You are on page 1of 1028



Intergroup Relations
Editorial Board
John M. Levine Michael A. Hogg
University of Pittsburgh Claremont Graduate University

Managing Editor
Danielle L. Blaylock
Claremont Graduate University

Associate Editors
Linda Argote Norbert L. Kerr
Carnegie Mellon University Michigan State

Marilynn B. Brewer Richard L. Moreland

Ohio State University University of Pittsburgh

John F. Dovidio Cecilia L. Ridgeway

Yale University Stanford University
1& 2


roup Processes
Intergroup Relations


University of Pittsburgh Claremont Graduate University
Copyright © 2010 by SAGE Publications, Inc.

All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical,
including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the
For information:
SAGE Publications, Inc.
2455 Teller Road
Thousand Oaks, California 91320
SAGE Publications Ltd.
1 Oliver’s Yard
55 City Road
London EC1Y 1SP
United Kingdom
SAGE Publications India Pvt. Ltd.
B 1/I 1 Mohan Cooperative Industrial Area
Mathura Road, New Delhi 110 044

SAGE Publications Asia-Pacific Pte. Ltd.

33 Pekin Street #02-01
Far East Square
Singapore 048763
Printed in the United States of America

Library of Congress Cataloging-in-Publication Data

Encyclopedia of group processes and intergroup relations/John M. Levine, Michael A. Hogg, editors.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-4129-4208-9 (cloth)
1. Social groups—Encyclopedias. 2. Intergroup relations—Encyclopedias. I. Levine, John M. II. Hogg, Michael A.,

HM716.E53 2010
302.303—dc22 2009026419

This book is printed on acid-free paper.

09   10   11   12   13   10   9   8   7   6   5   4   3   2   1

Publisher: Rolf A. Janke

Acquisitions Editor: Michael Carmichael
Editorial Assistant: Michele Thompson
Developmental Editors: Diana E. Axelsen, Carole A. Maurer
Reference Systems Manager: Leticia Gutierrez
Reference Systems Coordinator: Laura Notton
Production Editor: Kate Schroeder
Copy Editors: Bonnie Freeman, Jamie Robinson
Typesetter: C&M Digitals (P) Ltd.
Proofreaders: Kris Bergstad, Sandy Zilka Livingston, Penny Sippel
Indexer: Joan Shapiro
Cover Designer: Candice Harman
Marketing Manager: Amberlyn McKay

Volume 1
List of Entries   vii
Reader’s Guide   xi
About the Editors   xxi
Contributors   xxiii
Introduction   xxxiii

A 1 F 269
B 53 G 293
C 67 H 395
D 185 I 411
E 235 J 489

Volume 2
List of Entries   vii

K 507 S 723
L 511 T 897
M 549 U 941
N 591 V 947
O 613 W 953
P 635 X 959
R 675
Index   963
List of Entries

Action Research Collectivism/Individualism

Affect Control Theory Common-Identity/Common-Bond Groups
Affirmative Action Common Ingroup Identity Model
Ageism Common Knowledge Effect
Allport, Gordon Commons Dilemma
Ambivalent Sexism Communication Networks
Anticonformity Compliance
Anti-Semitism Computer-Mediated Communication
Apartheid Computer Simulation
Asch, Solomon Conformity
Assimilation and Acculturation Conservatism
Attachment Theory Conspiracy Theories
Attitudes Toward Women Scale Contingency Theories of Leadership
Attribution Biases Cooperation and Competition
Authoritarian Personality Cooperative Learning
Aversive Racism Cross-Categorization
Banality of Evil Crowds
Black Sheep Effect Cults
Boundary Spanning Culture
Brainstorming Culture of Honor
Bystander Effect
Categorization Dehumanization/Infrahumanization
Charismatic Leadership Deindividuation
Children: Stereotypes and Prejudice Delphi Technique
Children’s Groups Depersonalization
Civil Rights Legislation Desegregation
Civil Rights Movement Deutsch, Morton
Cliques Deviance
Coalitions Discrimination
Cognitive Consistency Distributive Justice
Collaboration Technology Diversity
Collective Guilt Dogmatism
Collective Induction Dominance Hierarchies
Collective Movements and Protest Dyads
Collective Self Dynamical Systems Approach

viii List of Entries

Emergent Norm Theory Hate Crimes

Entitativity Hidden Profile Task
Escalation of Commitment Holocaust
Essentialism Homophily
Ethnicity Homophobia
Ethnolinguistic Vitality Identification and Commitment
Eugenics Identity Control Theory
Evolutionary Psychology Ideology
Experimentation Idiosyncrasy Credit
Extended Contact Effect Illusion of Group Effectivity
Illusory Correlation
Fads and Fashions Immigration
False Consensus Effect Implicit Association Test (IAT)
Families Implicit Prejudice
Faultlines Inclusion/Exclusion
Feminism Informational Influence
Festinger, Leon Ingroup Allocation Bias
Free Riding Initiation Rites
Frustration-Aggression Hypothesis Innovation
Institutionalized Bias
Gangs Interactionist Theories of Leadership
Gender and Behavior Interaction Process Analysis
Gender Roles Interdependence Theory
Genocide Intergroup Anxiety
Graduated Reciprocation in Tension Reduction Intergroup Contact Theory
(GRIT) Intergroup Emotions Theory
Great Person Theory of Leadership Intergroup Empathy
Group Boundaries Intergroup Violence
Group Cohesiveness Interindividual–Intergroup Discontinuity
Group Composition Islamophobia
Group Development
Group Dissolution J-Curve Hypothesis
Group Ecology Jigsaw Classroom Technique
Group Emotions Job Design
Group Formation Juries
Group Learning Justice
Group Memory Just World Hypothesis
Group Mind
Group Motivation Köhler Effect
Group Performance
Group Polarization Language and Intergroup Relations
Group Position Theory Leader Categorization Theory
Group Potency Leader-Member Exchange (LMX) Theory
Group Problem Solving and Decision Making Leadership
Group Socialization Legitimation
Group Structure Leniency Contract
Group Task Levels of Analysis
Groupthink Lewin, Kurt
List of Entries ix

Linguistic Category Model (LCM) Realistic Group Conflict Theory

Linguistic Intergroup Bias (LIB) Reference Groups
Looking-Glass Self Referent Informational Influence Theory
Loyalty Relational Cohesion Theory
Relational Model of Authority in Groups
Mediation Relative Deprivation
Mergers Research Methods and Issues
Minimal Group Effect Reverse Discrimination
Minority Coping Strategies Right Wing Authoritarianism
Minority Groups in Society Ringelmann Effect
Minority Influence Roles
Modern Forms of Prejudice Role Transitions
Modern Racism Romance of Leadership
Modern Sexism Rumor
Moscovici, Serge
Multiculturalism Scapegoating
Multiple Identities Schisms
Mutual Intergroup Differentiation Model Self-Categorization Theory
Nationalism and Patriotism Self-Fulfilling Prophecy
Need for Belonging Self-Managing Teams
Need for Closure Self-Stereotyping
Need for Power Sensitivity Training Groups
Negotiation and Bargaining Sexism
Normative Influence Sexual Harassment
Norms Shared Mental Models
Obedience to Authority Sherif, Muzafer
Opinion Deviance Slavery
Optimal Distinctiveness Social Class
Organizations Social Comparison Theory
Ostracism Social Compensation
Outgroup Homogeneity Effect Social Darwinism
Social Decision Schemes
Path–Goal Theory of Leadership Social Deviance
Perceived Group Variability Social Dilemmas
Personality Theories of Leadership Social Dominance Theory
Personnel Turnover Social Entrainment
Pluralistic Ignorance Social Exchange in Networks and Groups
Power Social Facilitation
Power–Dependence Theory Social Identity Model of Deindividuation Effects
Prejudice Social Identity Theory
Prisoner’s Dilemma Social Identity Theory of Leadership
Procedural Justice Social Impact Theory
Process Consultation Social Loafing
Process Gain and Loss Socially Shared Cognition
Protestant Work Ethic Social Mobility
Social Networks
Racial Ambivalence Theory Social Relations Model
Racism Social Representations
x List of Entries

Socioemotional and Task Behavior Team Negotiation

Sociometer Model Team Performance Assessment
Sociometric Choice Team Reflexivity
Sports Teams Teams
Stanford Prison Experiment Territoriality
Status Terrorism
Status Characteristics/Expectation States Theory Terror Management Theory
Status Construction Theory Therapy Groups
Stepladder Technique Tokenism
Stereotype Threat Transactional Leadership Theories
Stereotyping Transactive Memory Systems
Stigma Transformational Leadership Theories
Subjective Group Dynamics Trust
Sucker Effect Ultimate Attribution Error
Support Groups Uncertainty-Identity Theory
Survey Methods
Symbolic Interactionism Vertical Dyad Linkage Model
Symbolic Racism Virtual/Internet Groups
System Justification Theory Weightism
System Theory Work Teams

Tajfel, Henri Xenophobia

Team Building
Reader’s Guide

Cognitions and Feelings Group Polarization

Affect Control Theory Group Position Theory
Ageism Group Potency
Ambivalent Sexism Group Problem Solving and Decision Making
Anti-Semitism Groupthink
Attachment Theory Hidden Profile Task
Attitudes Toward Women Scale Homophobia
Attribution Biases Identity Control Theory
Aversive Racism Ideology
Brainstorming Illusion of Group Effectivity
Categorization Illusory Correlation
Children: Stereotypes and Prejudice Implicit Association Test (IAT)
Cognitive Consistency Implicit Prejudice
Collective Guilt Informational Influence
Collective Induction Intergroup Anxiety
Collective Self Intergroup Emotions Theory
Collectivism/Individualism Intergroup Empathy
Common Knowledge Effect Islamophobia
Conservatism Justice
Conspiracy Theories Language and Intergroup Relations
Cross-Categorization Leader Categorization Theory
Culture Linguistic Category Model (LCM)
Culture of Honor Linguistic Intergroup Bias (LIB)
Decategorization Modern Forms of Prejudice
Dehumanization/Infrahumanization Modern Racism
Deindividuation Modern Sexism
Depersonalization Multiple Identities
Distributive Justice Need for Closure
Dogmatism Outgroup Homogeneity Effect
Entitativity Perceived Group Variability
Essentialism Pluralistic Ignorance
Ethnocentrism Prejudice
False Consensus Effect Procedural Justice
Group Cohesiveness Protestant Work Ethic
Group Emotions Racial Ambivalence Theory
Group Learning Racism
Group Memory Relational Cohesion Theory
Group Mind Rumor

xii Reader’s Guide

Self-Categorization Theory Group Structure

Self-Esteem Inclusion/Exclusion
Self-Fulfilling Prophecy Ingroup Allocation Bias
Self-Stereotyping Interdependence Theory
Sexism Interindividual–Intergroup Discontinuity
Shared Mental Models Jigsaw Classroom Technique
Social Comparison Theory Justice
Social Decision Schemes Mediation
Social Dominance Theory Minimal Group Effect
Socially Shared Cognition Minority Influence
Social Representations Moscovici, Serge
Sociometer Model Need for Power
Stereotype Threat Negotiation and Bargaining
Stereotyping Norms
Subtyping Opinion Deviance
Symbolic Racism Power
System Justification Theory Power–Dependence Theory
Team Reflexivity Prisoner’s Dilemma
Transactive Memory Systems Procedural Justice
Ultimate Attribution Error Relational Cohesion Theory
Weightism Relational Model of Authority in Groups
Xenophobia Relative Deprivation
Sensitivity Training Groups
Conflict and Cooperation Within Groups Sexual Harassment
Anticonformity Sherif, Muzafer
Asch, Solomon Social Decision Schemes
Black Sheep Effect Social Dilemmas
Cliques Sports Teams
Coalitions Stanford Prison Experiment
Collaboration Technology Subjective Group Dynamics
Commons Dilemma Tajfel, Henri
Conformity Team Negotiation
Cooperation and Competition Therapy Groups
Cooperative Learning Trust
Deutsch, Morton Virtual/Internet Groups
Deviance Work Teams
Distributive Justice
Dominance Hierarchies
Emergent Norm Theory Group Decision Making
Escalation of Commitment Anticonformity
Evolutionary Psychology Asch, Solomon
Faultlines Brainstorming
Frustration-Aggression Hypothesis Cliques
Graduated Reciprocation in Tension Reduction Coalitions
(GRIT) Collective Induction
Group Cohesiveness Common Knowledge Effect
Group Emotions Computer Simulation
Group Problem Solving and Decision Making Conformity
Reader’s Guide xiii

Delphi Technique Group Performance and Problem Solving

Dominance Hierarchies Boundary Spanning
Escalation of Commitment Brainstorming
Graduated Reciprocation in Tension Reduction Charismatic Leadership
(GRIT) Collaboration Technology
Group Cohesiveness Communication Networks
Group Composition Computer-Mediated Communication
Group Emotions Contingency Theories of Leadership
Group Memory Cooperative Learning
Group Mind Deindividuation
Group Motivation Delphi Technique
Group Performance Distributive Justice
Group Polarization Diversity
Group Problem Solving and Decision Making Dominance Hierarchies
Group Structure Dynamical Systems Approach
Group Task Emergent Norm Theory
Groupthink Escalation of Commitment
Hidden Profile Task Faultlines
Idiosyncrasy Credit Free Riding
Inclusion/Exclusion Gender and Behavior
Informational Influence Gender Roles
Juries Great Person Theory of Leadership
Leniency Contract Group Boundaries
Mediation Group Cohesiveness
Minority Influence Group Composition
Moscovici, Serge Group Development
Need for Closure Group Dissolution
Negotiation and Bargaining Group Ecology
Normative Influence Group Emotions
Norms Group Formation
Obedience to Authority Group Learning
Opinion Deviance Group Mind
Power Group Motivation
Power–Dependence Theory Group Potency
Reference Groups Group Problem Solving and Decision Making
Referent Informational Influence Theory Group Socialization
Relational Cohesion Theory Group Structure
Relational Model of Authority in Groups Group Task
Shared Mental Models Groupthink
Sherif, Muzafer Hidden Profile Task
Social Decision Schemes Identification and Commitment
Social Networks Illusion of Group Effectivity
Status Inclusion/Exclusion
Status Characteristics/Expectations Initiation Rites
States Theory Innovation
Status Construction Theory Interactionist Theories of Leadership
Team Negotiation Interaction Process Analysis
Trust Job Design
Work Teams Justice
xiv Reader’s Guide

Köhler Effect Transformational Leadership Theories

Leader Categorization Theory Trust
Leader-Member Exchange (LMX) Theory Vertical Dyad Linkage Model
Leadership Virtual/Internet Groups
Loyalty Work Teams
Negotiation and Bargaining
Norms Group Structure
Obedience to Authority Affirmative Action
Organizations Apartheid
Path–Goal Theory of Leadership Assimilation and Acculturation
Personality Theories of Leadership Attachment Theory
Personnel Turnover Boundary Spanning
Power Cliques
Power–Dependence Theory Coalitions
Procedural Justice Common-Identity/Common-Bond Groups
Process Consultation Communication Networks
Process Gain and Loss Computer-Mediated Communication
Relational Cohesion Theory Crowding
Relational Model of Authority in Groups Crowds
Ringelmann Effect Deviance
Roles Diversity
Role Transitions Dominance Hierarchies
Romance of Leadership Dynamical Systems Approach
Self-Managing Teams Emergent Norm Theory
Shared Mental Models Faultlines
Social Compensation Gender and Behavior
Social Entrainment Gender Roles
Social Exchange in Networks and Groups Group Boundaries
Social Facilitation Group Cohesiveness
Social Identity Model of Deindividuation Effects Group Composition
Social Identity Theory Group Development
Social Identity Theory of Leadership Group Dissolution
Social Impact Theory Group Formation
Social Loafing Group Position Theory
Socially Shared Cognition Group Socialization
Socioemotional and Task Behavior Group Structure
Sports Teams Homophily
Status Inclusion/Exclusion
Status Characteristics/Expectations States Theory Initiation Rites
Status Construction Theory Job Design
Stepladder Technique Legitimation
Sucker Effect Mergers
Team Building Norms
Team Negotiation Opinion Deviance
Team Performance Assessment Organizations
Team Reflexivity Ostracism
Teams Personnel Turnover
Transactional Leadership Theories Relational Cohesion Theory
Transactive Memory Systems Roles
Reader’s Guide xv

Role Transitions Minority Groups in Society

Schisms Multiple Identities
Shared Mental Models Mutual Intergroup Differentiation Model
Slavery Nationalism and Patriotism
Social Class Need for Belonging
Social Mobility Optimal Distinctiveness
Social Networks Perceived Group Variability
Social Relations Model Reference Groups
Sociometric Choice Referent Informational Influence Theory
Sports Teams Self-Categorization Theory
Status Self-Esteem
Status Characteristics/Expectation States Theory Self-Stereotyping
Status Construction Theory Social Identity Model of Deindividuation Effects
Stepladder Technique Social Identity Theory of Leadership
SYMLOG Sociometer Model
System Theory Subjective Group Dynamics
Team Building Symbolic Interactionism
Territoriality Tajfel, Henri
Tokenism Uncertainty-Identity Theory
Virtual/Internet Groups
Work Teams
Influence and Persuasion
Identity and Self Asch, Solomon
Assimilation and Acculturation Bystander Effect
Black Sheep Effect Charismatic Leadership
Categorization Cognitive Consistency
Collective Guilt Collective Induction
Collective Movements and Protest Collective Movements and Protest
Collective Self Common Knowledge Effect
Collectivism/Individualism Compliance
Common-Identity/Common-Bond Groups Conformity
Common Ingroup Identity Model Contingency Theories of Leadership
Cross-Categorization Culture
Deindividuation Deviance
Depersonalization Dominance Hierarchies
Ethnicity Dynamical Systems Approach
Ethnolinguistic Vitality Emergent Norm Theory
Extended Contact Effect Fads and Fashions
Gender and Behavior False Consensus Effect
Gender Roles Festinger, Leon
Group Position Theory Gender and Behavior
Identification and Commitment Great Person Theory of Leadership
Identity Control Theory Group Cohesiveness
Ingroup Allocation Bias Group Mind
Interindividual–Intergroup Discontinuity Group Polarization
Looking-Glass Self Group Problem Solving and Decision Making
Loyalty Groupthink
Minimal Group Effect Hidden Profile Task
Minority Coping Strategies Identification and Commitment
xvi Reader’s Guide

Idiosyncrasy Credit Transformational Leadership Theories

Inclusion/Exclusion Trust
Informational Influence Vertical Dyad Linkage Model
Interactionist Theories of Leadership
Leader Categorization Theory Intergroup Relations in Society
Leader-Member Exchange (LMX) Theory Affirmative Action
Leadership Ageism
Leniency Contract Allport, Gordon
Lewin, Kurt Ambivalent Sexism
Loyalty Anti-Semitism
Mediation Apartheid
Minority Influence Assimilation and Acculturation
Moscovici, Serge Attitudes Toward Women Scale
Need for Closure Authoritarian Personality
Need for Power Aversive Racism
Negotiation and Bargaining Banality of Evil
Normative Influence Black Sheep Effect
Norms Children: Stereotypes and Prejudice
Obedience to Authority Civil Rights Legislation
Opinion Deviance Civil Rights Movement
Ostracism Collective Guilt
Path–Goal Theory of Leadership Collective Movements and Protest
Personality Theories of Leadership Common Ingroup Identity Model
Pluralistic Ignorance Conspiracy Theories
Power Cooperative Learning
Reference Groups Cults
Referent Informational Influence Theory Desegregation
Relational Cohesion Theory Deviance
Ringelmann Effect Discrimination
Romance of Leadership Distributive Justice
Rumor Diversity
Sherif, Muzafer Ethnicity
Social Compensation Ethnocentrism
Social Decision Schemes Ethnolinguistic Vitality
Social Exchange in Networks and Groups Eugenics
Social Facilitation Evolutionary Psychology
Social Identity Model of Deindividuation Effects Extended Contact Effect
Social Identity Theory Feminism
Social Identity Theory of Leadership Frustration-Aggression Hypothesis
Social Impact Theory Gangs
Social Loafing Genocide
Status Graduated Reciprocation in Tension Reduction
Status Characteristics/Expectation States Theory (GRIT)
Status Construction Theory Group Emotions
SYMLOG Group Position Theory
Team Negotiation Hate Crimes
Terrorism Holocaust
Therapy Groups Homophobia
Transactional Leadership Theories Ideology
Reader’s Guide xvii

Immigration Symbolic Racism

Ingroup Allocation Bias System Justification Theory
Institutionalized Bias Tajfel, Henri
Intergroup Anxiety Territoriality
Intergroup Contact Theory Terrorism
Intergroup Emotions Theory Terror Management Theory
Intergroup Empathy Tokenism
Intergroup Violence Weightism
Islamophobia Xenophobia
J-Curve Hypothesis
Jigsaw Classroom Technique
Justice Methodology
Just World Hypothesis Action Research
Language and Intergroup Relations Ambivalent Sexism
Linguistic Category Model (LCM) Attitudes Toward Women Scale
Linguistic Intergroup Bias (LIB) Authoritarian Personality
Mergers Computer Simulation
Minimal Group Effect Experimentation
Minority Coping Strategies Implicit Association Test (IAT)
Minority Groups in Society Interaction Process Analysis
Modern Forms of Prejudice Levels of Analysis
Modern Racism Modern Racism
Modern Sexism Modern Sexism
Multiculturalism Need for Closure
Mutual Intergroup Differentiation Model Need for Power
Nationalism and Patriotism Research Methods and Issues
Prejudice Right Wing Authoritarianism
Procedural Justice Social Relations Model
Protestant Work Ethic Sociometric Choice
Racial Ambivalence Theory Survey Methods
Realistic Group Conflict Theory
Relative Deprivation
Right Wing Authoritarianism Organizations
Scapegoating Affirmative Action
Schisms Assimilation and Acculturation
Self-Categorization Theory Boundary Spanning
Self-Fulfilling Prophecy Charismatic Leadership
Sexism Cliques
Sexual Harassment Coalitions
Sherif, Muzafer Collaboration Technology
Slavery Collective Self
Social Class Communication Networks
Social Darwinism Computer-Mediated Communication
Social Deviance Contingency Theories of Leadership
Social Dominance Theory Cooperation and Competition
Social Identity Theory Culture
Stereotype Threat Deviance
Stigma Distributive Justice
Subjective Group Dynamics Diversity
xviii Reader’s Guide

Dominance Hierarchies Schisms

Dynamical Systems Approach Self-Managing Teams
Escalation of Commitment Sexual Harassment
Faultlines Social Compensation
Free Riding Social Exchange in Networks and Groups
Gender and Behavior Social Facilitation
Gender Roles Social Impact Theory
Great Person Theory of Leadership Social Loafing
Group Boundaries Socially Shared Cognition
Group Cohesiveness Social Networks
Group Composition Socioemotional and Task Behavior
Group Ecology Status
Group Learning Status Characteristics/Expectation States Theory
Group Motivation Status Construction Theory
Group Performance Sucker Effect
Group Socialization System Theory
Group Structure Team Building
Group Task Team Negotiation
Homophily Team Performance Assessment
Identification and Commitment Team Reflexivity
Ideology Teams
Initiation Rites Tokenism
Innovation Transactional Leadership Theories
Interactionist Theories of Leadership Transformational Leadership Theories
Job Design Trust
Justice Vertical Dyad Linkage Model
Leader Categorization Theory Virtual/Internet Groups
Leader-Member Exchange (LMX) Theory Work Teams
Mergers Theory
Multiple Identities Affect Control Theory
Negotiation and Bargaining Allport, Gordon
Norms Ambivalent Sexism
Obedience to Authority Asch, Solomon
Organizations Attachment Theory
Path–Goal Theory of Leadership Authoritarian Personality
Personality Theories of Leadership Aversive Racism
Personnel Turnover Charismatic Leadership
Power Cognitive Consistency
Power–Dependence Theory Contingency Theories of Leadership
Procedural Justice Deutsch, Morton
Process Consultation Dynamical Systems Approach
Process Gain and Loss Emergent Norm Theory
Protestant Work Ethic Eugenics
Relational Cohesion Theory Evolutionary Psychology
Relational Model of Authority in Groups Festinger, Leon
Roles Frustration-Aggression Hypothesis
Role Transitions Great Person Theory of Leadership
Romance of Leadership Group Position Theory
Reader’s Guide xix

Identity Control Theory System Justification Theory

Interactionist Theories of Leadership System Theory
Interdependence Theory Tajfel, Henri
Intergroup Contact Theory Terror Management Theory
Intergroup Emotions Theory Transactional Leadership Theories
Justice Transformational Leadership Theories
Just World Hypothesis Uncertainty-Identity Theory
Leader Categorization Theory Vertical Dyad Linkage Model
Leader-Member Exchange (LMX) Theory
Lewin, Kurt
Modern Forms of Prejudice Types of Groups and Subgroups
Modern Racism Children’s Groups
Modern Sexism Cliques
Moscovici, Serge Coalitions
Mutual Intergroup Differentiation Model Collective Movements and Protest
Need for Belonging Common-Identity/Common-Bond Groups
Need for Closure Communication Networks
Need for Power Computer-Mediated Communication
Optimal Distinctiveness Cooperative Learning
Path–Goal Theory of Leadership Crowds
Personality Theories of Leadership Cults
Power–Dependence Theory Dominance Hierarchies
Racial Ambivalence Theory Dyads
Realistic Group Conflict Theory Ethnicity
Referent Informational Influence Theory Families
Relational Cohesion Theory Gangs
Relational Model of Authority in Groups Jigsaw Classroom Technique
Right Wing Authoritarianism Juries
Self-Categorization Theory Minority Groups in Society
Sherif, Muzafer Organizations
Social Decision Schemes Reference Groups
Social Dominance Theory Schisms
Social Identity Model of Deindividuation Effects Self-Managing Teams
Social Identity Theory Sensitivity Training Groups
Social Identity Theory of Leadership Sports Teams
Social Impact Theory Stepladder Technique
Sociometer Model Support Groups
Status Characteristics/Expectation States Teams
Theory Therapy Groups
Status Construction Theory Tokenism
Subjective Group Dynamics Virtual/Internet Groups
Symbolic Interactionism Work Teams
About the Editors

John M. Levine is a professor of psychology and Kingdom and at the University of Queensland in
senior scientist in the Learning Research and Australia. He is a fellow of the Society for
Development Center at the University of Pittsburgh Personality and Social Psychology, the Society of
and Honorary Professor of Social Psychology at Experimental Social Psychology, the Society for
the University of Kent in the United Kingdom. He the Psychological Study of Social Issues, the
did his undergraduate work at Northwestern Western Psychological Association, and the
University and received his PhD in psychology Academy of the Social Sciences in Australia. He
from the University of Wisconsin. Levine’s research received his PhD from Bristol University and has
focuses on small group processes, including inno- taught at Bristol University, Macquarie University,
vation in work teams, group reaction to deviance the University of Melbourne, the University of
and disloyalty, majority and minority influence, Queensland, and Princeton University, and he
and group socialization. He has published widely has been a visiting professor at the University of
on these and related topics. Levine has served as California in Los Angeles, Santa Cruz, and Santa
associate editor of the Journal of Research in Barbara and at City University, Hong Kong. His
Personality, both associate editor and editor of research on group processes, intergroup rela-
the Journal of Experimental Social Psychology, tions, social identity, and self-conception is
and executive committee chair of the Society of closely associated with the development of social
Experimental Social Psychology. He is a fellow of identity theory. He has published extensively on
the Society of Experimental Social Psychology, the these and related topics. A former associate edi-
Society for Personality and Social Psychology, tor of the Journal of Experimental Social
the American Psychological Society, and the Psychology, he is foundation coeditor with
Midwestern Psychological Association. Dominic Abrams of the journal Group Processes
and Intergroup Relations and senior consultant
Michael A. Hogg is professor of social psychol- editor for the Sage Social Psychology Program.
ogy at Claremont Graduate University in Los Current research foci include leadership, devi-
Angeles and an honorary professor of social psy- ance, uncertainty reduction, extremism, and sub-
chology at the University of Kent in the United group relations.


Susanne Abele Bruce J. Avolio Alison J. Bianchi

Miami University University of Nebraska University of Iowa

Christopher L. Aberson Mahzarin R. Banaji Amy Blackstone

Humboldt State University Harvard University University of Maine

Dominic Abrams Andrew Baron Steven L. Blader

University of Kent at Harvard University New York University
Robert S. Baron Robert D. Blagg
Laura Aikens University of Iowa Claremont Graduate University
University of Georgia
Laura Barron Irene V. Blair
Kira M. Alexander Rice University University of Colorado
University of Pittsburgh
Brock Bastian Danielle L. Blaylock
Catherine E. Amiot University of Melbourne Claremont Graduate
Université du Québec à University
Montréal Andrew S. Baum
University of Texas at Michelle C. Bligh
Grace L. Anderson Arlington Claremont Graduate
University of California, University
Santa Barbara Julia C. Becker
University of Marburg, Ana-Maria Bliuc
Linda Argote Germany University of Sydney
Carnegie Mellon University
Lane Beckes Brittany Bloodhart
Arthur Aron University of Minnesota Pennsylvania State University
State University of New York
at Stony Brook Van Beck Hall Renata Bongiorno
University of Pittsburgh Murdoch University
Holly Arrow
University of Oregon / London Bradford S. Bell Joseph A. Bonito
Business School Cornell University University of Arizona

Blake E. Ashforth Jennifer L. Berdahl Martin J. Bourgeois

Arizona State University University of Toronto Florida Gulf Coast University

xxiv Contributors

Richard Y. Bourhis Adrienne R. Carter-Sowell Karen M. Douglas

Université du Québec à Purdue University University of Kent
Bettina J. Casad John F. Dovidio
Katie M. Bowen California State Polytechnic Yale University
University of Chicago University, Pomona
John Duckitt
Clint Bowers Jennifer A. Chatman University of Auckland
University of Central Florida University of California,
Berkeley Jennifer L. Dunn
Nyla R. Branscombe Southern Illinois University
University of Kansas Jacqueline M. Chen Carbondale
University of California, Santa
Marilynn B. Brewer Barbara Alice H. Eagly
Ohio State University Northwestern University
Margaret S. Clark
Celia A. Brownell Yale University Shaha El-Geledi
University of Pittsburgh Université du Québec à
Alain Clémence Montréal
Camille Buckner University of Lausanne
Marymount University Dina Eliezer
J. Christopher Cohrs University of California, Santa
Jerry M. Burger University of Jena Barbara
Santa Clara University
Brian Colwell Naomi Ellemers
Peter Burke University of Missouri Leiden University
University of California,
Riverside D’Lane R. Compton Nicholas Emler
University of New Orleans University of Surrey
Shauna Burke
University of Western Ontario Joel Cooper Ralph Erber
Princeton University DePaul University
Gary M. Burlingame
Brigham Young University Christian S. Crandall Victoria Esses
University of Kansas University of Western
David F. Caldwell Ontario
Santa Clara University William D. Crano
Claremont Graduate Mark A. Ferguson
Jan Cannon-Bowers University University of Kansas
University of Central Florida
Richard J. Crisp Thomas A. Finholt
Allison Cantwell University of Kent University of Michigan
University of California,
Riverside John Darley Gregory W. Fischer
Princeton University Duke University
Peter J. Carnevale
University of Southern Carsten K. W. De Dreu Donelson R. Forsyth
California University of Amsterdam University of Richmond

Albert Carron C. Nathan DeWall Susan Fussell

University of Western Ontario University of Kentucky Cornell University
Contributors xxv

Samuel L. Gaertner Craig Haney Crystal L. Hoyt

University of Delaware University of California, Santa University of Richmond
Amber M. Gaffney Jeffrey R. Huntsinger
Claremont Graduate Uriel J. Haran Loyola University Chicago
University Tepper School of Business
Chester A. Insko
Donna Garcia Chad Hartnell University of North Carolina
University of Western Ontario Arizona State University at Chapel Hill

Alexandra Gerbasi Simon Pierre Harvey Elizabeth Jacobs

California State University, Université du Québec à Loyola University Chicago
Northridge Montréal
Cathryn Johnson
Daniel Gigone Nick Haslam Emory University
Montana State University University of Melbourne
John T. Jost
Howard Giles S. Alexander Haslam New York University
University of California, Santa University of Exeter
Barbara Charles M. Judd
Michelle Hebl University of Colorado Boulder
Francesca Gino Rice University
University of North Carolina Lee Jussim
P. J. Henry Rutgers University
Peter Glick DePaul University
Lawrence University Tatsuya Kameda
John P. Hewitt Hokkaido University
George R. Goethals University of Massachusetts at
University of Richmond Amherst Martin F. Kaplan
Northern Illinois University
Kenneth Goh Miles Hewstone
Carnegie Mellon University University of Oxford Steven J. Karau
Southern Illinois University
Paul S. Goodman Michael A. Hogg
Carnegie Mellon University Claremont Graduate University Alian S. Kasabian
California State Polytechnic
Stephanie A. Goodwin Zachary P. Hohman University, Pomona
Purdue University Claremont Graduate University
Yoshihisa Kashima
Jeff Greenberg Andrea B. Hollingshead University of Melbourne
University of Arizona University of Southern
California Janice R. Kelly
Martin S. Greenberg Purdue University
University of Pittsburgh John G. Holmes
University of Waterloo Jared B. Kenworthy
Ana Guinote University of Texas at
University College London Ann E. Hoover Arlington
Purdue University
David L. Hamilton Norbert L. Kerr
University of California, Santa Matthew Hornsey Michigan State University,
Barbara University of Queensland University of Kent
xxvi Contributors

Nicolas Kervyn Edward J. Lawler Stephen Loughnan

Catholic University of Louvain Cornell University University of Melbourne

Jaeshin Kim Andrea Lawson Robert B. Lount , Jr.

University of Massachusetts, University of Western Ohio State University
Amherst Ontario
Jeffrey W. Lucas
William Klein Colin Wayne Leach University of Maryland
University of Pittsburgh University of Connecticut,
University of Sussex Anne Maass
Richard Klimoski University of Padova
George Mason University Naomi Lee
Georgetown University Diane M. Mackie
Jeffrey C. Kohles University of California, Santa
California State University, San Jürgen Leibold Barbara
Marcos Georg-August University
Göttingen Namrata Mahajan
Steve W. J. Kozlowski Claremont Graduate University
Michigan State University Edward P. Lemay
University of New Hampshire Angela T. Maitner
David Krackhardt
University of Kent
Carnegie Mellon University
Shana Levin
Claremont McKenna College Brenda Major
Roderick M. Kramer
University of California, Santa
Stanford University
John M. Levine Barbara
Arie W. Kruglanski University of Pittsburgh
University of Maryland, Elizabeth Mannix
College Park George Levinger Cornell University
University of Massachusetts,
Gillian Ku Amherst John Markoff
London Business School University of Pittsburgh
Jacques-Philippe Leyens
Manwai C. Ku Université of Louvain at Mitchell Lee Marks
Stanford University Louvain-La-Neuve San Francisco State University

Matthew B. Kugler E. Allan Lind José M. Marques

Princeton University Duke University University of Porto

Kathy J. Kuipers Patricia W. Linville Robin Martin

University of Montana Duke University Aston University

Jonathon LaPaglia Glenn E. Littlepage Miriam Matthews

University of Minnesota Middle Tennessee State Claremont Graduate
University University
Dora C. Lau
Chinese University of Hong Zayra N. Longoria Allan Mazur
Kong Purdue University Syracuse University

Patrick R. Laughlin Fabio Lorenzi-Cioldi Debra Theobald McClendon

University of Illinois University of Geneva Brigham Young University
Contributors xxvii

Craig McGarty Ian Newby-Clark Michale J. Platow

Murdoch University University of Guelph Australian National University

Jamie G. McMinn Leonard S. Newman Tom Postmes

Westminster College Syracuse University University of Groningen

Marisa Mealy Bernard Nijstad Anthony R. Pratkanis

Central Connecticut State University of Amsterdam University of California
Julia D. O’Brien Felicia Pratto
Avital Mentovich University of Maryland University of Connecticut
New York University
Katharine Ridgway O’Brien Radmila Prislin
Charles E. Miller Rice University San Diego State University
Northern Illinois University
Shigehiro Oishi Marina Rachitskiy
Norman Miller University of Virginia University of Surrey
University of Southern
California Greg R. Oldham Lisa Slattery Rashotte
University of Illinois at University of North Carolina,
Ella Miron-Spektor Urbana-Champaign Charlotte
Carnegie Mellon
Michael A. Olson Stephen Reicher
Fathali M. Moghaddam University of Tennessee University of St. Andrews
Georgetown University
Sabine Otten Scott A. Reid
Susan Albers Mohrman University of Groningen University of California, Santa
University of Southern Barbara
California Ernest S. Park
Cleveland State University C. Lausanne Renfro
Benoît Monin New Mexico State University
Stanford University Craig D. Parks
Washington State University Cecilia L. Ridgeway
Margo Monteith Stanford University
Purdue University Paul B. Paulus
University of Texas at Jason E. Rivera
Richard L. Moreland Arlington Claremont Graduate
University of Pittsburgh University
Samuel Pehrson
Keith Murnighan University of Limerick Dawn T. Robinson
Northwestern University University of Georgia
Müjde Peker
David G. Myers University of Kent Kristie M. Rogers
Hope College Arizona State University
Thomas Fraser Pettigrew
Paul R. Nail University of California, Santa Adam Rutland
University of Central Arkansas Cruz University of Kent

Todd D. Nelson Nia L. Phillips Claudia A. Sacramento

California State University University of Kansas Aston University
xxviii Contributors

Eduardo Salas Stacey Sinclair Colleen H. Stuart

University of Central Florida Princeton University University of Toronto

Fabio Sani Robert E. Slavin Jenny C. Su

University of Dundee Johns Hopkins University University of Minnesota

Leonard Saxe Eliot R. Smith Jerry Suls

Brandeis University Indiana University University of Iowa

Toni Schmader J. Allegra Smith Robbie M. Sutton

University of Arizona University of Colorado University of Kent

Christiane Schoel Joanne R. Smith Janet K. Swim

University of Mannheim University of Exeter Pennsylvania State University

Janet Ward Schofield John J. Sosik Donald M. Taylor

University of Pittsburgh Pennsylvania State University McGill University
at Great Valley
Deborah J. Terry
David A. Schroeder
University of Queensland
University of Arkansas Marija Spanovic
University of Southern Yvonne Thai
Nicholas G. Schwab California University of California,
University of Wyoming
Charles Stangor
David O. Sears University of Maryland R. Scott Tindale
University of California, Los Loyola University Chicago
Angeles Rebecca Starkel
Loyola University Chicago Justine Tinkler
William T. Self Stanford University
University of California, Garold Stasser
Berkeley Miami University Sarah Sachiko Martin
Gün R. Semin Sofia Stathi University of California, Santa
Free University Amsterdam University of Kent Barbara

Viviane Seyranian Barry Staw Linda R. Tropp

Claremont Graduate University Haas School of Business University of Massachusetts,
Daniel B. Shank Sarah Stawiski
University of Georgia Loyola University Chicago Lisa Troyer
University of Connecticut
Kim Shapcott Walter G. Stephan
University of Western New Mexico State University Franziska Tschan
Ontario University of Neuchâtel
Jan E. Stets
Anna C. Sheveland University of California, Marlene E. Turner
University of Maryland Riverside San Jose State University

Jeffry A. Simpson Danu Anthony Stinson Rhiannon N. Turner

University of Minnesota University of Waterloo University of Leeds
Contributors xxix

Thomas R. Tyler Alberto Voci Gwen M. Wittenbaum

New York University University of Padova Michigan State University

Kaat Van Acker Joel Vuolevi Anna Woodcock

Katholieke Universiteit Leuven Free University Amsterdam Purdue University

Ilja van Beest Ulrich Wagner Anita Williams Woolley

Leiden University Philipps-University Marburg Carnegie Mellon University

Joseph A. Vandello Iain Walker Stephen C. Wright

University of South Florida Murdoch University Simon Fraser University

Eric van Dijk Fred O. Walumbwa C. Wesley Younts

Leiden University Arizona State University University of Connecticut

Paul A. M. van Lange Laurie R. Weingart Mary E. Zellmer-Bruhn

Vrije Universiteit Amsterdam Carnegie Mellon University University of Minnesota

Mark van Vugt Michael A. West Stephanie Zerwas

University of Kent Aston University University of North Carolina

Graham M. Vaughan Tim Wildschut Andreas Zick

University of Auckland University of Southampton University of Bielefeld

Theresa K. Vescio Kipling D. Williams Philip G. Zimbardo

Pennsylvania State University Purdue University Stanford University

Penny S. Visser James H. Wirth Sabrina Zirkel

University of Chicago Purdue University Mills College

Scarcely anyone would quarrel with the assertion of particular structural features (e.g., norms or
that humans are social animals or with the corol- roles); group members’ agreement on shared goals;
lary assumption that, in order to understand peo- patterns of interaction between members (e.g.,
ple’s thoughts, feelings, and behaviors, it is communication, reciprocal influence, coordinated
necessary to understand the role that groups play action); members’ emotional bonds to the group as
in human affairs. We all are members of large a whole and/or one another (cohesiveness); and
social categories based on such criteria as race, members’ identification with the group. Which
gender, ethnicity, and nationality, and most of us criterion, then, is the most important? As suggested
also belong to a range of groups and organiza- by the eminent group theorist Joseph McGrath,
tions (e.g., families, friendship cliques, work this question assumes that the goal is to distinguish
teams, religious congregations, political parties). groups from all other social aggregates (non-
These various social aggregates (which will be col- groups), which may be a futile exercise. He argues,
lectively referred to as “groups”) profoundly and we agree, that it is more productive to construe
affect our well-being and life trajectory through groupness as a continuum, such that a given social
their impact on such basic needs as physical sur- aggregate is more or less “groupy” depending on
vival, belongingness and intimacy, accurate knowl- how many of the above criteria it satisfies.
edge about the world, and sense of self and The second definitional issue is raised by the title
identity. The groups that people belong to vary on of the encyclopedia. What do we mean by “group
several dimensions, including size, voluntariness processes” and “intergroup relations”? In general,
of membership, composition (the types of people the term group processes refers to what happens
who belong), structure (e.g., norms, roles, status within groups, that is to how members of a group
systems), collective goals, level of conflict, reward- think, feel, and act toward others who belong to
ingness, relations with outgroups, and so on. the same group. Topics that are typically subsumed
Notwithstanding this diversity, groups matter, under the heading of group processes include the
and matter a great deal, to their members. impact of member diversity on team performance,
Before discussing the goals and organization of the development and operation of group norms,
this encyclopedia, two definitional issues need to be the conditions under which numerical minorities
considered. First, what do we mean by a group? As can produce innovation, the characteristics of
suggested above, groups can vary in many ways, effective leaders, the factors that influence whether
and so providing a succinct definition is no easy negotiators reach mutually acceptable agreements,
task. In fact, some observers have suggested, only the conditions under which team members fail to
partly in jest, that there are as many definitions of work hard on collective tasks, and the causes of
groups as there are researchers who study them. poor decision making in groups that are under
Rather than adding our imprimatur to one of these stress. Defining intergroup relations is more com-
definitions, we will mention some of the major plex. In most cases, this term refers to what hap-
criteria that group researchers have emphasized in pens between groups, that is to how members of a
defining groups (all of which are used, either group think, feel, and act toward others who
explicitly or implicitly, by some authors in this belong to a different group. Topics that are typi-
encyclopedia). These criteria include the presence cally subsumed under the heading of intergroup

xxxii Introduction

relations include the role of categorization in ste- what were formerly two distinct perspectives on
reotyping and prejudice, the impact of social roles groups.
on gender stereotypes, the function that language When we (John Levine and Michael Hogg) were
serves in maintaining stereotypes, the conditions approached by Michael Carmichael of Sage with
under which members of stigmatized groups engage the idea of editing an Encyclopedia of Group
in collective action against those who discriminate Processes and Intergroup Relations, we were ini-
against them, the circumstances under which inter- tially reluctant to take on a task of this magnitude.
group contact does and does not reduce prejudice, Nevertheless, a combination of M. Carmichael’s
and the justifications that dominant groups use to persistence and charm and our propensity to over-
rationalize discrimination against subordinate commit ourselves eventually led to a round of
groups. However, the term intergroup relations is handshakes and then a contract. We knew that the
not restricted solely to what happens between only way to make the venture a success was to con-
groups. It can also apply to what happens within vince other overcommitted colleagues to help. We
groups (i.e., to how members of a group think, feel, had three criteria for choosing associate editors—
and act toward others who belong to the same disciplinary background (we wanted people from
group) as long as these responses are influenced by the three key disciplines of social psychology, soci-
the broader context of intergroup relations. An ology, and organizational behavior), substantial
example is the black sheep effect, in which group expertise and visibility in their respective fields, and
members respond more negatively toward ingroup excellent judgment. With only a little arm twisting,
members who deviate from group norms (and we were able to assemble a Dream Team of associ-
more positively toward ingroup members who con- ate editors, which included Linda Argote, Marilynn
form to these norms) when the ingroup feels threat- Brewer, Jack Dovidio, Norb Kerr, Dick Moreland,
ened by an outgroup than when it does not. and Cecilia Ridgeway. With their help and
Moreover, the term intergroup relations can also guidance, we developed a set of approximately
apply to some forms of self-directed thought, feel- 300 topics for inclusion in the encyclopedia and
ing, and action. An example is the phenomenon of then selected experts to write the relevant entries.
stereotype threat, in which group members who Almost all of these people accepted our invitations,
are reminded that outgroups hold negative stereo- and those who did produced excellent entries.
types of their ability in certain domains then per- Either we or one of the associate editors evaluated
form poorly in these domains. each entry and provided feedback to the author(s).
Many entries were revised to increase their acces-
sibility to the general reader. The invitation and
Development of the Encyclopedia
reviewing process was orchestrated by our superb
Over the past 75 years, there has been a tremen- managing editor, Danielle (Dani) Blaylock, who
dous amount of theoretical and empirical work on was a PhD student at Claremont Graduate
group processes and intergroup relations by schol- University and is now a postdoctoral fellow at
ars from various disciplines. Until recently, these Queens University in Belfast, Ireland.
two lines of work were quite distinct, and few
efforts were made to bring them together in a single
Intended Audience
volume (or, as in the present case, pair of volumes).
This situation began to change in the 1980s due in In choosing topics for inclusion, selecting authors,
large part to the influence of the social identity and editing entries, we were mindful that the ency-
approach to groups championed by Henri Tajfel clopedia is not intended for professionals or
and John Turner. A major indicator of and impe- experts with extensive knowledge of social psy-
tus for the integration of group processes and inter- chology, sociology, or organizational behavior.
group relations was the establishment of a journal Instead, it is intended for general readers who
by this name in 1998, co-founded by Dominic want state-of-the-art information about group
Abrams and Michael Hogg. In an important sense, processes and intergroup relations that is presented
then, this encyclopedia is a reflection of the grow- in a clear and accessible manner. Our instructions
ing (though still far from complete) integration of to authors were designed to achieve these goals.
Introduction xxxiii

We stated that, “The main target audience is Influence and Persuasion

undergraduate students in various disciplines (e.g.,
Intergroup Relations in Society
psychology, sociology, business, political science,
education), but the encyclopedia should also prove Methodology
useful to graduate students and faculty as well as
high school students . . . entries must accurately
convey what behavioral scientists know about Theory
how people think, feel, and act toward ingroup
Types of Groups and Subgroups
and outgroup members . . . entries must be written
so that readers with little or no behavioral science
background can understand what is being said.”
Thanks to the hard work of the authors and the Acknowledgments
editorial team, we believe that these goals were We owe a great debt to the people at Sage who
accomplished. encouraged and assisted us at each stage of the
project. We thank Michael Carmichael for his
enthusiastic support at the beginning and Rolf
Organization of the Entries Janke for clearing obstacles that arose as we went
The entries are listed in alphabetical order, begin- along. Diana Axelsen did an excellent job of keep-
ning with Action Research and ending with ing our feet to the fire when progress lagged, Kate
Xenophobia. The lengths of the entries reflect our Schroeder was extremely patient during the
judgment regarding topic breadth and importance. copyediting phase (as were our two fine copy
The longest entries are approximately 4,000 editors—Bonnie Freeman and Jamie Robinson),
words, and the shortest are about 1,500 words. and Carole Maurer made the “end game” as pleas-
Many of the shorter entries provide detailed dis- ant as possible. Finally, Leticia Gutierrez and
cussions of topics that are only briefly covered in Laura Notton were very helpful with the Sage
the longer entries. To help readers locate the top- Reference Tracking system, which greatly facili-
ics in which they are most interested, we prepared tated our work.
a Reader’s Guide (see below) that organizes the We also want to thank the people who worked
entries into 12 general categories. Note that many most closely with us during the process of edit-
entries appear in more than one category. At the ing the encyclopedia. They include our associate
end of each entry, readers will find cross-references editors—Linda, Marilynn, Jack, Norb, Dick, and
to other entries and a short list of Further Readings. Cecilia—who devoted a great deal of time to the
project, and our managing editor—Dani—who
Reader’s Guide Headings
was efficient and good humored even when we
were not. Finally, we want to express our gratitude
Cognitions and Feelings to the authors of the entries. They met or exceeded
our expectations for both the quality and clarity of
Conflict and Cooperation Within Groups their entries, and we appreciate their efforts.
Group Decision Making Last but not least, we thank Jan and Alison for
their patience and support as we worked on the
Group Performance and Problem Solving encyclopedia.
Group Structure
John M. Levine
Identity and Self Michael A. Hogg
the role of the researcher. It also shows how action
Action Research research differs from traditional scientific research,
exploring the common elements and core processes
Action research is a process of participatory of action research as well as the nature of the
inquiry aimed at generating knowledge to guide knowledge that is created and applied.
action in pursuit of the participants’ goals. It can
be compared with traditional scientific research,
History and Forms of Action Research
which seeks to find the “truth” through methods
that are highly controlled by the researcher. Action Action researchers trace their philosophical roots
research, in contrast, seeks to generate knowledge to Aristotle’s notion of goal-directed action (praxis)
that solves specific problems and enables the as one of the key activities of human beings, dis-
actors in the situation to achieve their goals. It is tinct from theorizing (theoria), and crafting things
carried out collaboratively by these actors, includ- (poiesis). They also refer to his notion of practical
ing researchers, who engage in a mutual inquiry wisdom (phronesis), that is, the ability to reflect
process that takes into account the perspectives, and determine the appropriate ends to which to
knowledge, and purposes of all involved. Thus the direct one’s life. This combination of reflecting on
conduct of action research is heavily dependent on appropriate purposes and learning how to achieve
the process through which the various people them is central to all forms of action research.
involved in the action research project interact to Action research involves groups, or communities,
create new ways of understanding their situation of individuals reflecting and learning for action.
and new paths forward. Action research may Individuals represented in the action research group
include traditional, scientific data-gathering and may have different perspectives and goals, thus,
analysis approaches. But it is more generally char- effective action research requires group processes
acterized by inquiry approaches that build on for reflection, learning, and consensus building.
and create the knowledge of practice. It employs The modern philosophical roots of action
different knowledge-generating approaches— research may be traced to John Dewey’s early
approaches that place practitioners and their 20th-century discussions of learning through a
knowledge front and center. The ultimate test of cycle of reflective thinking about problems, formu-
the knowledge that is generated through action lating hypotheses about what might solve them,
research is whether the action taken because of and testing them through practical action. The
the knowledge accomplishes the purposes of the term action research as an approach to social sci-
participants. ence research is often traced to the field research
This entry provides a short history that illus- tradition begun by Kurt Lewin during the Second
trates the varieties of action research and discusses World War. In this approach, people in various

2 Action Research

real-life settings such as work organizations par- Differences From Traditional

ticipate in research to discover better ways to Social Science Research
accomplish their goals. This action research tradi-
There are several ways in which action research dif-
tion emphasizes the formulation of theory that can
fers from more traditional research. These include
be tested and refined through experiments whose
the purpose of the research—whether the researcher
results have an impact on practice—so that the
is out to discover scientific “truth,” or whether the
requirements of practice are met and systematic
researcher aims to help people accomplish their
knowledge is furthered. Lewin’s belief that the best
purposes. Other differences concern where the
way to understand something is to change it has
research is conducted, and the methodologies that
been echoed by action researchers ever since.
are used. These differences have implications for the
Lewin’s approach sponsored the development
relationship between the researcher and the people
of the sociotechnical systems (STS) tradition for
in the real-life settings being studied.
improving work systems. Early STS research dis-
covered that productivity is enhanced by leader-
ship styles and work systems that empower
employees to be actively involved in making deci- Traditional social science research is based on a
sions about how to run their own work units. search for the “truth” about the phenomena being
These core STS ideas were expanded and refined investigated—whether they are aspects of the
through many action research projects in which physical world, investigated by hard scientists, or
the participants in a work setting such as a factory aspects of the social world, investigated by social
or a mine collaboratively designed their own work scientists. Scientific truths, as scientists have been
setting to be technically and socially effective. able to discover them, are embodied in theories
Researchers involved in this stream of action that yield predictions for further investigations to
research provide theoretical input and a process confirm and expand theoretical understanding. In
for design and planning. They study the group traditional social science research, the quest is to
processes through which the diverse members of find the truth about the social behavior of indi-
the setting—managers, supervisors, and workers— viduals, groups, organizations, and societal institu-
develop a new way of working together, the tions. The social entities being studied are treated
choices they make about how to organize them- as the objects of the research and it is considered
selves, and the outcomes of putting these choices poor form to engage with them, as the scientist is
into action. expected to remain disinterested in outcomes and
Since the 1980s, the Norwegian democratic dia- purposes and to retain objectivity in explaining
logue approach has emphasized the gathering of behavior.
various stakeholders (including management, the Purposes are central to action research, as the
workforce, government, and unions) in confer- quest of action research is to create knowledge that
ences in which they can speak with each other as can help participants accomplish their purposes.
equals about how to move forward on issues such Various traditions of action research differ in the
as work organization. The underlying principle is extent to which they emphasize the use of aca-
to move from traditional adversarial approaches demic theoretical knowledge in the processes
to cooperation through democratic dialogue, and through which purposes are defined and action is
the purpose is to build relationships and establish crafted. At one extreme are action researchers who
a new way of making decisions that take the inter- believe there is no generalizable truth, and that all
ests of all parties into account. At about the same knowledge is created in specific situations. Their
time, throughout the world, social justice has purpose is not to discover truth, but to introduce
become the focus of action research activities that frameworks for interaction that enable partici-
emphasize gender and race issues, and of emanci- pants to gather information, make choices, and
patory work in poor nations that is based on take action to accomplish their purposes. This is
empowering ordinary people by helping them the position of the Norwegian democratic dialogue
develop the capabilities to generate their own advocates. The action researcher’s role in the con-
knowledge as a basis for action. versation is as a member of the group who brings
Action Research 3

knowledge of how to set up dialogues, reflection, successive setting presented different challenges and
and learning. Yet even these action researchers are different opportunities for learning. The partici-
guided by their own values and purposes, such as pants in each setting were interested in creating
achieving democratic dialogue or emancipation. their own solutions, yet the action researcher
Other action researchers, such as STS research- brought useful experience and knowledge from
ers, have as their purpose contributing to knowl- previous research. For the action researcher in a
edge that is generalizable across settings, in addition new setting to establish enough trust for the other
to helping participants in an action setting learn participants to learn from this previous work, he
how to generate actionable knowledge. These or she has to be open to the uniqueness of each
researchers bring relevant theoretical constructs and setting and of the group of participants engaged in
the cumulative knowledge of the social sciences— the participative design process—and open to dif-
sometimes called content knowledge—to bear in ferent design choices and resulting action. Only in
the processes by which participants define and this way can the research be truly participative and
solve problems. The researchers’ knowledge of the researcher learn how the unique factors of the
theory and the participants’ knowledge of practice setting contribute to general knowledge.
are combined to yield solutions to problems and
designs for action. Through the action research,
Research Methodologies
the practical knowledge of the participants and the
theoretical knowledge of the social sciences are Beyond differences concerning randomization
both expanded. These action researchers bring versus in situ focus, action research methodologies
expertise to diagnose problems and to intervene in differ from traditional social science research in
a way that helps participants solve them. They other ways, including the amount of control the
may train, educate, and facilitate the group, thus researcher has over the research and the data-
assuming a central position in the group. gathering methods. Traditional researchers con-
duct research in a carefully controlled manner to
Research Conducted In Situ eliminate alternative explanations for the results
that they find. This generally means that the
Traditional researchers carrying out studies in objects of the research are unaware of the purpose
the field try to avoid focusing on only one organiza- of the research, the research questions, and the
tion. They seek random selection of the populations hypotheses being tested. This is believed to be nec-
being studied—sometimes into treatment and con- essary so that they do not behave in a manner that
trol groups—in order to randomly distribute exter- distorts the findings—either by trying to act in a
nal factors that might otherwise distort the findings. manner that fits the expectations of the researcher
Action research, in contrast, always is situated in or by trying to prove the opposite. Highly con-
and generates knowledge about a particular social trolled approaches fit a model where the researcher
system that has expressed interest in engaging in is seen as having a privileged knowledge-creating
action research—for example, a work unit, an role in society and is given permission to study oth-
industry, a community, or a subpopulation. It ers. People and organizations may agree to be part
focuses on creating the knowledge-generating and of such research to further science, but often do
action-taking capability in that particular system. not believe that it will yield knowledge useful to
An action research project enables researchers to their personal purposes.
learn about one system, and to test and expand The members of the action research community
theory in only that system. Action researchers who are, in contrast, co-investigators. Purposes are
are interested in building widely applicable knowl- transparent and they are codetermined by the
edge do so through a succession of action research action researcher and other participants. The
projects in different settings. For example, STS research questions are often co-defined by the par-
researchers created cumulative knowledge about ticipants, because these questions have to do with
the participative design of manufacturing systems their real-life situation. If there are hypotheses
for high performance through a succession of action guiding the research, these also will be formulated
research interventions in different factories. Each and influenced both by the researchers’ knowledge
4 Action Research

from theory and the other participants’ knowledge including the feelings and experiences of partici-
from practice. Finding common purposes and pants and the meanings they attribute to their
hypotheses to guide the inquiry and action plan- interactions.
ning process entails finding a process to come to Interpretation of these rich and diverse data is
agreement despite differing experience bases, central to the inquiry process. The group members
knowledge, and preferences. attach meaning not only to systematically collected
The action researcher is one of the participants data but also to their interactions, including those
in this community of co-investigators. Like all par- between the researcher and the practitioners.
ticipants in any group, action researchers face Academic interpretation is only one perspective in
challenges in defining and achieving perceived the process of attributing meaning. Given that the
legitimacy for their role in the group. The group group is working to agree on different ways of
members are being guided to behave in ways they operating and different outcomes, the academic
may not be used to—putting aside rank and biases interpretation may have the least impact because
and listening to and building on the perspectives of the participants’ criteria are usefulness and rele-
all. The members of the group may only appreciate vance. Both the process and content knowledge
the power of a truly participative inquiry process brought to the group’s collective sense-making
after experiencing it. Only then may they under- process by the action researcher will be interpreted
stand collaboration and appreciate the researcher’s in conjunction with the full set of knowledge
contribution. brought by the members of the group.
The action researcher who claims to have con-
tent expertise relevant to the group’s purpose, and
Common Elements in Action Research
who aims to further that knowledge through the
action research, faces the additional challenge of The broad assortment of approaches that are
achieving legitimacy and trust for that expert role labeled action research share some defining attri-
within the group. Theoretical knowledge is likely butes: a discourse-based learning cycle, an expanded
to be rejected unless the researcher engages with definition of knowledge, and an inherently politi-
the group, and accepts the importance of combin- cal nature. Each of these places strong require-
ing theoretical knowledge with the group mem- ments on the action research group’s interaction
bers’ knowledge of practice, to yield an approach patterns.
that is tailored to the situation and purposes at
hand. A Discourse-Based Learning Process
Traditional social science research is often char-
acterized by data-gathering methods such as sur- Action research is a discourse-based inquiry and
veys, questionnaires, and structured observations reflection process through which stakeholders and
that are coded, counted, and analyzed statistically participants in the real-life situation come together
to discover patterns of relationships between vari- to make choices, plan, and take action. If the
ables predetermined by the researchers. For exam- action research group is able to establish itself as
ple, researchers may be interested in whether the an ongoing learning community, the action and its
purposes of low-status group members are less consequences feed back into the learning of that
likely to be voiced and achieved in an action plan- community, establishing a cycle of experiencing,
ning process; they may measure the status of each reflecting, planning, and action taking. Common
member and ask the group members individually steps include:
to what extent they felt their ideas were taken into
account. Although such traditional methods may •• establishing the group to collectively engage in
be part of an action research project, action communication designed to raise consciousness
researchers generally feel that these methods are and increase mutual understanding and to create
insufficient to capture the complexity of human a sense of common purpose
interaction. These researchers are likely to intro- •• inquiring by gathering relevant data and
duce a variety of ways of understanding the system knowledge from each other and other sources,
and to encourage the consideration of rich data, sometimes including scientific knowledge and a
Action Research 5

formal data-gathering process applying formal orientation and collectively describe and create
scientific approaches their real-life situation as they would like it to be.
•• interpreting and reflecting on the meaning of the Different action research groups and their mem-
information and knowledge assembled, and its bers may begin the process with different compe-
relationship to purposes tencies in and orientations to these different kinds
•• deciding on and planning action focused on of knowledge. In work settings, managers and
solving the problems being addressed and technical employees may be heavily steeped in
achieving the purposes that the group has technical knowledge, and may see relational and
collectively defined reflective knowledge as unimportant to achieving
•• reflecting on the results of the action that feed their purposes. First-line employees, however, may
back into an ongoing inquiry, reflection, and orient themselves to these latter forms of knowl-
action-taking cycle edge, which determine their trust in the process
and focus them on creating a workplace where
A major role of the action researcher is to facili- they experience meaningful interpersonal relation-
tate that process while modeling it, thereby increas- ships and where their purposes are taken seriously.
ing the capacity of the group to develop knowledge. Inherent in effective action research is developing
When this capacity has been developed, all mem- an appreciation for these different forms of knowl-
bers will function as action researchers. edge that allow a community of participants to
move forward together.
An Expanded Nature of Knowledge
Political Processes
Scientific knowledge deals with the theoretical
Integral to action research is the capacity of the
connections between variables—and is aimed at
group to create power dynamics where the mem-
answering questions such as whether carrying out a
bers of the action research group are all heard, and
particular action will lead to a particular outcome. A
their knowledge, preferences, and perspectives are
more diverse set of knowledge is required to define
taken into account. Words like participative,
effective practice, and making decisions regarding
equal, democratic, social justice, and emancipation
practice requires the group to interpret patterns of
are used by action researchers in different kinds of
information and pull together diverse knowledge
settings. All of these terms carry the notion that the
sets. Even with a firm grasp of what is objectively
formerly disempowered will become empowered
known, and even with deep know-how about how
to influence the choices made and directions taken.
to achieve particular outcomes, the group is still
Achieving this requires a process where those with
faced with the challenge of how diverse participants
formerly privileged knowledge and power, includ-
who may not start out knowing or trusting each
ing the action researchers themselves, do not
other can find consensus about how to proceed.
dominate the process. Ultimately the choice of
Beyond objective knowledge, two other kinds action is politically determined. The goal of action
of knowledge are required for effective action research is to ensure that the political process is
research. One is the knowledge participants participatory and builds on the knowledge and
develop of each other—relational knowledge— purposes of the members.
that enables them to understand and feel empa-
thy for the others’ points of view. This is the Susan Albers Mohrman
knowledge that allows the group members to go See also Cooperative Learning; Group Learning; Lewin,
beyond their experience of the world and engage Kurt; Process Consultation
in reflection and action planning that incorpo-
Further Readings
rates the views and purposes of others. The sec-
ond kind of knowledge is the reflective knowledge Adler, N., Shani, A. B., & Styhre, A. (Eds.). (2003).
that comes from a truly collaborative inter- Collaborative research in organizations. Thousand Oaks,
change, and that equips the group to be critical CA: Sage.
of the status quo and to reformulate purpose. It Argyris, C. (1970). Intervention theory and method:
enables the group to get beyond a problem-solving Behavioral science view. Reading, MA: Addison-Wesley.
6 Affect Control Theory

