You are on page 1of 36

Conceptual foundations of safety culture

and safety climate measurement:


A snapshot review

Trang Vu and Helen De Cieri

3 December 2015
Research report#: 060-1215-R04

This research report was prepared by


Dr Trang Vu and Professor Helen De Cieri, Monash Business School, Monash University, P O Box
197, Caulfield East, Victoria 3145.
For WorkSafe Victoria.
This report is a companion to:
Vu, T., & De Cieri, H. (2015). Safety culture and safety climate measurement tools: A systematic
review. Melbourne, Australia: Institute for Safety, Compensation and Recovery Research (ISCRR),
Monash University. ISCRR Research Report #060-1215-R03.

Acknowledgements
The research team would like to thank WorkSafe Victorias project reference group, in particular Mr
Ross Pilkington and Mr Jamie Swann, for their input and valuable assistance in facilitating contact
with stakeholders of the research.
Funding for this study was provided by WorkSafe Victoria through the Institute for Safety,
Compensation and Recovery Research.

Disclaimer
The information provided in this document can only assist an individual or organisation in a general
way. Monash University accepts no liability arising from the use of, or reliance on, the material
contained on this document, which is provided on the basis that Monash University is not thereby
engaged in rendering professional advice. Before relying on the material, users should carefully
make their own assessment as to its accuracy, currency, completeness and relevance for their
purposes, and should obtain any appropriate professional advice relevant to their particular
circumstances.

Contact
Dr Trang Vu
Monash Business School, Monash University, P O Box 197, Caulfield East, Victoria 3145, Australia.
Email trang.vu@monash.edu
Website: www.iscrr.com.au
ISCRR is a joint initiative of WorkSafe Victoria (WSV), the Transport Accident Commission and Monash University. The
opinions, findings and conclusions expressed in this publication are those of the authors and not necessarily those of the
WSV or ISCRR.

ISCRR Research Report#: 060-1215-R04

Page 2 of 36

Contents
Executive summary .......................................................................................................................... 4
1.

Introduction ................................................................................................................................ 9

2.

Defining safety culture and safety climate .................................................................................. 9

3.

Methods ..................................................................................................................................... 9

4.

Conceptual models for understanding and measuring safety culture and safety climate .......... 10

5.

6.

7.

4.1

Scheins model of organisational culture ........................................................................... 10

4.2

Glendon and Stantons model for conceptualising safety culture and safety climate ......... 11

4.3

Guldenmunds safety culture framework ........................................................................... 11

4.4

Coopers reciprocal safety culture model .......................................................................... 12

4.5

Choudhry, Fang and Mohameds model of construction safety culture ............................. 14

4.6

Flins model of safety climate and injury outcomes for healthcare industry ....................... 14

4.7

Gellers Total Safety Culture model .................................................................................. 15

4.8

Griffin and Neals model of safety climate ......................................................................... 15

4.9

Helmreich and Merritts safety culture model .................................................................... 16

4.10

Kennedy and Kirwans safety culture model ..................................................................... 16

4.11

Reasons safety culture model .......................................................................................... 17

4.12

Reasons `Swiss Cheese model ...................................................................................... 18

4.13

Theory of high reliability organisation ................................................................................ 19

4.14

Turners theory of man-made disaster (1978) ................................................................... 19

4.15

Zohar and Lurias (2005) multilevel model of safety climate .............................................. 20

4.16

Summary remarks ............................................................................................................ 22

4.17

Implications for measurement ........................................................................................... 22

Approaches used to measure safety culture ............................................................................ 23


5.1

Measuring safety culture using a specific conceptual model or framework ....................... 24

5.2

Measuring safety culture using a pragmatic approach ...................................................... 27

Approaches used to measure safety climate............................................................................ 28


6.1

Measuring safety climate based on specific conceptual models ....................................... 28

6.2

Measuring safety climate based on a pragmatic approach ............................................... 29

Conclusion ............................................................................................................................... 29

References ..................................................................................................................................... 31

ISCRR Research Report#: 060-1215-R04

Page 3 of 36

Executive summary
This report has been written as a companion to the systematic review of safety culture and safety
climate measures (Vu & De Cieri, 2015). That review systematically identified and evaluated safety
culture and safety climate measures according to pre-defined selection and evaluation protocols.
Key theories and models used by researchers in the development of safety culture and safety
climate measures included in that review are examined in the current report. The aims of the current
report are to provide

a summary of key theories and conceptual models that underpin the development of safety
culture and safety climate measures included in the review by Vu and De Cieri (2015); and
an examination of approaches to measurement of safety culture and safety climate and
identification of key issues in the measurement of safety culture and safety climate.

