rigorously for over 50 years in an attempt to
define the traits and styles of good leaders (Yukl, 1989). Still, many questions remain unan-
swered, making leadership more an art than a science. Several decades of research, how-
ever, have turned up some important answers, which we will now apply to safety.
Many psychologists consider the characteristics that offset leaders to be permanent
and inborn personality traits (Kirkpatrick and Locke, 1991), but I prefer to consider them
response styles or personality states that can be taught and cultivated. If action plans or
interventions can be developed to promote styles typical of the best leaders, then the num-
ber of effective safety leaders in an organization can be increased.
Passion. The most successful leaders show energy, desire, passion, enthusiasm, and
constant ambition to achieve. Passion to achieve a Total Safety Culture can be fueled by
clarifying goals and tracking progress. Put a positive spin on safety, make it something to
be achieved—not losses to be controlled. Then, employees will be motivated to achieve
shared safety goals just like they work toward production and quality goals. Marking
progress leads to the genuine belief that the process works. This fires up employees to con-
tinue the process.
Honesty and integrity. Effective leaders are open and trustworthy. A Total Safety
Culture depends on open interpersonal conversation. This obviously requires honesty,
integrity, and trust. It is quite useful for work groups to discuss ways to nurture these qual-
ities in their culture. Take a look at certain environmental conditions, policies, and behav-
iors. Some arouse suspicions of hidden agendas, politics, and selfish aims. You can work to
eliminate some of these trust-busters by first identifying them, discussing their purpose,
and devising alternatives. While frankness is important in increasing trust, it is important
to be tactful when communicating an honest opinion, as not shown in Figure 19.4.
Motivation. Because most people really care about reducing personal injuries, even
to people they do not know, the motivation to lead others will spread naturally throughout

Figure 19.4 Candor should be delivered tactfully.

820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 455

Chapter nineteen: Obtaining and maintaining involvement 455

a work culture when people believe they can have personal control over injuries. This
occurs when they learn effective techniques to prevent injuries (as presented in Sections 3
and 4 of this text) and feel empowered to apply them (as covered in Section 5).
Self-confidence. Effective leaders trust in their own abilities to achieve (Baron, 1995;
Kirkpatrick and Locke, 1991). Education helps convince people they can achieve, but they
need ongoing support and recognition for their efforts. For example, the self-confidence
needed to give safety feedback can be initiated with appropriate education and training,
and can be maintained with coaching, communication, and recognition. Notice that
enhancing the three empowerment factors—self-efficacy, personal control, and opti-
mism—as discussed in Chapter 16, builds self-confidence.
Thinking skills. Successful leaders can integrate large amounts of information,
interpret it objectively and coherently, and act decisively as a result (Baron, 1995;
Kirkpatrick and Locke, 1991). Constructive thinking skills evolve among team members
when objective data are collected on the progress of safety interventions and used to refine
or expand these processes and develop new ones.
When teams work through the DO IT process (as covered in Section 4), participants
develop skills to evaluate behavioral data and use the information to make intervention
decisions. This is basic scientific thinking, the key to substituting profound knowledge for
common sense. This mindful learning (Langer, 1997) and critical thinking leads to special
expertise.
Expertise. To achieve a Total Safety Culture, everyone needs to understand the prin-
ciples behind policies, rules, and interventions to improve safety. When employees teach
these principles to coworkers, they develop the level of profound knowledge, expertise,
and responsibility needed for exemplary leadership.
Flexibility. Successful leaders size up a situation, and adjust their style accordingly
(Hersey and Blanchard, 1982). At times, some groups and circumstances call for firm
direction—an autocratic style. At other times, the same people might work better under
a nondirective, hands-off approach—a democratic style. The best leaders are good at
assessing people and situations, and then matching their behavior to fit the need (Zaccaro
et al., 1991).
Recall the discussion in Chapter 13 about directive, supportive, motivational, or dele-
gating conversations. Effective leaders size up the situation, especially the relative com-
mitment and competence of the participants, and then direct, support, motivate, or
delegate depending on this assessment (see Figure 13.16).
As discussed in Chapter 17, work groups or teams progress through four development
stages: forming, storming, norming, and performing (Tuckman and Jensen, 1977). During
the early stages of forming and storming, there is a need for structure, clear vision, and a
sense of mission. Autocratic (or directive) leadership is often most appropriate—though it
is good to get input from group members before the first meeting. When group members
become familiar with each other and start implementing their assignments, the norming
and performing stages, democratic (or delegating) leadership is usually called for.
However, there is always room for supportive leadership.

Safety management vs. safety leadership

Safety professionals seem to use the term “management” more than “leadership.” More
titles of safety books and professional development seminars reflect management than
leadership. Do these terms mean the same thing?
I heard Dr. Tom Krause say at the “1999 Best Practice in Safety Management” confer-
ence sponsored by the American Society of Safety Engineers that leaders “inspire people to
want to do something,” as opposed to managers who “hold people accountable for doing
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 456

456 Psychology of safety handbook

something.” This distinction reflects the difference I discussed in Chapter 9 between other-
directed and self-directed behavior (see Figure 9.9).
What is your focus? Managers are typically held accountable for outcome numbers,
and they use these numbers to motivate others. In safety, the outcome numbers are based
on the relatively rare occurrence of an injury. They are reactive, reflect failure, and are not
diagnostic for prevention.
Safety leaders hold people accountable for accomplishing activities that can prevent
injuries. When people see improvement in the process numbers, they are reinforced for
their efforts and develop a sense of personal responsibility for continued contributions and
continual improvement.
Do you educate or train? In industry, “training” is a more common term than “edu-
cation.” This reflects the concern that employees know exactly what to do in order to com-
plete a particular task effectively and safely. With a “training” mindset, however, managers
can come across as demanding a certain activity because “I said so” rather than because “it
is the best way to do it.”
Leaders educate—offering rationale and examples rather than policy and direction.
This enables individuals or work teams to select procedures that best fit their situation. In
the process of refining a set of procedures, people assume ownership and follow through
from a self-directed or responsible perspective.
Do you speak first or listen? Under pressure to get a job done, managers often speak
first and then listen to concerns or complaints. This is a reasonable strategy for efficient
action. After all, managers must make things happen according to an established plan, and
this requires specific directives and a mechanism for motivating compliance. After describ-
ing an action plan and accountability system, managers answer a lot of questions from
workers who want to make sure they will do the right thing.
Leaders take time to learn another person’s perspective before offering direction,
advice, or support. Proactive listening is key to diagnosing a situation before promoting
change or continuous improvement. This is not the most efficient approach to getting a job
done. It requires patience and a communication approach that asks many questions before
giving advice (as detailed in Chapter 9). This way, an individual or work team can cus-
tomize an action plan or process for achieving a particular outcome.
Do you promote ownership? When the development of an action plan involves the
people expected to carry out that plan, ownership for both the process and the outcome is
likely to develop. Then, leaders give a reasonable rationale for a desired outcome and, then,
offer opportunities for others to customize methods for achieving that outcome. They facil-
itate a special kind of motivation. This motivation comes from inside people—it is not
directed by others. It is internal or self-directed motivation (as discussed in Chapter 11). In
this state, people participate because they want to, not because they have to.
Managers direct by edit for efficiency. While they might get compliance, they might
also stifle self-directed motivation. Behaviors performed to comply with a prescribed stan-
dard, policy, or mandate are other-directed. They are accomplished to satisfy someone else
and are likely to cease when they cannot be monitored. This happens, for example, when
personal protective equipment is used at work but not at home for similar or even riskier
behaviors.
Do you leave room for choice? All mindful behavior starts because someone asked
for it. The important issue is whether behavior remains other-directed or advances to self-
directed (Chapter 9). This depends to some extent on how you ask.
Managers favor mandates, reflected in regulatory compliance issues and the common
slogan, “Safety is a condition of employment.” Mandated behavior is likely to require con-
stant “management.”
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 457

Chapter nineteen: Obtaining and maintaining involvement 457

Leaders use expectations rather than mandates to instil self-directed motivation. What
is the difference? Expectations imply choice. There is room for individual and group
decisions regarding procedures and methods. When people realize what is expected
of them but perceive some personal control in how to reach specific goals, they are
more likely to own the process and move from depending on the directions of others to
directing themselves.
Do you show some uncertainty? Langer (1989) proposes that a leadership quality
conducive to promoting innovation and initiative is displaying a degree of “uncertainty.”
She suggests leaders show confidence that a particular job will get done but without being
sure of the best way to do it. This allows employees room to be alert, innovative, and self-
motivated. They feel their involvement is a necessary part of the process. They feel impor-
tant (Carnegie, 1936).
People are less likely to hide mistakes when they work for a confident but uncertain
leader. It is okay to be imperfect if the boss is not sure exactly how the job should be done.
In this context, workers are more willing to suggest ways to improve a process. In Langer’s
words, “admission of uncertainty leads to a search for more information, and with more
information there may be more options” (page 143).
To test her theory about confident but uncertain leadership, Langer assessed the gen-
eral level of confidence among supervisors at a particular company and, then, asked them
how many of their daily decisions have absolutely correct answers. In addition, the
employees completed surveys that evaluated their work relationships with their supervi-
sors. Results showed that the confident but relatively uncertain supervisors were viewed
as allowing more independent judgment and innovative action.
Relating these findings to industrial safety, a leader should show confident expectation
that appropriate precautions will be taken to prevent injuries. However, they should not
pretend to know exactly how the injury-free job should be accomplished.
Leaders realize the employees of a work team are the true safety experts. Line workers
know what hazards need to be eliminated or avoided, and what safety-related behaviors
need to be improved. They are also more likely to be effective at encouraging safe behav-
iors and discouraging at-risk behaviors at their job sites. Thus, it seems a leadership qual-
ity of “confident uncertainty” can be instrumental in empowering employees to go beyond
the call of duty for safety and health.
Certainty and familiarity also contribute to fatigue and burnout (Langer, 1989). When
the job is seen as a mindless routine, energy and enthusiasm wane. Workers lose interest
and a sense of choice and personal control. It is easy to get lulled into a false sense of secu-
rity that doing things the way they have always been done will continue to work. This cer-
tainty mindset not only hinders creative innovation, it also contributes to feelings of fatigue
and burnout, which in turn puts people at-risk for personal injury.
Must it be measured? Managers focus on the numbers, and in safety that means
injury records and compensation costs. When I discuss behavior-based safety principles
and procedures with managers, I inevitably get the question, “What’s the ROI or return on
investment?” They want to know how much the process will cost and how long will it take
for the numbers to improve. This analytical approach is inspired by the popular manage-
ment principle, “You can only manage what you can measure.”
Leaders certainly appreciate the need to hold people accountable with numbers,
yet also understand you cannot measure everything. As I discussed in Chapter 18,
there are some things you do and ask others to do because it is the right thing
to do. Leaders believe, for example, it is important to increase self-esteem, self-efficacy,
personal control, optimism, and a sense of belonging throughout a work culture (see
Chapter 15).
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 458

458 Psychology of safety handbook

Now and then it is a good idea to assess whether certain actions are influencing peo-
ple’s subjective feelings in a desired direction. This can be done informally through per-
sonal interviews, unaided by a score card. It is a given that certain interpersonal and group
activities are useful. For example, genuine one-to-one recognition increases trust and feel-
ings of importance, behavior-based goal-setting builds feelings of empowerment, and
group celebrations facilitate a sense of belonging. You need to perform and support these
sorts of activities without expecting to see an immediate change in the numbers of a safety
accountability system.
Leaders do not need a monitoring scheme to motivate their attempts to help people feel
valuable and a part of an important team effort. This kind of leadership is self-directed and
responsible and helps to inspire self-directed responsibility in others.
Do you put people in categories? We tend to give global labels for people, such as
student, patient, homosexual, union representative, safety professional, athlete, or home-
less person. Each label prompts a particular image and a set of characteristics. Similarly, the
current best-seller, Men Are from Mars, Women Are from Venus (Gray, 1992) teaches us generic
differences between men and women which we presumably need to understand if we want
to be more successful at dating, loving, and sustaining a marriage. This leads to the kind of
stereotyping illustrated in Figure 19.5.
The general label we give people influences how we view them, judge them, and react
to their communication with us. This is the kind of destructive bias or premature cognitive
commitment (Chapter 5) that leads to prejudice, interpersonal conflict, and sometimes even
hate crimes.
Efforts to combat prejudice focus on teaching people that everyone should be consid-
ered equal and that categorizing people is wrong. In other words, to decrease discrimina-
tion and its accompanying problems we are told to stop discriminating. Langer (1989)
believes this is the wrong approach. Categorizing people and things according to discern-
able characteristics is a natural learning process. It is how we come to know and under-
stand our surroundings.

Figure 19.5 Sometimes we live up to our stereotypic labels.

820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 459

Chapter nineteen: Obtaining and maintaining involvement 459

Managers
• Hold people accountable
• Focus on outcomes
• Train
• Speak first, then listen
• Answer questions
• Promote compliance
• Direct by edict
• Use unconditional statements
• Mandate rules and policies
• Manage what’s measured
• Limit choice
• Enable mindlessness
• Follow a directive approach
Leaders
• Build responsibility
• Focus on process
• Educate
• Listen first, then speak
• Ask questions
• Promote ownership
• Inspire by example
• Use conditional statements
• Set expectations
• Facilitate intangibles
• Encourage choice
• Facilitate mindfulness
• Instruct, support, coach, or delegate

Figure 19.6 Management is not the same as leadership.

The key to reducing prejudice is to make more, not fewer, distinctions among people.
When people become more attentive to the numerous differences among individuals and
understand how these differences vary according to the environmental or interpersonal
context (Chapter 5), it becomes increasingly difficult to put individuals into universal cat-
egories. It becomes impossible to view people and their behavior as black or white, normal
or abnormal, masculine or feminine, safe or unsafe.
Leaders put people’s attributes and skills on a continuum. A person is not good or bad,
skilled or unskilled, safe or unsafe at a particular task but rather is a particular degree of
good, skilled, and safe. Plus, one’s quality level for a certain attribute can fluctuate dra-
matically from one situation to another. Thus, leaders make more distinctions between peo-
ple and fewer global stereotypes. This enables objective and fair linkage between people’s
talents and job descriptions and facilitates the kind of interpersonal trust needed for a Total
Safety Culture (Geller, 1999).
In summary The distinctions between managing and leading presented here are
summarized in Figure 19.6. The bottom line is that safety management is necessary at times
to motivate people to do the right things for injury prevention, but this is not sufficient to
achieve a Total Safety Culture. Safety managers need to know when to become safety lead-
ers and build personal responsibility rather than hold people accountable. Most important,
whether or not you hold a safety management position, you can be a safety leader and help
people transition from an other-directed to a self-directed motivational state. Remember,
few manage but many must lead.

Communication to sell the process

As introduced in Chapter 13, how we talk about safety influences whether people will con-
tribute their leadership skills to a safety process. Indeed, our language can determine
acceptance or rejection of the entire process.

Words are magical in the way they affect the minds of those who use
them . . . words have power to mold men’s thinking, to canalize
their feelings, to direct their willing and acting.

This quote from Aldous Huxley’s “Words and Their Meanings” (Hayakawa, 1978,
page 2), reflects the power of words to shape our feelings, expectancies, attitudes, and
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 460

460 Psychology of safety handbook

behaviors. When people use expressions like, “Say that enough times and you’ll start
to believe it,” “Can’t I talk you into it,” or “Do as I say, not as I do,” they acknowledge the
influence of words on behavior.
Words also affect feeling states. Years ago, when my two daughters discussed horse
manure at the dinner table, I lost my appetite. Likewise, using negative, uninspiring words
to describe ourselves or everyday events can contribute to losing our appetite or passion
for daily life.
Figure 19.7 illustrates how a simple word change can influence a feeling state and,
then, more behavior. The scenario might seem far-fetched at first, but it is really not if you
consider that the child is responding to the parents’ reactions. Initially, mom and dad are
unhappy with the word “crib” and their body language contributes to the child’s negative
reactions. However, the change of language made the parents happier. This is perceived by
the child and leads to his positive reaction.
What does this have to do with safety? I think some words we use in the safety and
health field are counter-productive. Let me point out a few that should be eliminated
from our everyday language if we want to “sell” safety and increase involvement to pre-
vent injury.
“Accident” implies “chance.” Earlier in Chapter 3, I made the case for removing the
word “accident” from all safety talk. The word accident implies chance or loss of control.
Workplace accidents are usually unintentional, of course, but are they truly chance occur-
rences? There are usually specific controllable factors, such as changes in the environment,
behaviors, and/or attitudes, that can prevent “accidents.”
We want to develop the belief and expectation in our work culture that injuries can be
prevented by controlling certain factors. “Accident,” then, is the wrong word to use when
referring to unintentional injuries. It can reduce the number of people who believe with
true conviction that personal injuries can be prevented.

Figure 19.7 Language can influence attitude and behavior.

820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 461

Chapter nineteen: Obtaining and maintaining involvement 461

Restraints do not invite use. For almost two decades, I have been urging trans-
portation and safety professionals to stop using the terms “occupant restraints” and “child
restraints” for vehicle safety belts and child safety seats. These terms imply discomfort and
lack of personal control and fail to convey these devices’ true function. “Seat belt” is better
than “occupant restraint,” but this popular term is not really adequate because it does not
describe the function or appearance of today’s lap-and-shoulder belts. We need to get into
the habit of saying “safety belt” and “child safety device.”
Priority or value. Priority implies importance and a sense of urgency, and safety
professionals are often quick to say “safety should be a priority.” This seems appro-
priate because my New Merriam-Webster Dictionary defines “priority” as “taking
precedence logically or in importance” (page 577). However, everyday experience
teaches us that priorities come and go. Depending upon the demands of the moment,
one priority often gets shifted for another. Do we really want to put safety on such shifting
ground?
I believe a Total Safety Culture requires safety to be accepted as a value. The relevant
definition of “value” in my New Merriam-Webster Dictionary is “something (as a principle
or ideal) intrinsically valuable or desirable” (page 800). Safety should be a “value” that
employees bring to every job, regardless of the ongoing priorities or task requirements.
Do not say “behavior modification.” Over recent years, I have seen “behavior mod-
ification” used many times for titles of safety presentations at regional and national con-
ferences. I have heard trainers, consultants, and employees use the term to describe
behavior-based safety. In fact, I have often been introduced at conferences as a specialist in
“behavior modification.” This is the wrong choice of words to use if we want acceptance
and involvement from the folks who are to be “modified.” Who wants to be “modified”?
This lesson was learned the hard way more than 30 years ago by the behavioral scien-
tists and therapists who developed the principles and techniques of “behavior modifica-
tion.” Whether it applied to teachers, students, employees, or prisoners, the term “behavior
modification” was a real turn-off. It conveyed images of manipulation, top-down control,
loss of personal control, and “Big Brother.” For example, in 1974, my colleagues and I
developed a behavior-change process and training program for the Virginia prison system
(Geller et al., 1977). Our innovative and very effective plan was never fully implemented,
partly because the inmates and guards associated “behavior modification” with brain-
washing and lobotomies.
Actually, the term “behavior” alone carries negative associations for many—as in
“Let’s talk about your behavior last night”—but I cannot see any way around using it. We
need to teach and demonstrate the benefits of focusing on behaviors, especially desirable
behaviors. We do not have to link “modification” with behavior; it only adds to the nega-
tive feelings.
“Behavior analysis” is the term used by researchers and scholars in this area of applied
psychology. This implies that behavior is analyzed first (Chapter 9) and, if change is called
for, an intervention process is developed with input from the clients (Chapters 11 to 13).
Given that “analysis” can sound cold or bring to mind Freud, I have recommended the
label “behavior-based approach” for several years. This contrasts nicely with the “person-
based approach” that focuses on attitudes, feelings, and expectancies. As I have repeat-
edly emphasized, a Total Safety Culture requires us to consider both behavior-based and
person-based psychology.
I hope the basic message is clear. We need to understand that our language can activate
feelings and even behaviors we do not want. Figure 19.8 illustrates what I mean. If we want
to communicate in order to “sell” the process, we need to consider how our language will
be perceived by the “customer.”
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 462

462 Psychology of safety handbook

Figure 19.8 The words we use can increase participation or resistance.

As we end this discussion on language, I would like you to ponder the following word-
attitude associations. Which terms are more likely to facilitate employee involvement?

“Peer pressure” or “peer support”?

“Loss control managers” or “safety facilitators”?
“Compliance” or “accomplishment”?
“Meeting OSHA standards” or “fulfilling a corporate mission”?
“30 days without an injury” or “30 safe days”?
“I must meet this deadline” or “I choose to achieve another milestone”?
“I’ve got to do this” or “I get to do this.”

It is a good personal or group exercise to consider the ramifications of using these

terms and phrases and adding alternatives to this list is even more beneficial.
Understanding the critical relationship between words, attitudes, and deeds is only one-
half the battle. We need to change verbal habits and this is easier said than done. When we
communicate with greater passion and optimism about safety we will attract more people
to our safety mission. We will also reduce resistance to change.

Overcoming resistance to change

“How do we deal with people who resist change?”
“How do we get more people to participate?”