Greenwood, D. J., & Levin, M. (1998). Introduction to identities, behaviors, settings, and emotions. The
action research: Social research for social change. labels in turn evoke affective meanings that are
Thousand Oaks, CA: Sage. shared with a larger culture. These affective
Lewin, K. (1951). Field theory in social science: Selected meanings are conceptualized and measured using
theoretical papers. New York: Harper & Row. three universal dimensions of meaning that
Pasmore, W. (1988). Designing effective organizations: Charles Osgood found to account for a substan-
The sociotechnical systems perspective. New York: tial amount of the variation in the lexicons of
John Wiley. over 20 language cultures. First, evaluation is a
Reason, P., & Bradbury, H. (Eds.). (2001). Handbook of
measure of a concept’s goodness or badness mea-
action research: Participative inquiry & practice.
sured on a continuum from bad, awful to good,
Thousand Oaks, CA: Sage.
nice. Second, potency is a measure of a concept’s
Shani, A. B., Mohrman, S. A., Pasmore, W. A., Stymne,
power and ranges on a continuum from power-
B., & Adler, N. (Eds.). (2007). Handbook of
collaborative management research. Thousand Oaks:
less, weak, small to powerful, strong, big. Third,
CA: Sage.
activity is a measure of a concept’s liveliness or
quietness and ranges from slow, quiet, old to
fast, loud, young. ACT refers to the affective
meanings measured on these dimensions as senti-
Affect Control Theory ments. Sentiments are trans-situational, general-
ized affective responses to specific symbols that
Affect control theory (ACT) is a mathematical are widely shared in a culture (or subculture).
theory of social interaction developed by David These three fundamental dimensions of mean-
Heise in the 1970s. Based on symbolic interac- ing serve as cultural abbreviations that describe
tionist ideas, ACT explains how interpersonal important affective information about all elements
interactions are constrained by the symbolic cul- of an interaction—identities, behaviors, emotions,
ture contained in language and the meanings it and settings. These dimensions are core to our
associates with things. The theory describes how understanding of intra- and intergroup processes.
actors cognitively and affectively negotiate these The evaluation (good–bad) dimension helps char-
cultural meanings to maintain a “working defini- acterize processes like status and affiliation at the
tion” of the situation. It also makes predictions interpersonal level and solidarity and cohesion at
about behaviors, emotions, and identity attribu- the group level. The potency dimension character-
tions that occur in culturally situated interactions. izes power relations between social actors and
Thus affect control theory helps us understand between social groups. The activity dimension
both inter- and intragroup processes. characterizes the expressiveness of identities and
ACT proposes that interactions confirming cul- interpersonal behaviors as well as feelings of
tural meanings require minimal cognitive process- excitement or quiet.
ing because such situations feel normal and The three dimensions of meaning operate cross-
expected. In contrast, when an interaction does culturally, but the sentiments associated with par-
not confirm standard cultural meanings, people ticular labels are specific to a culture or subculture.
attempt to cognitively interpret it, while being sig- ACT researchers have empirically compiled senti-
naled by their emotions that the situation is unex- ments associated with hundreds of identities,
pected. A core proposition in the theory is the behaviors, setting, emotions, and traits into cul-
control principle, which says that people attempt tural dictionaries. Cultural dictionaries have been
to restore the meanings in a situation after devia- compiled for the United States, Canada, Japan,
tion from the cultural standard, typically by gener- China, Germany, and Northern Ireland, and in the
ating new social behaviors. future this work will be extended to include other
cultures. All elements of an interaction (identities,
behaviors, settings, emotions, traits) are indexed
Cultural Sentiments
along the same three dimensions of meaning. This
ACT assumes that individuals understand social provides a common metric for use in the theory’s
events by labeling their elements—including equations that describe social interaction. Every
Affect Control Theory 7

label evokes culturally specific amounts of good- sheriff is engaging in an interaction that suggests
ness, powerfulness, and activity. In the U.S. cul- that the view of this outlaw should be better than
tural dictionary, for example, the identity of the typical cultural view of an outlaw. This incon-
outlaw is quite bad, slightly powerful, and some- gruence in the direction of increased goodness
what active. The behavior reward is extremely means the outlaw will experience positive emo-
good, quite powerful, and slightly active. tions, such as being thankful or relieved in response
to the interaction.
Second, actors experiencing this deflection,
which was initially signaled by emotions, will try
Although all social concepts evoke cultural senti- to restore the normative definition to the situation
ments, the meaning of a particular element of an through additional interaction. The theory predicts
interaction may change as the interaction develops. that actors will strive to restore the definition of
ACT proposes that individuals label elements of a the situation even if the emotion experienced was
social interaction with concepts common to their positive. Thus, in the example of the outlaw who
culture, and those concepts have sentiments associ- rewarded the sheriff, the outlaw could yell at the
ated with them. In addition, these concepts provide a sheriff or the sheriff could convict the outlaw to
reference point throughout the interaction, allowing restore cultural sentiments. ACT suggests that
any observer of the interaction to determine if the behavioral responses are often the easiest method
interaction deviates from culturally normal behavior. for controlling the inconsistencies created during
ACT contains equations that specify how each social interaction.
element of an interaction will contribute to altering Third, for interactants who cannot restore the
the sentiments of those elements. For example, if an definition of the situation behaviorally and for
outlaw rewards a sheriff, observers will change their observers who are not participating in the interac-
impressions of that particular outlaw, that particular tion, the situation can be resolved cognitively. For
sheriff, and that particular rewarding behavior. On a mild deflection, accepting the transient impres-
the evaluation dimension the outlaw will no longer sion may restore a working definition of the situa-
be seen as bad, but much closer to neutral, and the tion (in this case, deciding that this outlaw is not
rewarding will be seen as only slightly good. The ele- as bad as other outlaws). For more extreme deflec-
ment’s altered sentiments in a situation are referred tions, relabeling elements of the situation is another
to as transient impressions. possibility (e.g., the actor is not an outlaw but
Differences between the cultural sentiments and merely a rival, or the behavior was not really
the transient impressions reveal the degree to rewarding but taunting). ACT does not make spe-
which a situation is culturally normative. In affect cific predictions about when a behavioral or cogni-
control theory, differences are called deflection; tive approach to resolving the incongruency will
higher levels of deflection suggest less culturally take precedence, but it does suggest that a working
normative events. An example of such an event is definition of the situation must be restored for
an outlaw rewarding a sheriff, because according individuals to make sense of their interactions and
to the affect control equations it is not normal for the larger social world.
a bad person to do something good for a good
person. The theory suggests that when an interac-
Mathematical Foundation
tion is not harmonious with cultural expectations,
people experience unusual emotional, behavioral, ACT is a mathematical model with the theoretical
and cognitive responses. principals encoded in equations. Impression-formation
First, actors will experience this deflection, or equations specify the transient impressions of ele-
incongruence in meaning, emotionally—with larger ments after an interaction occurs. Similarly, labeling
incongruencies generally producing more intense equations can indicate how elements of interaction
emotions. An actor’s emotion will be positive or could be redefined by an observer. Behavior-prediction
negative depending on both how positive the event equations lead to predictions of what actions interac-
is and whether the incongruence is more positive tants might take to restore a working definition to the
than that actor’s identity. An outlaw rewarding a situation. Emotion equations predict emotions the
8 Affirmative Action

interactants are likely to experience as the result of an Smith-Lovin, L., & Heise, D. R. (1988). Analyzing social
interaction. interaction: Advances in affect control theory. New
All of the ACT equations are generated from York: Gordon & Breach.
empirical data for a particular culture. These equa-
tions as well as the cultural dictionaries have been
implemented in computer programs such as
Affirmative Action
ACT), which allows simulations and predictions of
interactions with all the ACT equations. These Affirmative action refers to efforts to provide
predictions can then be used to make specific equal opportunities for all in employment and
hypotheses about emotional, behavioral, and cog- education. This entry focuses on affirmative action
nitive reactions that are then subject to empirical in the United States because it has been the pri-
testing. mary site for social science research on the issue.
Researchers have made fruitful use of affect con- Affirmative action policies and programs take
trol theory to study stereotyping and intergroup measures to increase the representation of women
relations, the dynamics of therapeutic support and racial/ethnic minorities in employment and
groups, leadership structures within task groups, higher education through the use of targeted
political identification and action, and responses recruiting and training, formalizing personnel
to injustice. This research relies on a variety of practices, preferential treatment in hiring and edu-
methodological approaches, including laboratory cational admissions, and sometimes the use of
experiments, formal cross-cultural comparisons, quotas. The motivation for affirmative action is to
ethnographic studies, and survey research. redress historical inequalities between social
groups by “leveling the playing field” for groups
that are disadvantaged by past and current dis-
Dawn T. Robinson and Daniel B. Shank
crimination. Affirmative action has led to impor-
tant changes in intergroup relations, and its
See also Leadership; Social Identity Theory; Support history serves to highlight both the effectiveness
Groups; Symbolic Interactionism and limitations of laws aimed at changing existing
relations between social groups that differ in
power and status.
Further Readings
Heise, D. R. (1979). Understanding events: Affect and the
construction of social action. New York: Cambridge
University Press. Throughout its 45-year history, affirmative action
Heise, D. R. (2007). Expressive order: Confirming
has been met with controversy and debate. While
sentiments in social actions. New York: Springer.
presidential committees since the 1940s had been
wrestling with nondiscrimination clauses in federal
contracts, the first mention of the term affirmative
MacKinnon, N. J., & Bowlby, J. W. (2000). The affective
action came in 1961 from Executive Order 10925
dynamics of stereotyping and intergroup relations.
Advances in Group Processes, 17, 37–76.
issued by President John F. Kennedy. Executive
Osgood, C. E., May, W. E., & Miron, M. S. (1975). Order 10925 was the first legal mandate requiring
Cross-cultural universals of affective meaning. Urbana: organizations that do business with the federal
University of Illinois. government (federal contractors) to “take affirma-
Ridgeway, C., & Smith-Lovin, L. (1994). Structure, tive action” to ensure that hiring and promotion
culture, and interaction: Comparing two generative practices are free of discrimination. Following the
theories. Advances in Group Processes, 11, Civil Rights Act of 1964, President Lyndon B.
213–239. Johnson issued Executive Order 11246 (E011246)
Robinson, D. T., & Smith-Lovin, L. (2006). Affect in 1965, creating the first affirmative action policy
control theory. In P. J. Burke (Ed.), Social psychology. to be enforced enough to provoke controversy and
Palo Alto, CA: Stanford University Press. debate. E011246 applies to the federal government
Affirmative Action 9

and to federal contractors with a contract of at training, formalizing job posting procedures to
least $50,000 and 50 or more employees. promote equal access to hiring information, and in
Initially, the policy was targeted at eliminating certain cases, giving additional weight or assigning
discriminatory barriers for racial/ethnic minor- extra points to race- and gender-disadvantaged
ities, but it was modified in 1967 to protect applicants in hiring and admissions decisions.
groups based on color, religion, sex, and While much of the controversy surrounding
national origin. While E011246 only requires affirmative action has focused on the use of quo-
that the federal government and the businesses tas, the law forbids the use of quotas except in
that contract with the government have affir- circumstances in which courts order it as a remedy
mative action plans, many noncontracting for cases of blatant discrimination. In 1978, the
organizations have adopted policies that Supreme Court ruled against the use of explicit
enhance diversity and provide evidence against quotas in the case of the Regents of the University
potential discrimination lawsuits. of California v. Bakke. In this case, Allan Bakke, a
Affirmative action can be distinguished from White applicant who was not admitted to the
equal opportunity policies that simply prohibit medical school at the University of California,
discrimination by its call for actions to eliminate Davis, sued when several minority applicants were
barriers to equal opportunity. The presumption accepted to the medical school despite having
behind affirmative action is that even race- and lower grades and test scores than he did. In this
gender-neutral policies can operate in ways that case, the university was reserving 16% of admis-
advantage some groups over others. As Johnson sions spots for minority applicants and evaluating
relayed in his speech justifying E011246, “You the qualifications of the White students separate
do not wipe away the scars of centuries by saying: from the minority students. The Supreme Court
‘now, you are free to go where you want, do as ruled that this was unconstitutional, but in addi-
you desire, and choose the leaders you please’ . . . tion wrote that schools could treat the minority
and still justly believe you have been completely status of applicants as one among other character-
fair.” This inequality in access may be the result istics in making admissions decisions. This ruling,
of current and past discrimination, institutional then, made explicit quotas illegal but certain forms
forms of racism and sexism that bias measures of preferential treatment permissible.
of merit, and/or the tendency of people to hire Since the Bakke case, there have been a num-
those they know or who have similar back- ber of other important legal cases that have lim-
grounds. ited the methods by which colleges and universities
To ensure that equal opportunity exists, affir- can implement affirmative action plans. Appellate
mative action policies require employing organi- courts ruled that the admissions plans for the
zations and schools to allocate resources toward University of Texas Law School and the University
(1) evaluating workforce and enrollment statis- of Georgia violated the equal protection clause of
tics and (2) taking proactive measures to bal- the 14th Amendment. In a landmark case in
ance the representation of women and racial/ 2003, the U.S. Supreme Court ruled that the
ethnic minorities with respect to their availabil- University of Michigan’s point system for under-
ity for hire or admission. In evaluating statistics, graduate admissions made race too prominent of
employers and educational institutions evaluate a factor, but the law school’s practice of consid-
the proportion of qualified women, African ering race, but not assigning a specific weight to
Americans, Native Americans, Latinos/as, and it, was permissible. This case was important
Asian Americans and compare this proportion because the justices affirmed that broad social
to the number employed or admitted and retained. value can be gained from diversity in the class-
When any target group is underrepresented relative room. In addition to the court rulings, California
to their availability, federal contractors are required and Washington have passed propositions ban-
to develop affirmative action plans that include ning any form of preferential treatment based on
goals and timetables for making good-faith efforts race, color, sex, or national origin, and Florida
to remedy the problem. Goals for meeting affirma- has banned race-based preferential treatment in
tive action plans include targeted recruitment and college admissions.
10 Affirmative Action

Impact of Affirmative Action people of color, which in turn reduced the earnings
on Workplace Composition gap between Whites and people of color. In addi-
tion, federal contractors have granted more promo-
Has affirmative action been effective at increasing tions to people of color than have noncontractors.
the number of women and people of color in the Early studies comparing federal contractors to
workplace? To answer this question, many studies noncontractors underestimated the effectiveness of
have compared the proportion of women and peo- affirmative action because many noncontractors
ple of color in the federal workforce, where affir- implement voluntary affirmative action plans.
mative action is required, to that in the private More recent studies have accounted for this by
workforce, where affirmative action is not required. comparing the workplace composition of firms
Several studies have consistently shown that the that report having an affirmative action plan to
percentage of women, Hispanics, and Blacks in the those that report not having one. The results are
government workforce is higher than the percent- consistent with previous research, showing that
age in the private workforce. In addition, research organizations with an affirmative action plan have
studies have shown that women and minorities are more women and minorities, a higher proportion
more likely to advance to management in the public of women and minorities in high prestige jobs, and
sector, and that occupational advance has led to a smaller earnings gap.
smaller race- and gender-related earnings gaps in One of the major controversies surrounding the
the public sector. While these studies suggest that debate about affirmative action involves the claim
affirmative action may increase workplace diver- of reverse discrimination. Opponents of affirma-
sity, the broader differences between private and tive action argue that in giving advantages to
public sector jobs make it difficult to draw conclu- women and people of color, qualified White men
sions about the independent effect of affirmative are discriminated against. To assess the validity of
action. this argument, some researchers have looked for
There are other studies that more successfully evidence that when women and minorities are
isolate the impact of affirmative action by compar- hired through affirmative action plans, their quali-
ing the growth of women and minority employ- fications for the job are less than what is needed to
ment among organizations with federal contracts perform well. Economists have approached this
to that among similar organizations with no affir- question by examining whether the redistribution
mative action requirements. Studies of changes in of workers has come at the expense of quality and
the period in which affirmative action was most productivity. While the difficulties in assessing pro-
stringently enforced (1974–1980) showed the ductivity across organizations limits the available
employment of Black men and women growing at evidence, econometric studies conducted during
a faster rate and the employment of White men the late 1970s and late 1980s, when affirmative
growing at a slower rate among federal contrac- action was most strictly enforced, showed that the
tors than among similar establishments without industries under the most pressure to comply with
federal contracts. Studies also showed that compli- affirmative action plans were no less productive
ance reviews are an important form of enforce- than other industries. In addition, company-level
ment. Federal contractors that had undergone a analyses showed that affirmative action obliga-
compliance review had twice as much Black male tions and changes in workplace composition had
employment growth as such businesses that had no negative effect on company profits. In fact,
not been the subject of a compliance review. more recent research has shown that companies
Another important factor in understanding the that employ the highest proportion of women and
impact of affirmative action is employment growth minorities enjoy higher returns on their stocks
more generally. Federal contractors that had many than the market average, while those that employ
job openings were more likely to increase their the lowest proportion of women and minorities
representation of Black employees than contractors had stocks that underperformed relative to the
with less growth in the early years of affirmative market average.
action. A number of studies have also shown that One of the most commonly relied upon ways
affirmative action raised the occupational levels of that opponents of affirmative action assess whether
Affirmative Action 11

affirmative action results in reverse discrimination deal of controversy and has been criticized in a
is to compare the qualifications of employees who number of publications for its methods and the
benefit from affirmative action to the qualifica- author’s interpretations. In any case, research has
tions of those who do not. While this approach shown that even when students do not finish law
enjoys the most media attention, it is poor science school, simply attending increases annual earnings
because measures of merit are often intrinsically and thus serves to boost the life outcomes of peo-
tied to institutional forms of sexism and racism. ple of color.
Subjective measures are subject to implicit and One of the factors that the Supreme Court has
explicit race and gender bias, while objective mea- taken into account in ruling on affirmative action
sures like standardized testing have been shown to is the social value gained by diversity. A large body
be culturally biased and poor predictors of perfor- of research has shown that diversity leads to posi-
mance. Research using data from employers in tive learning outcomes for both Whites and people
Atlanta, Los Angeles, Boston, and Detroit has of color. Research has shown that interactions in
revealed that among employers most committed to diverse settings improve the ability to take the per-
affirmative action plans, women and minorities spective of others, and that heterogeneous groups
had lower average qualifications than White males, outperform homogeneous groups when members
but their performance on the job was no different. perceive their contribution to be important. In
addition to the direct benefit of diversity, ethnic
minorities are more likely than Whites to use their
Impact of Affirmative Action
education to benefit society—professionally and
on Diversity in Higher Education
through civic engagement. Thus, increasing the
Affirmative action in education has been even representation of minorities in higher education
more controversial than in employment, experi- institutions has been shown to have long-term
encing more legal challenges on charges of reverse positive social outcomes.
discrimination. The appellate and supreme courts
have ruled in favor of plaintiffs charging reverse
Impact of Affirmative Action
discrimination, and in so doing have effectively
on Intergroup Relations
eliminated quotas and severely limited race-based
preferences in college admissions. Because affirmative action is unique in its proactive
Research has suggested that prior to the current approach to reducing inequality, there are a num-
restrictions, affirmative action led to more diver- ber of important ways in which attitudinal responses
sity in higher education. Recent empirical research to the law and its impact expand our understand-
analyzing student data from 28 elite colleges and ing of intergroup relations. Attitudes about affir-
universities in 1951, 1976, and 1989 showed that mative action vary significantly according to how it
race-based affirmative action significantly increased is defined. Research has shown that people tend to
the number of Blacks admitted to and attending be the most supportive of outreach programs and
elite institutions. In addition, findings showed that formalized job postings, while there is greater resis-
ethnic minorities graduated at the same rate as tance to preferential treatment practices. Attitudes
Whites. Another recent study of the University of about affirmative action also vary according to the
Michigan Law School showed similar findings. As gender, race, political ideologies, and prejudice lev-
in the employment arena, the evidence—that stu- els of individuals. Research has shown that women,
dents admitted as a result of race- and gender- people of color, political liberals, and those who
sensitive policies graduate at the same rate as those hold the least prejudiced attitudes tend to be more
admitted without affirmative action—seriously supportive of affirmative action. The popular press
hampers claims of reverse discrimination. While has characterized affirmative action as a racially
most studies have provided this evidence, one polarizing policy that divides Whites and Blacks.
study has found that Blacks admitted to elite law Research has shown that Whites do tend to be less
schools entered with lower average credentials supportive of affirmative action than are Blacks,
than White students and were less likely to gradu- but the extent of polarization has been exagger-
ate and pass the bar. This study has sparked a great ated. Both Blacks and Whites tend to oppose quota
12 Affirmative Action

systems and support outreach programs. Though concern, showing that when affirmative action is
Whites resist preferential treatment practices more mentioned to people prior to their being asked to
than do minorities, there is a significant proportion evaluate women and men job applicants, women
of Whites and minorities who support and oppose are rated as less competent. In addition, laboratory
such forms of affirmative action. research has shown that when people believe they
People express opposition to special preferences have been granted preferential treatment or are led
in hiring and admissions because such preferences to believe others believe this, their general and
are perceived to violate norms of fairness and jus- task-specific performance is lower. Recent research
tice. While attitudinal survey research has shown has linked this disempowering effect to resistance
that people who oppose affirmative action believe it to affirmative action, showing that when political
is unjust, it has also shown that such a concern ideology, support for gender-based affirmative
affects opposition differently depending on the race action, symbolic racism, and perceived discrimina-
and gender of the group benefiting from the policies, tion are accounted for, racial minorities are more
as well as the race and gender of the respondents. likely to oppose special hiring preferences for their
For example, concerns about justice drive resistance own group when they have a close friend who is
to race-based affirmative action more than sex- White. This finding suggests that when the percep-
based affirmative action. In other words, people’s tion of the dominant society is close to home,
social locations and attitudes about other groups resistance to affirmative action is greater.
have greater explanatory power regarding resistance The history and impact of affirmative action
to affirmative action than do people’s adherence to serve to highlight the nature of modern race and
fairness and justice norms. One study also found gender relations. Efforts to eliminate barriers to
that prejudice levels mediated people’s tendency to equal opportunity have improved the educational
misconstrue affirmative action programs as justice and labor market outcomes of women and minor-
violating when they were explicitly designed not to ities. In spite of these gains, affirmative action
advantage certain groups over others. faces a formidable battle in winning over the sup-
In moving beyond explanations rooted in prin- port of those who stand to gain from the current
ciples of fairness, social psychologists have developed system of inequality. So long as Whites and males
a number of different theories for under­standing oppose equalizing policies like affirmative action,
variations in attitudes about affirmative action. those who benefit from the policies also incur the
Scholars who place primary importance on the role costs by being perceived as less worthy than others
of racial prejudice have theorized that individualist of their successes. This lag in attitudinal change is
values, which lead to resentment against Blacks for both a reason for affirmative action and an unfor-
their struggles to succeed economically (symbolic tunate consequence of laws aimed at forcing
racism), conflicting interests between social groups, changes in existing status hierarchies.
and the preference for social dominance, drive
resistance to affirmative action. All of these theories Justine E. Tinkler
differ in important ways, but they share the basic
notion that dominant groups oppose affirmative See also Civil Rights Legislation; Discrimination; Group
action because it threatens their privileges. These Position Theory; Justice; Racism; Sexism; Symbolic
theories provide the most purchase for understand-
ing why Whites and males resist affirmative action,
but the theories do not adequately address why Further Readings
groups that stand to benefit from affirmative action Bobo, L., & Kluegel, J. R. (1993). Opposition to race-
policies sometimes also oppose them. targeting: Self-interest, stratification ideology, or racial
Given the controversy surrounding affirmative attitudes? American Sociological Review, 58,
action and the widely publicized complaints about 443–464.
reverse discrimination, those who benefit from affir- Bowen, W. G., & Bok, D. (1998). The shape of the river:
mative action may be concerned about perceptions Long-term consequences of considering race in
that their success is not merit based. A number of college and university admissions. Princeton, NJ:
laboratory studies have provided support for this Princeton University Press.
Ageism 13

Crosby, F. J. (2004). Affirmative action is dead: Long live jokingly presented) is “I’m sorry to hear you’re
affirmative action. New Haven, CT: Yale University another year older” and that it is bad to get old.
Crosby, F. J., Iyer, A., & Sincharoen, S. (2006).
Understanding affirmative action. Annual Review of Forms of Ageism
Psychology, 57, 585–611. Ageism takes many forms. It affects how some
Leonard, J. S. (1990). The impact of affirmative action people speak to older adults. This language style,
regulation and equal employment law on black though grounded in good intentions, is experi-
employment. Journal of Economic Perspectives, 4,
enced by many older people as patronizing and
condescending. On the basis of stereotypes about
Regents of the University of California v. Bakke, 438 U.S.
the loss of cognitive abilities as we age, younger
265 (1978).
adults will often be overly polite, speak louder and
slower, exaggerate their intonation, speak in a
higher pitch, and use simple sentences. This “baby
talk” has been found to be the same type of speech
Ageism style that people use to talk to children, pets, and
even inanimate objects. Interestingly, some older
When we perceive other people, there are three adults don’t mind being spoken to in this way.
primary criteria upon which we automatically and Research has shown that older adults who are not
initially categorize them: race, gender, and age. functioning at a healthy level (physically, emotion-
This categorization process follows from the natu- ally, or cognitively) actually prefer this speech style
ral tendency of the mind to categorize objects in because it communicates a feeling that the younger
its environment to facilitate everyday cognition person perceives them as needing to be taken care
and action. The categorization of others on these of, and this dependency relationship is comforting
dimensions becomes so well learned that it is to the older adult. Older adults who are healthy,
automatic in social perception. Unfortunately, however, may find such speech styles offensive.
while categorizing people according to these char- Research has indicated that people have very
acteristics does indeed facilitate social cognition, it negative attitudes toward aging and adults over
is also the first step in stereotyping of, and preju- 55, and particularly over 75. However, when
dice toward, groups. While researchers have long asked about their attitudes toward their elderly
studied racism and sexism, they know compara- boss, or grandfather, or neighbor, respondents in
tively little about prejudice against someone based research studies have a positive attitude toward the
on their age, referred to as ageism. While it is cer- specific older adult. This led to confusion among
tainly true that people have prejudices and stereo- some early researchers who were not really sure if
types about virtually any age group, the vast prejudice against older adults existed. As it turns
majority of research on age prejudice has focused out, it does, and people have many different and
on the most common form of ageism: prejudice at times contradictory views of the older adult (for
toward older people, particular those over 74 example, sometimes as a “sage” or “perfect grand-
years of age. parent,” and sometimes as “impaired” or a
One reason that ageism has been underinvesti- “shrew” or “curmudgeon”). One contributing rea-
gated by researchers is that it is institutionalized son for this may be that we tend to think about
within American and many other Western cultures. stereotyped outgroups along two dimensions:
In other words, negative views of older people are warmth and competence. We tend to view older
very much a part of our everyday shared experi- people with whom we are familiar (such as family,
ences and lives, and older adults tend to buy into friends, coworkers) as warmer but less competent
the truth underlying the stereotypes, that those than other older people. Research indicates that
who experience ageism are not perceived as “vic- we regard other elderly people as cold and either
tims.” An example of how ageism is institutional- incompetent or competent.
ized can be found in greeting card stores. In the These divergent ways of treating older people
birthday card section, the basic message (though according to age stereotypes also can be linked to
14 Ageism

two different types of ageism. With benevolent age- that such a fate will befall them, and their anxiety
ism, the perceiver believes that older people need recedes.
help and are dependent, and that younger people Though more research is needed on the motiva-
have an obligation to care for older people. The tions behind age stereotypes and prejudice, this
motivation and attitude toward older people is theory has the most current empirical support and
kind, helpful, and positive. In contrast, malignant is highly regarded by many ageism researchers.
ageism rests on the belief that older people are
worthless, negative, and a burden on society. The
Internalization of Ageism
motivation and attitude of these perceivers are quite
negative and hostile toward older adults. These If a whole society is communicating to you that old
very different attitudes toward older people can age is bad, that it is something to be feared, that
lead to different perceptions of their warmth and your cognitive and physical abilities are declining
competence, and those perceptions, in turn, can with every day, and that your worth to society is
lead to very different beliefs, stereotypes about, and fairly low (because you are no longer working),
behavior toward elderly people. you, as an older person, may start to believe it.
This can have negative effects on self-concept and
self-esteem and may even influence a person’s lon-
Motivation for Ageism
gevity. One study found that older adults who had
A fundamental question is, “Why are people age- more positive self-perceptions of aging lived an
ist?” Older people are a unique group for prejudice average of 7.5 years longer than those with a more
researchers to study because, just by living long negative view of their aging. Interestingly, research
enough, these people move from the ingroup to a has shown that the self-esteem of older people is
stereotyped outgroup. Given this fact, why would not affected by ageism in society and age stereo-
a younger person be motivated to insult and deni- types and prejudice. In fact, some studies have
grate a group to which he or she will eventually shown the self-esteem of older adults to be double
belong? Though there are many potential, contrib- that of those of high school age. Again, if older
uting motivations, one theory has shown substan- adults believe that the ageist behavior of others is
tial and compelling empirical evidence supporting not prejudicial, but rather is merely communicat-
the idea that ageism is motivated by fear. Terror ing a societal, commonly understood “truth”
management theory suggests that culture and reli- about older adults, then older adults may not per-
gion are creations that impose meaning and order ceive anything negative about their ageist treat-
on our world. This order helps us keep at bay our ment by younger people.
feelings of fear about our mortality and the ran- This is an important point to discuss in a bit
dom nature of the universe. According to the the- more detail. Researchers have found utility in distin-
ory, as we go through childhood, we associate guishing between the “young–old” (ages 55–74)
good behavior with being rewarded and protected and the “old–old” (ages 75 and higher). Most of the
by our parents. This good feeling about ourselves, negative stereotypes about aging and older people
our self-esteem, therefore forms a buffer against are derived from our perceptions of the old–old.
fears of our eventual death. Research on terror These two groups of older adults react to ageist
management theory consistently has shown that treatment very differently. In a recent survey of over
when people are reminded about their mortality, 850 older adults, respondents were asked about
they feel more anxiety. Because older people their experiences with ageism and how it made
remind us of our mortality, we may avoid or even them feel. The young–old noted several incidents of
denigrate them to help us deny the possibility that ageist behavior directed at them, and it made them
we too will eventually get old (and die). Several very angry (because they do not think of themselves
studies therefore have shown that older people are as “old”). The old–old, however, were either unwill-
stereotyped and discriminated against by younger ing to admit they’d experienced ageism, or they just
people so they can cognitively distance themselves did not interpret that behavior as ageist (because, as
from their elders and blame them for their “sorry mentioned earlier, they perceived it as reflecting a
state” (being old). In so doing, young people deny true state of affairs—they were dependent and they
Allport, Gordon 15

were failing in their cognitive and physical abilities). becoming more ageist. Recent research suggests that
If they did mention they experienced ageism, they as Eastern cultures become more industrialized, and
said they were not bothered by it. more like the West—due to trade, tourism, and
increasing global connectedness—they may tend to
adopt more Western views of death, aging, and the
Pervasiveness of Ageism
role of the older person in society.
Research has found that ageism is so pervasive in
society that even those who work in helping profes-
Reducing Ageism
sions show ageist attitudes. In medical schools, little
training has typically been devoted to gerontology How then can ageism be reduced? From an early
or geriatrics, because it is not seen as an exciting age, children must learn that getting older does not
field in which to specialize. Older people are viewed mean one will eventually be a witch or a bad or
by some doctors as rigid, depressed, senile, or grumpy person (as most fairy tales suggest). Society
untreatable. Some doctors view treating older needs to educate children, employers, policymak-
patients as futile or as a waste of time, because they ers, and health care professionals about the perva-
are going to die soon anyway. Indeed, studies have siveness of ageism and how it has very real, harmful
shown that some doctors are less willing to pursue effects on older adults. Opportunities for older
expensive treatments and aggressive procedures or people to contribute to their community should be
therapies with older patients, and are more likely to created, and contact between younger and older
order pain medication to stabilize them until they people should be encouraged. Older people should
die. Other studies have shown that doctors regard be regarded with respect. In so doing, society will
the same disease (e.g., cancer) as a surprise and a enhance the quality of life for older adults and
tragedy in a 5-year-old but not in an older adult. enhance intergenerational interactions.
Researchers have referred to this as “healthism.”
Todd D. Nelson
Some mental health professionals may shy away
from accepting older clients because they view See also Discrimination; Perceived Group Variability;
older people as not really having serious problems, Prejudice; Stereotyping; Stigma; Terror Management
but rather as just feeling lonely and wanting to talk Theory
to someone. On a positive note, these biases in the
medical and psychological professions are indeed
Further Readings
changing as increasing attention is devoted to train-
ing doctors and psychologists in gerontology, in Bugental, D. B., & Hehman, J. A. (2007). Ageism: A
response to the growing demand for such training review of research and policy implications. Social
brought about by the retiring baby boomers. Issues and Policy Review, 1, 173–216.
At the extreme, malignant ageism can result in Martens, A., Godenberg, J. L., & Greenberg, J. (2005). A
exploitation, neglect, or abuse of older adults, and terror management perspective on ageism. Journal of
even in violent behavior toward them that leads to Social Issues, 61(2), 223–240.
their injury or death. Unfortunately, this type of Nelson, T. D. (2002). Ageism: Stereotyping and prejudice
abuse is on the rise, and it tends to be overlooked against older persons. Cambridge: MIT Press.
Nelson, T. D. (Ed.). (2005). Ageism: Prejudice against
because (1) physicians have, until only fairly
our feared future self. Journal of Social Issues, 61(2),
recently, been less acquainted with this form of
abuse and (2) the elderly victim is too embar-
rassed or afraid to report it. Elder abuse is not
restricted to the United States, as researchers have
uncovered such abuse in Japan, Puerto Rico, and
other cultures.
Allport, Gordon
Though ageism is most prevalent in the United (1897–1967)
States (with exceptions such as traditional Hawaiians,
who revere their elders) and other Western nations, Gordon Willard Allport is renowned for his work
other countries around the world are increasingly on the psychology of prejudice and his formulation
16 Allport, Gordon

of the highly influential contact hypothesis. His The Nature of Prejudice

work pioneered a focus on the cognitive anteced-
Allport’s 1954 book, The Nature of Prejudice,
ents of prejudice; it demonstrated how social psy-
contains his most influential theoretical contribu-
chological research can address important social
tion to social psychology. Focusing on intergroup
issues and have a tangible impact on policy and
conflict and in particular on interracial relations in
practice. This entry looks at his life and works.
the United States, the book provides a broad per-
Allport was born in Montezuma, Indiana, in
spective on defining, explaining, and reducing
1897. His father was a physician, and his mother
was a former schoolteacher. When he was 6 years
Allport was one of the first theorists to focus on
old, his family moved near Cleveland, Ohio, where
the cognitive antecedents and processes that con-
he spent all his school years. Allport had three
tribute to expressions of prejudice. He argued that
older brothers, one of whom, Floyd, was also a
stereotyping and categorization per se are func-
social psychologist and contributed to the estab-
tional aspects of people’s thinking processes, but
lishment of modern experimental social psychol-
that when combined with social inequalities they
ogy. Gordon Allport completed his bachelor’s
can propagate biased attitudes and evaluations of
degree at Harvard in 1919, and he then spent a
others. Thus, through social comparison with out-
year teaching English and sociology at Robert
groups, people locate themselves and their group
College, in Istanbul, Turkey. Returning to Harvard,
in the world. Allport’s analysis suggests that
he was awarded a PhD in 1922; his doctoral dis-
although the cognitive mechanisms involved in
sertation was entitled “An Experimental Study of
social categorization and stereotyping may some-
the Traits of Personality: With Special Reference to
times lead to negative intergroup attitudes, this is
the Problem of Social Diagnosis.”
not inevitably the case.
Following completion of his PhD, Allport was
This led to the important observation that if
awarded a fellowship to study in Europe. He spent
more general psychological processes relating to
one year in Berlin and Hamburg, Germany, where
categorization are involved in the formation of
he was introduced to the Gestalt theory of mind.
negative intergroup attitudes, then encouraging
He then spent a year in England at Cambridge
people to shift their conceptualizations of group
University before returning to United States and
membership from strictly defined criteria, such as
eventually becoming a faculty member of Harvard
race, to more inclusive categories, like common
University from 1930 to his death in 1967. Allport
humanity, may weaken antagonistic relations and
also served as president of the American Psycho­
prejudice between ethnic groups.
logical Association.
The Contact Hypothesis
Allport’s Work
A milestone theoretical contribution of The
Opposed to the strictly one-sided psychoanalytic Nature of Prejudice was Allport’s formulation of
and behaviorist approaches to the study of personal- the contact hypothesis. Allport considered whether
ity, Allport emphasized the uniqueness of the indi- simply bringing together members of groups that
vidual and argued that problems need to be treated differ in terms of race, religion, or national origin
in terms of present circumstances instead of child- could reduce stereotyping and prejudice. He
hood experiences. His theoretical views on person- argued that, in many cases, contact on its own
ality resulted in two books, Personality: A might not be sufficient to improve intergroup
Psychological Interpretation (1937) and Pattern attitudes. Rather, there are prerequisite situa-
and Growth in Personality (1961). In The Individual tional conditions that enable intergroup contact
and His Religion (1950), Allport discussed the experiences to result in positive attitude change.
development of religious attitudes and ideologies, as The “four necessary conditions” Allport identi-
well as the relationship between religion and inter- fied were equal status during contact, the exis-
group attitudes and behavior. This work led to what tence of common goals, cooperation in achieving
is perhaps Allport’s most important contribution to such goals, and institutional support (e.g., laws,
social psychology. authorities, customs).
Ambivalent Sexism 17

Over the past 50 years, a great deal of research Pettigrew, T. F. (1999). Gordon Willard Allport: A
has been devoted to testing and amending the basic tribute. Journal of Social Issues, 55, 415–427.
principles of the contact hypothesis, and contact is Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic
now one of the most widely used psychological test of intergroup contact theory. Journal of
interventions for reducing prejudice and improv- Personality and Social Psychology, 90, 751–783.
ing intergroup relations. Much of this research
initially focused on extending Allport’s four key
conditions for positive contact outcomes, leading
some to suggest that the approach had too many Ambivalent Sexism
such conditions to prove workable.
A recent meta-analysis conducted by Thomas Ambivalent sexism, a subtle but effective method
Pettigrew and Linda Tropp directly addressed this of keeping the gender equality gap from shrink-
criticism. The meta-analysis identified 515 studies ing, contains two complementary belief systems
conducted between 1949 and 2000 with 713 sam- about women that have the contrasting valences
ples and a total of 38 participating nations. The of subjective benevolence and hostility. Benevolent
result of this meta-analysis was a robust and statis- sexism masks the more overt hostile sexism by
tically significant negative effect of contact on giving seemingly caring reasons for discriminatory
prejudice, an effect that remained even for contact behaviors toward women. Thus, ambivalent sex-
that did not meet any of Allport’s initial four condi- ism can be a difficult prejudice to root out.
tions. Contemporary research on contact has, Historical conceptions of sexism assume the
therefore, begun to examine other issues, including hostile belief that women are inferior to men and
which forms of contact best reduce prejudice. For unfit for positions of leadership, especially those
instance, researchers have found that a unique involving power over men. In this view, women
form of contact, cross-group friendship, is more who adhere to traditional roles are undervalued
effective at improving outgroup attitudes than less and viewed with contempt, while those who chal-
intimate forms of contact. They also have discov- lenge such ascribed codes of behavior are resented
ered that indirect forms of contact, where contact as overstepping natural and cultural boundaries.
is experienced vicariously through others or through While this notion of sexism has prevailed for a long
simply imagining a positive outgroup encounter, period of time, a more recent conceptualization
can have a positive effect on outgroup attitudes. reveals that traditional beliefs about women may
More than 50 years after the first publication of be more complicated than previously assumed.
The Nature of Prejudice, its core ideas continue to Rather than depicting women in only openly
inspire and guide scholars and policymakers focused hostile ways, more recent depictions show that
on the assessment, explanation, and attenuation of most people (both men and women alike) tend to
prejudice, and this is Allport’s enduring legacy. hold dual conceptions about women: benevolent
and hostile sexism. Acting together, these aspects
Richard J. Crisp and Sofia Stathi
of ambivalent sexism reward women for avoiding
situations that make them seem nonfeminine and
See also Discrimination; Intergroup Contact Theory; for choosing situations that make them seem femi-
nine. Or, as Glick and Fiske have described, the
two components act as “carrot and stick” to
Further Readings encourage women to “remain in their place.”
Allport, G. W. (1954). The nature of prejudice. Garden Women receive rewards (i.e., the carrot) when they
City, NY: Doubleday. follow the rules, but are punished (i.e., receive the
Dovidio, J. F., Glick, P. G., & Rudman, L. (Eds.). (2005). stick) when they do not.
On the nature of prejudice: Fifty years after Allport.
Malden, MA: Blackwell.
Pettigrew, T. F. (1998). Intergroup contact: Theory,
research and new perspectives. Annual Review of Hostile sexism is the belief that women are by
Psychology, 49, 65–85. nature inferior and thus unfit for and incapable of
18 Ambivalent Sexism

holding positions of authority, especially over system. Benevolent sexism is patronizing care
men. Hostile sexists tend to see a power struggle taken of someone assumed to be unable to make
between the sexes and express resentment of decisions or act as an independent adult; it is
women for manipulating men—whether by blud- related to paternalism and the presumed benevo-
geoning men with “feminist demands” or control- lent authority parents exercise over their chil-
ling men through sexual seduction. Accordingly, dren. This implicitly treats women as children,
hostile sexists (both men and women) may experi- advocating that men be guardians of women’s
ence anger toward “feminist” women who chal- minds and bodies, exerting a protective influence
lenge prescribed gender roles and/or shirk “their over women because of their alleged vulnerabil-
moral and biological duty” of acting as a subordi- ity. In a complementary fashion, hostile sexism
nate to a presumably stronger male counterpart. reinforces patriarchal assumptions that men
Because laws, organizational policies, and norms should be in charge. However, hostile sexism
of social desirability often serve to protect women, more directly asserts women’s presumed inferior-
expressions of hostile sexism may have diminished ity (e.g., viewing them as too emotional to
in recent years. lead).
Benevolent sexism also works against the pro- A second component reflects biological and
motion of women as equals, but in a very different social gender differentiation. Men’s physical
way. This construct idealizes women as mothers, power is often equated with social power. Sex is
wives, and caregivers. In addition, benevolent sex- a fundamental biological and social category
ism assumes that women both have a purity that that tends to foster sharp social distinctions in
men do not and also need protection, as they are most societies (e.g., gender stereotypes), form-
too weak and good to defend themselves against ing the basis for a division of labor. Women
those who might otherwise do them wrong. Rather (due to their greater biological ties to reproduc-
than lowering the status of women by directly tion) are associated with nurturing and domes-
characterizing them as less competent, benevolent tic life, whereas men are associated with more
sexism subtly reinforces the idea that women are powerful societal roles and leadership positions.
more fragile and should be protected and provided These roles reinforce both benevolent sexism
for by men. In return, women are expected to con- (e.g., viewing women as warm and expressive—
fine themselves to a social sphere in which they can traits linked to the nurturing role) and hostile
nurture the next generation, serve as counterparts sexism (e.g., viewing women as less competent
to their adoring husbands, and create comfortable because they less often occupy leadership
homes. roles).
Hostile and benevolent sexism work to balance A third component of ambivalent sexism is het-
each other and together function (more effectively erosexuality, the premise that both sexes need a
than hostility would alone) to relegate women to a heterosexual romantic relationship to be fulfilled.
second-class status. Benevolent sexism may be Sexuality affords women a dual role. On one hand,
more palatable to most people (especially women) women may be viewed as good wives and mothers,
than hostile sexism because it appears to reflect or agents and targets of intimacy and affection
good intentions rather than antagonism. That is, (i.e., benevolence). On the other hand, women
women who allow themselves to be patronized may be viewed as seductresses, using their sexual
reap some benefits and earn the adoration of their power to take control over men and attempting to
male protectors, while women who do not con- emasculate them (i.e., hostile). The presumption
form to the model are subjected to the negative that women use sex as a tool by which to control
consequences provided by hostile sexism, includ- men elicits hostile resentment and attitudes that
ing censure, hate, and resentment. sexually demean women. However, because sex is
rewarding and fosters emotional intimacy, subjec-
tively benevolent views romanticize women (e.g.,
as fair and pure princesses). Thus, ambivalent sex-
Three structural foundations underlie ambivalent ism encourages polarized categorizations of some
sexist beliefs. The first is acceptance of a patriarchal women as “sluts” and others as “angels.”
Anticonformity 19

Measuring Ambivalent Sexism potential limits everyone in society. When women

are forced out of the workforce by prejudice,
The most commonly used measure of ambivalent
beaten in their homes for expressing ideas contrary
sexism is the Ambivalent Sexism Inventory, which
to those of their husbands, and treated as second-
separately assesses each component of sexism. The
class citizens on the basis of gender, an environ-
measure was created by Glick and Fiske in 1996,
ment is created where neither women nor men can
and since then it has been widely validated in more
thrive. While the process of change involves a cul-
than 25 countries (e.g., the U.S., Turkey, Brazil,
tural shift in attitudes toward women, recognizing
the Netherlands) with samples ranging from 200
the consequences of sexist behavior is an impor-
to 2,000. The measure goes beyond previous
tant first step in achieving equality. Ambivalent
boundaries by examining sexism as more than the
sexism complicates the prospects for exposing the
traditional, hostility-based view in which sexist
negative effects of sexism because of its subjec-
behaviors are solely motivated by a dislike of
tively positive component, which leads many peo-
women. Thus, an example of benevolent sexism is:
ple to view sexism as not as bad as other forms of
“Many women have a quality of purity that few
prejudice. Specifically, women are more accepting
men possess.” And an example of hostile sexism is:
of benevolent sexism (due to its apparent favorabil-
“Women seek to gain power by getting control
ity toward women) and, in turn, are more willing
over men.” The ambivalent nature of sexism can
to accept hostile sexism because it is “softened” by
be seen in the duality of women’s roles expressed
benevolent sexism.
here. Women are described as both pure beings
and power-hungry creatures. Michelle Hebl and Katharine Ridgway O’Brien
Research using the measure has shown that the
constructs of benevolent sexism and hostile sexism See also Discrimination; Leadership; Prejudice; Racism;
are positively correlated with each other (correla- Sexism
tions range across samples from .37 to .74), and
tend to be more highly correlated in women than in Further Readings
men. Hostile sexism is also correlated with other
measures, such as “protestant work ethic” and Glick, P., & Fiske, S. T. (1996). The ambivalent sexism
“social dominance,” while benevolent sexism has inventory: Differentiating hostile and benevolent
been correlated with “right wing authoritarian- sexism. Journal of Personality and Social Psychology,
70, 491–512.
ism.” When hostile sexism is statistically controlled,
Glick, P., & Fiske, S. T. (2001). An ambivalent alliance:
benevolent sexism is often no longer a significant
Hostile and benevolent sexism as complementary
predictor for constructs like modern sexism or tra-
justifications for gender inequality. American
ditional sexism. Such an effect may be taken as
Psychologist, 56, 109–118.
further evidence that benevolent sexism is a unique
construct that gives sexist behaviors a protectionist
aura. In these studies, men tend to score higher Anticonformity
than women on the hostile components of sexism,
but there are rarely gender differences on the
Anticonformity refers to behavior that is deliber-
benevolent sexism component, with both groups at
ately designed to go against the position advocated
least partially endorsing this behavior.
by one or more others. Also known as counterfor-
mity, anticonformity most typically occurs in
group settings when an individual rebels against
the dominant or majority opinion. It conjures up
The consequences of ambivalent sexism can be the image of the maverick or deviant who pur-
severe. As with most forms of subtle discrimina- posefully disagrees publicly with the positions of
tion, the slow buildup of unfair treatment over others in the group, even when he or she agrees
time exacerbates the impact of any one sexist inter- privately with these same positions.
action. Any belief system that systematically keeps Anticonformity stands in contrast to two other
one group from living and working at its full important types of response to social pressure:
20 Anticonformity

conformity and independence. Conformity and and social influence was measured by the change
anticonformity are essentially opposites. Whereas in participants’ ratings toward or away from the
with conformity, an individual is motivated to rating given by the “partner.” Argyle found that
cooperate, follow, and fit in with the group, with most participants, 58%, were uninfluenced by
anticonformity, he or she is motivated to disagree their partner; they showed independence by stick-
with, disrupt, and oppose the group. In research ing with their original opinion. Another 35%
settings, conformity is usually measured by move- showed conformity by moving toward the part-
ment toward the majority opinion, and anticon- ner’s position. The remaining 8% of participants,
formity by movement away from it. Independence however, showed anticonformity; they became
can look like anticonformity, but its motives are even more extreme in their disagreement with their
different. With independence, the goal is simply partner.
to be true to one’s self, regardless of how one’s In the early 1960s, Richard Crutchfield and
views might be received by others. The term non- Richard Willis published the earliest theoretical
conformity encompasses both anticonformity and work on the distinction between anticonformity,
independence. conformity, and independence. Working indepen-
A real-life example of anticonformity was widely dently, both proposed that although anticonfor-
reported in the American media during the mity and conformity are opposites in terms of
Christmas season in 1994. A man from Little underlying motives and measurement, they are,
Rock, Arkansas, was served with a court injunc- ironically, quite similar conceptually in that both
tion, raised at the behest of his neighbors, ordering are determined by the group’s position. Thus, both
him to remove some of the over 3 million lights are properly regarded as forms of dependent
from the Christmas display at his home. The dis- behavior. Both stand in contrast to independence,
play was attracting too many sightseers and too therefore, where the individual is not influenced
much traffic to his exclusive residential neighbor- one way or the other by social forces. Crutchfield
hood for his neighbors’ liking. The man could have and Willis concluded that anticonformity, confor-
conformed to his neighbors by removing some of mity, and independence should not be conceptual-
his lights. He could have shown independence by ized and measured merely by different degrees of
neither adding nor removing any lights. Instead, positive or negative movement along a single-
the man chose to anticonform. Soon after the dimension line segment, the standard practice of
injunction was served, he defied it by increasing Argyle and other early researchers. Rather, the
the number of lights in his display. This entry three responses should be seen as falling at the
examines the origins of anticonformity, describes vertices of a triangle.
some theories, and offers a few examples.
Theories of Anticonformity
Background and History
Social scientists have proposed a number of motives
In 1957, British psychologist Michael Argyle pub- that attempt to explain why anticonformity may
lished what was probably the first study to demon- occur in certain situations. One motive, first for-
strate anticonformity under controlled conditions. mally identified by Jack Brehm, is known as psy-
Argyle asked male students to evaluate a painting chological reactance. It is based on an individual’s
by Marc Chagall, Poète Allongé, which was cho- perceived rights and freedoms. When people are
sen deliberately because of its “unusual and members of a group, they can come to believe that
ambiguous character.” Participants were told that their rights as individuals are being eliminated or
they were working with a student partner, when in threatened with elimination. Under such condi-
fact their partner was Argyle’s confederate. In one tions, Brehm proposed, people may react by taking
condition, each student learned that his opinion of steps to restore their freedom.
the painting had been rejected by his partner (e.g., One clear way to reclaim a freedom is to do the
“What you say is trivial, for the picture is so mean- opposite of what the source of the threat suggests;
ingful as a whole”). Participants were then given that is, to anticonform. So if people in a neighbor-
an opportunity to rate the painting a second time, hood group say to one of their members, “Surely,
Anticonformity 21

you must agree with us that we have too much and opposition to the Nazis, in perhaps the only
traffic at night in our neighborhood. You need to way available to him under the circumstances, his
take down some of your Christmas lights,” the offer reflects anticonformity.
target individual might respond in words or deeds, This example is important because it illustrates
“Who gave you the right to tell me what to do on that although anticonformity is usually measured
my own property? Actually, I think my display by movement away from a group’s position, it
would benefit by adding even more lights.” Another can sometimes be indicated, ironically, by move-
example of anticonformity consistent with reac- ment toward the group, provided that such
tance motivation is known as the Romeo and Juliet movement is excessive. This brand of anticonfor-
effect. As in Shakespeare’s tragic drama, attempts mity was first identified by Willis and dubbed
by parents in Western cultures to restrict their overconformity.
teenagers’ freedom to date may backfire, leading
to increased dating. The Anticonformist
Other recognized motives for anticonformity
include the desires to (a) promote change and Anticonformity refers to a type of behavior. Yet, a
innovation; (b) establish or project one’s indi- person who consistently engages in anticonformity
viduality or uniqueness; (c) avoid bad group deci- across time and settings can be regarded as a type
sions (i.e., groupthink); (d) avoid the appearance of person—the anticonformist. Most evidence sup-
of sycophancy; (e) disconfirm another’s negative porting the existence of anticonformists is anec-
expectations regarding one’s skills, attributes, or dotal. Nevertheless, there have been a few
abilities; and (f) distance oneself or group from systematic attempts to identify the characteristics
dissimilar, disliked, or unattractive others. and etiology of anticonformists. One provocative
An example of this distancing occurred in the account was offered by historian of science, Frank
1930s following the rise of Nazism. Prior to Sulloway. Based on archival records, Sulloway
World War II, a type of swastika had been the found significant evidence that innovators, icono-
official insignia of the U.S. Army’s 45th Infantry clasts, and rebels in the history of science, religion,
Division. The swastika of the 45th was chosen and politics tend to be later-borns.
initially because it had been an ancient cosmic and To explain these findings, Sulloway proposed
religious symbol in many cultures (e.g., Navajo that because of firstborns’ typical role as surrogate
Indian culture). After a mirror-image swastika parents, and through the normal process of sibling
was adopted by Hitler and the Nazis in 1935, competition for parental attention, firstborns gen-
however, ranking officers in the 45th Division felt erally identify with their parents. Firstborns, there-
obliged to change their insignia in order to dissoci- fore, are predisposed to conformity and
ate the 45th from anything related to Nazism. The conventionality. Later-born children, in contrast,
division’s swastika was replaced in 1939 by a are outsiders to an established group from birth—
thunderbird. their parents and older siblings. Thus, they are
primed to rebel against the establishment, particu-
larly against the seemingly arbitrary authority that
Anticonformity by Overconformity is typically exerted over them by elder siblings, and
Another probable example of anticonformity based hence tend toward anticonformity.
on disassociation from the Nazis was demon- The primary force that drives change in history,
strated by Freud following the Nazi annexation of therefore, is not located between families divided
Austria in 1938. According to a biographer, the by social class, as Marx proposed. Rather,
82-year-old Freud was allowed to emigrate to Sulloway argued, it is located within families
England, but only after he had signed an affidavit divided by birth order, a function of small-group
stating that he had been under no pressure from dynamics.
Nazi authorities. After signing, Freud offered to add Paul R. Nail
“I can recommend the Gestapo to anyone,” but his
offer was turned down. Given that the intent of See also Conformity; Groupthink; Innovation; Minority
Freud’s offer was to register his disagreement with Influence; Self-Fulfilling Prophecy; Social Deviance
22 Anti-Semitism

Further Readings of 1096, the expulsion of Jews from England in

1290, the Spanish Inquisition, the expulsion from
Argyle, M. (1957). Social pressure in public and private
Spain in 1492 and from Portugal in 1497, and the
situations. Journal of Abnormal and Social
Psychology, 54, 172–175.
Holocaust of Nazi Germany.
Brehm, J. W. (1966). A theory of psychological reactance. Anti-Semitism seems to be prototypical of a
New York: Academic Press. number of prejudices. It encompasses nearly every
Crutchfield, R. S. (1962). Conformity and creative aspect of prejudice toward an outgroup. Anti-
thinking. In H. E. Gruber, G. Terrell, & Semitism has many individual-level facets, ranging
M. Wertheimer (Eds.), Contemporary approaches to from subtle anti-Semitic stereotyping and antipa-
creative thinking (pp. 120–140). New York: Atherton. thies to blatant expressions of anti-Semitic racism
Nail, P. R., MacDonald, G., & Levy, D. A. (2000). and discrimination. In many societies, the collec-
Proposal of a four-dimensional model of social tive memory has retained anti-Semitic racial stere-
response. Psychological Bulletin, 126, 454–470. otypes (Jewish character or appearance), religious
Sulloway, F. J. (1996). Born to rebel: Birth order, family stereotypes (Anti-Christ), secular stereotypes (prof-
dynamics, and creative lives. New York: Pantheon. iteer), and political anti-Semitic stereotypes (Jewish
Willis, R. H. (1963). Two dimensions of conformity- conspiracy). Some modern expressions of anti-
nonconformity. Sociometry, 26, 499–513. Semitism, which have been a subject of contro-
versy since the 1990s, are Islamist anti-Semitism
and a critique of Israeli policies that is fed by anti-
Semitic prejudices.
Anti-Semitism is prejudice toward Jews and Jewish
Psychological Foundations
culture. From a social psychological perspective, it Stereotypes and images of the “collective Jew” are
is a devaluation of the group of Jews and their very persistent. Cultures transport and transfer
culture or a devaluation of a Jewish person, anti-Semitic stereotypes and myths through social
because she or he is a member of the social cate- representations that are part of the collective
gory. A common definition refers to anti-Semitism memory. The social psychology of anti-Semitism
as hostile beliefs—expressed by attitudes, myths, focuses on individual causes within social contexts,
ideology, folklore and imagery, discrimination, reaching from prejudices to genocide.
and violence—which destroy the worth of Jews Early psychodynamic approaches attribute its
and Jewish culture. In its most comprehensive causes to intrapsychic crises and conflicts, resulting,
sense, it is hostility toward Jews as “Jews,” and for example, from feelings of guilt or infirm ego
thus an expression of devaluation and of inequal- strengths. The Frankfurt School’s project on the
ity between groups. Anti-Semitism is expressed by authoritarian personality by Theodore Adorno and
individuals, groups, or institutions against Jewish his colleagues had a particularly significant influ-
people, groups, or culture through the categoriza- ence on research on anti-Semitism. This personality
tion of Jews as negatively different or Jewish cul- approach refers to psychodynamic processes and
ture as strange. Jews are seen not as individuals explains anti-Semitism by reference to the individ-
but as a collective that brings problems to a com- ual trait of an authoritarian personality, which is
munity, often in a secret way. developed through punitive socialization and char-
The group-focused nature of anti-Semitism acterized by obedience. The researchers believed
links it to other expressions of prejudice, such as that this personality primes individuals to be per-
anti-immigrant prejudice, prejudice against suaded by propaganda and anti-Semitism. Several
Muslims, and sexism, all within a syndrome of studies have shown that authoritarianism predicts
group-focused enmity. The special importance of anti-Semitism. Current studies also demonstrate
anti-Semitism is derived from two features. First, that dominance-orientated people are prone to be
anti-Semitism has occurred worldwide for centu- anti-Semitic. Social dominance theory criticizes the
ries. Second, its most destructive expression has psychodynamic approach of the Frankfurt School.
been reflected in persecution: the German Crusade Authoritarianism is a pathological condition that
Anti-Semitism 23

does not explain institutional behavior and ideo- to a “group of regulars.” People gain
logical processes in society. Prejudices are legiti- recognition by others through expressing the
mizing ideologies for social hierarchies between normatively “correct” opinion of the group.
groups within a society, and people who are high
4. Anti-Semitism fulfills a knowledge function. It
in social dominance orientation are motivated to
explains what is going on, and why things
keep groups like Jews in lower status positions.
happen. For example, a belief in Jewish
The social identity approach taken by Henri
conspiracy explains why some groups suffer.
Tajfel and his colleagues gives a clearer picture of
Myths about the Zionist threat and the
the link between individual and contextual causes
conspiracy of Judaism that wants to rule the
of prejudices. From this perspective, anti-Semitism
world are extreme examples. These functions
is explained as a group-focused devaluation in the
are fed by anti-Semitic stereotypes, which bind
context of intergroup relations. The social-cognitive
these beliefs together and relate them to other
categorization of Jews and Jewish culture as an
outgroup is thought to be responsible for the devel-
opment of some anti-Semitic sentiments. Primary 5. Anti-Semitism may alleviate feelings of guilt
reference groups (ingroups), which define the social about matters of historical fact. An often-
identity of an individual group member, communi- reported expression of anti-Semitic sentiments is
cate anti-Semitism. Members of an ingroup differ- blaming the victims.
entiate themselves from Jews and Jewish culture
and demand conformity to the ingroup’s norms In addition to such cultural and individual causes,
and ideologies. The ingroup’s social-cognitive char- several other contextual factors permit or promote
acterization of Jews as an outgroup is thought to be anti-Semitism. Studies show that stereotypical
responsible for the development of anti-Semitic media presentations of Jews and Jewish culture
beliefs and attitudes. Anti-Semitic sentiments in over time, a denial of collective anti-Semitism by
extremist groups clearly show this dynamic, but it political and cultural elites, and lack of contact
can also be detected in anti-Semitism of peer groups and experiences with Jews and Jewish culture also
or familial socialization. are critical causes of anti-Semitism. The power of
The satisfaction of several overlapping needs the old anti-Semitism can be evoked by those who
and motives by anti-Semitism is linked to its group- rely on threats to groups and sentiments that can
focused nature. Five such needs and motives are: be linked to stereotypes kept in the historical
memory. Studies in Europe show that anti-Semitism
1. Anti-Semitism functions to reinforce self-esteem is a regular part of right wing populism, together
derived from group membership. Anti-Semitism with xenophobia and authoritarian orientations.
can strengthen social identities, such as those These attitudes are especially exerted by populists
defined by the ideology of a homogeneous who make use of freedom of speech and rely on
nation. Difference and differentiation can have the assumption that the majority feels and thinks
a detrimental effect on self-esteem through the same way. Right wing populists frequently
social identity processes, and may trigger challenge and break laws against anti-Semitism,
prejudices—especially if social identities are felt and anti-Semitic racism is a core element of right
to be threatened. Anti-Semitic stereotypes of wing extremism. However, groups and individuals
Jewish conspiracies keep such threats alive. who are aware of the norm against anti-Semitism
also sometimes fall back on anti-Semitic stereo-
2. Anti-Semitism fulfills the function of
types, for example, in the manner in which they
legitimizing devaluation of those who compete
criticize Israeli policies.
or are perceived to compete with the ingroup,
contributing to the suppression of outgroups
and the superiority of the ingroup. Racist New Anti-Semitism
images of Jews serve this function.
Current research on anti-Semitism is characterized,
3. Anti-Semitism can bind individuals to groups in particular, by controversies about the difference
and their opinions. Anti-Semitism links people between the old and new forms of anti-Semitism.
24 Anti-Semitism