Safety culture and safety climate are concepts that are often used interchangeably. In this report,
however, they are considered as two distinct concepts. In the absence of consensus definitions and
for the purpose of this report, safety culture is defined following the (US) National Occupational
Research Agenda (NORA) Construction Sector Council (2008) and safety climate is defined
following Wiegmann, Zhang, von Thaden, Sharma, and Mitchell (2002) (see Box 1).
Box 1. Safety culture and safety climate definitions for the purpose of this report
Authors

Definition

NORA Construction
Sector Council (US)
(2008, p.65)

Safety culture reflects the attitudes, values, and priorities of management


and employees and their impact on the development, implementation,
performance, oversight, and enforcement of safety and health in the
workplace.

Wiegmann et al.
(2002, p.10)

Safety climate is the temporal state measure of safety culture, subject to


commonalities among individual perceptions of the organisation. It is
therefore situationally based, refers to the perceived state of safety at a
particular place at a particular time, is relatively unstable, and subject to
change depending on the features of the current environment or prevailing
conditions.

Fifteen theories and models frequently used in the development of safety culture and safety climate
measures are examined in this report. They are summarised in Table 1 below. Each theory or
model summarised below identifies a slightly different set of potentially relevant attributes or
dimensions for measurement. It is therefore important that the measurement of safety culture or
safety climate should begin with a clear understanding of what the construct of interest is, what
should be included in the construct domain, whether industry-specific contexts are relevant in the
measurement and how the construct relates to safety outcomes. Additionally, since safety culture
and safety climate are likely to be influenced by interactional processes at multiple levels, it is
worthwhile to identify the key drivers of safety culture and safety climate at each level and
investigate cross-level influences using separate measurement scales.

ISCRR Research Report#: 060-1215-R04

Page 4 of 36

Table 1. A snapshot of the conceptual foundations of safety culture and safety climate
measurement
Theory or model presented in
this report

Brief description

Scheins model of
organisational culture
(Schein, 1985)

Describes three layers of organisational culture, from least to


most visible: basic assumptions, beliefs and values, and
artefacts. Basic assumptions represent the core component of
organisational culture and the ultimate source of behaviour of
an organisation.

Glendon and Stantons


model for conceptualising
safety culture and safety
climate (Glendon & Stanton,
2000)

Locates safety culture within its parent concept of


organisational culture and safety climate within its parent
concept of organisational climate. Proposes that safety culture
and climate are related yet distinct entities, with safety climate
being a component of safety culture.

Guldenmunds safety culture


framework (Guldenmund,
2000)

Views safety climate as a manifestation of safety culture.


Distinguishes three levels of safety culture:
The `core: basic underlying assumptions which are
unconscious and unspecified;
The middle layer: represents safety climate and
encompasses espoused values regarding hardware,
software, people and behaviour.
The outermost layer: are artefacts which are visible
manifestations of safety culture such as inspections, use or
non-use of personal protective equipment.
Describes three key components of safety culture:
Person (safety climate): individual and group beliefs,
attitudes, perceptions and values about safety.
Job: observable safety-related behaviours.
Situation: objective situational factors such as organisation
policies, management systems, operating procedures and
communication styles.
Proposes a safety culture model specific for the construction
industry by incorporating construction project elements, such
as construction site safety conditions, into the `situation
component of Coopers reciprocal safety culture model (Cooper
1993, 1997, 2000).

Coopers reciprocal safety


culture model (Cooper, 1993,
1997, 2000)

Choudhry, Fang and


Mohameds model of
construction safety culture
(Choudhry et al. 2007)

Flins model of safety climate


and injury outcomes for
healthcare industry (Flin,
2007)

Proposes pathways for worker and patient injury. Distinguishes


practices of senior managers from workgroup supervisors and
depicts errors as a consequence of workers unsafe behaviours
and as a predisposing factor for both worker and patient injury.