I frequently hear these questions at training seminars and workshops. First, let us face real-
ity. Change is unpleasant for many people, and some are apt to react poorly. Change often
threatens our “comfort zones”—those predictable daily routines we like to control. In fact,
it takes a certain amount of personal security and leadership to try something new. A cer-
tain kind of risk taking is needed to lead change, and some people want no part of explor-
ing the unknown.
We have all been in unfamiliar situations where we are not sure how to act. We feel
awkward and uncomfortable. If someone gives us direction, helps increase our sense of
control, it is easier to adjust. We might even help others deal with the change. Without lead-
ers and adequate tools to cope with change, we might retreat, withdraw from the situation,
or even actively resist the change.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 463

Chapter nineteen: Obtaining and maintaining involvement 463

So how do we deal with resistance? Simply put, we should teach people the skills and
give them the tools to handle change, plus support those who set the right examples. This
seems logical and intuitive, but it does not always happen. Instead, managers too often try
to identify the malcontents and discipline them for not participating.
Let us try to better understand resistance by considering one of the classic awkward,
strange situations thrust upon many of us—our first school dance. Remember it? For me,
it was a high school homecoming dance in 1957.
Remember your first dance. Attending your first dance is like a rite of passage. If
you were anything like me, you were a bit nervous about this change in your social world.
You might have been prepared for it. Family, friends, and teachers probably told you what
to expect. Maybe, you even had dance lessons but these “tools” did not make it any easier
for some of us to participate.
Not for me, anyway. I did not participate, but I wanted to. Before the dance, I practiced
how to ask a girl to dance. I took four, two-hour dance lessons at an “Arthur Murray Dance
Studio.” I felt ready, but never once did I dance that night. I did not feel too embarrassed,
though, because there were so many others not dancing. As was the custom, boys stood on
one side of the gym and the girls on the other.
As illustrated in Figure 19.9, some kids were dancing and seemed to be having a
great time. They danced almost every number and tried to lure others out on the floor.
I could not be enticed, though. I hung back in my comfort zone, but, at least, I was in the
dance hall.
As illustrated in Figure 19.10, some students stood around in the parking lot, talking,
drinking, and smoking cigarettes. These were the resisters. Some were active resisters. They
stayed in their cars, never intending to enter the dance hall. Now and then, these guys
started up their cars and cruised around town for awhile, and then returned to the parking
lot. They would persuade others to hop in their car, try some beer, smoke a cigarette, fool
around, or cruise.
Levels of participation. There are essentially five ways of reacting to change—call
them levels of participation—and they were all on display at the dance. First, there are the
true leaders who get totally involved. They are the innovators—those who view change as
necessary and an opportunity to improve. At the dance, they were the teenagers on the
floor for almost every number. They had the most fun. They did not necessarily know what
they were doing when it came to dancing, but they got out there and tried. They took a risk.
They got totally involved and benefited most from the occasion. A dance might start with
only a few of these “risk takers,” but they often persuaded a number of others to get
involved as the night wore on.

Figure 19.9 Different reactions to change can be seen at the high-school dance.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 464

464 Psychology of safety handbook

Figure 19.10 Some actively resist change while others follow.

Some people want to change but need direction and support. They are motivated to
participate but need models or leaders. At the dance, these were the kids who hung back
at first. With a little encouragement, they danced a few numbers. By the end of the evening,
you could not get them off the dance floor. They were now totally involved.
Most of us are at the third level of participation. We are ready to get involved, but we
will stay in our comfort zones until we are directed and motivated to participate. It might
look like we are resisting change, but not really. Call us neutral when it comes to our atti-
tude about change. We just are not sure what to do. We need self-confidence that we can
handle the change. We also need genuine support (positive recognition) when we try to
participate. Once in a while, just getting started, or “breaking the ice,” is enough to turn a
passive observer into an active participant. For the most part, however, we stand on the
sidelines and watch. This level of participation was represented by the boys and girls who
lined each side of the gym.
Types of resistance. The final two levels of participation are passive and active resist-
ance. Passive resisters perceive change as a problem. They complain a lot. They are critical
and untrusting of something new imposed on them. They seem to see only the negative
side of a new program, policy, or challenge. They rationalize their position by gathering
with others at their same level of nonparticipation, and they grumble and whine about pro-
posed changes or about others who are participating in a change effort. Their whining and
complaining usually stops when participation in a new process is clearly enjoyed by the
majority. Passive resisters are followers, and they will do what they see most people doing.
These are the teenagers who came to the dance because everyone else would be there,
but they felt so insecure or anxious they did not enter the building. They looked for others
hanging around outside and made fun of the silly dancing going on inside. Sometimes,
these nonparticipants ran into an active resister.
Fortunately, active resisters are few in number, but it does not take many of these char-
acters to slow down a change process. These individuals view change as a threat or an
opportunity to resist. They see any change effort that was not their idea as a potential loss
of personal control, and they often exert countercontrol to assert their control or freedom.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 465

Chapter nineteen: Obtaining and maintaining involvement 465

Figure 19.11 Deviant behavior is often a statement of countercontrol.

Psychologists call this “psychological reactance” (Brehm, 1966, 1972), a phenomenon most
parents observe when their children reach the teens. Teenagers want to feel independent
and, at times, will disobey their parents’ directions—break the rules—to gain a sense of
independence or self-control.
I am convinced some of the piercing and tattooing illustrated in Figure 19.11 occurs as
a way to react to top-down control and assert individual freedom. Today, this behavior is
so common it has become a social norm in some settings. So, now, many teens get pierced
and tattooed to conform rather than to assert their individuality.
We all feel overly controlled at times and, perhaps, react to regain independence or
assert personal freedom. Sometimes, our reactions are not thoughtful, caring, or safe.
Active resisters feel the need to resist change, the status quo, or authority much of the time.
This is partly because their contrary behavior brings them special attention—recognition
for resisting.
Who gets the attention? Active resisters stand out and attract attention. Nonpartici-
pants use them to rationalize their own commitment to comfort zones. Managers monitor-
ing the workplace often hit them with discipline, but this can backfire. This makes the
top-down control more obvious for resisters. Discipline builds their resentment of the sys-
tem and makes it even less likely they will join the change process.
For some individuals, disciplinary attention only fuels their burning desire to exert
independence and resist change. As a result, they might become more vigorous in recruit-
ing others to oppose change. As I discussed in Chapter 11, top-down discipline (actually
punishment) should be used sparingly if the ultimate purpose is total participation in an
improvement process.
How were resistant teenagers brought inside to the dance? The harsh warnings of the
school principal shouting from the steps did not work; neither did the one-on-one con-
frontation between one of the adult chaperones and the “leader of the pack.” Whenever I
saw a resister come inside, it was always the result of urging by another teenager. Peer pres-
sure (or peer support) is still the most powerful motivator of human behavior.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 466

466 Psychology of safety handbook

Figure 19.12 When a critical mass of the culture changes, others follow.

Therefore, the best way to deal with resistance is usually to arrange for situations that
enable or facilitate peer influence. Eventually, some of the resistant teenagers came into
the building so they would not miss something. As depicted in Figure 19.12, they saw
from their remote comfort zones that the people inside were really enjoying themselves,
and they chose to participate.
The dance party will become more enticing as more and more teenagers dance. To
increase such active involvement, the right kind of encouragement and support is needed.
Will motivational lectures from a teacher, counselor, concerned parent, or outside consult-
ant make that happen? It might make a temporary difference but not over the long haul.
The best way to deal with nonparticipation is usually to set up situations that allow for peer
influence. This could mean managers do nothing more than support the change process
and let peer pressure or support occur naturally.
Power of peer influence. So how do you facilitate peer influence? The best way is
through empowerment, but this is easier said than done. You can give people more respon-
sibility, such as the challenge to lead others in a change process, but they must feel respon-
sible. As I discussed in Chapter 15, they need to have sufficient self-efficacy (“I can do it”),
personal control (“I am in control”), and optimism (“I expect the best”).
Some people already have sufficient self-efficacy, personal control, and optimism
when assigned leadership responsibilities. As discussed in Chapter 15, they feel empow-
ered (“I can make a difference”). Still, these individuals may need some basic training in
communication, social recognition, and behavior-change principles. Others may lack one
or more of the three person states that facilitate feelings of empowerment. So, in addition
to education and training, they need a support system to build their sense of self-efficacy,
personal control, and optimism.
As discussed in Chapter 16, there is no quick fix for increasing perceptions of empow-
erment. If you ask people to define policies, settings, interactions, and contingencies that
influence these three person states, you are on your way to developing action strategies to
improve them. This, in turn, increases people’s readiness to be empowered.
Remember the bottom line is that resistance to change is overcome most effectively
through peer influence. This requires, of course, that people are willing to accept the
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 467

Chapter nineteen: Obtaining and maintaining involvement 467

leadership position of change agent (Chapter 12). Certain individuals naturally rise to the
occasion and welcome opportunities to lead. Others may be committed to continuous
improvement through change, but they need some direction and encouragement from the
natural leaders. Both of these groups need training and practice in behavior-change prin-
ciples and social influence strategies. Then, management needs to give these people direc-
tion and the opportunity to work with those folks standing on the sidelines in need of a
nudge to leave their comfort zones.
It is usually best to ignore the resisters. Do not give them too many opportunities to say
“no.” The more often they publicly refuse, the more difficult it will be for them to change
their minds and participate. So, do not pressure these folks. Invite them to contribute
whenever they feel ready to achieve success with those willing to try something new. When
the majority buys in and eventually celebrates their accomplishments, resisters will choose
to come on board. The key is for them to perceive that they chose to get involved rather than
being forced by a top-down mandate.
My teenage daughter’s experience at her first high-school dance several years ago
allows me to use this analogy once more to make a final point. Karly was well prepared for
the dance but, perhaps, not for her first date. She was quite talented at performing the lat-
est dances, she was wearing a new outfit, and she had planned to meet a number of her
girlfriends who were also attending their first homecoming dance with a date.
After the football game, Karly was to meet her date at the dance. She arrived at the gym
before him, and instead of waiting outside, she went inside. She, then, got totally involved
in the dance, heading out to the floor for just about every song with whomever was avail-
able and willing, boy or girl. In this situation, Karly was an innovator and a leader. She felt
empowered enough to lure others onto the dance floor so they could join in the fun. When
Karly’s date finally arrived, he rushed up to her, led her off the dance floor, and admon-
ished her for not waiting outside. Why should she have fun without him?
It is important to realize that even leaders need support for their leadership efforts.
Those who benefit from a leader’s inspiration or coaching should share their appreciation.
Karly’s date actually punished her for her initiative and total involvement. Fortunately,
Karly had sufficient self-efficacy, personal control, and available support from friends to
ignore her date’s reprimand. In fact, she went back to the dance floor and participated with
the support group that had evolved before her date arrived. She chose to make her first
date with that boy her last. That empowerment was sure appreciated by her dad!

Planning for safety generalization

I would like to discuss another factor that can help us build a Total Safety Culture—the
heavily researched phenomenon of generalization. What is it? Generalization has to do
with the spread of behaviors, and it occurs in two ways.
Stimulus generalization refers to the spread of influence from one setting or environ-
ment to another.
Response generalization refers to the occurrence of one behavior leading to another
(Kimble, 1961).
For example, if a safety training process increases the use of personal protective equip-
ment on the job and at home, stimulus generalization has occurred. On the other hand,
response generalization occurs when an increase in one safety behavior, such as vehicle
safety-belt use, is accompanied by an increase in another task-related behavior, like
turn-signal use.
Obviously, generalization is a desired outcome of safety efforts. Safety leaders hope
safe operating procedures used in one setting will spread to other situations, including
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 468

468 Psychology of safety handbook

employees’ homes (stimulus generalization), and to other procedures or behaviors

(response generalization). Psychological principles already discussed in this text (espe-
cially, Chapter 16) suggest ways to increase both types of generalization. Interventions set
up to improve the driving safety of pizza deliverers support these principles and demon-
strate practical ways to increase generalization.
Case study: generalizing pizza-delivery safety. When behavioral scientists evaluate
the impact of an intervention, they typically measure the quantity and/or quality of a tar-
get behavior before, during, and after the intervention. As discussed earlier in Chapter 8,
this is the basic research design used for more than 40 years to demonstrate the impressive
impact of behavior-based psychology and to refine intervention procedures and develop
new ones. Unfortunately, this method cannot be used to evaluate generalization because
usually only one situation is observed and only the behavior targeted by the intervention
is measured. A few years ago, Tim Ludwig, a former graduate student, and I measured
changes in behaviors that were not targeted by a safety intervention, and the results were
informative (Ludwig and Geller, 1991, 1997).
In the first project (Ludwig and Geller, 1991), our goal was to boost the safety-belt use
of pizza deliverers working out of two stores in southwest Virginia. Before, during, and
after a safety-belt campaign, we unobtrusively measured the deliverers’ safety-belt use
when entering and exiting the store parking lots. The results showed remarkable benefits
of the safety-belt intervention. We also discovered response generalization. Observing
the drivers’ daily use of turn signals, we found that the use of both safety belts and turn sig-
nals increased after the safety-belt campaign, which did not include any mention of turn
signal use.
Ludwig and I were actually surprised to find the marked increase in turn-signal use
(response generalization) after a campaign to increase only safety-belt use. We
concluded that certain aspects of the intervention process promoted a sense of personal
control, commitment, and group ownership, which theoretically should promote general-
ization. Ludwig’s Ph.D. dissertation, conducted two years later, verified these conclusions.
Our safety-belt campaign included a one-hour group discussion on the value of vehi-
cle safety belts and ways to support each other’s use of safety belts. Then, buckle-up prom-
ise cards were distributed to each participant and signed as a personal commitment to
buckle up consistently for two months. Signed promise cards were entered into a random
drawing for a $20 sweatshirt. Everyone signed the pledge. In addition, the group decided
to do a few things on the job to promote safety-belt use. Specifically, buckle-up reminder
signs were posted in the two stores, and the dispatchers agreed to remind drivers to buckle
up when giving them their pizzas to deliver. In an analogous study (Streff et al., 1993),
employees signed a promise card to use safety glasses. Subsequently, these employees
increased their use of safety glasses on the job and their use of vehicle safety belts when
entering and exiting the plant parking lot.
For his Ph.D. dissertation research, Ludwig varied the safety intervention strategies
between two pizza stores to test whether the choice, commitment, and involvement aspects
of this intervention process were critical to obtaining generalization (Ludwig and Geller,
1997). Goal setting and feedback were employed at each store to increase the target behav-
ior—complete vehicle stopping at intersections. During group discussion among the
employees at one store about the benefits of always stopping completely at intersections,
the manager noted that he had observed 55 percent complete stops during the previous
week. Then, the group selected a goal of 75 percent for each of the following four weeks.
The manager agreed to post biweekly percentages of safe intersection stopping obtained
from his periodic observations of vehicles leaving the parking lot.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 469

Chapter nineteen: Obtaining and maintaining involvement 469

Goal setting and feedback strategies were also used at the second pizza store, but the
process of implementing these behavior-change techniques was different. Instead of intro-
ducing the procedure through interactive discussion, Ludwig and the store manager lec-
tured about the benefits of complete intersection stopping, emphasizing the same points
brought out through group discussion at the other store. Then, the manager assigned this
group the same goal chosen by the employees at the other store. He also indicated he
would post biweekly percentages of safe stopping from his daily observations. The same
weekly feedback percentages were posted at each store and were not calculated from actual
observations. The results were provocative and instructive.
The pizza deliverers at both stores significantly increased their percentage of complete
intersection stops during the intervention period, but response generalization was
observed only among employees involved in the open discussion and group goal setting.
In addition to systematically evaluating vehicle stopping, our field observers recorded
unobtrusively (from store windows across the street) whether the pizza deliverers buckled
up and used their turn signals when leaving the parking lot. As we expected, only the pizza
deliverers who chose their own safe-stopping goal following interactive discussion sig-
nificantly increased their use of both safety belts and turn signals.
Involvement made the difference. This study demonstrates that safety generaliza-
tion is most likely to occur when people feel a sense of commitment and ownership for the
goals of a behavior-based intervention process. When this happens, some people remind
themselves to perform the target behavior in various situations (stimulus generalization)
and to perform other safe behaviors related to the target behavior (response generaliza-
tion). Individuals who are told they must comply with a safety policy or mandate might
only do so because the consequences of not complying could mean a penalty, not because
they believe and “own” the reason for the safe behavior. These people will probably not
show generalization to other situations or behaviors.
The important lesson here is that people who believe in the mission of a safety inter-
vention are usually willing to extend their commitment to safety across situations and
behaviors. Buying into the mission, they recognize the inconsistency and futility of limit-
ing safety to only certain conditions and behaviors. This leads to stimulus and response
generalization. The key to gaining commitment and ownership is involvement. When we
participate in discussing the rationale and goals of an intervention process, we are apt to
develop internal justification and support for the intervention process—and beyond.
Slight differences in the way Ludwig and the store managers implemented the behav-
ior-based intervention apparently resulted in different feelings of personal commitment
and group ownership. Involving employees in a discussion of the intervention process,
rather than lecturing information, and allowing them to choose a group safety goal, rather
than assigning one, led to response generalization.
These findings bring to mind the discussion of “choice” in Chapter 16 and the results
of my manipulations of students’ opportunity to choose stimulus materials in the labora-
tory and reading assignments in the classroom. The implications are critical for obtaining
and maintaining involvement in a safety process.
For the benefits of an intervention process to generalize across situations and behav-
iors, participants need to do more than comply with the specifics of a mandate. They need
to believe in the goals and the methods used to reach those goals. Theory and research indi-
cate that a prime way to develop this personal commitment and ownership is to involve
the participants in deciding goals and ways to attain them. Perceptions of choice and con-
trol conducive to personal commitment, ownership, and involvement can be increased by
applying the three support processes—leadership, communication, and recognition.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 470

470 Psychology of safety handbook

Figure 19.13 A shift in momentum can be good or bad.

Building and sustaining momentum

My copy of The American Heritage Dictionary (1991) defines “momentum” as “impetus in
human affairs” (page 809). Sports fans and writers use the term “momentum” to describe
a certain kind of performance in athletic competition. A team with momentum is doing
things right and on the verge of cinching a win. Then, when we see the opposing team gain-
ing ground and looking like they could win, we say “the momentum” is shifting. A shift in
momentum can also be used to explain an injury, as depicted in Figure 19.13.
A common coaching strategy is to call “time out” when it appears the opposing team
has momentum. For example, Mace et al. (1992) systematically analyzed 14 college basket-
ball games during the 1989 National Collegiate Athletic Association tournament and found
that basketball coaches called time-out from play when being outscored by their opponents
an average of 2.63 to 1.0. Calling time-outs from play was usually effective at stopping the
other team’s momentum. The rate of successful plays during the three minutes immedi-
ately after a time-out was nearly equal for both teams.
Now, my dictionary also gives a physics definition of momentum, which is “the prod-
uct of a body’s mass and linear velocity” (The American Heritage Dictionary, 1991, page 809).
Velocity is the speed or rate a mass is traveling and when the mass of a body is large, its
velocity is relatively unaffected by an external force. In other words, the greater the mass
and its velocity, the more difficult it is to stop the momentum. In sports, the more players
(mass) with a high rate or fluency of success (velocity), the greater the momentum and the
more difficult it is for the other team to make a comeback.

Relevance to industrial safety and health

Are you wondering what all of this has to do with the psychology of safety? Of course, the
notion that sports teams and safety has momentum is metaphorical. This momentum
metaphor does have intuitive appeal, and it can be useful when analyzing behavior and
making decisions about safety intervention. It captures two general and separate aspects of
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 471

Chapter nineteen: Obtaining and maintaining involvement 471

an intervention process—the number of people participating (mass) and the rate or fluency
of relevant behaviors or activities (velocity).
So, the more individuals participating in a safety-improvement process and the greater
the rate of process-related behavior, the greater the momentum. The greater the momen-
tum, the more likely the process will be sustained and contribute to the ultimate vision—a
Total Safety Culture.
Let us consider factors relevant to increasing momentum. We can return to the sports
analogy for intuitive answers to the question, “How can we build and maintain momen-
tum?” I think you will agree from personal experience that three factors are critical:
achievement of the team, atmosphere of the culture, and attitude of the coaches and
team leaders.

Achievement of the team

It is obvious that success builds success. Good performance is more likely after a run of suc-
cessful behaviors than failures. In sports, a succession of winning plays or points scored
creates momentum. When the fans notice a series of successful behaviors from their team,
they say, “The momentum has shifted to our side.”
Sports psychologists talk about momentum as a gain in psychological power, includ-
ing confidence, self-efficacy, and personal control, that changes interpersonal perception
and attitude, and enhances both mental and physical performance. It all starts with notic-
ing a run of individual or team achievements.
If momentum requires people to recognize sequences of small wins, then a scoring sys-
tem is needed that can provide ongoing objective measurement of the participants’ per-
formance. Sports events provide us with scores linked directly to individual or team
behavior and, thus, we can readily notice and support momentum. I am convinced this is
a prime reason we like to watch or participate in sporting events. We get objective and
fair feedback regarding ongoing performance. As a result, we can celebrate a win based
on observable and equitable appraisal. Sometimes, we can use these measures as feedback
to improve subsequent performance and increase the probability of more success and
continued momentum.
Hence, a key to continuous safety improvement is finding an ongoing objective and
impartial measure of performance that allows for regular evaluations of relative success.
This is why advocates of behavior-based safety emphasize the need to

• Develop up-stream process measures, such as number of audits completed or per-

centage of safe behaviors (Chapter 18).
• Set process-oriented goals that are specific, motivational, achievable, relevant, track-
able (Chapter 10).
• Discuss safety performance in terms of achievement—what people have done for
safety and what additional achievement potential is within their domain of control
(Chapter 7).
• Recognize individuals appropriately for their accomplishments (Chapter 13).
• Celebrate group or team accomplishments on a regular basis (Chapter 17).