The old or classical anti-Semitism is an overt The German study group on “group-focused
devaluation of Jews that refers to negative, racist enmity” proposes that the presence of one or more
stereotypes (e.g., racist images or stereotypes asso- of the following four criteria indicates that a c�����
ciated with the Anti-Christ or Devil) and is often cism of Israel may be considered anti-Semitic:
tabooed and outlawed. The new anti-Semitism is
based on traditional anti-Semitic stereotypes that 1. The denial of the right of Israel to exist and the
are expressed in claims about current societal right of its self-defense (i.e., anti-Zionism);
events such as a worldwide Jewish conspiracy or
2. a historical comparison between Israeli policy
Israeli terror, which is interpreted to be Jewish. For
concerning Palestine and the persecution of Jews
example, the new anti-Semitism is represented by
in Nazi Germany;
a specific Islamist, anti-Zionist anti-Semitism
founded on myths of conspiracy. Several studies 3. the evaluation of Israeli policy with double
show that since 2000 (the Second Intifada), anti- standards (i.e., political measures are criticized
Semitism has been on the increase in Muslim com- in Israel but not in other countries); and
munities. Large parts of this Islamist anti-Semitism
4. the transference of anti-Semitic stereotypes to
are being justified by claims that Muslims are vic-
Israel and, in turn, the transformation of Israel
tims of Israeli policies, which are represented and
into the myth of “the collective Jew.”
mythologized as Jewish. Another facet of the new
anti-Semitism is the “secondary anti-Semitism,”
occurring in Germany and other European coun- If criticism of Israel does not meet any of these
tries, that involves denying historical anti-Semitic criteria, it is not considered anti-Semitic. Criticism
events, such as the genocide at Auschwitz, and of Israeli policies in Palestine is possible with-
demanding a Schlussstrich (“final closure”) to the out anti-Semitic sentiment, but analyses show
history of the Holocaust. Other facets of the new that it seems to be very difficult to criticize Israel
anti-Semitism are the positions that Jews benefited without referring to one of these components of
from exploiting their suffering during the Holocaust anti-Semitism.
and that other communities suffered more from
World War II than the Jews did. These arguments
entail denying the persecution of the Jews and their
status as victims. Polls show that this secondary The group-focused enmity criteria mentioned ear-
anti-Semitism is increasingly spreading into the lier give a basis for detecting new expressions of
mainstream of many civil societies. anti-Semitism from a nonideological point of view.
Often anti-Semitism is hidden by a critique of Unfortunately, the discourse on anti-Semitism has
Israel. This new expression of anti-Semitism is always been charged by ideological positions. This
found in right wing populism, Islamist propa- partly explains why current surveys show that it is
ganda, and sometimes left wing ideologies. Israeli difficult for people to speak about Jews and Jewish
policies against Palestinians are sometimes defined culture without referring to stereotypes. In many
as “Jewish” and are thus attributed to religious societies, such anti-Jewish sentiments are misused
rather than nationalistic causes. This anti-Semitic for propaganda. In Europe, anti-Semitism has
critique is linked to two other themes: first, a com- become a critical part of right wing populism.
parison of Israeli policies to the crimes of the Nazis Additional elements tied to anti-Semitism are
in the Third Reich; and second, a separatist ideol- anti-immigrant prejudices and authoritarian orienta-
ogy categorizing Jews as a strange community that tions, which are often precursors of attacks on Jews,
is not part of society. A topic causing serious dis- synagogues, and Jewish schools. In many European
putes and ideological debate is the question of cities, Jewish buildings still have to be protected by
which criticisms of Israeli politics represent anti- police. Also, innumerable efforts are being made to
Semitism. For example, some argue that any criti- combat traditional and modern anti-Semitism.
cism of Israel represents anti-Semitism, whereas Above all, programs focus on the education of
others claim that virtually no criticism of Israel has schoolchildren and young adults, but anti-Semitism
anti-Semitic roots. is still prevalent among elderly people. However,
Apartheid 25

although many organizations and countries sup-

port campaigns against anti-Semitism, evidence Apartheid
about the effectiveness of these approaches is rare.
Social psychological research offers evidence Apartheid is an Afrikaner word that means “sep-
showing that actions that promote positive inter- arateness” or “apartness.” It represents a cluster
group contacts and self-esteem, lower intergroup of policies that were designed to achieve “total
threats, and strengthen empathy and perspective separation” between races in South Africa, the
taking can reduce prejudices like anti-Semitism. effect of which was to preserve the economic and
However, evidence is needed to establish the effec- political privilege of the White minority. The
tiveness of such interventions specifically with application of apartheid led to a vast program of
anti-Semitism. In addition to analysis of interven- social engineering that lent constitutional legiti-
tionist approaches, more substantive research on macy to the subjugation of the non-White major-
other aspects of anti-Semitism is needed. Although ity. In this entry, the theory, practice, demise, and
many scientists agree that anti-Semitism still exists legacy of apartheid will be discussed, with a focus
and poses a severe threat to democracy, some fun- on its effects on intergroup thoughts, feelings, and
damental questions have to be answered. For behaviors.
example, rising anti-Semitism in Eastern Europe
has to be explained more exactly. The rise and dis-
Historical and Theoretical Context
semination of anti-Semitic stereotypes, anti-
Semitism in elderly people, Islamist anti-Semitism, Apartheid in South Africa cannot be understood
the anti-Semitism of elites, and many more phe- without being placed in its historical context.
nomena need to be understood. And over and For centuries, descendents of the Dutch settlers
over again, we have to explain the unexplainable: (the Afrikaners) coexisted uneasily with native
Auschwitz. African tribes who were being displaced by
Afrikaner territorial expansion. Afrikaners also
Andreas Zick found themselves increasingly in competition
with the British, who began to assume political
See also Authoritarian Personality; Dehumanization/
Infrahumanization; Genocide; Prejudice; Racism;
and economic control of much of southern
Scapegoating; Social Dominance Theory; Social Africa. The tension between the two imperial
Identity Theory forces reached a head during the Boer Wars,
which entrenched British influence and extin-
guished the political independence of the
Further Readings Afrikaner republics. The period after this defeat
was marked by the growth of a distinct Afrikaner
Fein, H. (Ed.). (1987). The persisting question:
identity, which gradually reasserted itself cul-
Sociological perspectives and social contexts in
turally, linguistically, and politically under
modern anti-Semitism. Berlin: de Gruyter.
British rule.
Laqueur, W. (2006). The changing face of anti-Semitism:
As South Africa became increasingly urban-
From ancient times to the present day. Oxford, UK:
Oxford University Press.
ized, Afrikaners began drifting into the cities,
Newman, L. S., & Erber, R. (Eds.). (2002).
where they perceived themselves to be the vic-
Understanding genocide: The social psychology of the tims of British racism and cultural imperialism.
Holocaust. Oxford, UK: Oxford University Press. A new class of urban Afrikaner poor emerged
Strauss, H. A., & Bergmann, W. (Eds.). (1987). Error that had to compete with cheap labor from Black
without trial: Psychological research on anti-Semitism. migrants. Traditional racial hierarchies were
Berlin: de Gruyter. realigning around class, and many poor White
Zick, A., & Küpper, B. (2005). Transformed anti- Afrikaners found their traditional privileges to
Semitism—a report on anti-Semitism in Germany. be under threat. The fear was that British capi-
Journal für Konflikt- und Gewaltforschung/ talist imperialism would result in Afrikaners
International Journal of Conflict and Violence being “lumped together” with other minority
Research, 7, 50–92. ethnic groups and afforded the same kind of
26 Apartheid

second-class citizenship that Black, Colored, and How It Worked

Indian South Africans had received. Political
sympathies began drifting toward segregation- In practice, the policy of apartheid comprised two
ists, who worked to revive the fortunes of the separate programs: “grand apartheid” and “petty
Afrikaners relative to the British colonizers and apartheid.” Grand apartheid involved an ambi-
the ethnic minorities. tious and brutal process of social engineering.
The policy that became known as apartheid Black immigration into White areas was halted,
was designed to entrench Afrikaner power relative and many Black migrants considered “surplus” to
to these two traditional threats. The model for economic requirements were deported to “home-
race relations in South Africa (and many other lands.” Mass relocations of ethnic minorities in
nations) in the early 20th century was a British South Africa resulted in hundreds of thousands of
imperialist model, in which Blacks and Whites people being forcibly removed from their homes.
were geographically segregated within a single The “homelands” offered Blacks were dispropor-
polity. Whites ruled over Blacks politically, and tionately small and arid.
Blacks were expected to assimilate to White cul- Slums that had grown up after World War II
ture in order to become competitive within the were demolished and replaced with “townships,”
socioeconomic system. where Blacks had no permanent property rights.
Apartheid theorists in the 1940s argued that Many of the townships were placed just inside
this horizontal system of White supremacy was Black homeland borders, and White-run industrial
unsustainable because it would breed frustration, plants were relocated just outside the borders. This
violence, and rebellion from the ethnic minorities. was designed to encourage Blacks to migrate to the
Under apartheid, Afrikaners, Anglos, Coloreds, homelands, at the same time as offering South
and various Black tribes would be given separate African industrialists access to “foreign” labor
homelands, which would coexist within the nation that was not entitled to the same rights offered to
of South Africa. By giving each ethnic group its those in Afrikaner areas.
own political and cultural space, it was argued that Services for Black people were boosted in the
racial conflict would be reduced because each eth- Black homelands but dramatically cut in “White
nic group would be free to develop its own politi- areas,” a strategy designed to coax Blacks to settle
cal and cultural identity independent of the in the Black homelands. This epic program of
others. relocation required that Blacks be under close sur-
As an intellectual abstraction, apartheid is con- veillance and that their movements be closely
sistent with “dual identity” models of intergroup regulated. Black workers needed permits to leave
relations, whereby subcultures are encouraged to the homelands to seek work and to live in Black
foster a distinct identity while at the same time townships: Hundreds of thousands of Blacks were
embracing what they share at the superordinate imprisoned for not having a pass or for traveling
(national) level. Indeed, much of the rhetoric that to a place without permission. Institutions were
was used to promote apartheid focused on its manipulated to prevent the desegregation of the
potential to liberate Afrikaners from British dom- races. For example, school curricula were rede-
ination and to reduce interracial conflict. signed to actively discourage economic assimila-
In reality, though, the implementation of apart- tion of Blacks. Interracial marriage and even
heid reinforced the type of horizontal White sexual relations between races were prohibited
supremacy it defined itself against. Rather than by law.
reducing racial conflict, it dramatically deepened Those Black and Colored South Africans who
inequities between White and Black South Africans, remained in White areas were segregated from the
and intensified the frustration, violence, and rebel- White population. This policy—known as petty
lion that it was designed to diminish. Rather than apartheid—involved racial segregation of services
allowing for subcultures to flourish, apartheid and facilities such as parks, public transportation,
became an intellectual masquerade that allowed and restaurants. The policy was essentially a for-
both Afrikaners and British descendents to main- malized version of the segregation policies that
tain their traditional racial privilege. existed in many countries in the early- to mid-20th
Apartheid 27

century, such as those used in the United States From the mid-1970s, a number of insurrections
during the Jim Crow era. “Grand apartheid” was broke out in poor Black townships. These expres-
the most dramatic and distinctive manifestation of sions of people power were often poorly organized
the policy of total separation, but for outsiders, it and easily crushed. But images of Black protest
was “petty apartheid” that came to symbolize the and heavy-handed attempts by police to quell the
injustices of South African race relations. Although revolts increased pressure on the international
the complexities of grand apartheid were difficult community to coerce South Africa into reform. In
to capture and communicate to international audi- the 1970s, economic sanctions and sporting boy-
ences, the “White only” signs associated with petty cotts turned South Africa into a pariah state. A
apartheid provided images that pricked the con- gulf developed between mainstream Whites within
sciences of liberal Whites around the world. and outside South Africa. To outsiders, apartheid
South Africa’s policy of institutionalized segre- was illegitimate, irredeemable, and morally repug-
gation (masquerading as the defense of cultural nant. In contrast, many Afrikaners perceived
identity) emerged at about the same time that rac- themselves to be a misunderstood last line of
ist practices were being actively contested and defense against chaos, communism, terrorism, and
overthrown in many parts of the Western world. godlessness. The apartheid debate became severely
As a result, South African apartheid became a polarized within and outside South Africa.
high-profile cause among international activists In the 1980s, the energies of the ANC gradually
who campaigned for civil liberties and the disman- moved from armed resistance to collective protest
tling of institutionalized racism. and mobilization. South Africa experienced an
unprecedented wave of marches, riots, and boycotts,
this time with significant support from Indian South
Opposition and Demise
Africans and international media and activists. In the
For Black South Africans, the introduction of face of social and economic decline, the National
apartheid resulted in economic marginalization, Party diluted some of the more interventionist aspects
disempowerment, humiliation, and organized of apartheid, before formally negotiating ways to
resistance. Intellectuals such as Steve Biko drew resolve the 40-year conflict. In 1994, multiracial elec-
inspiration from the Black Power movement in the tions were held for the first time, and Mandela
United States and worked to develop Black pride became the first Black president of South Africa.
and nonviolent opposition to apartheid in Black Since then, apartheid has been morally and
South Africans. Advocates of Black consciousness intellectually discredited within South Africa as
reinforced the notion that Blacks must stop their well as outside it. Morally, it is considered indis-
psychological subservience to and economic depen- putable that governments need to protect the rights
dency on Whites, and that Blacks should ultimately of all its citizens, not just those of racial elites. The
rule South Africa. The psychological transforma- intellectual case for apartheid has been dismantled by
tion was buttressed by a military operation, largely social psychological work on the contact hypothesis,
coordinated by the African National Congress which argues that intergroup relations are best man-
(ANC). Led by Nelson Mandela, the ANC coordi- aged when members of different cultures are allowed
nated underground cells of militia who carried out to interact with equal status, and with support from
sabotage attacks and assassinations. norms and institutional authorities that protect
In response, White South Africa was galvanized against racism. Today, the term apartheid lives on as
in their antipathy toward what they perceived to a metaphor that is occasionally invoked to describe
be agents of terrorism and communism. An army and condemn any policy that is seen to segregate and
of police, intelligence agents, and conscripts was promote inequities between social groups.
built up to crush resistance. A covert civil war
developed between Black militias and the Matthew J. Hornsey
Broederbond, a secretive society of pro-Afrikaner
advocates who engaged in their own military See also Aversive Racism; Civil Rights Movement;
resistance with the blessing of the South African Desegregation; Discrimination; Intergroup Contact
government. Theory; Minority Groups in Society; Prejudice; Racism
28 Asch, Solomon

Further Readings Asch was not the first psychologist to be inter-

Beinart, W., & Dubow, S. (Eds.). (1995). Segregation and ested in how people perceive others, but his
apartheid in twentieth-century South Africa. London: approach was radically different from that of
Routledge. previous researchers. Earlier scholars were inter-
Goboda-Madikizela, P. (2002). A human being died that ested primarily in percetual accuracy—whether
night: A South African woman confronts the legacy of people could accurately guess the personalities
apartheid. New York: Houghton Mifflin. of other individuals, whereas Asch was more
Harvey, R. (2003). The fall of apartheid: The inside story interested in process—in learning how people
form Smuts to Mbeki. New York: Palgrave form impressions of others. He conducted
Macmillan. research designed to answer three questions
Louw, P. E. (2004). The rise, fall, and legacy of apartheid. about impression formation, which were derived
Westport, CT: Praeger. from Gestalt theory. First, when people receive
Mandela, N. (1995). Long walk to freedom: The items of information about an individual, do
autobiography of Nelson Mandela. London: Abacus. they form a coherent and unified impression of
that individual? Second, do some items of infor-
mation organize the overall impression? And
third, do early items influence how later items
Asch, Solomon are interpreted?
Fundamental Questions
Solomon Asch was born in 1907 in Warsaw, To answer his first question, Asch gave par-
Poland, and emigrated in 1920 to the United States. ticipants the following list of traits characterizing
He remains one of the most influential social psy- a fictitious “person X”: intelligent, a hard-
chologists of the 20th century. His research on worker, skillful, warm, determined, practical,
impression formation and social influence consti- and cautious. Participants then wrote a sketch of
tuted innovations that revolutionized the field of person X and answered questions about other
social psychology. The questions he sought to characteristics (e.g., generous, friendly) of that
answer, namely how people form impressions of person. Asch found that participants formed a
others and when they are influenced by others, coherent and positive impression of person X
continue to inspire research to this day. based on the traits they were given.
His ideas about impression formation and To answer his second question, Asch gave par-
social influence, borrowed from the domain of ticipants another list of traits with a single change:
vision, are examined in this entry and illustrated warm was replaced by cold. This time participants
with some of his most famous experiments. The formed a negative impression of person X. When
importance of his work for the subfield of group Asch replaced the traits warm and cold with blunt
processes and the more general field of social psy- and polite, nothing happened. Thus, in regard to
chology is also discussed. his second question, Asch found that certain traits
(warm and cold ) were central for organizing par-
ticipants’ impressions of person X, whereas other
Impression Formation
traits (blunt and polite) were not.
Asch’s research was based on German Gestalt To answer his third question, Asch gave partici-
theory, which can be translated as the theory of pants one of two lists in which the order of the
the “whole.” According to Gestalt theory, when traits was reversed (either intelligent, hard-worker,
we see a face, we do not first perceive one eye, impulsive, critical, envious or envious, critical,
then the other eye, then the mouth, and so on. impulsive, hard-worker, intelligent). He found that
Instead, we immediately see the entire face (a participants’ impressions of person X were more
gestalt), and this face is more than the sum of its favorable when they received the first list than the
parts (e.g., if an eye and the mouth changed second, revealing a primacy effect in which early
places, we would perceive a very different face). traits in the list guided participants’ interpretation
Asch, Solomon 29

of later traits (e.g., impulsive may be understood to States were more likely to agree with the sentence
mean spontaneous in the first list and aggressive in when it was attributed to Jefferson than to Lenin.
the second list). One could interpret this result as evidence that
admiration for Jefferson generalized to the sen-
tence when it was attributed to him, whereas dis-
Explaining Attitudes
dain for Lenin generalized to the (same) sentence
The results Asch obtained suggest that people when it was attributed to him.
have implicit theories about others. For instance, This is not the best explanation for what Asch
we may believe that if someone is warm, then he found, however. Rather than terminating his study
or she is also generous. Such implicit theories may after participants expressed their level of agree-
help to explain certain stereotypes. For example, ment with the sentence, Asch also asked them the
we may believe that if X is a gypsy, then he or she meaning of the sentence. He found that this mean-
is also a musician. Asch’s theory-driven approach ing differed depending on the ostensible author.
to impression formation did not go unchallenged. When Jefferson was the author, rebellion was
For example, Norman Anderson argued that interpreted to mean peaceful political change.
when forming an overall impression of an indi- When Lenin was the author, rebellion was inter-
vidual, people use a data-driven approach in preted to mean violent revolution. In line with the
which they evaluate each trait associated with the Gestalt perspective, Asch concluded that changing
individual (intelligent, hardworking, etc.) and the ostensible author of the sentence did not
then combine (e.g., through adding or averaging) change participants’ attitude toward the statement,
these evaluations. but rather the meaning of the statement.
The controversy between theory-driven and
data-driven impression formation went on for
some years, but was finally resolved by Susan Fiske Surprising Results
and Steven Neuberg in 1990. According to these In all the experiments summarized so far, the
scholars, people’s first tendency is to place others stimuli that participants judged were rather ambig-
into a familiar category (e.g., French). If the cate- uous (i.e., there were no clear-cut right and wrong
gorization does not fit the evidence, and if people answers). In subsequent studies, Asch sought to
are motivated to obtain a better fit and have the determine whether he could obtain the same results
cognitive capacity (and time) to do so, they will go using unambiguous stimuli. In these studies, he
through additional steps. First, they will try to showed two cards to participants. One card con-
confirm their initial categorization. If this fails, tained three lines of different lengths (a, b, and c).
they will try to recategorize the person in a way The other card contained a single (standard) line
that makes sense of most of his or her characteris- that was the same length as one of the lines on the
tics. Finally, if this fails, they will default to “piece- first card. Participants’ task was easy: They simply
meal integration,” which involves simply adding had to say which line on the first card was the
or averaging all of the person’s characteristics. same length as the standard line. The stimuli were
unambiguous, as indicated by the fact that partici-
pants tested alone hardly ever made errors.
Social Influence
Asch was interested, however, in whether par-
Social influence is another domain in which Asch ticipants tested in a group situation where other
had an indelible impact. Imagine a sentence assert- people made incorrect judgments about the line
ing that a little rebellion now and then is a good lengths would still answer correctly. So, he created
thing and is as necessary in the political world as a situation in which a single naïve participant was
storms are in the physical world. In addition, confronted by several people (experimental con-
imagine that this sentence is attributed either to federates) who gave unanimously incorrect answers
U.S. President Thomas Jefferson (the real author) on several trials of the line judgment task. Asch
or to Vladimir Lenin, one of the leaders of the expected that the incorrect majority would have
Communist Revolution in Russia. Not surpris- little or no influence on participants’ judgments,
ingly, Asch found that participants in the United but his prediction turned out to be wrong.
30 Assimilation and Acculturation

Participants conformed to the erroneous majority Leyens, J.-Ph., & Corneille, O. (1999). Asch’s social
answer about one third of the time. This finding psychology: Not as social as you may think.
surprised Asch, but turned out to be one of the Personality and Social Psychology Review, 3,
most influential findings in social psychology. 345–357.
Asch did subsequent experiments to clarify the
conditions under which people do and do not con-
form to group pressure. For instance, he varied the
number of confederates and their level of unanim-
ity. He found, for example, that the presence of a
Assimilation and
single confederate who gives correct answers sub- Acculturation
stantially reduces the group’s tendency to yield to
the majority. Later research by others demon- During much of the 19th and early 20th centuries,
strated that conformity can also be affected by such the term assimilation was used to describe the
factors as the publicness of participants’ responses process by which immigrants inevitably gave up
and their liking for other group members. The their culture of origin for the sake of adopting the
impact of these variables has often been explained mainstream language and culture of their adopted
in terms of two motives: the desire to respond accu- country. However, by the late 20th century, the
rately and the desire to be liked. Asch’s research on term acculturation was adopted by scholars to
conformity also inspired other important work on describe the more fundamental process of bidirec-
social influence. Two examples are Stanley tional change that occurs when two ethnocultural
Milgram’s studies on obedience to authority and groups come into sustained contact with each
Serge Moscovici’s research on minority influence. other. From this latter perspective, assimilation is
only one of the many acculturation strategies that
immigrant and national minorities may adopt as
His Legacy they strive to adapt to mainstream society.
The enduring legacy of Asch’s work is due to sev- Such strategies have become more and more
eral factors. His theoretical perspective was ele- necessary as immigration, legal or illegal, has
gant, and his results were clear-cut. More important, become increasingly common across the globe.
the two phenomena he studied—impression for- Through immigration and the recognition of the
mation and social influence—are everyday occur- rights of indigenous and national minorities, most
rences and play a major role in interpersonal and 19th-century nation-states have been transformed
intergroup relations. Although Asch was not a from being more or less unicultural to being mul-
highly prolific writer during his lifetime, the fact ticultural, multi-ethnic, and multilingual states.
that his 1952 textbook is still widely cited provides Following the height of nation building in 19th-
strong evidence for his influence in the field. century Europe, the term host majority was
ascribed to the “core founding members” of a
Jacques-Philippe Leyens nation who constituted the dominant ancestral
community in control of the state.
See also Anticonformity; Conformity; Informational Traditionally, host majorities expected immi-
Influence; Minority Influence; Normative Influence; grants to assimilate to the culture and values of the
Obedience to Authority receiving society. Host majorities have found it
easier to assimilate immigrants when their cultural
differences were reduced to exotic manifestations
Further Readings such as ethnic restaurants, music, and dance.
Asch, S. E. (1952). Social psychology. Oxford, UK: However, host societies have found it difficult to
Prentice Hall. share jobs, housing, and welfare with immigrants,
Levine, J. M., & Russo, E. M. (1987). Majority and whom they often see as unentitled to compete for
minority influence. In C. Hendrick (Ed.), Review of such limited resources and as contributing to the
personality and social psychology: Group processes growing cultural and physical insecurity of the soci-
(Vol. 8, pp. 13–54). Newbury Park, CA: Sage. ety. At stake is whether or not host communities
Assimilation and Acculturation 31

wish to accept, nurture, assimilate, or reject the majority, which by ignorance, indifference, or
distinctiveness of immigrants as members of cul- design has sought to subjugate immigrant or indig-
tural communities. Ultimately, will dominant enous communities through forced assimilation,
majority members allow immigrant minorities not segregation, cultural genocide, ethnic cleansing, or
only to maintain their distinctive culture and lan- extermination.
guage, but also to transform the institutions, cul-
ture, and values of the host society?
From Uni-Dimensional to
This entry examines various models of accul-
Bi-Dimensional Models of Acculturation
turation and how strategies of acculturation may
be linked to political views, socioeconomic charac- In Western societies, much of the early accultura-
teristics, and personality traits. tion research focused on the adaptation strategies
of immigrant minorities as they interacted with
the dominant host majority. This almost exclusive
Acculturation and Deculturation
focus on the acculturation process of immigrants
From the cross-cultural psychology perspective, imposed a form of “collective dispositional bias,”
acculturation implies that both immigrants and which often blamed immigrants for not suffi-
host majority members are influenced and trans- ciently or successfully adapting to the culture,
formed by their intercultural contact and are habits, and values of the receiving society.
expected to modify some aspects of their respective Furthermore, traditional models of immigrant
cultures. Host majority members enjoy some con- acculturation were uni-dimensional, as they pro-
trol over the degree of contact they have with posed that during immigrants’ lifetime, they shifted
immigrants and may experience acculturation from exclusive grounding in their culture of origin
either through direct interpersonal contacts in to a bicultural phase reflecting maintenance of the
school and at work or through indirect contacts heritage culture and adoption of the host culture,
via mass media portrayals. However, relative to to complete assimilation to the dominant host
dominant majorities, cultural minorities are more majority culture.
likely to be transformed by such intergroup con- Criticisms of the uni-dimensional model led to
tacts. Immigrants and national minorities have in the development of bi-dimensional models of
common their vulnerability to the tolerance or acculturation. In his bi-dimensional model, John
intolerance of dominant host majorities, whose Berry proposed that the maintenance of the immi-
demographic strength, prestige, and institutional grant culture and adoption of the host majority
power within the national state can result in much culture could be portrayed as independent dimen-
acculturative pressure. sions instead of contrasting points on a single con-
The following types of minorities are likely to tinuum of cultural change. Thus, whether
experience much acculturation pressure: first- and immigrants achieve competence in the host major-
second-generation immigrants, sojourners, refu- ity language could have little to do with their
gees, asylum seekers, and national minorities. An maintenance of their heritage language. An adap-
extreme case of acculturation pressure was that of tation of the Berry model asserts that immigrants
South and North American aboriginals in the 17th and national minorities may endorse five accul-
through 19th centuries, as they had no control turation orientations, including the assimilationist
over the unwanted, massive, and sustained immi- strategy proposed in traditional uni-dimensional
gration of Northern Europeans whose demo- models. Immigrants with an integrationist orienta-
graphic, economic, technological, and military tion want to maintain certain aspects of their cul-
supremacy physically decimated their indigenous ture of origin while also adopting key features of
communities while causing acculturation pressures the culture of the host community. Those with a
that often resulted in outright deculturation. The separatist perspective seek to maintain their lan-
term deculturation is used to describe the cultural, guage and culture of origin while rejecting key
linguistic, religious, psychological, and health aspects of the host community culture. Immigrants
breakdown that occurs in minority communities who adopt the assimilationist strategy want to
that experience sustained contact with a dominant abandon their culture and/or language of origin
32 Assimilation and Acculturation

for the sake of adopting the culture and/or lan- origin, type, and rate of immigration accepted
guage of the host community. The marginalized within their boundaries, public policies designed to
feel alienated from their culture of origin and expe- facilitate the integration of immigrants and national
rience sustained rejection by members of the minorities within mainstream society remain the
dominant host majority, a double jeopardy often exception rather than the rule. State integration
leading to anomie. Immigrants may also endorse policies consist of the approaches adopted by
an individualist acculturation orientation as they national, regional, and municipal governments to
define themselves and others on the basis of their help immigrants and host communities adapt to
personal characteristics and achievements rather the growing ethnic, linguistic, and religious diver-
than on their group membership. Such individual- sity of modern states.
ists are not concerned with maintaining the immi- The IAM proposes four clusters of ideologies that
grant culture or adopting the host culture, as they can shape the integration policies adopted by demo-
are more involved with achieving their personal cratic governments of multiethnic states. As a heu-
goals in their country of adoption. ristic for analyzing integration policies, these four
clusters can be placed along a continuum ranging
from the pluralism and civic ideologies at one end of
The Interactive Acculturation Model
the continuum to the assimilationist and ethnist ide-
It is only in the last decade that researchers have ologies at the other end. Depending on political,
focused their attention on the acculturation orien- economic, demographic, and military events occur-
tations held by host communities, which by virtue ring at the national and international levels, state
of their dominant position and control of immigra- integration policies can shift from one ideological
tion and integration policies have a substantial orientation to the other. The IAM proposes that
impact on the acculturation orientations adopted adoption of state integration policies may reflect
by immigrant and national minorities. The interac- and also shape host community acculturation orien-
tive acculturation model (IAM) was proposed by tations, as well as more general opinions concerning
Richard Bourhis to better account for the inter- the ideal or preferred ways of integrating minorities
group processes that characterize relations between within mainstream society. Political tensions may
host majority members and cultural minorities. emerge between factions of the host majority hold-
The IAM framework includes the following ing rival ideological views on immigration and inte-
elements: (a) immigration and integration policies gration issues. The polarization of ideological
that can affect the climate of intergroup relations positions regarding such issues may lead to the for-
between immigrant and host communities, mation of political parties whose main platform is
(b) acculturation orientations adopted by host to change state policies on immigration and integra-
community members toward specific groups of tion issues. While left wing parties may endorse
immigrants, (c) acculturation orientations adopted public policies at the pluralist pole of the ideological
by immigrants within their country of adoption, continuum, right wing nationalist or religious par-
and (d) interpersonal and intergroup relational ties may advocate integration policies situated at the
outcomes that are the product of combinations of assimilationist or ethnist side of the continuum.
immigrant and host community acculturation ori- The IAM proposes that the acculturation ori-
entations. As a complement to other acculturation entations of dominant host majority members can
frameworks, the IAM focuses on the cultural have a major impact on the acculturation orienta-
adoption strategies of immigrant and host major- tions of immigrant minorities. Dominant host
ity members rather than on their dual group identi- community members may endorse five accultura-
ties or desires for intergroup contact. tion orientations they wish immigrants to adopt:
The IAM takes into account how public policies integrationism, assimilationism, segregationism,
regarding immigration and integration relate to exclusionism, or individualism. These accultura-
the acculturation orientations endorsed by host tion orientations are measured using the validated
majority and immigrant group members. While Host Community Acculturation Scale (HCAS).
most democratic states have formulated and Integrationism is endorsed by host community
applied immigration policies regulating the national members who accept that immigrants maintain
Assimilation and Acculturation 33

some aspects of their heritage culture, and also conflictual relational outcomes. Intergroup rela-
accept and value that immigrants adopt impor- tional outcomes include cross-cultural and bilin-
tant features of the host majority culture. gual communications, interpersonal and
Integrationists value a stable biculturalism/bilin- intergroup misunderstanding, prejudice and ste-
gualism among immigrant communities, which, reotyping, social and institutional discrimination
in the long term, may contribute to cultural and in employment, housing, education and interper-
linguistic pluralism as an enduring feature of the sonal relations. Harmonious relational outcomes
host society. Assimilationism corresponds to the include optimal intergroup understanding and
traditional concept of absorption, whereby host can be expected when immigrants and host
community members expect immigrants to relin- community members both adopt the integration-
quish their language and cultural identity for the ist and individualist acculturation orientations.
sake of adopting the dominant culture and lan- Problematic relational outcomes are expected
guage of the host majority. Segregationism is when the acculturation orientations of host
exemplified by host community members who majority members and immigrants are partially
accept immigrants’ maintenance of their heri- concordant or discordant. For instance, problem-
tage culture, as long as the immigrants keep atic outcomes, including intergroup misunder-
their distance from host members, as they do standing and miscommunication, may emerge
not wish immigrants to transform, dilute, or when immigrants endorse integrationism while
“contaminate” the host culture and value sys- host community members endorse assimilation-
tem. Host community members who adopt this ism for immigrants. Problematic outcomes may
orientation discourage cross-cultural contacts also emerge when host majorities represent immi-
with immigrants, prefer immigrants to remain grants as endorsing mainly separatism, while
together in separate urban or regional enclaves, immigrants perceive the host majority to be
and are ambivalent regarding the status of mainly segregationist or exclusionist. Conflictual
immigrants as rightful members of the host soci- relational outcomes including discrimination,
ety. Exclusionism is adopted by members of the hate crimes, and intergroup violence can be expected
host majority who deny immigrants the right to from host majority members who endorse segrega-
adopt features of the host community culture. tionism or exclusionism, especially for immigrants
Exclusionists also deny immigrants the choice to perceived as threatening. Faced with systemic dis-
maintain their heritage language, culture, or crimination and hostility from host majority mem-
religion and believe that some immigrants have bers who are segregationist and exclusionist,
customs and values that can never be socially immigrants who adopt separatism or marginaliza-
incorporated within the host community main- tion may eventually resort to outright conflict
stream. Individualism is an orientation endorsed strategies through civil disobedience, rioting, crim-
by host community members who define them- inal activity, armed struggle, or terrorism.
selves and others as individuals rather than as
members of group categories such as immigrants
Studies of Host Community
or host community members. Because it is per-
Acculturation Orientations
sonal qualities and individual achievements that
count most, individualists will tend to interact Numerous empirical acculturation studies have
with immigrants in the same way they would with been conducted with dominant host community
other individuals who happen to be members of undergraduates, thus controlling for the educa-
the host community. tional and socioeconomic status of respondents in
The IAM proposes that acculturation orienta- urban centers such as Los Angeles, Montreal, Paris,
tions endorsed by host community members may Brussels, Geneva, and Tel Aviv. These studies have
be concordant or discordant with those held by shown that individualism and integrationism are
members of specific immigrant communities. The the most strongly endorsed acculturation orienta-
degree of concordance between the acculturation tions toward immigrants. Endorsement of welcom-
orientations of host community members and immi­ ing acculturation orientations such as individualism
grants may result in harmonious, problematic, or and integrationism may reflect the meritocratic
34 Assimilation and Acculturation

and individualistic university organizational cul- of threat and cultural insecurity toward devalued
ture, which favors the equal treatment of individu- immigrants as a way of maintaining mobilization
als, regardless of race, color, or creed. Studies have in favor of their respective nationalistic causes.
shown that assimilationism, segregationism, and Right wing nationalist parties gain much of their
exclusionism are the least endorsed acculturation support from host majority electorates by nurtur-
orientations among college students, though in ing feelings of symbolic and realistic threats espe-
recent years endorsement of segregationism by cially from the presence of “devalued” immigrants
students has increased somewhat in both Québec whose demographic presence is often portrayed as
and France. overwhelming and out of control.
Overall, undergraduates endorsed more wel- Even though host majorities may endorse each
coming acculturation orientations toward “valued” acculturation orientation to a different degree
immigrants than toward “devalued” immigrants or cross-culturally, the social psychological profile of
national minorities. For instance, undergraduate each acculturation orientation remains similar
students in Tel Aviv more strongly endorsed the regardless of the national background of respondents.
individualism and integrationism orientations Study results obtained in Montreal, Los Angeles,
toward Jewish immigrants from Russia and Ethiopia Paris, Geneva, Brussels, and Tel Aviv suggest that
than toward the devalued Israeli Arab national this is the case. Individualism and integrationism
minority in Israel. Conversely, Jewish undergradu- are two “live and let live” acculturation orienta-
ates more strongly endorsed the segregationism and tions the correlates of which were quite similar
exclusionism orientations toward Israeli Arabs cross-culturally. Individualists and integrationists
than toward Jewish immigrants from Russia and felt comfortable with immigrants, wanted close
Ethiopia. relations with both valued and devalued immi-
Does left wing versus right wing political affili- grants, including as best friends, and felt that
ation influence acculturation orientations toward immigrants in general wanted good relations with
devalued groups? In Israel, left wing Labour iden- members of the host majority. Individualists and
tifiers more strongly endorsed the individualism integrationists did not endorse the authoritarian or
and integrationism orientations toward Israeli social dominance orientation and ethnocentric ide-
Arabs than did Likud party identifiers. Conversely, ologies, and they were more likely to identify with
Likud Party identifiers more strongly endorsed the “left of center” political parties in their respective
segregationism and exclusionism orientations sociopolitical settings.
toward Israeli Arabs than did Labour party identi- Assimilationists, segregationists, and exclusion-
fiers. Most important, right wing Likud Party ists all rejected immigrants and their culture,
identifiers were unique in more strongly endorsing endorsed the social dominance orientation and
the segregationism and exclusionism orientations authoritarian and ethnocentric ideologies, and
than the individualism and integrationism orienta- were more likely to identify with right wing politi-
tions toward Israeli Arabs. Political parties are cal parties. Importantly, they were more likely to
created and remain popular to the degree that they feel that their ingroup identity was threatened by
offer “solutions” to the fears and aspirations of the presence of immigrants, especially “devalued”
their electorate. The right wing Likud Party plat- ones. They were also more likely to feel insecure
form nurtures a sense of threat to the vitality and culturally, linguistically, and economically as mem-
national security of the Jewish majority in Israel. bers of their own group, while wishing to avoid
Threats felt from the presence of Israeli Arabs immigrants as colleagues at work, as neighbors, or
make it particularly difficult for Likud Party sym- as best friends. In each cultural setting, specific social
pathizers to accept any type of relationship with psychological variables differentiated the assimila-
Israeli Arabs, and “justifies” keeping Arabs segre- tionist, segregationist, and exclusionist acculturation
gated and excluded from the Jewish dominant orientations. Taken together, these social psycho-
majority. logical correlates of acculturation orientations
Right wing nationalist parties in other settings, attest to the construct validity of the HCAS and
such as France and Québec, also nurture feelings also support some basic premises of the IAM.
Assimilation and Acculturation 35

Studies of Immigrant foreign diplomas are not recognized in the country

Acculturation Orientations of settlement. Immigrant women who seek more
egalitarian sex roles in their country of settlement
Empirical cross-cultural studies have examined are more likely to experience acculturative stress
acculturation orientations endorsed by immigrants than men, especially when sex roles in the country
and national minorities using variants of the of origin were quite traditional. While accultura-
Immigrant Acculturation Scale (IAS) developed by tive stress is more likely to be experienced by
John Berry and his colleagues. In many cultural immigrants who settle at an older age in their
settings, immigrants endorse integrationism to a country of adoption, personality factors such as
greater degree than assimilationism and separat- introversion or extraversion, internal or external
ism, and marginalization is rarely endorsed. locus of control, and degree of self-efficacy have
Exceptions to these findings are Turks in Germany, also been linked to acculturative stress.
lower economic status Turks in Canada, and some As developed by Colleen Ward, ethnocultural
indigenous minorities in various parts of the world, identity conflict (EIC) stems from identity conflict
who endorse separatism more than integrationism. occurring when the multiple social identities devel-
Overall, feelings of being the victim of prejudice oped as a result of emigration become incompati-
and discrimination are the most important corre- ble with each other. EIC can be prevalent for
lates of separatism and marginalization. While immigrant youth, who experience difficulties in
newly established immigrants may at first adopt harmonizing the traditional values of their parents
integrationism or assimilationism, sustained expe- with the modern values of their host majority age
rience of discrimination and exclusion in the host peers. Infrequent contact with host majority peers,
society may shift acculturation orientations to interethnic tensions, threats to cultural continuity,
separatism or marginalization. Acculturation ori- and perceived discrimination are aggravating fac-
entations can also be endorsed differently in the tors that contribute to EIC. Furthermore, immi-
public and private domains. In the public domain, grants who endorse the separation, assimilation,
immigrants may endorse linguistic integration and marginalization acculturation orientations are
through bilingualism and assimilation at work, more likely to experience EIC than those who
whereas in the private domain they may practice endorse integrationism.
separatism through religiously and ethnically Despite the pressures of acculturative stress and
endogamous marriages. ethnocultural identity conflict, immigrants can be
Acculturative stress may be experienced as a quite resilient in their psychological and sociological
result of intercultural contacts that highlight differ- adaptation to their country of adoption. While psy-
ences between the heritage culture of immigrants chological adaptation refers to good mental health
and that of the dominant host majority. This is more and a sense of well-being, sociocultural adaptation
likely to occur when the “cultural distance” between involves a set of social competencies that enable
the heritage culture of immigrants and that of the minority individuals to live successfully in their
receiving society is large. As proposed by Anthony intercultural world. Studies with immigrants showed
Richmond, immigrants may suffer more accultura- that the relationship between psychological and
tive stress when their migration was involuntary sociocultural adaptations increased over time and
(reactive emigration) than in cases where individuals tended to be stronger in cases where the cultural
voluntarily chose to emigrate to a country to which distance between the immigrant culture and that of
they were attracted (proactive emigration). For most the host community was small rather than large.
immigrants, acculturative stress is related to the The complementary link between psychological and
experience of culture loss and anxieties about how sociocultural adaptation was stronger for immi-
to adapt to the country of settlement. grants with the integrationism and assimilationism
While higher education is associated with less orientations than for those with the separation and
acculturative stress, immigrants who suffer an marginalization acculturation orientations.
important drop in occupational status can suffer A recent comparative study of immigrant youth
much acculturative stress, especially when their from 13 countries showed that better psychological
36 Assimilation and Acculturation

and sociocultural adaptation was related to relational outcomes by adopting separatism

endorsement of the integrationism orientation but rather than integrationism or assimilationism.
not very related to endorsement of assimilationism Finally, marginalization is the acculturation ori-
and separatism, and least related to the marginal- entation associated with the least desirable psycho-
ization acculturation orientation. Results also logical and sociocultural correlates. Marginalization
showed that perceived discrimination against is associated with neuroticism, anxiety, closed-
ingroup members was the single strongest predic- mindedness, and unsociability. Similarly, a link has
tor of poor psychological and sociocultural adapta- been found between marginality, alienation, ano-
tion. The 13-country study showed that immigrant mie, deviance, and psychosomatic stress. Being a
youth who endorsed the integration orientation victim of discrimination was found to be the single
experienced less ethno-cultural identity conflict, most important predictor of marginalization.
less anxiety and depression, and fewer psychoso-
matic symptoms than their peers who endorsed
assimilationism, separatism, and especially the
marginalization acculturation orientation. Immi­ Much fundamental and applied research remains
grant endorsement of integrationism was also to be done to do justice to the complexity and
shown to be positively correlated with the traits of subtlety of immigrant–host community relations in
extraversion, emotional stability, sociability, agree- multi-ethnic settings. In line with the interactive
ableness, sensation seeking, and open-mindedness. acculturation model (IAM), more empirical studies
In addition, endorsement of integrationism was are needed to explore how concordant and discor-
found to be related to higher self-esteem, which in dant acculturation orientations between immigrant
turn was a strong predictor of immigrant adapta- and host communities can result in harmonious,
tion. Immigrant youth who endorsed integration- problematic, or conflictual relational outcomes,
ism were those whose social identification was not only in regard to intercultural communication
dual, who were more likely to be bilingual, and and prejudicial attitudes but also in behavioral
who had both ingroup and outgroup peer con- outcomes such as prosocial behaviors, employ-
tacts. Conversely, separatism was positively cor- ment equity, discrimination, intergroup conflicts,
related with neuroticism, anxiety, impulsivity, and hate crimes.
sensation seeking, and aggressiveness, and nega- Emerging research is currently exploring the
tively correlated with extraversion, sociability, acculturation orientations of immigrant communi-
self-assurance, and self-esteem. ties toward coexisting, competing, or rival “other”
Studies found that assimilationism is positively immigrant communities, either long established
related to task-coping and emotion-coping orien- following earlier immigration cycles or more
tations, and thus contributes to the reduction of recently arrived as a result of current immigration
emotional distress associated with stressful situa- waves. Likewise, in culturally divided societies,
tions. Personality traits that were associated with acculturation orientations endorsed by national
assimilationism were agreeableness and sociabil- minorities toward the dominant majority are being
ity, as well as neuroticism, anxiety, closed-mind- explored. More acculturation research should be
edness, and field dependence. In immigrants’ conducted with sojourners, refugees, and asylum
quest to endorse integrationism and assimila- seekers as they adapt to increasingly multi-ethnic
tonism, they may also adopt the less desirable and multilingual receiving societies. Multiple iden-
habits and customs of the host majority. For tity research dealing with the interplay of subna-
instance, one study showed that immigrant youth tional, national, supranational, and transnational
who endorsed integration and assimilation were identities across the world also calls for more com-
at higher risk than separatists of adopting health- plex elaborations of current acculturation models.
compromising behavior such as smoking and The very premise of considering host societies as
drinking alcohol. In line with the IAM, immi- being composed of single or dual host communi-
grants who are confronted by mainly segregation- ties may already be an oversimplification, both
ist and exclusionist host majority members may conceptually and empirically. Host societies of the
reduce acculturative stress and avoid conflictual future may well be constituted of multiple host
Attachment Theory 37

communities, all of which are ethnic and linguistic shapes the goals, working models (that is the inter-
minorities sharing two or more official national personal attitudes, expectancies, and cognitive
languages but no obvious core founding majority. schemas), and coping strategies that she or he uses
when emotion-eliciting events happen in relation-
Richard Y. Bourhis and Shaha El-Geledi ships. This entry examines Bowlby’s original ideas
and the evolution of his theory among later
See also Culture; Desegregation; Discrimination;
Diversity; Immigration researchers.

Normative Features of Attachment Theory

Further Readings
Bowlby’s fascination with the emotional ties that
Berry, J. W. (1997). Immigration, acculturation and
bind humans to one another began with an astute
adaptation. Applied Psychology: An International
observation. In all human cultures and indeed pri-
Review, 46, 5–34.
mate species, young and vulnerable infants display
Berry, J. W., Phinney, J. S., Sam, D. L., & Vedder, P. H.
a specific sequence of reactions following separa-
(Eds.). (2006). Immigrant youth in cultural transition:
tion from their stronger, older, and wiser caregiv-
Acculturation, identity and adaptation across national
contexts. Mahwah, NJ: Lawrence Erlbaum.
ers. Immediately following separation, infants
Bourhis, R. Y., Moïse, C., Perreault, C., & Senécal, S. protest vehemently, typically crying, screaming, or
(1997). Towards an interactive acculturation model: A throwing temper tantrums as they search for their
social psychological approach. International Journal caregivers. Bowlby believed that vigorous protest
of Psychology, 5, 1–18. during the early phases of caregiver absence is a
Sam, D. L., & Berry, J. W. (Eds.). (2006). The Cambridge good initial “strategy” to promote survival, espe-
handbook of acculturation psychology. Cambridge, cially in species born in a developmentally imma-
UK: Cambridge University Press. ture and very dependent state. Intense protests
often draw the attention of caregivers to their
infants, who would have been vulnerable to injury
or predation during evolutionary history if left
Attachment Theory unattended.
If loud and persistent protests fail to get the
Attachment theory was developed by John Bowlby caregiver’s attention, infants enter a second stage,
to explain personality and social development known as despair, during which they usually stop
from the cradle to the grave. The theory focuses on moving and become silent. Bowlby believed that
the experience, expression, and regulation of emo- from an evolutionary standpoint, despondency is a
tions at both normative (species-typical) and indi- good second strategy to promote survival. Excessive
vidual difference (person-specific) levels of analysis. movement could result in accident or injury, and
This focus is not surprising given how important loud protests combined with movement might
emotions and affect regulation are to interpersonal draw predators. According to this logic, if protests
functioning in all types of close relationships. fail to retrieve the caregiver quickly, the next best
Bowlby believed that the attachment system survival strategy would be to avoid actions that
serves two primary functions: (1) to protect vulner- might increase the risk of self-inflicted harm or
able individuals from potential threats or harm and predation.
(2) to regulate negative affect following threatening After a period of despair, infants who are not
or harmful events. The normative component of reunited with their caregivers enter a third and
attachment theory specifies the stimuli and con- final stage—detachment. During this phase, the
texts that normally evoke and terminate different infant begins to resume normal activity without the
kinds of emotions, as well as the sequence of emo- caregiver, gradually learning to behave in an inde-
tions usually experienced following certain rela- pendent and self-reliant manner. Bowlby believed
tional events. The individual difference component that the function of emotional detachment is to
addresses how an individual’s personal history of allow the formation of new emotional bonds with
receiving care and support from attachment figures new caregivers. He reasoned that emotional ties
38 Attachment Theory

with previous caregivers must be relinquished using the Strange Situation. The Strange Situation
before new bonds can fully be formed. In terms of involves a sequence of separations and reunions of
evolution, detachment allows infants to cast off caregivers (usually mothers) and their children. It
old ties and begin forming new ones with caregiv- assesses how children regulate negative emotions
ers who might be able to provide the attention and vis-à-vis their caregivers when the children are
resources needed for survival. Bowlby also conjec- upset. Even though most children are distressed
tured that these normative stages and processes when left alone at this age, securely attached chil-
characterize reactions to prolonged or irrevocable dren tend to reduce their negative emotions by
separations in adult attachment-based relation- using their caregivers as a “secure base,” and they
ships, which might also have evolutionary adap- resume other activities fairly quickly after reunit-
tive value in terms of maintaining, casting aside, or ing with them in the Strange Situation. Anxious–
forming new romantic pairs. resistant children, by comparison, remain distressed
In addition to identifying the course and func- and often exhibit anger or resentment toward their
tioning of these three distinct stages, Bowlby also caregivers during reunions episodes. Anxious–
identified several normative behaviors that infants avoidant children, who display fewer overt signs of
commonly display in attachment relationships. distress but usually have elevated heart rates,
Such hallmark behaviors include sucking, clinging, remain distant and emotionally detached from
crying, smiling, and following the caregiver, all of their caregivers during reunions, opting to calm
which serve to keep the infant or child in close themselves in a self-reliant manner.
physical proximity to the caregiver. Bowlby also During later stages of development, one of the
documented unique features of the caregiver and key differences between secure and different types
his or her interaction with the infant that are likely of insecure individuals is how their negative emo-
to promote attachment bonds. The features include tions are regulated and controlled based on their
the competence with which the caregiver alleviates specific beliefs and expectancies about the avail-
the infant’s distress, the speed of responsiveness ability of comfort and support from their attach-
of the caregiver to the infant, and the familiarity of ment figures. Highly secure individuals have
the caregiver. These behaviors and features are also learned from past caregiving experiences to follow
believed to be critical to the development of adult “rules” that permit distress to be acknowledged
attachment relationships. Debra Zeifman and and motivate them to turn toward attachment
Cynthia Hazan, for example, have noted that most figures as sources of comfort and support. Highly
romantically attached adults repeatedly engage in avoidant adults, in contrast, have learned to fol-
hallmark attachment behaviors such as sucking, low rules that limit the acknowledgment of dis-
clinging, prolonged eye contact, and extensive tress and encourage the use of self-reliant tactics
belly-to-belly body contact. Mario Mikulincer and to control and reduce negative affect when it
Phillip Shaver have documented the importance of arises. Highly anxious people have learned to use
the responsiveness of romantic partners in the for- rules that direct their attention toward the possible
mation of attachment security. source of distress, to ruminate about it, and to
worry that their attachment figures will never
fully meet their persistent needs for comfort and
Individual Difference Features
of Attachment Theory
Mikulincer and Shaver have recently proposed a
Attachment theorists after Bowlby have proposed process model that outlines the sequence of events
that different attachment patterns (in children) and that underlie the emotional coping and regulation
attachment styles or orientations (in adults) reflect strategies of people who have different attachment
different ways of regulating affect, particularly histories. For example, when stress or a potential
controlling or dampening negative affect in stress- threat is perceived, highly secure individuals remain
ful, threatening, or overly challenging situations. confident that their attachment figures will be
Individual differences in patterns of attachment in attentive, responsive, and available to meet their
12- to 18-month-old children were first docu- needs and help them lower their distress and anxi-
mented by Mary Ainsworth and her colleagues ety. These beliefs, in turn, should increase their
Attitudes Toward Women Scale 39

feeling of security, which should deactivate their early attachment experiences and later attachment-
attachment systems, allowing them to use con- based relationships in early adulthood, just as
structive, problem-focused coping strategies that Bowlby anticipated.
over time are likely to solve their problems.
Highly insecure individuals follow different
pathways. When highly anxious individuals Conclusion
encounter attachment-relevant stress or threats, In conclusion, attachment theory was developed to
they are uncertain as to whether their attachment account for different patterns of personality and
figures will be sufficiently attentive, available, and social development across the entire life span.
responsive to their needs. Such worries should sus- According to Bowlby, understanding the experi-
tain their distress and keep their attachment ence, expression, and regulation of emotion—par-
systems activated, resulting in the use of emotion- ticularly negative emotion in response to events
focused coping strategies such as hypervigilance to that activate the attachment system—is essential to
signs of possible relationship loss and ruminating understanding how and why individuals with dif-
over worst-case scenarios. When highly avoidant ferent attachment histories behave as they do in
individuals feel stressed or threatened, they experi- their close relationships.
ence—but may not consciously acknowledge—
anxiety at a physiological level. To keep their Jeffry A. Simpson and Lane Beckes
attachment systems deactivated, highly avoidant
persons work to inhibit and control their See also Dyads; Families; Interdependence Theory; Levels
emotional reactions by using avoidant coping of Analysis; Need for Belonging; Social Relations
strategies. Model; Trust
These three emotion regulation/coping strategies—
problem-focused, emotion-focused, and avoidance- Further Readings
focused strategies—are the source of many of the Bowlby, J. (1969). Attachment and loss: Vol. 1.
interesting cognitive and behavioral outcomes that Attachment. New York: Basic Books.
have been discovered in people who have different Bowlby, J. (1973). Attachment and loss: Vol. 2.
attachment styles or orientations. More securely Separation: Anxiety and anger. New York: Basic
attached individuals, for instance, typically experi- Books.
ence more intense and mild positive emotions in Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss.
their romantic relationships and fewer intense and New York: Basic Books.
mild negative emotions, whereas the reverse is true Mikulincer, M., & Shaver, P. R. (2003). The attachment
of more insecurely attached persons. Recent longi- behavioral system in adulthood: Activation,
tudinal research has also documented connections psychodynamics, and interpersonal processes. In M. P.
between an individual’s early attachment pattern Zanna (Ed.), Advances in experimental social
(being classified as secure or insecure in the Strange psychology (Vol. 35, pp. 53–152). San Diego, CA:
Situation at age 1) in relation to his or her mother Academic Press.
and emotions experienced and expressed with a
romantic partner 20 years later. In addition, indi-
viduals classified as insecure (either anxious–
avoidant or anxious–resistant) in the Strange
Attitudes Toward Women
Situation at age 1 are rated by their teachers as less Scale
socially competent during early elementary school.
Lower social competence, in turn, predicts greater The Attitudes Toward Women Scale (AWS),
likelihood of being rated as insecurely attached to developed by Janet T. Spence and Robert Helmreich
same-sex friends at age 16, which in turn predicts in the early 1970s, measures attitudes about the
both the experience and expression of greater nega- rights and roles of women—relative to men—in
tive affect in relationships with romantic partners occupational, educational, and relational domains.
when individuals are in their early 20s. Thus, indi- As an attitude measure focusing on gender roles,
rect but theoretically meaningful links exist between the AWS assesses opinions about the behavioral
40 Attitudes Toward Women Scale

patterns deemed appropriate for men and women competent women in stereotypically masculine
in society. Examples include believing that men domains and whether this would relate to their
should be more responsible for supporting their gender-role attitudes. Although there were gender-
families, whereas women should be more respon- role attitude measures already in existence, such as
sible for nurturing their children. Clifford Kirkpatrick’s Belief-Pattern Scale for
Spence and Helmreich created versions of the Measuring Attitudes Toward Feminism published
AWS with 55 items, 25 items, and 15 items, which in 1936, the items were relatively outdated. In
were published in 1972, 1973, and 1978, respec- need of a more contemporary means of measuring
tively. Sample items on the AWS are as follows: gender-role attitudes, Spence, along with her col-
“There are many jobs in which men should be league Robert Helmreich, developed the original
given preference over women in being hired or 55-item version of the AWS. They then discovered,
promoted” and “Under modern economic condi- to their surprise, that male and female college stu-
tions with women being active outside the home, dents, even those with more traditional gender-role
men should share in household tasks such as wash- attitudes, formed positive impressions of compe-
ing dishes and doing the laundry.” Respondents tent women with masculine interests.
indicate their level of agreement with each state-
ment on a four-option scale. A summary score is
Significance of the AWS
created across all scale items such that higher num-
bers indicate more egalitarian gender-role atti- Though neither the first, nor the most recent,
tudes. More than three decades of research have measure of gender-role attitudes, the AWS is the
demonstrated all three versions of the AWS to be most widely cited and used, serving as a reference
reliable, consistently yielding the same results, and point for more recently developed measures.
valid, accurately measuring what they are intended Spence has attributed the popularity of the AWS
to measure. These properties have added to the to its emergence as one of the first gender-role
usefulness and importance of the scale. attitude measures in the early 1970s, when inter-
This entry addresses the background of the est in gender research was growing exponentially
AWS, the significance of the scale, changes over in psychology.
time in gender-role attitudes, and new directions in Because the AWS has been used so widely to
their measurement. measure gender-role attitudes, comparisons of these
attitudes can be made across time (as discussed in
the next section) and across cultures. Investigators
Background and History
have used the AWS in at least 15 different countries
When discussing the history of the AWS, it is inter- (including Brazil, Canada, Croatia, Czech Republic,
esting to note the relevance of the career history of India, Philippines, South Korea, and Spain) on five
its primary founder, Janet Spence. Earning her different continents (Asia, Australia, Europe, North
PhD in 1949, Spence was a pioneering figure for America, and South America). In addition, the
women in psychology at a time when the field was AWS used on college students has also been adapted
largely dominated by men. In 1984, with the for use with adolescents and with the general popu-
American Psychological Association approaching lation in the United States. Although findings from
its centennial, she served as its sixth female presi- these studies are diverse, consistent patterns emerge.
dent, and in 1988, she served as the first member- For example, women and female adolescents typi-
elected president of the American Psychological cally report more egalitarian gender-role attitudes
Society (now the Association for Psychological than their male counterparts. Though the gender-
Science). role attitudes of parents and their children show a
In the 1970s, during the second wave of the moderate degree of association, students consis-
feminist movement, Spence’s research interests tently report more egalitarian attitudes than their
turned to gender. In response to research findings parents or grandparents. In addition, more tradi-
that people liked competent, academically success- tional gender-role attitudes are reported by those
ful males more than incompetent ones, Spence lower in socioeconomic status and those stronger in
became interested in how people would perceive religious affiliation.
Attitudes Toward Women Scale 41

Change Over Time this problem, these ceiling effects are, in and of
themselves, a phenomenon of interest. That is,
The consistent use of the AWS over a period of because these ceiling effects are a more recent
more than three decades has allowed gender phenomenon occurring mostly on certain AWS
researchers to track changes in gender-role atti- items in female college students, researchers can
tudes over time. It is interesting—and no acci- gain a better understanding of change over time
dent—that the AWS appeared at a time when in gender-role attitudes and can also make inter-
women in the United States were more forcefully esting comparisons between male and female
asserting their rights to attain the same educational students.
and employment status as men. Given that the Over time, it has become less socially acceptable
actual behaviors and roles of U.S. women were in the United States to express negative attitudes
changing as they increasingly entered traditionally toward women openly. Because the AWS is an
male-dominated domains, it became important to overt measure of attitudes, it could be argued that
examine whether there were corresponding changes some of the egalitarian responses on the AWS
in societal attitudes about male and female gender might not accurately reflect the respondents’ true
roles. Research does, indeed, indicate that gender- beliefs. This has led to the construction of more
role attitudes are becoming less traditional over subtle measures of gender-role attitudes, such as
time, especially among college students in the the Modern Sexism Scale developed by Janet Swim
United States For example, Janet Spence and Eugene and her colleagues. This scale assesses the extent to
Hahn compared gender-role attitudes (using the which respondents deny that discrimination against
AWS) in four different student cohorts assessed at women still exists, and it has been shown to be a
the same university in 1972, 1976, 1980, and different kind of gender-role attitude measure than
1992, finding the most egalitarian attitudes in more overt measures. It might seem sensible, there-
1992 and the least egalitarian attitudes in 1972. In fore, to discontinue the use of overt measures of
a more comprehensive examination of changes in gender-role attitudes in favor of more subtle mea-
gender-role attitudes that included 71 different sures; however, this action would be short-sighted
samples of U.S. college students, Jean Twenge in the end. Both overt and subtle gender-role atti-
found that gender-role attitudes became steadily tude measures are vital because they serve distinct
more egalitarian in both male and female students research purposes. Most importantly, because the
over a 25-year period from 1970 to 1995. In addi- AWS has been used consistently in research since
tion, although males had more traditional attitudes the early 1970s, its continued use will allow gender
than females at every point in time, these gender researchers to examine how gender-role attitudes
differences decreased in size from 1986 to 1995. change over time well into the future.