ISCRR Research Report#: 060-1215-R04

Page 5 of 36

Theory or model presented in


this report

Brief description

Gellers total safety culture


model (TSC) (Geller, 1994)

Proposes three dynamic and interactive factors: person,


environment and behaviour. Argues that changes in one factor
will eventually impact the other two. Theorises that an actively
caring factor - employees acting to optimise the safety of other
employees - is an essential part of safety culture.

Griffin and Neals model of


safety climate (Griffin & Neal,
2000)

Proposes that knowledge, skill and motivation mediate the


relationship between safety climate and safety performance.
Distinguishes compliance to safety rules and regulations which
are active practices from safety participation which are
proactive practices and represent citizenship behaviours.

Helmreich and Merritts


safety culture model
(Helmreich & Merritt, 1998)

Distinguishes four types of culture: national culture,


professional culture, organisational culture and safety culture.
Theorises that national culture is the least amenable-to-change
component that has a direct influence on both organisational
culture and safety culture through communication style,
leadership, style of training and nature of training delivery, and
reverence for procedures.

Kennedy and Kirwans safety


culture model (Kennedy &
Kirwan, 1998)

Proposes a four-layer model in which organisational culture is


the inner most layer, safety culture and safety climate are the
middle layers, and safety management practices represent the
outer most layer. Hypothesises that safety culture failures will
manifest themselves in safety management failures.

Reasons safety culture


model (Reason, 1997)

Describes safety culture as a combination of five subcultures:


an informed culture, a reporting culture, a just culture, a flexible
culture and a learning culture. Argues that safety culture is
much more than the sum of its parts.

A multilevel model of safety


climate (Zohar & Luria, 2005)

Defines safety climate as a multi-level construct


operationalised at both organisational and group levels, with
organisational-level safety climate (perceptions of senior
managers) predicts group-level safety climate (perceptions of
workgroup supervisors) which in turn predicts employee safety
behaviour.

Reasons `Swiss Cheese


model (Reason, 1997)

Explains the path from workplace hazards to organisational


safety incidents. Theorises that several layers of mutually
supporting defence are required to guard against
organisational safety incidents. Theorises that these layers,
and the safety system as a whole, are shaped by an
organisations safety culture.

ISCRR Research Report#: 060-1215-R04

Page 6 of 36

Theory or model presented in


this report

Brief description

Theory of high reliability


organisation (HRO) (Weick &
Sutcliffe, 2001)

Describes five characteristics of HROs:


Preoccupation with failures rather than successes.
Reluctance to simplify interpretations.
Sensitivity to operations (ensuring early detection and
prompt rectification of deficiencies).
Commitment to resilience. Deference to expertise (allowing
people with the most expertise to deal with problems within
their area of expertise).
Theorises that man-made disasters develop through six distinct
stages:
Stage 1: notionally normal starting point with risks at an
acceptable level.
Stage 2: incubation period in which warning signals
accumulate but go unnoticed or are misunderstood.
Stage 3: precipitating event in which a dramatic event
occurs, creating a large scale disruption.
Stage 4: accident1 onset leading to direct and
unanticipated consequences.
Stage 5: rescue and salvage with rapid and ad hoc situation
analyses undertaken.
Stage 6: cultural readjustment resulting in norms being
adjusted and a new level of precautions being established.

Turners theory of man-made


disaster (Turner, 1978)

Two main approaches have been used to develop safety culture and safety climate measures:

a model-based approach in which the framework used to guide the development of a


measure is a specific conceptual model or theory of safety culture, safety climate or a related
construct; and
a pragmatic approach in which elements of different conceptual models and theories of
safety culture, and/or safety climate and/or related construct(s) are used to inform the
development of a measure.

The model-based approach has the advantage of being guided by specific theoretically grounded
models, enabling a defined set of attributes or dimensions to be identified and considered in the
development of measures. Another advantage of this approach is that it helps focus the
measurement process.
The pragmatic approach allows researchers/practitioners to design their measures to meet specific
measurement needs. However, since the conceptual domain is not specifically defined, measures in
Word used by author. Consistent with standards in safety literature, in this report we avoid using the term
`accidents wherever possible, unless it was specifically used by authors in their publications, because this term
implies an unavoidable event (Davis & Pless, 2001). Instead, the term `incident is used to denote a range of events
that may cause unintended injuries.
1