Atmosphere of the culture

Many sports fans are fickle. When their team has a winning record, they fill the stadium.
When their team shows momentum toward winning the competition, they cheer loudly
and enthusiastically. However, teams with a losing record often play in front of a much
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 472

472 Psychology of safety handbook

smaller audience, and when the home team is not doing very well in a particular game, the
crowd is quiet. When the opposing team shows momentum, sports announcers say, “The
home fans have been taken out of the game.”
Having the home-team advantage means the fans are available to support momentum
when it occurs. Sports fans can also get a losing team going by optimistically cheering when
there is only a successive approximation to success. This atmosphere can motivate the team
to try harder, and partial success can lead to total success and then momentum to achieve
more. This stimulates the home crowd to cheer more, and momentum is supported.
I hope the relevance to safety is clear. The atmosphere surrounding the process
influences the amount of participation in a safety-improvement effort. Is the work cul-
ture optimistic about the safety effort or is the process viewed as another “flavor of the
month”? Do the workers trust management to give adequate support to a long-term inter-
vention or is this just another “quick fix” reaction that will soon be replaced by another
“priority”?
Before helping a work team implement a behavior-based safety process, my partners
at Safety Performance Solutions insist everyone in the work culture learn the princi-
ples underlying the process. Everyone in the culture needs to learn the rationale behind
the safety process, even those who will not be involved in actual implementation. This
helps to provide the right kind of atmosphere or cultural context (Chapter 14) to support
momentum.
When the vision of a work team is shared optimistically with the entire workforce, peo-
ple are likely to buy in and do what it takes to support the mission. When this happens,
interpersonal trust and morale builds, along with a winning spirit. People do not fear fail-
ure but expect to succeed, and this atmosphere fuels more achievement from the process
team. Having the right kind of leadership is necessary, of course, to help people understand
what cultural assistance is needed for momentum and, then, to help mobilize such support.

Attitude of the leaders

The coach of an athletic team can make or break momentum. Coaches initiate and support
momentum by helping both team members and the team as a whole recognize their accom-
plishments. This starts with a clear statement of a vision and attainable goals. Then,
the leader enthusiastically holds individuals and the team accountable for achieving
these goals.
A positive coach can even help members of a losing team feel better about themselves
and give momentum a chance. The key is to find pockets of excellence to acknowledge,
thereby building self-confidence and self-efficacy (Chapter 15). Then specific corrective
feedback will be heard and accepted as key to being more successful and building more
momentum (Chapter 12).
It does little good for safety leaders to reprimand individuals or teams for a poor
safety record, unless they also provide a method people can use to do better. The leader
must explain and support the improvement method with confidence, commitment, and
enthusiasm.
For momentum to build and continue, support means more than providing necessary
resources. It means looking for success stories to recognize and celebrate (Chapter 13). This
helps to develop feelings of achievement among those directly involved (the team) and an
optimistic atmosphere from others (the work culture). These are the ingredients for safety
momentum, as summarized in Figure 19.14. Keep these in place and your momentum will
be sustained. Then, you can truly expect the best from your safety improvement efforts.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 473

Chapter nineteen: Obtaining and maintaining involvement 473

Achievement Atmosphere
. Success Focus
of Team
. Shared Vision
of Culture
. Engaged in Process . Optimism
. SMARTS Goals . High Morale
. Celebrate Wins . Trust

Attitude

. Vision and Goals

of Leaders

. Commitment
. Confidence
. Enthusiasm
Figure 19.14 Three “A” factors build and maintain momentum.

In conclusion
This chapter began with a list of guidelines to initiate and sustain a culture-change process
aimed at achieving a Total Safety Culture. Critical challenges include

• Gaining sufficient top management support.

• Creating a Safety Steering Team.
• Developing valid evaluation procedures.
• Establishing an education and training process.
• Sustaining the culture change with activators, consequences, evaluation techniques,
and follow-up training.
• Dealing with outside contractors.
• Trouble-shooting, fine-turning, and updating the various process procedures.

Three support processes were identified to maintain employees’ long-term commit-

ment and involvement in a culture-change effort—leadership, communication, and recog-
nition. Each of these processes were discussed in various forms throughout previous
chapters. Aspects covered in this chapter were characteristics of effective leaders; safety
language that increases resistance and should be avoided; levels of resistance that can be
influenced by leadership, communication, and recognition; the concept of generalization
as a desired outcome of an intervention process that draws on leadership, communication,
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 474

474 Psychology of safety handbook

and recognition strategies; and three key ingredients for building and sustaining momen-
tum in a safety-improvement process.
Psychologists find the best leaders are enthusiastic, honest, motivated, confident, ana-
lytical, informed, and flexible. Although it is common to see these characteristics described
as permanent personality traits, it is certainly reasonable to assume they can be increased
through education, communication, recognition, and involvement in a successful safety
process. Thus, while it is useful to look for “natural” leaders when selecting members of a
Safety Steering Team, it is important to realize that leadership qualities could be sup-
pressed in some people by their lack of empowerment or sense of belonging. New safety
processes and eventual culture change might bring out leaders you did not know existed
in the workforce.
Involvement is key to so many aspects of building a Total Safety Culture and it can be
increased many ways. Start by getting into the habit of using more positive safety lan-
guage. Focus less attention on the active resisters. There are five levels of involvement in
any change effort that you should recognize

1. Total involvement from innovators who see change as an opportunity to improve.

2. Individuals committed but not totally involved until direction and support are given.
3. People, usually the majority, ready but on the sidelines until prodded and encour-
aged by others.
4. Doubters who see change as a problem and use learned helplessness and cynicism
as excuses to remain detached.
5. The active resisters who see change as an opportunity to resist, complain, and pro-
mote mistrust.

Active and passive resisters (categories 4 and 5) should be ignored, if possible.

Recognize and support those willing to try the new process. Employees totally involved in
the process (category 1) need to help individuals committed but not yet totally immersed
(category 2). Then these two groups can work with the majority (category 3) who need
examples to follow. You can see why it is important to cultivate leadership, communication,
and recognition skills among the “true believers” in innovation. Turn these leaders loose
and they will be your best recruiters to build the base of support for a Total Safety Culture.
One of the best ways to develop champions of change is to give true believers oppor-
tunities to teach others the principles and procedures of a safety process. When people teach
stuff they believe in, they increase their personal commitment and become continuous role
models and visible leaders of the process. A “teach-the-teacher”* process is critically impor-
tant to achieve a Total Safety Culture. First, identify your potential change agents. Then,
teach them the right principles and procedures and how to teach coworkers. The final chap-
ter reviews the principles presented in this text. They are the building blocks for a teach-the-
teacher process that integrates behavior-based and person-based psychology.

References
The American Heritage Dictionary, Houghton Mifflin Boston, MA, 1991.
Baron, R. A., Psychology, 3rd ed., Allyn & Bacon, Boston, MA, 1995.
Brehm, J. W., A Theory of Psychological Reactance, Academic Press, New York, 1966.

*A more standard term is “train-the-trainer,” but because both education and training are needed (as defined ear-
lier), I believe “teach-the-teacher” is a more appropriate label for this process.
820049_CRC1_L1540_CH19 11/10/00 2:10 PM Page 475

Chapter nineteen: Obtaining and maintaining involvement 475

Brehm, J. W., Responses to Loss of Freedom: a Theory of Psychological Reactance, General Learning Press,
New York, 1972.
Carnegie, D., How to Win Friends and Influence People, Simon & Schuster, New York, 1936.
Drebinger, J. W., Jr., Mastering Safety Communication, 2nd ed., Wulamoc Publishing, Galt, CA, 2000.
Geller, E. S., Interpersonal trust: key to getting the best from behavior-based safety coaching. Prof.
Saf., 44(4), 16, 1999.
Geller, E. S., Johnson, D. F., Hamlin, P. H., and Kennedy, R. D., Behavior modification in a prison:
issues, problems, and compromises, Crim. Just. and Behav., 4, 11, 1977.
Gray, J., Men Are from Mars, Women Are from Venus, HarperCollins, New York, 1992.
Hayakawa, S. I., Language in Thought and Action, 4th ed. Harcourt Brace Jovanovich, New York,
1978.
Hayes, S. C., Rincover, A., and Solnick, J. V., The technical drift of applied behavior analysis. J. Appl.
Behav. Anal., 13, 275, 1980.
Hersey, P. and Blanchard, K., Management of Organizational Behavior, 4th ed., Prentice-Hall,
Englewood Cliffs, NJ, 1982.
Kimble, G. A., Hilgard and Marquis’ Conditioning and Learning, Appleton-Century-Crofts, New York,
1961.
Kirkpatrick, S. A. and Locke, E. A., Leadership: Do traits matter?, Acad. Manage. Execut., 5(2), 48,
1991.
Langer, E. J., Mindfulness, Perseus Books, Reading, MA, 1989.
Langer, E. J., The Power of Mindful Learning, Perseus Books, Reading, MA, 1997.
Ludwig, T. D. and Geller, E. S., Improving the driving practices of pizza deliverers: response gener-
alization and moderating effects of driving history, J. Appl. Behav. Anal., 24, 31, 1991.
Ludwig, T. D. and Geller, E. S., Improving driving practices of professional pizza deliverers:
participative versus assigned goal setting and response generalization, J. Appl. Psychol., 82, 253,
1997.
Mace, F. C., Lalli, J. S., Shea, M. C., and Nevin, J. A., Behavioral momentum in college basketball,
J Appl. Behav. Anal., 25, 657, 1992.
New Merriam-Webster Dictionary, Merriam-Webster Inc., Springfield, MA, 1989.
Streff, F. M., Kalsher, M. J., and Geller, E. S., Developing efficient workplace safety programs:
observations of response covariation, J. Organ. Behav. Manage., 13(2), 3, 1993.
Tuckman, B. W. and Jensen, M. A. C., Stages of small group development revisited, Group Organ.
Stud., 2, 419, 1977.
Yukl, G., Leadership in Organizations, 2nd ed., Prentice-Hall, Englewood Cliffs, NJ, 1989.
Zaccaro, S. J., Foti, R. J., and Kenny, D. A., Self-monitoring and trait-based variance in leadership:
an investigation of leader flexibility across multiple group situations, J. Appl. Psychol., 76, 308,
1991.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 477

chapter twenty

Reviewing the principles

This Handbook summarizes principles for understanding the human dynamics of safety. When you
use these principles to design, execute, evaluate, and continuously improve interventions to improve
safety-related behaviors and attitudes, you are well on your way to achieving a Total Safety Culture.

“If you want to get a good idea, get a lot of ideas.”—Linus Pauling

“How should we translate these concepts into real-world application?”

“Would you please put your theory into procedures or practices we could follow in
our plant?”

I have heard questions like these at each of the Deming workshops I attended. They seemed
to disappoint Deming (1991, 1992), who would assert that the purpose of the seminar was
to teach theory and principles, not specific procedures. It was up to the participants to return
to their own organizations and devise specific methods and procedures that fit their culture.
Deming stressed the need to start with theory and then customize practices.
This text also downplays a “one-size-fits-all” solution. Packaged programs are not the
answer to safety problems, though they can be found everywhere. Sure, the quick fix might
work for a while. Short-term success follows a familiar pattern. Injuries reach unacceptable
levels, management hires a consultant, employees react with interest, injuries go down, sta-
tistics improve. Improvement is not difficult if injury rates are bad to begin with, but as I
discussed earlier, this superficial approach only improves the worst organizations for a
limited time period. For companies with a good safety record, it does not work even for the
short term.
As Deming well knew, and as I have discussed throughout this text, lasting improve-
ment is built on specific procedures that fit the culture of an organization. Outside con-
sultants can be invaluable, teaching appropriate principles and facilitating the
implementation process, but if most of the employees do not understand and believe in the
principles to begin with, well-intentioned efforts never take root.
That is why throughout this Handbook I have presented theory and principles from psy-
chological research to help you design safety-improvement interventions. The successful
applications of these principles that I have described in brief case studies are not intended
as step-by-step procedures to follow but, rather, examples to consider when customizing a
process for your culture.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 478

478 Psychology of safety handbook

The 50 principles
It all starts with theory. In this final chapter, I pull together 50 important principles that
summarize the psychology of safety and lay the groundwork for building a Total Safety
Culture. I know 50 sounds like a long list, but do not worry, I shall be brief. The principles
will be familiar to you, having come from information already covered.
Hopefully, you will find the list useful as a review and as a starting point for develop-
ing your safety-enhancement process. Some of the principles focus on design and imple-
mentation. Others explain why we often fail in safety. Most can be used as guidelines for
checking potential long-term benefits of a specific safety-improvement procedure. All will
help you appreciate the complex human dynamics of safety and health promotion.
This is not a priority list. Do not read anything into the order of principles. What I hope
is that you teach them to others. You can make a difference and bring about constructive
culture change.

Principle 1 Safety should be internally—not externally—driven.

It is common to hear employees talk about safety in terms of OSHA—the Occupational

Safety and Health Administration. It often seems they “do” safety more to satisfy the man-
dates of this outside regulatory agency than for themselves. This translates into percep-
tions of top-down control and performing to avoid penalties rather than to achieve success.
Ownership, commitment, and proactive behaviors are more likely when we work
toward our own goals, not the government’s. As discussed in Chapter 19, how we define
programs and activities can influence attitudes that shape involvement. It makes sense to
talk about corporate safety as a mission owned and achieved by the very people it benefits.

Principle 2 Culture change requires people to understand the principles and how to
use them.

In Chapter 9, I distinguished between education and training and emphasized that

long-term culture change requires both. Education focuses on theory or principles.
Training gets into the specifics of how to turn principles into effective action. Role playing
or one-to-one interaction is very important for training because participants get direct feed-
back on how they are executing procedures or processes.

Principle 3 Champions of a Total Safety Culture will emanate from those who teach
the principles and procedures.

When people teach, they “walk the talk” and become champions of change. After more
than 30 years of safety consulting, it is clear to me that success depends on the presence of
these leaders. I have seen no better way to develop champions of a campaign than to first
teach relevant theory and method, then show how others can be instructors, and finally
allow opportunities for colleagues and coworkers to teach each other.

Principle 4 Leadership can be developed by teaching and demonstrating the charac-

teristics of effective leaders.

Just because you believe in something does not guarantee you will be an effective
champion of the cause. Leaders have certain characteristics, as discussed in Chapter 19,
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 479

Chapter twenty: Reviewing the principles 479

that can be taught to and cultivated in others. People need to understand the principles
behind good leadership and the behaviors that reflect good leadership qualities. You can
also learn by observing the leadership skills of others. When you see leaders in action,
reward their exemplary behavior with quality recognition or rewarding feedback.

Principle 5 Focus recognition, education, and training on people reluctant but will-
ing, rather than on those resisting.

As discussed in Chapter 19, people resist change for many reasons. Some feel insecure
leaving their comfort zones. Some mistrust any change in policy or practice that was
not their idea. Others balk for the special attention they get by resisting. It is usually a waste
of time trying to force change on these folks. In fact, resistance hardens as more pressure
is applied.
It saves time to prioritize. First, focus on those who “want to dance,” the ones willing
to get involved. Then, turn these leaders loose on the folks who “came to the dance” but
are reluctant to get involved. At least these people are willing to consider a change pro-
posal. Peer instruction can cultivate change champions (Principle 3) as well as increase par-
ticipation. When a critical mass of individuals gets involved and achieves success as a
result of change, many initial resisters will join in—out of choice, not coercion.

Principle 6 Giving people opportunities for choice can increase commitment, own-
ership, and involvement.

A basic reason for preferring the use of positive over negative consequences to moti-
vate behavior (Chapter 11) is that people feel more free. They perceive more choice when
working to achieve rewards than when working to avoid penalties (Skinner, 1971). As illus-
trated in numerous laboratory experiments and field applications, increasing perceptions
of choice leads to more motivation and involvement in the process (Chapter 16).
Personal choice also implies personal control-enhancing empowerment and willing-
ness to actively care for others (Chapter 15). It is important to realize that eliminating the
perception of choice—by imposing a top-down mandate that restricts or constraints work
behavior, for example—can sap feelings of ownership, commitment, and empowerment,
and inhibit involvement.

Principle 7 A Total Safety Culture requires continuous attention to factors in three

domains: environment, behavior, and person.

Early on, I introduced the “Safety Triad” with behavior and person sides representing
the psychology of safety. That is the focus of this book, but do not overlook the need for
environmental change. The environment includes physical conditions and the general
atmosphere or ambiance regarding safety. The behavioral safety analysis presented in
Chapter 9 started with addressing ways to simplify the task through re-engineering. Thus,
before addressing behavior change, it is critical to improve environmental conditions that
can make a job more user-friendly and ergonomically sound.
Environment, behavior, and person factors are dynamic and interactive; a change in
one eventually impacts the other two. For example, behaviors that reduce the probability
of injury often involve environmental change and lead to attitudes consistent with the safe
behaviors. In other words, when people choose to act safely, they act themselves into safe
thinking and this often results in some environmental change.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 480

480 Psychology of safety handbook

Principle 8 Do not count on common sense for safety improvement.

Most common sense is not common. It is biased by our subjective interpretation of

unique experiences. As a researcher of psychological principles for more than 35 years, I
have become quite committed to this basic principle (discussed in Chapter 2). Indeed, I
have dedicated most of my career to discovering principles of human behavior through
systematic application of the scientific method.
You have probably noticed by now that I get quite disturbed when I read or hear pop
psychology based on unfounded intuition or common sense. Many statements I have read
or heard relating to the psychology of safety sound good but are incorrect. Profound
knowledge comes from rigorous research and theory development, and often runs counter
to common sense.

Principle 9 Safety incentive programs should focus on the process rather than
outcomes.

One of the most frequent common-sense mistakes in safety management is the use of
outcome-based incentive programs. Giving rewards for avoiding an injury seems reason-
able and logical, but it readily leads to covering up minor injuries and a distorted picture
of safety performance. The basic activator–behavior–consequence contingency (see
Chapter 8 and Principle 18) demonstrates that safety incentives need to focus on process
activities, or safety-related behaviors.
Several years ago, I consulted with a chemical plant well-known for exemplary safety
performance. The annual number of OSHA recordables among approximately 550 employ-
ees had varied from 3 to 10 over several years. At the start of 1995, management initiated
an outcome-based incentive program to reach a “step improvement” in safety. Specifically,
20 percent of a year-end bonus, amounting to $800 per employee, hinged on having 6 or
fewer OSHA recordables at the end of the year.
By mid-August 1995, the plant had experienced seven OSHA recordables. Everyone
lost that $800. Needless to say, morale for safety plummeted to an all-time low. To boost
spirits, employees were promised a significant surprise reward “that will warm your
hearts” if they could go the rest of the year without a single OSHA recordable. When I vis-
ited this plant no OSHA recordable had been reported for 117 days. The plant was on its
way to achieving the goal.
These outcome-based incentive programs were clearly well-intentioned, but I hope
you are suspicious about the result. My discussion with employees at this facility verified
that minor injuries were being covered up. In fact, one line worker remarked, “It’s only
common sense, isn’t it, that when you put so much pressure on a person to not have an
injury, they’ll be motivated to conceal it if they can.”
Yes, some people’s common sense is correct. As illustrated in Figure 20.1, it is easy to get
over focused on outcome measures and overlook processes needed to achieve the outcome.

Principle 10 Safety should not be considered a priority but a value with no compromise.

As discussed in Chapter 3, this is the ultimate vision. Safety becomes a value linked to
every priority in the workplace or wherever we find ourselves. Priorities change according
to circumstances; values are deep-seated personal beliefs beyond compromise.
Actually, it is common for people to affirm they “hold safety as a core value.” “Great,”
I say, when I hear this from an employee, “Now I can count on you to facilitate any process
that could increase the occurrence of safe behavior or decrease the occurrence of at-risk
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 481

Chapter twenty: Reviewing the principles 481

Figure 20.1 Focus on process to improve outcome statistics.

behavior.” Given the Principle of Consistency (Chapter 16), pointing out a discrepancy
between a value and a behavior should lead to behavior change.

Principle 11 Safety is a continuous fight with human nature.

I know people who meet the behavioral criteria for holding safety as a value—they
practice safety, teach it, go out of their way to actively care for the safety of others, but their
numbers are few. Why? Because human nature (or natural motivating consequences,
Chapter 11) typically encourages at-risk behavior.
The soon, certain, positive, and natural consequences of risky behavior are hard to
overcome. We are talking about comfort, convenience, and expediency. Now, consider safe
alternatives which often mean discomfort, inconvenience, and inefficiency. The inconven-
iences involved in safely locking out equipment are illustrated in Figure 20.2. When you
compete with natural supportive consequences in order to teach, motivate, or change
behavior, you are fighting human nature.

Principle 12 Behavior is learned from three basic processes: classical conditioning,

operant conditioning, and observational learning.

Through naturally occurring consequences and planned instructional activities, we

learn every day and we develop attitudes and emotional reactions to people, events, and
environmental stimuli. The mechanisms for learning voluntary and involuntary behavior
and emotions were reviewed in Chapter 8.
The critical aspect of this principle is that our actions and feelings result from what we
learn by experience, both planned and unplanned. Basic learning principles can be applied
to change what we do and feel. Experience and practice develops habits that are hard to
break, however. Morever, it is possible that natural contingencies and social influences
support a bad habit or negative attitude. So, learning new behaviors and attitudes often
requires another fight with human nature.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 482

482 Psychology of safety handbook

Figure 20.2 Safety often requires the choice of the more inconvenient and less efficient
option.