Camille E. Buckner
Limitations and New Directions
Despite its usefulness, the AWS has limitations. See also Ambivalent Sexism; Feminism; Gender Roles;
Certain items on the scale appear outdated (e.g., Modern Sexism; Sexism
“It is ridiculous for a woman to run a locomotive
and for a man to darn socks”). Another criticism
levied against the scale, which applies particularly Further Readings
to more recent samples of female college students, Beere, C. A. (1990). Gender roles: A handbook of tests
is that it shows ceiling effects. Ceiling effects occur and measures. Westport, CT: Greenwood Press.
when most or all responses on a measurement McHugh, M. C., & Frieze, I. H. (1997). The
scale cluster around the high (or, in the case of the measurement of gender-role attitudes: A review and
AWS, the more egalitarian) end. This lack of vari- commentary. Psychology of Women Quarterly, 21,
ability in scale responses can lead to difficulty in 1–16.
determining what relationships the scale shows Spence, J. T., & Hahn, E. D. (1997). The Attitudes
with other variables (e.g., educational level or reli- Toward Women Scale and attitude change in college
giosity), an important goal of research. Despite students. Psychology of Women Quarterly, 21, 17–34.
42 Attribution Biases

Spence, J. T., & Helmreich, R. L. (1978). Masculinity and on a scientific analysis of how people should
femininity: Their psychological dimensions, correlates, explain, or attribute, their own or others’ behavior
and antecedents. Austin: University of Texas Press. by using the available information in a systematic
Spence, J. T., Helmreich, R., & Stapp, J. (1973). A short manner. Heider and Kelley investigated the locus
version of the Attitudes Toward Women Scale (AWS). of causality, whether behavior is caused by some-
Bulletin of the Psychonomic Society, 2(4), 219–220. thing internal or external to the actor (the person
Swim, J. K., & Cohen, L. L. (1997). Overt, covert, and performing the behavior). Later work, by Bernard
subtle sexism: A comparison between the Attitudes Weiner, identified three further causal dimensions
Toward Women and Modern Sexism Scales.
in terms of which attributions can be classified:
Psychology of Women Quarterly, 21, 103–118.
stability, the extent to which causes are stable and
Twenge, J. M. (1997). Attitudes toward women,
permanent versus temporary and fluctuating; con-
1970–1995: A meta-analysis. Psychology of Women
trollability, the extent to which causes can be influ-
Quarterly, 21, 35–51.
enced by the actor; and globality, whether a cause
is global in nature or specific to a given situation.
Of most relevance to the issue of intergroup
attribution biases is locus of causality. An internal
Attribution Biases attribution is any explanation that locates the
cause as being internal to the person, such as per-
Attribution refers to the way in which people sonality, mood, abilities, attitudes, and effort. An
explain their own behavior and that of others. An external attribution is any explanation that locates
attribution bias occurs when someone systemati- the cause as being external to the person, such as
cally over- or underuses the available information the actions of others, the nature of the situation,
when explaining behavior. There is evidence that social pressures, or luck. Thus, if people see a
when we are making judgments about the behav- mother shouting at her child and decides that she
ior of our own group (the ingroup) and that of is doing this because she is an aggressive person,
other groups (outgroups), we show attributional they are making an internal attribution. In con-
biases that favor the ingroup. Specifically, where trast, if they decide that she was reprimanding the
ingroup members are concerned, we explain posi- child for behaving badly, they are making an
tive behaviors in terms of internal characteristics external attribution.
(e.g., personality) and negative behaviors in terms
of external factors (e.g., illness). Conversely,
where outgroup members are concerned, we Individual Attribution Biases
explain positive behaviors in terms of external Kelley’s model is a rather idealized account of how
characteristics and negative behaviors in terms of people make causality judgments. Given that we
internal characteristics. The study of attribution normally have limited time and resources, we have
biases is an essential aspect of group processes and a tendency to use heuristics, or shortcuts, when
intergroup relations because these biases can fuel making social judgments, rather than taking into
negative relations between opposing groups. account all of the available information. As a
Understanding how and why attribution biases result, researchers have observed a number of sys-
arise, however, facilitates the development of tematic biases that are made when people are
interventions to reduce them. assessing the causes of behavior.
This entry outlines the basic theory, discusses There are three well-documented attribution
how it applies in individual and group contexts, biases. The correspondence bias refers to the fact
and describes research showing how attribution that behavior is often viewed as a reflection of an
bias may be mitigated. actor’s corresponding internal disposition even
when it was actually caused by situational factors.
The actor–observer bias arises when we attribute
Attribution Theory
other people’s behavior to internal causes and our
Following the pioneering work of Fritz Heider, Harold own behavior to external causes. Both of these
Kelley developed a theory of causal attribution based effects can be explained by perceptual salience.
Attribution Biases 43

The people being observed are the most salient helped or failed to help the participant when he or
aspect of the situation, as they are actually per- she had fallen off a bike). Among Muslim partici-
forming the action—they and their behavior appear pants, positive behavior of a Muslim (an ingroup
to go together, so an internal attribution is made. member) and negative behavior of a Hindu (an
In contrast, when making self-attributions, we are outgroup member) tended to be attributed to
focused outward and the situation is salient, and causes rated as internal, stable, uncontrollable by
thus we attribute causality for our behavior to others, and global. In contrast, positive behavior
external factors. of a Hindu and negative behavior of a Muslim
The self-serving attribution bias refers to our were typically attributed to causes rated as exter-
tendency to make internal attributions for our nal, unstable, controllable by others, and specific.
successes and external attributions for our fail- Notably, Hindu participants showed considerably
ures. If students excel in an exam, for example, less intergroup bias in attributions, suggesting that
they are likely to think this is because they are these biases are stronger among majority groups
very intelligent, but if they fail, they may attribute than minority groups.
this to the poor quality of their teacher. In con- Research has also considered whether there are
trast to the perceptual processes underlying cor- biases in attributions made for the historical
respondence and actor–observer biases, the actions of entire outgroups. (Non-German) Jewish
self-serving attribution bias has a motivational and (non-Jewish) German participants were asked
basis. We are motivated to view ourselves in a why they thought Germans mistreated Jewish
positive light, to have high self-esteem. Attributing people during the Second World War. Jewish par-
success to internal causes boosts our feelings of ticipants were more likely to attribute the behavior
self-worth, whereas attributing our failures to of the Germans to internal characteristics such as
external causes protects us from feeling bad when German aggression than were German partici-
we do not do well. Together, these processes pants. In a further study, Dutch participants were
enable us to maintain and enhance our self- asked to make internal or external attributions for
esteem. Extending these findings, research has behavior in two historical contexts: Dutch behav-
shown that as well as making attributions that ior toward Indonesians during the colonization
favor the self, we are also motivated to make attri- period (negative ingroup behavior) and German
butions that favor groups to which we belong behavior toward the Dutch during the Second
over groups to which we do not. World War (negative outgroup behavior). Partici­
pants were more likely to make internal attribu-
tions about negative outgroup behavior than
Intergroup Attribution Biases
negative ingroup behavior, and more likely to
Intergroup attribution refers to the ways in which make external attributions about negative ingroup
members of different social groups explain the behavior than negative outgroup behavior.
behavior of members of their own and other social Finally, there is evidence for linguistic inter-
groups. A person attributes the behavior of another group attribution biases. People tend to use rela-
person not simply to individual characteristics, but tively abstract terms to describe the negative
also to characteristics associated with the group to behavior of an outgroup member and the positive
which the other person belongs. Moreover, the behavior of an ingroup member, because this
group membership of the perceiver, or attributor, implies that the behavior is generalized to the
can also affect the intergroup attribution process. personality of the actor. In contrast, people use
Social psychologists have investigated how we relatively concrete terms to describe the negative
make attributions in an intergroup context. Hindus behavior of an ingroup member and the positive
(a minority group) and Muslims (a majority group) behavior of an outgroup member because this
in Bangladesh read scenarios about an individual implies that the behavior is specific to a particu-
from either their ethnoreligious group or the other lar context.
group, and they were instructed to imagine that To summarize, in an intergroup context, we
this person had behaved in either a positive or a tend to make attributions regarding locus of cau-
negative way toward them (e.g., a passerby either sality that favor the ingroup over the outgroup.
44 Attribution Biases

This is a form of self-serving attribution bias, but nationality. This cross-categorization creates four
instead of enabling us to view ourselves in a positive groups. For a Bangladeshi Muslim, the double
light compared to other individuals, it enables us ingroup refers to those who share both group
to view the groups to which we belong positively memberships (other Bangladeshi Muslims), the
compared to other groups. Specifically, we tend to partial ingroups are those who share one group
explain the positive behavior of ingroup members membership (Bangladeshi Hindus and Indian
in terms of internal characteristics but the positive Muslims), and the double outgroup refers to those
behavior of outgroup members in terms of exter- who share neither group membership (Indian
nal characteristics. In contrast, we tend to explain Hindus). People tend to favor double ingroup
the negative behavior of ingroup members in terms members and show the greatest discrimination
of external characteristics, but the negative behav- against the double outgroup. Intergroup bias
ior of outgroup members in terms of internal char- against partial ingroup members, however, is
acteristics. We also have also a tendency to make reduced compared to the double outgroup. Thus,
biased intergroup attributions based on linguistics, seeing an outgroup member as being an ingroup
globality, stability, and controllability. member on a second dimension has benefits for
So why do we make these intergroup attribu- intergroup relations. Research on intergroup attri-
tion biases? According to social identity theory, we bution biases mirrors these findings. Bangladeshi
tend to favor our own group over other groups to Muslim study participants made the most positive
maintain a positive perception of the ingroup and attributions about a Bangladeshi Muslim protago-
therefore maintain a high level of self-esteem. We nist and the most negative attributions about an
make intergroup attribution biases to ensure that Indian Hindu protagonist. Attributions made
our group is perceived in a positive light compared about Bangladeshi Hindus and Indian Muslims
to other groups. Three findings support this social were, however, significantly more positive than
identity explanation. First, making group member- those made about Indian Hindus.
ship salient prior to completing an intergroup In sum, intergroup attributional biases arise
attribution task increases the extent to which par- because of our motivation to maintain a positive
ticipants show intergroup attribution biases. social identity, and these biases contribute to the
Second, intergroup attribution biases are stronger maintenance and exacerbation of conflict between
among participants who highly identify with their groups. Research has shown, however, that chang-
ingroup. Third, it has been demonstrated that ing our perceptions of intergroup categories
making internal attributions about ingroup mem- through cross-categorization can lead to reduc-
bers and making global attributions about the tions in intergroup attribution biases. This research
negative behavior of outgroup members predicts therefore makes an important contribution to our
higher self-esteem. understanding of how intergroup relations can be

Reducing Intergroup Attribution Biases Rhiannon N. Turner and Miles Hewstone

According to social identity theory, making our See also Cross-Categorization; Discrimination; Social
group membership salient increases intergroup Identity Theory
bias, as we are motivated to maintain a positive
perception of our own group relative to other
groups. To reduce attributional bias, it is therefore Further Readings
necessary to change the nature of categorization. Doosje, B., & Branscombe, N. R. (2003). Attributions for
One way of doing this is cross-categorization, the negative historical actions of a group. European
which involves crossing a dichotomous categoriza- Journal of Social Psychology, 33, 235–248.
tion with a second categorization. In the case of Hewstone, M. (1990). The “ultimate attribution error”?
Hindus and Muslims in Bangladesh, for example, A review of the literature on intergroup causal
it is possible to introduce a second categorization, attribution. European Journal of Social Psychology,
the distinction between Bangladeshi and Indian 20, 311–335.
Authoritarian Personality 45

Islam, M. R., & Hewstone, M. (1993). Intergroup homemakers, longshoremen, civic volunteers, vet-
attributions and affective consequences in majority erans, psychiatric patients, and prisoners, among
and minority groups. Journal of Personality and Social others, from the West Coast of the United States
Psychology, 64, 936–950. For this reason, the research taught Americans much
about their own authoritarianism and prejudice.
Authoritarian Personality The Frankfurt/Berkeley school, as the group
was called, viewed following hateful authorities as
being at least as problematic as hateful leadership
Why would a progressive society line up behind a
itself, for without assent and cooperation, what
ruler who invades other nations unprovoked? What
power does a leader have? Their approach was
would lead ordinary people to carry out orders that
thus one of the first to prioritize understanding
risked their nation’s future in order to commit geno-
mass political psychology. According to the author-
cide? Nazi Germany posed such questions to many
itarian personality theory (APT) that Adorno,
social scientists. Authoritarian personality theory
Frenkel-Brunswik, Levinson, and Sanford pub-
(APT), based in psychodynamic theory, was devel-
lished in The Authoritarian Personality in 1950,
oped to explain these behaviors and their psycho-
three elements are necessary to produce an author-
logical underpinnings. Studies based on APT have
itarian personality: (1) being raised in a culture
shown that prejudice is related to the outlook of the
that vilifies certain groups (e.g., European anti-
people who hold such views rather than to charac-
Semitism and U.S. racism), (2) needing to be loved
teristics of the groups they disdain. Thus the social
by one’s parents, and (3) having parents who are
significance, testable hypotheses, and intellectual
punitive and unaffectionate.
ambition of APT has drawn much attention and
The psychodynamic process states that when
criticism and inspired a wide variety of new research.
parents scorn their children, children adopt the
In addition, the cross-culturally robust association
prejudices of their parents and society in an
of authoritarianism with prejudice, stereotyping,
attempt to become pleasing to their parents. As
political attitudes and behavior, and social and
children try to gain moral acceptability by obeying
political values continues to inspire research in per-
authorities who are prejudiced, they adopt the pre-
sonality and social psychology, political science,
dilections for conformity, blind submissiveness to
sociology, and political psychology. This entry
authority, and intolerance of difference. This
examines the concept, supporting evidence, criti-
makes them especially vulnerable to messages
cisms, and responses to these critiques.
from authorities that denigrate the weak and the
deviant. In expressing such prejudices, children
Historical and Theoretical Context
can view themselves as acceptable. Hence, the
During World War II, scholars Theodor Adorno combination of psychological motivations, the cul-
and Elsa Frenkel-Brunswik, who were German tural context of prejudicial ideologies, and particu-
refugees, joined American psychologists Daniel J. lar family practices account for how cultures
Levinson and R. Nevitt Sanford at the University transmit prejudice across generations.
of California, Berkeley. The group was given fund-
ing from the American Jewish Committee to
Evidence of the Authoritarian Personality
research the psychological roots of anti-Semitism.
However, their study became substantially broader, Interview studies of adults by Elsa Frenkel-
representing the intellectual ambition to solve Brunswik attempted to explicate these psychody-
major societal problems by understanding the namic processes. Such evidence is now met with
interplay of human development, psychology, and skepticism because of concerns about retrospective
societies. Their theorizing incorporated two pre- memory and interviewer biases. However, the
dominant schools of thought: psychodynamics, existence of an authoritarian “personality,” or
and culture and personality, and it addressed rela- syndrome of traits, including conformity, submis-
tions within families, between groups, and between sion, and intolerance, was demonstrated with stan-
leaders and their societies. Participants in this dard personality techniques, including interviews,
research included professional men and women, projective tests, and extensive scale development.
46 Authoritarian Personality

Adorno and his colleagues’ research, especially but also because of APT’s moral and political
that of Daniel Levinson, showed that people differ implications and of developments in psychological
reliably from one another in the general tendency theorizing and research methods.
to be prejudiced. That is, those scoring high on the One major criticism of APT concerns the valid-
F-scale (fascism scale) also tended to score high on ity of its personality measures. Richard Christie
anti-Semitism scales; on generalized ethnocentrism and Marie Jahoda, among others, have noted that
scales that tap prejudice against “Negroes,” the scales developed by Adorno and his colleagues
Mexicans, Japanese, “Okies,” immigrants, and are not balanced with equal numbers of protrait
foreigners; and on scales measuring patriotism and and contrait items. That is, the scales contain only
political–economic conservatism. In fact, preju- items with which someone highly authoritarian
diced individuals are likely to endorse logically would agree. For this reason, it is unclear whether
contradictory statements, so long as the statements the scales simply measure authoritarianism as
indicate culturally normed disdain for members of response acquiescence (the tendency to simply
excluded groups. For example, Adorno and his agree with statements), or whether the contents of
colleagues found that authoritarians are likely to the scales matter. Further, this measurement prob-
endorse both of these anti-Semitism scale state- lem can inflate correlations among different scales
ments: “Districts containing many Jews always because if some participants are “yay-sayers” and
seem to be smelly, dirty, shabby, and unattractive” others are “nay-sayers,” that would produce posi-
and “Jews seem to prefer the most luxurious, tive correlations among different scales regardless
extravagant, and sensual way of living.” Such find- of item content.
ings suggest that authoritarianism does not stem Another kind of criticism is both political and
from rational beliefs, but rather from motivation theoretical. People who view patriotism and conser-
or cognitive style. vatism as prosocial and moral may be discomfited
Prejudiced people often feel that their percep- by the finding that patriotism and conservatism cor-
tions and feelings about denigrated groups stem relate strongly with forms of prejudice the West
from qualities of denigrated groups themselves. came to disapprove of following the Nazi genocide,
But research on authoritarianism has documented such as anti-Semitism and anti-Black racism. Scholars
that people prejudiced against one group tend to including Edward Shils and Milton Rokeach think
be prejudiced against other groups who are disre- that extreme intolerance could be exhibited on both
spected in their societies. This finding provides a ends of the political spectrum, and they believe that
completely different interpretation of the cause of communists, for example, should score high on
prejudice than that of naïve psychology. Rather intolerance. In fact, Adorno and his colleagues had
than prejudice being due to properties of the vili- postulated the same idea, but they found no empiri-
fied group, such as their immorality or crudeness, cal evidence for left wing authoritarianism. In 1960,
it means that prejudice stems from the psychologi- Milton Rokeach proposed the D-scale (dogmatism
cal outlook of the perceiver. This finding therefore scale) as an alternative to the F-scale in an attempt
provides psychology with an agenda to research to capture intolerance among both right and left
what, exactly, that outlook is and how it works. wingers, but found comparable empirical results for
the F-scale and the D-scale.
Another criticism of APT is that there is very
Criticisms of Authoritarian Personality Theory
little evidence for the psychodynamic processes it
In fact, the finding that certain individuals are posits. In fact, there is evidence against the hypoth-
more robustly prejudiced, conservative, insecure, esis that punitive parents produce authoritarian
and punitive than others had been empirically children. From the 1950s on, as psychology came
documented early in the 20th century by pro-Nazi to insist on empirical evidence and experimenta-
researchers (e.g., Jaensch) and anti-Nazi (e.g., tion and reject the unobservable unconscious
Lenski) researchers in Europe and the United States processes associated with Freudianism, more psycho­
APT attracted both more attention and more criti- logists were ready to disregard Elsa Frenkel-
cism than that research. This may be because both Brunswik’s psychodynamic research and Theodor
world wars showed the costs of intergroup hate, Adorno’s rejection of positivism.
Authoritarian Personality 47

Finally, the research focus of APT on individual it does not show parallel patterns with RWA.
differences produces a logical contradiction: How Reviews by William Stone corroborate the fact that
can normative prejudice in a society be explained by authoritarianism and the tendency to be prejudiced
features that only some people in that society have? against groups characterizes conservatives more
Tom Pettigrew’s 1958 dissertation examined racism than liberals. In the Soviet Union and satellite
among South African and American Whites and nations, Walter Stephan and his colleagues have
showed that authoritarianism does little to explain shown that those higher on RWA were more likely
racist behavior when racism is normalized in a cul- to endorse communism and the Communist Party,
ture. This criticism led many social psychologists to consistent with the idea that authoritarians con-
reject consideration of individual differences in preju- form to the norms promoted by authorities in their
dice. It can also be said that APT does not answer societies. Recently, Alain van Hiel and his col-
important questions about the culture of prejudice, leagues found that extreme left wing activists in
such as why particular groups get vilified in the first Western Europe have very high scores on a new left
place, why prejudice against the same group rises and wing authoritarianism scale, and this correlates
falls, nor how prejudicial ideologies may change or with liberal economic views.
spread outside of socialization, although answers to
such questions are still being generated. Despite Psychodynamics and Personality Development
including some progressive theorizing about gender,
The Authoritarian Personality shares a cultural short- Consistent with the psychodynamic view, but
coming with its contemporaries: It did not develop a also with other socialization and genetic develop-
sexism scale as an aspect of group prejudice. ment theories, Altemeyer showed that people within
families have similar levels of authoritarianism.
Responses to Criticisms More recently, Christopher Weber and Christopher
Federico have shown that RWA corresponds with
After decades of neglect, the major questions raised anxious attachment. In terms of general personality,
by authoritarian research concerning the normativ- studies in several countries show that RWA corre-
ity of prejudice, how psychological motives lead to lates with being less open to experience and more
prejudice, how prejudice is socialized, and the rela- conscientious. Thus, although APT appears wrong
tion of culture to intergroup relations have been in the particulars concerning socialization, authori-
reconsidered by a wide variety of scholars. tarianism does correspond with people’s orientations
toward close relationships and their temperament.
Scale Redevelopment
Canadian psychologist Bob Altemeyer reviewed Other Conceptualizations
research on authoritarianism and determined that
Rather than accept the description of authori-
the most central elements of the authoritarian syn-
tarianism as a prejudice syndrome or as essentially
drome were authoritarian submission, convention-
political, theorists have sought to identify its core
alism, and authoritarian aggression. He developed
psychology by emphasizing either a socioemo-
balanced scales called Right Wing Authoritarianism
tional or cognitive orientation. Those with a
(RWA) that include all three concepts and are reli-
socioemotional orientation have conceived of
able, valid, and widely used around the world.
authoritarianism as whether an individual is tough-
RWA corresponds to the public’s political behav-
minded or tender-minded (Ted Goertzel, Hans
ior, including voting in many societies. The scales
Eysenck), is easily threatened (David Winter, Bill
have distinguished among the voting records of
Petersen), is uncomfortable and unsuccessful with
North American legislators, evidence that connects
personal autonomy (Detlef Oesterreich), sees con-
authoritarianism with national leadership.
formity as the means to social order (Stanley
Feldman), or perceives the world to be a dangerous
Scale Balance and Left Wing Authoritarianism
place (John Duckitt). The psychological habits that
Altemeyer also tried to develop a left wing may underlie generalized conservatism and preju-
authoritarian measure, but in convenience samples dice may include fear of uncertainty (Michael
48 Aversive Racism

Hogg), intolerance of ambiguity (Glenn Wilson), Duckitt, J. (2001). A dual-process cognitive-motivational

cognitive rigidity (Richard Christie), need for theory of ideology and prejudice. In M. P. Zanna
cognitive closure (Arie Kruglanski), and closed- (Ed.), Advances in experimental social psychology
mindedness (Milton Rokeach). John Duckitt’s (Vol. 33, pp. 41–113). San Diego, CA: Academic
conception of authoritarianism also addresses Press.
how different elements of cultural context may Eckhardt, W. (1991). Authoritarianism. Political
be important in the development of authoritari- Psychology, 12, 97–124.
anism. Duckitt argues that in more collectivist Marcus, G. E. (Ed.). (2005). Authoritarianism [Special
issue]. Political Psychology, 26.
societies, conformity is emphasized, and along
with the presumption that the world is threaten-
ing, this context should especially lead to the
development of authoritarianism. Correlational
models in several countries are consistent with Aversive Racism
this theory. Finally, Phil Tetlock and Jim Sidanius
have each developed approaches to political cog- Aversive racism is a form of bias that is not overtly
nition to account for extremes on both the right expressed but may reflect the attitudes of a sub-
and the left. stantial portion of people in societies that have
strong egalitarian traditions and norms. Much of
the research on aversive racism has focused on the
Authoritarianism and Culture orientation of Whites toward Blacks in the United
Authoritarianism in individuals correlates robustly States, but similar attitudes have been found
across nations with their right wing political affili- among members of dominant groups in other
ation and voting, prejudice against women, gays, countries with strong contemporary egalitarian
immigrants, foreigners, and subordinated ethnic values but discriminatory histories or policies. In
and religious groups. Cultures can also be consid- contrast to the traditional form of racism, which
ered more or less authoritarian, and can become is expressed openly and directly, aversive racism
more or less authoritarian depending on how inse- operates in subtle and indirect ways. For example,
cure they are. For example, cultures that privilege the negative feelings that aversive racists have
conformity and hierarchicality are considered by toward Blacks do not manifest themselves in open
some to be more authoritarian, and periods of war hostility or hatred. Instead, aversive racists’ reac-
may produce more authoritarian behavior, as tions may involve discomfort, anxiety, and/or
seen in content analyses of popular culture, fear. That is, they find Blacks “aversive,” while at
endorsement of leaders, and voting patterns. the same time rejecting any suggestion that they
Authoritarianism may be conducive to certain might be prejudiced. Despite its subtle expression,
aspects of group living, such as cooperation and aversive racism has consequences that are as sig-
ingroup identification. nificant and pernicious (e.g., the restriction of
minorities’ economic opportunities) as those of
Felicia Pratto the traditional, overt form. This entry provides a
fuller description of aversive racism and its expres-
See also Conservatism; Dogmatism; Need for Closure; sions, then looks at strategies for combating it.
Prejudice; Right Wing Authoritarianism

Nature of the Attitudes

Further Readings A critical aspect of the aversive racism framework
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & is the conflict between aversive racists’ denial of
Sanford, R. N. (1950). The authoritarian personality. personal prejudice and underlying unconscious
New York: Norton. negative feelings toward and beliefs about particu-
Altemeyer, B. (1988). Enemies of freedom: Understanding lar minority groups. For example, because of cur-
right wing authoritarianism. San Francisco: Jossey- rent cultural values in the United States, most
Bass. Whites have strong convictions concerning fairness,
Aversive Racism 49

justice, and racial equality; however, because of a but do not have significant implicit prejudice or
range of normal cognitive, motivational, and stereotypes. Consistent with the aversive racism
sociocultural processes that promote intergroup framework, whereas the majority of Whites in the
biases, most Whites also have some negative feel- United States appear “nonprejudiced” on self-
ings toward or beliefs about Blacks. They may be report (explicit) measures of prejudice, a very large
unaware of these feelings or try to deny them to proportion of Whites also demonstrate implicit
retain a self-image as unprejudiced, but when racial biases. Overall, studies have found that
engaged in social categorization, for example, they Whites’ generally negative implicit attitudes and
will find that their cultural stereotypes are sponta- stereotypes are largely dissociated from their typi-
neously activated. cally more positive overt expressions of their atti-
tudes and beliefs about Blacks.
The aversive racism framework also helps to
Identifying Aversive Attitudes
identify when discrimination against Blacks and
Generally, aversive racists may be identified by other minority groups will or will not occur.
a constellation of characteristic responses to racial Whereas old-fashioned racists exhibit a direct and
issues and interracial situations. First, aversive rac- overt pattern of discrimination, aversive racists’
ists, in contrast to old-fashioned racists, endorse actions may appear more variable and inconsis-
fair and just treatment of all groups. Second, tent. Sometimes they discriminate (manifesting
despite their conscious good intentions, aversive their negative feelings), and sometimes they do not
racists unconsciously harbor feelings of uneasiness (reflecting their egalitarian beliefs). Nevertheless,
toward those of other races (e.g., Blacks) and thus their discriminatory behavior is predictable.
try to avoid interracial interaction. Third, when
interracial interaction is unavoidable, aversive rac-
Predicting Aversive Behavior
ists experience anxiety and discomfort, and conse-
quently they try to disengage from the interaction Because aversive racists consciously recognize
as quickly as possible. Fourth, because part of the and endorse egalitarian values and because they
discomfort that aversive racists experience is due truly aspire to be unprejudiced, they will not act
to a concern about acting inappropriately and inappropriately in situations with strong social
appearing prejudiced to themselves and others, norms when discrimination would be obvious to
aversive racists strictly adhere to established rules others and to themselves. Specifically, studies have
and codes of behavior in interracial situations that shown that when they are presented with a situa-
they cannot avoid. Finally, their feelings will get tion in which the normatively appropriate response
expressed, but in subtle, unintentional, rationaliz- is clear, in which right and wrong are clearly
able ways that disadvantage minorities or unfairly defined, aversive racists will not discriminate
benefit the majority group. Nevertheless, in terms against Blacks. In these contexts, aversive racists
of conscious intent, aversive racists intend not to will be especially motivated to avoid feelings,
discriminate against people of color—and they beliefs, and behaviors that could be associated
behave accordingly when it is possible for them to with racist intent. Wrongdoing of this type would
monitor the appropriateness of their behavior. directly threaten their image of themselves as non-
Recent research in social cognition has yielded prejudiced.
new techniques—such as the Implicit Association Aversive racists still possess unconscious nega-
Test, which uses response times to pairs of stimu- tive feelings and beliefs, however, which will even-
li—for tapping the “implicit” stereotypic or evalu- tually be expressed in subtle, indirect, and
ative (e.g., good–bad) associations that people rationalizable ways. For instance, discrimination
have toward other groups, but possibly without will occur in situations in which the normative
full awareness. These techniques are very useful structure is weak, the guidelines for appropriate
for distinguishing between aversive racists, who behavior are vague, or the basis for social judgment
endorse egalitarian views and unprejudiced ideolo- is ambiguous. In addition, discrimination will occur
gies but harbor implicit racial biases, and unpreju- when an aversive racist can justify or rationalize a
diced people, who also endorse egalitarian values negative response on the basis of some factor other
50 Aversive Racism

than race. Studies show that under these circum- bias that focus on the immorality of prejudice and
stances, White aversive racists may engage in illegality of discrimination are not effective for
behaviors that ultimately harm Blacks, but in ways combating it. Aversive racists recognize that prej-
that allow the racists to maintain their self-image as udice is bad, but they may not recognize that they
unprejudiced and that insulate them from recogniz- are prejudiced.
ing that their behavior is not color-blind. Nevertheless, aversive racism can be addressed
Evidence in support of the aversive racism with techniques aimed at its roots at both the indi-
framework comes from a range of paradigms. For vidual and collective levels. At the individual level,
instance, White bystanders who are the only wit- strategies to combat aversive racism can be directed
nesses to an emergency (and thus are fully respon- at unconscious attitudes. For example, extensive
sible for helping) are just as likely to help a Black training to create new, counterstereotypical associa-
victim as a White victim. However, when White tions with social categories (e.g., Blacks) can inhibit
bystanders believe that others also witness the the unconscious activation of stereotypes, an element
emergency (distributing the responsibility for help- of aversive racists’ negative attitudes. In addition,
ing), they are less likely to help a Black victim than aversive racists’ conscious attitudes, which are already
a White victim. In personnel or college admission favorable, can be instrumental in motivating change.
selection decisions, Whites do not discriminate on Allowing aversive racists to become aware, in a non-
the basis of race when candidates have very strong threatening way, of their unconscious negative atti-
or weak qualifications. Nevertheless, they do dis- tudes, feelings, and beliefs can stimulate self-regulatory
criminate against Blacks when the candidates have processes. Such processes not only elicit immediate
mixed qualifications. In these circumstances, aver- deliberative responses reaffirming conscious unpreju-
sive racists weigh the positive qualities of White diced orientations (such as increased support for
applicants and the negative qualities of Black policies that benefit minority groups), but also pro-
applicants more heavily in their evaluations, which duce, with sufficient time and experience, reductions
provide justification for their decisions. in implicit negative beliefs and attitudes.
Analogously, aversive racists have more difficulty At the intergroup level, interventions may be
discounting incriminating evidence that is declared targeted at processes that support aversive racism,
inadmissible when evaluating the guilt or innocence such as ingroup favoritism. One approach, repre-
of Black relative to White defendants in studies of sented by the common ingroup identity model,
court decisions. In interracial interactions, Whites’ generally proposes that if members of different
overt behaviors (e.g., verbal behavior) primarily groups are induced to conceive of themselves more
reflect their expressed, explicit racial attitudes, as an alternative single, superordinate group rather
whereas their more spontaneous and less controllable than as two separate groups, attitudes toward for-
behaviors (e.g., their nonverbal behaviors) are related mer outgroup members will become more positive
to their implicit, generally unconscious attitudes. through processes involving pro-ingroup bias. Thus,
changing the basis of categorization from race to an
alternative dimension can alter who is grouped as
Combating Aversive Racism
“us” and who is grouped as “them,” undermining
Traditional prejudice-reduction techniques have been a contributing force for contemporary forms of rac-
concerned with changing conscious attitudes—old- ism, such as aversive racism.
fashioned racism—and obvious expressions of bias. For instance, Black interviewers are more likely
Attempts to reduce this direct, traditional form of to obtain the cooperation of White respondents
racial prejudice have typically involved educational when they emphasize their common group mem-
strategies to enhance knowledge and appreciation of bership (such as shared university identity, as indi-
other groups (e.g., multicultural education programs), cated by insignia on their clothes) than when they
emphasize norms that prejudice is wrong, and involve do not. Intergroup interaction within the guide-
direct (e.g., mass media appeals) or indirect (disso- lines of the contact hypothesis (i.e., the idea that
nance reduction) attitude change techniques. How­ contact between groups improves intergroup rela-
ever, because aversive racism is pervasive, subtle, and tions) and antibias interventions with elementary
complex, the traditional techniques for eliminating schoolchildren that emphasize increasing their circles
Aversive Racism 51

of inclusion can also reduce bias through the processes See also Common Ingroup Identity Model;
outlined in the common ingroup identity model. Discrimination; Implicit Association Test (IAT);
The manifestations of aversive racism are more Modern Racism; Prejudice; Racial Ambivalence
subtle than are those of old-fashioned racism, but Theory; Racism; Symbolic Racism
aversive racism has consequences as significant as
blatant bias. Even though it is expressed in indirect
and rationalizable ways, it operates to systemati- Further Readings
cally restrict opportunities for Black members of
other traditionally underrepresented groups. In Dovidio, J. F., & Gaertner, S. L. (2004). Aversive racism.
addition, because aversive racists may not be aware In M. P. Zanna (Ed.), Advances in experimental social
of their implicit negative attitudes and only dis- psychology (Vol. 36, pp. 1–51). San Diego, CA:
criminate against Blacks when they can justify their Academic Press.
behavior on the basis of some factor other than Gaertner, S. L., & Dovidio, J. F. (1986). The aversive
form of racism. In J. F. Dovidio & S. L. Gaertner
race, they will commonly deny any intentional
(Eds.), Prejudice, discrimination, and racism
wrongdoing when confronted with evidence of
(pp. 61–89). Orlando, FL: Academic Press.
their bias. To the extent that minority group mem-
Gaertner, S. L., & Dovidio, J. F. (2000). Reducing
bers detect expressions of aversive racists’ negative
intergroup bias: The common ingroup identity model.
attitudes in subtle interaction behaviors (e.g., non-
Philadelphia: Psychology Press.
verbal behavior) and attribute the consequences of Greenwald, A., McGhee, D., & Schwartz, J. (1998).
aversive racism to blatant racism, aversive racism Measuring individual differences in implicit cognition:
also contributes substantially to interracial distrust, The Implicit Association Test. Journal of Personality
miscommunication, and conflict. and Social Psychology, 74, 1464–1480.
Nevertheless, aversive racism can be addressed Kovel, J. (1970). White racism: A psychohistory.
by encouraging increased awareness of uncon- New York: Pantheon.
scious negative feelings and beliefs, emphasizing Pettigrew, T. F., & Meertens, R. W. (1995). Subtle and
alternative forms of social categorization around blatant prejudice in western Europe. European
common group membership, and providing Journal of Social Psychology, 25, 57–76.
appropriate intergroup experiences to support Son Hing, L. S., Chung-Yan, G., Hamilton, L., &
the development of alternative implicit attitudes Zanna, M. P. (2008). A two-dimensional model that
and stereotypes and reinforce common identities. employs explicit and implicit attitudes to characterize
prejudice. Journal of Personality and Social
John F. Dovidio and Samuel L. Gaertner Psychology, 94(6), 971–987.
evil actions would manifest. They seem like rather
Banality of Evil ordinary people.
Assume that this realization is true. The disturb-
A recent morning’s newspaper had a charming ing consequence of this realization is that large-
snapshot of a laughing soldier playing an accor- scale actions of evil, such as genocide, do not
dion, surrounded by equally cheerful, laughing require large armies of evil individuals to carry
women, all clearly having a carefree time of it. But them out. Instead, ordinary people can become
the reader soon discovers a shocking fact. It is a entrained in the processes that produce the evil
photo of a playful off-duty moment for the staff of outcomes and thereby make those terrible out-
the Auschwitz death camp, which came from a comes possible. Two questions need to be asked
photo album documenting many such ordinary here: How is it that these ordinary people become
moments. The article’s author, Neil Genzlinger, enlisted in the process, and what are the conse-
comments that “yes, the genocide was conducted quences for them of their enlistment? Psychological
by real human beings who kicked back after a theory and research provide at least partial answers
day’s work, flirted with the ladies, shared a joke, to both of these questions.
played with the dog” (2008, B14). In psychology, the “banality of evil” notion has
The reader’s shock is much like the reaction become linked to the famous experiments by
that Hannah Arendt had observing the war crimes Stanley Milgram, which are described in the entry
trial of the Nazi Adolph Eichmann. Arendt was “Obedience to Authority.” Coincidentally, both
surprised, disconcerted, deeply unsettled. She Eichmann in Jerusalem and the first reports of the
expected the man, who had sent countless Jews to Milgram research appeared in 1963, so the world
their deaths, to look and act evil, to embody faced the ordinariness of the Nazi functionary
“evilness.” But instead she was struck by his ordi- Eichmann at about the time it faced the fact that,
nariness, his depthless normality. Her book if ordered to do so, many ordinary citizens of New
Eichmann in Jerusalem characterized this as the Haven, Connecticut, were willing to administer
“banality of evil.” presumably dangerous, perhaps life-threatening,
The book initially generated a storm of contro- shocks to another person in a learning experiment.
versy from those who thought Arendt was trivial- (The use of presumably here is important. Many
izing the evils that the Third Reich committed. descriptions of the Milgram experiment ignored
Now it is read as calling attention to the fact that various cues that it was, in fact, acceptable for the
when acts of evil are committed by an organiza- respondent to continue giving the shocks because
tion, the various actors in the organization will not the experimenter had asserted that the shocks
show all, or even many, of the personal character- would be painful “but cause no permanent
istics that a single individual perpetrating similar damage.”) Later, Philip Zimbardo’s well-known

54 Banality of Evil

Stanford prison experiment (described in the entry really being hurt, and this sort of reassurance from
of that name) was read as showing that under- a person who was assumed to have expertise about
graduate students, role-playing prison guards in a this enabled some respondents to continue. The
realistic simulation of a prison environment, were importance of this expert reassurance was demon-
also willing to engage in quite cruel actions toward strated by Milgram himself. In one condition,
other students who were in the prisoner role. This Milgram arranged to have the regular experi-
added a distressing new element to the findings of menter called out of the room while the earlier
the Milgram studies. In the Milgram studies, the shocks in the series were given. The experimenter
respondents’ actions were those that they were asked another respondent, who had been doing a
directed to take by the experimenter, who was record-keeping job, to take his place and continue
present to see that they obeyed his orders. In con- the experiment. That person, who was actually a
trast, the students in Zimbardo’s experiment them- confederate of Milgram’s, told the teacher to con-
selves designed and then independently enacted tinue administering the shocks. At some point in
many of the cruel actions that were taken. That the learner’s protests, the typical respondent refused
evil actions were within the repertoire of at least to continue; this experimenter did not have the
many people seemed well established. expertise to provide valid assurances that continu-
Certainly Milgram thought so. In a letter to the ation was safe. A further interesting response was
sponsor of his research, Milgram wrote: shown by several of these respondents. The replace-
ment experimenter stepped to the shock generator,
In a naïve moment some time ago, I once won- announcing that if the respondent would not con-
dered whether in all of the United States a vicious tinue the process, then he would. These respon-
government could find enough moral imbeciles dents moved to restrain the experimenter from
to meet the personnel requirements of a national doing so, thereby protecting the learner from
system of death camps, of the sort that were harm! Suddenly a very different image of the
maintained in Germany. I am now beginning to respondents emerges. In a similar fashion, one
think that the full complement could be recruited could point to several elements of Zimbardo’s
in New Haven. A substantial proportion of peo- study that were critical to producing the disturbing
ple do what they are told to do, irrespective of behavior of his student guards.
the content of the act, and without pangs of con- So it is possible to contend that the Milgram
science, so long as they perceive that the com- and Zimbardo studies’ subjects were not display-
mand comes from a legitimate authority. (quoted ing the sort of full-blown evil behavior that
in Blass, p. 100) Eichmann displayed. This said, a disturbing possi-
bility remains. Had those individuals remained in
For Americans, this unpleasant conclusion has the experimental contexts over longer time peri-
received intermittent confirmation: My Lai and ods, they might have become habituated to what
Abu Ghraib are familiar names. However, many they were doing, and their ethical concerns might
social psychologists would contest a generalization have faded away as they adapted to their tasks. Or
about the ease with which people could be brought they might have dehumanized the victims by
to harm others. They would point to some more attaching labels such as “rodents” to them. Or
particular aspects of the Milgram experiments that they might have cast them as members of out-
perhaps allowed the respondent to continue to groups who threatened the existence of the ingroup.
administer the shocks because those shocks were By means of these mechanisms, they would eventu-
not going to do permanent harm to the person ally do voluntarily and independently what they
receiving them. One key to understanding the were coerced to do by the situations in which they
Milgram situation is this: When the learner pro- found themselves. They might, in other words,
tested and asked that the experiment be stopped, have moved toward becoming evil-doers.
the experimenter heard this as well as the respon- An insight can be drawn from this. The hierar-
dent, and the experimenter instructed that the chical social components of organizations can be
experiment continue. Respondents would proba- structured in ways that bring about tremendously
bly read this as telling them that the learner wasn’t destructive harm to individuals. Of course, we
Black Sheep Effect 55

know that this is possible when the organization’s Darley, J. M. (1996). How organizations socialize
leaders seek to bring about this harm. That is what individuals into evildoing. In D. M. Messick &
the Nazi regime of Adolph Hitler and his collabo- A. E. Tenbrunsel (Eds.), Codes of conduct: Behavioral
rators has shown us. What we must now realize is research into business ethics (pp. 13–43). New York:
that organizations can drift into harm-doing prac- Russell Sage.
tices even without most individuals in the organi- Genzlinger, N. (2008, April 26). Smiling, everyday faces
zations desiring that this come about. of the killers at Auschwitz [Television review]. New
How this occurs is a topic that social psycholo- York Times, p. B14.
Milgram, S. (1974). Obedience to authority: An
gists are now studying, both experimentally and in
experimental view. New York: Harper & Row.
observational studies. They begin with the under-
Zimbardo, P. (2007). The Lucifer effect: Understanding
standing that people seek to discover how to con-
how good people turn evil. New York: Random House.
duct themselves in organizations by observing the
actions of the people around them. This is a ver-
sion of what social psychologists call the situa-
tional perspective. In the organizational instance
of this perspective, people look to other, more Black Sheep Effect
experienced people to “see how we do things
here.” When they do this, how others behave may In everyday language, a “black sheep” is a group
lead them ethically astray, and they may not member who is undesirable and stands out from
become aware of it. Organizations can fragment the group in such a way as to attract disapproval
tasks and parcel them out among different indi- from the rest of the group. In social psychology,
viduals so that no one individual is aware that the the term black sheep effect, coined by José
sum of the fragments produces harmful outcomes. Marques, refers to a more specific phenomenon in
Also, contagion phenomena can exist. Suppose which someone who is socially undesirable (unlik-
others are “making their numbers” by committing able) is liked less if he or she is a member of your
actions that the new person thinks are ethically group (an ingroup member) than if he or she is a
bad. But suppose that the new person also sees member of a group to which you do not belong
that these rule-bending individuals are getting (an outgroup member). Conversely, someone who
bonuses and promotions from their superiors. It is socially desirable (likable) is liked more if he or
will be very hard for the new person to resist the she is a member of your ingroup rather than a
temptation to bend the rules. This in turn may lead member of an outgroup. Put another way, socially
to escalating pressures in which rule bending turns desirable ingroup targets are judged more favor-
to rule breaking. ably than socially desirable outgroup targets,
while socially undesirable ingroup targets are
John Darley judged less favorably than socially undesirable
See also Conformity; Dehumanization/Infrahumanization;
outgroup targets. Likable and unlikable ingroup
Ingroup Allocation Bias; Intergroup Violence; Minimal members are judged more extremely than likable
Group Effect; Obedience to Authority; Stanford Prison and unlikable outgroup members.
Experiment The black sheep effect usually emerges in asso-
ciation with a more favorable evaluation of the
ingroup than the outgroup as a whole (ingroup
Further Readings bias). The black sheep effect is also more pro-
Arendt, H. (1963). Eichmann in Jerusalem: A report on nounced (a) if the person doing the judging identi-
the banality of evil. New York: Viking. fies strongly with his or her ingroup, (b) if the
Blass, T. (2004). The man who shocked the world: The dimension on which the target is evaluated is
life and legacy of Stanley Milgram. New York: Basic highly relevant to establishing an overall favorable
Books. evaluation of the ingroup in comparison to a rele-
Darley, J. M. (1992). Social organization for the vant outgroup, and (c) if the situation is one in
production of evil. Psychological Inquiry, 3, which intergroup differentiation in favor of the
199–218. person’s ingroup is important.
56 Black Sheep Effect

Social Identity and Ingroup Favoritism between ingroup and outgroup categories. In pur-
suit of positive differentiation of the ingroup as a
Explanation of the black sheep effect draws on social
whole from the outgroup, people adopt a descrip-
identity theory and self-categorization theory. One
tive focus, in which they focus on and emphasize
basic assumption of these theories is that, with few
intergroup differences and pay little attention to
exceptions, when group membership is salient (when
intragroup differences. A descriptive focus directs
it is the psychological basis of information process-
attention to those characteristics of a person (e.g.,
ing, self-conception, and behavior), people engage in
born in Lisbon, Portugal) that allow category
ingroup-serving perceptual and judgmental biases
assignment (e.g., this person is more likely to be
and hold partisan intergroup attitudes. This ingroup
Portuguese than British) rather than to characteris-
favoritism generates ethnocentrism and leads people
tics that allow one to determine whether the
to prefer ingroup members and features to outgroup
person is likable or unlikable.
members and features. It is not surprising, then, that
However, in some situations, there may be
people evaluate ingroup members in general, and
salient ingroup members who display behaviors
socially desirable ingroup members in particular,
that conflict with the expectations associated with
more favorably than similar outgroup members.
a positive ingroup image—for example, dishonest
From this, one might suppose that overall one
members of a political party, traitors, or otherwise
would favor a socially desirable or undesirable
socially undesirable ingroup members. In this con-
ingroup member over a socially desirable or unde-
text, people adopt a prescriptive focus. This focus
sirable outgroup member; after all, ingroup mem-
directs attention to characteristics that are imbued
bers should be favored over outgroup members.
with value but are not associated with a particular
This is where the black sheep effect is counterintui-
social category (e.g., being dishonest). Thus, a pre-
tive and in conflict with social identity theory—as
scriptive focus allows one to determine whether a
indicated above, socially undesirable ingroup
person is socially desirable or undesirable but does
members are disliked more than socially undesir-
not allow one to determine the social category to
able outgroup members.
which he or she belongs (e.g., whether the person
Proponents of the black sheep effect believe that
is British or Portuguese).
this conflict is only apparent. They argue that
While adopting a prescriptive focus, people
socially desirable ingroup members support the
react favorably toward ingroup members whose
ingroup’s overall positive image and thus attract
behavior supports their conviction that the ingroup
positive reactions from other members of the group,
is “right” or is “better than” the outgroup.
whereas socially undesirable ingroup members
Concomitantly, people react negatively toward
undermine such an image and thus attract negative
“deviant” ingroup members because their behavior
reactions. Rather than being in contradiction to the
departs from standards that sustain the perceived
social identity framework, the black sheep effect
relative superiority of the ingroup (i.e., the ingroup’s
corresponds to a more sophisticated form of
subjective validity). Outgroup members, whether
ingroup favoritism. By derogating unlikable ingroup
socially desirable or socially undesirable, are much
members, people are protecting, and thus promot-
less relevant to the definition of the ingroup’s iden-
ing, the positive image of the ingroup as a whole.
tity, and thus they invite less extreme reactions.
The black sheep effect, therefore, ensues from an
Subjective Group Dynamics
internalized social influence process in which peo-
José Marques, Dario Paez, and Dominic Abrams ple subjectively reinforce their confidence in the
have recently developed the subjective group ingroup by upgrading normative members and
dynamics model to explain the cognitive and moti- derogating deviant members in an attempt to rees-
vational antecedents of the black sheep effect. This tablish the group’s positive social identity.
model proposes that people’s reactions toward
socially undesirable (or deviant) ingroup members
Empirical Evidence
involve two interrelated processes.
First, and as proposed by social identity theory, Research has supported the above analysis. For
people attempt to establish a clear-cut difference instance, in one study it was found that Belgian
Boundary Spanning 57

students, presented with likable or unlikable ingroup members who uphold such norms are par-
Belgian or North African target students, judged ticularly favorably evaluated, while those who
the likable Belgian targets more positively than the deviate from the norms are particularly unfavor-
likable North African targets and the unlikable ably evaluated.
Belgian targets more negatively than the unlikable
North African targets. Also, law students evalu- José M. Marques
ated a good performance by another law student See also Conformity; Deviance; Group Cohesiveness;
more positively than a similar performance by a Group Socialization; Norms; Opinion Deviance; Self-
philosophy student and evaluated a poor perfor- Categorization Theory; Social Identity Theory;
mance by a law student more negatively than a Subjective Group Dynamics
similar performance by a philosophy student.
In a similar vein, high school students evaluated
likable students from their own school more posi- Further Readings
tively than similar students from a rival school and
Marques, J. M., Abrams, D., Paez, D., & Hogg, M. A.
evaluated unlikable students from their own school (2001). Social categorization, social identification, and
more negatively than similar students from a rival rejection of deviant group members. In M. A. Hogg &
school. In addition, the black sheep effect generally R. S. Tindale (Eds.), Blackwell handbook of social
co-occurred with more favorable evaluations of psychology: Group processes (Vol. 3, pp. 400–424).
the ingroup than the outgroup as a whole. Oxford, UK: Blackwell.
Research has also shown that the black sheep Marques, J. M., Abrams, D., Paez, D., & Martinez-
effect occurs only when what defines group mem- Taboada, C. (1998). The role of categorization and
bers’ social desirability is deemed relevant for gen- ingroup norms in judgments of groups and their
erating positive differentiation between the ingroup members. Journal of Personality and Social
and outgroup. For example, if the ingroup is a soc- Psychology, 75(5), 976–988.
cer team, being a good athlete will matter more to Marques, J. M., Abrams, D., & Serôdio, R. G. (2001).
its members than being a good musician. Other Being better by being right: Subjective group dynamics
research has shown that participants who identify and derogation of in-group deviants when generic
strongly with their group are more favorable norms are undermined. Journal of Personality and
toward likable and less favorable toward dislik- Social Psychology, 81(3), 436–447.
able ingroup members than toward similar kinds Marques, J. M., & Paez, D. (1994). The “black sheep
of outgroup members. This is not the case with effect”: Social categorization, rejection of ingroup
participants who either identify weakly with their deviates, and perception of group variability. In
group or are allowed to subjectively leave that W. Stroebe & M. Hewstone (Eds.), European review
group by disidentifying with it. of social psychology (Vol. 5, pp. 38–68). Chichester,
Other research directly supports tenets of the UK: John Wiley.
Marques, J. M., Paez, D., & Abrams, D. (1998). Social
subjective group dynamics model. For example,
identity and intragroup differentiation as subjective
there is evidence that people upgrade likable others
social control. In S. Worchel, J. F. Morales, D. Paez, &
and derogate unlikable others when these others
J.-C. Deschamps (Eds.), Social identity: International
are perceived as ingroup members. The effect does
perspectives (pp. 124–142). London: Sage.
not occur when the targets are described as simply
interpersonally similar to, or different from, the
participants or are described as outgroup members.
This supports the idea that the black sheep effect
indeed emerges as a consequence of the perceived Boundary Spanning
interdependence of self and ingroup members for
the maintenance of a positive social identity. Every group or social system has a boundary that
In a similar vein, there is evidence that evalua- separates it from other groups or systems and
tions are particularly extreme in the case of norms defines who is in the group and who is outside of it.
that are central to and legitimize the group’s exis- Boundary spanning represents the actions that are
tence—in comparison to outgroup members, taken by members of a group to manage those
58 Boundary Spanning

boundaries, particularly interactions with outsid- thus, some of the most fine-grained analyses of
ers. Boundary-spanning activities may be a formal boundary-spanning activities have been conducted
part of some individuals’ jobs or informally carried with these groups.
out by members of the group. How a group man- One comprehensive study of boundary-spanning
ages its boundaries has implications for other group activities in product development teams identified
processes and the performance of the group. four categories of activities. One set of activities,
Early research on groups within organizations termed ambassador activities, was directed pri-
identified two broad functions of boundary marily toward upper levels of the organization and
spanning—information processing and external focused on protecting the team from outside pres-
representation. Information processing involves sure, persuading others to support the team, and
gathering and assessing data from outside the lobbying for resources. The second set of activities,
group and providing summaries and conclusions labeled task coordinator activities, included such
to members of the group. External representation things as resolving product specifications, obtain-
has to do with providing information to outsiders ing feedback on the new design, and negotiating
and trying to shape their perceptions of the group. solutions for coordination problems. The third set,
More recent research has divided these broad cat- described as scout activities, related to general
egories into specific functions, such as mapping scanning of the external environment as opposed
(building a model of the external environment), to interactions aimed at addressing a particular
filtering (keeping troubling information from the issue. In product development groups, examples of
group), and negotiating (developing goals and scout activities included collecting general techni-
schedules). Scholars have argued that the optimal cal information and investigating broad market
amount of boundary spanning is related to the trends inside and outside the organization. The
uncertainty of environment in which the group final set of activities, termed guard activities, rep-
operates and the nature of the technology it uses. resented internal activities designed to prevent
One of the important effects of boundary span- information from leaving the group.
ning concerns the permeability of the boundary of Product development teams that concentrated
the group. Permeability refers to the ease with on ambassador and task coordinator activities and
which information and resources can pass into and displayed relatively low levels of scout activities
out of the group. The permeability of the group’s were most successful in completing their tasks.
boundary can affect the group’s processes and Groups with other profiles of activities were less
potentially its performance. On the one hand, if a successful. These results illustrate two important
group has too few or overly restrictive interactions points about boundary spanning. First, it is not
with outside groups, the group can become “over- simply the amount of external interaction that
bounded.” If this is the case, the group may contributes to successfully managing interdepen-
become isolated and detached from the environ- dencies. Second, some types of external interac-
ment. This can lead to isolation, inaccurate percep- tions may contribute to success, while others may
tions of other groups, and ineffective internal not. In particular, very general and unfocused
group processes, such as “groupthink.” On the interactions that continue throughout the group’s
other hand, if a group becomes “underbounded” life may make it difficult for the group to concen-
by having such a porous boundary that interac- trate on the specific external interactions necessary
tions with outsiders become predominant, the to complete work that is highly interdependent.
group may have difficulty developing cohesive- Some jobs within organizations have formal
ness, establishing shared commitments, and agree- boundary-spanning responsibilities. The boundar-
ing on a common course of action. ies may be between different groups within an
Not surprisingly, much of the research on organization or between the organization and out-
boundary spanning has been conducted with side entities. These types of jobs frequently involve
groups in organizations. For many such groups, simultaneous negotiations with both entities. A
such as engineering groups or product develop- number of factors can influence the approaches
ment teams, interactions with other groups are boundary spanners take in dealing with their
critical for successfully performing their tasks; responsibilities, including the boundary spanner’s
Brainstorming 59

relations with his or her own group or organization follows a set of guidelines or procedures. Research
(for example, how much latitude the boundary on brainstorming has focused on comparing the
spanner is given), the relationship between the effectiveness of group brainstorming and individ-
boundary spanner and the outsider (for example, ual brainstorming. This has led to the develop-
the history of the interactions between the entities), ment of theoretical models of the social and
and the personal characteristics of the boundary cognitive processes involved in group idea genera-
spanner (for example, his or her sensitivity to social tion. These models have relevance for an under-
cues and ability to adapt to different situations). standing of creative processes in a wide range of
Research has shown that individuals who hold groups and teams. Research has also provided
boundary-spanning positions often experience much useful information for improving the prac-
stress and dissatisfaction because of the conflicting tice of brainstorming in organizations.
and ambiguous demands they face. However, such Brainstorming provides an interesting example
positions offer benefits to individuals in terms of of the various factors that may influence group
career advancement and the opportunity to build task performance. One of the major issues in
relationships within the organization. studying groups has been comparing the function-
ing of groups to appropriate individual perfor-
David F. Caldwell mance baselines to determine whether groups are
beneficial or harmful for performance. This has
See also Group Boundaries; Group Cohesiveness; Group
Performance; Groupthink; Work Teams been one of the focal issues in group brainstorming
research. The research has also provided important
insights for optimizing work team performance
Further Readings that involves the sharing of information or ideas.
Brainstorming research has three major goals.
Adams, J. S. (1976). The structure and dynamics of
One is to discover the causes of production losses
behavior in organization-boundary roles. In
in groups. Another is to develop theoretical models
M. D. Dunnette (Ed.), Handbook of industrial and
of the group creative process. The third is to gener-
organizational psychology (pp. 1175–1199). Chicago:
ate ways to enhance brainstorming in groups and
Rand McNally.
overcome productivity losses. After discussing the
Aldrich, H., & Herker, D. (1977). Boundary spanning
roles and organizational structure. Academy of
foundations of brainstorming, this entry describes
Management Review, 2, 217–235. all three areas of research.
Ancona, D., & Bresman, H. (2007). X-teams: How to
build teams that lead, innovate, and succeed. Boston: History and Background
Harvard Business School Press.
Ancona, D., & Caldwell, D. (1992). Bridging the Brainstorming was initially popularized by Alex
boundary: External activity and performance in Osborn, an advertising executive who published
organizational teams. Administrative Science several books in the 1940s and ’50s on the creative
Quarterly, 37, 634–655. process. In these books, Osborn highlighted the
Tushman, M. (1977). Special boundary roles in the use of brainstorming in coming up with novel
innovation process. Administrative Science Quarterly, ideas. He argued that it is important for creativity
22, 587–605. that one refrain from being critical during the idea
generation process. To be creative, one has to let
one’s ideas flow in an unrestrained manner with-
out concern for quality. Expressing a lot of ideas
Brainstorming no matter how wild they are increases the likeli-
hood that good ideas will be generated. Any type
Brainstorming is a technique for idea generation in of negative or evaluative feedback might kill the
which the focus is on generating as many ideas as motivation for unbridled creativity.
possible on a topic in a noncritical fashion. The Osborn also assumed that group brainstorming
term is often used for any interaction that involves is effective because it allows individuals to build on
sharing of ideas, but formal brainstorming typically the ideas of others. Osborn’s various books helped
60 Brainstorming

popularize brainstorming, and it is still the founda- Another factor that may be important in inhib-
tion for much of what passes as group creativity iting performance in groups is concern about oth-
exercises or procedures. For example, IDEO cor- ers’ reactions to one’s ideas, or evaluation
poration, a top product development company, apprehension. Although brainstorming rules spec-
bases its product development sessions on the prin- ify that group members should not criticize each
ciples outlined by Osborn. other’s ideas and should withhold judgment about
Osborn’s ideas were based on his own intuitions their quality, individuals in groups may still be
and experiences of working with groups and indi- concerned about the impression they make on oth-
viduals in the advertising field. He was also well ers. People generally have a desire to make a posi-
versed in the general literature on the group creative tive impression and hence often censor what they
process, and many of his ideas still have merit. say to maintain such an impression. This kind of
However, controlled studies have challenged one of behavior is another reason why group brainstorm-
his claims. He suggested that group brainstorming ing may be less effective than individual brain-
would be twice as effective as individual brainstorm- storming.
ing. At first glance, this prediction seems sensible,
because several individuals are likely to be able to
Social Comparison and Brainstorming
generate more ideas than a single individual can.
However, a fairer comparison involves a group of One inevitable feature of group brainstorming
people who generate ideas together versus the same is that it allows individuals to compare their rate
number of people who generate ideas by themselves of performance and overall quality of ideas with
(what is known as a nominal group). Research has those of others. Several studies have highlighted
consistently revealed that interacting groups gener- some interesting consequences of this social com-
ate fewer ideas than do nominal groups. For exam- parison process. One result of such a process is
ple, a group of four brainstormers might only that the performance of group members may con-
generate half as many unique ideas as the total num- verge in both rate and quality. That is, group
ber of unique ideas generated by four individual members may become more similar over time in
brainstormers. The poorer performance of interact- their pace of idea generation and the types of ideas
ing groups in comparison to nominal groups is they generate. Such convergence is, of course, con-
termed a production loss, because interaction is trary to the goal of encouraging a wide range of
associated with a loss of productivity relative to the ideas and divergent thinking. The social compari-
potential of the group members working alone. son process can also be involved in production
losses. In laboratory groups there appears to be a
tendency for social comparison to lead individuals
Explanations for Production Losses
to reduce their efforts and thus lower the overall
A variety of factors have been implicated as performance of the group. Another interesting
responsible for productivity losses in brainstorm- result of social comparison is that group members
ing groups. Research by Michael Diehl, Wolfgang may perceive their performance as more positive
Stroebe, and Bernard Nijstad has supported the than it really is. Whereas people working alone do
role of production blocking. These researchers not have a reference point for evaluating their per-
propose that the major problem in group brain- formance and thus may feel uncertain about the
storming is that only one person can speak at a quality of their performance, people working in
time, blocking the others from sharing their ideas. groups may discover that they are doing at least as
While one group member is speaking, the other well as others. This in turn may cause them to have
group members may forget their ideas, or they may an inflated view of their performance.
become less motivated to share these ideas because Research by Bernard Nijstad has suggested that
of the blocking experience. Studies have examined another reason for this inflated perception of per-
different aspects of this blocking effect and have formance is the general flow of ideas in groups. In
concluded that not being able to present ideas as individual brainstorming there is likely to be peri-
they occur is a critical factor to the relatively poor odic difficulty in coming up with ideas. In contrast,
performance of interacting groups. in groups there are fewer times when no ideas are
Brainstorming 61

being generated since there are multiple potential additional ideas. This assumption underlies several
contributors to the process. Several researchers cognitive models of group brainstorming. Paul
have sought to demonstrate that group idea Paulus and Vincent Brown have proposed a
exchange can stimulate a higher level of creativity semantic network model to explain the potential
than just individual brainstorming. These research- benefits of brainstorming. This model assumes
ers generally assume either that social-motivation that concepts are stored in long-term memory and
factors or cognitive stimulation can overcome the are related in a network so that some concepts are
inhibiting effects of group interaction. more strongly related to one another than to other
concepts. Ideas that are shared and have some
degree of relatedness to a concept can stimulate
Models of Brainstorming that concept and thus make it available for addi-
Social-Motivational Strategies tional ideas.
Some concepts may be more accessible than
Whenever group members collaborate on gener- others and may thus be more easily retrieved dur-
ating ideas, various social pressures may motivate ing the idea generation process. As a result, it is
them in their search for more creative ideas. likely that the initial phases of idea generation will
Although, as noted above, social comparison can focus on these. However, once these concepts have
reduce performance, the tendency of individuals to been tapped to some extent, individuals may
compare their performance with that of others can increase efforts to search their knowledge base,
also motivate them to work harder. Several studies leading to exploration of somewhat less accessible
have found that when individuals feel individually concepts or categories of ideas. This cognitive
accountable or are aware of how well they are model assumes that both attention to shared ideas
doing in comparison to other group members, and memory of these ideas are critical for contin-
there is increased motivation and increased genera- ued generation of creative ideas in groups.
tion of ideas. Group members can also be moti- Several studies have demonstrated the impor-
vated by intergroup competition when they find tance of both of these factors in enhancing the
out that the performance of their group is lower stimulating potential of shared ideas. For example,
than that of some referent group. when ideas are presented that are relatively com-
An important factor in the success of groups is mon or typical for a topic, they are remembered
the extent to which they set high goals. When better and stimulate the generation of additional
someone in authority sets high goals, these goals ideas more than do relatively unique or atypical
significantly increase the brainstorming perfor- ideas. This model also suggests that some of the
mance of both groups and individuals. One prac- real benefits of group brainstorming may only be
tice that may also be helpful to groups is the use of evident after individuals leave the group and have
trained facilitators or explicit rules that can serve as some time to reflect on the shared ideas. The
a substitute for facilitators (such as encouraging full memory of these ideas may stimulate additional
participation and discouraging talking about non- ideas during such a time of solitary reflection. This
task-relevant issues). Although it has been found type of effect is sometimes called incubation, since
that group creativity can be enhanced by various it is presumed that ideas stimulated in group inter-
motivational factors, similar benefits accrue for action may not be fully realized until the individual
individual brainstormers. No studies have found has had time to let this stimulating effect continue
that the level of motivation in groups allows them throughout his or her semantic network.
to exceed the performance of nominal groups. Bernard Nijstad and Wolfgang Stroebe have also
developed a cognitive model of brainstorming that
focuses on search for ideas in associative memory
Cognitive Models
(SIAM). These researchers emphasize the impor-
The prediction that group brainstorming should tant role of both long-term memory and working
be more effective than individual brainstorming memory—a temporary storage system where one
derives from the notion that ideas shared in the deals with the stored material. This involves a cen-
group should stimulate group members to think of tral executive system that actively searches for cues
62 Brainstorming

(the idea retrieval phase) that can then be the basis programmed to function in a variety of ways and
for a flow of ideas (the idea production loop). can include voting and decision-making features.
When blocking occurs in groups, this may interfere This is sometimes labeled a group decision support
with both stages of this cognitive process. system. For example, one system provides partici-
Although it has some conceptual overlap with the pants with a window at the bottom of the computer
Paulus and Brown model, the SIAM model makes screen in which to generate an idea and then send
some unique predictions. The SIAM model it to the pool of ideas. Another window at the top
assumes that transitions from one idea to another displays ideas generated by the group. These ideas
within a category will be quicker than transitions are arbitrarily divided into folders, and participants
from one category to another. Therefore, external are exposed to the ideas in one of these folders each
stimulation should be helpful for facilitating time they generate an idea. Several studies have
the change to a new category but not for within- found that this kind of exchange of ideas in groups
category ideation. This has been found. Also, when avoids the productivity loss problem. This is not
individuals experience failures in the search process difficult to explain, since there is little production
(by having difficulty coming up with new ideas), blocking using this procedure—participants can
they may become dissatisfied with the process and type ideas as they occur. In fact, as groups increase
stop. This is less likely in groups in which individu- in size, there tends to be an enhanced productivity
als have fewer failures in coming up with new ideas compared to similar-size nominal groups. Most
and therefore enjoy the process and stay with it. studies indicate that this enhancement occurs when
group size reaches eight or more.
Although the reduction of production blocking
Ways to Enhance Brainstorming
appears to be a straightforward explanation for
Based on the two cognitive models, it is not sur- part of the benefit of brainwriting and electronic
prising that approaches to brainstorming that limit brainstorming, it cannot explain the production
cognitive interference and maximize attention to gains. However, these gains can be explained by
ideas from others are most beneficial. That would social motivational and cognitive mechanisms.
explain the positive results obtained with brain- The brainwriting procedure enhances the degree of
writing, which involves the exchange of ideas on accountability and may produce a sense of compe-
pieces of paper. For example, in one study partici- tition in groups, which in turn may motivate group
pants in groups of four sitting around a table members to work harder and persist longer than
wrote ideas on pieces of paper. Participants passed solitary brainwriters. Electronic brainstormers are
each piece of paper to the person on their right for typically anonymous, so the accountability factor
reading and the addition of another idea. When cannot explain their productivity. There may,
each piece of paper contained four ideas, it was however, be an increased sense of competition as
placed in the center of the table. Paul Paulus and group size increases with the larger number of
Huei-Chuan Yang found that this procedure ideas being shared in the folders.
increased the number of ideas generated by groups In most of these studies, the participants were in
by 40% compared to nominal groups of partici- the same room, so social facilitation may have
pants who wrote ideas alone. Even more striking, increased the motivation level in the larger groups.
when group participants continued writing ideas It is also possible that the shared ideas in brain-
in a subsequent solitary session, the group brain- writing and electronic brainstorming provide cog-
storming experience led to 89% more ideas than nitive stimulation. With increased group size, there
the nominal experience. So instead of showing a is increased exposure to novel ideas. A factor that
production loss due to group brainstorming, this may be important is the extent to which the proce-
study demonstrated a significant production gain. dure insures attention to the shared ideas. The
A more popular approach to overcoming the brainwriting procedure encourages attention
production loss of group brainstorming has been because it requires participants to monitor the
electronic brainstorming, which involves generat- ideas. In electronic brainstorming this is not neces-
ing ideas by exchanging them on a computer net- sary. Participants can ignore the shared ideas and
work. The advantage of this system is that it can be focus only on generating their own.
Bystander Effect 63

When participants are encouraged to attend to

others’ ideas because they expect to be tested on Bystander Effect
their memory of these ideas, electronic brainstorm-
ing with only four participants leads to the genera- The bystander effect refers to the inhibiting influ-
tion of more ideas than nominal groups generated. ence of the presence of others on a person’s willing-
This benefit is maintained in a subsequent solitary ness to help someone in need. Researchers have
ideation session, just as in the case of brainwriting. found that, even in an emergency, a bystander is
Thus far, research has not clearly determined the less likely to extend help when he or she is in the
relative importance of the motivational and cogni- real or imagined presence of others than when he
tive factors in brainwriting and electronic brain- or she is alone. Moreover, the number of others is
storming. It seems likely that both factors important, such that more bystanders lead to less
contribute to their benefits. assistance, although the impact of each additional
bystander has a diminishing impact on helping.
Paul B. Paulus The bystander effect is well illustrated by the events
See also Group Performance; Social Comparison Theory; surrounding the brutal murder of Kitty Genovese
Social Facilitation; Social Loafing in 1964, which had a major impact on bystander
intervention research in particular and helping
research in general. After summarizing the Kitty
Further Readings Genovese story, this entry reviews the sequence of
Derosa, D. M., Smith, C. L., & Hantula, D. A. (2007). decisions a bystander may engage in when encoun-
The medium matters: Mining the long-promised merit tering a person in need of help, as well as the pro-
of group interaction in creative idea generation tasks cesses of social influence and diffusion of
in a meta-analysis of the electronic group responsibility that may affect these decisions.
brainstorming literature. Computers in Human
Behavior, 23, 1549–1581.
Nijstad, B. A., & Stroebe, W. (2006). How the group
Kitty Genovese and Bystander Intervention
affects the mind: A cognitive model of idea generation In the early morning on March 13, 1964, a woman
in groups. Personality and Social Psychology Review, named Kitty Genovese was returning home late
10, 186–213. from work when a man with a knife viciously
Osborn, A. F. (1957). Applied imagination. New York: attacked and sexually assaulted her in the parking
Scribner. lot of her apartment complex. As reported in the
Parnes, S. J., & Meadow, A. (1959). Effects of New York Times, for over half an hour 38 respect-
brainstorming instructions on creative problem solving able, law-abiding people heard or saw this man
by trained and untrained subjects. Journal of attack her three separate times. The voices and
Educational Psychology, 50, 171–176.
lights from the bystanders in nearby apartments
Paulus, P. B. (2007). Fostering creativity in groups and
interrupted the killer and frightened him off twice,
teams. In J. Zhou & C. E. Shalley (Eds.), The
but each time he returned and stabbed her again.
handbook of organizational creativity (pp. 165–188).
None of the 38 witnesses called the police during
Boca Raton, FL: Taylor & Francis Group.
Paulus, P. B., & Brown, V. (2003). Ideational creativity
the attack, and only one bystander contacted
in groups: Lessons from research on brainstorming. In
authorities after Kitty Genovese died.
P. B. Paulus & B. Nijstad (Eds.), Group creativity: While some details of this story as initially
Innovation through collaboration (pp. 110–136). New reported have since been contradicted by evidence,
York: Oxford University Press. the story has become a modern parable for the
Paulus, P. B., & Brown, V. R. (2007). Toward more creative powerful psychological effects of the presence of
and innovative group idea generation: A cognitive-social others. The story serves as an extreme example of
motivational perspective of brainstorming. Social and how people sometimes fail to react to the needs
Personality Compass, 1(1), 248–265. of others and, more broadly, how behavioral ten-
Sutton, R. I., & Hargadon, A. (1996). Brainstorming dencies to act prosocially are greatly influenced by
groups in context: Effectiveness in a product design the situation. Moreover, this tragic event sparked
firm. Administrative Science Quarterly, 41, 685–718. a good deal of research on prosocial behavior,
64 Bystander Effect

which examined when people do and do not emergencies and provide assistance, whereas nega-
extend help (referred to as bystander intervention). tive moods, such as depression, inhibit helping.
The seminal research on bystander intervention However, some negative moods, such as sadness
was conducted by two social psychologists, Bibb and guilt, have been found to promote helping. In
Latané and John Darley. These researchers, and addition, some events, such as someone falling
subsequently others, found that bystanders do care down a flight of stairs, are very visible and hence
about those in need of assistance, but nevertheless attract bystanders’ attention. For example, studies
often do not help them. Whether they extend help have demonstrated that victims who yell or scream
depends on how they make a series of decisions. receive help almost without fail. In contrast, other
events, such as a person suffering a heart attack,
often are not highly visible and so attract little
Bystander Decision Making
attention from bystanders. In these latter situa-
The circumstances surrounding an emergency in tions, the presence of others can have a substantial
which an individual needs help tend to be unique, impact on bystanders’ tendency to notice the situa-
unusual, and multifaceted. Most people have never tion and define it as one that requires assistance.
encountered such a situation and have little experi- In situations where the need for help is unclear,
ence to guide them during the pressure-filled bystanders often look to others for clues as to how
moments when they must decide whether or not to they should behave. Consistent with social com-
help. Latané and Darley’s decision model of parison theory, the effect of others is more pro-
bystander intervention elaborates the sequence of nounced when the situation is more ambiguous.
decisions leading to a bystander’s response. For example, when other people act calmly in the
According to Latane and Darley, before helping presence of a potential emergency because they are
another, a bystander progresses through a five-step unsure of what the event means, bystanders may
decision-making process. A bystander must notice not interpret the situation as an emergency and
that something is amiss, define the situation as an thus act as if nothing is wrong. Their behavior can
emergency or a circumstance requiring assistance, cause yet other bystanders to conclude that no
decide whether he or she is personally responsible action is needed, a phenomenon known as plural-
to act, choose how to help, and finally, implement istic ignorance. But when others seem shocked or
the chosen helping behavior. Failing to notice, distressed, bystanders are more likely to realize an
define, decide, choose, and implement leads a emergency has occurred and conclude that assis-
bystander not to engage in helping behavior. tance is needed. Other social comparison variables,
Examinations of the bystander effect have focused such as the similarity of other bystanders (e.g.,
mainly on the role social influence plays in indi- whether they are members of a common ingroup),
viduals noticing something is wrong and defining can moderate the extent to which bystanders look
the situation as an emergency and on how the pres- to others as guides in helping situations. In sum,
ence of others can cause diffusion of the responsi- when the need for help is unclear, bystanders look
bility to help. Research indicates that social influence to others for guidance. This is not the case when
and diffusion of responsibility are the fundamental the need for assistance is obvious.
processes underlying the bystander effect during
the early steps of the decision-making process.
Diffusion of Responsibility
If a person does notice a situation and defines it as
Informational Social Influence
requiring assistance, he or she must then decide if
If a bystander is physically in a position to notice a the responsibility to help falls on his or her shoul-
victim, factors such as the bystander’s emotional ders. Thus, in the third step of the bystander
state, the nature of the emergency, and the presence decision-making process, diffusion of responsibil-
of others can influence his or her ability to realize ity rather than social influence is the process
something is wrong and that assistance in required. underlying the bystander effect. Diffusion of
In general, positive moods, such as happiness responsibility refers to the fact that as the number
and contentment, encourage bystanders to notice of bystanders increases, the personal responsibility
Bystander Effect 65

that an individual bystander feels decreases—and influence their decision. Researchers have demon-
as a consequence so does his or her tendency to strated the effect of situational expectations on
help. Thus a bystander who is the only witness to helping behavior by presenting people with an
an emergency will tend to conclude that he or she emergency in an area they have been told not to
must bear the responsibility to help, and in such enter. Bystanders previously warned not to enter an
cases people typically do help. But bystanders dif- area where an emergency was occurring were far
fuse responsibility to help when others are present. less likely to help than were those told they could
In the case of Kitty Genovese, it seems that all the enter the area. Thus, when an emergency occurs,
bystanders made the assumption that others were the social context can be a powerful determinant of
present and would intervene, and so they felt little bystanders’ decision to intervene.
or no personal responsibility to help. Diffusion of
the responsibility to help is reduced, however,
when a bystander believes that others are not in a
position to help. For example, in one study, par- The bystander effect refers to the socially inhibiting
ticipants who believed that the only other witness presence of others on helping. When it is unclear
to an emergency was in another building and whether there is a need for help, the presence of oth-
could not intervene were much more likely to help ers tends to influence the first two steps of the
a victim than were participants who believed that bystander decision-making process (i.e., noticing
another witness was equally close to the victim. something is wrong and defining the circumstance
Diffusion of the responsibility to help is increased as requiring assistance). Whether or not a bystander
when others who are viewed as more capable of will feel personally responsible to help is influenced
helping (e.g., a doctor or police officer) are present. by the number of others actually present (or assumed
Research suggests that in emergency situations to be present) and their ability to help. Latané and
where a vicitm will suffer greatly if help is not forth- Darley’s initial investigations of the bystander effect
coming, bystanders relieve themselves of responsi- sparked a wealth of research on helping behavior,
bility by asking “experts,” such as firefighter or which has expanded beyond emergency situations
paramedics, for assistance, thus indirectly helping. to include everyday forms of helping. By illuminat-
But when the costs of helping and not helping are ing the power of situations to affect individuals’
both high, bystanders feel a strong conflict between perceptions, decisions, and behavior, study of the
the desire to act and the fear of helping. For exam- bystander effect continues to influence the course of
ple, in the case of Kitty Genovese, bystanders may social psychological theory and research.
have felt the need to help because the cost of not
helping would be her death, but the possibility of Robert D. Blagg
being hurt or killed themselves deterred them from See also Conformity; Crowds; Informational Influence;
acting. Bystanders often resolve this conflict by con- Normative Influence; Norms; Pluralistic Ignorance;
cluding that someone else will help (i.e., diffusing Social Comparison Theory; Social Facilitation
responsibility), thereby psychologically reducing
the perceived cost of not helping the victim.
Further Readings