ISCRR Research Report#: 060-1215-R04

Page 7 of 36

this category sometimes incorporate dimension(s) which may not be universally accepted as being
related to safety culture or safety climate.
The conceptual domain used to inform the development of a measure is an important consideration
in the choice of a safety culture or safety climate measure because this domain would likely
influence the nature and scope of measurement. Organisations interested in safety culture or safety
climate measurement should make sure that this conceptual domain is consistent with their
definition of the construct of interest.
It should be noted that this report is a snapshot review of theories and models frequently used in the
development of safety culture and safety climate measures. The report is intended to accompany
the review of safety culture and safety climate measures (Vu & De Cieri, 2015) and is not a
systematic review of theories and models on safety culture and safety climate.

ISCRR Research Report#: 060-1215-R04

Page 8 of 36

1. Introduction
Safety culture and safety climate have been extensively researched in the past 30 years by scholars
from diverse disciplines and numerous theories and models have been proposed to conceptualise
safety culture and safety climate and describe the relationships these constructs have with each
other, with related constructs and with safety outcomes. These theories and models enable a set of
potentially relevant attributes or dimensions to be identified and considered in the development of
safety culture and safety climate measures. A consensus on a suitable set of dimensions, however,
is still lacking and it is necessary to clarify conceptual issues before considering measurement
approaches and measures selection. The aims of this report are to provide:

a summary of key theories and conceptual models related to the development of safety
culture and safety climate measures; and
an examination of approaches to measurement of safety culture and safety climate.

This report has been written to be a companion to the systematic review of safety culture and safety
climate measures (Vu & De Cieri, 2015) commissioned by WorkSafe Victoria (WSV). These
publications are part of a larger research program supported by WSV via the Institute for Safety
Compensation and Recovery Research (ISCRR) to investigate safety culture and safety climate.

2. Defining safety culture and safety climate


Debates continue in the safety science literature regarding what the concepts of safety culture and
safety climate are, and whether they are distinct or interchangeable terms (e.g., Blewett, 2011;
Choudhry et al., 2007; Guldenmund, 2000). Researchers, however, appear to have increasingly
more clarity about safety climate and its measurement and thus a stream of research has
distinguished safety climate from safety culture, while still recognising that they are related concepts
(e.g., Cox & Flin, 1998; Flin, 2007; Zohar, 2010).
In the absence of a consensus on definitions and for the purpose of this report, safety culture and
safety climate are considered as two distinct concepts. Safety culture is defined following the (US)
NORA Construction Sector Council (2008) and safety climate is defined following Wiegmann et al.
(2002) (see Box 1).

3. Methods
Vu and De Cieri (2015) identified and evaluated safety culture and safety climate measures
according to pre-defined selection and evaluation protocols. The evaluation protocol follows
guidelines for scale development (Hinkin, 1995; MacKenzie, Podsakoff & Podsakoff, 2011) and
assesses the content validity, construct validity and reliability of included measures. As part of the
evaluation of the content validity of included measures, the theoretical basis of each measure was
evaluated and documented. Key theories and models identified through this process are examined
in the current report. This report provides a snapshot, not an exhaustive review, of theories and
models used in the development of safety culture and safety climate measures. It should be noted
that not all authors of measures included in the review by Vu and De Cieri (2015) reported the
conceptual foundation of their measures.

ISCRR Research Report#: 060-1215-R04

Page 9 of 36

4. Conceptual models for understanding and measuring safety culture and safety
climate
Fifteen theories and models of safety culture, safety climate and related constructs are examined in
this report. Eleven of these theories and models are specifically about safety culture and/or safety
climate.
A description of each of the 15 key theories and models is provided below. Scheins (1985)
model of organisational culture is presented first, followed by the presentation of theoretical
models/frameworks based on Scheins model. From then on, theories and models are presented in
alphabetical order by author, commencing with Coopers (1993) reciprocal safety culture model.