Principle 13 People view behavior as correct and appropriate to the degree they see
others doing it.

Because personal experience often convinces us that “it’s not going to happen to me,”
we need a powerful reason to perform safely when personal injury is improbable. So con-
sider this: everyone who sees you acting safely or at risk either learns a new behavior or
thinks what you are doing is okay. Now, consider the vast number of people who observe
your behavior every day. Our influence as a social model gives us special responsibility to
go out of our way for safety.

Principle 14 People will blindly follow authority, even when the mandate runs
counter to good judgment and social responsibility.

This principle was discussed in Chapter 5 as a potential barrier to safe work practices.
The fact that people often follow top-down rules without regard to potential risk is alarm-
ing. This puts special responsibility on managers and supervisors who give daily direction.
These front-line leaders could signal, even subtly, the approval of at-risk behavior in order
to reach production demands. People are apt to follow even implicit demands from their
supervisor to whom they readily delegate responsibility for injury that could result from
at-risk behavior.

Principle 15 Social loafing can be prevented by increasing personal responsibility,

individual accountability, group cohesion, and interdependence.

This principle was introduced in Chapter 16 when discussing ways to increase

group productivity and synergy. Giving up personal responsibility for safety to another
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 483

Chapter twenty: Reviewing the principles 483

person (Principle 14) could be due to the social mechanisms presumed to influence social
loafing.
It is possible to decrease blind compliance to rules that foster at-risk behavior by
manipulating factors found to decrease social loafing. Thus, workplace interventions and
action plans need to be implemented with the aim of increasing an individual’s perception
of individual accountability and personal responsibility, including their sense of group
cohesion and interdependence (Chapter 17).

Principle 16 On-the-job observation and interpersonal feedback are key to achieving

a Total Safety Culture.

Critical behavior checklists (Chapter 8) and communicating the results of checklist

observations (Chapter 12) put this principle to work. Unlike the situation depicted in
Figure 20.3, the observation and feedback process must be positive. Only then will this
basic improvement tool spread throughout a work culture. The more people giving and
receiving interpersonal feedback related to safety, the greater the improvement in safety-
related behaviors and the more injuries prevented.

Principle 17 Behavior-based safety is a continuous DO IT process with D  Define

target behaviors, O  Observe target behaviors, I  Intervene to improve behaviors,
and T  Test impact of intervention.

Good intentions cannot work without appropriate method. As Deming (1991) put it,
“Goals without method—what could be worse?” The four-step DO IT process enables con-
tinuous improvement through an objective behavior-focused approach.
As detailed in Chapter 8, people need to decide on critical target behaviors to observe.
After baseline observations are taken, an intervention is developed and implemented. By
continuing to observe the target behaviors, the impact of the intervention program can be
objectively evaluated. Results might suggest a need to refine the intervention, carry out

Figure 20.3 While some negative feedback seems warranted, it is unlikely to have a ben-
eficial effect.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 484

484 Psychology of safety handbook

Figure 20.4 We can add activators to our environment for self-direction.

another one, or define another set of behaviors to work on. The next four principles pro-
vide guidance for designing behavior-change interventions.

Principle 18 Behavior is directed by activators and motivated by consequences.

External or internal events occurring before behavior (referred to as activators) only

motivate to the extent that they signal or specify consequences. Intentions and goals can
motivate behavior if they stipulate positive or negative consequences. Understanding this
principle is critical to developing effective behavior-change techniques.
Chapter 10 showed how this principle guides the development of more effective acti-
vators and Chapter 11 outlined procedures for improving the motivational power of con-
sequences. In Chapter 9, I introduced the concept of self-directed behavior, implying that
we can provide our own activators to direct our behavior, as exemplified in Figure 20.4.
Then, when we comply with a self-arranged activator, we can use positive self-talk as a
motivating consequence.

Principle 19 Intervention impact is influenced by the amount of response informa-

tion, participation, and social support, as well as external consequences.

In Chapter 12, I discussed ways to maximize the immediate impact of an intervention.

Interventions that give specific instructions (response information) and get participants
actively involved are likely to influence behavior and attitude change. If the intervention
facilitates support from others, such as coworkers or family members, it can have lasting
effects. Figure 20.5 shows undesirable external consequences from an activator with mis-
leading information.
The role of external consequences in intervention design is a bit tricky. Pioneering
research by Freedman (1965) demonstrated the need to limit external consequences if we
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 485

Chapter twenty: Reviewing the principles 485

Figure 20.5 Misleading activators can lead to undesirable external consequences.

want people to develop internal motivation. Freedman used a mild or severe threat to pre-
vent seven- to nine-year-old boys from playing with an expensive battery-controlled robot.
In the Mild Threat condition, the boys were merely told, “It is wrong to play with the
robot.” Alternatively, the boys in the Severe Threat condition were told, “It is wrong to play
with the robot. If you play with the robot, I shall be very angry and will have to do some-
thing about it.” There were four other toys available for the boys to play with when the
experimenter left the room.
From a one-way mirror, researchers observed that only 1 of 22 boys in each condition
touched the robot. About six weeks later, a young woman returned to the boys’ school and
took them out of class one at a time to perform in a different experiment. She made no ref-
erence to the earlier study, but instructed each boy to take a drawing test. While she scored
the text, she told the boy he could play with any toy in the room. The same five toys from
the previous study, including the robot, were available.
Of the boys from the Severe Threat condition, 17 (or 77 percent) played with the robot,
compared to only 7 (33 percent) from the previous Mild Threat condition. Presumably,
more boys in the Mild Threat condition developed an internal rationale for avoiding the
robot and, as a result, avoided this toy when the external pressure was not available.
Other researchers have followed up this study and demonstrated that people are more
apt to develop internal motivation when external rewards or threats are relatively small
and insufficient to completely justify the target behavior (Riess and Schlenker, 1977). This
phenomenon has been referred to as the “less-leads-to-more effect” (Baron, 1995) and is
most likely to occur when people feel personally responsible for their choice of action and
the resulting consequences (Cooper and Scher, 1990; Goethals et al., 1979; Lepper and
Green, 1978). Recall the discussions of choice, customization, and ownership throughout
Chapter 16.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 486

486 Psychology of safety handbook

Principle 20 Extra and external consequences should not overjustify the target
behavior.

The various examples of positive consequences presented in Chapters 11 and 12

(from thank-you cards to the privileges, commendations, and small tangibles listed in
Figure 12.2) are not large nor expensive. For the reasons discussed previously, rewards
should not provide complete justification for desired behavior. We do not want people
complying with safety rules only to gain a reward or avoid a penalty. If that is the case,
what happens when we take away the consequence, good or bad? We take away the rea-
son to comply. This is why people wear PPE at work, but rarely at home.

Principle 21 People are motivated to maximize positive consequences (rewards) and

minimize negative consequences (costs).

This principle offers another reason why people are not likely to follow safe operating
procedures in the absence of external controls or behavior-consequence contingencies. As
reflected in Principle 11, natural external consequences usually support risk-taking at the
expense of safe alternatives which are usually more inconvenient, uncomfortable, or time
consuming.
Of course, this principle relates to many behaviors. In Chapter 14, it was used to
explain why people often do not rush to help in a crisis. If there are more perceived costs
than benefits to intervening, actively caring behavior is unlikely. Therefore, a prime strat-
egy for increasing safety and actively caring behaviors is to overcome the costs (negative
consequences) with benefits (positive consequences). Various kinds of consequences are
defined by the next principle.

Principle 22 Behavior is motivated by eight types of consequences: positive vs. neg-

ative, natural vs. extra, and internal vs. external.

Understanding these characteristics (as explained in Chapter 11) can enable significant
insights into the motivation behind observed behavior. Appreciating these various conse-
quences can also suggest whether external intervention is called for to change behavior and
what kind of intervention to implement.
Can you define the type of consequences motivating the biker in Figure 20.6? The
odometer provides external and natural immediate feedback to the exerciser as he pedals.
When he talks to himself while pedaling, he adds internal motivating feedback to the
situation. His evaluation of the feedback determines whether the feedback is positive or
negative.
It is possible the natural external consequences supporting ongoing at-risk behavior
cannot be overcome with extra external consequences. In this case, long-term behavior
change requires the modification of the natural consequences or the application of tech-
niques discussed in Chapter 16 to alter internal consequences. Throughout this text, I have
downplayed the use of negative consequences and the reasons are reflected in the next
principle.

Principle 23 Negative consequences have four undesirable side effects: escape,

aggression, apathy, and countercontrol.

How did you feel the last time you received a reprimand from a supervisor? Maybe,
you felt like slinking away or taking a swipe at him. Chances are you did not go back to the
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 487

Chapter twenty: Reviewing the principles 487

Figure 20.6 Natural immediate consequences can be very motivating.

job charged up. Perhaps, you wanted to do something to make him look bad. These and
other undesirable side effects of using negative consequences are discussed in Chapter 11.

Principle 24 Natural variation in behavior can lead to a belief that negative conse-
quences have more impact than positive consequences.

As detailed in Chapter 12, behavior fluctuates from good to bad for many reasons.
Peak performance seldom can be sustained, and poor performance is almost bound to get
better at some point. So if you praise someone and their performance falters, do not swear
off positive feedback. Do not overestimate the power of your reprimand if it gets some
immediate results. Keep things in perspective. Remember, only with positive conse-
quences can both behavior and attitude be improved.

Principle 25 Long-term behavior change requires people to change “inside” as well

as “outside.”

The psychology of safety requires us to consider both external behavior and internal
person factors. Chapter 15 focused on the role of person states in influencing people to
actively care for another person’s safety and health. Chapter 16 showed how outside fac-
tors can be manipulated to influence these person states and, thus, increase actively caring
behavior. A Total Safety Culture requires integrating both behavior-based and person-
based psychology. The next several principles focus on understanding “inside” factors.

Principle 26 All perception is biased and reflects personal history, prejudices,

motives, and expectations.

Appreciating this principle is key to understanding people and realizing the importance
of actively listening to others before intervening. It also supports the need to depend on
objective, systematic observation for knowledge rather than common sense (Principle 8).
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 488

488 Psychology of safety handbook

It is important to realize the reciprocal relationship between perception and behavior.

Perceptions influence actions and, in turn, actions influence perceptions. If we perceive risk,
we will act to reduce it; by acting to reduce risk, we will become more aware of other risks.
The increased popularity of safety perception surveys in industry reflects an increased
awareness of how perceptions impact safety performance. These surveys can help pinpoint
issues that need attention and activities in need of an intervention. They also can be used
to assess the person factors influenced by a particular intervention program.

Principle 27 Perceived risk is lowered when a hazard is perceived as familiar, under-

stood, controllable, and preventable.

When people perceive a new risk, they adjust their behavior to avoid it. Call it “fear of
the unknown.” The reverse is also true. As discussed in Chapter 5, research has shown that
hazards perceived as familiar, understood, controllable, and preventable are viewed as less
risky. This is why many hazards are underestimated by employees.

Principle 28 The slogan “all injuries are preventable” is false and reduces perceived
risk.

Frankly, I believe telling people all injuries are preventable insults their intelligence. They
know better. It is difficult enough to anticipate and control all environmental and behavioral
factors contributing to injuries, but controlling factors inside people is clearly impossible.
Such a slogan can make it embarrassing to report an injury and could influence a cover-
up. “If they think all injuries are preventable, they will think I was really stupid to have this
injury, so I had better not report it.” The most critical problem with this popular slogan is
that it can reduce the perception of risk. Hazards considered controllable and preventable
are perceived as relatively risk free (Sandman, 1991; Slovic, 1991).

Principle 29 People compensate for increases in perceived safety by taking more risks.

As reviewed in Chapter 5, researchers have shown that some people will compensate
for a decrease in perceived risk by performing more risky behavior. In other words, some
people increase their tolerance for risk when feeling protected with a safety device (Wilde,
1994). As shown in Figure 20.7, high technology safety engineering can give a false sense
of security. This is not the case for people who hold safety as a value (Principle 10).

Principle 30 When people evaluate others they focus on internal factors; when eval-
uating personal performance, they focus on external factors.

As discussed in Chapter 6, this principle is termed “The Fundamental Attribution

Error.” It contributes to systematic bias whenever we attempt to evaluate others, from com-
pleting performance appraisals to conducting an injury investigation. Because we are
quick to attribute internal (person-based) factors to other people’s behavior, we tend to
presume consistency in others because of permanent traits or personality characteristics. To
explain injuries to other persons, we use expressions like, “He’s just careless,” “She had the
wrong attitude,” and “They were not thinking like a team.”
On the other hand, when evaluating our own behavior, we point the finger to external
factors. Figure 20.8 illustrates this bias in a context many readers can relate to from personal
experience. This should make us stop and realize the many external variables that can be
observed and often changed to increase everyone’s safety-related behavior and reduce
injuries throughout a culture.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 489

Chapter twenty: Reviewing the principles 489

Figure 20.7 Technology can cause reduced perception of risk and increase at-risk behavior.

Principle 31 When succeeding, people over attribute internal factors, but when fail-
ing, people over attribute external factors.

This research-based principle is referred to as the “self-serving bias” (see Chapter 6)

and is sure to warp injury analyses formerly called “investigations” (Chapter 9). Placing
blame for a mistake on outside variables is just a basic defense to protect one’s self-esteem.

Figure 20.8 It feels better to project our imperfections on outside factors.

820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 490

490 Psychology of safety handbook

In most organizations, even a minor injury is perceived as a failure. As a result, the victim
is sure to avoid discussing inside, person factors contributing to the mishap.
Statements like “I was fatigued,” “I didn’t know the proper procedure,” or “My mind
was on other things,” are far less probable than “The work demands were too severe,” “The
trainer didn’t show me the correct procedure,” or “Excessive noise and heat distracted
me.” My advice is to accept the self-serving bias and allow people their ego-protecting
excuses. Then, search for measurable external factors (including behaviors) that can be
changed to reduce the probability of another injury.

Principle 32 People feel more personal control when working to achieve success than
when working to avoid failure.

The sense of having control over life events is one of the most important person states
contributing to our successes and failures. When we feel in control, we are more motivated
and work harder to succeed. We are also more likely to accept failure as something we
can change. Thus, the value of increasing people’s sense of personal control over safety is
obvious.
A prime way to increase perceptions of control over injuries is to develop scoring pro-
cedures for safety achievements, rather than focusing on the number of reported injuries as
a measure of success. This puts the emphasis on measuring process activities that can lead
to loss control or injury prevention as detailed in Chapter 18.

Principle 33 Stressors lead to positive stress or negative distress depending on

appraisal of personal control.

When we believe we can do things to reduce our stressors—work demands, interper-

sonal conflict, boredom—we are more motivated to take control. As discussed in Chapter
6, this is positive stress, an internal person state not nearly as detrimental to safety as dis-
tress. We feel distress when we believe there is little we can do about current stressors. This
state can lead to frustration, exhaustion, burnout, and dangerous behavior. As shown in
Figure 20.9, even an uncontrollable telephone message can elicit frustration and distress.
It is important to recognize states of distress in others and attempt to help them. After
actively listening to another person’s concerns, you might be able to offer constructive
suggestions. Sometimes it is useful to help people distinguish between the stressors they
can control and the ones they cannot. We can be concerned about a lot of things, but we can
only control some of these. Helping people focus on the stressors they can reduce builds

Figure 20.9 A simple telephone message can cause frustration and distress.
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 491

Chapter twenty: Reviewing the principles 491

their sense of self-efficacy, personal control, and optimism. These are the person states that
imply empowerment—“I can make a difference”—and increase one’s willingness to
actively care (Chapter 15).

Principle 34 In a Total Safety Culture everyone goes beyond the call of duty for the
safety of themselves and others—they actively care.

Here, we have a primary theme of this Handbook. While behavior-based psychology

provides methods and techniques to improve the human dynamics of safety, principles
from person-based psychology need to be considered to assure the behavior-based tools
are used. The ultimate aim is to integrate behavior-based and person-based psychology so
everyone participates in efforts to achieve a Total Safety Culture. In the ideal culture, every-
one actively cares for the safety and health of others.

Principle 35 Actively caring should be planned and purposeful and focus on the
environment, person, or behavior.

Section 5 of this text is all about actively caring for safety, from understanding why
people resist it (Chapter 12) to implementing strategies that increase it (Chapter 14). We
need to plan ways to enable and nurture as much actively caring behavior as possible,
rather than sit back and wait for “random acts of kindness.”
By considering the three domains of actively caring focus, we can sometimes get more
benefit from an act of kindness. In particular, including behavior-focused actively caring can
often result in the most benefit. For example, it is often possible to include specific behav-
ioral advice, direction, or motivation with a donation (environment-focused actively caring)
and with crisis intervention and proactive listening (person-focused actively caring).

Principle 36 Direct, behavior-focused actively caring is proactive and most challeng-

ing and requires effective communication skills.

Some acts of caring are relatively painless and effortless—contributing to a charity,

sending a get-well card, or actively listening to another person’s problems. Telling some-
one how to change his behavior can be confrontational and challenging, especially when it
is direct. Think of a parent telling a child, “I want to give you some feedback about your
behavior.” This is the type of active caring we are most likely to avoid, which is unfortu-
nate because it is the most beneficial. Even parents will pass up chances to talk about
behavior with their kids. Why?
You often hear parents lament that they had no training in how to raise children. Our
resistance is partly owing to the lack of confidence in our communication skills (Chapter
13). Proper training and practice as a safety coach increases our ability to actively care for
safety in this most beneficial way.

Principle 37 Safety coaching that starts with Caring and involves Observing,
Analyzing, and Communicating, and leads to Helping.

The basic components of effective safety coaching were presented in Chapter 12, with
each letter of COACH signifying a label for the sequence of events in the process. The
coaching process should start with an atmosphere of interpersonal Caring and an
agreement that the coach can Observe an individual’s performance, preferably with
a behavioral checklist. Then, the coach Analyzes the observations from a fact-finding,
system-level perspective. Subsequently, the results are Communicated in one-to-one
820049_CRC1_L1540_CH20 11/10/00 2:11 PM Page 492

492 Psychology of safety handbook

actively caring conversation, with the sole purpose to Help another individual reduce the
possibility of personal injury.

Principle 38 Actively caring can be increased indirectly with procedures that

enhance self-esteem, belonging, and empowerment.

This principle reflects one of the most innovative and important theories presented in
this book (Chapter 15). Substantial research is available to support each component of this
principle, but prior to my journal editorial (Geller, 1991), no one had combined these com-
ponents (or person states) into one actively caring model.
Procedures that enhance a person’s sense of self-esteem (“I am valuable”), belonging (“I
belong to a team”), and empowerment (“I can make a difference”) make it more likely that a
person will actively care for the safety or health of another person. Nourishing each of these
person states leads to the actively caring belief that, “We can make valuable differences.”

Principle 39 Empowerment is facilitated with increases in self-efficacy, personal con-

trol, and optimism.

This principle was mentioned earlier when reviewing the distinction between stress
and distress (Principle 33). When people’s sense of self-efficacy (“I can do it”), personal
control (“I am in control”), or optimism (“I expect the best”) is increased, they are less apt
to experience distress and more likely to feel empowered (“I can make a difference”). In
addition, empowerment increases one’s inclination to perform actively caring behaviors.
Notice that empowerment does not necessarily result from receiving more authority or
responsibility. In order to truly feel more empowered, people need to perceive they have
the skills, resources, and opportunity to take on the added responsibility (self-efficacy),
believe they have personal impact over their new duties (personal control), and expect
the best from their efforts to be more responsible (optimism). However, as illustrated in
Figure 20.10, what may seem like empowerment to one person might not feel like empow-
erment to another.

Principle 40 When people feel empowered, their safe behavior spreads to other situ-
ations and behaviors.

In a Total Safety Culture, people go beyond the call of duty for safety. This means they
perform safe behaviors in various situations. More specifically, they show both stimulus
generalization—performing a particular safe behavior in various settings—and response
generalization—performing safe behaviors related to a particular target behavior. Figure
20.11 depicts both stimulus and response generalization of actively caring.
Both types of generalization occur naturally when safety becomes a value rather than
a priority (Principle 10). Obviously, we need to intervene in special ways to promote safety
as a value. As reviewed in Chapter 19, our field research with pizza deliverers has shown
that facilitating empowerment is one special way to increase generalization and cultivate
safety as a value.

Principle 41 Actively caring can be increased directly by educating people about fac-
tors contributing to bystander apathy.

In Chapter 16, I discussed strategies for encouraging actively caring behavior directly.
This principle expresses the most basic procedure for doing this. Research has shown that
820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 493

Chapter twenty: Reviewing the principles 493

Figure 20.10 Empowerment is in the eyes of the beholder.

educating people about the barriers to helping others can remove some obstacles and
increase the probability of actively caring behavior. Similarly, I have found that discussing
the barriers to safe behavior can motivate people to improve safety, provided they also
learn specific techniques for doing this.

Figure 20.11 The best interventions spread their effects to other behaviors and environ-
mental settings.
820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 494

494 Psychology of safety handbook

Principle 42 As the number of observers of a crisis increases, the probability of help-

ing decreases.