Normative Social Influence Dovidio, J. F., Piliavin, J. A., Schroeder, D. A., & Penner,
L. A. (2006). The social psychology of prosocial
A bystander’s decision regarding his or her per- behavior. Mahwah, NJ: Lawrence Erlbaum.
sonal responsibility to help may be affected by situ- Latané, B., & Darley, J. M. (1970). The unresponsive
ational norms and expectations for behavior. For bystander: Why doesn’t he help? New York: Meredith.
example, in a library patrons are expected to be Latané, B., & Nida, S. (1981). Ten years of research on
quiet and in a classroom students may speak up in group size and helping. Psychological Bulletin, 89(2),
a respectful and orderly way, but at a party people 308–324.
may be much less inhibited. When bystanders in an Penner, L., Dovidio, J., Piliavin, J., & Schroeder, D.
emergency situation assess their personal responsi- (2005). Prosocial behavior: Multilevel perspectives.
bility to act, social expectations for behavior may Annual Review of Psychology, 56, 365–392.
that is central to the way people understand things
Categorization in the world. Indeed, some writers use the term
category in almost the same way they use concept.
Categorization is the process of understanding things Categorization takes on an added importance in
by knowing what other things they are equivalent to social psychology, since people categorize them-
and different from. It is a process that is widely stud- selves and other people. In fact, discussions of
ied in cognitive and social psychology and in phi- categorization are central to discussions of group
losophy and linguistics. This entry offers an overview processes and intergroup relations because groups
of categorization and outlines its functions, then are based on categories of people.
describes the major views on category structure and The major contemporary theories of group pro-
representation and on category learning, formation, cesses and intergroup relations—social identity
and use (drawn mainly from cognitive psychology). and self-categorization theories, along with more
It also examines social psychological perspectives on specific models such as the common ingroup iden-
biased stimulus processing and sense making, and tity model, optimal distinctiveness theory, and the
concludes by considering a range of ongoing debates ingroup projection model—therefore all address
and controversies in the field. the process of categorization quite directly.
Fundamentally, categorization is about attain-
ing knowledge or understanding. The central con-
Describing the Process
cerns of social psychology relating to social norms,
The categorization process could also be called influence, attitude change, and socially shared cog-
classification or grouping, but it is helpful to pre- nition suggest that knowledge and understanding
serve the term classification for the formal, behav- develop consensually. That is, different people tend
ioral process of separating things and categorization to develop related understanding of the same
for the psychological process and experience. Thus things. In large part, this rests on their developing
a biologist who separates life forms into different compatible ways of categorizing the self and other
species is engaged in a formal scientific process of people and things.
classification (taxonomy), but the psychological People can also categorize the same things in
process that the biologist uses to perform this task radically different ways. Other people could see
is categorization. Importantly, it is also the same your furniture as their firewood or your pet dog as
process people use to decide that the person doing their lunch. This potential for disagreement, how-
the classification is a biologist and not a psycholo- ever, actually points to the success of members of
gist, physicist, or astrologer. human societies in coordinating their categoriza-
As the previous example illustrates, categoriza- tions in harmonious ways. The triumphs and
tion is widely regarded as the cognitive process achievements of human civilization (and not just

68 Categorization

social ills and conflict) rest upon the widespread new inferences and predictions about events and
consensus that emerges in categorization, despite objects that he or she is likely to encounter. In this
the infinitely variable number of ways that any way, categorization allows perceivers to go beyond
object (let alone sets of objects) can be categorized. the information given in order to develop an
Categorization can be understood as a blend of understanding, rather than an exact reproduc-
three elements, including background knowledge tion, of the world.
(i.e., expectancies and explicit or implicit beliefs), More than 50 years after their publication, these
perceived equivalence (i.e., how objects are seen to principles remain fresh and vitally applicable, even
be the same as some other objects but different though categorization is intensely researched and
from others), and category use (i.e., explicit label- the research undergoes rapid change. In the field of
ing). These three things can be understood as ele- intergroup relations, part of this vitality stems
ments that constrain each other. Thus it is possible from the incorporation of the ideas of veridical
that explicitly using a category label like “Black” perception into self-categorization theory and into
or “men” may affect the degree to which we see social cognitive work on construct accessibility.
people as equivalent to each other. This can in turn
affect our store of knowledge about relations, so
Allport’s Analysis
that after using or developing an explicit category
label, we may not only come to see those people as The other key contribution that has remained
equivalent but also come to know that Blacks or fresh and vital for a similar period of time has been
men are similar to each other in certain ways. This Gordon Allport’s analysis of categorization as a
knowledge may constrain our future perceptions simplification process involving the reduction of
of equivalence and use of labels. information. Allport argued that categorization is
a response to the overwhelming burden of infor-
mation that confronts perceivers (especially when
The Functions of Categorization
they perceive other people). Following his think-
There are two major views on the functions of ing, we have so many daily encounters that we are
categorization. The first is the view of categoriza- forced to group or type others in order to cope. We
tion as a sense-making process that involves over- therefore tend to group stimuli into the largest
laying meaning on elements. This approach is most categories that we can get away with. That is,
famously associated with Jerome Bruner, who according to Allport, we tolerate abstract, impre-
developed seven principles of veridical perception. cise categories as a response to a complex world.
In the process of simplification, many of these
abstract, imprecise categories, such as group mem-
Bruner’s View
berships, can end up being highly irrational and
Bruner argued that perception is a decision distorted. They may contain a “kernel of truth,”
process and that decisions involve discriminating but they have a high probability of being wrong—
between stimuli and between categories of stim- and such irrational, errant categories can form just
uli. This decision process follows a chain of infer- as easily as more rational, truthful categories.
ence from detecting cues to determining the According to this view, and later adaptations of it,
categorical identity of the stimulus (which could categorization can be seen as an errorful process
lead to the search for further cues). These catego- that is used in preference to more accurate indi-
ries comprise sets of specifications about which viduated or piecemeal perception only due to the
stimuli can be grouped as equivalent. A category limitations of our processing capacity.
is most likely to be used when it is accessible to Allport’s view was subsequently championed in
perception (because it matches the perceiver’s the social cognitive approach to intergroup rela-
past experience or current goals) and where there tions that was dominant in the latter part of the
is high degree of fit between the category and the 20th century. It starkly contrasts with Bruner’s view,
stimuli viewed. Veridical perception involves the which has been championed by self-categorization
perceiver’s taking stimulus input and forming cat- theorists and others. One way of highlighting the
egories, and then using those categories to form contrast between these views is to consider an
Categorization 69

everyday interaction such as waiting for and then Prototype View

getting on a bus.
The prototype view of the nature of categories
In the Brunerian view, when we are confronted
was popularized in the work of Eleanor Rosch,
with stimulus cues such as people sitting on a
who argued that categories are based on an
bench and someone in a uniform driving a bus, we
abstract summary or prototype. Thus, less typical
are likely to search for cues that suggest the exis-
examples of a category are more different from the
tence of known categories, or to form new catego-
prototype than are more typical examples. Rosch
ries to make inferences and predictions. One such
also argued that categorical systems are hierarchi-
categorization might involve distinguishing between
cally organized into a system of levels, where more
bus passengers and the bus driver to make accurate
abstract or general categories include more specific
inferences about the roles and actions of members
categories. That is, there is a basic level of catego-
of each group. The Allportian view would suggest
rization for a category system.
that it is the presence of a large number of people
The basic level for a category system is the level
that leads us to form categories that allowed us to
at which the instance is spontaneously named and
treat the members of those categories in identical
recognized most rapidly (e.g., when we are shown
ways, but at the cost of inevitable error.
a picture of a dining chair and then asked to iden-
tify what it is, we say “chair” rather than “piece of
The Structure of Categories furniture” or “dining chair” because “chair” is the
basic level for this system). The idea of category
Cognitive psychologists have identified three major
abstraction is central to self-categorization theory
long-standing approaches to category structure
and the common ingroup identity model, but the
and representation: the classical view of categories,
idea of a fixed basic level that can be applied to
the prototype approach, and the exemplar view.
social categories is more controversial.
There are also more recent mixed and variable
structure approaches, and all of these views can
also be observed in social psychology, including in Exemplar View
analyses of phenomena such as outgroup homoge- The exemplar view of category structure is that
neity and stereotype change. categories are stored as a collection of instances of
the larger category. According to this view, there is
Classical View
no abstract summary, but merely a set of stored
representations or exemplars of the category. Thus,
The classical view of categories is that they have a social category such as men or Blacks would be
all-or-none defining features. The philosopher stored in the memory of the perceiver as a set of
Ludwig Wittgenstein illustrated the implausibility traces about the members of that category who
of this view through the example of the category had been encountered.
“game.” He demonstrated that despite the fact There has been a great deal of debate in cogni-
that people have no trouble using this type of cat- tive psychology about the merits of the exemplar
egory in everyday life, it is actually very difficult to and prototype views, but there are more recent
come up with a set of all-or-none defining rules for attempts to show that categories can have both
the category “game” that includes things as diverse exemplars and prototypes (in particular for deal-
as children’s amusements and the Olympic games. ing with exceptions). There is also research that
In the 1970s, the classical view was also demon- suggests that categories can vary from having a
strated to be inconsistent with evidence that many form that is closer to an exemplar to one that is
categories are fuzzy and share relatively vague closer to a prototype representation.
family resemblances with others. Furthermore,
category members vary in the degree to which they
Category Learning, Formation, and Use
are typical of a category. Thus, a hijacker is an
atypical example of the category “criminal,” and a The major approaches to category learning, for-
bank robber is a typical (good) example, yet they mation, and use have been based on the ideas
are both criminals in an absolute sense. of similarity-based formation and theory-based
70 Categorization

formation. The purely similarity-based approaches Henri Tajfel and his colleagues, include the intra­
will not be discussed in this entry, as they seem class (assimilation) effect (i.e., increased perceptions
inadequate as accounts of the ways that categori- of similarity within categories) and the interclass
zation takes account of meaning and experience. (contrast) effect (i.e., increased perceptions of dif-
Category formation appears to be based on inter- ference between categories). In social psychological
nally coherent theories about the function and terms, categorization effects are applied by group
nature of things rather than just on the basis of members through the development of standards
surface appearances or gross similarities. and norms that are different from the ones devel-
Indeed, cognitive researchers have shown that oped by other relevant groups. Within the group,
objects could be seen as similar or different depend- members converge on those standards and norms
ing on the order in which they are presented. Thus (usually through the processes of social influence).
Mexico might be seen as like the United States, but The nature of these effects remains debatable. In
the United States might be seen as different from particular, the question is whether categorization
Mexico. Evidence such as this can lead us to ask, effects are distortions and/or biases or accurate but
as Douglas Medin did, if something is said to be contextually variable reflections of reality.
similar to something else in certain respects, then The dominant social cognitive approach has
where do those respects come from? argued that categorization is used because limited
The answer that has had the most extensive capacity leads to biased processing, which in turn
implications for the study of social groups is con- creates distortions. Categorization is seen as an
tained in psychological essentialism, the idea that application of selective attention to both stimuli
categories may have a coherent core or essence that occur spontaneously and constructs that are
that binds category members together. This idea already stored in our cognitive systems. In other
has been applied to social categories in research words, it is believed that categorization takes place
over the last two decades. Psychological essential- because of the need to be selective when detecting
ism for social categories is often equated with the stimuli, and categorization is in fact designed to
notion that some categories are natural and have a make stimulus information that is overwhelming
biological (e.g., genetic) core, while other catego- for the system (or too diffuse) manageable.
ries lack this core. The belief in the psychological However, self-categorization and other sense-
essence of a group or category has therefore been making approaches argue that categorization is a
linked to analyses of race and racial prejudice. veridical, context-dependent, sense-making pro-
Psychological essentialism complements and is cess that produces accurate perceptions from a
sometimes confused with Donald Campbell’s con- current perspective (which follows from the pro-
cept of entitativity. This is the idea that social cess of salience). Henri Tajfel argued that categori-
groups may be perceived to a greater or lesser zation allows perceivers to derive meaning by
degree as entities or things (a point rejected by indi- making different classes of stimuli (social and non-
vidualist stances that claim only individual things social stimuli) coherent. In self-categorization the-
are real). Entitativity was developed from the gestalt ory, perceptions about the self are organized in
principles of perception to argue that a group is hierarchical categories that are contextually acti-
perceived to be more real and coherent to the extent vated through the process of salience (the extent to
that its members are perceptually alike, are close which a particular self-categorization becomes
together, share a common fate, and form a coherent psychologically proponent).
figure or form. Much of the research effort in this It is important to note that self-categorization
area has focused on seeking to demonstrate that theorists believe that variations in self-perceptions
certain types of groups (e.g., families or sporting reflect real changes, as experienced from the point
teams) are more entitative than others (e.g., people of view of the perceiver. The sense-making
waiting at a bus stop and sports crowds). approaches share the idea that categorization is a
Social psychologists have been more interested process that helps perceivers understand their sur-
than cognitive psychologists in categorization roundings, but in each of these treatments, there is
effects on judgment (e.g., in research on multiple a particular key construct that plays the major
categorization). These effects, first studied by role: coherence (in optimal distinctiveness theory),
Categorization 71

explanation (in the social judgability approach), increase the match between perception and real-
and assimilation and contrast (in approaches based ity. According to this model, all perception is
on categorization effects and construct activation). categorical, and categorization allows people to
represent reality from their particular point of
view. Thus self-categorization theorists argue that
Ongoing Debates
apart from the perceiver’s belief in the accuracy of
The contrast of these perspectives sets the scene for his or her perception, accuracy of perception can
some debates and unresolved issues. The first only be established with respect to the degree to
debate revolves around how information-processing which some relevant ingroup member agrees with
capacity affects categorization processes. The idea the judgment.
of capacity limitations is of central importance to The third debate revolves around whether cate-
social cognitive approaches to categorization. gories are stored and retrieved from memory (when
Here, categorization and stereotyping are viewed they are accessible) or spontaneously created anew
as necessary but imprecise shortcuts for people for each situation. The self-categorization theory
overloaded with information from their environ- version of the sense-making account tends to
ments. Accurate judgments are only considered assume that categories are constructed on the spot
possible when people have time and are willing to for each new social encounter. What is assumed to
put in the effort to attend to the unique features of be relatively stable is long-term background knowl-
the individuals they encounter. edge about the world that is used to help construct
This understanding of categorization has been an infinite number of possible categorizations to
challenged by the alternative meaning-seeking view suit each new social context. It is less clear from
espoused by self-categorization theorists. From this this perspective what form this long-term back-
perspective, instead of leading to errors, the pro- ground knowledge takes if it is not categorical.
cess of categorization is seen to provide a meaning- The alternative view put forward by the selective
ful framework for understanding and predicting attention perspective argues that categories are
the behavior of people in different social contexts. stored and retrieved from memory. This approach
The second major debate regarding these two can more easily explain why categories that have
approaches revolves around the issue of whether recently been activated, or are chronically accessi-
categorizations produce distortions or accurate ble, are more likely to be used than older, less
reflections of reality from a perspective. The selec- chronically accessible categories. To resolve this
tive attention model suggests that categorization debate, it will be necessary to determine whether
and stereotyping produce distortions of reality remembering a category is the same as reusing it.
because these processes are seen to lead people to The fourth debate relates to whether some cat-
exaggerate equivalences of members of particular egories are pervasive and constantly applicable or
groups. In this view, the accuracy of group judg- whether all categories are utterly context depen-
ments can only be determined with reference to the dent. Many social cognitive researchers argue that
characteristics of the unique individuals who make there are a small number of basic and primary
up the group. categories, such as gender, ethnicity, and age that
Alternatively, several accuracy-oriented are automatically and unconsciously activated in
approaches propose that stereotypes are based on real interaction. Alternatively, self-categorization theo-
characteristics of social groups. The reflection– rists argue that there is no basic level or set of cat-
construction model argues that beliefs about cate- egories that is primary. Indeed, self-categorization
gories reflect, modify, and influence the construction theory emphasizes that categorization is highly
of reality. According to this model, the accuracy of variable and context dependent and that it would
particular stereotypes will determine whether be a mistake to suggest that gender, ethnicity, or
categorization enhances, or reduces, the accuracy any other category will always be used to catego-
of perception. An alternative accuracy-oriented rize others in interactions. A possible resolution of
approach to stereotyping has been derived from this issue is to view basic and primary categories as
self-categorization theory, in which it is proposed operating at a background level of perception and
that appropriate categorizations always serve to intruding into judgments about members of those
72 Charismatic Leadership

categories even when those categories are not social and historical context, has received an enor-
explicitly salient. mous amount of attention over the past several
One final issue that social models of categoriza- decades. In this entry, charismatic leadership is
tion will need to resolve is whether categories are defined, along with the conditions conducive to its
based on hierarchical structures much like those appearance and typical features of its communica-
shown in tree diagrams. The assumption of hierar- tion. An illustrative example is then presented,
chical structures is adapted from cognitive research followed by some of the limitations of this form of
and appears in self-categorization theory, the com- leadership.
mon ingroup identity model, and the ingroup pro-
jection model. Some recent evidence from cognitive
Defining the Concept
psychology suggests that the idea of hierarchical
structures does not seem to capture the way people First introduced by German sociologist Max Weber,
make inferences about natural object categories. the word charisma comes from the Greek word
This raises similar questions about social catego- χα′ρισμα (kharisma), meaning “gift” or “divine
ries that often do not have clear tree-like struc- favor.” Following this original definition, charisma
tures. For example, Texas is part of America, but refers to an extraordinary quality of a person that
we know that not all Texans are Americans. allows him or her to charm and influence others. In
the common vernacular, charisma is often treated
Craig McGarty, Renata Bongiorno, as a powerful personal appeal or magnetism that
and Ana-Maria Bliuc captivates others. The charismatic leadership
See also Common Ingroup Identity Model; Cross-
approach emphasizes heroic leaders with forceful,
Categorization; Decategorization; Entitativity; dramatic personalities and widespread appeal;
Essentialism; Optimal Distinctiveness; Self- recent examples include leaders such as Jack
Categorization Theory; Social Identity Theory Welch, Bill Clinton, and Arnold Schwarzenegger.
However, charisma is very much in the eye of
the beholder, suggesting that charismatic leader-
Further Readings ship is more accurately understood as a relation-
Crisp, R. J., & Hewstone, M. (2007). Multiple social ship between leaders and followers. Charismatic
categorization. In M. P. Zanna (Ed.), Advances in leadership emphasizes the importance of symbolic
experimental social psychology (Vol. 39, pp. 163–254). behaviors, emotional appeals, and the role of the
Orlando, FL: Academic Press. leader in making events meaningful for followers.
Gaertner, S. L., & Dovidio, J. F. (2000). Reducing It focuses on understanding how a leader can influ-
intergroup bias: The common ingroup identity model. ence followers to make sacrifices, commit to diffi-
Philadelphia: Psychology Press. cult or seemingly impossible objectives, and achieve
Lakoff, G. (1987). Women, fire, and dangerous things: much more than was initially expected. Charismatic
What categories reveal about the mind. Chicago: leadership is, therefore, not solely a property of the
University of Chicago Press. leader’s charisma, and focusing on the leader alone
McGarty, C. (1999). Categorization in social psychology. ignores the unique circumstances that are crucial
London: Sage. in each instance of charismatic leadership. While a
Turner, J. C., Hogg, M. A., Oakes, P., Reicher, S. D., & leader with extraordinary gifts and qualities is a
Wetherell, M. S. (1987). Rediscovering the social critical element in this relationship, aspects of the
group: A self-categorization theory. Oxford, followers and the situation are also important in
UK: Blackwell. understanding why some leaders are viewed as
charismatic and others are not.
A significant element is that followers who are
frightened, threatened, or uncertain are more likely
Charismatic Leadership to view a leader as a charismatic savior. Followers
are more susceptible to a charismatic leader and his
Charismatic leadership, combining a leader with or her vision if they are insecure, alienated, and
powerful personal magnetism and a particular fearful about their physical safety or economic
Charismatic Leadership 73

security, or if they have low self-esteem or a weak gifts, and his flare for dramatic gestures such as fast-
self or social identity. Therefore, charismatic leader- ing, simple dress, and spiritual rituals produced a
ship is more likely to emerge during a crisis or dur- heroic persona that challenged current norms and
ing a situation of desperation or uncertainty. The conventional behavior of the time. Second, Gandhi
leader presents a vision or set of ideas promising a came to power in response to a desperate situation:
solution to the crisis and a better future (whether the crisis in India in the wake of British colonization
achievable or not), and followers are attracted to and oppression. Third, Gandhi offered a powerful
this gifted person and come to believe in his or her vision of a better future, including a self-governing
exceptional powers and vision for a better future. India where its citizens would live under their own
To retain his or her followers, however, the laws free from outside interference. Fourth, Gandhi’s
leader must also demonstrate or convince others of vision, rooted in the basic values of Indian culture,
his or her ability to deal with the crisis or threat and attracted a tremendous following. His dynamic
move followers in the direction of a better future. vision created a contagious atmosphere that inspired
Accordingly, charismatic leadership is a process his followers to seek independence. Gandhi’s fol-
that resonates in the exceptional personal attributes lowers were also motivated by his repeated smaller
of the leader, as well as the fit between those attri- successes along the way to independence, including
butes and the needs of followers, against the back- tax reforms and the recognition of non-Christian
drop of a crisis or undesirable state of affairs. marriages. Finally, Gandhi led the country to inde-
Research has emphasized the critical importance pendence from the British, although his ultimate
of the charismatic leader’s vision, including aspects vision of Hindus and Muslims living in peaceful
of both the content and delivery style of the leader’s cohabitation was never achieved. His principles of
message. Charismatic leaders are able to distill com- nonviolence transformed his own country and
plex future ideals into simple messages with wide- inspired future leaders all over the world.
spread and emotional appeal, such as Martin Luther Overall, Gandhi’s followers realized many of
King, Jr.’s “I Have a Dream” speech. They also use the beneficial outcomes of charismatic leadership,
a variety of communication techniques to appeal to as Gandhi is credited with broadening and elevat-
their followers, such as using colorful, vivid lan- ing the interests of his followers, generating aware-
guage and imagery rooted in shared values. ness and commitment to the mission of the group,
Charismatic leaders also frequently employ and motivating followers to go beyond their own
symbols to their advantage, including the site of interests and sacrifice for the good of the Indian
their speech, visual symbols, props, music, and people as a whole. This example illustrates that
lighting to enhance the appeal of their message and charismatic leadership involves the actual charac-
increase the level of excitement and emotion in fol- teristics of the leader as well as characteristics that
lowers. Similarly, they use tone of voice, inflection, are attributed to leaders by followers within a
pauses, and gestures to increase the intensity and given situation. If the leader’s extraordinary quali-
emotional meaning of their message. Charismatic ties do not clearly resonate with the values and
speeches frequently incorporate analogies, repeti- needs of followers, he or she is likely to be rejected
tion, metaphors, and stories to bring the vision as a radical or mocked as delusional. However, if
“alive” for followers, and communicate optimism the leader’s extraordinary qualities provide a good
that, together, group members can achieve the fit for, or proposed solution to, followers’ needs
promise of a better future if they are willing to sup- and anxieties, followers may even exaggerate how
port the leader and are prepared to make sacrifices exceptional their leader is and idolize him or her.
for his or her cause. In this way, a charismatic leader may attain a
seemingly magical or superhuman persona in the
eyes of his or her followers.
A Historical Illustration
A familiar example of an admired leader, Mohandas
Limitations and the Darker Side
Gandhi, illustrates the ingredients of the charismatic
leadership relationship. First, Gandhi’s followers Charismatic leadership has been defined as a rare
often credited him with possessing extraordinary form of leadership in which the leader is often
74 Charismatic Leadership

irreplaceable because of followers’ dependence on benefit. Other more attractive and credible leaders
the leader’s skills. Therefore, social movements may challenge the vision or the effectiveness of the
and change processes are often difficult to main- current leader.
tain once the charismatic leader is gone, and fol- In addition, charismatic leaders may become
lowers struggle to carry on the momentum of victims of their own success. As the change move-
change. In addition, because charismatic leader- ment grows larger and more powerful, or as their
ship often results in unquestioning obedience to organizations expand in scope and influence, char-
the leader’s directives, followers themselves may be ismatic founders may become increasingly irrele-
less likely to develop leadership skills and experi- vant or lack the expertise to deal with new
ence that will sustain the change process after the challenges. There is also the possibility that their
leader is gone. unconventional behaviors may become distracting
In addition, charismatic leadership also has a or counterproductive as the focus shifts from cre-
darker side. None of the ingredients of the charis- ating change to implementation and performance.
matic leadership relationship guarantee that the
leader’s vision will be morally defensible or ethical, Michelle C. Bligh and Jeffrey C. Kohles
nor do they prohibit followers from carrying out See also Contingency Theories of Leadership; Great
unethical or violent behaviors in pursuit of the Person Theory of Leadership; Interactionist Theories
leader’s vision. Charismatic leaders of the past such of Leadership; Leader-Member Exchange (LMX)
as Adolf Hitler and Jim Jones were able to create Theory; Leadership; Path–Goal Theory of Leadership;
widespread appeal for their visions of racial supe- Personality Theories of Leadership; Social Identity
riority and revolutionary suicide, respectively. Due Theory of Leadership; Transactional Leadership
to the extraordinary, even heroic qualities that fol- Theories; Transformational Leadership Theories;
lowers may attribute to such charismatic leaders, Vertical Dyad Linkage Model
these leaders may begin to believe that they really
are exceptionally qualified to determine the fate of
Further Readings
their followers, resulting in increasingly autocratic,
manipulative, dictatorial leadership that is intoler- Bligh, M. C., Kohles, J. C., & Meindl, J. R. (2004).
ant of dissent or alternate points of view. Charisma under crisis: Presidential leadership,
In addition, due to their often powerful appeal rhetoric, and media responses before and after the
and nonconventional or radical visions for the September 11th terrorist attacks. Leadership
future, charismatic leaders are often extremely Quarterly, 15(2), 211–239.
divisive figures that create bitter enemies as well as Conger, J., & Kanungo, R. N. (1987). Toward a
devoted followers. In some situations, the charis- behavioral theory of charismatic leadership in
matic leader may prove so divisive that he or she organizational settings. Academy of Management
Review, 12(4), 637–647.
paralyzes the change process. In other situations,
Howell, J. M., & Shamir, B. (2005). The role of followers
charismatic leaders become public targets for the
in the charismatic leadership process: Relationships
opposition, and consequently leaders such as
and their consequences. Academy of Management
Gandhi, King, and John F. Kennedy were ulti-
Review, 30, 96–112.
mately assassinated for their beliefs. Hunt, J. G., Boal, K. B., & Dodge, G. E. (1999). The
Finally, it is important to note that charismatic effects of visionary and crisis-responsive charisma on
leadership is often a fleeting phenomenon that can followers: An experimental examination of two kinds
be gained or lost as circumstances change. of charismatic leadership. Leadership Quarterly,
Charismatic leadership can be lost if the crisis 10(3), 423–448.
ends, or if followers become more confident and Shamir, B., House, R. J., & Arthur, M. B. (1993). The
feel that they are capable of solving problems on motivational effects of charismatic leadership: A self-
their own. Also, charismatic leaders may make concept based theory. Organization Science, 4,
decisions that result in failure, seem to betray their 577–594.
core vision or followers’ needs, or focus more on Yukl, G. (1999). An evaluation of conceptual weaknesses
obtaining power and influence for themselves in transformational and charismatic leadership
than on supporting the greater cause for followers’ theories. Leadership Quarterly, 10(2), 285–305.
Children: Stereotypes and Prejudice 75

Evidence suggests that such social categoriza-

Children: Stereotypes tion can result in preference for one’s own group.
and Prejudice Psychologists have investigated this in studies
using a visual preference task. In this task, infants
Prejudice, the holding of negative attitudes toward are presented with examples from two racial cate-
others based on the groups to which they belong gories simultaneously, and how long they look at
and the stereotypes attached to these groups, con- each example is used to indicate preference. Studies
tinues to be a major source of strife and conflict using this task show that by the time they are
throughout the world. Do children show prejudice 3 months old, infants prefer to look at faces of
and have stereotypes? If so, how do stereotypes their own racial group rather than those of other
and prejudice arise in early life? What are the dif- racial groups. This preference is not typically
ferent ways that prejudice and stereotypes emerge shown by newborns and is only present in 3-month-
in childhood, and what forms can they take? These old infants living in a predominantly racially
are the questions that are addressed in this entry. homogeneous environment. These findings suggest
People commonly think that young children are that the early development of own-race preference
innocent and devoid of stereotypes and prejudice. in infancy is linked to living in an environment that
However, research in developmental and social exposes the children only to own-race individuals.
psychology has shown that children exhibit many Developmentally, infants have social categoriza-
types of biases at an early age. These can be based, tion ability and, depending on their environment,
for example, on someone’s gender, race, ethnicity, this may result in certain group preferences.
nationality, or body type. The findings suggest that However, at this stage they do not have the ability
understanding the origins and nature of prejudice to express prejudice or stereotypes due to their
in children should be a high priority if we are to limited cognitive and linguistic abilities. As infants
establish effective policy for combating its negative become young children, they often express biases
consequences. Given that stereotypes and preju- directly or explicitly in the words they use to
dice are hard to change in adulthood, most psy- describe different social groups, or more indirectly
chologists agree that interventions must be or implicitly by forming mental associations link-
implemented early in life to be successful. ing their own group (rather than other groups)
with positive experiences, emotions, or attributes
(i.e., showing implicit biases).
Origins of Prejudice
Developmental and social psychologists have sus-
Implicit Biases in Children
pected that prejudice in children may originate
from the child’s early ability to categorize the Recent studies have provided evidence for when
social world. Children develop the ability to recog- implicit biases emerge in childhood. Using the
nize characteristic features of people from their Implicit Associations Test (IAT), researchers have
own group and other groups, and then use this shown that adults have biases toward others,
information to cluster individuals together in which they are not aware of, based solely on race.
social categories. Adults are known to be more This is evidenced by the fact that White adults
accurate at recognizing a face from their own more quickly associate negative words with out-
racial group than from an unfamiliar racial group. group (Black) faces than with ingroup (White)
That this type of face processing is not present in faces. These findings have spawned an industry of
infants at 3 months of age suggests that it is not IAT studies, with much debate about the extent to
innate. In fact, it emerges by 6 months in infants which such associations reflect prejudice or bear on
living in an environment with little racial diversity behavior, such as discrimination toward others.
(i.e., a racially homogeneous environment). This Some developmental studies have examined
indicates that facial input from the infant’s visual implicit bias in childhood using IAT-type method-
environment is a key contributing factor. Therefore, ologies. One study of White British children, using
at an early age children show categorization ability a child-friendly pictorial IAT, found that implicit
based on race. racial and national biases were present from 6 to
76 Children: Stereotypes and Prejudice

16 years of age. This children’s version of the IAT Developmental research has examined prejudice
measured the relative strength of association in the form of explicit preference for one social
between a concept (e.g., “White British” or “Black group over another. Early work on this form of
British”) and an attribute (e.g., “good” or “bad”). explicit bias included the doll test, which showed
Implicit bias was judged to be present if the chil- that Black American children in segregated schools
dren showed faster reaction times for stereotypical preferred White dolls to Black dolls. This research
(e.g., “White British” and “good”) than counter­ was influential in Brown v. Board of Education,
stereotypical (e.g., “Black British” and “good”) the 1954 Supreme Court case that outlawed school
associations. Studies have also found that these segregation in the United States. Other measures
implicit biases remain in older children who typi- have examined explicit forms of prejudice by ask-
cally show reduced explicit bias. This research sug- ing children to attribute positive (e.g., “clean,”
gests that implicit biases based on associations “smart”) and negative (e.g., “mean,” “dirty”)
made between different categories and their attri- traits to a White child or a Black child or to both
butes (either positive or negative) are established children. Explicit forms of ethnic prejudice based
early in life and remain stable, relatively hard to on the above measures are known to develop from
change, and less open to conscious control. 4 to 5 years of age among ethnic majority children.
Other studies have looked at how children Unlike implicit prejudice, which remains relatively
evaluate the intentions of an individual from a stable once it appears in childhood, explicit ethnic
different racial group, in an ambiguous but bias typically declines in ethnic majority children
familiar everyday peer encounter, to determine from approximately 7 years of age.
whether children have implicit biases. The studies Some psychologists explain this developmental
found that 6- to 9-year-old European American trend in ethnic majority children’s prejudice by
children attributed more positive intentions to a attributing it to young children’s poor cognitive
White child than to a Black child in potential ability to judge people from different groups using
“pushing” and “stealing” peer encounters on the individual characteristics (e.g., friendly, hard-
playground. working). Instead, these researchers argue that
Does this suggest that implicit biases are auto- cognitively immature children judge an individual
matic in children? The answer would appear to be based on the group to which he or she belongs (e.g.,
“no”—since these biases were found only among boys or girls). Thus, according to this analysis, chil-
European American children in racially homoge- dren show prejudice because they cannot see the
neous schools; European American children of the similarities between individuals in different groups
same age, in the same school district, but enrolled and the differences between individuals in the same
in heterogeneous schools did not attribute more group. However, recent studies have found a weak
positive intentions to their ingroup than the out- relationship between this cognitive ability and chil-
group; in fact, race was not used to attribute nega- dren’s prejudice, and there is an extensive literature
tive intentions. showing that cognitively mature adults can still
show prejudice. So a cognitive explanation does
not seem adequate for explaining why explicit
Explicit Prejudice in Children
prejudice declines during middle childhood.
Prejudice takes many forms in childhood. This is
not surprising, given that what counts as prejudice
Self-Presentation and Explicit Prejudice
changes as children develop both cognitively and
socially. The nature and complexity of a 5-year-old Some researchers have argued that the develop-
child’s group-related attitude will be manifestly mental decline in explicit ethnic prejudice reflects
different from a 14-year-old adolescent’s attitude. children’s concern about social desirability, in par-
So implicit biases are only one form of childhood ticular their increasing awareness of how others
prejudice, and there is an extensive history of might perceive them and how they can promote a
research in developmental psychology revealing positive impression of themselves to significant
other forms of prejudice in childhood that pro- others. This self-presentation account contends
vides a broader picture. that children develop the ability to strategically
Children: Stereotypes and Prejudice 77

control their expression of prejudiced attitudes which often show a different developmental pat-
and behavior. Developmental research certainly tern from the explicit forms of prejudice described
indicates that by middle childhood, children earlier.
have the social–cognitive capacity to understand
self-presentational motives and engage in self-
Explicit Judgments About Social Exclusion
presentational behavior.
Importantly, children’s self-presentation of prej- Researchers have recently begun to assess chil-
udice requires attention to prevalent social norms dren’s evaluations and reasoning about the exclu-
about the explicit expression of prejudice. Social sion of ingroup and outgroup peers within everyday
norms prescribe cultural expectations regarding group settings. This research provides a broader
attitudes, values, and behavior. These expectations insight into the development of prejudice and use
may be specific to social groups or more wide- of stereotypes in childhood. Developmental
spread within society. Once children understand research has demonstrated that from a young age,
these social norms, they may strategically present children emphasize moral reasoning (e.g., fairness)
themselves as acting in accordance with them, when judging social exclusion based on group
thereby giving a positive impression of themselves membership, such as gender or race, in straightfor-
to relevant and significant others. ward or unambiguous situations (e.g., “Is it all
In support of this argument, developmental psy- right or not all right to exclude a boy from a ballet
chologists have recently shown that group norms club?”). However, children often resort to stereo-
affect the self-presentation of young children’s types to justify exclusion when complexity or
explicit ethnic prejudice and the development of ambiguity is added. For example, who should a
children’s prejudiced exclusion of peers. This ballet group pick when only one space is available
research has found that increasing children’s and two children want to join—one who matches
accountability to their peer group, in that their the stereotype or one who does not? Moreover,
actions are visible and may have to be defended, this research has found that with age, children also
causes children to increase or decrease their preju- begin to use more socially conventional reasoning
diced judgments in line with the dominant norm in (e.g., the need to adhere to social norms or stereo-
their group. types, concerns about effective group functioning)
For example, one study found that 5- to 16-year- to justify exclusion of their peers based on either
old White British children who were highly aware race or gender. Such referencing of social norms
of the social norm against expressing explicit rac- and group functioning may serve as a proxy for
ism spontaneously showed little explicit racial prejudice and stereotypes.
prejudice. In contrast, children with little aware- For example, one recent study found that, when
ness of this norm inhibited their racial prejudice asked to make decisions about exclusion in com-
only when their group’s antiprejudice norm was plex situations, the majority of children between 6
made salient by increasing their accountability to and 12 years of age increasingly justified racial or
the group. Still other studies have shown that chil- gender exclusion using socially conventional rea-
dren can increase rather than inhibit their preju- sons, such as mentioning how an individual chal-
dice based on national group membership when lenges the gender stereotype (“It will be weird to
they are made accountable to their national have a boy wearing leotards, so they should choose
ingroup. This finding fits with studies showing the girl.”). Moreover, a range of studies have
that national prejudice is often seen as more legiti- shown that children’s identification with the per-
mate or acceptable than racial prejudice. These son who is the excluder is related to their justifying
studies show that the self-presentation process racial or gender exclusion based on socially con-
operates to encourage prejudice in the domain of ventional reasons. These findings indicate that
nationality, whereas with race, children typically older children, who are less likely to show the
self-present by inhibiting their prejudice. explicit biases described in the previous section, are
The next section describes another facet of prej- nonetheless willing to condone prejudice and social
udice recently studied by developmental scientists, exclusion using stereotypical reasoning. Together,
namely explicit judgments about social exclusion, these studies suggest that with age, children in
78 Children: Stereotypes and Prejudice

everyday situations find less direct and more subtle seriously attempts to reduce these problematical
ways of showing bias and stereotyping. phenomena in childhood. Indeed, research suggests
Other recent developmental research has docu- that reducing prejudice in children can be accom-
mented age-related increases in social exclusion plished through a variety of methods, including the
judgments, with older children excluding peers promotion of intergroup contact, social–cognitive
who challenge both moral principles and their own skills, empathy, moral reasoning, and tolerance.
group norms about loyalty to the group. This
increased social exclusion of deviant (i.e., “black Adam Rutland
sheep”) peers by older children reflects the grow- See also Black Sheep Effect; Categorization; Children’s
ing importance of group identity as children age Groups; Discrimination; Implicit Association Test
and advance their understanding of how groups (IAT); Inclusion/Exclusion; Prejudice; Stereotyping;
operate. Specifically, this research has found that Subjective Group Dynamics
younger children evaluate the actions of their peers
only in terms of morality, that is, adherence to
moral norms such as fairness or equality, or as self- Further Readings
ish, that is, acting out of self-interest. Older chil- Aboud, F. E. (1988). Children and prejudice. Oxford,
dren, however, judge peer group members in terms UK: Blackwell.
of both their morality and their adherence to Abrams, D., & Rutland, A. (2008). Subjective group
group norms about what a genuine group member dynamics approach: Applications to children and
should do in different settings. adults. In S. Levy & M. Killen (Eds.), Intergroup
relations: An integrative developmental and social
psychological perspective. New York: Oxford
University Press.
Research indicates that children develop prejudice Bigler, R. S., & Liben, L. S. (2007). Developmental
and stereotypes from an early age. The foundation intergroup theory: Explaining and reducing children’s
of this prejudice and stereotyping is the develop- social stereotyping and prejudice. Current Directions
ment of social categorization during infancy. in Psychological Science, 16, 162–166.
Infants begin to show categorization ability in the Brown, R. J. (1995). Prejudice: Its social psychology.
domain of race and also show a visual preference Cambridge, MA: Blackwell.
for faces from their own group. Implicit biases are Katz, P. A. (2003). Racists or tolerant multiculturalists?
known to appear around 6 years of age and remain How do they begin? American Psychologist, 58,
relatively stable through childhood into adoles- 897–909.
cence, though some evidence suggests implicit Killen, M. (2007). Children’s social and moral reasoning
racial attitudes are more negative in ethnically about exclusion. Current Directions in Psychological
Science, 16, 32–36.
homogenous environments. Explicit racial preju-
Killen, M., Lee-Kim, J., McGlothlin, H., & Stangor, C.
dice measured through preference or trait attribu-
(2002). How children and adolescents evaluate gender
tion tasks appears around 4 to 5 years of age in
and racial exclusion. Monographs of the Society for
racial majority group children, but generally
Research in Child Development, 67(4, Serial No. 271).
decreases from middle childhood. In contrast, Nesdale, D. (2001). Development of prejudice in children.
children’s reasoning and judgments about social In M. Augoustinos & K. J. Reynolds (Eds.),
exclusion can become more biased with age, as Understanding prejudice, racism and social conflict
children begin to use stereotypes and conventional (pp. 57–72). London: Sage.
reasons to justify social exclusion in everyday com- Rutland, A. (2004). The development and self-regulation
plex situations. Psychologists have typically of intergroup attitudes in children. In M. Bennett &
explained developmental trends in children’s preju- F. Sani (Eds.), The development of the social self
dice using either a cognitive development or self- (pp. 247–265). East Sussex, UK: Psychology Press.
presentational account. Rutland, A., Cameron, L., Milne, A., & McGeorge,
Prejudice and stereotyping begin in early child- P. (2005). Social norms and self-presentation:
hood, often making it difficult to change these atti- Children’s implicit and explicit intergroup attitudes.
tudes in adulthood. For this reason, we need to take Child Development, 76, 451–466.
Children’s Groups 79

experiences and in the peer group itself of chil-

Children’s Groups dren’s behavior, skills, and social understanding
that contribute to such individual differences; and
Peer groups constitute the primary source of chil- to explain peer group processes that contribute to
dren’s socialization outside of the family. development and adjustment.
Competence in the peer group is critical for many
aspects of children’s development, including cog-
nitive development and school achievement as
The Peer Group in Development
well as social, emotional, and personality develop- Observations of children in family, child care,
ment. Within their groups, children are more or and laboratory settings, as well as cross-culturally
less accepted by their peers, more or less visible, in small villages of hunter–gatherer societies,
more or less dominant, and more or less con- show that peer relationships emerge late in the
nected to other group members. These individual second year of life and become increasingly com-
differences in peer group experiences are associ- plex and a more regular part of children’s social
ated with different socioemotional outcomes later experiences over the course of childhood. By
in development. As a result, much of the scholar- middle childhood, children spend a third or more
ship concerning children’s peer groups has been of their time in the company of their peers and
devoted to measuring and explaining the sources begin to express a preference for spending time
of individual differences in children’s integration with peers. Peer relationships are more egalitar-
into the peer group and the developmental conse- ian than those with adults, feature more play and
quences of these differences. More recent and less more conflict, and tend to be less stable and
extensive research literature addresses questions enduring than relationships in the family. Hence,
about children’s attitudes toward their own group children’s peers provide unique experiences not
and groups of other children. available in the family and unique opportunities
for acquiring a wide range of social and emo-
tional competencies.
Perhaps one of the most interesting characteris-
Contemporary research interest in children’s peer tics of children’s peer groups is that even with
relationships is usually traced to the discovery by children as young as 3 or 4 years of age, whenever
Harry Harlow in the 1950s that infant rhesus children assort themselves voluntarily, these groups
monkeys reared with their mothers but without are composed almost exclusively of same-sex chil-
exposure to other young monkeys failed to acquire dren. The amount of time children spend with
the social skills necessary for successful adaptation opposite-sex peers declines over childhood and
to group living. Harlow and his students also dis- remains short until adolescence. Sex segregation
covered that the profound social and emotional begins early and trumps any other characteristic
deficits resulting from early maternal deprivation on which children self-segregate, including race.
could be largely overcome by “peer therapy” pro- Girls precede boys in preferring to play with their
vided by extended play with younger monkeys. own sex during the toddler and early preschool
Later, psychologists studying the origins of years, but by childhood boys are more exclusion-
adult psychopathology began to find that difficul- ary. Children of both sexes establish and police the
ties in childhood peer relations were a regular boundaries of their sex-segregated groups, teasing
antecedent of serious problems later in life. Thus, or rejecting peers who play with children of the
children’s functioning in the peer group appeared opposite sex. Perhaps as a result, children who
to be formative in the development of competence spend larger amounts of time in opposite-sex peer
and to have long-lasting consequences. Since the groups than is typical for their age tend to be less
1970s, scholars have worked to measure and char- popular and well adjusted than children who
acterize development and individual differences in prefer same-sex group play.
children’s peer group experiences; to identify the Because sex segregation appears early, is robust,
correlates and causes of differential success in the and appears to be universal, scholars have endeav-
peer group; to determine the origins in family ored to explain why it occurs, how experiences
80 Children’s Groups

differ in boys’ and girls’ groups, and the functions Measurement

of sex-segregated play for development. Eleanor
Children’s peer group acceptance is most com-
Maccoby’s work has been influential in promoting
monly described using sociometric classifications.
and studying such questions. Maccoby has pro-
A frequently used method for indexing sociometric
posed that sex segregation occurs because of
status was developed by John Coie in the 1980s. In
incompatibilities in boys’ and girls’ play and inter-
this procedure, children are asked to select among
action styles. Boys’ play is predominantly active,
their peers those they like most and those they like
physical, rough, and competitive and occurs in
least from a roster of children in their group (e.g.,
large groups, whereas girls’ play is characteristi-
in a class or camp cabin). Children are then classi-
cally quiet, social, verbal, and cooperative, occur-
fied into categories based on how preferred they
ring in dyads or small groups. Therefore, to
are by their peers. Popular children are liked by
understand sex segregation, researchers must
many peers and disliked by few. Rejected children
explain how sex differences in play style develop
are disliked by many peers and liked by few.
as a function of biology and socialization. This is
Controversial children are liked by many peers but
currently an active area of inquiry in developmen-
disliked by many others. Neglected children are
tal psychology.
seldom singled out as being either liked or disliked.
Researchers have also documented that boys’
Average children are those who are liked by some
and girls’ groups differ in the structure and the-
and disliked or ignored by others, and generally do
matic content of their interactions, the density of
not fit any of the other categories.
their social networks, their relationship goals and
There is some debate about whether such selec-
concerns, the types and amount of peer stress
tions should be limited to same-sex peers or should
encountered in the group, and group members’
include both sexes, how large the reference group
interpersonal qualities (such as trust, validation,
should be (e.g., classroom vs. school), and the type
and closeness). There is much less research on
of cut-off score that should be used for creating the
how these different experiences relate to develop-
categories. Different methods yield more or less
ment. Some research suggests that the amount of
extreme groups, which are, in turn, more or less
time children spend in same-sex peer groups influ-
stable over time and may be more or less predictive
ences how sex-typed their behavior is, and sub-
of optimal or problematic outcomes.
stantial research points to sex differences in
Another frequently used procedure is to ask
adjustment, especially in adolescence. For exam-
children to rate each of their peers on a single scale
ple, boys are more likely to exhibit behavior prob-
that varies from “like very much” to “dislike very
lems, whereas girls are more likely to exhibit
much.” The average rating a child receives from
anxiety and depression. One reasonable hypothe-
his or her peers reflects how accepted the child is
sis is that different experiences in boys’ and girls’
in the peer group. This method does not yield dif-
peer groups contribute to sex differences in devel-
ferent sociometric categories, but because it per-
opmental outcomes. Research is currently address-
mits children to evaluate all of their peers, and not
ing this hypothesis.
only those liked most and least, some investigators
argue that it provides a more valid measure of peer
Individual Differences group acceptance.
in Peer-Group Acceptance
Causes and Correlates
Jacob Moreno’s work in the 1930s on the interper-
sonal forces of attraction, repulsion, and indiffer- Regardless of the method used, children who
ence set the stage for much of the research in the differ in peer group acceptance have been found to
latter part of the 20th century on children’s accep- have different behavioral profiles, self-perceptions,
tance by their peer group. This work has empha- socioemotional adjustment, patterns of school
sized that different children are perceived differently achievement, and attitudes and cognitions about
by their peers, with some children being well liked their peers. Initially, research identified broad
and widely accepted, while others are rejected, and dimensions of social behavior such as aggression,
still others are ignored. withdrawal, and prosocial behavior as correlates
Children’s Groups 81

of various sociometric status groups. For example, In addition to differing in social behavior, chil-
rejected children tend to be more hostile and dren from different sociometric groups also exhibit
aggressive toward their peers, whereas popular distinct patterns of social information processing.
children tend to be more friendly and cooperative. In the 1980s, Kenneth Dodge formulated an influ-
Experimental studies of playgroups formed by ential model of individual differences in how chil-
combining previously unacquainted children have dren encode, interpret, and act on social information
shown that such behavioral differences shape chil- during interactions with peers. Empirical research
dren’s sociometric status. has largely confirmed the basic tenets of this
In more recent years, researchers have uncov- model. For example, rejected children tend to per-
ered substantial complexity both in the sociomet- ceive and interpret ambiguous social behavior by
ric categories themselves and in the kinds of their peers as hostile, a characteristic termed the
behavior, cognitions, and attitudes associated with hostile attribution bias. They also endorse hostile
children’s sociometric status. There are thus sub- goals and coercive solutions to social problems,
types of rejected children, such as withdrawn chil- evaluate aggressive solutions positively, and over-
dren and especially aggressive children. Interestingly, estimate how well liked they are by peers.
not all types of aggression are associated with Popular children, in contrast, more accurately
rejection. Indeed, only about half of aggressive encode social cues, attribute benign intentions to
children are rejected by their peers. Experimentally ambiguous behavior by peers, and generate proso-
constructed groups of rejected and “unrejected” cial solutions to social problems. Research has
children have shown that in early childhood, peer shown that such differences in social information
rejection is associated with higher rates of instru- processing originate in representations or schemas
mental aggression to obtain desired objects or that are automatically activated during encounters
positions. Among older children, angry, impulsive with peers. Attempts to alter children’s social per-
aggression in response to provocation, as well as ceptions and cognitions and/or their social problem-
unprovoked, person-centered aggression, is also solving strategies have produced some limited
associated with rejection. changes in peer acceptance. This suggests that how
Furthermore, different types of aggression are children encode and interpret others’ behavior, and
associated with rejection in girls’ groups and in their goals and solutions when they encounter con-
boys’ groups. Whereas physical aggression is often flict or difficulty with peers, may be causally
characteristic of rejected boys, relational aggres- related to their peer group status.
sion in which children use nonphysical means such
as gossip to exclude, harass, or threaten others is
more characteristic of rejected girls. In addition,
the extent to which aggression causes peer rejec- A lack of acceptance by peers has been regularly
tion depends on how normative it is in the peer shown to predict a host of negative developmental
group; in peer groups where aggression is more outcomes for children, including poor school atti-
frequent, it is less likely to result in rejection. tudes and achievement, psychological maladjust-
Likewise, shyness and withdrawal are associated ment, and delinquency. Longitudinal studies have
with rejection only in later childhood when they shown that such associations are especially robust
become nonnormative, although the picture is for children who are stably rejected by their peers
more complicated for children from non-Western over time, particularly for those who are aggres-
cultures (e.g., China). sive and disruptive. It is as yet unknown to what
Similar distinctions have been made among extent these associations are driven by the negative
popular children. For example, some children per- peer experiences these children encounter, by their
ceived as popular by their peers are more domi- failure to acquire important skills from peer group
neering, assertive, and manipulative than they are socialization because of their lack of integration in
prosocial and cooperative. Prosocial behavior the group, by exposure to the socializing influences
tends to be associated with peer acceptance in of other deviant peers, or by some underlying qual-
groups where positive peer-directed behavior is ity of the children themselves or experiences out-
relatively common. side the peer group that feed both peer rejection
82 Children’s Groups

and other social and emotional problems. There is Additional measures are often derived to index
empirical evidence for each of these potential how central, visible, or influential such affiliative
mechanisms, as well as evidence for more complex networks or cliques are. Some are considered
theoretical models that posit multiple, interacting nuclear, others secondary, and others peripheral.
influences. Higher centrality groups tend to be larger than
For example, existing vulnerabilities in the child groups on the periphery. Children themselves can
that might predispose him or her to poor outcomes also be identified as more or less central within
(such as aggression or hyperactive, impulsive their own affiliative networks. The correlates of
behavior) appear to be exacerbated by negative centrality are different for boys and girls. For boys,
experiences with peers, such as rejection or victim- athletic ability, leadership, dominance, and per-
ization. Some research has found that children’s ceived popularity tend to be important for high
reputation in their peer group is self-perpetuating network centrality. For girls, academic skills and
and that peers are likely to discount information achievement, leadership, and perceived popularity
about a child when it runs counter to the peer tend to be correlated with network centrality.
group’s prevailing perceptions of the child. This There is little research as yet concerning how
contributes to the stability of children’s sociomet- group processes such as norm establishment and
ric status as well as their social and emotional maintenance, leadership, and cohesion might vary
experiences in their peer group, and makes it dif- as a function of group size or centrality.
ficult for children with low social status to improve In addition to declining in size with age, peer
their standing. networks become less exclusive and more perme-
able so that children increasingly belong to more
than one clique and the interconnections increase
Children’s Social Networks
among different groups. Boys’ groups generally
Children’s immediate peer groups are embedded in tend to be more interconnected and less exclusive
larger social networks of peer relationships that than girls’ groups, possibly because they are larger
provide unique experiences as well. Social network overall. Across short periods of time, up to 6
analysis identifies and examines patterns in chil- weeks, affiliative networks remain stable with rela-
dren’s specific peer affiliations. This contrasts with tively little change in members. Instability increases
the sociometric approach, which focuses on chil- over periods of a year or more, and younger chil-
dren’s acceptance by their larger peer group. dren’s groups are generally less stable than high
Research on children’s social networks is compara- school groups.
tively recent and has been influenced by concep- Theory and empirical research have focused on
tual frameworks from sociology. the factors and processes that underlie children’s
To identify their social networks, children are affiliative group formation and influence. Factors
asked questions such as who “hangs around with” important to group formation include propinquity,
whom, who belongs to which groups, and which familiarity, and similarity. Children are most likely
children do not belong to any group. Rosters are to affiliate with classmates and other close associ-
not used so that children are limited to reporting ates than with children in their school or neighbor-
on peers about whom they are knowledgeable. hood with whom they have less systematic contact.
Because of the verbal and cognitive demands of This appears to be especially true for younger chil-
this procedure, it is typically used only with chil- dren, whose classrooms are more likely to be self-
dren past third grade. By combining children’s contained. Familiarity is also an important
responses statistically, researchers can create a map determinant of group affiliation, often more impor-
of children’s affiliations with one another based on tant than dissimilarities in academic performance
shared perceptions across respondents. A child is or socioeconomic status.
typically considered to belong to a particular Nevertheless, a key force in peer group affilia-
group or network if so identified by 50% or more tion is similarity based on features such as age,
of his or her peers. On average, network size tends race, physical characteristics, academic achieve-
to increase between elementary school and middle ment, parents’ income, and qualities such as
school, then decrease over high school. aggressiveness or popularity. Some scholars have
Civil Rights Legislation 83

suggested that these shared attributes may provide outgroup prejudice may be weaker until middle
grounds for mutual validation and approval dur- childhood. Also like adults, children’s ingroup
ing a period when social identities are developing. favoritism depends on factors such as the status of
Similarity within children’s peer groups may also the child’s group; children in low-status and/or
reduce intragroup conflict and other threats to the minority groups are less positive about their own
group’s cohesiveness and integrity. group and sometimes even favor the outgroup.
Thus children choose to affiliate with peers who This research area has been concerned primarily
are like them, and within-group socialization pro- with explaining the childhood roots of social ste-
cesses can consolidate, amplify, or alter children’s reotyping and racial prejudice. It would be interest-
behavior and attitudes over time. Research on the ing and productive to integrate research and theory
socialization of children’s attitudes and behavior on children’s peer group affiliation, acceptance,
within their peer group is relatively recent and and socialization with work on the formation and
largely limited to adolescents. The bulk of this maintenance of their intergroup attitudes.
research has been devoted to peer group affilia-
tions among aggressive or delinquent adolescents. Celia A. Brownell
One of the strongest predictors of continuing See also Children: Stereotypes and Prejudice; Cliques;
aggression and problem behavior in adolescence is Gender and Behavior; Group Socialization; Inclusion/
affiliation with peers who also use illegal sub- Exclusion; Sociometric Choice
stances or engage in other risk-taking, violent, or
delinquent behaviors.
Thomas Dishion has coined the term “deviancy Further Readings
training” to describe the processes of reinforce-
Bigler, R., & Liben, L. (2007). Developmental intergroup
ment and approval within adolescent boys’ affilia-
theory: Explaining and reducing children’s social
tive networks that serve to maintain or increase
stereotyping and prejudice. Current Directions in
such behavior and associated normative beliefs. Psychological Science, 16, 162–166.
Although less researched, similar processes have Collins, W., & Steinberg, L. (2006). Adolescent
been shown to operate within affiliative networks development in interpersonal context. In W. Damon
formed on more positive attributes (such as aca- & N. Eisenberg (Eds.), Handbook of child psychology
demic achievement, in which members maintain or (6th ed., Vol. 3, pp. 1003–1067). Hoboken, NJ:
increase their similarity with respect to academi- John Wiley.
cally relevant behaviors). Some research has also Gifford-Smith, M., & Brownell, C. (2003). Childhood
shown that peer group affiliations can reduce a peer relationships: Social acceptance, friendships, and
child’s negative attitudes, such as racism, when the peer networks. Journal of School Psychology, 41,
peer group’s attitudes are more positive. Because 235–284.
of the heightened importance of peer group accep- Maccoby, E. (1998). The two sexes: Growing up apart,
tance for adolescents, coupled with the salience of coming together. Cambridge, MA: Harvard University
identity formation and general impressionability Press.
during this period, it is possible that peer group Rubin, K., Bukowski, W., & Parker, J. (2006). Peer
socialization effects are stronger or qualitatively interaction, relationships, and groups. In W. Damon &
different during adolescence than in adulthood. N. Eisenberg (Eds.), Handbook of child psychology (6th
This issue has yet to be addressed, however. ed., Vol. 3, pp. 571–645). Hoboken, NJ: John Wiley.
An independent research literature on children’s
attitudes toward their own groups versus other
groups has shown that, like adults, children have
more positive attitudes toward their own group Civil Rights Legislation
(ingroup favoritism) and more negative attitudes
toward other groups (outgroup prejudice). The Civil rights legislation is a broad term that may be
developmental picture is not yet clear, in part applied to any laws or legal rulings designed to
because of measurement limitations, but it appears protect the basic human rights of individuals any-
that ingroup favoritism develops by age 5, whereas where in the world. These rights include any of a
84 Civil Rights Legislation

range of principles that ensure freedoms, liberties, For example, the United States Congress passed
and general happiness to which all humans are a series of civil rights acts in 1957, 1960, and 1964
considered entitled from birth, such as free speech, that provided further protection against discrimi-
religious freedom, participation in electoral pro- nation for a wide range of groups, establishing
cesses, due legal processes in the court system, and protected classes such as race, sex, and nation of
so forth. The term civil rights legislation is most origin. The Voting Rights Act of 1965 outlawed
commonly used to refer to laws passed during the literacy tests and other barriers to voting, and the
civil rights movement of the 1960s in the United Fair Housing Act of 1968 outlawed discrimination
States, but the term could be applied more broadly in the housing market. In 1990 the Americans with
to describe legal action to protect human rights Disabilities Act added the physically disabled to
further back in history as well as across the globe. the list of protected classes.
For example, civil rights legislation in the United Laws that threaten civil rights also can be
States arguably began with the Constitution of deemed unconstitutional through judiciary deci-
1787, and civil rights are a concern for all nations sions such as those handed down by the United
interested in ensuring rights and freedoms for its States Supreme Court. For example, many schools
citizens. Thus, for example, the United Kingdom in the early 20th century were racially segregated
instituted affirmative action policies to ensure under the notion that the education system was
equal opportunities for Catholics and the poor, “separate but equal”; however, the Supreme Court
and South Africa established anti-apartheid mea- ruled this system unequal and unconstitutional in
sures to end to segregation. the 1954 Brown v. Board of Education case, set-
In the United States, civil rights legislation is ting the stage for the mandatory desegregation of
accomplished through bills passed into law, court American schools.
rulings, and executive orders. The most salient The president of the United States may encour-
examples of U.S. civil rights legislation have con- age civil rights through executive orders. For
cerned Blacks and women, but civil rights legisla- example, President Lincoln’s executive order of
tion has been extended to a wide range of groups 1862—the Emancipation Proclamation—set the
and people in society. Which groups deserve legal stage for the abolition of slavery, and President
protection against civil rights violations and what Truman’s executive order in 1948 formally deseg-
constitutes a “civil right” remain points of contro- regated the military, allowing Black and White
versy. Nevertheless, the ultimate goal of civil rights soldiers to serve in the same units, equally.
legislation is to provide equal freedoms and liber- The goal of American civil rights legislation is to
ties for those who are most threatened and who ensure civil rights for all citizens and guarantee
may not have the political voice for social change that those civil rights are applied equally across
without such legislation. This entry presents a his- different groups of people. Much of the civil rights
torical overview of civil rights legislation and legislation in the United States is designed to pro-
related controversies, considers the impact of such tect classes of individuals who may be at greater
laws, and describes monitoring and enforcement risk of discrimination or harm. Currently pro-
efforts. tected groups at the federal level include those
based on race, color, ethnicity, national origin, reli-
gion, sex, age (that is, those 40 and over), disabil-
Historical Background
ity, and veteran status. This protection is meant to
Civil rights in the United States are enacted both prevent discriminatory treatment in such areas as
by the passing of laws that promote civil rights and employment, the housing market, voting access,
by the overturning of laws that threaten civil and education.
rights. These legal actions can take place in the
legislative, judicial, and executive branches of gov-
Related Controversies
ernment at both the federal and local level.
Legislative bodies can pass laws that enact greater Controversies have emerged concerning the meth-
civil rights protection or remove laws that threaten ods for ensuring civil rights protection, who should
civil rights. be protected, and what constitutes “civil rights.”
Civil Rights Legislation 85