This principle, supported with substantial behavioral research, is probably the first
barrier to actively caring behavior that should be taught. It is strange but true, and means
that people cannot assume that someone else will intervene in a crisis. In fact, the most
common excuse for not acting is probably something like, “I thought someone else would
do it,” or “I didn’t know it was my responsibility.” This principle reflects the need to pro-
mote a norm that it is everyone’s responsibility to actively care for safety. We can never
assume someone else will correct an at-risk behavior or condition.
At the time of this writing (June 2000), more than 50 women were sexually assaulted
and mauled by numerous men in Central Park, New York. Many of the attacks were cap-
tured on videotape because they occurred in broad daylight. Also depicted on video were
several “off-duty” police officers sitting on a park bench within earshot of the various sex-
ual incidents which reportedly continued for over an hour. Victims reported they solicited
help from these and other police officers after their incidents, but none occurred. The lack
of actively caring in this situation, from police officers to numerous by-standers (including
several videographers) is certainly disheartening, but it is clearly in line with this principle
of social behavior.

Principle 43 Actively caring behavior is facilitated when appreciated and inhibited

when unappreciated.

Making an effort to actively care directly for someone else’s safety is a big step for
many people and deserves genuine recognition. Then, if advice is called for to make the
actively caring behavior more effective, corrective feedback should be given appropriately.
Be sure to make your deposits first. All actively caring behavior is well-intentioned but not
frequently practiced with the kind of feedback that shapes improvement. A negative reac-
tion to an act of caring can be quite punishing and severely discourages a person from try-
ing again. Consequently, much of the future of actively caring behavior is in the hands of
those who receive people’s attempts to actively care.

Principle 44 A positive reaction to actively caring can increase self-esteem, empow-

erment, and sense of belonging.

This is a follow-up to Principle 43 and supports the need to sincerely recognize occur-
rences of actively caring behavior. Although research in this area is lacking, it is intuitive
that feeling successful at actively caring behavior should lead to more active caring.
Success should enhance self-esteem, empowerment, and belonging and so, indirectly,
increase the probability of more caring acts. Thus, we have the potential for a mutually
supporting cycle of actively caring influence, provided the reactions to actively caring
behavior are positive.

Principle 45 The universal norms of consistency and reciprocity motivate everyday

behaviors, including actively caring.

These social influence norms have a powerful impact on human behavior. Sometimes,
people apply these norms intentionally to influence others. At other times, these norms are
activated without our awareness. Regardless of intention or awareness, behavior-change
820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 495

Chapter twenty: Reviewing the principles 495

techniques derived from these norms can be very effective. In Chapter 16, I discussed how
these social influences can be used to directly increase actively caring behavior. The next
three principles offer more direction.

Principle 46 Once people make a commitment, they encounter internal and external
pressures to think and act consistently with their position.

This is why I say you can act people into thinking differently or think people into
acting differently. If people act in a certain way on the “outside,” they will adjust
their “inside”—including perceptions, beliefs, and attitudes—to be consistent with their
behaviors. The reverse is also true, but throughout this text I have recommended targeting
behavior first because it is easier to change on a large scale. As presented in Chapters 10
and 11, we know much more about changing behavior than perceptions, beliefs, and atti-
tudes because behavior is easier to measure objectively and reliably.
Figure 20.12 depicts a humorous scenario of rational behavior preceding the internal
emotion of fear. Is this realistic? Is it reasonable to believe that this act of running from a
bear will come before an internal person state? In fact, this is likely what happens, as pre-
dicted by the James-Lange Theory of emotion. As James (1890) put it, “We feel sorry
because we cry, angry because we strike, and afraid because we tremble” (page 1066).

Figure 20.12 Behavior precedes emotion; we are afraid because we run.

820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 496

496 Psychology of safety handbook

Principle 47 The consistency norm is responsible for the impact of “foot-in-the-door”

and “throwing a curve.”

As detailed in Chapter 16, the “foot-in-the-door” and “curve ball” techniques of social
influence succeed because of the consistency norm (Principle 46). When an individual
agrees with a relatively small request, for example, to serve on a safety committee, you
have your foot in the door. To be consistent, the person is more likely to agree later with a
larger request, perhaps to give a safety presentation at a plant-wide meeting. Similarly,
when people sign a petition or promise card that commits them to act in a certain way, say
to actively care for the safety of others, they experience pressure from the consistency norm
to follow through.
The technique of throwing a curve occurs when a person is persuaded to make a par-
ticular decision because there is not much at stake. Then, the stakes are raised. Due to the
consistency norm, the individual likely will stick with the original decision.
Here is a safety application. An employee is asked to serve on a safety committee. No
big deal—the committee meets just once a month, but then the employee is asked to attend
more meetings because a special project has come up. To remain true to his decision, the
employee will probably stay on the committee and take on the additional work.

Principle 48 The reciprocity norm is responsible for the impact of the door-in-the-
face technique.

The reciprocity norm is a powerful determinant of human behavior. Its influence is

reflected in the popular expression—”one good turn deserves another” and the well-
known Golden Rule—”Do unto others as you would have them do unto you.” This is
another reason to actively care for safety. One good act will likely lead to another.
The success of the “door-in-the-face” technique depends on the reciprocity norm. If an
employee shuts the door on a major request, he is more likely to be open to a lesser request.
If you ask for something less imposing, costly, or inconvenient after the initial refusal, your
chances of being accepted are greater than if you started with the minor request. Your will-
ingness to withdraw the larger request sets up an obligation to reciprocate and accept the
smaller request. In Chapter 16, I discussed applications of this principle to promote actively
caring behavior.

Principle 49 Numbers from program evaluations should be meaningful to all partic-

ipants and direct and motivate intervention improvement.

The last two principles relate to the critical issue of program evaluation (Chapter 18).
In safety, the total recordable injury rate (TRIR) is the most popular evaluation number
used to rank companies for safety rewards. It is calculated by multiplying the number of
workplace injuries by 200,000 and dividing the answer by the total person-hours worked
in that time period (U.S. Department of Labor, 1994). What an obvious example of an
abstract number with little meaning. The most direct measure of ongoing safety perform-
ance comes from behavioral observations and, in Chapters 8, 12, and 18, I recommended
ways to obtain meaningful feedback numbers from such process evaluation.
Throughout this Handbook, I have presented various questionnaires that assess partic-
ular person states to gauge reactions to interventions. Such evaluation tools are not as
objective and directly applicable to process improvement as feedback charts from behav-
ioral observations. Results of surveys to measure perceptions, attitudes, or person states
820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 497

Chapter twenty: Reviewing the principles 497

Figure 20.13 At-risk behavior is embarrassing if it results in an injury.

can be meaningful to program participants if explained properly. If given before and dur-
ing an intervention process, questionnaires can reflect changes in the “inside” factors that
impact program acceptance, participation, and future success.

Principle 50 Statistical analysis often adds confusion and misunderstanding to eval-

uation results, thereby reducing social validity.

Complex statistics are appropriate and often necessary for research journals. If the pur-
pose of a program evaluation is to improve a safety process, we need to provide numbers
that give the most meaningful feedback to program participants—the people in the best
position to improve the process.
Recall also the lesson from Sandman (1991) and Slovic (1991) that group statistics have
minimal impact on risk perception (Chapter 5). If your objective is to increase risk aware-
ness and motivate safe behavior, the most influential evaluation tool you can use is actu-
ally anecdotal. The most moving feedback usually comes from the personal report of an
injured employee. However, as illustrated in Figure 20.13, the victim might want to cover
up at-risk behavior leading to an injury. The culture needs to support reporting personal
injuries, as well as discussing ways to prevent future incidents.
As illustrated in Figure 20.14, when people give personal testimony, the presentation
is more useful than a statistical analysis. We should probably spend less time calculating
summary injury statistics and more time eliminating the barriers to the personal reporting
and analysis of safety-related incidents—from near hits and first-aid cases to lost-time
injuries.

In conclusion
This chapter reviews the principles of human dynamics discussed throughout this book.
Founded on research published in scientific journals, they enable profound understanding
of the psychology of safety. Use them as guidelines to develop, implement, evaluate, and
820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 498

498 Psychology of safety handbook

Figure 20.14 Personal testimony is a more powerful motivator than group statistics.

refine safety-improvement programs and you will make a positive difference in the safety
of your organization, community, or culture.
Champions are needed to lead this process. Some are easy to find; others will evolve
when the principles reviewed here are taught. Give potential champions opportunities
to teach these principles and help develop interventions. Active participation increases
both belief in the principles and the empowerment to apply them to achieve a Total
Safety Culture.
There is no quick-fix to culture change. The journey is not to be without bumpy roads,
forced detours, and missed turns. These principles are your map to reach an enviable des-
tination, but be prepared to blaze new paths and traverse difficult terrain. Please do not for-
get to take a break now and then to appreciate journey milestones. Recognize behaviors
that contribute to a successful journey.
At the end of the second Deming workshop I attended, a participant raised his hand to
ask one final question. When acknowledged, he stood and walked to the nearest micro-
phone and stated, “Dr. Deming, you have taught us many important principles to consider
when designing procedures to transform a culture. But frankly, the challenge seems over-
whelming. Can we really expect to make a difference in our lifetime?”
W. Edwards Deming, at age 92, replied, “That’s all you’ve got!”

References
Baron, R. A., Psychology, 3rd ed., Allyn & Bacon, Boston, 1995.
Cooper, J. and Scher, S. J., Actions and attitudes: the role of responsibility and aversive conse-
quences in persuasion, in The Psychology of Persuasion, Brock, T. and Shavitt, S., Eds., Freedman,
San Francisco, CA, 1990.
Deming, W. E., Quality, productivity, and competitive position, four-day workshop presented in
Cincinnati, Ohio by Quality Enhancement Seminars, Inc, May 1991.
820049_CRC1_L1540_CH20 11/10/00 2:12 PM Page 499

Chapter twenty: Reviewing the principles 499

Deming, W. E., Instituting Dr. Deming’s methods for management of productivity and quality,
two-day workshop presented in Washington, D.C. by Quality Enhancement Seminars, Inc,
January 1992.
Freedman, J. L., Long-term behavioral effects of cognitive dissonance, J. Exp. Soc. Psychol., 1, 145,
1965.
Geller, E. S., If only more would actively care, J. Appl. Behav. Anal., 24, 601, 1991.
Goethals, G. R., Cooper, J., and Naficy, A., Role of foreseen, foreseeable, and unforeseeable behav-
ioral consequences in the arousal of cognitive dissonance, J. Personal. Soc. Psychol., 37, 1179,
1979.
James, W. J., Principles of Psychology., Holt, New York, 1890.
Lepper, M. and Green, D., Eds., The Hidden Cost of Reward, Erlbaum, Hillsdale, NJ, 1978.
Riess, M. and Schlenker, B. R., Attitude changes and responsibility avoidance as modes of dilemma
resolution in forced-compliance situations, J. Personal. Soc. Psychol., 35, 21, 1977.
Sandman, P. M., Risk  Hazard  Outrage: a Formula for Effective Risk Communication, videotaped
presentation for the American Hygiene Association, Environmental Communication Research
Program, Cook College, Rutgers University, New Brunswick, NJ, 1991.
Skinner, B. F., Beyond Freedom and Dignity, Alfred A. Knopf, New York, 1971.
Slovic, P., Beyond numbers: a broader perspective on risk perception and risk communication, in
Deceptable Evidence: Science and Values in Risk Management, Mayo, D. G. and Hollander, R. D.,
Eds., Oxford University Press, New York, 1991, 48.
U.S. Department of Labor, Recordkeeping Guidelines for Occupational Injuries and Illnesses Bureau of
Labor Statistics, Washington, D.C., September 1994.
Wilde, G. J. S., Target Risk, PDE Publications, Toronto, Ontario, 1994.
1540 name index 11/11/00 1:29 PM Page 523

Name Index
A Beaman, A.I., 307, 371
Beattie, R., 331
Aaltonen, M., 7 Beck, A.T., 361
Abramson, L.V., 428 Bell, P.A., 310
Adams, J.G.U., 82, 83 Bem, D.J., 390
Adler, Alfred, 27 Berger, M., 310
Adler, R.B., 367 Berkowitz, L., 120, 343, 375
Ajzen, I., 29 Berry, T.D., 176, 180, 186
Albert, M., 100 Beruvidesk, Mario, 419
Alexander, K., 186 Betz, N.E., 331
Alimpay, D.A., 74 Bickman, L., 310
Allen, J., 168 Bierhoff, H.W., 307, 308, 342
Allowy, L.B., 428 Bigelow, B.E., 184
Amabile, T.M., 103 Bird, Frank, 16, 219, 423, 425
American Cancer Society, 101 Björgvinsson, T., 193
Armstrong, T.J., 214 Bjurstrom, L.M., 6
Aronoff, J., 329, 342 Blair, E.H., 270
Asch, S.E., 63, 64 Blake, J.A., 344
Austin, J., 253 Blanchard, K., 276, 289–290, 455
Azrin, N.H., 214 Bobbitt, John Wayne, 78
Bolen, M.H., 331, 342
B Borgida, E., 309
Boschman, I., 360
Baer, D.M., 329, 433, 434 Boyce, T.E., 110, 163, 184
Bailey, J.S., 253 Branton, R., 9
Balcazar, F., 257 Brehm, J.W., 36, 84, 195, 465
Bandura, A., 99, 115, 119, 120, 271, 331 Brehony, K.A., 110
Barker, J.A., 296 Broadbent, D., 58, 338
Barling, J., 331 Broll, L., 309
Barlow, D.H., 436 Brown, A.E., 6
Barnes, P.J., 307, 371 Brown, H.J., Jr., 298, 299, 300, 301
Baron, R.A., 120, 216, 331, 455, 485 Brown, J.D., 331, 392
Barrett, L.C., 333 Brownstein, R., 379
Barry, P.Z., 33 Bruff, C.D., 176, 178
Bar-Tal, D., 328 Bryant, S.L., 184
Bassett, R., 379 Buckwald, Art, 366
Batson, C.D., 329, 331, 342 Buermeyer, C.M., 344
Batten, Joe, 33 Bugelski, B.R., 74
Baum, A., 310 Burger, J.M., 379
Baumeister, R.F., 313 Burke, A., 431
1540 name index 11/11/00 1:29 PM Page 524

524 Psychology of safety handbook

Byers, P.Y., 338 Cross, J.A., 331, 342

Byham, W.C., 331, 393 Crutchfield, R.S., 64
Byrne, D., 331 Cuddiky, K., 306
Curley, T., 304
C Cybriwsky, R., 183

Caciioppo, J.T., 379

D
Cadwell, C.M., 393
Cady, L.D., 7 Daniels, Aubrey C., 11, 168, 226, 380
Calef, R.A., 176, 186 Daniels, L.R., 375
Calef, R.S., 176, 186 Darby, B.L., 376
Cameron, J., 13, 203 Darley, J.M., 304, 305, 309, 311, 317, 371
Campbell, B.J., 83 Dashiell, J.F., 368
Campbell, F.A., 83 Davies, R.J., 16
Canfield, J., 332 Deaux, K., 316
Canon, L.K., 310 Deci, E.L., 207, 211, 271
Carlson, M., 308, 343 Delprata, D.J., 176
Carlson, N.R., 179 Deming, W. Edwards, 11, 21, 28, 34,
Carnegie, Andrew, 129 41–42, 164, 175, 197, 203, 204,
Carnegie, Dale, 203, 252, 291, 457 296, 415, 416, 419, 438, 441, 477,
Carnevale, P.J., 375 483, 498
Carr, C., 203 DePasquale, J.D., 29, 121, 122, 145, 163,
Carrington, P.I., 375 317
Carton, J.S., 13 De Santamaria, M.C., 111, 257
Carver, C.S., 99, 100, 331, 333, 338, 343 Deutsch, F.M., 373
Casa, J.M., 331
Dickinson, A., 203
Catalan, J., 376
Dickson, W.J., 253
Chance, P., 181, 215, 279
Dickson-Markman, F., 336, 338
Chapman, M., 372, 373
Ditzian, J.L., 373
Charlin, V., 308, 343
Dodson, J.D., 92
Chesner, S.P., 313
Dollard, J., 428
Chhokar, J.S., 111, 257
Doob, L., 428
Cialdini, R.B., 271, 274, 285, 374, 376, 377,
379 Dovidio, J.F., 309, 315, 316
Clark, R.D., III, 311, 314, 315, 316 Downey, G., 101
Clarke, S., 163, 170, 184, 277 Downs, R., 342
Clausen, G.T., 308 Drebinger, J.W., Jr., 273, 448
Cone, J.D., 195, 235 Dubbert, P.M., 101
Conger, J.A., 331 Duke, M.P., 338
Conner, C., 309 Dumais, S.T., 54
Connor, W.H., 343
Cooper, J., 485 E
Cooper, M.D., 257
Cooper, P.F., 58 Easley, A.T., 110
Coopersmith, S., 342 Eason, S.L., 253
Cope, J.G., 184 Eisenberg, N., 372
Covell, V.T., 76 Eisenhower, Dwight, 265
Covey, S.R., 21, 28, 75, 98, 203, 204, 241, Eisner, H.S., 6
256, 271, 296, 355, 363, 369, 374 Eisworth, R., 191
Cox, Valerie, 242, 243 Elder, J.P., 29
Coyne, J.C., 101 Eliot, George, 325
1540 name index 11/11/00 1:29 PM Page 525

Index 525

Emerson, Ralph Waldo, 53 422, 424, 428, 433, 440, 442, 459,
Endler, N.S., 64 461, 468, 492
England, K.J., 121, 122 Gergen, K.J., 328
Eskew, R.T., 333 Gergen, M.M., 328
Evans, L., 34, 76, 83, 181 Germain, G.L., 219, 423, 425
Everett, P.B., 29 Gershenoff, A.G., 344
Eysenck, H.J., 238, 338 Gewirtz, J.L., 329
Eysenck, M.W., 238, 338 Ghandi, Mahatma, 302
Gilligan, C., 329
Gilmore, M.R., 338
F Glenwick, D., 109, 235
Fadiman, C., 303 Goethals, G.R., 485
Fairbanks, D.F., 393 Goldstein, A.P., 29, 109
Farris, J.C., 191 Goleman, D., 345, 346, 347
Feather, N.T., 333 Goodman, J.B., 256
Federal Register, 76 Goranson, R., 375
Ferrari, J.R., 371 Gouldner, A.W., 374
Festinger, L., 29, 442 Grant, B.A., 193
Gray, D.A., 148, 440
Fire, L.J., 214
Gray, J., 458
Fishbein, M., 29
Green, D., 485
Fisher, J.D., 310, 331
Greene, B.F., 29, 109, 235, 372
Fitzgerald, P., 58
Greene, T., 310
Flora, S.R., 203
Grossnickle, W.F., 184
Folkman, N., 97
Grote, D., 220
Foti, R.J., 455
Gruder, C.L., 308, 343
Frankl, Viktor, 27, 302 Grusec, J.E., 372
Fraser, S.C., 379 Guarnieri, M., 33, 34
Freedman, J.L., 379, 484 Guastello, Stephen, 4–5, 6, 109, 337
Freedman, S.M., 7
Freud, Sigmund, 27
Fulghum, Robert, 204 H
Hackett, G., 331
G Haddon, W., Jr., 33, 191
Hall, E.T., 251
Gabrenya, W.K., Jr., 368 Hammer, C., 253
Gabriele, T., 310 Hansen, L., 415, 442
Gaertner, S.L., 315, 316 Hansen, M.V., 332
Gage, H., 82 Harkins, S., 368, 369
Garber, J., 100 Hartley, S., 64
Gardner, H., 345 Hartshorne, H., 328
Geller, E.S., 8, 13, 29, 40, 83–84, 100, 109, Harvey, J.H., 103
110, 111, 121, 122, 123, 130, 145, Hayakawa, S.I., 36, 459
147, 148, 150, 153, 157, 159, 163, Hayes, S.C., 195, 235, 451
167, 168, 170, 176, 178, 180, 184, Hearn, K., 306
186, 188, 191, 192, 195, 211, 219, Heberlein, T.A., 195
226, 227, 233, 234, 235, 236, 238, Heinold, W.D., 308, 314
240, 241, 253, 255, 257, 258, 267, Heinrich, H.W., 16, 33, 111, 219, 257, 425
277, 278, 285, 297, 306, 308, 327, Heinzmann, A.T., 40, 111, 253, 257
329, 337, 338, 344, 353, 354, 364, Hernandez, A.C.R., 309
371, 372, 374, 405, 407, 408, 421, Hersen, M., 436
1540 name index 11/11/00 1:29 PM Page 526

526 Psychology of safety handbook

Hersey, P., 455 Keller, J., 374

Hidley, J.H., 40, 141, 147, 257 Kelley, A., 191
Higgins, E.G., 331 Kello, J.E., 184
Hodson, S.J., 40, 141, 147, 257 Kennedy, John F., 233, 270
Hoffman, M.L., 373 Kenny, D.A., 455
Holmes, T.H., 93 Kerrigan, Nancy, 78
Holz, W.C., 214 Kessler, M.L., 253
Hopkins, B.L., 257 Kim, J., 253
Horcones, 207, 208 Kimble, G.A., 181, 364, 467
Hornstein, H.A., 307 Kirkpatrick, F.H., 124, 454, 455
Hovell, M.F., 29 Klein, R., 307, 308, 342
Howard, J.A., 315 Klentz, B., 307, 371
Hui, C.H., 308 Knapp, L., 186
Hunt, M.M., 333 Knight, P.A., 331
Hutson, T.L., 328 Koepnick, W., 39
Huxley, Aldous, 459 Kogan, N., 390
Kohn, A., 13, 203, 207, 211, 296
Komaki, J., 40, 111, 253, 257
I Korte, C., 328
Isen, A.M., 342, 343 Kramer, K.D., 110
Kramp, P., 307, 308, 342
Isenberg, D.J., 390
Krasner, L., 29, 109
Ivancevich, J.M., 7
Krause, T.R., 13, 40, 141, 147, 257, 337, 431
Iwata, B.A., 29, 109, 235, 329
Kreitner, R., 327
Krisco, K.H., 269, 270, 275
J Kroemer, K.H., 5
Kunz, P.R., 375
J.J. Keller and Associates, 258
Jacobs, B., 338
James, William, 291, 495 L
Janis, I.L., 101, 390
Lalli, J.S., 470
Jansson, W., 84 Lamberti, D.M., 373
Jason, L., 109, 235 Langer, E.J., 85, 130, 165, 274, 388, 455,
Jenkins, E., 306 457, 458
Jensen, M.A.C., 455 Lasorda, Tommy, 387
Johnson, R.P., 110 Latané, B., 304, 305, 306, 309, 311, 317,
Jones, J.W., 90 368, 369, 371, 411
Joule, R.V., 379 Latham, G., 197
Jung, Carl, 27 Lawson, L., 40, 111, 253, 257
Lazarus, A.A., 55
K Lazarus, R.S., 97, 192
Leal, F., 9
Kahn, A., 373, 375 Leger, J.P., 6
Kaisher, M.J., 184 Lehman, G.R., 40, 82, 110, 184
Kalsher, M.R., 40, 110, 184 Lepper, M., 485
Kamp, J., 431 Lerner, J.J., 80
Kanungo, R.N., 331 Lerner, M.S., 80
Karwasky, R.J., 7 Leventhal, H., 192
Katzenbach, J.R., 393 Levin, P.F., 342, 343
Katzev, R., 184, 379 Levine, J.M., 390
Kazdin, A.E., 138, 436 Lewin, Kurt, 21
1540 name index 11/11/00 1:29 PM Page 527

Index 527

Lewis, S.K., 376 Metzgar, C.R., 81

Ley, D., 183 Michael, J., 211, 329
Lieberman, M.A., 101 Michelini, R.L., 342
Lindsley, O.R., 435, 437, 438 Michener, James, 445
Lloyd, K.E., 195, 398, 408 Midlarsky, E., 342
Lloyd, M.E., 195 Milgram, S., 64–66
Lloyd, S.R., 393 Miller, D.T., 103
Locke, E., 197, 454, 455 Miller, J.A., 379
Lubbock, John, 69 Miller, N., 308, 343, 428
Ludwig, T.D., 110, 468 Mischel, Walter, 347
Lund, A.K., 83, 181 Monty, R.A., 364
Moreland, R.L., 390
M Moriarty, T., 313
Moss, M.K., 373
Ma, Hing-Keung, 308 Mowrer, O.H., 428
Mace, F.C., 470 Murphy, L.R., 7
Maddox, J.E., 338 Mussen, P., 372
Mager, R.F., 156, 158, 160 Myers, I.B., 328, 428
Maier, S.F., 99–100, 358
Makin, P.J., 257
Malott, M.E., 380 N
Malott, R.W., 146, 380 Nadler, A., 331
Manteufel, L., 309 Naficy, A., 485
Martin, C., 344 Nair, K., 303
Marwell, G., 331 National Academy Press, 76
Maslach, C., 94
National Highway Traffic Safety
Maslow, Abraham, 27, 302, 339
Administration (NHTSA), 34
Masuda, M., 93
National Safety News, 256
Mathews, K.E., 310
Nau, P.A., 193
Matteson, M.T., 7
Nemeth, C., 64
Mawhinney, T., 203
Neuringer-Benefiel, H.E., 331, 342
May, M.A., 328
Nevin, J.A., 470
Mayer, G.R., 205
Mayer, J.A., 29, 110 Newcomb, M.D., 309
Mayo, E., 253 Newman, B., 29
McCaulley, M.H., 328, 428 Newman, O., 183
McCusker, C., 308 Nida, S., 306
McGill, K.L., 331 Nimmer, J.G., 176, 178
McGovern, L.P., 373 Norman, D.A., 58, 156, 217
McGuire, W.J., 192 Notz, W.W., 360
McLaughlin, C., 310 Nowicki, S., 333, 338
McQuirk, B., 307, 371
McSween, T.E., 29, 40, 141, 147, 257 O
Melville, Herman, 295
Mercandante, B., 338 O'Neill, B., 191
Merrill, A.R., 21 Oborne, D.J., 9
Merrill, R.R., 21 Oliner, P.M., 329
Merton, R., 333 Oliner, S.P., 329
Messe, L.A., 342 Oliver, S.D., 216
Metalsky, G.I., 428 Orne, M.T., 432
Meter, K., 328 Ozer, E.M., 331
1540 name index 11/11/00 1:29 PM Page 528

528 Psychology of safety handbook

P Redler, E., 372

Rees, F., 393, 396, 408
Page, R.S., 373 Regan, D.T., 375
Palmer, M.H., 195 Reichel, D.A., 110
Panagis, D., 192 Rhodewalt, F., 91
Pardini, A., 184 Riccobono, J.E., 253
Parker, G.M., 408 Rice, J.C., 184
Parker, K., 58 Rice, P.L., 94, 97
Parsons, H.M., 253 Richardson, D., 120
Pauling, Linus, 477 Riche, C.V., 333
Pavlov, I.P., 116 Ricks, D., 27, 29
Peake, P.K., 347 Ridgeway, C.L., 336
Peale, N.V., 203, 296 Riess, M., 485
Pearlstein, R., 203 Rincover, A., 451
Peck, M.S., 335 Risley, T., 433
Peltzman, S., 81, 82, 83 Robbins, Anthony, 10
Penner, L.A., 309 Roberts, D.S., 226, 337, 338, 344, 354, 374
Perlmuter, L.C., 364 Roberts, M.C., 186
Peter, Lawrence, 15 Robertson, L, 191, 192
Peters, Tom, 10 Rocha-Singh, I.A., 331
Petersen, D., 5, 29, 34, 36, 111, 147 Rodin, J., 315
Peterson, C., 333, 358, 428 Roethisberger, F.J., 253
Pettinger, C., 163 Rogers, Carl, 27, 271, 329
Petty, R.E., 369, 379 Rogers, R.W., 338
Phares, E.J., 238, 331 Rogers, Will, 89
Phares, E.S., 333 Romer, D., 308, 343
Phillips, J.A., 253 Rosenberg, M., 338
Phillips, J.S., 7 Rosenhan, D.L., 329
Phillips, R.A., 257 Rosenthal, A.M., 304
Pierce, W.D., 13, 203 Ross, D., 120
Pierson, M.D., 253 Ross, H.L., 196
Piliavin, I.M., 309, 315 Ross, L., 103
Piliavin, J.A., 309, 315, 316 Ross, M., 103
Pipe, P., 156, 158, 160 Ross, S.A., 120
Popper, P., 27, 29 Rotter, J.B., 331, 332
Porter, B.E., 121, 122 Rundmo, T., 7
Post, D.S., 191 Rushton, J.P., 328, 332, 372
Postmes, T., 391 Rutherford, E., 372
Prentice-Dunn, S., 338 Rutkowski, G.K., 308, 343
Preusser, D.F., 83 Ryan, R.M., 207, 271
Pringle, D.R.S., 6
Probert, L.L., 34, 36
Pruitt, D.G., 375
S
Purcell, S., 178, 186 Saarela, K.L., 6, 7
Saari, J., 7
R Sanders, P.S., 313
Sandman, P.M., 76, 77, 216, 488, 497
Rabow, J., 309 Savage, R.E., 364
Radke-Yarrow, M., 372, 373 Scheier, M.F., 99, 100, 331, 333, 338, 343
Ragnarsson, R.S., 193 Schein, E., 378
Rasmussen, D., 344 Scher, S.J., 485
1540 name index 11/11/00 1:29 PM Page 529

Index 529

Schlenker, B.R., 155, 432, 485 Synder, S., 120

Schmitz, M.F., 420 Szymanski, K., 369
Schneider, W., 54
Schroeder, D.A., 309, 312, 315
Schwartz, I.S., 434
T
Schwartz, S.H., 308, 315, 328 Talton, A., 184
Sears, R.R., 428 Tax, S.S., 360
Seligman, M.E.P., 99–100, 223, 331, 333, Taylor, S.E., 333, 361, 391
358, 372, 415, 428 Taylor, S.P., 373
Selye, Hans, 92 Teed, N., 181
Senge, Peter, 3 Teger, A., 310, 379
Shafer, M., 192 Tel-A-Train, Inc., 258
Shea, M.C., 470 Tharp, R.G., 166, 170, 236, 359
Sherer, M., 338 Thomas, P.C., 7
Sherrod, D.R., 342 Thorndike, E.L., 279
Shiffrin, R.M., 54 Thyer, B.A., 176, 178, 186
Shipley, P., 9 Tice, D.M., 313, 375
Shoda, Y., 347 Torres, C., 393
Shotland, R.L., 308, 314 Towne, N., 367
Sidman, M., 168, 215 Tracey, Brian, 10
Silverstein, B.A., 214 Triandis, H.C., 242, 308
Simon, R., 337 Tucker, R.K., 338
Simon, S., 337 Tuckman, B.W., 289, 409, 455
Skinner, B.F., 17, 28, 38, 84, 109, 114, 117,
132, 168, 195, 203, 208, 211, 214,
215, 216, 241, 296, 479 U
Sleet, D.A., 433
Ullman, L.P., 29
Sloane, H.N., 216
Umstot, D.D., 432
Slovic, P., 77, 488, 497
Smith, D.K., 393
Smith, T.W., 91 V
Solnick, J.V., 451
Spears, R., 391 Van Houten, R., 29, 109, 193, 235, 372
Staub, E., 308, 311, 328, 329, 343, 372 Vaughan, M.E., 211
Steblay, N.M., 309 Vincent, J.E., 376
Steel, S., 420 Vojtecky, M.A., 420
Steiner, I.D., 365 Volk, D., 224
Steinmetz, J.L., 103 Vollmer, T.R., 329
Stewart, J.R., 83 von Buseck, C.R., 34, 83, 180
Stewart, T., 9
Stoner, J.A.F., 390 W
Storey, C., 419
Stovie, P., 76 Wagenaar, A.C., 83
Straumann, T.J., 331 Waitley, Denis, 10
Streff, F.M., 83–84, 110, 184, 468 Wallach, M.A., 390
Strickland, B.R., 238, 333 Waller, J.A., 76, 83
Suarez, Y., 257 Wallin, J.A., 111, 257
Sulzer-Azaroff, B., 40, 111, 147, 205, 211, Wandersman, A., 27, 29
253, 257 Wang, T.H., 184
Sunaday, E., 310 Wang, Y.E., 368
Sutherland, V.J., 257 Ward, S., 29
1540 name index 11/11/00 1:29 PM Page 530