One important controversy surrounds the issue of that are generally seen as deserving protection, and
whether civil rights legislation should be approached the term special rights used disparagingly to refer
in a “color-blind” or a “color-conscious” way. A to proposed legislation to protect groups that some
color-blind approach puts the focus on protection people believe do not to deserve such protection. It
of minorities through ensuring equal treatment could be argued that hate-crimes legislation, such
regardless of minority or majority status, while a as parts of the Violent Crime Control and Law
color-conscious approach puts the focus on prefer- Enforcement Act of 1994 that impose harsher sen-
ential treatment of minorities as a means of over- tencing for crimes against individuals based on
coming more subtle or institutional discriminatory their race, religion, and so on, is a kind of civil
barriers. rights legislation designed to protect minorities
Martin Luther King, Jr. famously endorsed a from crimes directed against them because of their
color-blind society where people would “not be minority status. However, such laws have been
judged by the color of their skin,” but he did so in criticized as unfairly endowing minorities with
an era when blatant prejudice was far more com- special rights or protections that should not differ
mon than it is now. Color-blind approaches today from one group to another. Similar arguments
may be problematic, especially because they often against special rights have been made concerning
involve the denial of real barriers that minorities various instantiations of affirmative action, as well
continue to face. Striking a balance between ensur- as gay rights propositions.
ing equal treatment while helping to surmount Another point of controversy concerns legisla-
existing barriers remains a challenge in the estab- tion that restricts civil rights during times of par-
lishment of civil rights legislation. ticular danger or threat. For example, shortly after
Other controversies have emerged concerning September 11, 2001, the Patriot Act was signed
which minority groups should be protected by civil into law, lifting restrictions on electronic surveil-
rights legislation. Some groups are explicitly denied lance and monitoring on the grounds of increasing
civil rights without much social outcry, such as national security. While these laws are intended for
convicted felons who are denied the right to vote the protection of the country, they are criticized for
in some states. However, the issue of protection compromising the civil rights of those who do not
can become quite controversial with respect to pose a threat.
many groups, and civil rights legislation is often
passed in the face of considerable public resistance.
Effects of Civil Rights Legislation
For example, the Civil Rights Act and Voting
Rights Acts of the 1960s were signed into law Sociologists in the early 20th century doubted that
despite considerable outspoken opposition. More civil rights legislation would affect public opinion,
recent times have seen resistance to the legal pro- as exemplified by William Graham Sumner’s obser-
tection of certain immigrant populations or minor- vation that “stateways do not make folkways.”
ities based on sexual orientation. What makes a This statement suggests that laws do not determine
group deserving of civil rights protection is a point public opinion or cultural views, but that instead
of ongoing, often heated, debate. For example, the public opinion and cultural views determine laws.
passing of Proposition 8, an initiative on the Some would even argue that civil rights laws do
November 8, 2008, California State ballot that not change people’s attitudes toward the group
denied same-sex couples the right to marry, contin- being protected and may even have a negative effect
ues to fuel heated conflict between liberals who for those groups. For example, such legislation may
opposed the proposition and many churches that not remove prejudice, but instead may change the
supported it. form of its expression from blatant forms to more
What is a “civil right” in the first place is also a subtle forms, as has been suggested by modern rac-
point of controversy and is reflected in the lan- ism and sexism theorists. Political scientists have
guage used to discuss civil rights. For example, argued that civil rights legislation from the 1960s
distinctions have been made between “civil rights” led to various backlashes, including the Southern
and “special rights,” with the term civil rights shift from predominantly majority Democratic to
describing proposed legislation to protect groups majority Republican support. Similarly, social
86 Civil Rights Movement

dominance theorists have argued that advances in international level, organizations like Amnesty
civil rights legislation for an oppressed group are International monitor civil rights activities through-
often balanced through countermeasures that out the globe, including the United States, where
ensure the continuing balance of social power in policies such as prisoner detainment without due
favor of majority groups. Other research on “shift- process and capital punishment practices have
ing standards” warns that the protection of various been criticized.
minorities may lead to the application of lower
standards for evaluating their performance, which P. J. Henry
can also have harmful consequences. See also Affirmative Action; Civil Rights Movement;
Nevertheless, positive shifts in American atti- Desegregation; Discrimination; Diversity;
tudes toward Blacks since the passage of civil Institutionalized Bias; Justice; Modern Forms of
rights laws in 1960s have been documented, mov- Prejudice
ing from majority endorsement of segregation and
beliefs in the inferior ability of Blacks to a majority
electing its first Black president in Barack Obama. Further Readings
Attitudes toward women also shifted toward
Crosby, F. J. (2004). Affirmative action is dead; long live
greater equality following civil rights legislation affirmative action. New Haven, CT: Yale University
protecting women. Press.
Further questions remain concerning the impact Faludi, S. (1992). Backlash: The undeclared war against
of civil rights legislation on the attitudes and American women. New York: Doubleday.
behaviors of the protected groups themselves. Katz, P. A., & Taylor, D. A. (Eds.). (1988). Eliminating
Does such legislation offer psychological relief to racism: Profiles in controversy. New York: Plenum Press.
those it protects? Or does it introduce new psycho- Schofield, J. W. (1986). Causes and consequences of the
logical challenges, such as the reinforcement of the colorblind perspective. In J. F. Dovidio & S. L.
stereotype that minority groups inherently need Gaertner (Eds.), Prejudice, discrimination, and racism
protection? These questions are being increasingly (pp. 231–253). San Diego, CA: Academic Press.
researched and debated in the social sciences. Schuman, H., Steeh, C., Bobo, L., & Krysan, M. (1997).
Racial attitudes in America: Trends and
interpretations. Cambridge, MA: Harvard University
Monitoring Activities Press.
Once civil rights legislation is enacted, it needs to Sumner, W. G. (1906). Folkways: A study of the
be monitored and enforced to be effective. The sociological importance of usages, manners, customs,
passage of legislative acts, judicial rulings, and mores, and morals. Boston: Ginn.
executive orders does not guarantee that civil
rights will be established and upheld. Consequently,
various public and private organizations have
emerged in the United States to help monitor and Civil Rights Movement
enforce civil rights legislation. At the federal level,
the 1957 Civil Rights Act established the United The term civil rights movement refers to the activ-
States Commission on Civil Rights to investigate ist efforts of Black Americans and their allies dur-
civil rights violations, and the Equal Employment ing a particular historical period (1955–1968) to
Opportunity Commission was created in 1961 claim certain basic civil rights previously withheld
to monitor employment discrimination in the pri- from Blacks and to end legalized segregation.
vate sector. These efforts were designed to overturn laws and
Private organizations, too, have emerged to customs of racial segregation, racialized disen-
monitor enforcement of civil rights laws, such as franchisement, and violence against Blacks. Thus
the American Civil Liberties Union and the the civil rights movement represents one of the
Southern Poverty Law Center, legal associations most comprehensive and concerted efforts by U.S.
committed to monitoring and prosecuting civil citizens to bring about social changes that would
rights abuses within the purview of the law. At the both directly improve the lives of Blacks and
Civil Rights Movement 87

expand intergroup contact and facilitate the devel- desegregation orders in places such as Little Rock,
opment of improved intergroup relations. Arkansas, as well as at the University of Alabama
From a broader perspective, the struggle for and the University of Mississippi, and many cities
civil rights in the United States did not begin or in the South chose to shut down their public school
end with the events of this period. A more thor- system for a year or longer rather than integrate
ough examination of the civil rights movement the schools. Years later, Thurgood Marshall, the
among Blacks would take into account a history of lead attorney who brought Brown v. Board of
Blacks’ efforts to secure civil rights from the Education to the Supreme Court, quipped: “Now
moment of their arrival in this country as slaves. It I know what ‘deliberate speed’ means—it means
would also include modern-day efforts to secure ‘very slowly.’” Impatience with the slow progress
equity in education, housing, health care, and all of the use of legislation and the courts to effect
areas of economic life. Moreover, civil rights change led directly to the civil rights movement.
efforts by other groups include the women’s move-
ment, the Chicano movement, the Native American
Historical Highlights
movement, and the gay liberation movement, and
civil rights efforts continue to this day and through- The civil rights movement was distinguished by a
out the world. This entry focuses first on the nar- shift away from the use of litigation as the prime
rower meaning of the term civil rights, looking at strategy for winning new rights toward a focus on
the history and impact of the efforts to claim Black the use of direct action—civil disobedience, non-
Americans’ civil rights, and then touches briefly on violent resistance, and mass mobilization—to effect
related efforts. social change. Rather than seeking support pri-
marily from legislators and judges, civil rights
activists sought broader support for the cause and
Background and Context
demanded the enforcement of laws that already
After the Civil War, many states passed a number of existed.
racially discriminatory laws, and racial violence
against Blacks was both brutal and widespread. In
The Montgomery Bus Boycott
1896, the Supreme Court held in Plessy v. Ferguson
that legalized segregation was constitutional as Trained in nonviolent civil disobedience, Rosa
long as separate but equal public facilities (e.g., Parks was arrested on December 1, 1955, for
schools) were provided to Blacks. Organizations refusing to give up her seat on the bus in
such as the Ku Klux Klan, which engaged in con- Montgomery, Alabama, to a White male passen-
certed and organized acts of violence against Blacks ger. In response, the Women’s Political Council and
that included murder, flourished in the early part of the NAACP organized a bus boycott in Montgomery
the 20th century. In this context, civil rights efforts to protest both Rosa Parks’s arrest and segregation
on the part of Blacks and other groups were focused and discrimination in the bus system. The boycott
primarily on legal efforts to overturn racially dis- was tremendously successful, and the Montgomery
criminatory laws and congressional lobbying to Improvement Association was formed to continue
secure legislative assistance at the federal level. the boycott. Dr. Martin Luther King, Jr. was
These efforts made some progress, culminating in elected to lead the Montgomery Improvement
the 1954 Brown v. Board of Education decision in Association, which sought broad change in
which the U.S. Supreme Court overturned Plessy v. Montgomery, including but not limited to its bus
Ferguson and held that separate schools for White system. The bus boycott lasted for a year, culmi-
and Black Americans could never be equal and thus nating in 1956 in a Supreme Court ruling in
were not constitutionally permissible. Browder v. Gayle that outlawed segregation in
Although the Court ordered the desegregation public buses.
of all public schools “with all deliberate speed,” King and other leaders of the Montgomery
school desegregation efforts at the elementary, sec- Improvement Association joined the group to other
ondary, and college levels were difficult to enforce. civil rights organizations in other areas of the
Federal military personnel were required to enforce South, leading to similar boycotts and eventually to
88 Civil Rights Movement

the formation of the Southern Christian Leadership House to push for passage of civil rights legisla-
Conference in January 1957. King’s leadership of tion. Kennedy agreed to support the legislation,
the Southern Christian Leadership Conference and but it was not clear he had the votes to pass it. The
the Montgomery Improvement Association, the bills that comprised this legislation—the Civil
success of the efforts in Montgomery, and the Rights Act (1964), Voting Rights Act (1965), and
expansion of those efforts across the U.S. South Fair Housing Act (1968)—were passed not during
made King a national figure. the Kennedy Administration, however, but during
During the early 1960s, organizations such as the succeeding Johnson Administration. The Civil
the Southern Christian Leadership Conference, the Rights Act of 1964 prohibited discrimination on
Congress of Racial Equality, and the Student the basis of race or sex in schools, public facilities,
Nonviolent Coordinating Committee engaged in government, and employment, effectively ending
broad efforts to end desegregation laws. Sit-ins Jim Crow segregation in the South. Initially fairly
were used effectively in many areas to protest and weak in enforcement capabilities, the law was
challenge laws enforcing racial segregation in pub- strengthened in subsequent Civil Rights Acts (e.g.,
lic facilities. In 1964, four organizations together Civil Rights Acts of 1968 and 1991).
(the NAACP, Congress of Racial Equality, Southern The Voting Rights Act prohibited literacy tests
Christian Leadership Conference, and Student to register to vote and provided federal oversight
Nonviolent Coordinating Committee) engaged in of voting registration in areas where there had
a broad voting rights campaign by bringing college been evidence of discrimination in voter registra-
students from around the country to Mississippi to tion. The Civil Rights Act of 1968 is commonly
register voters. referred to as the Fair Housing Act, in that it pro-
This activism was met with strong resistance. vided clearer enforcement provisions against dis-
Civil rights activists were subjected to beatings and crimination in the sale, rental, and financing of
brutal treatment and arrest by police, headquarters housing.
and meeting sites were bombed, and individual
activists were murdered. The strength of this resis-
Aftermath and Ongoing Struggles
tance played a role in the success of the civil rights
movement. Television images of police attacking The Civil Rights Act of 1964, the Voting Rights Act
nonviolent demonstrators with weapons, fire of 1965, the Fair Housing Act of 1968, and
hoses, and attack dogs, as well as images of gov- Executive Order 11246 of 1965, which ordered the
ernment officials attempting to bar Black students use of affirmative action to ensure the lack of dis-
from schools and colleges, played an important crimination in employment and federal contract-
role in changing broader public opinion. ing, gave many the sense that the issues for which
the movement had fought most strongly—voting
rights, antidiscrimination laws in employment,
The 1963 March on Washington
housing, and education, and increased attention to
In August 1963, a number of civil rights groups and therefore consequences for violent acts against
and leaders collaborated on the March on Blacks—had been largely resolved. Optimism was
Washington that took place at the National Mall high that the changed legal and social climate
in Washington, D.C. The march was organized to would lead to effective changes in the lives of
push for greater civil rights protections, including Blacks. However, 1968 was also significant because
greater legal protections in the South, fair housing it was the year in which both Martin Luther King,
and employment, a federal employment program, Jr. and Robert F. Kennedy were assassinated,
and voting rights. With more than 200,000 dem- thereby depriving the movement of its most visible
onstrators participating, it was a tremendous suc- leader and of an important political ally.
cess and widely televised. King delivered his most The optimism of the 1960s civil rights move-
famous speech, “I Have a Dream,” at the Lincoln ment victories gave way to the slow erosion of the
Memorial during this demonstration. hard-won gains over the coming decades. School
After the march, King and other leaders met desegregation efforts slowly dissolved beginning in
with President John F. Kennedy in the White the 1980s, as federal courts released pressure on
Cliques 89

districts to continue those efforts. Schools have Transnationally, social change efforts, and par-
slowly resegregated in many places, and today ticularly the use of nonviolent protest and direct
schools in many areas are as segregated as they action to secure civil rights, have been modeled on
were before Brown v. Board of Education. In the civil rights movement. Examples of such move-
2007, a more conservative Supreme Court over- ments include the successful efforts to end apart-
turned school desegregation efforts in Seattle and heid in South Africa, the efforts of students and
Louisville in Parents Involved in Community other activists to secure human rights in China and
Schools v. Seattle School District No. 1. In their Myanmar, and movements to secure human rights
decision, the justices effectively turned Brown v. and create economic change in the Soviet Union
Board on its head and argued that taking race into and the eastern European nations in the late 20th
account in assigning students to schools was century.
Affirmative action policies designed to improve Sabrina Zirkel
the access of underrepresented groups to colleges, See also Affirmative Action; Civil Rights Legislation;
employment, and federal contracting have also lost Collective Movements and Protest; Desegregation;
favor among many people, and recent legislative Intergroup Contact Theory; Racism
efforts and court rulings have resulted in many
affirmative action policies becoming illegal in sev-
eral states. Further Readings
The civil rights movement, as a formal, national
Browder v. Gayle, 142 F. Supp. 707 (M.D. Ala. 1956).
struggle, may have ended in the late 1960s, but
Brown v. Board of Education of Topeka, 347 U.S. 483
many groups have continued and expanded civil
rights efforts to the present day, both in the United Carson, C. (Ed.). (2001). The autobiography of Martin
States and throughout the world. In the United Luther King, Jr. New York: Warner.
States, efforts continue to create racial equity in Kluger, R. (1975). Simple justice: The history of Brown v.
health care, employment, primary and secondary Board of Education and Black America’s struggle for
education, and access to higher education. equality. New York: Random House.
Parents Involved in Community Schools v. Seattle School
Related Efforts District No. 1, 551 U.S. 701 (2007).
Plessy v. Ferguson, 163 U.S. 537 (1896).
Moreover, other movements have persisted in pur- Tushnet, M. V. (2001). Thurgood Marshall: His speeches,
suing civil rights for other groups. Examples writings, arguments, opinions, and reminisces.
include efforts to pursue employment rights by the Chicago: Lawrence Hill.
United Farm Workers of America, and recent dem- Williams, J. (1987). Eyes on the prize: America’s civil
onstrations of support for undocumented workers rights years, 1954–1965. New York: Penguin.
in the United States.
The gay rights movement has been one of the
longest standing civil rights movements in the
United States. The movement began in New York Cliques
with the Stonewall Rebellion, a series of rebellions
that emerged from frustration over police raids on A clique is a small, exclusive, tightly knit group
gay bars to arrest gay men for illegal sexual activ- of people. Membership in such groups usually
ity. Over a period of decades, these efforts have depends on social status and can have negative
succeeded in decriminalizing homosexual sexual connotations (e.g., “Goths,” “geeks,” or
activity across the United States. Efforts to secure “nerds”). However, the term need not be nega-
the right to same-sex marriage have been stronger tive; for instance, clique is used to refer to close
in recent years, and recent court cases have been groups of friends like those commonly seen in
examining antigay marriage statutes across the high schools, organizations, and neighborhoods.
county. One of these cases will likely go to the This entry looks at some of the research related
Supreme Court in the coming years. to this phenomenon.
90 Cliques

Cliques as Social Hierarchies Related Constructs

In early research, Warner argued that cliques are Among adults and adolescents, cliques are charac-
intimate informal groups of friends that represent terized by high levels of conformity. Research sug-
a triumph of class over democratic values in the gests that cliques exemplify homogeneity in
American school system. According to this per- academic performance, showing remarkable con-
spective, cliques are part of the social structure sistency in academic achievement, substance use,
that prevent people of lower social status from and aggression. These findings seem to reaffirm
socializing with those of higher social status. At the popular conception that cliques are breeding
the same time, the clique functions to include grounds of peer pressure and conformity. However,
members of the higher class with others of their research suggests that conformity can also result
kind. In other words, the clique system ordinarily from selection processes, with similar individuals
helps reward those who are higher in class and choosing to associate with one another.
punish those of a lower class. High levels of conformity go hand in hand with
Subsequent research has focused on the forma- the tendency for cliques to be highly cohesive
tion of cliques among children and in organiza- groups. Cohesive groups tend to exert more social
tions, and on the impact of cliques within larger influence, and their members are typically more
social structures on individuals’ satisfaction with committed to the group. There is an inverse cor-
those structures. In development, adolescents from relation between group size and cohesiveness. As
all social classes form cliques. However, students more and more members join a group, it becomes
from middle-class backgrounds tend to switch more difficult to maintain cohesion. This is consis-
friendship groups with shifting interests, whereas tent with the characterization of cliques as typi-
students from working-class backgrounds place cally small groups.
more emphasis on loyalty and stability.
Furthermore, perhaps surprisingly, while posi- Richard J. Crisp and Angela T. Maitner
tive perceptions of self and other are important
determinants of interaction, personality character- See also Conformity; Group Cohesiveness
istics are unrelated to clique formation. In con-
trast, IQ, social class, and how favorably children
are perceived by their teachers predict clique for-
mation in classroom settings. Further Readings
Adler, P. A., & Adler, P. (1995). Dynamics of inclusion
and exclusion in preadolescent cliques. Social
Gender Differences Psychology Quarterly, 58, 145–162.
Brown, B. B., & Klute, C. (2003). Friendships, cliques,
Cliques may have different meanings and func-
and crowds. In M. D. Berzonsky & G. R. Adams
tions for adolescent boys and girls. In adolescence,
(Eds.), Blackwell handbook of adolescence
girls appear more interpersonally competent and
(pp. 330–348). Malden, MA: Blackwell.
are more concerned with intimacy and exclusivity
Henrich, C. C., Kuperminc, G. P., Sack, A., Blatt, S. J., &
in their friendships than are boys. These differ- Leadbeater, B. J. (2000). Characteristics and
ences are reflected in friendship patterns, with homogeneity of early adolescent friendship groups:
preadolescent girls tending to form exclusive dyads A comparison of male and female clique and
or triads, and boys forming larger, more loosely nonclique members. Applied Developmental Science,
knit groups. Being in a clique can have negative 4, 15–26.
effects on girls’ self-esteem, with the clique encour- Moreland, R. L., & Levine, J. M. (2001). Socialization in
aging expressions of jealousy and competition. In organizations and work groups. In M. E. Turner (Ed.),
contrast, there can also sometimes be positive Groups at work: Theory and research (pp. 69–112).
effects of clique membership for girls, with recent Mahwah, NJ: Lawrence Erlbaum.
research by Henrich and his colleagues reporting Warner, W. L. (1942). Educative effects of social status.
its helpful effects on peer relationships and school Environment and education. Chicago: University of
adjustment. Chicago.
Coalitions 91

only obtain the positive outcome if they join

Coalitions forces. For example, in a three-person group, the
members A, B, and C can be assigned two, three,
A coalition is defined as two or more individuals and four votes, respectively. They may then learn
or groups who act jointly to affect their own and that $100 will go to the members of any coalition
others’ outcomes. Coalitions, which can be found that controls at least five votes.
at every level of human organization, arise in situ-
ations where people are in conflict over the alloca-
tion of scarce outcomes but need each other to
Static Theories of Coalition Formation
reach an agreement. This entry begins by describ- The resources that people control are important in
ing the game approaches used to study coalition two distinct ways. First of all, the resources have a
formation, and then examines some of the predom- strategic function in that they determine people’s
inant theories used to explain coalition-building power position or bargaining strength. For exam-
behavior. ple, people’s resources may determine how many
alternatives they have to form a coalition. In gen-
eral, people or parties with many resources will
Coalition Games
have more opportunities to form a winning coali-
To understand coalition formation as a group pro- tion. Second, the distribution of resources also has
cess, social psychologists have adopted an approach a normative function in that they determine how
that has its roots in game theory. In this approach, parties prefer the outcomes of the coalition to be
parties (individuals or groups) are called players, distributed. For example, the minimum resource
and the format in which these players negotiate theory assumes that the coalition members will
about forming a coalition is called a coalition want to distribute the coalition outcomes in pro-
game. Moreover, the outcome that is obtained portion to their initial resources, such that mem-
when forming a coalition is usually a quantitative bers with twice as many resources should obtain
measure such as money or points. The two main twice as much from the coalition outcomes.
questions that are addressed in this research are The essence of most theories of coalition forma-
“Which players will be included in the coalition?” tion is that the selection of coalition partners and
and “How will they distribute the outcomes of the the distribution of the coalition outcomes across
coalition?” its members are strongly related. The preferred
Coalition games have been categorized in a distribution of coalition outcomes thus determines
number of ways. One distinction between coali- who people prefer as their coalition partner.
tion games is whether all coalitions that are Combined with the assumption that people want
allowed to form also yield the same reward or to maximize their own outcomes, minimum
whether each possible coalition yields a different resource theory for example predicts that people
reward. The first type of game is called a simple want to be part of a coalition that contains as few
game. The second type of game is called a multi- resources as possible. In the example above, with
valued game. Another distinction between coali- five votes needed for a winning coalition, this
tion games is whether the individual parties are means that the members with two and three votes
assigned resources or not. These resources are will team up because this would yield each of them
comparable to votes in a political convention. a higher share of the coalition reward than would
Resources are assigned in most simple games but teaming up with the member controlling five votes.
are not assigned in most multivalued games. The members of the coalition with two and three
The coalition games most often investigated by votes would obtain 40% and 60%, respectively.
social psychologists are simple games in which When teaming up with the member controlling
three or more people are each assigned a number four votes, their shares would be lower—
of resources (e.g., votes), and learn that they need 33% (2/6) and 43% (3/7), respectively. Controlling
a certain total number of resources to obtain a many resources may thus not always be an advan-
positive outcome. Because this number exceeds the tage, a finding that is also referred to as the
number of resources of any individual, they can strength-is-weakness effect.
92 Coalitions

Whereas minimum resource theory assumes follow. To solve the disagreement, members are
that people take the distribution of resources as a expected to subsequently meet each other half-
basis for the distribution, other theories stress the way. Bargaining theory thus sees coalition forma-
importance of alternative dimensions. Minimum tion as a dynamic bargaining process that starts
power theory assumes that people primarily focus off with disagreement, and after bargaining, results
on the power dimension (i.e., on whether members in settling and agreement. When members outside
are really needed to form a winning coalition). A a coalition make a competing offer to tempt the
person’s pivotal power is defined by determining coalition members to leave it and form a new
how many winning coalitions would turn into los- coalition, the situation becomes even more
ing coalitions if the member withdrew. This theory dynamic, and new distributions will emerge
assumes that members want to distribute coalition during the process.
outcomes in proportion to the power of their posi- A final theory to consider here is equal excess
tions. Based on reasoning similar to minimum theory, which places greater emphasis on bargain-
resource theory, this leads to the prediction that ing strength. This theory assumes that the distribu-
people want to be part of a coalition that is mini- tion that members of a coalition initially demand
mal in terms of total pivotal power. is determined by what they can reasonably expect
to receive in their best alternative coalition. The
excess that the current coalition can offer over this
Dynamic Theories of Coalition Formation
best alternative will then be divided equally.
Static theories assume that members agree on the Suppose, for example, that two members, A and B,
basis for distribution. This gives the impression bargain over how to distribute the coalition out-
that coalition formation is a process where mem- comes of $100, while knowing that the best alter-
bers take the dimension for distribution as a given native for A pays $50, and for B pays only $30.
and simply calculate which coalition will then Because the best alternatives of A and B add up
yield them the highest outcomes. Dynamic theories to $80, the total excess in this case would be
of coalition formation are more process oriented $20 ($100 – $80). The expected distribution after
and acknowledge that members may disagree an equal split of the excess would then be
about the dimension that should be used to distrib- $60 for A ($50 + $10) and $40 for B ($30 + $10).
ute coalition outcomes. By assuming that people Similar to the process of counteroffers that underlies
are mainly self-interested and that their primary bargaining theory, other members may subsequently
aim is to maximize own outcomes, dynamic theo- tempt the coalition members into forming a new
ries of coalition formation resemble static theories. coalition, which implies that standards for distri-
The main difference between these theories is that bution will change during the bargaining process.
dynamic theories also assume that self-interest col-
ors the selection of dimensions for distribution.
Empirical Support and
Bargaining theory assumes two possible distri-
Theoretical Limitations
bution rules; the proportionality rule (that also
forms the basis of minimum resource theory) and The emphasis of coalition research has tradition-
equal distribution. According to this latter rule, the ally been to determine which of the proposed theo-
outcomes of the coalition should be distributed ries most accurately provides the answers to the
equally to all members of the coalition. Self-interest questions of who will be the members of the win-
may shape the preference for distribution rules. ning coalition and how the coalition outcomes will
Group members with many resources will prefer to be distributed over the members. In general, the
distribute outcomes in proportion to the resources dynamic theories have obtained more empirical
each member possesses. Such a rule would be to support than the static theories. Situational char-
the disadvantage of members with few resources; acteristics and personal characteristics may, how-
they would obtain higher outcomes if the coalition ever, strongly affect how people bargain. Such
outcomes were distributed equally. factors may therefore affect the predictive accu-
Bargaining theory states that self-interested racy of the theories. For example, experienced
members may initially disagree on which rule to bargainers seem less likely than inexperienced
Cognitive Consistency 93

bargainers to distribute the coalition outcomes See also Cooperation and Competition; Distributive
equally, and women appear to show a greater pref- Justice; Group Formation; Mergers; Need for
erence for equality than do men. Belonging; Ostracism; Power
A related and more general issue concerns the
primary motivation of bargainers. The theories cited
Further Readings
above—whether static or dynamic—all agree on
one thing: The primary motivation is self-interest. Gamson, W. A. (1964). Experimental studies of coalition
Parties are first and foremost assumed to strive to formation. In L. Berkowitz (Ed.), Advances in
maximize their own outcomes, and this in the end experimental social psychology (Vol. 1, pp. 81–110).
determines which coalitions will be formed and New York: Academic Press.
how the coalition outcomes will be distributed to Kahan, J. P., & Rapoport, A. (1984). Theories of
the coalition members. This focus on self-interest coalition formation. Hillsdale, NJ: Lawrence Erlbaum.
fits with the game-theoretic foundations of these Komorita, S. S. (1984). Coalition bargaining. In
theories and research on coalition formation. L. Berkowitz (Ed.), Advances in experimental social
However, social psychology also acknowledges psychology (Vol. 18, pp. 183–245). New York:
motives other than self-interest. Academic Press.
As social beings, we also care for the other Murnighan, J. K. (1978). Models of coalition behavior:
Game theoretic, social psychological, and political
people’s outcomes. The social utility model of
perspectives. Psychological Bulletin, 85, 1130–1153.
coalition formation has formalized this broader
Thibaut, J., & Kelley, H. H. (1959). The social
perspective on human motivation by distinguish-
psychology of groups. New York: John Wiley.
ing fairness as a separate motivator. The basic
Van Beest, I., & Van Dijk, E. (2007). Self-interest and
tenet of the model is that people assign a positive fairness in coalition formation: A social utility
value to both self-interest and fairness, but that approach to understanding partner selection and
situational and personal characteristics may affect payoff allocations in groups. European Review of
the weights of each of these motives. With this Social Psychology, 18, 132–174.
model, it is now also possible to explain why
sometimes people want to include others in a coali-
tion even if the coalition would be winning only if
these others were left out. Self-interest alone could
not explain such behavior, because coalition mem-
Cognitive Consistency
bers obtain higher outcomes if they share the coali-
tion payoff among few members rather than Cognitive consistency theory encompasses a broad
among many members. People may, however, con- group of theoretical statements whose central core
sider it to be unfair if others are left out. This view is that people prefer consonance among their cog-
accords with the notion that people are reluctant nitions, thoughts, feelings, and behaviors. These
to exclude and ostracize others. People with a theories seek to explain individuals’ discomfort
more prosocial motivation appear especially sensi- with inconsistency in their social lives. Although
tive to what happens to those who are excluded. theories of cognitive consistency initially focused
In a similar vein, research has shown that it on individuals as the unit of analysis, research has
matters whether outcomes are positive or negative. recently shown that cognitive consistency is a cen-
Whereas most research on coalition formation has tral component of intergroup and intragroup rela-
focused on positive outcomes, some research has tionships as well. This entry looks at a precursor
studied how coalitions form when the winning theory and then examines the development of cog-
coalition can allocate a negative outcome and so nitive consistency theory and its application to
determine which parties have to pay. Because both groups and intergroup processes.
people assign a greater weight to fairness when
outcomes are negative, people are more likely to
Balance Theory
include all parties in the coalition agreement.
Most influential of the early cognitive consistency
Eric van Dijk and Ilja van Beest models was Fritz Heider’s balance theory. The
94 Cognitive Consistency

principle of psychological balance can be applied to or counterproductive (the opposite of their actual
any set of cognitions, but it is described most easily beliefs), the action of passing by without giving
as a set of relations between a reference person (P), would have logically followed. As things are, the
another person, perhaps a friend (O), and an atti- action is dissonant.
tude object (X). These relations were said to be Holding two or more inconsistent cognitions
balanced when they were consistent. For example, arouses the state of cognitive dissonance, which is
if P likes O and both P and O like the object X, the experienced as uncomfortable tension. This ten-
cognitions are said to be consistent or balanced. sion has drive-like properties and must be reduced.
However, if person P likes O but they differ in One of the major innovations of dissonance theory
their evaluation of X, then the relationships are compared to prior models of consistency such as
imbalanced or inconsistent. The perception of cog- balance theory, however, is that it speaks in terms
nitive inconsistency produces a strain toward bal- of magnitude. Most dissonance research does not
ance in which P attempts to restore consistency by compare how people respond to consonant and
changing his or her attitude toward the object X or dissonant relationships among cognitions, but
toward the friend O. The steak lover who is friends rather how they respond to cognitions that are dis-
with the vegetarian experiences tension whenever sonant to varying degrees.
their food preferences are discussed; as much as Perhaps the most famous study in dissonance
these two people like each other, there is tension research makes this point very well. Festinger and
over their food preferences and a drive to establish his colleague, J. Merrill Carlsmith, asked research
balance. participants to perform a series of boring tasks:
Much foundational research was conducted turning pegs clockwise, then turning them counter-
under the rubric of balance theory. Balanced rela- clockwise, taking the pegs out of the peg board,
tionships are viewed more positively than imbal- putting the pegs back into the board. After an
anced ones, imbalanced relationships cause more extended period, the experimenter thanked the par-
tension than balanced ones, and there is a desire to ticipants, telling them that the task was completed.
bring imbalanced relations into balance. But bal- The study was about expectations and perfor-
ance theory had very little to say about the charac- mance, he said, and some participants had been
teristics of P, O, and X that would make imbalance told that the task was interesting and exciting,
most troubling or about the methods people would while others had been given no expectation. This
use to attempt to restore balance to dissonant rela- explanation of the study’s purpose was false, but it
tionships. It was in part to address this void that, was a necessary component of the real experiment.
in the late 1950s, there arose a new theory: cogni- The experimenter then said that the research assis-
tive dissonance. tant who was supposed to tell the next participant
that this dull, boring task was actually interesting
and exciting had failed to arrive. Would the par-
Cognitive Dissonance
ticipant be willing to help out and take the assis-
In 1957, Leon Festinger published A Theory of tant’s place?
Cognitive Dissonance. In some ways, dissonance The participant knew that the task was actually
theory is similar to balance theory. The state of boring and dull and that it was generally wrong to
cognitive dissonance occurs when people believe lie to people to raise their hopes. However, the
that two of their psychological representations are experimenter was doing important work and was
inconsistent with each other. Put another way, a even offering to pay the participant for this rela-
pair of cognitions is inconsistent if one cognition tively trivial task. Almost all participants agreed to
follows from the obverse of the other. For example, lie to the waiting stranger (actually a research
some people believe that they should give money assistant), and they then rated the pleasantness of
to the poor but refuse to give change to someone the peg-turning task. The researchers were inter-
who asks for it. These two cognitions are disso- ested in how participants who were highly paid to
nant because not giving alms follows from the lie rated the study compared to those who were
obverse of these people’s belief. Had they believed poorly paid. They found that those paid $20 to tell
that giving money to the poor was either wasteful the lie rated the study as having been more boring
Cognitive Consistency 95

than those only paid $1. This counterintuitive discussion group (which was intentionally made as
finding was well explained by dissonance theory: boring as possible) was in fact interesting.
Those paid $1 had less justification for lying to After having been made to suffer so intensely,
the waiting stranger and thus more need to dis- participants who had read lurid words to this
tort their impressions of the task to match their stranger needed to justify their effort by claiming
attitude-inconsistent statement. that the experience had been worth it. Their dislike
This experimental paradigm was later refined of the discussion group was in conflict with the
and simplified into an “induced compliance” effort they had expended to enter it, and they
method. In this method, participants are persuaded resolved this dissonance by changing their impres-
to make a counterattitudinal statement under con- sion of the group to be more consistent with their
ditions in which (1) they believe they are free to effort. They came to like the discussion better and
refuse (i.e., there are no prohibitively large induce- to like the group members better than those who
ments or punishments for refusal) and (2) the had not suffered to join the group. This work is
statement, if made, will have some aversive conse- relevant to hazing and initiation rituals, as making
quences. A classic example is advocating a noxious a group difficult to join apparently increases that
political position in an essay that will be shown to group’s attraction.
impressionable high school students. Compared to Other studies have shown that groups can alter
those who either are not free to refuse or believe the magnitude of cognitive dissonance and the
their actions will have no consequences or lesser direction that its resolution takes. One study
consequences, those who act under these condi- showed that people in groups can diffuse responsi-
tions come to agree more with their counterattitu- bility for their inconsistent actions and experience
dinal position. less dissonance. Individuals who were engaged in a
group project to write an essay inconsistent with
their attitude apparently convinced themselves that
Application to Groups
their own portion of the inconsistency was less
Shortly after the pioneering study described above, than that of the group, and therefore, they experi-
Elliot Aronson and Judson Mills applied disso- enced less discomfort as a result of the group proj-
nance theory to aspects of a social group. They ect. Another study in a group context showed that
tested whether an unlikable group would become people who act in a manner inconsistent with a
better liked if it had been difficult to join. They fundamental attitude of their group respond to
recruited female participants for a “sexual discus- inconsistency not in the usual manner of changing
sion group.” After an initial screening, participants their attitudes toward the issues, but rather by
were supposedly connected via speakers and micro- derogating the outgroup. For example, members
phones to a discussion of sexual behavior. In real- of the Republican Party in the United States who
ity, participants were made to listen to a tape of wrote an attitude-inconsistent essay advocating
people having a boring conversation about the the Democratic candidate for president resolved
secondary sexual characteristics of various insects. their inconsistency by derogating the outgroup—
Participants then rated how enjoyable the conver- that is, Democrats.
sation had been.
As in the peg-turning study, the key variable of
Intergroup Processes
interest was something unexpected. Before allow-
ing participants to listen to the discussion, the A more recent expansion of cognitive inconsistency
experimenters explained that many people found research has taken it in a direction relevant to inter-
discussions of sexual topics disturbing, and to show group processes. Working from both cognitive dis-
that they were comfortable discussing such topics, sonance theory and social identity theory—which
participants would have to read a list of words. For describes when and why individuals identify with,
some participants, the list was fairly mundane, and act as members of, social groups—researchers
for others, lurid and obscene. Those participants have begun to investigate what is termed vicarious
given the more extreme screening task were more dissonance. They have found that, in addition to
likely than the others to report that the subsequent experiencing personal cognitive dissonance when
96 Collaboration Technology

we make our own choices or act in ways that are Cognitive consistency research began in an indi-
counterattitudinal, we may also experience disso- vidual context, with occasional research studies
nance vicariously whenever members of our social examining people as members of social groups.
groups make choices or act in ways that are incon- This research on individuals focused on their suf-
sistent with their attitudes. Because we are in many fering from imbalanced relationships or dissonant
social groups, the opportunities to experience dis- cognitions as a function of mental representations
sonance on behalf of fellow group members are inside their heads. The current interest in social
numerous. The conceptual proposition that is at identity and cultural psychology, however, has led
the core of vicarious cognitive dissonance is that to an explosion of research examining consistency
dissonance brought about by the actions of a pro- from the perspective of the embeddedness of peo-
totypical member of a social group will lead us, as ple in their cultures and within the social groups
fellow group members, to change our attitudes, that comprise those cultures.
even though we have no direct responsibility for
the dissonance-inducing behavior. Joel Cooper and Matthew B. Kugler
In a typical vicarious dissonance study, a person See also Collective Self; Culture; Festinger, Leon; Social
is induced to evaluate speeches made by other Identity Theory
group members, typically members of the same
group or school as the participant. The participant
overhears the group member agree to make an Further Readings
attitude-inconsistent statement under conditions
that classically evoke dissonance. If the speech- Cooper, J. (2007). Cognitive dissonance: Fifty years of a
writer and the actual participant share a common classic theory. Thousand Oaks, CA: Sage.
group identity, then the participant too experiences Cooper, J., & Hogg, M. A. (2007). Feeling the anguish of
others: A theory of vicarious dissonance. In M. P.
attitude change, despite having done nothing that
Zanna (Ed.), Advances in experimental social
should provoke dissonance. Moreover, the magni-
psychology (Vol. 39, pp. 359–403). San Diego, CA:
tude of the dissonance is a function of the strength
Academic Press.
of the social identification that participants feel
Festinger, L. (1957). A theory of cognitive dissonance.
with their social group. Palo Alto, CA: Stanford University Press.
This finding has broad implications for inter-
group processes. For example, in cultures that
emphasize group harmony and cohesiveness, the
experience of vicarious dissonance should be quite
high. One recent study in that area found that East Collaboration Technology
Asians, who often do not show personal dissonance
effects, strongly experienced vicarious dissonance The emergence of the Internet over the past 40
when a fellow group member acted in an inconsis- years has created a rich new arena for group activ-
tent manner. This finding was the latest in a series ity. Specifically, where physical collocation was
that has shown that, while personal inconsistency is once a requirement for both group membership
dissonance provoking primarily in Western cul- and communication, computer networks now
tures, interpersonal dissonance also occurs in col- create the opportunity to form and maintain
lectivist cultures such as those of Japan and Korea. groups independent of time and space. These are
A representative study conducted with Japanese often called distributed groups. Tools for support-
and Canadians of European descent showed that ing distributed group work and play are collec-
European participants experience dissonance when tively referred to as collaboration technology.
making difficult choices for themselves as individu- Early forms of collaboration technology emerged
als but that Japanese experienced dissonance only in the 1960s and 1970s as by-products of the first
when making choices for fellow group members. computer networks, such as the ARPAnet. Over
These results suggest that group-based dissonance time, many applications (e.g., electronic mail, or
processes may be universally pervasive and therefore e-mail) designed to support remote computer
relevant to a wide range of people and situations. operations came to be valuable on their own and
Collaboration Technology 97

have become nearly ubiquitous as communication Some “being there” technologies have achieved
tools. The emergence of the Internet in the 1980s, tremendous success and have come to define
and the explosion of network use associated with entirely new genres of interaction and affiliation.
the rise of the Web in the 1990s, have accelerated Consider, for example, adolescents’ use of text
the development of collaboration technology. messaging and the proliferation of “textese.”
Today, this technology spans a wide array of appli- Other forms of “being there” technology, such as
cations and services. Some of these would be rec- data conferencing, have received broad adoption
ognizable to the builders of the ARPAnet (e.g., in certain contexts, such as within corporations to
e-mail and instant messaging), but others probably support distributed work teams. And even more
could not have been imagined by them (e.g., vir- novel “being there” technologies, such as virtual
tual worlds, such as Second Life, and social net- worlds, in some cases have attracted large numbers
working sites, such as Facebook). of users (e.g., the hundreds of thousands using
The development, adoption, and use of collabo- Second Life) who have created vibrant “inworld”
ration technology raise key questions about group economies. Despite this growth, virtual worlds
processes and intergroup relations. Specifically, in remain outside the mainstream of collaboration
traditional groups, physical proximity plays a cen- technology use.
tral role in group formation, maintenance, and
communication. In contrast, distributed groups
“Beyond Being There” Technology
mediate their activity through collaboration tech-
nology. Therefore, the success of distributed groups An equally important thrust in collaboration tech-
depends, to a large extent, on the ability of col- nology is an emphasis on asynchronous interaction.
laboration technology to allow distributed groups This allows people who are not in the same place at
to perform as well, or maybe even better than, col- the same time to engage in a collaborative or collec-
located groups. Much of the history of collabora- tive activity. Landmark instances of “beyond being
tion technology can be understood as attempts there” technology include (a) applications to col-
either to mimic the benefits of collocation (“being lect, process, and distribute user-contributed
there” technology) or to exploit certain features of content—Wikipedia is the best known example;
collaboration technology to create new benefits (b) applications to exploit, analyze, and visualize
(“beyond being there” technology). links and ties among individuals, such as the social-
networking site Facebook; and (c) applications to
process and distill patterns from collective behav-
“Being There” Technology
ior, such as the pagerank algorithm used by Google
An important thrust in collaboration technology is to sort the results of searches in terms of the fre-
an emphasis on real-time interaction that allows quency of pointers to a site. More recent instances
distributed participants to engage in activity at a of “beyond being there” technology include appli-
distance as if they were collocated. Landmark cations to aggregate small increments of human
instances of “being there” technology include attention and labor into large-scale efforts, such as
(a) applications to share common views and con- von Ahn’s “games with a purpose” (e.g., tagging all
trol of documents and drawings, now common in extant images on the Web), Amazon’s Mechanical
the form of data conferencing tools such as WebEx; Turk, NASA’s “clickworkers” (e.g., classifying cra-
(b) applications that provide greater awareness of ters from photographs of the Martian surface), and
distant group members (e.g., “busy” or “away from “crowdsourcing” (e.g., attempting to harness exper-
desk”), now popular in the form of instant messag- tise that is dispersed and difficult to locate, such as
ing tools such as MSN Messenger; and (c) applica- use of the Innocentive Web site).
tions to see and hear distant group members, such Some “beyond being there” technologies have
as videoconferencing. More recent instances of achieved tremendous success and have come to
“being there” technology include virtual worlds define entirely new modes of production and work
such as Second Life, which provide simulated geog- organization, such as Wikipedia or open source
raphies and built environments in which individu- software projects. Similarly, pagerank and Google
als interact via computer-generated avatars. have redefined how people seek information—“to
98 Collective Guilt

google” is now a recognized verb form. Other systems, such as those used by Netflix and Amazon,
forms of “beyond being there” technology, such as depends on users rating or evaluating a sufficient
crowdsourcing, have also produced notable suc- number of movies or books. Motivating these con-
cesses. For example, Mechanical Turk was used to tributions involves system designs that reduce free
accelerate the search for Jim Grey, a famous com- riding. In particular, people are more likely to make
puter scientist who was lost at sea. Grey’s col- contributions when they feel that others, especially
leagues used Mechanical Turk to get thousands of similar others, are also making contributions. In the
volunteers to prescreen tens of thousands of satel- future, the design of these “choice architectures”
lite images of open ocean searching for evidence of will be as significant to the success of “beyond being
Grey’s boat. Figuring out how to use crowdsourc- there” technologies as the design of the underlying
ing to integrate closely coupled, cognitively com- technologies and interfaces.
plex work has proved more elusive.
Thomas A. Finholt

New Directions in See also Communication Networks; Computer-Mediated

Collaboration Technology Communication; Virtual/Internet Groups

Current research suggests important directions for

the development of the next generation of collabo- Further Readings
ration technologies. In terms of “being there” tech-
Bainbridge, W. S. (2007). The scientific research potential
nology, for example, continuing increases in network
of virtual worlds. Science, 317, 472–476.
bandwidth, combined with advances in video reso-
Beenen, G., Ling, K., Wang, X., Chang, K., Frankowski,
lution, have resulted in videoconferencing systems
D., Resnick, P., & Kraut, R. E. (2004). Using social
that show life-sized views of distant group members
psychology to motivate contributions to online
with sufficient clarity to reveal subtle nuances of communities. In Proceedings of ACM CSCW 2004
communication, such as gaze direction and body Conference on Computer Supported Cooperative
language. Researchers have known for some time Work (pp. 212–221). New York: ACM Press.
that users dislike videoconferencing due to the Finholt, T. A. (2002). Collaboratories. Annual Review of
fuzziness of remote images, poor audio quality, and Information Science and Technology, 36, 73–107.
“choppiness” in video and audio transmissions. Hollan, J., & Stornetta, S. (1992). Beyond being there. In
Modern systems, such as Cisco’s Telepresence, Proceedings of ACM CHI 1992 Conference on
address these concerns by combining high defini- Human Factors in Computing Systems (pp. 119–125).
tion video, spatially located audio (i.e., voices come New York: ACM Press.
from the direction of the speaker’s image on screen), Mark, G. (2002). Extreme collaboration.
and uniform room furnishings to create the illusion Communications of the ACM, 45, 89–93.
that distant group members are all together in the Newell, A., & Sproull, R. F. (1982). Computer networks:
same space. The success of these systems may Prospects for scientists. Science, 215, 843–852.
depend as much on their technical performance as Silberman, S. (2007). Inside the high tech hunt for a
it does on successful elaboration of group proce- missing Silicon Valley legend. Wired, 15. Retrieved
dures and practices to accommodate the systems. February 5, 2009 from
Specifically, scheduled use of videoconferencing people/magazine/15–08/ff_jimgray
(e.g., formal meetings) does not create opportuni- Thaler, R. H., & Sunstein, C. R. (2008). Nudge:
ties for spontaneous encounters. By contrast, con- Improving decisions about health, wealth, and
tinuous video links between remote public spaces happiness. New Haven, CT: Yale University Press.
do allow for chance meetings, which are an impor-
tant way of coordinating work.
In terms of “beyond being there” technology,
advances in productivity and performance will require Collective Guilt
a better understanding of what motivates individual
contributions to contribution-based production Guilt is an unpleasant emotional reaction that
systems. For instance, the success of recommender occurs with the perception of having committed
Collective Guilt 99

some type of moral violation. Historically, psy- against an outgroup. This produces a perceptual
chological research on guilt has focused on the shift from thinking of oneself in terms of “I” or
feelings of guilt that arise when people feel per- “me” toward thinking of oneself in terms of “we”
sonally responsible for causing illegitimate harm or “us.” In this way, the self can be linked with
to others. Recent research has revealed that people past or present ingroup harm-doing. For instance,
can have similar feelings of guilt when their group contemporary Americans can certainly claim that
is perceived to be responsible for illegitimately they personally did not participate in slavery or the
harming members of another group. This collec- colonization of indigenous peoples. Nonetheless,
tive guilt results from sharing a social identity when contemporary Americans think about them-
with others whose actions represent a threat to the selves as part of the historical legacy, from which
positivity of that identity. Thus even though oth- they may even benefit in the present, they can
ers were responsible for the harm or moral viola- experience collective guilt based on the past actions
tion, and the individual is not directly implicated of the larger “we.”
in the harm-doing, the individual can still have The second factor that influences the extent to
feelings of collective guilt. which collective guilt is experienced is collective
A wide range of intergroup inequalities can responsibility. In order to feel collective guilt, it is
elicit these feelings, from the receipt of unearned important for people to perceive their group as
benefits or privileges that members of other groups responsible for the harm done to the outgroup.
do not receive to more extreme forms of harm- One basis for attributing responsibility to a group
doing, including genocide. Given the aversive is perceiving that group as having benefited from
nature of collective guilt, people are motivated to the harm done to the outgroup. For instance, exist-
avoid or decrease its intensity. There are several ing racial inequality can be framed in terms of the
methods for doing so; these generally involve dis- consequences that it has for outgroup members, or
torting perceptions of the ingroup’s behavior (e.g., in terms of the consequences that it has for ingroup
minimizing the extent of harm done, denying the members. Thus researchers framed racial inequal-
harmful actions entirely) or justifying its actions ity in the United States in terms of Black disadvan-
(e.g., because the victims deserved their outcomes tage or White privilege. The framing of racial
or the ingroup had legitimate reasons for causing inequality as “Black disadvantage” allowed White
the harm it inflicted). Use of all of these options participants to feel less collectively responsible for
can help to maintain a positive social identity the harm done to the outgroup, which lessened
when even the gravest of ingroup harm-doing is collective guilt. Framing racial inequality as “White
confronted. privilege” increased White participants’ feelings of
Despite its aversive nature, feeling collective collective responsibility for the harm done to the
guilt can lead to positive social outcomes, such as outgroup, leading to greater collective guilt.
reducing negative attitudes toward the harmed The third factor that influences the experience of
outgroup and promoting intergroup reconciliation collective guilt is the perception that the ingroup’s
through apologies or reparations. These benefits actions toward the outgroup were illegitimate.
are particularly likely when repairing the harm Collective guilt requires that people see their
done is perceived to be not too difficult or costly, ingroup’s actions as unjustified, immoral, or wrong
so that correcting the wrongs committed by the based on existing ingroup norms. Because it is
ingroup seems both feasible and worthwhile. threatening to conclude that one’s group has acted
unjustly, people will employ a number of strategies
that are aimed at justifying the ingroup’s actions. To
What Causes Collective Guilt?
the extent that the ingroup’s harmful actions can
Several factors influence whether, and how much, be interpreted as legitimate and reflecting noble
collective guilt is experienced in response to remind- intentions—especially those that can be construed
ers of ingroup harm-doing. First, one’s social iden- as protecting the ingroup from harm—collective
tity must be salient. For one to experience collective guilt for even the most severe harm will be lessened.
guilt, one must perceive oneself as a member of a The experience of collective guilt is not simply a
social group that has committed illegitimate harm function of empathy for those harmed. Rather, the
100 Collective Guilt

experience of collective guilt reflects the distress Israel; such claims undermine feelings of collective
that is aroused when the ingroup’s morality is guilt. The same is true of Americans. When
questioned. Two studies have directly tested Americans are reminded of the 9/11 terrorist
whether empathy for the victims or distress about attacks on the United States, they are less likely to
one’s own social identity determines the extent to feel collective guilt for subsequent harm done to
which collective guilt is experienced. In these stud- Iraqis—U.S. actions in Iraq are seen as a legitimate
ies, perceiving the ingroup (i.e., men) as responsible response to al Qaeda-sponsored terrorism. Another
for the inequality that harms the outgroup (i.e., way to legitimize the ingroup’s harmful treatment
women) was found to increase collective guilt via of outgroups is to dehumanize the victims. By por-
self-focused distress and not by empathy for the traying members of outgroups whom we harm as
outgroup. This not only discounts the view of guilt either animals or machines, we can make our
as stemming from empathy, but it also supports ingroup’s behavior appear natural and even neces-
the notion that guilt is a self-conscious emotion. sary. Such perceptions allow people to escape feeling
collective guilt for harm-doing that is inten­tionally
What Reduces Collective Guilt?
Fourth, people can focus on the benefits of the
There are numerous means by which collective guilt harm done to outgroup members, rather than the
can be undermined. For this reason, collective guilt costs. For instance, Dutch research participants
has been described as a fragile emotion. Collective who read a benevolent description of their ingroup’s
guilt can be lessened in at least four ways. historical colonization of Indonesia (e.g., “they
First, people can deny the ingroup’s harmful built roads and schools”) experienced less collec-
actions, or downplay the severity of the harm done tive guilt than those who read a less benevolent
to an outgroup. Perceiving fewer victims or even description of their ingroup’s colonization (e.g.,
fewer ingroup members as involved in the harm- “they annexed land and killed natives”). When the
doing can lessen collective guilt. harmful treatment of the outgroup is portrayed as
Second, people can deny the ingroup’s responsi- turning out positively, then people can escape col-
bility for harm done to an outgroup. For example, lective guilt for their group’s harmful actions.
men could blame women who are raped by sug- People’s feeling collective guilt may not be suf-
gesting that women somehow encourage the per- ficient to translate into their reconciliation with
petrators, whether through their appearance or the outgroup and future positive behavior toward
behavior. By blaming female victims for the harm outgroup members. In order to initiate action,
done to them, men can escape feeling any collec- people must also feel some amount of efficacy to
tive guilt for their group’s harmful treatment of bring about desired changes and believe that it is
women. Moreover, people can deny the existence possible to make up for the harm done. For
of collective responsibility, choosing instead to instance, when men were led to believe that the
claim that only individuals are responsible for their difficulty of making reparations to women for the
behavior. For instance, American soldiers who harm done to them (e.g., the economic disadvan-
served in Vietnam could deny collective responsi- tages women have suffered due to institutional-
bility for the harm done to Vietnamese civilians by ized sexism) would be very costly, collective guilt
suggesting that such harm was committed by a few was lower than when the cost was deemed to be
bad ingroup members and those alone should be more moderate and therefore potentially manage-
held responsible. When individual members of the able. Thus, when people believe they can bring
ingroup are singled out for blame, then people can about change that will result in more positive rela-
escape feeling collective guilt for the group’s harm- tions between groups, feelings of collective guilt
ful actions. are most likely to encourage reparations for past
Third, people can claim that their group’s harm.
behavior was legitimate. For instance, Jewish Although collective guilt is an aversive emotion,
research participants have reported that Israel’s it is predictive of a number of positive social con-
harm to Palestinians is justified because it is in sequences. A variety of studies have shown that the
response to Palestinian terrorist attacks against more White Americans feel collective guilt for
Collective Induction 101

racial inequality, the more they support affirmative An Experimental Procedure for Research
action programs for the harmed group. Feelings of
In studies of collective induction, group members
collective guilt also predict support for apologies
are gathered around a table and informed that
to the harmed group, as well as financial repara-
their objective is to induce a rule based on stan-
tions. Perhaps most important, feeling collective
dard playing cards with four suits (clubs, dia-
guilt for racial inequality can decrease racism.
monds, hearts, and spades) of 13 cards each. Aces
Nyla R. Branscombe and Mark A. Ferguson are given the numerical value 1, deuces 2, and so
on up to kings, which are given the value of 13.
See also Intergroup Emotions Theory; Self-Categorization The rule may be based on any of the characteristics
Theory; Social Identity Theory of the cards, such as suit (e.g., “diamonds”) or
number (e.g., “eights”), or any combination of
numerical and logical operations on suit and num-
Further Readings ber (e.g., “red queens,” “even diamonds alternate
Barkan, E. (2000). The guilt of nations: Restitution and with odd spades,” “red queens or black jacks”).
negotiating historical injustices. New York: Norton. The experimenter then places a card that is an
Branscombe, N. R. (2004). A social psychological process example of the rule (e.g., the eight of diamonds for
perspective on collective guilt. In N. R. Branscombe & the rule “diamonds”) on the table.
B. Doosje (Eds.), Collective guilt: International Each trial consists of three stages: (1) each
perspectives (pp. 320–334). New York: Cambridge group member records his or her hypothesis about
University Press. the rule, (2) the group proposes a hypothesis, and
Branscombe, N. R., & Miron, A. M. (2004). Interpreting (3) the group plays any one of the 52 cards. As
the ingroup’s negative actions toward another group: cards are played, the experimenter arranges them
Emotional reactions to appraised harm. In L. Z. on the table in a way that provides information
Tiedens & C. W. Leach (Eds.), The social life of about what the correct rule might be (with cards
emotions (pp. 314–335). New York: Cambridge that are examples of the rule placed beside one
University Press. another, and cards that are not examples placed
Wohl, M. J. A., Branscombe, N. R., & Klar, Y. (2006).
below the last card played). This procedure contin-
Collective guilt: Emotional reactions when one’s group
ues for 10 trials, at which point the group makes a
has done wrong or been wronged. European Review
final hypothesis and is informed whether it is cor-
of Social Psychology, 17, 1–37.
rect or incorrect.