530 Psychology of safety handbook

Wasielewski, P., 34, 83 Wilson, J.M., 393

Waterman, A.S., 329 Wilson, J.P., 311, 329, 342
Watson, D.L., 166, 170, 236, 359 Winett, R.A., 29, 109, 176, 192, 235, 372
Weary, G., 103 Winn, G.L., 34, 36
Weick, K.E., 186, 358, 362 Witmer, J.F., 195
Weinstein, N.D., 392 Wixom, C., 191
Weintraub, J.K., 333 Wolf, M.M., 433
Weiss, H.M., 331 Woolcott, M., 375
Wellins, R.S., 393 Word, L.E., 311, 314
Wells, J.D., 181 Wrightsman, L.S., 316
Wells, L.E., 331 Wylie, R.C., 191, 331
West, R.C., 216
Whaley, D.L., 380
Y
Wheeler, D., 376
Wheeless, L.R., 336, 338 Yerkes, R.M., 92
Wheeless, V.E., 336, 338 Yukl, G., 197, 454
White, R.W., 365
Whitehead, T.N., 253
Wilde, G.J.S., 82, 84, 112 Z
Willems, E.P., 420 Zaccaro, S.J., 455
Williams, A.F., 83, 181 Zahn-Waxler, C., 372, 373
Williams, J.H., 110, 150, 153, 163, 253, 255 Zigarmi, D., 276
Williams, K., 368, 369 Zigarmi, P., 276
Williams, M., 178, 186 Zimbardo, P.G., 391
Wilson, D.W., 373 Zuckerman, M., 338
1540 subject index 11/11/00 1:30 PM Page 501
Subject Index
A categorizing behaviors, 298–300
ABC, see Activator-behavior-conse-
quence model
ABC Network, 122
"Accident" as word choice, 44, 460
Accident investigation
implications of, 42–43
incident analysis approach
accountability, 44–45
broad understanding need, 43
communication improvement and,
43
elements of, 44
involvement of personnel, 43–44
systems solution application, 44
Accident proneness, 7, 33
Accountability system for a team,
395–396
Accountability vs. responsibility, 170
Activator-behavior-consequence (ABC)
model
DO IT process and, 132
human behavior and, 28, 167
operant conditioning and, 117
person factors and, 327
role in Total Safety Culture, 484
Activators, see also Activator-behavior-
consequence (ABC) model
behavior influenced by, 28, 132, 145,
167
intervening with, see Intervening
with activators
role in sustaining change, 450
types of approaches, 234
Active listening, 271
Actively caring
behavior-based efforts example,
300–302
belonging and, 335–337
consequences and, 296, 315–316,
373–374
context considerations
definition of context, 317
influence example, 318–319
influence of, 317–318
safety at work and, 319
summary of influences, 320
continuous improvement and,
296–297
decision process
education and training influences,
313–314, 371–373
influence of behavior of others,
311–312
intervention responsibility, 312–313
recognition of a problem, 309–311
described, 295
factors involved, 297–298
hierarchy of needs, 302–304
increasing, see Increasing actively car-
ing
person-based approach, see Person-
based approach to actively car-
ing
psychology of
affects on inclination to care,
306–308
bystander apathy examples,
304–305
bystander apathy research, 305–306
social responsibility norm, 307
role in Total Safety Culture, 491, 492,
494
summary, 320–321
"Actively Caring Thank-You Cards,"
344, 374
Active resisters, 464–466
Adjourning stage of teams, 412
1540 subject index 11/11/00 1:30 PM Page 502
502
Aggression and perceived control by
negative consequences,
215–216
Airline Lifesaver intervention example,
188–190
"All injuries are preventable" slogan,
57–58, 363, 488
All I Really Need to Know I Learned in
Kindergarten (Fulghum), 204
Altruistic suicide, 344
American cultural norms, 307
Apathy and perceived control by nega-
tive consequences, 216
Arousal and performance, 92
"As you know" phrase, 273, 274
At-risk behavior, see also Calculated
risks
discipline for safety lapses and,
218–219
feedback process, 145
human nature and, 54
inadvertent rewards, 158–159
motivational interventions and, 168
reduction of, 111–112
risk compensation and, 83
support from consequences, 133–134
Attitude, 56–57
Attributional bias
example of, 102–103
fundamental error of, 103–104
self-serving bias, 104–105
Authority's power over safety behavior,
63, 64–66, 482
Autobiographical bias, 274
Automatic vs. controlled processing, 54,
59, 166, 217
Automobile safety
risk compensation and, 83–84
seat belt use
Airline Lifesaver example, 188–190
buckle-up road signs example,
192–195
cost-benefit evaluation, 440
cost effectiveness, 148
DO IT process and, 146
Flash for Life example, 186–188
generalization study, 468–469
habituation and reminders exam-
ple, 180–181
language choice and, 461
risk compensation and, 83
Psychology of safety handbook
safe behavior promise example,
184–185
television's impact on, 122
television activator example,
191–192
standards for, 33
Avoidance contingency, 125
B
Baseline measures of an intervention,
114–115
BASIC ID, 55–57, 66
Behavior
behavioral feedback, 165
factors in safety culture, 25–26
factors in systems approach, 42
perceived risk and, 77
psychological dimension of, 56
Behavioral safety analysis, see also
Incident analysis
behavior-based training
feedback importance, 165
reasons to use, 162–163
techniques for, 164–165
training vs. education, 163–164,
165–166
critical behavior checklist use, 154
critical behavior identification, see
Critical behaviors identification
feedback importance, 154, 165
intervention and behavior change
accountability vs. responsibility, 170
flow of change, 169–170
intervention strategies, 167–169
types of behavior, 166–167
percent safe score, 153
reducing behavioral discrepancies
consequences used effectively,
159–160
inadvertent at-risk rewards,
158–159
inadvertent punishment, 157–158
in job matching, 161
quick fix assessment, 156–157
in skills, 160
task simplification, 155–156
in training, 160–161
summary, 171–172
Behavior-based approach/programs
actively caring and, 298–302
1540 subject index 11/11/00 1:30 PM Page 503
Index
behavior-focused instruction, 277–278
change intervention influences, 47
cost effectiveness of, 29
described, 5
development of fluency and habit, 12
direct assessment and evaluation,
114–115
human element considerations, 8–9
intervention by management and
peers, 115
intrinsic consequences example,
210–211
language choice and, 461
learning from experience
classical conditioning, 116–117
definition of learning, 115–116
observational learning, 119–124
operant conditioning, 117–119
overlapping of types, 124–125
motivation and, 204
myth of habit reinforcement, 12
myths concerning success elements,
13–14
need for, 39
overview, 109–110
person-based approach vs.
behavior-based, 28
cost effectiveness, 29
described, 26–27, 30
integration of, 29–30
primacy of behavior
at-risk behavior reduction, 111–112
increasing safe behavior, 112–114
treatment or prevention, 110
response specificity impact, 176
Total Safety Culture and, 40
training
feedback importance, 165
reasons to use, 162–163
techniques for, 164–165
training vs. education, 163–164,
165–166
validity of approach, 17
Behavior change from interventions
accountability vs. responsibility, 170
approaches, 169
discipline use
employee council use, 221
positive, 220
progressive, 219–221
503
punishment contingency considera-
tions, 217–219
unpleasantness of, 216
evaluation of, 114–115
flow of change, 169–170
intervention strategies, 167–169
types of behavior, 166–167
Behaviorists, 29
Behavior observation in coaching
critical behavior checklist
development, 244–246
feedback importance, 248–250
observation process features,
247–248
observation scheduling, 246–247
sample, 246
guidelines, 243
Beliefs and expectations and active car-
ing, 308
Belonging
actively caring behavior and, 335–337
emotional intelligence and, 346
group-based behavior and
barriers to work family concept, 370
interdependence, 369
personal responsibility impact, 369
social loafing, 368–369, 411, 482–483
synergy, 367
research support for actively caring
model, 343–344
Between groups research, 83
Beyond Freedom and Dignity (Skinner), 17
Bias
attributional, see also Perception
example of, 102–103
fundamental error of, 103–104
self-serving, 104–105, 155, 489
autobiographical, 274
in communication process, 357–358
contextual, 71–72
leadership and, 458–459
past experience and, 73–75
role in Total Safety Culture, 487–488
in supportive conversation, 274–275
Bird Triangle, 425
Blood donors, 344
Brain camp, 58, 217
Bringing Out the Best in People
(Daniels), 11
British cultural norms, 307
Buckle-Up Promise Cards, 184
1540 subject index 11/11/00 1:30 PM Page 504
504
Buckle-up road signs example, 192–195
Burnout, 94, 411, 457
Bystander apathy
affects on inclination to care, 306–308
examples, 304–305
intervention responsibility and, 312
research on, 305–306
role in Total Safety Culture, 492–493
Bystander effect, 306
C
Calculated risks, see also At-risk behavior
discipline for safety lapses and, 218
emotional intelligence and, 346
human error and, 61, 62
self-directed behavior and, 166, 168
Capture errors, 59–60
CBC, see Critical behaviors checklist
CBS Network, 121, 122
Challenge, 97
Checklist, see Critical behavior checklist
Children and education on caring,
372–373, 374
Chinese cultural norms, 307
Choice, personal
generalization study and, 469
as a motivator, 365
risk perception and behavior and, 77
role in Total Safety Culture, 479
"Circle of Concern," 363
"Circle of Influence," 363
Classical conditioning, 116–117
Coaching, see Safety coaching;
Supportive conversations
Cognition, 57
Cognitive failures and safety
"all injuries are preventable" slogan,
57–58, 363, 488
capture errors, 59–60
description errors, 60
human error causes, 61–62
loss-of-activation errors, 60
mistakes and calculated risks, 61
mode errors, 60–61
propensity for, 58
punishment contingency considera-
tions, 217
Collectivistic cultural norms, 307, 372
Commitment and consistency in self-
motivation
Psychology of safety handbook
personal/interpersonal pressures, 377
public and voluntary commitment,
377–379
role in Total Safety Culture, 495–496
Common sense
common beliefs refutation, 16
common myths and, 11–16
vs. scientific knowledge, 10–11
Communication in coaching, see also
Safety coaching; Supportive
conversations
bias and, 357–358
conversations, 277
to enhance others' self-esteem, 355
eye contact use, 252
feedback
group, 255
importance of, 252
individual, 253–255
research evidence, 253
for safety, 253
social comparison, 255–256
interpersonal zone respect, 251
name use, 252
Compensation for risk
described, 80–82
implications of, 84–85
research support for, 82–84
Competency states, 169
Complacency and risk, 78
Comprehensive ergonomics
described, 5, 35
success of, 8–9
Concurrent validity, 431
Conditioned response (CR), 116–117
Conditioned stimulus (CS), 116–117
Conformity, see Social conformity
Consciously competent, 146, 217
Consciously incompetent, 145, 147
Consensus building for teamwork,
396–397
Consequences, see also Activator-behav-
ior-consequence (ABC) model
activator interventions and
goal setting, 197–199
incentives vs. disincentives, 195–197
for actively caring, 315–316, 373–374
control by, 17
critical behavior checklist used as, 145
DO IT process and, 132–134
effectiveness of, 159–160, 181
1540 subject index 11/11/00 1:30 PM Page 505
Index
effects of, 28
emotional reactions to, 118–119
extrinsic, 159–160, 207–209
interpersonal factors and, 157–158
intervention using, see Intervening
with consequences
intrinsic example, 210–211
motivational interventions and, 168
role in sustaining change, 450–451
role in Total Safety Culture, 486
selection by, 117–118
types of approaches, 234
Construct validity, 430, 432
Content validity, 430
Contests for safety, 225
Context and actively caring
definition of context, 317
influence example, 318–319
influence of, 317–318
safety at work and, 319
summary of influences, 320
Contextual bias, 71–72
Continuous improvement evaluation
costs and benefits of safety, 439–441
intervention impact measurement de-
vices, 434
issues summary, 443
performance appraisals
effective use of, 416–417
limitations, 416
performance definition
individual, 417
system, 418–419
system vs. individual, 419–420
process development
documentation areas, 420–421
environmental conditions, 423–427
intervention effectiveness assess-
ment, 421
person factors, 427–430
process outcome cause-and-effect,
421
process purpose, 420
reliability and validity of, see
Validity and reliability
scope of measurements, 422–423
work practices, 427
role in Total Safety Culture, 488–489,
496–497
statistical test data
percentages use, 437–438
505
reliance on, 436
usefulness of, 435–436
value of immeasurable things,
441–442
Contractor involvement in safety pro-
gram, 452
Control
by consequences, 17
external vs. internal, 238
personal
choice as a motivator, 365
emotional intelligence and, 346
empowerment and, 332–333
research support for actively caring
model, 342
role in Total Safety Culture, 490
strategies for enhancing, 361–363
Controlled vs. automatic processing, 54
Convergent validity, 431
Conversation, supportive, see
Supportive conservation
"The Cookie Thief," 242, 243
Corrective action plan, 221
Corrective feedback
to enhance others' self-esteem,
355–357
openness to, 166
in supportive conversation, 271, 272
Correlation coefficient, 431
Cost benefits
of behavior-based programs, 29
benefits of safety, 439–441
savings from seat belt use, 148
Cost-reward model in actively caring,
315–316
Countercontrol
in buckle-up sign example, 195
perceived control by negative conse-
quences and, 84, 216
Couple at costume ball perception ex-
ample, 70–71
CR (conditioned response), 116–117
Criterion validity, 430
Critical behavior checklist (CBC)
driving behaviors example, 143–145
feedback process, 145, 146
as part of incentive program, 229
percent safe score, 153
in safety analysis, 154
in safety coaching
development, 244–246
1540 subject index 11/11/00 1:30 PM Page 506
506
feedback importance, 248–250
observation process features,
247–248
observation scheduling, 246–247
sample, 246
Critical behaviors identification
DO IT process
basic approaches, 147–150
consequences, 132–134
described, 130–131
interventions steps, 132
target behaviors, 131, 134–136
driving behaviors example
activators and consequences,
142–143
critical behavior checklist feedback,
145
critical behavior checklist use,
143–145
incompetence to competence, 147
unconscious to conscious behavior,
145–147
observing behavior
DO IT process and SOON, 139–140
measuring behavior, 140
properties of behavior, 140
recording observations, 140–142
target behaviors definition
behavioral outcomes, 136
defining behaviors, 136
describing behaviors, 137–138
DO IT process and, 131, 134–136
interobserver reliability, 138
multiple behaviors, 138–139
person-action-situation, 137
Total Safety Culture basis, 129–130
CS (conditioned stimulus), 116–117
CTDs (cumulative trauma disorders),
214
Culture, see also Culture change; Total
Safety Culture
and active caring, 308
conversation and, 267–268
Culture change, see also Total Safety
Culture
challenges of, 473
contractor involvement, 452
cultivating continuous support
management vs. leadership,
455–459
safety leader qualities, 453–455
Psychology of safety handbook
education and training process set up
course content, 448
effectiveness evaluation, 449
instructional process planning,
448–449
instructor selection, 447–448
evaluation procedures development,
447
involvement importance, 474
language choice and use, 459–462
management support need, 446
momentum and
leaders' attitudes, 472
program atmosphere, 471–472
relevance to occupational safety,
470–471
success evaluation, 471
teams impacted by, 470
overcoming resistance
active resisters, 465–466
level of participation, 463–464
peer influence, 466–467
teaching skills, 463
types of resistance, 464–465
paradigm shifts for
accident proneness mentality, 7, 33
achievement orientation need,
38–39
behavior focused approach, 39–40
bottom-up involvement, 40–41
continuous improvement, 46
corporate responsibility and, 37
fact finding focus, 42–45
interdependence, 41
old three Es, 33–34
paradigm definition, 36–37
proactive, 45
systems approach, 41–42
value based, 46
quick fix absence, 445–446
role in Total Safety Culture, 478, 487
safety generalization
research results, 468–469
types, 467–468
safety steering team creation, 446
sustaining the process
awareness/reminders, 450
follow-up, 451–452
incentives/rewards, 451
management support need, 450
ongoing measurement, 451
1540 subject index 11/11/00 1:30 PM Page 507
Index
performance feedback, 450–451
troubleshooting/fine tuning,
452–453
Cumulative trauma disorders (CTDs),
214
D
Danger compensation, see Risk, com-
pensation
Defensive working style, 204
Deindividuation, 391–392
Delegating conversations, 277
Demand characteristics, 432
Department of Transportation, U.S., 34
Description errors, 60
The Different Drum: Community
Making and Peace (Peck), 335
Diffusion of responsibility and active
caring, 306–307; see also
Responsibility for safety
Directive conversations, 271
Discipline as intervention
employee council use, 221
positive, 220
progressive, 219–221
punishment contingency considera-
tions, 217–219
unpleasantness of, 216
Discrepancies, reducing behavioral
consequences used effectively,
159–160
inadvertent at-risk rewards, 158–159
inadvertent punishment, 157–158
in job matching, 161
quick fix assessment, 156–157
in skills, 160
task simplification, 155–156
in training, 160–161
Discriminative stimulus, 124
Disincentive-penalty programs
activator effectiveness and, 195–197
behavioral-based safety and, 159
motivational interventions and, 168
Distress definition, 91
Divergent validity, 431
Documentation
of evaluation process, 420–421
importance of, 140–142
of safety team meetings, 402
DO IT process
507
basic approaches, 147–150
behavior observation and, 139–140
consequences, 132–134
described, 130–131
interventions steps, 132
role in Total Safety Culture, 483–484
target behaviors, 131
Driving behaviors
activators and consequences, 142–143
critical behavior checklist feedback,
145
critical behavior checklist use, 143–145
incompetence to competence, 147
skill maintenance training and, 161
unconscious to conscious behavior,
145–147
Drugs and safety, 57
DuPont STOP, 112
Duration property of behavior, 140
E
Education, see also Training
actively caring effected by, 371–373
approach to safety, 34
behavior focused approach, 277–278
children and actively caring and,
372–373, 374
culture change process
course content, 448
effectiveness evaluation, 449
instructional process planning,
448–449
instructor selection, 447–448
influence on actively caring behavior,
313–314
training vs., 163–164, 165–166
Emergencies and impulse to actively
care, 309–311
Emotional bank account concept, 241,
355
Emotional intelligence
impulse control and, 346–347
intra- vs. interpersonal, 345–346
relevance to occupational safety, 348
Emotional reactions to consequences,
118–119
Empathy, 157, 289
Employee discipline council, 221
Empowerment
approach to safety, 35
1540 subject index 11/11/00 1:30 PM Page 508
508
described, 331
optimism, 333–335
personal control, 332–333
role in Total Safety Culture, 492
self-efficacy, 331–332
Enforcement approach to safety, 34, 35
Engineering changes approach to safety
described, 6, 34
task simplification from, 155–156
Environmental audit, 149
Environmental factors
actively caring and, 297
evaluation conditions
checklist use, 423
property damage incidents, 423–427
safety culture role, 25–26
systems approach and, 42
Ergonomics approach to safety
described, 5, 35
success of, 8–9
Errors from cognitive failures
capture , 59–60
causes, 61–62
description , 60
judgement, 61
loss-of-activation , 60
mode , 60–61
Escape and perceived control by nega-
tive consequences, 215
Establishing operations, 329
Esteem, see Self-esteem
Evaluation approach to safety, 36
Event recording, 141
Execution factors and at-risk behavior
reduction, 111
Expectations and active caring, 308
Experience and bias, 73–75
Expertise and leadership, 455
External consequences, 209, 211
External locus of control, 238, 332
Extinction, 133
Extra consequences, 211
Extrinsic consequences, 160, 207–209
Extroverts, 238
ExxonMobile Chemical, 258, 438–439
Eye contact in coaching, 252
F research results, 468–469
Faking good, 432
Fear-arousing approach to safety, 192, 215
Psychology of safety handbook
Feedback
behavior-based, 157, 165
communication in coaching and
group, 255
importance of, 252
individual, 253–255
research evidence, 253
for safety, 253
social comparison, 255–256
corrective
to enhance others' self-esteem,
355–357
openness to, 166
in supportive conversation, 271, 272
from critical behavior checklist, 145,
146, 248–250
impact in buckle-up road sign exam-
ple, 193
importance to safety analysis, 154, 165
for intervention process, 115
job-specific focus need, 357
natural, 207
performance appraisal timing, 416
role in sustaining change, 450–451
role in Total Safety Culture, 483
skill maintenance training and, 161
from tasks, 208
on team assignments, 402–403
on team performance, 403–405
Fight-or-flight syndrome, 91
First Things First (Covey and Merrill),
21
Flash for Life example, 186
Flexibility and leadership, 455
Follow-up measures for intervention
process, 115
"Forgetting," 60
Forming stage of teams, 409
FOX Television Network, 120–121, 122
Frequency records, 141
Frustration, 428
"The Fundamental Attribution Error,"
488
G
Generalization, safety
types, 467–468
General Motors Research Laboratories,
180
1540 subject index 11/11/00 1:30 PM Page 509
Index
Genovese, Kitty, 304
Global percent safe score, 255–256
Goal setting, 197–199
Goal statements, 38
Golden Rule reconsideration, 15
Government safety regulations, 37–38,
112
Group-based behavior
active caring decision and, 311–312
belonging and
barriers to work family concept, 370
interdependence, 369
personal responsibility impact, 369
social loafing, 368–369, 411, 482–483
synergy, 367
consensus building for teamwork,
396–397
Group cohesiveness, 336
groupthink, 311–312, 390–392
ineffectiveness of contingencies,
222–223
problem solving, 6
recognition of safety achievement,
281, 286–289
role in Total Safety Culture, 494
Group polarization, 390–391
Groupthink, 311–312, 390–392
H
Habits, 12, 166; see also Habituation
Habituation
described, 178–179
overreliance on activators from, 182
radar detector use example, 181
safety-belt reminders example,
180–181
warning beepers example, 181–182
Hairston, Tywanii, 325
Harm, 97
Hawthorne Effect study, 253
Hazards
acceptable consequences and risk,
79–80
perceived risk and, 78–79
Heinrich's Law, 16, 111, 219, 257, 425
HELP acronym in coaching, 256–257
Hierarchy of Needs, 302–304
Hoechst Celanese safety program,
226–228
Honesty and integrity and leadership,
509
454
Human element in safety, see also
Person factors in behavior
barriers to safety, 66
cognitive failures
"all injuries are preventable" slogan,
57–58, 363, 488
capture errors, 59–60
description errors, 60
human error causes, 61–62
errors, 60
mistakes and calculated risks, 61
mode errors, 60–61
propensity for, 58
punishment contingency considera-
tions, 217
considerations in behavior-based pro-
grams, 8–9
human nature
at-risk behavior, 54
barriers to safety, 53–54
controlled to automatic processing,
54
dimensions of, 55–57
role in Total Safety Culture, 481
perceived risk and safety, 78
in workplace injury, 8–9
Humanism, 27, 28, 29, 210–211; see also
Person-based approach
Humor and coaching, 256
I
"I.M. Ready" program, 222–223
Imagery, 57
Impression management, 432
Impulse control, 346–347
Incentives/rewards programs
activator effectiveness, 195–197
active caring influenced by, 373
in behavior-based approach, 28, 159
material reward use considerations,
224
motivational interventions and, 168
myth of detrimental effects, 13
myth of recognition's value, 12
outcome focused programs need, 39
promotion use myth, 15
reciprocity principle and, 376
reinforcers vs. rewards, 380–381
reward definition, 380
1540 subject index 11/11/00 1:30 PM Page 510
510
reward suggestions, 235
role in Total Safety Culture, 480
safety rewards considerations
guidelines for effective approaches,
223–226
"Mystery Observee" program,
228–229
program example, 226
safety thank-you cards use, 226–228
types of ineffective approaches,
222–223
tangibles use, 282, 288, 451
for teamwork, 388–389
value of rewards, 381
Incident analysis, see also Behavioral
safety analysis
vs. accident investigation, 44
accountability and, 44–45
behavior-based, 162
broad understanding need, 43
communication improvement, 43
involvement of personnel, 43–44
systems solution application, 44
Increasing actively caring
direct effects on behaviors
commitment and consistency,
377–379
consequences and, 373–374
education and training, 371–373
reciprocity principle, 374–377
reinforcers vs. rewards, 380–381
indirect strategies use, 353
person state enhancement strategies
belonging, 367–371
choice as a motivator, 364–366
importance of feelings, 354
optimism, 366–367
personal control, 361–363
self-efficacy, 357–361
self-esteem, 354–357
summary, 381–382
Independence, 336
Individualistic cultural norms, 307
Injury prevention
"all injuries are preventable" slogan,
57–58, 363, 488
probability reduction, 213
property damage incidents and, 16,
423–425
risk compensation impact on, 82
Instructional intervention, 167
Psychology of safety handbook
Instructive conversations, 277–278
Instructor selection and culture change,
447–448, 478
Intelligence, emotional, see Emotional
intelligence
Intensity, 140
Interdependence and safety, 41
Intermittent reinforcement schedule and
motivation, 207, 208
Internal consequences, 208, 209, 211
Internal locus of control, 238, 332
International Safety Rating System
(ISRS), 6
Interpersonal communication, 265; see
also Supportive conservation
Interpersonal conversation, see
Supportive conservation
Interpersonal factors in safety, see also
Person factors in behavior
authority's power, 64–66, 482
barriers of conformity and authority,
63
consequences and, 157–158
peer influence, 63–64
psychological dimension of, 57
stressors questionnaire, 94
trust and, 41, 130
Interpersonal intelligence, 345
Interpersonal relationships, 336
Interpersonal zone respect in coaching,
251
Interpretation, 269
Interval recording, 141
Intervening with activators
behavior specificity principle, 176
consequence implication
activator effectiveness and, 195, 196
goal setting, 197–199
incentives vs. disincentives, 195–197
habituation vs. salience
described, 178–179
overreliance on activators from, 182
radar detector use example, 181
safety-belt reminders example,
180–181
warning beepers example, 181–182
message/sign variation
changeable design, 182
worker-designed slogan use, 183
poster/sign use, 175–176
target audience involvement
1540 subject index 11/11/00 1:30 PM Page 511
Index
Airline Lifesaver example, 188–190
Flash for Life example, 186–188
ownership-involvement principle,
183–184
safe behavior promise example,
184–185
timing and placement considerations
buckle-up road signs example,
192–195
message effectiveness and, 190–191
point-of-purchase placement, 191
television's impact, 191–192
Intervening with consequences
managing for safety
behavior-consequence contingen-
cies, 213–214
discipline and involvement,
216–221
negative consequence avoidance,
214–216
motivation principles, 203–204
power of consequences
behavior vs. humanistic perspective
example, 210–211
vs. external, 209
intrinsic vs. extrinsic, 207–209
in school, 205–207
simple rules importance, 204–205
types of consequences, 211–212
safety rewards considerations
guidelines for effective approaches,
223–226
"Mystery Observee" program,
228–229
program example, 226
safety thank-you cards use, 226–228
types of ineffective approaches,
222–223
Interventions
agents per target population, 238
as a behavior-change agent, see Safety
coaching
behavior change from, see Behavior
change from interventions
discipline used as
employee council use, 221
positive, 220
progressive, 219–221
punishment contingency considera-
tions, 217–219
unpleasantness of, 216
511
evaluation of, 114–115
levels of, 236–237
role in Total Safety Culture, 484–485
supportive conversation use, see
Supportive conservation
target audience involvement
Airline Lifesaver example, 188–190
Flash for Life example, 186–188
ownership-involvement principle,
183–184
safe behavior promise example,
184–185
timing and placement considerations
buckle-up road signs example,
192–195
message effectiveness and, 190–191
point-of-purchase placement, 191
television's impact, 191–192
Intrapersonal communication
described, 265
self-esteem and, 269
shifting from past to future to present,
359
"tune-out" filters and, 275
verbal commitment indications, 271
Intrapersonal intelligence, 345
Intrinsic motivation, 203, 207
Intrinsic vs. extrinsic consequences,
207–209, 210–211
Introverts, 238
Involvement, 130, 336
ISRS (International Safety Rating
System), 6
J
James-Lange Theory of Emotion, 495
Japanese cultural norms, 307
Job hazard analyses, 135
Job safety analysis records importance,
140–142
Job safety myths, 16
Judgment errors, 61
K
Knowledge factors and at-risk behavior
reduction, 111
L
1540 subject index 11/11/00 1:30 PM Page 512
512
Language choice and use, 459–462
Language in Thought and Action
(Hayakawa), 36
"Law of Effect," 279
Leadership
vs. management, 455–459
momentum and, 472
qualities of, 453–454
role in safety culture, 66
role in Total Safety Culture, 478–479
Leadership and the Art of Conversation
(Krisco), 269
Learned helplessness, 99–100, 223
Learning
from experience
classical conditioning, 116–117
definition of learning, 115–116
observational learning, 119–124
operant conditioning, 117–119
overlapping of types, 124–125
role in Total Safety Culture, 481
from successes vs. failures, 279
by trial and error, 279
"Less-leads-to-more effect," 485
Levels of interventions, 236–237
Life's Little Instruction Book (Brown), 298
Listening and coaching, 256
"Litter begets litter," 424
Locus of control, 332
Loss-of-activation errors, 60
Lottery incentive programs, 224
M principles of, 203–204
Maintaining involvement in safety, see
Culture change, sustaining the
process
Management
active caring and, 370
behavior-based intervention, 115
leadership vs., 455–459
personnel
Golden Rule reconsideration, 15
matching talent to jobs, 13–14, 161
myth of ability to do any job, 13
myths concerning, 14–15
talents of employees and, 14–15
safety support needed from, 446, 450
"Marshmallow Test," 347, 348
McKee, John, 325
Measuring behavior, 140
Psychology of safety handbook
Men Are from Mars, Women Are from
Venus (Gray), 458
Mental attitude and myth of negative
thoughts, 12
MIL (multiple intervention level) hierar-
chy, 236–237
Mine Safety and Health Administration
(MSHA), 37, 135
Mission statement development, 21–22,
24–26
Mistakes and calculated risks, 61
Mode errors, 60–61
Momentum in culture change
leaders' attitudes, 472
program atmosphere, 471–472
relevance to occupational safety,
470–471
success evaluation, 471
teams impacted by, 470
Monitoring achievement to increase safe
behavior, 113–114
Monitoring progress for intervention
process, 115
Mood states, 308, 310
Motivation
achievement orientation need, 38–39
choice as a motivator, 364–366
intermittent reinforcement schedule
and, 207, 208
internal, 485
leadership and, 454–455
myth of self-motivation only, 12–13
role in Total Safety Culture, 486
self-directed behavior
commitment and consistency in,
377–379
described, 166
quality recognition and, 284
requesting recognition, 285–286
self-motivation, 203, 210–211
self-reinforcement need, 209
supportive conversation used for, 278
Motivational intervention, 168–169
Motor vehicles, see Automobile safety
MSHA (Mine Safety and Health
Administration), 37, 135
Multiple intervention level (MIL) hierar-
chy, 236–237
Myers-Briggs Type Indicator, 27, 328,
427–428
1540 subject index 11/11/00 1:30 PM Page 513
Index
"Mystery Observee" program, 228–229
Myths vs. objective research, 11–16
N 149, 153
Name use in coaching, 252
National Highway Traffic Safety
Administration (NHTSA), 33,
123
Natural consequences
at-risk behavior and, 159
motivation and, 204
power of consequences and, 211
Natural feedback, 207
NBC Network, 121, 122
Near hits
at-risk behavior reduction and, 111
reporting, 8, 78, 90
target behaviors and, 135
Near miss incidents
approach to safety, 8
ratio triangle, 425–426
Negative attitudes, 112
Negative consequences
calculated risks and, 168
compliance with activators and, 133
cost-reward model and, 315, 316
personal control and, 238
reasons to avoid, 214–215
role in Total Safety Culture, 486–487
from tasks, 208
NHTSA (National Highway Traffic
Safety Administration), 33, 123
Nondirective psychotherapy, 271
Norming stage of teams, 410
NORPAC safety program, 228–229
O consequences acceptability and, 79–80
Obedience studies, 64–66
Observational learning
children and actively caring and, 372
example setting, 123
television and, 120–122
television and safety learning,
122–123
vicarious consequences, 119–120
Obtaining involvement in safety, see
Culture change
Occupational Safety and Health
513
Administration (OHSA), 37, 39,
112, 135
One-to-one safety coaching
critical behavior checklist use, 147,
feedback importance, 250
as part of incentive program, 229
Operant conditioning, 117–119
Operant learning, 372
Optimism
emotional intelligence and, 346
empowerment and, 333–335
research support for actively caring
model, 342–343
resistance to stress from, 99
strategies for enhancing, 366–367
Other-directed behavior, 166
Ownership, 130
Ownership-involvement principle,
183–184
P
Passion and leadership, 454
Passive reactions to danger, 311
Pavlov's dogs, 116
Peer influence
active resisters and, 464–466
behavior-based interventions and, 115
culture change process and, 466–467
on safety behavior, 63–64
in school, 206–207
Penalties, see Punishment approach to
safety
People factor, see Person factors in be-
havior
Perceived risk
choice and, 77
familiarity and complacency, 78
hazard qualities and, 78–79
publicity and, 78
real risk vs., 76–77
role in Total Safety Culture, 488
sense of fairness and, 80
sympathy for victims and, 78
Percentage of agreement, 138
Percentage of occurrence per opportu-
nity, 140
Percentages use for evaluation, 437
Percent safe behavior, 140, 150, 249, 255
1540 subject index 11/11/00 1:30 PM Page 514
514
Percent safe score, 150, 153
Perception, see also Bias
contextual bias and, 71–72
described, 69–70
of helplessness, 428
instructions' impact on, 70–71
past experience bias and, 73–75
perceived risk
choice and, 77
consequences acceptability and,
79–80
familiarity and complacency, 78
hazard qualities and, 78–79
publicity and, 78
real risk vs., 76–77
sense of fairness and, 80
sympathy for victims and, 78
premature cognitive commitment,
85–86, 274
relevance to safety culture, 75–76
risk compensation
described, 80–82
implications of, 84–85
research support for, 82–84
self-efficacy and, 357, 358–359
Performance appraisals
effective use of, 416–417
for individuals, 417, 419–420
limitations, 416
as stressors, 94
for teamwork, 404–405
timing for feedback, 416
Performing stage of teams, 411
Person-action-situation, 137
Personal control
choice as a motivator, 365
emotional intelligence and, 346
empowerment and, 332–333
research support for actively caring
model, 342
role in Total Safety Culture, 490
strategies for enhancing, 361–363
Personal protective equipment (PPE)
behavior-consequence contingencies
and, 213–214
CBC feedback process, 149–150
development of, 33
false sense of safety from, 82
Personal zone respect in coaching, 251
Person-based approach/programs
actively caring and, see Person-based
Psychology of safety handbook
approach to actively caring
behavior-based approach vs.
cost effectiveness, 29
described, 26–27, 30
integration of, 29–30
challenges of, 29
described, 27–28
motivation and, 204
Total Safety Culture and, 40
Person-based approach to actively car-
ing
ABC model, 327
behavior change basis, 326–327
emotional intelligence
impulse control and, 346–347
intra- vs. interpersonal intelligence,
345–346
relevance to occupational safety, 348
measuring states
efficacy of surveys, 337
regression results, 338
safety culture survey, 338, 428
survey questions, 338–339, 340,
428–430
overview, 297, 325–326
personality traits of helpful people,
328–329
research support for actively caring
model
belonging, 343–344
direct test of model, 344–345
optimism, 342–343
personal control, 342
self-esteem, 342
states of actively caring
belonging, 335–337
empowerment, 331–335
self-esteem, 329–331
theoretical support for actively caring
model, 339–341
traits vs. states, 328
Person factors in behavior, see also
Human element in safety;
Interpersonal factors in safety
cognitive failures
"all injuries are preventable" slogan,
57–58, 363, 488
capture errors, 59–60
description errors, 60
human error causes, 61–62
loss-of-activation errors, 60
1540 subject index 11/11/00 1:30 PM Page 515
Index
mistakes and calculated risks, 61
mode errors, 60–61
propensity for, 58
human nature and safety
at-risk behavior, 54
barriers to safety, 53–54
controlled to automatic processing,
54
dimensions of, 55–57
improvement evaluation process and,
427–430
interpersonal
authority's power, 64–66
barriers of conformity and author-
ity, 63
peer influence, 63–64
safety culture role, 25–26
social support factors, 238
stressors coping mechanism, 99–100
systems approach and, 42
Total Safety Culture and, 40
Personnel management
Golden Rule reconsideration, 15
matching talent to jobs, 13–14, 161
myth of ability to do any job, 13
myths concerning, 14–15
talents of employees and, 14–15
Person states
improvement evaluation and, 428
measuring
efficacy of surveys, 337
regression results, 338
safety culture survey, 338, 428
survey questions, 338–339, 340,
428–430
vs. traits, 328
Perverse compensation, see Risk, com-
pensation
Pessimist vs. optimist distinction, 100,
333
"The Peter Principle," 15
Physical demands of work stressors
questionnaire, 94
Physical factors and at-risk behavior re-
duction, 111
Physical fitness and stress, 100–101
Pluralistic ignorance, 311
Point-of-purchase advertising, 191
Polarization, group, 390–391
Pop psychology
common myths from, 11–16
515
humanism and, 27
praise in public philosophy, 281
self-help strategies shortcomings,
9–10
self-motivation and, 203
Positive consequences
actively caring and, 373
calculated risks and, 168
compliance with activators and, 133
cost-reward model and, 316
power of, 206
recognition of safety achievement
and, 279
Positive discipline as intervention, 220
Poster/sign campaigns and safety
impact of, 7
as part of incentive program, 224
as part of intervention, 175–176, 182,
183
seat belt use impacted by, 192–195
variation need, 182
worker-designed slogan use, 183
PPE, see Personal protective equipment
Praise use
coaching and, 256–257
praise in private philosophy, 281
Predictive validity, 431
Prejudgment filters, 274
Premature cognitive commitment,
85–86, 274
Primary appraisal of stressors, 97
Principle-Centered Leadership (Covey),
21
Principle-centered motivation, 204
Proactivity, 44, 112
Probability of injury, 213
Progressive discipline as intervention,
219–221
Promotion use myth, 15
Property damage incidents
impact on workplace, 423–425
as indication of personnel injury, 16
personal blame removal, 426–427
ratio triangle, 425–426
Proxemics, 251
Psychological approach to safety, 33
Psychological reactance research, 35, 84,
465
The Psychology of Everyday Things
(Norman), 58
Psychotherapy approach, 26
1540 subject index 11/11/00 1:30 PM Page 516
516
Publicity and risk, 78
Punishment approach to safety
contingency considerations, 217–219
inadvertent, 157–158
reasons to avoid, 35, 214–216
vs. reward approach, 117, 118, 124
Q feedback process, 145
Questionnaires, see Surveys
R reduction of, 111–112
Radar detector use and habituation, 181,
216
"Random acts of kindness" concept, 295,
300
Rate of a behavior, 140, 141
Reactive approach, 112
Real risk vs. perceived risk, 76–77
Reciprocity principle
actively caring behavior and, 374–377,
379
bias avoidance using, 274
recognition and, 285
role in Total Safety Culture, 494–495,
496
Recording behavior observations, see
Documentation
Red/white bead game, 418
Reliability, see Validity and reliability
Renewing a team, 407–408
Resistance to change, overcoming
active resisters, 465–466
level of participation, 463–464
peer influence, 466–467
teaching skills, 463
types of resistance, 464–465
Response frequency, 140, 141
Response generalization, 467
Response information, 238
Response specificity impact, 176
Responsibility for safety
accountability vs., 170
active caring and, 306–307, 312–313
bystander apathy and, 312
corporate responsibility and, 37
diffusion of, 306–307
group-based behavior and, 369
in Total Safety Culture, 313
Psychology of safety handbook
Restructuring a team, 406–407
Reward vs. punishment, 117, 118, 124;
see also Incentives/rewards
programs
Risk
at-risk behavior
discipline for safety lapses and,
218–219
human nature and, 54
inadvertent rewards, 158–159
motivational interventions and, 168
risk compensation and, 83
support from consequences,
133–134
calculated
discipline for safety lapses and, 218
emotional intelligence and, 346
human error and, 61, 62
self-directed behavior and, 166, 168
compensation
described, 80–82
implications of, 84–85
research support for, 82–84
role in Total Safety Culture, 488
perceived
choice and, 77
consequences acceptability and,
79–80
familiarity and complacency, 78
hazard qualities and, 78–79
publicity and, 78
real risk vs., 76–77
role in Total Safety Culture, 488
sense of fairness and, 80
sympathy for victims and, 78
Risk homeostasis, see Risk, compensa-
tion
"The risky shift," 390
"Road rage," 317
Role playing demonstrations, 138
Root cause of an accident, 43, 44
Rule governed behavior, 146
S
Safe behavior opportunity (SBO), 148,
149
"Safe Behavior Promise Card," 377, 379
Safe behavior promise example, 185
1540 subject index 11/11/00 1:30 PM Page 517
Index
Safety belt use, see Seat belt use
Safety celebrations guidelines, 286–289
Safety coaching, see also
Communication in coaching;
Supportive conversations
action plan guidelines, 261
consequences of, 240
description of coaching, 239–240
impact on behaviors, 257–259
intervention approach selection
increasing impact, 238
intervention levels, 236–238
reward suggestions, 235
types of approaches, 234
one-to-one
critical behavior checklist use, 147,
149, 153
feedback importance, 250
as part of incentive program, 229
process of
ABC analysis, 250
behavior observation, see Behavior
observation in coaching
caring attitude, 241–242
communication, see
Communication in coaching
purpose of helping, 256–257
role in Total Safety Culture, 491–492
self-appraisal of skills, 259, 260
Safety Culture Survey, 338–339, 340, 344,
428–430
Safety estimates, 153
Safety generalization
research results, 468–469
types, 467–468
Safety improvement programs
approach selection
common sense and common myths,
11–16
common sense vs. scientific knowledge,
10–11
human element, 8–9
research evaluation, 3–5, 24
research programs descriptions, 5–8
self-help strategies shortcomings,
9–10
seminar shortcomings, 9
research importance
behavioral science application,
16–18
common beliefs refutation, 16
517
Safety rewards, see Incentives/rewards
programs
Safety share
activator effectiveness and, 182
to increase safe behavior, 114
for team effectiveness, 401
Safety steering team
creation for culture change, 446
management support need, 450
Safety thank-you cards use, 226–228
Safety Training and Observation
Program (STOP), 112
Safety Triad
described, 25, 109
evaluation process and, 421
human nature and, 55
person-based approach and, 327
role in Total Safety Culture, 479–480
use in systems approach, 43
Salience and habituation, see
Habituation
Satisfaction and group cohesiveness, 336
SBO (safe behavior opportunity), 148,
149
Seat belt use
Airline Lifesaver example, 188–190
buckle-up road signs example,
192–195
cost-benefit evaluation, 440
cost effectiveness, 148
DO IT process and, 146
Flash for Life example, 186–188
generalization study, 468–469
habituation and reminders example,
180–181
language choice and, 461
risk compensation and, 83
safe behavior promise example,
184–185
television's impact on, 122
television activator example, 191–192
Secondary appraisal of stressors, 97–98
Second hand recognition use, 282–283
See-Think-Act-Reward (STAR), 145
Selection by consequences, 117–118, 296
Selective listening, 274
Selective sensation, see Perception
Self-actualization, 302, 340
Self-confidence and leadership, 455
Self-directed behavior
commitment and consistency in
1540 subject index 11/11/00 1:30 PM Page 518
518
personal/interpersonal pressures,
377
public and voluntary commitment,
377–379
described, 166
quality recognition and, 284
requesting recognition, 285–286
role in Total Safety Culture, 478
self-motivation, 203, 210–211
self-reinforcement need, 209
Self-efficacy
emotional intelligence and, 346
empowerment and, 331–332
strategies for enhancing
assign achievable tasks, 358
communication process, 357–358
focus of feedback, 357
focus on positive, 360–361
steps in, 359–360
Self-esteem
actively caring behavior and, 329–331
coaching and, 256, 275
emotional intelligence and, 346
research support for actively caring
model, 342
self-talk influences on, 268
strategies for enhancing
communication, 355
corrective feedback, 355–357
supportive conversation and, 256, 275
Self-feedback, 146
Self-fulfilling prophecy, 333, 358
Self-help strategies for safety, 9–10
Self-mastery, 99
Self-motivation, see Self-directed behav-
ior
Self-serving bias, 104–105, 155, 489
Self-talk, 265, 268; see also Self-directed
behavior
Self-transcendence, 302, 339
Seminar, safety, 9
Sensation, see also Perception
described, 69
psychological dimension of, 57
The Seven Habits of Highly Effective
People (Covey), 21
Sexual behavior on television, 121–122
Signs, see Poster/sign campaigns and
safety
Skill discrepancy in employees, 13–14,
161
Psychology of safety handbook
Skinner Box, 117
Small wins concept, 186, 358, 362
SMART goals, 197–198, 408
Social conformity and safety
comparison feedback, 255
peer influence, 63–64
social support factors, 101–102, 238
Social loafing, 368–369, 411, 482–483
Social responsibility norm, 307
Social validity, 433–435
SOFTEN, 251
SOON (Specific, Observable, Objective,
and Naturalistic), 139
Speed, 140
STAR (See-Think-Act-Reward), 145
"Start small and build" strategy, 377–379
States, person, see Person states
Statistical test data
percentages use, 437–438
reliance on, 436
usefulness of, 435–436
Steering team, safety
creation for culture change, 446
management support need, 450
Stimulus generalization, 467
Stimulus overload theory, 309–310
Stimulus-response relationship, 116–117
STOP (Safety Training and Observation
Program), 112
Storming stage of teams, 410
Stressors, see Stress vs. distress
Stress-related headaches, 90
Stress vs. distress
attributional bias
example of, 102–103
fundamental error, 103–104
self-serving bias, 104–105
constructive and destructive aspects,
91–92
key points, 97–98
overview, 89–90
person factors contributing to, 92–93
stress definition, 90–91
stress management, 7
stressors coping mechanism
examples, 98
person factors, 99–100
physical fitness, 100–101
role in Total Safety Culture, 490–491
social factors, 101–102
stressors identification
1540 subject index 11/11/00 1:30 PM Page 519
Index
burnout symptoms, 94
domains of work distress, 94
types of stressors, 93
work-related factors, 94
work stress profile, 94–98
Supportive conversations, see also
Communication in coaching;
Safety coaching
approach selection
competence and commitment role,
289–290
individual situation considerations,
289
conversation improvement considera-
tions
bias, 274–275
flow direction, 270
nondirective approach use, 271
question asking technique, 271–273
self-image support, 256, 275
summary checklist, 275–276
transition from nondirective to di-
rective, 273
verbal commitment, 271
motivation and, 278
power of conversation, 266–269
quality recognition effects on, 282
receiving recognition
guidelines, 283–286
quality recognition and, 283
recognition of safety achievement
emphasis on mistakes avoidance,
278–279
higher-level praise association, 281
personal approach, 280–281
positive reinforcement and learn-
ing, 279
praise in private philosophy, 281
second hand recognition use,
282–283
stand alone recognition use,
281–282
tangibles use, 282
timing importance, 279–280
safety and, 265
safety celebrations guidelines,
286–289
for safety management, 276–278
shifting from past to future to present,
359
Supportive intervention, 167–168
519
Surveys
ethics of, 432
limitations of, 430
person state measurement, 338–339,
340, 428–430
reliability and validity
construct validity, 430, 432
correlation coefficient, 431
described, 430
validity, 433–435
validity types, 430–431
safety culture assessment, 338–339,
340, 344
for safety team selection, 374
team meeting evaluation, 406
team performance assessment, 405
training evaluation example, 449
System performance evaluation,
418–420
Systems approach to safety, 41–42
T
Talent, matching to jobs, 13–14, 161
Tangibles use
incentives/rewards programs, 282,
288
role in sustaining change, 451
Target audience involvement in inter-
ventions
Airline Lifesaver example, 188–190
Flash for Life example, 186–188
ownership-involvement principle,
183–184
safe behavior promise example,
184–185
Target behaviors
definition
behavioral outcomes, 136
defining behaviors, 136
describing behaviors, 137–138
DO IT process and, 134–136
interobserver reliability, 138
multiple behaviors, 138–139
person-action-situation, 137
safe behavior promise example, 185
Target Risk (Wilde), 84
TEAM (Together Everyone Achieves
More), 395
Teams, safety
developmental stages
1540 subject index 11/11/00 1:30 PM Page 520
520
adjourning, 412
forming, 409
norming, 410
performing, 411
storming, 410
failures of teams
gambles, 390–392
groupthink, 390
overcoming groupthink, 392
ground rules for meetings, 398
groups and belonging and, 367–371
paradigm shifts for, 41
described, 388–389
summary table, 413
phases of teamwork
action plan development, 399–400
assignment clarification, 394–395
charter establishment, 395
consensus building, 396–397
disband, restructure, or renew,
405–408
evaluate performance, 400–403
meeting procedures set, 400–403
member selection, 393–394
mission statement, 399
Television
observational learning and, 120–122
safety learning and, 122–123
use as an activator, 191–192
Thank-you cards use, 226–228
Theory-based safety
behavior- vs. person-based ap-
proaches
behavior-based, 28
cost effectiveness, 29
described, 26–27, 30
integration of, 29–30
person-based, 27–28
elements of approach, 21
mission statement development,
21–22, 24–26
relevance to occupational safety, 24
theory as a map, 22–24
Thinking skills and leadership, 455
Threat, 97
"Throwing a curve," 379
Timing and placement considerations in
interventions
buckle-up road signs example,
192–195
message effectiveness and, 190–191
Psychology of safety handbook
point-of-purchase placement, 191
television's impact, 191–192
Together Everyone Achieves More
(TEAM), 395
Total recordable injury rate (TRIR), 44
Total Safety Culture, see also Culture
change
actively caring role in, 353
aim of, 46–47
critical behaviors identification,
129–130
evaluation procedures, 418, 420
factors involved, 25–26
leadership's role in, 66
mission statement development,
21–22
new three Es
empowerment, 35
ergonomics, 35
evaluation, 36
paradigm shifts for
accident proneness mentality, 7, 33
achievement orientation need,
38–39
behavior focused approach, 39–40
bottom-up involvement, 40–41
continuous improvement, 46
corporate responsibility and, 37
fact finding focus, 42–45
interdependence, 41
old three Es, 33–34
paradigm definition, 36–37
proactive, 45
systems approach, 41–42
value based, 46
perception's relevance to, 75–76
principles summary
ABC model, 484
actively caring, 491, 492, 494
"all injuries are preventable" slogan,
488
bias and perception, 487–488
bystander apathy, 492–493
commitment and consistency,
495–496
culture change requirements, 478
DO IT process, 483–484
empowerment, 492
evaluation process, 496–497
example setting, 482
external consequences in modera-
1540 subject index 11/11/00 1:30 PM Page 521
Index
tion, 486
feedback importance, 483
focus on receptive people, 479
group-based behavior, 494
human nature and safety, 481
incentive program focus, 480
instructor impact, 478
internal change importance, 487
internally driven safety, 478
internal vs. external evaluation fac-
tors, 488–489
intervention impact, 484–485
leadership and, 478–479
learning new behaviors, 481
motivation basis, 486
negative consequence effects,
486–487
perceived risk, 488
personal control, 490
power of authority, 482
power of choice, 479
reciprocity principle, 494–495, 496
risk compensation, 488
safety coaching, 491–492
Safety Triad, 479–480
social loafing prevention, 482–483
stressors coping mechanism,
490–491
value based, 480–481
sense of shared responsibility, 313
tolerance for risk and, 84–85
value of working safely, 46–47
Trained seal act perception example, 71
Training, see also Education
assessing type needed, 160–161
behavior-based
feedback importance, 165
reasons to use, 162–163
techniques for, 164–165
training vs. education, 163–164,
165–166
definition, 136
521
flexibility in job performance, 15
influence on actively caring behavior,
313–314
myth of ability to do any job, 13
myth of learning from mistakes, 12
overcoming resistance to change, 463
Traits
of helpful people, 328–328
improvement evaluation and, 427–428
vs. states, 328
Trial and error learning, 279
TRIR (total recordable injury rate), 44
Trust and reciprocity, 379
"Tune-out" filter, 274, 275
Type A personalities, 91
U
U.S. Department of Transportation, 34
Unconditioned response (UCR), 116–117
Unconditioned stimulus (UCS), 116–117
Unconscious competence, 146, 147
Unconsciously incompetent, 145, 217
Unions and active caring, 370
Upside-down consequences, 158
V
Validity and reliability
construct validity, 430, 432
correlation coefficient, 431
described, 430
social validity, 433–435
validity types, 430–431
Values
person factors and, 112
priorities vs., 461
role in Total Safety Culture, 480–481
Verbal commitment, 271
Vicarious consequences, 119–120
Violence on television, 120–121
1540 subject index 11/11/00 1:30 PM Page 522

522 Psychology of safety handbook

Vision communication, 270 Y

Yerkes-Dodson Law, 92
W
Walden Two (Skinner), 17 Z
Warning beepers and habituation,
181–182 Zero injuries safety goal, 197, 198
Welch, Martell, 304
Within subjects design, 83
Word, Deletha, 304
Work distress profile, 94–98
Work family concept, 370
Workplace myths, 12–16
Work underload, 94