Collective Induction Collective Versus Individual Induction

Groups solve these rule-learning problems better
Collective induction is the cooperative search for than the average individual (e.g., 20 four-person
rules and principles. For example, members of a groups versus 20 individuals). Beyond this, groups
scientific research team observe patterns and regu- have been compared to the best of an equivalent
larities in some domain (e.g., biology), propose a number of individuals (e.g., 20 four-person groups
theory to explain them, derive hypotheses from versus the best 20 of 80 individuals). In this
the theory, and use experiments or controlled research, the group or individual solves the prob-
observations to test the hypotheses. If the results lem from one array of hypotheses and card plays
of their research support the hypotheses, team (as described above), or from two, three, four, or
members become more confident in their theory. five arrays of hypotheses and card plays. In all
If the results fail to support the hypotheses, team cases, card plays can be viewed as “evidence”
members revise or reject their theory. This entry regarding the correct rule. Including two or more
discusses research on collective induction that arrays is useful, because this group models the
models the behavior of scientific research teams procedure of a team of scientists conducting two
and similar cooperative groups, such as auditing or more simultaneous experiments in a particular
teams, air crash investigators, and art experts. domain (e.g., synthesizing a compound).
102 Collective Induction

With one, two, three, or four arrays of hypoth- hypotheses are either plausible (consistent with the
eses and card plays, groups perform at the level of evidence to that point) or “nonplausible” (incon-
the second-best individuals. With five arrays, how- sistent with the evidence to that point). Research
ever, groups perform at the level of the best indi- indicates that groups use an interesting and orderly
viduals. This is interesting, because it is relatively process of choosing group hypotheses from mem-
uncommon for groups to perform at this level. bers’ individual hypotheses. If at least two group
Studies varying both the number of hypotheses members propose correct and/or plausible hypoth-
and the number of card plays (or evidence) demon- eses, the group selects one of them. However, if
strate that card plays are relatively more important only one member or no member proposes a correct
than hypotheses in helping groups determine the or plausible hypothesis, the group selects among
correct rule. Although group members are gener- all the proposed hypotheses (correct, plausible, or
ally able to generate and propose an adequate nonplausible). If a majority of members proposes
number of hypotheses, they need sufficient evi- the same hypothesis, the group chooses this hypoth-
dence to test and evaluate them. esis, but if there is no majority, the group takes
turns in choosing each member’s hypothesis over
Simultaneous Collective successive trials. On approximately 20% of the
and Individual Induction trials, the group proposes an emergent hypothesis
that no member has proposed, but these emergent
Many scientific research teams conduct experi-
hypotheses are rarely correct.
ments and also exchange hypotheses and/or evi-
If one group member proposes the correct
dence with independent individual scientists who
hypothesis on a trial, the final group hypothesis
are working on the same problem. For example, a
will be correct with a very high probability (.99),
team of virologists at the Centers for Disease
but if no group member proposes the correct
Control may exchange hypotheses and experimen-
hypothesis, the final group hypothesis will be cor-
tal results with independent researchers in other
rect with a very low probability (.01). Thus, groups
laboratories. In research on such simultaneous col-
are remarkably good at recognizing and adopting
lective and individual induction, a group and one
a correct hypothesis if it is proposed by a member,
or more individuals solved the same problem at the
but they are remarkably poor at forming correct
same pace in separate rooms.
emergent hypotheses that no group member has
In four conditions, the group and individual(s)
(1) exchanged hypotheses and card plays on each
trial, (2) exchanged hypotheses only, (3) exchanged
card plays only, or (4) solved the problem indepen- Conclusion
dently without exchange. Groups performed better
Research on collective induction yields several gen-
than the individuals, and both exchange of hypoth-
eral conclusions. Collective induction is superior to
eses and exchange of card plays improved group
the induction of the average individual and equal
and individual performance. Moreover, group per-
to the induction of the best individual when the
formance was improved relatively more by
group is given a large amount of information from
exchange of card plays, whereas individual perfor-
many arrays of hypotheses and card plays. Group
mance was improved relatively more by exchange
induction benefits more from increasing evidence
of hypotheses. This again shows that groups are
than from increasing hypotheses. Groups influence
able to generate sufficient numbers of hypotheses
individuals more than individuals influence groups
but need evidence to test and evaluate them.
in simultaneous collective and individual induc-
Additional analyses indicated that, across succes-
tion. Groups follow an orderly process in forming
sive pairs of trials, groups influenced individuals
hypotheses from the hypotheses of their members.
more than individuals influenced groups.
And finally, groups rarely form correct emergent
hypotheses that no member has proposed, but
The Processes of Collective Induction
groups can eventually recognize and adopt correct
After the correct rule is chosen by the experimenter hypotheses if they are proposed by a member.
and an initial example is designated, all proposed These results increase our understanding of how
Collective Movements and Protest 103

cooperative groups, such as scientific research and urbanization had brought increasing numbers
teams, auditing teams, air crash investigators, and of people together in new ways and added new
art experts, engage in collective induction. tactics and modes of organization to long familiar
forms of popular protest. The emergence of demo-
Patrick R. Laughlin cratic politics impelled the educated and well-off
to try to understand the thinking of the large num-
See also Group Performance; Group Problem Solving and
Decision Making bers of their fellow citizens who were acquiring the
right to vote, and to understand popular participa-
tion in transgressive as well as routinized forms of
Further Readings political action.
One very influential interpretation of popular
Laughlin, P. R. (1996). Group decision making and
collective action came to be known as the “collec-
collective induction. In E. Witte & J. H. Davis (Eds.),
tive behavior school” by virtue of its emphasis on
Understanding group behavior: Vol. 1. Consensual
the ways in which the actions of people in collec-
action by small groups (pp. 61–80). Mahwah, NJ:
tives seemed to defy what one would expect of a
Lawrence Erlbaum.
rational individual. This approach was developed
Laughlin, P. R. (1999). Collective induction: Twelve
postulates. Organizational Behavior and Human
by several important writers in late-19th-century
Decision Processes, 80, 50–69. France and continued by U.S. writers well into the
20th century. Writers in this tradition saw unusual
fads, senseless panics, riotous crowds, and even
social revolutions not merely as separate curiosi-
Collective Movements ties but as phenomena with common properties
subject to common explanation. Some writers in
and Protest this tradition stressed the ways in which the inter-
actions of people could lead to a surrender of the
The study of collective movements and protest has capacity for realistic assessment of the conse-
roots in the 19th century and has long been part quences of action. In this line of thought, group
of the social sciences curriculum. In the 1970s, members would uncritically imitate each other,
new theoretical approaches and new empirical buoyed up by group approval, with collectively
methodologies revitalized the field. The next several irrational results. This variant made social psycho-
decades witnessed an efflorescence of research by logical processes central to their explanations. The
practitioners of various social science disciplines— French writer Gabriel Tarde was one of the foun-
especially sociology, but also political science, dational figures with his stress on the sources and
history, and anthropology. Some researchers consequence of “contagion” as otherwise puzzling
focused on the internal dynamics of collective actions spread from one person to another. In the
mobilizations, including interpersonal processes; United States, a major figure was Herbert Blumer,
others addressed the ways broad social and politi- who classified crowds into distinctive varieties and
cal contexts shaped movements and were shaped for whom crowd behavior went through a succes-
by them. Building on the scholarly advances of the sion of lawlike stages.
previous 30 years, researchers in the early 21st A second approach was to stress the significance
century have been raising new questions. of social context. A very common argument in this
tradition was that social transformations in the
modern world, especially rural to urban migration,
Historical Background
industrialization, and access to mass communica-
Social and political transformation in the recent tions, broke down the traditions that had held
past and anticipated future led Americans and people in their conservative grip. Consequently,
Europeans in the 19th century to reflect on collec- they disrupted the networks of family and village
tive movements and social protest. The social sci- that had socialized the young and monitored the
ences emerged at this time, in the wake of the actions of adults, and exposed people to an unfamil-
American and French Revolutions. Industrialization iar social world in the growing towns and the new
104 Collective Movements and Protest

routines of an industrializing order. As people irrationality was intuitively appealing to those

moved from village to town, such processes both afraid of riotous crowds, bemused by odd fash-
removed the inhibitions that had been built into ions and fads, saddened by the erosion of valued
rural life and exposed the urban newcomers to mes- traditions, and/or appalled by revolutions. The
sages from manipulative elites that they were unable rise of fascist movements after World War I
to evaluate. The combined result of such processes seemed to lend additional credence to this picture,
was dangerously irrational behavior. This argument since these movements were seen as both irratio-
about modernizing contexts as a source of collective nal and popularly supported, and since they were
irrationalism could be combined with the social commonly understood as originating in modern,
psychological properties of collectives by contend- mass societies battered by the shock of war fol-
ing that those shorn from the familiar and custom- lowed by economic crises. The general notion that
ary order were prone to seek companionship in collective behavior was a type of temporary psy-
mass organization and to listen to leaders promising chopathology remained strong.
simple solutions to the ills of modern life. Pioneering
French sociologist Émile Durkheim influentially New Evidence and New Ideas
described modernizing change in Europe as a break-
down of customary norms and practices. The During the 1960s, the accumulation of a wide vari-
notion of breakdown seemed to several generations ety of new evidence called into question some of the
of social scientists to explain why contemporary prevailing theories, and new theoretical approaches
Europe and North America fostered apparently were soon developed. Over the next several decades,
irrational movements despite the advances of sci- there was an explosive growth in research in the
ence and technology. In the 1960s, American soci- field. Good research into the participants in the so-
ologist Neil Smelser synthesized this line of thinking called ghetto disturbances in many U.S. cities
in his book Theory of Collective Behavior. strongly indicated that the events tended to involve
Other approaches were also being developed, long-time residents of those cities, not newcomers
both in theory and in empirical observation. torn loose from some other way of life. New
Students of European labor movements, including research by social historians of France and England
Marxists as well as others, thought at least some revealed that participants in many urban distur-
kinds of protest derived from accurate understand- bances in the 18th and 19th centuries were people
ings of material conditions on the part of workers without criminal records, with stable occupations,
in the modern industrial sectors. Investigations and with families. Research by social historians and
into labor movements sometimes drew upon con- anthropologists on migrants from countryside to
siderable empirical research both on the contexts town in the past and present showed strong pro-
of work and on such collective actions as strikes. pensities to maintain ties to village life rather than
Labor movements acquired a rather special posi- to experience complete ruptures. Protest began to
tion for some observers. This was especially so in seem a form of political action rather than a col-
Europe, where it became common to distinguish lectively induced bout of irrationality, more to be
two classes of collective phenomena, the “social” thought of among other political phenomena than
movement and the “national” movement, the for- in the company of fads and crazes, and deserving of
mer stressing actors whose common class interests more scrutiny by political scientists and less scru-
infused their self-representation, identity, and tiny by students of mental aberration.
action and the latter acting on the basis of real or At about the same time, social scientists were
imagined commonalities of history, language, cul- suggesting four new models for the explanation of
ture, and geography. This mode of classification protest, drawing on extensive empirical investiga-
suggested on the one hand that labor movements tion for support and building conceptual bridges
were the core of social movement studies, and on to other fields of social research.
the other that nationalism was the proper subject
Resource Mobilization
matter of other scholars with other tools.
Nonetheless, the general sense that there was To the pioneers of this approach, John McCarthy
a range of phenomena that suggested collective and Meyer Zald, the core actors in protest and
Collective Movements and Protest 105

other forms of collective action were organizations action seemed explicable as evolving organizational
engaged in the strategic deployment of available strategy. When Suzanne Staggenborg and David
resources, rather than individuals committed to Meyer analyzed social conflicts through the lens of
irrational actions. Such organizations had actual movement–countermovement interaction, they
and potential access to such resources as the time were drawing on this young, lively theoretical tra-
of supporters of organizational goals or the funds dition. Social movement organizations could be
supplied by sympathizers. They might be con- seen as one kind of contender for influence, along
strained in various ways by resource deficits; an with the interest groups that had long been noted
organization that could rely on the time of pas- by students of democracy. The concept of “oppor-
sionate enthusiasts might well make different stra- tunity” became central to such interpretations,
tegic decisions from one that relied on donations since movements were seen as adapting to available
from cooler sympathizers. One organization might possibilities. Through the political process model,
launch street protests and another hire lobbyists. social movement theory was able to draw on ideas
One might develop mechanisms to arouse and sus- developed by political scientists.
tain commitment to highly risky personal actions,
while another recruited professional specialists in
New Social Movement (NSM) Theory
making fundraising appeals. McCarthy and Zald
paid a good deal of attention to the ways in which Scholars, at first particularly in Europe, were
movement organizations might permit a profes- paying attention to what protesters were demand-
sionalized stratum to make a career in the move- ing and noted a reduced emphasis on the class-
ment. The ways in which movements might based themes with which certain branches of social
“frame” issues and actions could be seen by adher- movement scholarship had been preoccupied. A
ents of this model as rational strategies for induc- variety of new concerns seemed to dominate recent
ing commitments of resources from actual or protest politics—feminism, human rights, critiques
potential contributors of time or money or as of consumer culture, concerns over alternate
devices to achieve various goals in dealing with lifestyles—and these concerns were often carried
opponents, other movement organizations, or the by less hierarchically structured and less centrally
public. Through the resource mobilization model, controlled organizations than the long-familiar
social movement theory was able to appropriate unions and socialist parties, and with significant
ideas drawn from the sociology of organizations. middle-class participation. The new movements
seemed as much about expressing identities as
about satisfying interests, and as much aimed at
Political Process
altering cultures as at altering the distribution of
Emerging at about the same time, the political material well-being, often describing themselves as
process model looked at movement organizations something new. Scholars like Alain Touraine and
as engaged in strategic interactions with their envi- Alberto Melucci saw the explanation in broad
ronment, including potential allies and opponents, changes in European social context after World
governments, and the public. Movement organiza- War II. The achievement of working-class prosper-
tions were seen as modifying their strategies and ity in postfascist Europe not only reduced the
tactics in light of the responses of their environ- intensity of the class-based conflicts of the past,
ments and of their successes and failures in attain- but also incorporated working-class politics within
ing goals. For organizations that were engaged in the framework of securely democratic political
extremely contentious causes or that embraced for- processes and extensive social welfare institutions.
bidden tactics, simply surviving in the face of oppo- But new issues emerged and energized a different
nents’ counteraction was likely to be among the kind of social movement activism on behalf of
key goals. As researchers came to chart the ways in other goals.
which organizations altered their actions in response Critics who took a broader historical or geo-
to success and failure and to other actions by allies graphical view were quick to point out that the
and opponents, phenomena that had to earlier gen- sorts of identity issues that postwar new social
erations seemed instances of irrational belief and movement (NSM) theorists held to be new could be
106 Collective Movements and Protest

found in the working-class mobilizations of the Since it is evident that there is a great deal of
industrializing era or that environmental issues had social movement activism, either these dilemmas
emerged not only among the well-off citizens of are in practice solved in some way or the issues are
prosperous democracies but in the slums of places poorly posed. Some social movement scholars
like Rio de Janeiro. While some social movement essentially accept the agenda of trying to explain
scholars tended to be simply dismissive of NSM how rational individuals come to participate in
theorists, however, their attention to identity, cul- social movement activism. Some argue that there
ture, the diversity of participants, and the ways are many ways of resolving these dilemmas and
goals and organizational forms shift in response to that participants do not invariably experience
long-term broad changes in economy and polity them as a problem to be overcome. (If, for exam-
were drawn on by many others, even if specific ple, people experience the solidarity of shared
claims about novelty and geography proved empir- danger as a reward, then the apparent cost of risky
ically untenable. In addition, through new social collective action is simply an observer’s error.)
movement theory, the study of movements was able Through collective action theory, whether
to draw on ideas from the sociology of culture. embraced, modified, or rejected, social movement
scholars came to refine their sense of the rewards
and costs of action.
Collective Action
This fourth innovation took the rational indi-
The Standard Agenda
vidual actor as central. The “collective action
problem,” as understood by economists like Since the 1960s, these four approaches, separately
Mancur Olson, was explaining how individual or in combination, have informed the interpreta-
human beings could manage to come together for tion of data gathered by an increasingly numerous
collective goals. This was initially only incidentally body of researchers using a wide array of empirical
a question of how they could come together in materials. Archival research, statistical analyses,
social movement activism in particular. sample surveys, ethnographies, and other tech-
This could be thought of as several separable niques have been used in the study of social move-
problems. ments. Even more than by the embrace of new
theories, social movement scholarship for the past
•• The free rider problem. Why should a rational several decades has been distinguished from that of
individual participate in collective action for a the 19th century and earlier 20th century by draw-
collective objective (even if that individual were ing on superior research.
to benefit from success), rather than sit back and Together these approaches constituted an
let others take the risks and pay the costs? agenda for the study of protest movements that
•• The problem of trust. In circumstances under became the main line of approach for the next
which the participation of many would be several decades. The standard agenda saw move-
essential to achievement of the goal, why should ments as episodes in which actors of various sorts
a rational individual trust those essential others developed strategies and tactics that made the best
to participate, especially if the risks of failure of available resources, reacted to threats or seized
would be to leave one significantly less well off opportunities they for the most part did not them-
than before (for example, dead, imprisoned, fired, selves create, and struggled to frame issues in
or poorer)? This can be formulated in terms of ways that would energize supporters, attract
the well-known game prisoner’s dilemma. allies, neutralize opponents, appeal to powerhold-
•• The problem of personal efficacy. To the extent ers, or justify their actions to themselves. Different
that one could trust that many others would elements of this agenda appealed variously to
participate, why would one pay the costs and scholars committed to “structuralist,” “cultural-
run the risks of participation unless one thought ist,” or “rationalist” modes of explanation, who
that one’s own addition to the collective action invested much energy debating whose mode of
would be the crucial difference between success explanation was more fundamental. For those of
and failure? a more structuralist bent, a good explanation
Collective Movements and Protest 107

would be one that would delineate the institutions Expanded Geography

that endowed various actors with differing inter-
In addition to shining the light of scholarship
ests and resources; culturalists tried to grasp pre-
into the past of the countries that became wealthy
vailing values within which movement activists
democracies, scholars took note of the great geo-
maneuvered to demonstrate the worthiness of
graphic concentration of high-quality empirical
their purposes and methods; and rationalists tried
research within the wealthier parts of the world.
to show how actors endowed with particular
These were places where data collection was easier,
interests would hit on particular strategies to
and they were the places social movement scholars
defend and advance them.
lived and worked. But this meant a restriction on
Other researchers, however, found the stan-
the range of social contexts within which protest
dard agenda too constraining and in increasing
and collective action more generally were being
numbers were urging the field to move into new
studied. Beginning in the 1970s in Mediterranean
Europe, the greatest wave of democratization in
history opened up new opportunities to conduct
Recent Directions research and posed an array of interesting com-
Expanded History parative questions for students of protest and col-
lective action. Spanish and Portuguese scholars
Increased attention to the interactions of move-
explored the role of protest mobilizations in those
ments with their contexts, including other organi-
countries. The role of social movements in the ter-
zations and especially governments, moved beyond
mination of military regimes in the 1980s was one
short-term studies of particular episodes or cam-
important focus of new study in Latin America.
paigns to long-term historical change. Movement
With the development of the ensuing democratiz-
scholars explored the ways patterns of mobiliza-
ing regimes, students of that region’s protest pat-
tion, expression of demands, targets, tactics, and
terns were examining which sorts of movements,
modes of movement organization changed in
grievances, and actions were continuing those of
response to large social, economic, and political
the military period and which were new. They
changes like urbanization, the growth of effective
were addressing as well which aspects of mobiliza-
states, democratization, or, in the late 20th cen-
tions grew weaker in the democratic period. With
tury, globalization. Charles Tilly argued that there
the fall of European Communist regimes in 1989
are “repertoires of contention” that are available
and beyond, the democratic transition in South
to participants in conflict and that these repertoires
Africa of the 1990s, and democratic movements
are formed at particular historical junctures. The
and transitions in Asia, similar sorts of questions
“demonstration,” for example, entered the reper-
were framing research in those places as well.
toire of contention in Europe in the early 19th
Those aiming at large theoretical statements were
century and then spread far and wide. So the ways
coming to have a much greater range of well-
in which people engage in conflict tends to be situ-
researched cases on which to draw, and this was as
ated within a limited array of culturally available
much a challenge to generalization as it was a spur
forms, although at important moments new forms
to new theory. Some scholars were arguing for a
are invented or old forms stop being used.
broader geographical extension still, bringing
Historians have made enormous contributions to
together into a comparative framework the forms
the study of particular forms, such as the “food
of social movement activism of wealthy democra-
riot” that was of great importance in English and
cies and the forms of protest characteristic of non-
French history from the 17th century into the 19th
democratic regimes.
century. By extending the terrain of research back
before the era of democracy, it became possible to
Transnational Processes
conduct empirical research into how subsequent
democratization reshaped social conflict. These One rather special aspect of the geographic
forays into the past by social movement scholars expansion of protest studies was attention to
dovetailed with the movement among historians to movements that themselves moved on a large geo-
write “history from below.” graphic stage. Students of protest and collective
108 Collective Movements and Protest

action were paying a lot of attention to movements people into protest, studying recruitment into a
that in some sense crossed borders, which could very wide variety of organizations and causes,
come in many guises: from peaceful protest carried out in public view to
violent acts prepared by clandestine organizations.
•• participants traveling to distant sites for Participants were interviewed while demonstra-
concerted action tions were taking place, in prison after arrest, or
•• organizational ties that crossed national frontiers years after various events took place. Researchers
•• protesters targeting institutions of transnational collected participants’ accounts of the processes
governance like the International Monetary Fund that led them to activism and of their lives while
or the World Bank they were activists. This research suggested that
•• immigrants becoming involved as participants or recruitment was frequently a network phenome-
as objects of protest non. People often went to their first meeting or
•• transnational institutions having an impact on first demonstration because a friend was going.
national politics (which became increasingly Although people sometimes chose movements out
common within the European Union) of deep affinity for a cause, it also often happened
•• transnational identities playing a conspicuous that deep ideological affinity only developed as a
role in social conflict (for example, in Islamicist consequence, not a cause, of participation.
movements) Particularly in the subset of causes known as “high
•• protesters in one part of the world learning from risk activism,” participation itself led to increased
protesters far off through vivid television footage trust in fellow activists and increased distance
and exchanging ideas through e-mail and the from previous acquaintances outside the move-
Internet ment. Activist groups could thus develop distinc-
tive subcultures.
Bringing Back Emotion
Beyond Organizations and Beyond the State
In rejecting the older collective behavior model
of irrational action induced by collective processes, Following the standard agenda, a great deal of
some scholars were suggesting that the field had research had accumulated on the ways in which
gone too far in the direction of seeing movement organizations recruited members, crafted strategies
organizations and individual participants engaged and tactics, and dealt with each other in forming
in consciously calculated behavior toward speci- alliances, competing for resources, and opposing
fied ends. Some scholars were now arguing that one another as they struggled to shape state poli-
protest and activism were as much matters of the cies. Thus the movement organization and the
heart as of the head. In considering the claims of state became central concepts for the analysis of
injustice that so often were a part of movement protest. Some researchers aimed at reconceptual-
life, could one do without taking into account situ- izing the study of protest in ways that made one or
ations that elicit anger, disgust, and humiliation? both of these concepts less central.
Could one explain why one appeal succeeds in
Movement Subcultures and Milieu
mobilizing participation and another does not
without taking into account the ways in which Some scholars, like David Meyer, suggested that
shame can become pride? Could one understand there were movement subcultures formed around
how social activism was sustained over time with- adherents of change, who took their propensity to
out considering the pleasure social activists some- protest from cause to cause, and from organization
times take in the fellowship formed in working to organization, and whose careers as protesters
together for a common purpose? and activists were an important object of study.
The formation and sustenance of such subcultures
seemed a new subject matter not exhausted by the
Microprocesses in Context
study of the strategies of movement organization.
Researchers were using varied methodologies to Others suggested that one could identify venues
address the interpersonal mechanisms that brought where activists, potential activists, and nonactivists
Collective Movements and Protest 109

met and created protest-oriented places that func- •• Relevant data. Since data on policy change are
tioned as sources of protest identity and recruitment, generally going to be quite different from data
yet were distinguishable from the organizations on movement actions, the incorporation of
into which they fed. Particular bars, cafes, con- appropriate data will often mean a very much
certs, parks, or neighborhoods could be places larger research endeavor.
strongly flavored by certain social and political
attitudes and outlooks, shaping and nurturing pro- Despite these difficulties, scholars were recogniz-
test but readily overlooked by an exclusive focus ing the study of movement impact as a necessary
on organizations that acted. addition to the research agenda.

Reconsidering the State

Some held that the same sorts of processes that
protesters deployed in relation to states occurred Four decades after new theory and superior empir-
in relation to other defined sites of hierarchical ical research challenged the kinds of analysis that
power. Others went further, suggesting that if the had developed in the 19th and early 20th centu-
structures that sustain social hierarchies are located ries, the study of protest, collective action, and
inside each of us through beliefs taken in as chil- social movements is livelier than ever, moving in
dren or fostered by social interactions as adults, new directions, and with the kinds of disagree-
then one could think of action directed at changing ments among practitioners that, in raising basic
such attitudes, socialization processes, and interac- questions, inspires new research.
tions as crucial sites of change. In this perspective, John Markoff
a group of people meeting in each others’ homes
for the purpose of mutual support in inner change See also Civil Rights Movement; Crowds; Free Riding;
could be thought of as participating in social activ- Ideology; Multiple Identities; Nationalism and
ism even if no publicly visible protest against pub- Patriotism; Organizations; Relative Deprivation;
lic policies ensued. The field of protest studies, in Virtual/Internet Groups
fact, has significant disagreement over how to
define a social movement. For some, the existence
of publicly visible actions in interaction with a Further Readings
state is critical, but for others any concerted action,
Kriesi, H., Koopmans, R., Duyvendak, J. W., & Giugni,
whether publicly visible or hidden away some-
M. G. (1995). New social movements in western
where, and whether directed at a state or not, Europe: A comparative analysis. Minneapolis:
constitutes a sign of social movement activism. University of Minnesota Press.
Lichbach, M. I. (1995). The rebel’s dilemma. Ann Arbor:
Consequences University of Michigan Press.
McAdam, D., McCarthy, J. D., & Zald, M. N. (Eds.).
While much of protest and movement activism
(1996). Comparative perspectives on social
involves claims of acting on behalf of change, the movements: Political opportunities, mobilizing
study of the effectiveness of collective action in structures, and cultural framings. Cambridge, UK:
bringing about change has emerged as a relatively Cambridge University Press.
neglected area of research. The difficulties are of McAdam, D., Tarrow, S., & Tilly, C. (2001). Dynamics
several sorts: of contention. Cambridge, UK: Cambridge University
•• Conceptual. What sorts of change should be Meyer, D. S. (2007). The politics of protest: Social
looked for? And on what time scale? movements in America. New York: Oxford University
•• Analytic. Since movements are generally Press.
happening at the same time as many other Meyer, D. S., & Staggenborg, S. (1996). Movements,
things, demonstrating the precise impact that countermovements, and the structure of political
movements have had on policy changes can be a opportunity. American Journal of Sociology, 101,
difficult task. 1628–1660.
110 Collective Self

Smelser, N. (1963). Theory of collective behavior. scholars have argued that there are two kinds of
New York: Free Press. self, the individual self and the collective self. Each
Smith, J. (2008). Social movements for global democracy. has its peculiar body of self-knowledge and both
Baltimore: Johns Hopkins University Press. provide a means of self-definition.
Tarrow, S. (2005). The new transnational activism. The individual self is organized around a set of
Cambridge, UK: Cambridge University Press. personal attributes that, like a fingerprint, are
Tilly, C. (2008). Contentious performances. Cambridge, unique and differentiate a person (“I”) from all
UK: Cambridge University Press. others (“you”). So, for example, if I am kind but
Zald, M. N., & McCarthy, J. D. (Eds.). (1987). Social
fastidious, hardworking but temperamental, shy
movements in an organizational society. New
but reliable, then such attributes are what make
Brunswick, NJ: Transaction.
me a different person from you. In contrast, the
collective self is constructed from attributes that
are shared with other members of the ingroup. Its
function is to differentiate the person from others
Collective Self who are not group members, thus distinguishing
“we” from “they.”
Collective self is a term used in social psychology People belong to a variety of groups, some of
to refer to those aspects of self that derive from which are important in defining a number of col-
the groups a person identifies with or belongs to lective selves, such as ethnicity, race, gender, and
as a member. When the collective self is active, a nationality. One’s membership in groups can vary
person construes the self as “we” based on attri- in salience at different times. If someone is Swiss
butes that are shared with other group members, and crosses into Italy, it is highly likely that nation-
rather than simply as “I” based on attributes that ality, in this case based on attributes shared with
are uniquely personal. Viewed historically, this others who are Swiss, will be highly accessible (i.e.,
psychological focus on a social self rather than an easily brought to mind).
individual self has been slow in coming, though By learning to identify with a particular group,
the concept has long been recognized in sociology. a person acquires a specific collective self and a set
The collective self taps into group membership of group norms that can guide how to think and
and is activated when a particular group becomes act appropriately.
salient at a moment in time. When this happens,
the person draws on a relevant group membership
Collective Action
to make inferences of self-worth and sometimes to
take action in concert with others. Research and Because the individual self and the collective self
theory in this field is at the intersection point of are aspects of the self-concept, both are located
social psychology and sociology, though there within the person and function at that level. Unlike
have been some recent contributions within polit- the individual self, the collective self provides the
ical psychology. person with a mechanism for responding to a
social situation in unison with others. It can both
motivate and guide behavior in one-on-one encoun-
Two Kinds of Self
ters with members of outgroups, or when a person
The idea that the self-concept develops exclusively is acting as part of a larger group (e.g., as in a
from individual experience held sway in social psy- political demonstration). The collective self can
chology for many years, ignoring much of what therefore serve as a platform for collective action,
early sociologists such as G. H. Mead and Charles which social identity theory (SIT) equates with
Cooley had to say about the interplay between indi- social change.
viduals and their society, and about how the self is As conceived by Henri Tajfel, social change is
socially constructed. More recently, social psychol- predicated on a belief system in which intergroup
ogists have come to accept that in characterizing boundaries are thought to be impermeable to
the self as a purely individual entity a great deal of “passing,” for example, the Hindu caste system in
psychological meaning is lost. Many contemporary India. As a result, a lower status individual can
Collective Self 111

improve his or her self-esteem defined by caste associates developed this theory to bring ideology
only by challenging the legitimacy of the higher to center stage—an ideology that favors the status
status group’s position. Whether or not some form quo, even when this conflicts with the interests of
of group action is possible depends on how the the individual or of the group. Ironically, lower
status quo (the existing status and power hierar- status group members sometimes subscribe to this
chy) is perceived. Is it secure or insecure? If lower ideology even when it legitimizes their current
status individuals think it is stable, legitimate, and status and maintains their position of disadvan-
therefore secure, it may be impossible to imagine a tage. Members of a subordinate group may do
different world—an alternative social structure. this to reduce uncertainty by assuming that it is
In such circumstances there is no path for col- better to live in disadvantage and be certain of
lective action. However, there are socially creative one’s place than to challenge the status quo and
strategies that groups can use to foster a positive enter unknown territory.
self-image for the collective self:
Collective Self-Esteem and Ethnicity
•• Subordinate groups can make intergroup
comparisons on novel or unorthodox dimensions A group-based definition of the self can lead to
that favor their group. For example, in a French depressed self-esteem, for example, among mem-
study by Gerard Lemaine, children took part in bers of disadvantaged minorities. Studies of young
an intergroup competition to build the best hut. children’s ethnic identity in the United States and
Lemaine found that when groups were provided elsewhere have provided supporting evidence. A
with poor building materials, and therefore series of American studies of ethnic choices made
could not win, they became creative by by children from the late 1930s to the early 1960s
emphasizing how good a garden they had made. showed that
•• Groups can try to change the consensual value
attached to ingroup characteristics (e.g., •• White children identified with and preferred to
scientists with insufficient funds to purchase be with White children;
sophisticated equipment can highlight the •• Black children identified with and preferred to
conceptual advances of their work). be with White children;
•• They can compare themselves favorably with •• Black children had lower self-esteem.
other groups also of lower status (e.g., working-
class Hispanics pinpointing ways in which they In 1954, the eminent Black American researcher
are superior to working-class Whites). Kenneth Clark appeared as a witness in the land-
mark United States Supreme Court case Brown v.
The contexts in which collective action occurs Board of Education. He testified that generations
require a belief system with distinctive features. of Black children’s self-esteem had been exten-
First, lower status people need to believe that the sively damaged over time. Flowing from this case,
status quo is illegitimate, unstable, and thus inse- the legal decision to outlaw school segregation was
cure. Second, they need to have cognitive alterna- instrumental in legitimizing the civil rights move-
tives (i.e., conceive of a different and more ment in the United States.
promising social order). If both of these conditions Clark’s research was criticized for its assump-
are met, direct social competition is likely. This tion that doll preferences reflected children’s self-
manifests itself in actual intergroup conflict. Such esteem levels. However, Graham Vaughan found
conflict could take several forms, including civil two stable trends in international ethnic identity
rights action, political lobbying, public demonstra- studies that are consistent with Clark’s conclu-
tions, and even terrorism, revolution, or war. As sions: (1) ethnic minorities that are disadvantaged
Stephen Reicher has observed, social movements (educationally, economically, politically) are typi-
typically emerge under these circumstances. fied by lowered self-esteem when intergroup com-
SIT incorporates both social psychological con- parisons are made and (2) social change in the
cepts and social structural (macro) concepts, as status relationship between ethnic groups leads to
does system justification theory. John Jost and his a significant improvement in minority pride and
112 Collectivism/Individualism

individuals’ feelings of self-worth. By 1970, the

ethnic choices made by Black children in the Collectivism/Individualism
United States reflected the phenomenon of “Black
Is Beautiful” that followed the success of the Collectivism is a societal orientation toward promot-
American Black Power movement in the late ing the well-being of the collective, whereas individu-
1960s. There was a similar trend in a series of New alism is a societal orientation toward the well-being
Zealand studies of cohorts tested in different time of individuals. Collectivism and individualism are
periods. Young Maori children switched from pro- important concepts in group processes and inter-
White choices to pro-Maori choices against the group relations because these concepts capture two
backdrop of a Brown Power movement. fundamental motivations of homo sapiens: “getting
Findings from ethnic identity research testify to along” and “getting ahead.” Because humans are
the power of the collective self as a focal point in social animals living in a group that often has com-
the processes of social movements and social plicated internal structures (e.g., an organization),
change. Collective self-esteem can be linked to psy- we must get along with other group members while
chological health. A heightened sense of pride in simultaneously trying to get ahead of them.
one’s ethnicity or gender, or any other valued Collectivism and individualism can be understood
social category, can contribute significantly to a as two significant ways to deal with these two fun-
positive view of oneself defined in group terms. damental human needs at the level of society. Society
consists of multiple groups that compete with one
Graham M. Vaughan another for resources for survival. Thus, in addition
to an individual’s need for survival, the group has its
See also Civil Rights Movement; Collective Movements
own need for survival. Collectivism can then be con-
and Protest; Collectivism/Individualism; Ethnicity; Self-
sidered a solution that attempts to maximize har-
Esteem; Social Identity Theory; Symbolic
Interactionism; System Justification Theory
mony and solidarity among group members (i.e.,
“getting along”), while minimizing the potentially
destructive effect on the group as a whole of egotistic
behaviors on the part of individual group members.
Further Readings
In contrast, individualism can be considered as a
Brewer, M. B., & Gardner, W. (1996). Who is this “we”? solution that attempts to maximize the pursuit of
Levels of collective identity and self representations. self-interest (i.e., “getting ahead”), while minimizing
Journal of Personality and Social Psychology, 71, the oppressive effect of the group on individuals.
83–93. When faced with competition from other groups,
Gaertner, L., Sedikides, C., Vevea, J. L., & Iuzzini, J. a group is likely to adopt a collectivistic approach
(2002). The “I,” the “we,” and the “when”: A meta- because intergroup competition requires a degree
analysis of motivational primacy in self-definition. of solidarity among members within the group.
Journal of Personality and Social Psychology, 83,
When a group does not face significant competi-
tion from other groups and has sufficient resources
Jost, J. T., Kay, A. C., & Thorisdottir, H. (Eds.). (2009).
to share, it should be more receptive to an indi-
Social and psychological bases of ideology and system
vidualistic approach. Ultimately, collectivism and
justification. New York: Oxford University Press.
Reicher, S. D. (2001). The psychology of crowd
individualism are two adaptive approaches to the
dynamics. In M. A. Hogg & R. S. Tindale (Eds.),
prevailing intra- and intergroup conditions. This
Blackwell handbook of social psychology: Group entry looks at the history of these two social orien-
processes (pp. 182–207). Oxford, UK: Blackwell. tations and describes different ways that each may
Sedikides, C., & Brewer, M. B. (Eds.). (2001). Individual be expressed in society.
self, relational self, and collective self. Philadelphia:
Psychology Press.
Historical Background
Tajfel, H., & Turner, J. C. (1979). An integrative theory
of intergroup conflict. In W. G. Austin & S. Worchel The Confucian system, which developed in China
(Eds.), The social psychology of intergroup relations around the 5th century BCE, represents a collec-
(pp. 33–47). Monterey, CA: Brooks/Cole. tivistic solution to the potential tension between
Collectivism/Individualism 113

“getting along” and “getting ahead.” Under this provided, over the past two decades, powerful
system, harmony within five cardinal relation- frameworks for understanding cross-national vari-
ships is emphasized: father–son, husband–wife, ations in the self-concept, interpersonal relation-
elder–younger, emperor–subject, and friend– ships, and various other social behaviors. However,
friend. The emphasis is placed on understanding these constructs have also been subject to some
one’s roles, fulfilling one’s duties, and showing criticism, including the charge of conceptual ambi-
deference to authority. These, in turn, strategically guity and problems over the existence of various
reduce within-group competition, smooth inter- subtypes. Most notably, Daphna Oyserman and
personal relationships within a group, and main- her colleagues published a meta-analysis (i.e.,
tain group solidarity. quantitative analysis of all the published studies on
The Greek philosophers of the 5th century BCE self-reported individualism–collectivism) in 2002,
were some of the earliest individualist philoso- which concluded that the difference in individual-
phers. Aristotle’s Nichomachean Ethics is mainly ism–collectivism of North Americans and East
concerned with how an individual can live a virtu- Asians is negligible. Since the publication of the
ous life and attain the highest possible human goal meta-analysis by Oyserman and her colleagues,
of eudaimonia, or personal well-being. This is in many prominent researchers in the field have
stark contrast with the Confucian emphasis on called for alternative conceptual frameworks and
societal well-being and how to create a nation methodologies to be adopted in cultural and cross-
characterized by cohesion and interpersonal har- cultural psychology. However, it should be also
mony. Although some Western philosophers have noted that when response sets are statistically con-
endorsed a form of collectivism (e.g., Jean-Jacques trolled, self-reported individualism scores were
Rousseau, Karl Marx, Émile Durkheim), many highly correlated with Hofstede’s original results.
other Western philosophers explicitly endorse An additional critique of individualism–collec-
various forms of individualism (e.g., Adam Smith, tivism research revolves around its relatively static
John Locke, Thomas Hobbes). view of culture. With globalization and increasing
Geert Hofstede’s 1980 book, Culture’s international immigration, there is substantial cul-
Consequences, ignited an interest in collectivism and tural fusion. Ronald Inglehart and his colleagues
individualism in psychology. In this book, Hofstede collected survey data on values and attitudes from
examined work-related values of IBM workers from around the world for the last 20 years, concluding
40 nations and identified four important cultural that, in many parts of the world, people’s values
dimensions: “masculinity,” “power distance,” have shifted from traditional attitudes (akin to col-
“uncertainty avoidance,” and “individualism.” In lectivism) to self-expression (a feature of individu-
this survey, the United States ranked ahead of all alism) as a reflection of modernization. However,
other nations on individualism. In the 1980s, Harry some of the cross-national differences have
Triandis and his colleagues conducted numerous remained relatively unchanged over the last two
cross-cultural studies comparing the United States decades. For instance, although Japanese have
with so-called collectivist nations such as Japan, become more individualistic during that time, so
China, and Korea, and legitimized the study of col- have Americans. Thus, the magnitude of cultural
lectivism and individualism in academic psychology. difference between Japan and America has not
In 1991, two prominent social psychologists, decreased.
Hazel Markus and Shinobu Kitayama, published
an influential paper that specified systematic psy-
Varieties of Collectivism
chological differences between collectivist and
individualist nations in terms of people’s self- The Confucian style of collectivism is primarily
concept, motivation, cognition, and emotion. This concerned with how to maintain harmonious rela-
work rendered collectivism and individualism one tionships at various dyadic levels. Because of its
of the most popular research topics in social psy- emphasis on dyadic relationships, the idea is that
chology in the 1990s. group members tend to have strong ties with other
Two related constructs, individualism–collectivism members, and their interpersonal relationships are
and independent and interdependent self have as important as each member’s relationship with
114 Collectivism/Individualism

the group as a whole. Whereas the Confucian style and experienced. Because harmony is emphasized,
of collectivism emphasizes hierarchy and dyadic similarity rather than uniqueness is valued. The
relationships (i.e., hierarchical and relational col- private self—its desires, opinions, inclinations, and
lectivism), other forms of collectivism emphasize so on—may be kept cleanly separate from and seen
collective actions and cooperation in general (i.e., as irrelevant to public interactions (e.g., in Japan,
horizontal and group collectivism). this is expressed by two terms: honne, the private
For instance, in an agricultural society that self, and tatemae, the public self).
requires an irrigation system, a large number of The context-dependent nature of the self is
members have to work together to accomplish one apparent here. While an individualist might make a
common goal. In this context, one’s identification context-independent claim such as “I am hard-
with the entire group becomes more important working,” a collectivist might say “As a teacher, I
than dyadic relationships between group members. am outgoing, but as a husband, I am quiet.” The
In addition, in a situation in which collective self in collectivist societies is experienced as part of
action and cooperation are required, a large pro- a network of social roles and obligations, and
portion of group members share the same status, friendship is no exception. Opportunities to create
and therefore, relationships are not as hierarchical friendships tend to be fewer in collectivist societies,
as a Confucian-style collective organization. Sports as they are created based on preexisting networks.
teams (e.g., football, soccer) represent this type of In fact, friendships are not “created” so much as
collectivism. The distinction between relational “given” in collectivist societies such as Ghana. In
and group collectivism is similar to Ferdinand contrast to individualist cultures, where it is more
Tönnies’s distinction between a small village-like common and desirable to have many friendships,
community (gemeinschaft) and an association- with various levels of personal intimacy, friendships
based society (gesellschaft). in collectivist cultures tend to be more binding,
In relational collectivism, relationships are the with a variety of obligations. The depth and obliga-
self’s defining features, and either personal goals tion of collectivist friendship also limits the sheer
are deemphasized in favor of interdependent goals number of friends one can maintain. In Ghana,
or no distinction between personal goals and sig- most people view having many friends as “foolish”
nificant others’ goals is made. Similarly, in group or naïve, given the impossible level of commitment
collectivism, the fluidity of the self in its relation- and strain that they place on one’s resources.
ship to the group as a whole is expressed as a sense As indicated above, collectivist groups tend to
of sharing a common fate with other group mem- be less permeable: They are difficult to freely enter
bers and in the belief that the experience of one and exit. As people’s social ties are relatively set,
ingroup member affects all other members of the social goals focus on maintaining the relationships
group. The fulfillment of one’s collective responsi- one already has instead of forming new relation-
bility by promoting the interest of one’s group or ships. Consequently, people in a collectivist society
significant others is of primary concern in both tend to draw a sharper distinction between mem-
forms of collectivism. The emphasis on group bers of one’s ingroup and strangers than people in
goals and interdependence is associated with an individualist society, where strangers are
greater exertion of effort and higher levels of coop- regarded as potential future friends. Indeed, in col-
eration among group members, who not only lectivist cultures, the emphasis on harmony, coop-
identify closely with their groups but also regard eration, and cohesion tends to be limited to
their individual contributions as vital to their ingroup relationships. Interactions with strangers
groups’ accomplishments and overall well-being. and outgroup members may be characterized by a
Instead of striving explicitly to actualize a “true” lack of concern or even hostility and discrimi-
self, collectivist individuals strive to cultivate and nation during intergroup conflicts. Furthermore,
maintain deep bonds with significant others and/or differences between ingroup and outgroup com-
important ingroup members. This involves main- munication have been noted in collectivist societ-
taining harmony and adjusting to others’ (or the ies, where generally speaking communication with
group’s) needs, which in turn affects the type of emo- ingroup members is more effective than communi-
tions (e.g., calm, contentment) that are idealized cation with outgroup members.
Collectivism/Individualism 115

Another area where collectivists pay greater boundary. Personal goals are defining characteristics
attention to distinctions between an ingroup and of this autonomous entity and may be in opposition
outgroup than do individualists is in the allocation to the social world. In such cultures, importance is
of available resources. Given that relationships placed on self-reliance, personal cultivation, per-
with ingroup members are of great importance, sonal choice, uniqueness, and self-expression.
material resources tend to be shared equally within Like collectivism, individualism manifests in
collectivist groups: A tribe’s catch can be distrib- different forms. One common typology that has
uted and shared among its members; a dish bought been used to describe different types of individual-
at a restaurant can be distributed among those at ism is horizontal versus vertical individualism.
the table. However, when the task involves distrib- Both types of individualism are characterized by
uting material resources to strangers, decisions are the emphasis on the autonomy and uniqueness of
based on the recipient’s level of contribution, which individuals. However, horizontal individualism
is more typical of exchanges among individualists. tends to stress equality between individuals,
Collectivists also behave more like individualists in whereas vertical individualism emphasizes compe-
favoring contribution over membership when goals tition and being the best in comparison to others.
of productivity and competition take precedence Although some forms of individualism are con-
over goals of interdependence and solidarity. sidered antagonistic toward society (e.g., rugged
In terms of work-related behaviors within col- individualism), the pursuit of self-interest can be
lectivist contexts, working with ingroup members beneficial both to individuals and society. For
tends to result in better performance than working instance, in the 19th century Alexis de Tocqueville
alone. Individual performance of collectivists is observed that Americans, despite their individual-
also enhanced when respected ingroup members ism, help strangers much more willingly than do
are involved in assigning specific goals and tasks. the French, precisely because Americans under-
Compared to their individualist counterparts, lead- stand that many others do not have anyone else to
ers and superiors in collectivist cultures are more rely on; he called this “the doctrine of self-interest
likely to provide nurturance and guidance to sub- properly understood.” In other words, when self-
ordinates, who are expected to show loyalty and interest is pursued with moderation, individualism
devotion in return. does not necessarily interfere with the well-being
Collectivism also has important implications for of society.
group decision making and conflict resolution. For The emphasis on the development and expres-
example, the focus on maintaining ingroup cohe- sion of one’s unique, “true” self constitutes a cor-
sion and harmony means that relationships among nerstone of the formation and experience of
ingroup members may be valued more than the friendship in individualistic societies. According to
tasks themselves during the decision-making pro- most forms of individualism described above,
cess. Also, greater attention is paid to the goals and friendship involves the meeting of two distinct and
concerns of the collective than to the needs of indi- fundamentally separate beings, whose connection
vidual members. Conflicts within the ingroup tend must be created via a mutual sharing of unique
to be avoided whenever possible. When such con- and authentic selves and/or common interests.
flicts do arise, attempts are made to reach agree- Romantic companionship should also be a celebra-
ment or consensus within the group. This often tion and exploration of exciting and unique traits
requires willingness on the part of group members in this context. With different friends reserved for
to modify their own preferences in order to con- the sharing of specific interests, friendships in such
form to the group’s position. societies also become more compartmentalized.
True individualist friendship, like the true self,
should be genuine and spontaneous. Friendships
Varieties of Individualism
should feature spontaneous acts of affection,
The defining characteristic of individualism is the should not arise out of obligation, and may involve
priority it accords to individual goals over group but should not be predicated on practical or mate-
goals. This worldview features the individual as a rial support. Friendships have varying degrees of
self-contained entity with a well-defined self–other intimacy, and those that require self-censorship or
116 Collectivism/Individualism

too much obligation can easily be deemphasized or connections between people are loose, task con-
even left. This ideal is reflected in the high cerns prevail over relationship concerns when mak-
American divorce rate. Indeed, many Americans ing group decisions. Without having to worry as
show a permissive, noncommittal view of love. much about harmony and cohesion within the
Relationships with groups and associations group, members tend to be more outspoken about
also reflect the above model of low-cost, low- their feelings, attitudes, and opinions. It is consid-
commitment, personally expressive relationships. ered acceptable for individuals to place personal
Groups in individualist cultures are more perme- goals ahead of group goals. When disagreements
able. That is, they are easy to join and leave, arise within the group, maintaining one’s own views
requiring little obligation (e.g., a book discussion and problem solving via open and direct communi-
group). With high intergroup mobility, there are cation is seen as helpful, whereas conforming to
many potential ingroups, and it is relatively easy to group pressure is considered a personal weakness.
stay with those that meet one’s personal needs and In addition to particular views toward friend-
leave those that do not. Intergroup mobility is sup- ships and groups, individualism has been associ-
ported by Americans’ high level of general trust in ated with specific emotional experiences. For
others, and this facilitates social exploration and example, pride, a socially disengaging emotion,
making friends. clearly sets the self apart from surrounding others.
With the primary focus of attention on the pur- Among individualists, positive emotions are
suit of personal goals, needs, and desires, the dis- desired, regardless of whether they are socially
tinction between ingroup and outgroup does not engaging or disengaging. This is not the case
take on the same degree of importance in individu- among collectivists, because pride signals a disrup-
alist cultures as it does in collectivist cultures. In tion in harmony with others.
other words, people who are individualist oriented The ethos of individualism is liberating to the
tend to treat friends and strangers more equally. individual, but it is also associated with anxiety and
Their communication with ingroups and out- identity crises. In the United States, for instance,
groups tends to be similar, not varying much in increased personal freedom to choose one’s occu-
overall effectiveness. Also, because individualists pation, spouse, and system of values over the last
experience greater opportunities and need to inter- 50 years has been accompanied by an increased
act with outgroup others, they tend to be more anxiety; this has been referred to as the age of the
skilled than collectivists are at communicating “me” generation and as an “age of anxiety.”
with strangers. Young Americans report experiencing significant
Research also indicates that an individualist ori- pressures to actualize their potential and define
entation is associated with a phenomenon known their identity, and they are plagued by the impend-
as social loafing, which refers to the tendency of ing fears and guilt of failing to choose the best
individuals to exert less effort when working as opportunities available to them. In this sense, the
part of a group than when working alone. The personal autonomy and independence that is cen-
higher occurrence of social loafing among individ- tral to American individualism has its downside.
ualists may partly be due to the greater value being
placed on personal outcomes as opposed to collec- Shigehiro Oishi and Jenny C. Su
tive welfare. It is also known that individualists
See also Collective Self; Common-Identity/Common-Bond
perform better when they are given a choice in the
Groups; Culture; Diversity; Group Socialization;
type of tasks that they will undertake.
Group Structure
Individualism also affects how important group
decisions, such as the allocation of material
resources, are made. In most situations, people in
individualist cultures are inclined to allocate rewards Further Readings
based on group members’ individual contributions Brewer, M. B., & Chen, Y.-R. (2007). Where (who) are
rather than their membership because, in their collectives in collectivism? Toward conceptual
view, group achievement reflects the sum of clarification of individualism and collectivism.
individual inputs. In individualist cultures, where Psychological Review, 114, 133–151.
Common-Identity/Common-Bond Groups 117

Hofstede, G. (1980). Culture’s consequences: Background

International differences in work-related values.
Beverly Hills, CA: Sage. The distinction between common-identity and
Markus, H. R., & Kitayama, S. (1991). Culture and the common-bond groups is reflected in the two pri-
self: Implications for cognition, emotion, and mary and competing approaches to the study of
motivation. Psychological Review, 98, 224–253. groups within social psychology. Twentieth-century
Nisbett, R. E. (2003). The geography of thought: How social psychology heavily emphasized the study of
Asians and Westerners think differently . . . and why. the individual. The focus of much of this research
New York: Free Press. was how the person impacted his or her group.
Oyserman, D., Coon, H. M., & Kemmelmeier, M. Kurt Lewin defined a group as the sum of its inter-
(2002). Rethinking individualism and collectivism: personal bonds, implying that attraction to group
Evaluation of theoretical assumptions and meta- members forms the basis for groups. The social
analyses. Psychological Bulletin, 128, 3–72. identity perspective challenged this definition,
Schimmack, U., Oishi, S., & Diener, E. (2005). arguing that groups can be formed under minimal
Individualism: A valid and important dimension of circumstances and members can identify with
cultural differences. Personality and Social Psychology groups based on shared attributes. Thus, the social
Review, 9, 17–31. identity perspective asserts that attraction to a
Tocqueville, A. de (2003). Democracy in America. New group can involve attraction to its identity rather
York: Penguin. (Original work published 1835) than to its individual members.
Triandis, H. C. (1989). The self and social behavior in The distinction between common-identity and
differing cultural contexts. Psychological Review, 96, common-bond groups is relevant to the distinction
between the individualistic and social identity
Yuki, M. (2003). Intergroup comparison versus
perspectives. Research on common-identity and
intragroup relationships: A cross-cultural examination
common-bond groups focuses on the different
of social identity theory in North American and East
functions that each kind of group provides for its
Asian cultural contexts. Social Psychology Quarterly,
66, 155–183.
members as well as the different processes that
occur within each kind of group.

Differences Between
the Two Kinds of Groups
Common-Bond Groups The distinction between common-identity and
common-bond groups has been applied to a vari-
ety of real-world groups (e.g., college eating clubs,
Some group memberships are based on sharing a online communities). In their original study of
category membership (e.g., women), while others attachment to different types of groups, Prentice
are based on attraction to fellow group members and her colleagues speculated that common-identity
(e.g., groups based on friendships). Based on this and common-bond groups differ in terms of fair-
distinction, in 1994 Deborah Prentice, Dale Miller, ness, longevity, and conflict. Recent research has
and Jenifer Lightdale identified two primary types demonstrated that these groups may also differ in
of groups to which people may belong. This typol- members’ responses to group norms and the fac-
ogy allows researchers to make predictions of how tors they use in forming impressions of themselves,
behaviors differ between the two types of groups. other group members, and their social identity.
Common-identity groups comprise members who
share a social category and are attracted to the
group as a whole as well as its overarching identity.
More specifically, members of common-identity Prentice and her colleagues predicted that mem-
groups are attracted to the group’s norms, goals, bers of common-identity and common-bond groups
activities, and other defining features. In contrast, would differ in their preferences for how rewards
common-bond groups comprise members who are should be distributed among members. They argued
attracted to one another as individuals. that, from a social identity perspective, group
118 Common-Identity/Common-Bond Groups

members who are made aware of their social iden- group type that may have consequences for the
tity should perceive members as interchangeable longevity of the group. These researchers found
and homogeneous. Hence, within common-identity that women were attached to groups in which they
groups, members should demonstrate a preference felt close to the other members, while men rated
for resources to be distributed equally among all groups as important when they were attached to
members. In contrast, people in common-bond individual members and the group as a whole. If
groups, who are attached to individuals within the the common bonds in a group disappear, the group
group, should prefer equity relationships in which may no longer be valuable for women, whereas the
members receive rewards in proportion to their common identity of the group would allow men to
contributions (i.e., each individual gets what he or remain attracted to it. Thus some men’s groups
she deserves on the basis of his or her inputs). may last longer than women’s groups because of
In a later study, Sonja Utz and Kai Sassenberg the greater importance of group identity.
found that members of common-bond groups
operate according to an egocentric principle of
maximizing their rewards and minimizing their
costs with respect to their own contributions. This Prentice and her colleagues also suggested that
suggests that in common-bond groups, personal members of common-identity and common-bond
goals are more salient than group goals. In con- groups may differ in how they react to conflict.
trast, Utz and Sassenberg found that members of They reasoned that if there is internal conflict
common-identity groups operate according to an within the group, common-identity groups might
altruistic principle of maximizing the group’s be less affected by it than common-bond groups
rewards and minimizing its costs through their because the former are not dependent on harmony
own contributions. Thus, consistent with a social among members. Conflict within common-bond
identity perspective, these results suggest that in groups, however, can have serious implications for
common-identity groups, group goals are more the group’s existence. But common-bond groups
salient than personal goals because the group pro- are not always damaged by internal conflict.
vides an important source of identity. Prentice and her colleagues suggested that external
threat to a common-bond group may encourage
members to come together for a common cause
and thereby transform the group into a common-
Another difference between common-identity identity group. Indeed, in field studies, Simon
and common-bond groups is related to the longev- Bernd and his colleagues found that people with a
ity of the group. Common-identity groups should common identity are more likely to participate in
last longer than common-bond groups because group activities, such as collective action, and less
members’ commitment to the former is determined likely to leave the group than are members of
by their attraction to the group’s identity and groups that are based on mutual attraction among
norms rather than to their interpersonal relation- members.
ships with other members. Commitment to
common-identity groups should be stronger than
Group Norms
commitment to common-bond groups because
identity and norms tend to fluctuate very little, Research on chat rooms by Sassenberg and his
whereas interpersonal relationships frequently colleagues elucidated some differences in the pre-
change and even dissolve, and members may leave cursors of norm adherence within common-identity
the group. Using observations of real-world groups, and common-bond groups. This work demon-
researchers have found that lasting communities strated that individuals in common-identity groups
have strong group identities and discourage mem- exhibit higher levels of identification with their
bers from having strong interpersonal relation- chat rooms than do individuals in common-bond
ships that could negatively affect allegiance to their groups. Moreover, identification predicted indi-
group. For example, Elizabeth Seeley and her col- viduals’ adoption and expression of group norms,
leagues found sex differences in preferences for implying that differences in common-identity and
Common Ingroup Identity Model 119

common-bond group members’ adherence to group the group identity) and demonstrate important dif-
norms are explainable by identification with the ferences in processes that occur within these two
group. This research demonstrates the importance types of groups.
of identification processes in adopting and express-
ing group norms. Identification with the group Amber M. Gaffney and Namrata Mahajan
appears to be a key component in understanding See also Collectivism/Individualism; Self-Categorization
when and why members of these two types of Theory; Social Identity Model of Deindividuation
groups adopt group norms. Further, identification Effects; Social Identity Theory
processes may differ due to the reasons people join
groups, affecting a variety of group behaviors and
group norms. Further Readings
Prentice, D. A., Miller, D. T., & Lightdale, J. R. (1994).
Impression Formation Asymmetries in attachments to groups and to their
members: Distinguishing between common-identity
How group members form impressions of them- and common-bond groups. Personality and Social
selves and others is another distinguishing feature of Psychology Bulletin, 20, 484–493.
common-bond and common-identity groups. Sassenberg, K. (2002). Common bond and common
Related to social identity theory, self-categorization identity groups on the internet: Attachment and
theory posits that people’s tendency to categorize normative behavior in on-topic and off-topic chats.
themselves as group members is influenced by the Group Dynamics: Theory, Research, and Practice,
social context (e.g., the extent to which their group 6, 27–37.
identity is made salient through comparison with a Spears, R., Postmes, L. M., & Watt, S. E. (2001).
relevant outgroup). According to self-categorization A SIDE view of social influence. In J. Forgas &
theory, when categorized as members of a group, K. D. Williams (Eds.), Social influence processes:
people see themselves in terms of their group’s Direct and indirect influence (pp. 331–350).
shared attributes. Accordingly, members of com- New York: Psychology Press.
mon-identity groups should describe themselves in Utz, S., & Sassenberg, K. (2002). Distributive justice in
terms of their similarity to other group members common-bond and common-identity groups. Group
and to shared group attributes. This should not be Processes & Intergroup Relations, 5, 151–162.
true for members of common-bond groups, which
emphasize members’ personal identities rather than
their shared identity. Russell Spears and his col-
leagues developed the social identity model of dein- Common Ingroup
dividuation effects, or SIDE model, which predicts
that because common-bond groups are based on Identity Model
mutual attraction of members, people in these
groups should be concerned with distinguishing The common ingroup identity model represents a
themselves from other members. These differing strategy for reducing prejudice that assumes that
motivations, in turn, should produce differences in intergroup biases are rooted in fundamental, nor-
how members of common-identity and common- mal psychological processes, particularly in the
bond groups form impressions of other members. universal tendency to simplify a complex environ-
More specifically, while members of common- ment by classifying objects and people into groups
identity groups use group attributes and the group or categories. This process of categorization often
identity to form impressions of one another, mem- occurs spontaneously on the basis of physical
bers of common-bond groups use information similarity, proximity, or shared fate. When people
about individual members to form impressions. or objects are categorized into groups, actual dif-
Studies of common-identity and common-bond ferences between members of the same category
groups clarify two differing perspectives on the tend to be perceptually minimized, and differences
sources of people’s attachment to groups (i.e., attach- between members of different groups become
ment to individual members versus attachment to exaggerated and overgeneralized.
120 Common Ingroup Identity Model

Social Categorization and Bias forces that drive ingroup favoritism to increase
positive attitudes toward others who were previ-
Social categorization, the categorization of people
ously seen primarily in terms of their outgroup
into different groups, has another unique feature.
When a person categorizes others into groups,
The development of a common ingroup identity
these categories are fundamentally differentiated
does not necessarily always require each group to
between groups to which the perceiver belongs
forsake its less inclusive group identity completely.
(ingroups) and groups to which the perceiver does
It is possible for members to conceive of two
not belong (outgroups). Because people derive
groups (for example, parents and children) as dis-
their self-esteem in part from the prestige of groups
tinct units within the context of a superordinate
to which they belong, members are motivated to
(i.e., family) identity (as “subgroups within one
regard their ingroup in a positive light compared
group” or a “dual identity” representation). When
to other groups. Upon social categorization, people
group identities and the associated cultural values
typically express more positive beliefs, feelings,
are central to members’ self-identification or when
and behaviors toward ingroup members than
they are associated with high status or highly visi-
toward outgroup members. Hence, social categori-
ble cues to group membership, it may be very
zation can enable ingroup favoritism to service
threatening for people to be asked to relinquish
ego-enhancing motivations as long as the situa-
these group identities or, as perceivers, to be
tional context reinforces the importance of the
“color-blind.” Indeed, demands to forsake ethnic
categorical distinction between the groups.
or racial group identities to adopt a color-blind
Although a preferential ingroup orientation can
ideology would likely arouse strong reactance and
evolve into a more destructive, anti-outgroup atti-
result in especially poor intergroup relations.
tude (i.e., prejudice), the mechanisms of ingroup
There is support, however, for the idea that if
favoritism can also provide a means to reduce
people continued to regard themselves as members
prejudice and discrimination. This latter assertion is
of different groups—but all playing on the same
the essence of the common ingroup identity model.
team (i.e., part of the same inclusive entity), inter-
group relations between the “subgroups” would
be more positive than if members only considered
Social Categorization and Recategorization
themselves “separate groups.”
The common ingroup identity model recognizes The common ingroup identity model acknowl-
the fluidity of social categorization processes and edges that other approaches to addressing social
the reality that people belong to a variety of groups categorization can also affect intergroup relations.
that are hierarchically organized in terms of inclu- For instance, decategorizing people—seeing them
siveness. For example, people are members of in an individuated or personalized way instead of
families, neighborhoods, cities, regions, and as members of a group—can also reduce inter-
nations. Modifying goals, motives, expectations, group bias. However, whereas recategorization of
or factors within the immediate situation provides people within a common ingroup identity reduces
the opportunity to shift the level of category inclu- intergroup bias by enhancing evaluations of origi-
siveness that will be dominant. This fluidity of nal outgroup members, decategorization reduces
social categorization, and consequently, the salience intergroup bias by reducing positive regard to
of a particular social identity, are important original ingroup members through eliminating the
because of the implications for altering the way forces of social categorization that promote ingroup
people think about others in terms of their ingroup favoritism in this circumstance.
or outgroup membership, and therefore, how
positively they feel about them. Specifically, the
Functional Relations,
common ingroup identity model proposes that
Categorization, and Bias
inducing people to recategorize ingroup and out-
group members within a common boundary inclu- The common ingroup identity model recognizes
sive of both groups (e.g., a school, a city, a nation, that although social categorization can be suffi-
humanity) redirects the cognitive and motivational cient to produce intergroup bias, intergroup threat,
Common Ingroup Identity Model 121

competition, and conflict can further contribute to This process was particularly exemplified when
negative relations between groups. In particular, one of the boys exclaimed after the achievement of
realistic group conflict theory argues that the moving the truck, “We won the tug-of-war against
major cause of intergroup bias and prejudice the truck.” Subsequent laboratory studies offer
involves zero-sum competition between the groups support for this interpretation of how cooperation
over valuable resources. More broadly, whether reduces intergroup biases by producing recategori-
relations between groups are conflictual or harmo- zation within a common ingroup. In addition,
nious is determined by the functional relations other bias-reducing interventions that rely on
between the groups. When the relationship is pri- related principles of the functional approach (e.g.,
marily competitive, intergroup relations would be techniques such as the jigsaw classroom and some
expected to be prejudice-ridden and conflictual, forms of cooperative learning) may operate psy-
whereas when the relations between groups are chologically through common identity as well as
primarily cooperative, relations between the groups through functional relations directly.
would be harmonious.
In the classic Robbers Cave study, for example,
Empirical Support for the
Sherif and his colleagues studied 12-year-old, middle-
Common Ingroup Identity Model
class boys at a 3-week summer camp in an experi-
ment about the creation and reduction of intergroup Formally, the common ingroup identity model
bias and conflict. These boys were initially assigned identifies potential antecedents and outcomes of
to two groups. To permit time for group formation recategorization, as well as mediating processes.
within each group (e.g., norms and a leadership The general framework specifies the causes and
structure), these groups were kept completely apart consequences of a common ingroup identity.
for the first week. During the second week, the Specifically, different types of intergroup interde-
investigators introduced intense competitive rela- pendence (e.g., cooperative or competitive rela-
tions between the groups in the form of repeated tions) and cognitive, perceptual, linguistic, affective,
competitive athletic activities—centering on tug-of- and environmental factors can alter individuals’
war, baseball, and touch football—in which only perceptions of the different groups. These resulting
members of the winning group received rewards. As cognitive representations (i.e., one group, two-
expected by the functional relations account, the subgroups within one group—a dual identity, two
introduction of competitive activities generated separate groups, or separate individuals) are then
derogatory stereotypes and very physical, hostile proposed to result in specific cognitive, affective,
conflict between these groups. In the third week, and overt behavioral consequences involving inter-
only after the functional relations between the group attitudes. Thus, the causal factors that
groups were altered by introducing a series of goals include features specified by contact theory (i.e.,
(e.g., finding leaks to the camp’s water supply, col- cooperation, equal status, opportunity for self-
lecting money to watch a popular movie, and mov- revealing interaction, and egalitarian norms sup-
ing a stalled truck carrying lunch up a hill to the ported by local authorities) are proposed to influence
dining area)—goals that could not be achieved with- members’ cognitive collective representations of the
out cooperating with each other—did the relations memberships that then, at least in part, mediate the
between the groups become more harmonious. relationship between the causal factors and inter-
Although there was no formal assessment of the group attitudes (i.e., feelings, beliefs, and behav-
psychological processes involved in just how coop- iors). In addition, a common ingroup identity may
eration between these groups reduced intergroup be achieved by increasing the salience of existing
animosity, from the perspective of the common common superordinate memberships or by intro-
ingroup identity model, intergroup cooperation ducing factors (e.g., common goals or fate) that are
likely reduced intergroup conflict because working perceived to be shared by the memberships.
together toward their common goal changed the The common ingroup identity model has
boys’ perceptions of their intergroup boundaries received considerable empirical support. In an
from an “us” and “them” orientation to a more early exploration of the causal role of common
inclusive “we” (i.e., common ingroup) orientation. ingroup identity in reducing bias, members of two
122 Common Ingroup Identity Model

separate laboratory-formed groups of American confidence in suggestions for innovation, and even
college students who were homogeneous with forgiveness. In addition, creating a common ingroup
respect to sex and racial composition were induced identity has been found to increase positive forms of
through various structural interventions (e.g., seat- behavior, such as self-disclosure and helping, across
ing arrangement, dress) to maintain their original original group lines and to be effective for improving
group identities (i.e., conceive of themselves as dif- relations between groups, such as ethnic and racial
ferent groups), recategorize themselves as one groups, that have extended histories of intergroup
group, or decategorize themselves (i.e., to conceive bias. In a particularly dramatic example, emphasiz-
of themselves as separate individuals). The manip- ing common humanity increased Jewish students’
ulations to encourage recategorization and decat- willingness to forgive Germans for the Holocaust.
egorization each reduced bias, and as predicted,
did so in different ways. Specifically, recategorizing
the memberships into one group reduced bias by
increasing the attractiveness of former outgroup Overall, the research evidence reveals consistent
members, whereas decategorizing members of support for the key principle of the common ingroup
these groups reduced bias by decreasing the attrac- identity model: Successfully inducing ingroup and
tiveness of former ingroup members. Identical outgroup members to adopt a more inclusive, one-
findings were obtained when the recategorization group representation inclusive of both groups
and decategorization manipulations were imple- reduces the groups’ bias toward one another.
mented with Black and White Portuguese children, Furthermore, this fundamental principle has been
who have different social status and a shared his- supported across studies that used a variety of
tory of intergroup conflict. methodological approaches, involving participants
Also, manipulations that have been demon- of different ages, races, and nationalities. Conse­
strated to reduce prejudice, by inducing cooperative quently, this model can inform the development of
interaction or structuring positive intergroup con- powerful prejudice-reducing interventions.
tact in ways specified by contact theory, have been
shown in laboratory research to reduce bias by Samuel L. Gaertner and John F. Dovidio
changing intergroup cognitive representations from See also Categorization; Cooperation and Competition;
two separate groups to one group. Specifically, Ingroup Allocation Bias; Jigsaw Classroom Technique;
intergroup cooperation leads to stronger inclusive, Prejudice; Sherif, Muzafer; Stereotyping
one-group representations, which in turn predicts
more favorable outgroup member evaluations. In
addition, investigations across a variety of inter- Further Readings
group settings offer converging support for the idea
Dovidio, J. F., Gaertner, S. L., Validzic, A., Matoka, K.,
that the features specified by contact theory (i.e.,
Johnson, B., & Frazier, S. (1997). Extending the
cooperation, equal status, opportunity for self-
benefits of re-categorization: Evaluations, self-
revealing interaction, and egalitarian norms sup-
disclosure and helping. Journal of Experimental Social
ported by local authorities), reduce intergroup bias
Psychology, 33, 401–420.
because they transform members’ representations Gaertner, S. L., & Dovidio, J. F. (2000). Reducing
of the memberships from separate groups to a sin- intergroup bias: The common ingroup identity model.
gle, more inclusive group. Participants in these stud- Philadelphia: Psychology Press.
ies included students attending a multi-ethnic high Gaertner, S. L., Mann, J. A., Dovidio, J. F., Murrell, A.
school, banking executives who had experienced a J., & Pomare, M. (1990). How does cooperation
corporate merger involving a wide variety of banks reduce intergroup bias? Journal of Personality and
across the United States, and college students from Social Psychology, 59, 692–704.
blended families with households composed of two Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., &
formerly separate families trying to unite into one. Sherif, C. W. (1988). The Robbers Cave experiment:
Common ingroup identity not only increases Intergroup conflict and cooperation. Hanover,
positive evaluations of others, but also increases NH: Wesleyan University Press. (Original work
cross-group friendship selection, helpfulness, trust, published 1961)
Common Knowledge Effect 123

reach a different decision depending on whether

Common Knowledge Effect important information is initially shared or
unshared. This difference is especially problematic
The common knowledge effect describes the when the shared information favors an incorrect
impact on group decision making of whether decision alternative and the unshared information
knowledge relevant to a decision is shared by all favors a different, correct decision alternative. In
group members prior to discussion. Specifically, such a case, known as a hidden profile, groups are
laboratory studies have shown that information unlikely to discover and decide on the correct
known by everyone prior to discussion has a more alternative. Instead, even after discussion, groups
powerful influence on decisions than information are likely to decide on the incorrect alternative that
not shared by everyone. The common knowledge was favored by the information shared by all of
effect demonstrates that an irrelevant factor—the the members prior to discussion.
number of members who know a particular piece
of information—can affect group decisions. If a
piece of unshared information is crucial to making
Theoretical Explanation
a correct decision, the result may be an incorrect Why does information have more influence on the
decision. This entry looks at the common knowl- group decision when all members know it prior to
edge effect and some possible explanations for the group discussion than when only some mem-
such outcomes, then discusses what research has bers know it? One explanation focuses on the con-
shown about promoting better decisions. tent of group discussion, that is, which information
group members actually tend to discuss. Arguably,
discussion of unshared information is more impor-
The Decision-Making Process
tant than discussion of shared information because
When a group comes together to make a decision, unshared information may alter the opinions of
the members must combine and process the infor- members who lacked the information prior to the
mation that is relevant to the decision. Typically, the discussion. Discussion of initially shared informa-
group has a discussion during which members men- tion is less important because all of the group’s
tion information that they believe is relevant to the members have had the opportunity to consider that
decision. The information that the members bring information prior to the group discussion.
to the discussion may initially be known by all of In decisions involving both shared and unshared
the members (shared) or known by only one or a information, however, group discussion tends to
few members (unshared). A primary goal of group focus on the shared, rather than the unshared,
discussion is to inform members of information that information. A particular piece of information is
they did not know prior to the discussion, that is, to more likely to be mentioned if all group members
discuss the information that was initially unshared. know about it prior to the discussion than if only
Ideally, in the end, the group’s decision should one or a few group members know about it prior
reflect all of the relevant information, whether that to the discussion. Initially unshared information is
information was initially shared or unshared. often not mentioned at all. In hidden profile tasks,
However, research shows that information that for example, groups tend to discuss the shared
was shared prior to the discussion tends to influ- information, which favors the incorrect alterna-
ence group decisions more than information that tive, whereas they often fail to mention the
was unshared prior to the discussion. Thus, the unshared information, which would allow the
same fact will likely have more influence on the group to discover the correct alternative. Conse­
group’s final decision if it is initially known by all quently, if the group bases its decision on the infor-
of the group’s members than if it is initially known mation that is actually discussed, that decision will
by only one group member, regardless of how per- be affected more by the initially shared informa-
tinent the fact is to the decision at hand. This is tion because the group discusses more shared than
known as the common knowledge effect. unshared information.
The common knowledge effect can negatively But the tendency of groups to discuss more
affect the quality of group decisions. A group may shared than unshared information does not fully
124 Common Knowledge Effect

explain the relatively strong influence of shared shared information may bolster members’ confi-
information on group decisions. Even when the dence in their initial opinions. Members may also
group discusses unshared information, that infor- interpret new information to be consistent with
mation does not necessarily affect the group’s deci- their opinions.
sion. For example, in hidden profile tasks, groups In addition, unshared information cannot be
that discuss more unshared information are not confirmed by other members, so members who
always more likely to discover the correct alterna- hear the new information may doubt its validity.
tive. Although the tendency of groups to discuss Individuals who discuss shared information are
more shared than unshared information is an evaluated more positively by other group mem-
important factor in the common knowledge effect, bers, and members who know a large amount of
it is not the only factor. shared information tend to have more influence on
The group members’ prediscussion opinions are the group decision. Moreover, full consideration of
another factor in the common knowledge effect. In new information requires time and cognitive effort.
addition to affecting a group’s decision directly, Groups may be motivated to process unshared
through group discussion, information also affects information superficially to reach consensus more
a group’s decision indirectly, by affecting the initial quickly, particularly if they do not believe that
opinions of the group’s members. When members more effortful processing of that information will
first consider decision-relevant information, prior lead to a significantly improved group decision.
to any discussion, they tend to form opinions Thus, the common knowledge effect seems to
about the correct decision. During group discus- be explained primarily by two phenomena: the
sion, a member learns the opinions of the other tendency to discuss more shared than unshared
members. In fact, the members’ opinions are often information and the tendency to base the ultimate
the first thing the group discusses, prior to the dis- decision heavily on the group members’ initial
cussion of any specific information. Those initial opinions, which are based on the information
opinions, in turn, often have a strong influence on members have prior to discussion. Groups do,
the group’s final decision. however, sometimes overcome the common knowl-
Groups often appear to use relatively simple edge effect. To do so, they must both discuss and
methods of combining their members’ initial opin- carefully consider their unshared information.
ions into a group decision, such as averaging those
opinions or choosing the alternative that is initially
Fostering Better Decisions
favored by a majority of the members. Initially
shared information can affect all of the members’ Research has identified a number of conditions that
prediscussion opinions. Initially unshared informa- increase the likelihood that groups will discuss and
tion, however, can only affect one member’s opin- be influenced by information that is unshared prior
ion prior to the group discussion. When the group to the discussion. For example, groups are more
combines all of its members’ opinions into a group likely to decide on the correct alternative in a hidden
decision, the shared information will tend to have profile task if they rank order all of the alternatives,
more influence on the group decision because it has rather than simply choosing the single alternative
influenced the opinions of more group members. that they think is best. Rank ordering forces the
Why do groups base their decisions primarily group to consider all of the alternatives, which may
on their members’ initial opinions, even when lead to the discussion of unshared positive informa-
unshared information is revealed during the group tion about an initially rejected alternative.
discussion? After all, when several members learn Groups that believe they are making a decision
new facts, they should realize that their initial that has an objective correct answer tend to discuss
opinions were based on an incomplete subset of more unshared information and make better deci-
the relevant information and change their opin- sions than groups that believe they are making a
ions. However, when unshared information is dis- decision that does not have a single correct answer.
cussed, it tends to come up late in the discussion, Groups whose members initially disagree about
after the group has already discussed some amount the correct decision are also more likely to discuss
of shared information. That early discussion of and use unshared information. All of these factors
Commons Dilemma 125

encourage a group to discuss more information,

including information that was unshared prior to Commons Dilemma
discussion, and to base its final decision on all of
the relevant information, rather than settling too The commons dilemma is a specific class of social
quickly on a decision that was favored by the dilemma in which people’s short-term selfish
group members’ initial opinions. interests are at odds with long-term group inter-
But laboratory research may overestimate the ests. The commons dilemma, also known as the
severity of the problem posed by the common common pool resource (CPR), the resource
knowledge effect. In everyday decisions, shared dilemma, or the take-some dilemma, was inspired
and unshared information may generally favor the by the metaphor of the Tragedy of the Commons
same alternative, or the most relevant information that Garrett Hardin discussed in his seminal
might typically be shared. In such cases, the addi- 1968 Science article. This story describes a group
tional time and effort that would be required to of herders with open access to a common parcel
discuss and consider unshared information might of land on which to let their cows graze. It is
not result in a better decision. Moreover, real in each herder’s interest to put as many cows
groups may be more motivated and have more time as possible onto the land, even if the commons
than experimental groups to discuss and consider is damaged as a result. The herder receives all
more of the available information before commit- the benefits from the additional cows, but the
ting to a decision. Further research is needed to damage to the commons is shared by the entire
understand the impact of the common knowledge group. Yet if all herders make this individually
effect on important real-world decisions. rational decision, the commons is destroyed and
all will suffer.
Daniel Gigone The commons dilemma stands as a model for a
See also Group Performance; Group Problem Solving and great variety of resource problems in society today,
Decision Making; Hidden Profile Task; Informational such as water, land, fish, and nonrenewable energy
Influence; Normative Influence; Social Decision sources like oil and coal. When water is used at a
Schemes; Socially Shared Cognition higher rate than the reservoirs are replenished, fish
consumption exceeds its reproductive capacity, or
Further Readings oil supplies are exhausted, we face a tragedy of the
Gigone, D., & Hastie, R. (1993). The common In the 1980s, researchers created an experimen-
knowledge effect: Information sampling and group tal game version of the commons dilemma involv-
judgment. Journal of Personality and Social ing a common resource pool (filled with money or
Psychology, 65, 959–974. points that could be converted into money or lot-
Greitemeyer, T., & Schulz-Hardt, S. (2003). Preference- tery tickets) from which a group of individuals
consistent evaluation of information in the hidden
could harvest. If the sum of their harvests per
profile paradigm: Beyond group-level explanations for
round of the game is lower than the replenishment
the dominance of shared information in group
rate, the pool is maintained. The individuals are
decisions. Journal of Personality and Social
each tempted to harvest as much as possible, but if
Psychology, 84, 322–339.
Kerr, N. L., & Tindale, S. R. (2004). Group performance
they do, all suffer and the resource is depleted,
and decision-making. Annual Review of Psychology,
upon which the game ends.
55, 623–655.
Stasser, G., & Titus, W. (1985). Pooling of unshared Factors Promoting Conservation
information in group decision making: Biased in Commons Dilemmas
information sampling during discussion. Journal of
Personality and Social Psychology, 48, 1467–1478. Commons dilemma researchers have studied con-
Wittenbaum, G. M., & Stasser, G. (1996). Management ditions under which groups and communities are
of information in small groups. In J. L. Nye & A. B. likely to under- or overharvest the common
Brower (Eds.), What’s social about social cognition? resource in both the laboratory and field. Research
(pp. 3–28). London: Sage. programs have concentrated on a number of
126 Commons Dilemma

motivational, strategic, and structural factors that pool also increases harvesting. The most likely
might be conducive to commons management. explanation is that people have an optimistic bias.

Motivational Solutions Strategic Solutions

The research shows that some people are more Strategic factors also matter in commons dilem-
motivated than others to manage the common mas. One often-studied strategic factor is the order
resource responsibly. Using the commons dilemma in which people take harvests from the resource. In
game, researchers found that people with prosocial simultaneous play, all people harvest at the same
value orientations harvest less from a resource dur- time, whereas in sequential play people harvest
ing a period of scarcity. Prosocial individuals are from the pool according to a predetermined
also more inclined to engage in sustainable envi- sequence—first, second, third, and so on. There is
ronmental behaviors such as taking public trans- a clear order effect in the latter games: The harvests
portation (instead of the car) and conserving energy of those who come first—the leaders—are higher
and water, as well as to explain their decisions in than the harvest of those coming later—the follow-
terms of environmental impact. ers. The interpretation of this effect is that the first
Motivation to conserve a common resource is players feel entitled to take more. Whereas with
also promoted by people’s group ties. When people simultaneous play, people may adopt an equality
identify with their group, they are more likely to rule, with sequential play, individuals adopt a “first
exercise personal restraint, as well as to compen- come, first served” rule. Another strategic factor is
sate for greedy harvest decisions of ingroup mem- the ability to build up reputations. Research found
bers more than for those of outgroup members. that people take less from the common pool in
Similarly, in the field strongly knit communities are public situations than in anonymous private situa-
usually better at managing resource shortages than tions. Moreover, those who harvest less gain
communities with weak social ties. It might be that greater prestige and influence within their group.
group identity promotes a long-term perspective
on resource management that makes it easier for
Structural Solutions
people to sacrifice their immediate interest on
behalf of their local community. It could also be Much research has focused on when and why
that group identification increases the social inter- people would like to structurally rearrange the
dependencies between community members so commons to prevent a tragedy. Hardin stated in
that they care more for the social rewards and his analysis of the Tragedy of the Commons that
punishments of their community. This needs fur- “freedom in a commons brings ruin to all.” One of
ther investigation. the proposed solutions is to appoint a leader to
The state of the common resource can also regulate access to the commons. Not surprisingly,
shape motivations. One motivational factor is groups are more likely to endorse a leader when a
people’s attributions of the state of the commons. common resource is being depleted and when
Research has manipulated the reasons people were managing a common resource is perceived as a dif-
given for resource overuse. When greedy people ficult task. Interestingly, groups prefer leaders who
were seen as causing the depletion, participants are elected, democratic, and prototypical of the
were greedier than when there was deemed to be a group, and these leader types are more successful
natural cause (like a sudden drought). Resource in enforcing cooperation. There is a general aver-
uncertainty further contributes to overharvesting. sion against autocratic leadership—although it is
In commons dilemmas, uncertainly about the pool quite an effective solution—possibly because of the
size tends to increase individual harvesting and fear of power abuse and corruption.
expectations about how much other people har- Another structural solution is the privatization
vest. When there is uncertainty, people overesti- of the commons, and this has been shown in
mate the size of the resource and perceive greater experimental and field research to be very effec-
variability in how much other people take. Similarly, tive. However, it is difficult to imagine how com-
uncertainty about the replenishment rate of the mon movable resources such as fish, water, and
Communication Networks 127

clean air can be privatized. Privatization also raises See also Cooperation and Competiton; Leadership;
concerns about social justice, as not everyone may Prisoner’s Dilemma; Social Dilemmas; Social Identity
be able to get an equal share. Finally, privatization Theory; Trust
might erode people’s personal and social motiva-
tions to cooperate in preserving a resource. Further Readings
The provision of rewards and punishments
might also be effective in preserving common Foddy, M., Smithson, M., Schneider, S., & Hogg, M.
resources. Selective punishments for overuse can (1999). Resolving social dilemmas. Philadelphia:
be effective in promoting domestic water and Psychology Press.
energy conservation, for instance, through install- Hardin, G. (1968). Tragedy of the commons. Science,
ing water and electricity meters in houses. Selective 162, 1243–1248.
rewards also work, provided that they are open to Messick, D. M., Wilke, H. A. M., Brewer, M. B.,
Kramer, R. M., Zemke, P. E., & Lui, L. (1983).
everyone. An experimental carpool lane in the
Individual adaptations and structural change as
Netherlands failed because car commuters did not
solutions to social dilemmas. Journal of Personality
feel they were able to organize a carpool. Hence,
and Social Psychology, 44, 294–309.
they showed a reaction against this pro-environment
Ostrom, E. (1990). Governing the commons: The
intervention. evolution of institutions for collective action.
There has been much field research on commons Cambridge, UK: Cambridge University Press.
dilemmas that has combined solutions obtained in van Vugt, M. (2001). Community identification
experimental research. The seminal work of Elinor moderating the impact of financial incentives in a
Ostrom and her colleagues is worth mentioning. natural social dilemma. Personality and Social
They looked at how real-world communities man- Psychology Bulletin, 27, 1440–1449.
age communal resources such as fisheries, land van Vugt, M., van Lange, P. A. M., Meertens, R. M., &
irrigation systems, and farmlands and came up Joireman, J. A. (1996). Why structural solutions to
with a number of factors conducive to successful social dilemmas might fail: A field experiment on the
resource management. One factor is the resource first carpool priority lane in Europe. Social Psychology
itself. Resources with definable boundaries (e.g., Quarterly, 59, 364–374.
land) can be preserved much more easily than can
resources without such boundaries. A second fac-
tor is resource dependence. There must be a per-
ceptible threat of resource depletion, and it must be Communication Networks
difficult to find substitutes. The third is the pres-
ence of a community. Small and stable populations The term communication networks describes the
with a thick social network and social norms pro- structure and flow of communication and infor-
moting conservation do best. A final condition is mation between individuals within a group. Within
that there are appropriate community-based rules many groups (e.g., a typical office), formal and
and procedures in place with built-in incentives for informal communication patterns are often deter-
responsible use and punishments for overuse. mined in a top-down, hierarchical fashion, where
members direct communication to others at the
Conclusion same level or below but not above. Much of the
research on the effects of communication networks
As populations grow and resources become scarcer,
was conducted in the 1950s; after a long hiatus,
there is a need for policies to avoid commons trag-
researchers are again exploring the effects of com-
edies. It is encouraging that commons dilemma
munication networks. The primary foci of classic
research is increasingly applied to local and global
research on communication networks had been to
environmental problems. The emphasis is shifting
measure how network structure affects informa-
from pure laboratory research toward research
tion flow, and how position within the networks
testing combinations of motivational, strategic,
may affect an individual’s status within the group.
and structural solutions.
More recent work has focused on how communi-
Mark van Vugt cation networks affect group-level properties. This
128 Communication Networks

entry discusses both classic and more recent communication could take place within the struc-
approaches to the study of communication net- ture of a wheel, with one central member (the hub)
works within groups. through which all communications must pass. One
consistent result in such early research was that the
centralization of a communication network (i.e.,
Why Study Communication Networks?
the degree to which some members of the group
The overwhelming majority of research on group had more communication partners than others;
dynamics has studied interacting groups in which e.g., the wheel is more centralized than the circle
communications from each member are sent to the network) was a strong predictor of the efficiency
entire group, with no constraints on communica- of problem solving; that is, the more centralized a
tion (a prototypical example would be a jury sit- network, the more efficient the group was at solv-
ting around a table). This type of communication ing problems. Later research by Marvin E. Shaw
network is common in many small real-world qualified this finding to show that centralized
groups, such as juries; however, it is not common groups solve relatively simple problems better than
in larger groups and institutions. In larger groups decentralized groups, but when problems become
communication is likely asymmetrical, where dif- more complex, centralization can hamper problem
ferent individuals receive and transmit information solving. Another consistent finding of this research
heterogeneously across the entire group. First, was that more centralized group members are
many groups do not necessarily meet in a specific more satisfied with the group process than are
location at the same time, but rather consist of a more peripheral members.
number of asynchronous communications within
subsets of group members. Second, as group size
Recent Advances
increases, there may be evolutionary constraints
on the optimal number of group members per- Most of the research on small group decision mak-
forming any given task. Robin Dunbar has specu- ing in the 1950s through the 1980s was conducted
lated that as groups increase in size, the brain’s in groups with symmetrical communication net-
processing capacity may constrain the number of works, in which each member’s communication
individuals with whom one can optimally commu- was received by the entire group. Therefore, there
nicate. This evolved constraint likely leads to were few advances within the field of communica-
recurrent and structural patterns across all groups tion networks during this time period. Bibb Latané
in terms of communication networks. Therefore, it and his colleagues revived interest in communica-
is important to study the development and effects tion networks in the late 1980s, pointing out that
of communication networks within groups, espe- in large groups, individual group members cannot
cially among larger groups (for example, within necessarily communicate with the entire group at
organizations). the same time. Latané’s dynamic social impact
theory includes a principle of immediacy, which
assumes that influence between any two members
Classic Research
in a group is predicted by the likelihood that they
The first systematic research on communication can easily share communications.
networks was conducted by Harold J. Leavitt of Latané and his colleagues tested the implications
MIT and Alex Bavelas of Bell Laboratories in the of his dynamic social impact theory by conducting
1950s. This work was stimulated by formal math- computer simulations, in which agents were situated
ematical models derived from graph theory. By in a two-dimensional space, where the strongest
placing partitions between participants seated at a influence between agents occurred with immediate
table, Leavitt and Bavelas manipulated communi- neighbors. Each agent was randomly assigned a
cation structures within groups of varying sizes. binary opinion on an issue (e.g., Republican vs.
For example, in a five-person group, members Democrat). Following from other assumptions of
could communicate within a circle structure, in dynamic social impact theory, individual agents in
which each person can only share messages with the simulations also varied in strength (i.e., some
those on either side of him or her. Alternatively, were more influential than others), and agents were
Communication Networks 129

influenced by the number of other agents sharing or groups. Recent empirical research has provided
opposing their preferences. further support for the “six degrees of separation”
After simulating a number of rounds of com- idea popularized by Milgram.
munication, in which each agent’s opinion was Another solution to the small-world problem
compared to the opinions of fellow agents, the involving computer simulation with communication
researchers found that opinions either were main- networks has been provided by Albert-Laszlo
tained or changed as a function of the strength, Barabasi and his colleagues. Barabasi has shown that
immediacy, and number of other agents, and sev- communication networks within large groups share
eral group-level phenomena emerged. Opinions in properties with what are known as “scale-free” net-
the group typically consolidated (or polarized); works. In a scale-free network, some individuals
that is, whichever opinion was most commonly within the larger group have many more communi-
held within the group became even more common cation partners than others; in the terms of earlier
after simulated communication. Because commu- work on communication networks, such members
nication networks constrained communication, can be said to be more centralized. Scale-free net-
opinions also became regionally clustered, such works are another way to solve the small-world
that agents shared opinions with other agents problem; when a small number of members within a
who were physically close to them in the two- large group have a large number of communication
dimensional space, and who were thus able to partners, it takes a relatively small number of links
exert greater influence over them. Latané and his to join any two randomly chosen group members.
colleagues then tested whether these phenomena
that emerged in simulations also occurred within
actual groups discussing issues in communication
networks configured via e-mail exchanges. Both The field of communication networks, a classic
group-level phenomena observed in the computer area of research within group dynamics, recently
simulations—consolidation and clustering—also has been reactivated, partly as a function of
emerged within groups of people discussing issues. advances in computer science. The fact that many
Subsequently, Latané and his colleagues have group decisions are made by subsets of members,
shown that the “geometry” of communication without all group members present at any given
networks—how they are organized—can deter- time, creates a need for more research in this area.
mine the extent to which groups’ opinions will
consolidate and cluster as a function of communi- Martin J. Bourgeois and Nicholas G. Schwab
cation. For example, as communication networks See also Culture; Dynamical Systems Approach; Group
become more “clumpy” or hierarchical, consolida- Polarization; Social Impact Theory; Social Networks
tion and clustering of opinions tend to increase.
Mathematicians and physicists have also recently
shown interest in using computer simulation to Further Readings
test some of the implications of constrained com-
Barabasi, A.-L. (2003). Linked. New York: Penguin.
munication networks within large groups. One
Bavelas, A. (1950). Communication patterns in task-
recent line of evidence was provided by Duncan oriented groups. Journal of the Acoustical Society of
Watts and his colleagues. Watts used computer America, 22, 725–730.
simulation to solve the “small-world problem” Collins, B. E., & Raven, B. H. (1969). Group structure:
posited by Stanley Milgram: If most people com- Attraction, coalitions, communication, and power. In
municate with others within local networks (as E. Artonson & G. Lindzey (Eds.), Handbook of social
social impact theory assumes), how can any two psychology. Reading, MA: Addison-Wesley.
randomly chosen people within the larger group be Latané, B., & Bourgeois, M. J. (2001). Dynamic social
connected by a surprisingly small number of links impact and the consolidation, clustering, correlation,
(“six degrees of separation”)? Watts showed that and continuing diversity of culture. In M. A. Hogg &
simply adding a small number of random commu- R. S. Tindale (Eds.), Blackwell handbook of social
nication links between people can create such psychology: Group processes (pp. 235–258). Oxford,
small world networks even within extremely large UK: Blackwell.
130 Compliance

Leavitt, H. J. (1951). Some effects of certain she may not agree with her supervisor, she is likely
communication patterns on group performance. Journal to comply because the supervisor controls impor-
of Abnormal and Social Psychology, 46, 38–50. tant rewards (e.g., salary) and punishments (e.g.,
Nowak, A., Szamrej, J., & Latané, B. (1990). From the power to have her fired) and is able to monitor
private attitude to public opinion: A dynamic theory her response.
of social impact. Psychological Review, 9, 362–376. Groups may represent another source of com-
Shaw, M. E. (1964). Communication networks. Advances pliance, such as when a holdout juror yields to
in Experimental Social Psychology, 1, 111–147. group pressure despite privately disagreeing with
Watts, D. J. (2003). Six degrees: The science of a
the verdict. In this instance, “going along to get
connected age. New York: Norton.
along” reflects normative influence. In contrast, if
the individual believes that the verdict is correct,
the influence would reflect conversion because his
private beliefs would be consistent with his public
Compliance behavior. The influence in this case would reflect
informational influence.
Compliance is one of a variety of ways in which
people can be influenced by others. Two meanings
have been attached to the term. Investigators such Why Is It Important to Distinguish
as Leon Festinger, Herbert Kelman, and Paul Nail Between Compliance and Private Acceptance?
define compliance as a change in public behavior By knowing the basis for someone’s acceptance
without private acceptance. (As discussed below, of influence—compliance versus conversion—one
when others influence both public behavior and can better predict when the response is likely to be
private acceptance, the form of social influence is performed. In the case of compliance, the influenc-
called conversion or internalization.) More recently, ing agent (e.g., a group) must retain control of
Robert Cialdini has offered a second definition of resources valued by the target person and be able
compliance that ignores the distinction between to monitor whether or not the person complies.
public and private. He defines compliance as However, if the person privately accepts the influ-
acquiescence to a request. Each definition has gen- encing agent’s position, the agent’s ability to
erated a different set of questions and research reward or punish the person and to maintain sur-
findings. This entry examines both definitions. veillance are unnecessary for the response to be
performed. The person will perform the behavior
Compliance as a Change in Public Behavior because it is internalized, that is, it is consistent
with her private beliefs.
Research on compliance as a change in public
behavior has addressed such questions as (a) What
causes people to comply? (b) Why is it important Can Compliance Lead to Private Acceptance?
to distinguish between compliance and private
Numerous studies have shown that compliance
acceptance? and (c) Can compliance lead to private
can lead to private acceptance via a variety of
mechanisms. Perhaps the best documented mecha-
nism is the one proposed by Leon Festinger in his
What Causes People to Comply?
theory of cognitive dissonance. According to this
Research shows that for people to comply, two theory, inconsistency between public behavior and
conditions must be present. First, people must private beliefs produces a tension known as cogni-
believe that the influencing agent has the ability to tive dissonance. This tension motivates people to
reward them for compliance or punish them for reduce their discomfort by changing their private
noncompliance. And, second, they must believe beliefs to be more consistent with their public
that the influencing agent has the ability to moni- behavior. Studies show that belief change is most
tor their compliance or failure to comply. As an likely to occur when there is minimal pressure to
example, consider a worker’s decision to comply comply and the compliance is public. When too
with her supervisor’s request to work faster. While much force is used to gain compliance, there will
Compliance 131

be less cognitive dissonance and therefore less form accurate perceptions of reality, (2) to develop
pressure to change private beliefs to be consistent and preserve meaningful social relationships, and
with public behavior. (3) to preserve a positive self-concept.
Another theory that can account for the impact
of compliance on private beliefs is Daryl Bem’s
Motivation to Form Accurate
self-perception theory. According to this theory,
Perceptions of Reality
people use their behavior to infer their private
beliefs, particularly when they are uncertain about People want to be effective decision makers. To
these beliefs. For example, when a group subtly meet this goal, they need to have accurate percep-
induces a member to contribute to a charitable tions of reality because inaccurate perceptions are
cause and the individual is later questioned about likely to produce poor decisions. Capitalizing on
his feelings about the charity, he might reason that, this motivation, compliance professionals have
because he made the donation, he must have posi- developed several effective strategies for gaining
tive beliefs about the charity. compliance. One such strategy involves presenting
A third explanation for how compliance pro- oneself as an authority or expert on the subject
duces private acceptance relates to the finding that matter under consideration. The strategy relies on
compliance can provide people with the opportu- targets’ tendency to rely on heuristics, or simple
nity to gain new information that could change rules of thumb, to make decisions. In this particu-
their private beliefs. Evidence for this mechanism lar case, target persons would be relying on the
can be found in studies of role playing where indi- heuristic “experts know what’s best.” Examples of
viduals were induced to write an essay favoring a this strategy include an advertisement showing a
position counter to their beliefs. Results show that NASCAR driver recommending a brand of motor
this experience modifies individuals’ private beliefs oil and a waiter informing dinner guests that coq
in the direction of the position taken in the essay, au vin is his favorite item on the menu.
because writing the essay forces them to acquire To increase the perceived value of their product,
new information that can potentially change their compliance professionals also make use of the
beliefs. In sum, then, there are multiple ways in notion of scarcity. They attempt to capitalize on the
which compliance to pressure can ultimately lead belief in many people’s minds that the quality of a
to private acceptance. product is a function of its scarcity—the less avail-
able the product, the better its quality. Examples of
this strategy include stating that there is a “limited
Compliance in Response to a Request
supply” of the product, or it is available only to the
Research on compliance with a request has focused first 50 callers, or “it’s a limited time offer.”
on the strategies often used by professionals to Another tactic used by professionals to increase
gain compliance from potential clients. These pro- the perception that a product is a “good deal”
fessionals include fund-raisers, salespeople, adver- involves the “that’s-not-all” technique. The com-
tisers, political lobbyists, and recruiters, to name a pliance agent first makes an offer and then imme-
few. Their requests may be explicit, such as an diately sweetens the deal by lowering the price or
invitation to donate to a charitable cause, or increasing the benefits. Sometimes this tactic is
implicit, such as an advertisement touting the used in combination with the scarcity tactic, as
advantages of owning a brand of clothing without when the target is informed that the special carving
directly asking for a purchase. knife can be purchased for $30, but if “you
Relevant research has addressed two questions: respond in the next hour you will also receive a
(1) What underlying motives of people do com- free set of paring knives and cutting board.”
pliance professionals attempt to capitalize on?
(2) Based on these underlying motives, what spe-
Motivation to Develop and Preserve
cific compliance strategies do professionals employ?
Meaningful Social Relationships
According to Robert Cialdini, those who employ
compliance-gaining strategies capitalize on three People often want to form new relationships
basic motivations of the target audience: (1) to and maintain existing relationships. It is therefore
132 Compliance

not surprising that people are more likely to com- compliance strategy, “the foot-in-the-door” strat-
ply with a request from someone they like than egy, capitalizes on the need for consistency. People
from someone they dislike. The positive feelings are first asked to comply with a small request,
underlying compliance may be based on the simi- which is usually granted. This is followed by a
larity or attractiveness of the person making the larger request.
request. A compliance tactic based on the motiva- Research shows that complying with the smaller
tion to preserve meaningful social relationships is request makes people more likely to comply with
used by the American Cancer Society. This tactic the larger request. For example, one study showed
involves enlisting group members, such as neigh- that people who first agreed to put a small sign in
bors, to solicit donations from fellow neighbors. their window were more likely to agree to a subse-
Evidence for the effectiveness of such tactics was quent request by a different solicitor to place a
obtained in a study showing that attractive solici- large sign on their lawn than were people who
tors for the American Heart Association produced were not first given the small request. Presumably,
almost twice the amount of compliance as unat- people complied with the larger request because
tractive solicitors. they wanted to be consistent with their image of
Norms, or agreed upon rules of conduct, are an themselves as helpful. Groups often employ this
important part of maintaining social relationships. tactic when socializing new members. For example,
Compliance professionals frequently make use of the group may initially request that the new mem-
one such norm, the norm of reciprocity, to elicit ber make only a minimal contribution to group
compliance. The strategy involves providing the effort, which is followed by escalating demands.
target person with a gift, such as address labels, A second compliance strategy, known as “low-
greeting cards, or a calendar, accompanied by a balling,” capitalizes on targets’ desire to view
request for a donation. The recipient, feeling themselves positively by adhering to commitments.
indebted to the donor, feels obligated to recipro- Take the case in which a customer in an auto show-
cate by making a donation. room makes an initial commitment to purchase a
Another compliance tactic that capitalizes on car. Lowballing occurs when the salesperson ini-
the norm of reciprocity is the “door-in-the-face” tially agrees to sell the car at a lower price than she
strategy. A requester makes an extreme request intends to get. Subsequently, after “consulting”
(e.g., a $200 donation), which is certain to be with her manager, she ups the price by informing
rejected. This is followed by a more moderate the customer that features that were supposedly
request (e.g., a $25 donation). Target persons, feel- included in the original offer, such as undercoating
ing obligated to make a reciprocal concession, tend and power steering, will cost extra. Customers
to comply by agreeing to the more moderate often comply with the additional request, presum-
request. A substantial amount of research docu- ably because they want to view themselves as the
ments the efficacy of this strategy. While the pre- type of person who adheres to commitments.
vailing view is that reciprocal concessions are the
mechanism that makes this work, some research-
ers contend that the key mechanism is “perceptual
contrast.” That is, the more moderate request is Groups commonly pressure members to comply
accepted because, in comparison to the extreme with their wishes. This entry examined two mean-
request, it appears to be a very minimal request. ings attached to the term compliance. Early
researchers defined compliance as public behavior
without private acceptance. Research in this tradi-
Motivation to Preserve a Positive Self-Concept
tion has produced important insights about the
Finally, people desire to maintain a positive conditions under which compliance leads to private
view of themselves, and compliance professionals acceptance. More recently, researchers have
often attempt to capitalize on this motivation. employed a definition of compliance that ignores
Viewing oneself as a person who is consistent and the distinction between public behavior and private
who adheres to commitments contributes to one’s acceptance. Instead, compliance is defined simply
positive self-perception. One well-documented as acquiescence to a request. This definition has
Computer-Mediated Communication 133

stimulated a great deal of research on the motiva- and a wide variety of less familiar tools. CMC
tional underpinnings of compliance and the tactics allows people to communicate at a distance,
that groups and individuals use to induce people to across both space and time. These technologies
comply. Research resulting from these two lines of can alter the ways that people converse with one
inquiry has enriched our understanding of social another in both positive and negative ways. It is
influence in both dyadic and group contexts. important to group processes and intergroup rela-
tions in two ways: First, the use of CMC instead
Martin S. Greenberg of face-to-face conversation can alter the ways
that people interact with each other, affecting such
See also Cognitive Consistency; Conformity;
Informational Influence; Normative Influence; Norms; things as group performance and people’s liking
Obedience to Authority for one another. Second, CMC itself makes it pos-
sible for people to interact with a more diverse set
of individuals, spanning many time zones and
Further Readings countries. Thus, CMC technologies can broaden
Burger, J. M. (1999). The foot-in-the-door compliance
social networks, helping people to know individu-
procedure: A multiple-process analysis and review. als from around the world.
Personality and Social Psychology Review, 3, 303–325. The appropriateness of any particular CMC
Cialdini, R. B. (2001). Influence: Science and practice tool for group communication will depend on
(4th ed.). Boston: Allyn & Bacon. what the group is trying to accomplish. Researchers
Cialdini, R. B., & Goldstein, N. J. (2004). Social are continuously working to create newer and bet-
influence: Compliance and conformity. Annual Review ter CMC technologies that will better support
of Psychology, 55, 591–621. group processes and intergroup relations. This
Cialdini, R. B., & Trost, M. R. (1998). Social influence: entry examines the various types of CMC tech-
Social norms, conformity, and compliance. In nologies, their effects on the conversational pro-
D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The cess, and their impact on social networks.
handbook of social psychology (4th ed., Vol. 2,
pp. 151–192). Boston: McGraw-Hill.
Festinger, L. (1953). An analysis of compliant behavior. CMC Technologies
In M. Sherif, & M. O. Wilson (Eds.), Group relations There are many types of CMC technologies, each
at the crossroads (pp. 232–256). New York: Harper. of which can have different effects on group pro-
Kelman, H. C. (1958). Compliance, identification, and cesses. These technologies can be differentiated
internalization: Three processes of attitude change. along several dimensions. First, CMC technologies
Journal of Conflict Resolution, 2, 51–60. vary according to the mode of communication sup-
Nail, P. R. (1986). Toward and integration of some ported, such as typing, voice communication, facial
models and theories of social response. Psychological
expressions, or gestures. A common way of talking
Bulletin, 100, 190–206.
about this is in terms of richness—the more aspects
of face-to-face communication (words, intonation,
facial expressions, etc.) available in a CMC tech-
Computer-Mediated nology, the richer that technology is said to be.
Thus, video is richer than telephony, which in turn
Communication is richer than instant messaging. Second, CMC
technologies vary according to the type of social
Computer-mediated communication (CMC) refers group they are intended to support. Some, such as
to the exchange of messages via computer tech- IM, are used predominantly by pairs of colleagues,
nologies, as opposed to face-to-face communica- friends, and family members. Others, such as tele-
tion. It includes familiar technologies, such as conferencing and online chatrooms, can support
e-mail or telephony; less common, but also popu- small groups. Yet others, such as blogs and e-mail
lar technologies, such as video conferencing, distribution lists, can support communication
instant messaging (IM; exchanging short text mes- among hundreds, thousands, or even millions of
sages), and blogging (writing in an online diary); people. Third, CMC technologies vary according
134 Computer-Mediated Communication

to whether they are intended to support real-time How well CMC supports group activities
(synchronous) communication, in which speakers depends in large part on the type of task people are
and listeners are both present at the same time and doing. When the goal is informal chitchat, most
can respond immediately to each other’s messages, any kind of synchronous communication will suf-
or whether they are intended to support different- fice, including the telephone and IM. The reason-
time (asynchronous) communication, such as when ing here is that nonverbal and vocal cues are less
someone reads and responds to an e-mail that was important for such conversations than they are for
sent the day before. tasks requiring more social delicacy, such as nego-
The characteristics of a given type of CMC tiations. For these more delicate tasks, richer
(mode of communication, type of social group, media like video conferencing are more appropri-
synchronous vs. asynchronous) have implications ate (in some cases people may actually travel long
for the types of group processes that can be sup- distances to conduct these conversations face-to-
ported. For example, e-mail can be a good method face). Tasks that involve talking about physical
for exchanging documents or notifying others of objects, such as maps, architectural diagrams, or
an upcoming event such as a party or presentation, pieces of technology, also benefit from CMC tools
but it is not as effective as the telephone for mak- that include video. Here, however, the kind of
ing rapid decisions. Instant messaging can be video that is most useful does not show a partner’s
slower than a telephone call, because people typi- face, but rather a view of the work space, so that
cally type more slowly than they speak, but it people can share a view of the objects they are
provides a written record of the discussion that can talking about.
be referenced in the future. At the level of conversational processes, research-
ers have found that specific attributes of media
influence how people communicate with one
Effects of CMC on Conversational Processes
another. For example, when people can’t see their
Face-to-face conversation is an essential part of partners, they are more likely to say things that are
most all group processes; it is the basic mechanism impolite, a phenomenon known as flaming. A
by which people exchange ideas, come to agree- common explanation for this is that when people
ment, negotiate outcomes, and perform other can’t see a partner they feel less constrained by
group tasks. Three key aspects of conversational social norms that prohibit rude behavior. They
processes are often altered when people communi- don’t have to look at a partner’s upset facial
cate via CMC: the types of communicative behav- expression or listen to him or her yell in response.
iors people perform, the role of conversational Flaming is particularly likely when people are
context, and turn taking. anonymous—for example, when they use fictitious
screen names in chat rooms—and thus cannot be
held personally accountable for what they say.
Words, Intonation, and Nonverbal Behavior
Communication via CMC has also been studied
In conversations, listeners attend to the words with respect to interpersonal deception. In text-
being spoken, but they also listen to how these based media like e-mail or IM, not only are visual
words are spoken, and they watch facial expres- cues missing (e.g., “shifty eyes”) but also the per-
sions, hand gestures, and other nonverbal behav- son who is lying has more time to carefully com-
iors for insight into what the speaker means. pose falsehoods. In fact, police agencies can create
Imagine, for example, hearing someone say, “I’m entirely false personas in text chat rooms, as is the
so happy!” versus “I’m so happy.” With the rising case when agents pretend to be young girls in order
intonation suggested by an exclamation mark, the to catch sexual predators. The lack of visual cues to
speaker sounds much happier. Similarly, listeners’ the typist’s identity and the additional time avail-
nonverbal behaviors are important sources of able to carefully craft typed messages create oppor-
information for speakers. Speakers monitor lis- tunities to deceive on a much larger scale than that
teners’ gaze and facial expressions to assess commonly available in face-to-face groups.
whether or not they are listening and understand Perhaps because lying is easier when text-based
the message. media are used, members of groups that interact
Computer-Mediated Communication 135

virtually often show less trust in each other than people to multitask, switching between multiple
do members of collocated groups. Trust among conversations or between a conversation and some
members of virtual groups can be increased if other activity. Because of the invisibility of the con-
people interact informally with one another first, text, conversational partners can’t see that they do
either face-to-face or via CMC. Once people get to not have a person’s undivided attention.
know one another, trust levels in virtual and col- When context is lacking, the careful negotiation
located groups are similar. of engagement in an interaction is disrupted. When
people are face to face, they use a series of nonver-
bal cues (looking at the other person from a dis-
tance to see if he or she is busy, establishing eye
Conversations take place within a larger set- contact, moving into speech range, etc.) to initiate
ting, such as a workplace, school, or home, and a conversation. In virtually all types of CMC, the
this setting helps shape what kinds of messages information needed for this engagement process to
are appropriate and how they should be under- go smoothly is missing. As a result, it is easy to
stood. In CMC, the sender of a message may not interrupt people when they are busy. For example,
know the context of the recipient, and thus acci- when making a telephone call there is no way to
dentally violates social norms. One example of determine if the called person is sleeping, cooking,
this is sending jokes or other personal IM mes- painting, or otherwise engaged—the person is sim-
sages to recipients who are projecting their com- ply interrupted in the middle of whatever he or she
puter screen to an audience. In such circumstances, was doing. The same holds true for senders of IM
the entire audience will see the possibly inappro- messages, although some IM clients allow people
priate message. to indicate their availability for communication
The absence of context also affects people’s (e.g., setting their status as “away”). E-mail spam-
interpretations of events. It becomes easy for peo- mers take advantage of this lack of context by
ple to misinterpret others’ responses or failures to sending hundreds or even thousands of unwanted
respond. If someone fails to respond to an e-mail messages each day.
message, it could be because he or she is intention-
ally ignoring it, or it could be because the person is
Turn Taking and Conversational Participation
very busy. When cell phones become disconnected
in the heat of an argument, it could be because the In a typical conversation, people take turns
signal was dropped or it could be because the speaking and listening in an orderly fashion, using
other party hung up in anger. Several studies have turn-taking signals such as tone of voice; questions
found that when context is lacking, people tend to like, “What do you think, John?”; and eye gaze. In
make what social psychologists call the fundamen- CMC, many if not most of these cues are missing.
tal attribution error: They overattribute the causes This has both negative and positive effects on
of events to their partner’s dispositional character- group interaction. On the negative side, it can be
istics (e.g., laziness, lack of interest) when those more awkward to change speakers in CMC, espe-
causes actually involve situational constraints, cially in large groups such as multiparty audio
such as the quality of the technology (e.g., bad conferences. On the positive side, however, shier
connection). people find it easier to speak up when they do not
A flip side of the ambiguity of CMC is a phe- have to attend to these turn-taking cues. In fact,
nomenon known as plausible deniability. People one of the earliest goals of text-based CMC tools
are aware that their communication partners can’t was to increase the evenness of participation
see what they are doing, and they can thus pretend among members of group by making everyone
to miss messages when they really just don’t want anonymous. That way, it was reasoned, people of
to respond. Plausible deniability is risky, however, lower status or with greater fears of speaking
for the reasons stated above: The sender of the mes- would contribute more. These systems were fairly
sage is most likely to assume that a nonresponse is successful, but as one might imagine, people did
due to partner characteristics, not the situation. not like having anonymous partners and the tools
The ambiguity of context in CMC also enables were never used to any great degree.
136 Computer-Mediated Communication

Using CMC to Broaden Social Networks cultures, particularly the United States, Canada,
and Europe. An important area for future work is
In addition to its effects on individual conversa-
to understand how CMC changes group process
tional processes, CMC can help grow people’s
among different cultural groups and in intercul-
social networks—the set of people they know and
tural communication.
who those people know. But CMC can also intro-
duce problems into these new social relationships.
Longer Term Relationships
Meeting New People
An intriguing aspect of virtual relationships is
When people are all collocated in the same that people are more likely to disclose personal
physical area, they tend to bump into one another information about themselves to those they meet
in common areas such as cafeterias, hallways, or online than they are to disclose such information
coffee shops. Social norms dictate greeting to face-to-face acquaintances. This can have both
acquaintances in such settings, so unplanned con- benefits and costs. On the positive side, disclosure
versations often arise. These conversations of personal information, especially when recipro-
strengthen interpersonal bonds. The majority of cated, can speed up the development of deep and
CMC technologies do not support this kind of long-lasting relationships. On the negative side,
informal communication, though there are some however, disclosure of personal information to
exceptions. For example, there are thousands of unfamiliar others can be risky and in some cases
online chat rooms where one can go to discuss has led to serious consequences such as identity
topics of common interest with people one has theft or predatory behavior.
never met in real life. In some cases, longer term Although CMC can broaden social networks
friendships and even marriages have developed and lead to deep friendships, it can also introduce
from chat room interactions. Another place that problems into social relationships. Earlier, this
people encounter new potential friends is in vir- entry discussed phenomena that surround a single
tual social environments such as Second Life, interaction, either face-to-face or via CMC. These
which even includes large (virtual) public events phenomena have a way of adding up over time,
like concerts or presentations to bring people shaping the quality of the interactions among a
together. Social networking sites like Facebook group of people and influencing the outcomes of
and MySpace also allow people to extend their their activities. The erroneous overattribution of
networks by facilitating connections between problems to others’ personal characteristics rather
friends of friends. At the same time, these sites than technological limitations is one example.
allow people to maintain existing close relation- When failures to respond are attributed to ill will
ships after life changes, such as moving away for or laziness on the part of one’s partners, this
college. can’t help but negatively impact future interac-
One of the values of meeting new people online tions. Similarly, confusions that arise because
is that these people tend to be more diverse than visual or auditory cues are missing in CMC may
those one would meet in one’s neighborhood. In not be noticed until much further down the line,
one chat room, for example, a typical discussion when they have already had significant conse-
included people from the United States, Canada, quences. Perhaps for these reasons alone, social
Mexico, Germany, and Australia. Over time, many interactions via CMC are often less successful
of these people became close friends and some than those among collocated individuals, at least
even traveled to other countries to meet face-to- in the short run. With more time and experience
face. Interaction across cultures brings interesting with CMC, groups are often able to overcome
new challenges for many people, because cultures these problems.
differ in communication styles, politeness norms,
opinions about what topics are appropriate versus Susan Fussell
inappropriate, and norms for who should take the
lead in discussions. Most of what we know about See also Culture; Social Identity Model of Deindividuation
CMC is the result of studies conducted in Western Effects; Virtual/Internet Groups; Work Teams
Computer Simulation 137

Further Readings validity can be evaluated by comparing its output

Clark, H. H., & Brennan, S. E. (1991). Grounding in to empirical observations. Structurally, the theory is
communication. In L. B. Resnick, R. M. Levine, & embodied in the choices of attributes and behaviors
S. D. Teasley (Eds.), Perspectives on socially shared that are included in the model. Dynamically, the
cognition (pp. 127–149). Washington, DC: American theory is captured by the functional rules. These
Psychological Association. rules describe how agents’ attributes (knowledge,
Hancock, J. T. (2007). Digital deception: When, where, mood, etc.) affect their behaviors and how the
and how people lie online. In K. McKenna, behavior of each agent changes the attributes of
T. Postmes, U. Reips, & A. N. Joinson (Eds.), Oxford others, and thus their subsequent behaviors. Using
handbook of Internet psychology (pp. 287–301). the capacity and flexibility of computational mod-
Oxford, UK: Oxford University Press. eling, one can combine many theoretical ideas
Hinds, P., & Kiesler, S. (Eds.). (2002). Distributed work. across different levels of analysis (individual, small
Cambridge: MIT Press. group, organizational, and community) into a uni-
Spears, R., Lea, M., & Postmes, T. (2001). Social fied and coherent working model.
psychological theories of computer-mediated
communication: Social pain or social gain. In W. P.
Robinson & H. Giles (Eds.), The new handbook of An Example: Mate Selection
language and social psychology (pp. 601–623). Consider modeling how potential mates choose
Chichester, UK: John Wiley. partners. Suppose the domain of interest is mate
Weisband, S. (2007). Leadership at a distance: Research selection in a heterosexual, monogamous commu-
in technologically supported work. Mahwah, NJ:
nity of unattached agents. Characterizing the
Lawrence Erlbaum.
domain in this way implies that mating occurs
Whittaker, S. (2003). Theories and methods in mediated
when a female and a male agree to form a union
communication. In A. Graesser, M. Gernsbacher, & S.
and leave the pool of unattached agents. Hence,
Goldman (Eds.), The handbook of discourse processes
each agent’s gender is a necessary attribute and a
(pp. 243–286). Mahwah, NJ: Lawrence Erlbaum.
necessary functional rule is that same-sex agents
will not mate. These structural components of the
model are necessary but often are not of primary
theoretical interest. For example, one might develop
Computer Simulation the model based on the following propositions:

Computer simulation uses the computational and Attractiveness proposition: Each agent has a level
storage capacity of computers to model complex of attractiveness that summarizes her or his value
and dynamic systems of behaviors. In a typical as a mate.
application, a model of group process includes:
Process assumptions: (1) a meeting is a temporary
•• agents (group members) who act and interact; pairing of two agents, (2) a proposal to a mate may
•• attributes that describe these agents; occur if the two parties in a meeting are of the
•• behaviors that the agents can display (behavioral opposite gender, (3) either party may initiate a
repertoire); and proposal, and (4) for a proposal to result in mating,
•• functional rules that specify how agents’ the recipient of a proposal must accept.
attributes affect their behaviors and how other
agents’ behaviors affect these attributes. Proposal proposition: The probability that a person
initiates a proposal is positively related to the level
Through a series of computational steps, the model of attractiveness of her or his partner in a
generates the interaction of agents as it unfolds over meeting.
time. At any point, the state of the interaction is
described by the current values of agents’ attributes. Acceptance proposition: The level of attractiveness
Such a computational model is a theory. It sym- of the proposer is positively related to the probability
bolically represents a real-world process, and its that a proposal will be accepted.
138 Computer Simulation

Rejection proposition: The more often a person’s random processes superimposed on the processes
proposals are rejected, the more likely the person that are explicitly incorporated in the model.
will be to initiate and accept a proposal.
Computational Models in Social Psychology
These propositions and assumptions comprise a
simple model of mate selection. The simplicity is In the 1968 Handbook of Social Psychology,
imposed for exposition purposes. For example, the Robert Abelson reviewed the early history of com-
model as outlined assumes that unions once formed putational modeling in social psychology. His and
remain intact. More realistically, some attached subsequent works include many examples of com-
partners may reenter the pool, but the process that puter simulations of group behavior. In 1988,
determines this reentry would have to be specified. Thomas Ostrom edited a special edition of the
Moreover, representing attractiveness as one- Journal of Experimental Social Psychology that
dimensional is simplistic. In practice, attractiveness featured computational models of social behavior.
is determined by many attributes, which are Ostrom characterized computational models as the
weighted differently by different agents. Also, the “third symbol system” for theorizing—verbal lan-
domain of interest could be expanded to include guage and mathematical systems being the other
homosexual and bisexual agents. This expansion two. He argued that computational models combine
would require adding additional attributes and the precision and deductive power of mathematical
functional rules to the process. Adding such com- systems with the flexibility and complexity of verbal
plexity is limited primarily by the creativity and expression. In the 2000 Handbook of Research
ingenuity of the modeler. The medium of computer Methods in Social and Personality Psychology, Reid
simulation easily accommodates complexity. Hastie and Garold Stasser presented two examples
Even within this simple model of mate selection, that illustrate the process (and possible pitfalls) of
translating the theoretical propositions into compu- computational modeling: the IMP (impression and
tational steps adds complexity. Consider the pro- memory processor) model of impression formation
posal proposition. Two agents, M and F, meet. and the DISCUSS model of small group discussion
Each has a specified level of attractiveness and these and decision making. Each model is presented in
are stored in the variables Mattr and Fattr. Does F sufficient detail to illustrate how one constructs and
propose a union? There are two general approaches evaluates models and how computational models
to translating attributes into behavior. One might interact with verbal and mathematical theories and
define a deterministic functional rule: If Mattr empirical observations. In 2000, Daniel Ilgen and
exceeds a critical threshold of attractiveness, then F Charles Hulin compiled a set of papers that describe
will propose. This approach requires an additional applications of computational models and commen-
attribute, namely an attractiveness threshold for tary on these applications. Their book documents
each agent. Alternatively, one might define a sto- the richness and variety of models in organizational
chastic rule that adjusts the probability of F propos- and group behavior.
ing as a positively increasing function of Mattr. This
stochastic approach requires that the program
Computational Modeling of Group Process
complete three steps: (1) compute the probability
of the event as a function of relevant attributes, An attractive feature for the study of group process
(2) sample a value from a probability distribution, is that a model can represent multiple levels of
and (3) compare the sampled number to the com- analysis: the individual, the small group, the orga-
puted probability of the event. If the sampled ran- nization, the community, and society. Bridging
dom number is greater or equal to the probability levels of analysis permits scholars to explore the
of the event (in this case, F proposing), then the implications of what happens at one level for what
event occurs; otherwise, it does not. A stochastic is observed at another level. For example, S. M.
approach implicitly recognizes that variables other Kalick and T. E. Hamilton used a model similar to
than those represented in the computational steps the foregoing example, in which individuals prefer
may determine whether the event of interest occurs, attractive mates, and showed that such a process
and these extraneous contributions are modeled as resulted in mates that were similar in attractiveness.
Conformity 139

They concluded that similar levels of attractiveness Further Readings

within couples does not imply that people seek Hastie, R., & Stasser, G. (2000). Computer simulation
mates who are similar to themselves in attractive- methods in social psychology. In H. Reis & C. Judd
ness. It is possible that an apparent preference for (Eds.), Handbook of research methods in social and
matching attractiveness is an emergent feature of personality psychology (pp. 85–114). Cambridge, UK:
the process, not a preference of the individual Cambridge University Press.
agents. Bibb Latané and his colleagues provided Ilgen, D. R., & Hulin, C. L. (2000). Computational
another example of bridging levels. A major tenet modeling of behavior in organizations: The third
of social impact theory is that the influence exerted scientific discipline. Washington, DC: American
on a person increases as the number of sources of Psychological Association.
influence increases. One implication is that major- Mason, W. A., Conrey, F. R., & Smith, E. R. (2007).
ity positions are more likely to gain than lose Situating social influence processes: Dynamic,
adherents. As a result, it seems that minority dissent multidirectional flows of influence within social
would disappear over time, causing opinions in a networks. Personality and Social Psychology Review,
group or community to converge on a shared posi- 11, 279–300.
tion. However, Latané and his colleagues demon- Ostrom, T. M. (1988). Computer simulation: The third
strated that another major tenet of social impact symbol system. Journal of Experimental Social
theory modulates the process of majority influence: Psychology, 24, 382–392.
The impact of a source depends on its psychologi-
cal immediacy. One component of immediacy is
proximity in a communication network. They
modeled communities of agents that were con- Conformity
nected by various configurations of communication
channels. Over the course of several rounds of Social influence, defined as changing one’s percep-
“talking,” agents learned their neighbors’ opinions tions, opinions, or behaviors in response to real or
and adjusted their individual opinions toward the imagined pressure from others, is a fundamental
opinions expressed by their neighbors. Latané and aspect of group life. Various forms of social influ-
his colleagues noted three emergent features at the ence have been identified, including compliance
level of the community. First, the number of agents (going along with a direct request from others),
holding minority opinions decreased (consolida- obedience (following the orders of a legitimate
tion). Second, neighborhoods of agents who agreed authority figure), and conformity (changing one’s
emerged (clustering) and disparity of opinions sur- perceptions, beliefs, or actions in the direction of
vived across neighborhoods (continuing diversity). a perceived group norm). This entry focuses on
That is, neighborhoods of minority opinion sur- conformity, examining some of the definitional
vived within an overall pattern of majority influ- and measurement issues that researchers face, the
ence. Third, when multiple issues were tracked motivations that lead people to conform, the
simultaneously, opinions across issues became impact of having a partner (social supporter) on
increasingly correlated over time (correlation), an resisting group pressure, and the role of individual
unintuitive by-product of consolidation and clus- differences in determining conformity.
tering. Thus, the dynamics of social impact, mod-
eled as communication and influence between
Classic Studies
individuals, generated neighborhoods of like-
minded people, pockets of minority dissent, and Two lines of classic research had a great impact on
characteristic belief profiles across multiple issues. how social psychologists think about conformity.
In one set of studies published in 1935, Muzafer
Garold Stasser and Susanne Abele Sherif demonstrated the power of social influence
to change people’s perceptions of highly ambigu-
See also Dynamical Systems Approach; Group Problem ous stimuli. Sherif made use of a perceptual illusion
Solving and Decision Making; Minority Influence; called the autokinetic effect, which occurs when
Research Methods and Issues; Social Impact Theory people are asked to concentrate on a stationary
140 Conformity

point of light in a dark room. Under these circum- groups elicit conformity from their members.
stances, people (who are not informed the light is Scores of studies have been conducted on this topic
actually stationary) perceive movement in the light. in the years since their groundbreaking research
Some think it moves only a little; others think it was published, and much has been learned.
moves a lot. Sherif found that, when groups of
three people were brought together and asked to
Definitional and Measurement Issues
say out loud how far the light moved, their judg-
ments gradually converged over trials. In other As noted above, we define conformity as change in
words, they developed a group norm about the a person’s perceptions, beliefs, or actions in the
distance the light moved. Moreover, this norm had direction of a perceived group norm. Although
a lasting impact on participants’ perceptions. seemingly straightforward, this definition masks
When later asked to make estimates alone, their several complexities regarding how conformity is
responses continued to be influenced by the group conceptualized and measured.
estimate. Subsequent research demonstrated that
conformity to the norm group was still evident a
Movement Versus Agreement Conformity
year later. It is important to note that, in these stud-
ies, there was initially no norm to which partici- Defining conformity in terms of change is use-
pants could conform. Instead, they created this ful, because it allows us to differentiate conformity
norm through mutual social influence, and it then from behavioral uniformity, which involves inde-
influenced their private responses. pendent agreement in the absence of perceived
A second set of studies, published by Solomon group pressure. Simply knowing that a person
Asch in 1951, demonstrated the power of social agrees with a group norm at one point in time does
influence to change people’s perceptions of highly not allow us to make a confident judgment about
unambiguous stimuli. Thus, in contrast to Sherif, the source of that agreement. Perhaps the person
Asch was interested in the conditions under which independently arrived at the group’s position with-
people would yield to group pressure even though out any knowledge of the group norm or any
the group was obviously incorrect. To answer this desire to adhere to it—an instance of behavioral
question, Asch assembled groups of seven to nine uniformity. In contrast, knowing that the person
people for a study on visual perception. The exper- disagreed with the group at Time 1 and then
imental task, which involved matching the length shifted toward it at Time 2 would increase our
of a standard line against three comparison lines, confidence that the group exerted influence on the
was quite easy. Each group contained one naïve person—an instance of conformity. This is espe-
participant who answered next-to-last. The remain- cially true if others who shared the person’s initial
ing “members” were confederates of the experi- position, but were not exposed to group pressure,
menter and gave unanimously incorrect answers failed to change their position.
on 12 of 18 trials. Asch found, to his surprise, that Although the criterion of movement is useful in
conformity occurred even in a situation where the defining conformity, it has potential pitfalls. For
majority gave clearly erroneous answers. example, in some cases a person who indepen-
Participants’ responses agreed with the erroneous dently agrees with a group norm but is tempted to
majority approximately one third of the time, and abandon it may fail to take this action because of
27% of participants conformed on at least 8 trials. group pressure. Here, conformity is revealed by
In contrast, control participants (who made judg- refusal to change. Response inhibition as a reac-
ments privately) gave incorrect answers less than tion to group pressure has also been discussed
1% of the time. Although the level of conformity under the rubric of “conformity by omission,”
that Asch obtained may seem surprising, it is which is contrasted with the more commonly stud-
worth noting that participants’ responses were ied “conformity by commission.” In the commis-
correct approximately two thirds of the time and sion case, conformity involves performing a
24% of participants never conformed. behavior because of group pressure that one would
Together, Sherif’s and Asch’s studies stimulated not otherwise perform (e.g., saying a prayer one
a tremendous amount of interest in when and why does not believe in because classmates are saying
Conformity 141

it). In the omission case, conformity involves fail- sanctions from outgroup members for taking a
ing to perform a behavior because of group pres- public stand agreeing with the group), or both
sure that one would otherwise perform (e.g., not compliance and acceptance (e.g., because he or she
saying a prayer one does believe in because class- accepts the group’s position and wants to encour-
mates are not saying it). age others to adopt this position). Whether confor-
Another potential problem with the movement mity reflects compliance and/or acceptance has
criterion involves the temporal relationship between implications for how a person will behave when
exposure to group pressure and response to this the group is absent. For example, a person who
pressure. We have implicitly assumed that confor- conforms at the public but not at the private level
mity occurs immediately after pressure is exerted, is unlikely to endorse the group’s position when
but this is not always the case. One counter example responding privately. In contrast, a person who
is anticipatory conformity, in which a person conforms at both levels is likely to endorse the
expects future group pressure and responds to it by group’s position even when the group is absent.
moving toward the group norm before the pressure The distinction between compliance and accep-
is applied. Another counter example is delayed con- tance is applicable to nonconformity as well as
formity, in which a person experiences group pres- conformity. Several forms of nonconformity can
sure, is unable or unwilling to conform immediately, be distinguished, but two of the most important
but moves to the group norm at some later time. In are independence and anticonformity. Independence
both cases, there is a causal link between pressure occurs when a person initially disagrees with a
and conformity, but this link is hard to detect. group and exhibits neither compliance nor accep-
tance after being exposed to group pressure. In
other words, the person stands fast when faced
Public Versus Private Conformity
with disagreement. In contrast, anticonformity
Our discussion so far has emphasized overt occurs when a person initially disagrees with a
(behavioral) responses to group pressure. However, group and moves even further away from its posi-
conformity can involve covert (attitudinal or per- tion (at the public and/or private level) after being
ceptual) responses as well. Two general categories exposed to pressure. In other words, the person
of conformity have therefore been distinguished— becomes more extreme in his or her initial position
public agreement (compliance) and private agree- when faced with disagreement. Ironically, then,
ment (acceptance). If conformity is defined as anticonformers are just as responsive to group
movement toward a group norm, then compliance pressure as are conformers, but they manifest their
refers to overt behavioral change in the direction susceptibility in a very different way (by moving
of that norm, whereas acceptance refers to covert away from, rather than toward, the group).
attitudinal or perceptual change. For example, if
an individual initially refused to sign a petition
Motivational Bases of Conformity
advocating abortion rights, learned that a group
advocated these rights, and then signed a petition People conform to group pressure because they are
favoring these rights, the person would be showing dependent on the group for satisfying two impor-
compliance. In contrast, if an individual privately tant goals. One is the desire to have an accurate
believed that abortion should be outlawed, learned perception of reality, and the other is the desire to
that a group advocated abortion rights, and then be accepted by other members.
changed his or her private opinion about these
rights, the person would be showing acceptance.
Informational Influence
The relationship between compliance and accep-
tance is potentially complex. An individual could There is a great deal of evidence that people
exhibit compliance but not acceptance (e.g., want to hold correct beliefs about the world,
because he or she fears group reprisal for deviance because such beliefs lead to actions that maximize
but does not privately accept the group’s position), the probability of rewarding outcomes. Some of
acceptance but not compliance (e.g., because he or our beliefs about the world (e.g., oatmeal cooks
she privately accepts the group’s position but fears better in hot water than cold water) can be verified
142 Conformity

by using objective tests (e.g., leaving oatmeal in 19 expected negative group evaluation. In con-
hot vs. cold water for 5 minutes and then tasting trast, of 24 people who conformed a great deal, 17
it). In contrast, other beliefs (e.g., the U.S. should expected positive evaluation from other group
maintain its nuclear arms capability; the federal members. Additional evidence demonstrates a fac-
government should institute stronger environmen- tual basis for nonconformers’ fear of punishment.
tal regulations) cannot be verified using objective Many studies of reaction to deviance show that
standards and hence must be verified using social group members do indeed reject people who devi-
tests, namely comparing our beliefs to those of ate from group consensus, depending on such fac-
other people whose judgment we respect. If these tors as the extremity and content of the deviate’s
others agree with us, we gain confidence in our position, the presumed reason for the deviate’s
beliefs. If they disagree with us, we lose confidence. behavior, and the deviate’s status in the group.
Because disagreement is disturbing, we are moti- Because actual or anticipated rejection is so dis-
vated to eliminate it, and one way to do so is to turbing, people try to minimize it by conforming to
conform to group norms. group norms.
According to this analysis, people sometimes According to this analysis, people sometimes
conform to groups because they are uncertain conform to groups because they are motivated to
about the correctness of their beliefs and believe be liked (or at least not disliked) and believe that
the group is more likely to be correct than they are. other members will feel more kindly toward
This kind of conformity reflects what Morton them if they conform to rather than deviate from
Deutsch and Harold Gerard labeled informational group norms. This kind of conformity reflects
influence. In general, informational influence pro- what Deutsch and Gerard labeled normative
duces private acceptance as well as public compli- influence. In general, normative influence pro-
ance. This is illustrated in the work of Muzafer duces public compliance but not private accep-
Sherif, discussed earlier. His research indicated that tance. This is illustrated in the work of Solomon
people judging an ambiguous stimulus exhibited Asch, discussed earlier. Few of Asch’s partici-
both compliance (when they made judgments in pants reported changing their perceptions of the
others’ presence) and acceptance (when they later experimental stimuli during the group pressure
responded privately). situation, and subsequent studies indicated that
Because informational influence is based on participants’ private responses after leaving the
insecurity about one’s beliefs, we would expect it situation often differed from their public responses
to be more common when an individual feels in that situation.
dependent on others for information. Consistent Because normative influence is based on insecu-
with this assumption, people conform more when rity about one’s acceptance, we would expect it to
they are working on ambiguous tasks than on be more common when an individual feels threat-
unambiguous tasks. In addition, they conform ened for deviating from group norms. Consistent
more when they have doubts about their own task with this assumption, group members conform
competence and when they think other group more when working for a common goal rather
members are highly competent on the task. than individual goals, presumably because they
believe that deviance on their part will be punished
more severely in the former case. As might be
Normative Influence
expected, however, conformity in common goal
In addition to wanting to hold correct beliefs groups is substantially reduced if members believe
about the world, people are motivated to be that this behavior will lower the group’s probabil-
accepted by other group members. The desire for ity of attaining a positive outcome. Another factor
social acceptance is very powerful in a wide range that increases normative influence is surveillance
of situations and explains why people are typically by other group members. People who are con-
quite uncomfortable if they think others currently cerned about others’ evaluations ought to conform
reject them or are likely to do so in the future. For more when their behavior is public rather than
example, the results of one study indicated that, of private, and conformity is in fact higher in the
25 people who conformed very little or not at all, former condition.
Conformity 143

Mixed Cases The effectiveness of social support can be

explained in terms of the supporter’s ability to
Although informational and normative influ-
reduce informational and/or normative influence.
ence have been discussed here as though they are
In the case of informational influence, social sup-
mutually exclusive, they occur simultaneously in at
porters can lower participants’ dependence on the
least some group situations. This is a major prem-
majority for information about reality. For exam-
ise of social identity theory, which seeks to explain
ple, in one study, participants received support
a range of social influence phenomena, including
from a partner who either had normal vision (and
conformity. This theory assumes that disagreement
hence could see the stimuli clearly) or wore
with others produces uncertainty only when one
extremely thick glasses and failed a “vision test” in
expects to agree with these people. For this reason,
the participant’s presence. Consistent with an
disagreement with ingroup members produces
informational influence explanation, the compe-
more uncertainty than disagreement with outgroup
tent supporter was more effective in reducing con-
members. In addition, the theory assumes that
formity. In the case of normative influence, social
some ingroup members are more influential than
supporters can lower participants’ fear of punish-
others. More specifically, a member’s influence
ment for deviation from the group norm. Research
depends on how much his or her position embod-
indicates that people who dissent from majority
ies what is unique about the group—the norm that
consensus with a supporter are much less appre-
differentiates the ingroup from outgroups.
hensive about being rejected than are those who
Members who are closer to this norm are more
dissent alone. The presence of a supporter may
influential than those who are further from it.
reduce participants’ fear of retaliation because they
Finally, the theory assumes that conformity involves
believe the supporter will absorb some of the
private acceptance as well as public compliance,
majority’s anger toward deviates. This should not
because people believe that ingroup norms provide
be the case, however, if majority members are
valid evidence about reality. A substantial amount
assumed to dislike the supporter, for example
of research is consistent with the social identity
because they are prejudiced against his or her racial
explanation of conformity.
group. In such cases, participants may expect that
a perceived alliance with the supporter will increase,
Social Support and Conformity Reduction
rather than decrease, the majority’s hostility toward
Asch investigated the impact of group unanimity them and hence may continue to conform.
on conformity by having a single confederate dis-
sent from the erroneous majority by giving correct
The Role of Individual Differences
answers on the line-judging task. The presence of
this social supporter reduced conformity dramati- Does everyone who enters a group pressure situa-
cally, from 33% to 6%. In later research, Asch tion respond in exactly the same way to it? Of
found that participants who were opposed by an course not. Every conformity experiment that has
eight-person majority and had a supporter con-