4.1

Scheins model of organisational culture

Safety culture and safety climate are often discussed by reference to Schein's model of
organisational culture. Schein is a key figure in the body of literature on organisational culture.
Schein (1985, p. 9) defines organisational culture as collective ways of acting, thinking and feeling:
The pattern of shared basic assumptions - invented, discovered or developed by a given
group as it learns to cope with its problems of external adaptation or internal integration that has worked well enough to be considered valid and, therefore, to be taught to new
members as the correct way to perceive, think and feel in relation to those problems.
According to Schein (1985) three levels of organisational culture may be discerned: artefacts,
beliefs and values, and basic assumptions. Artefacts are the most visible manifestations of
organisational culture and include elements such as dress codes, symbols, structures, written
documents, and observable rituals, rewards and ceremonies. The next level of manifestations of
organisational culture is stated beliefs and values which may be found in an organisations
mission/vision statement or strategy. These beliefs and values relate to what is judged to be
important, what is valued and what is considered to be acceptable or unacceptable. Basic
assumptions represent the least visible, albeit core component of organisational culture and are
formed mainly from beliefs and values that have been internalised through shared experience.
These assumptions are largely unspoken and less subject to discussion than espoused beliefs and
values. According to Schein, the influence of these assumptions on perceptions and reasoning is
the ultimate source of behaviour of an organisation.
Schein (1985) theorises that when all levels of organisational culture are in agreement, a
homogeneous organisational culture can be observed. Schein further theorises that an organisation
with a homogeneous culture should perform better than an organisation that is culturally divided.
Scheins later work (1992) suggests the existence of three sub-cultures within an organisation that
are important to distinguish: the operator culture, the engineer culture and the executive culture.
The operator culture is associated with workers who are involved in frontline service delivery and is
characterised by team work, communication and trust. The engineer culture involves `designers and
technocrats who develop the processes by which an organisation delivers its products and
services, and maintains itself. This subculture tends to rely on technological solutions to create
reliable and efficient processes, and values connections with professional groups outside the
organisation. The executive culture is associated with `executive management, the chief executive
officer and his or her immediate subordinates, who are responsible for the survival of the
organisation. Cost, efficiency and market share are the key viewpoints of this subculture. Changes

ISCRR Research Report#: 060-1215-R04

Page 10 of 36

in the workplace can threaten the stability of one or more of these subcultures and potentially create
conflicts; therefore, it is important to recognise the dynamics between these subcultures.
Schein's three-layer model of organisational culture has been used as the basis on which to develop
models to conceptualise safety culture and safety climate by a number of authors, including
Glendon and Stanton (2000) and Guldenmund (2000). Both of these models are safety culture
models; they are summarised below.
Scheins more recent work (1999, p. 3) pays attention to the reciprocal concept of organisational
climate, declaring that organisational culture and organisational climate are two crucial building
blocks for organizational description and analysis. Schein (1999, p.3) considers organisational
climate as a cultural artifact resulting from espoused values and shared tacit assumptions. To fully
understand climate one must dig deeper and examine values and assumptions.
According to Schein (and other scholars such as Schneider, Ehrhart & Macey, 2013), it is important
to differentiate the concept of organisational climate from the concept of organisational culture for
both practical and empirical reasons. Practical considerations relate to research designs,
measurement methods and approaches that can be used to evaluate organisational culture and
organisational climate. Some of the empirical considerations discussed by Schein include the
relationship between organisational culture, organisational climate and organisational performance
and organisational change in the context of organisational culture. It should be noted, however, that
not all authors believe that it is necessary to make a distinction between the two concepts (Blewett,
2011; Hopkins, 2006). For example, Blewett (2011, p. 7) states in her review of organisational
culture, This paper takes the middle ground and uses the terms culture and climate (both
organisational and safety) interchangeably.

4.2

Glendon and Stantons model for conceptualising safety culture and safety climate

Glendon and Stanton (2000) modified Scheins model of organisational culture by adding:

a horizontal breadth axis to depict the extent to which organisational culture is shared or
localised;
a time axis to represent the progression of organisational culture through time; and
arrows to depict the dynamics between organisational culture and organisational climate.

Glendon and Stantons (2000) conceptualisation of safety culture and safety climate is based on this
modified model, with safety culture being located with its parent concept of organisational culture
and considered to overlap but be distinct from safety climate. Additionally, the authors argue that
the dynamics between safety culture and safety climate are bidirectional. Thus, Glendon and
Stantons model addresses the limitation of Guldenmunds (2000) linear model (see Section 4.3).
The inclusion of breadth and time axes, and the incorporation of bidirectional relationship between
safety culture and safety climate mean that a sophisticated approach to measurement is required.
For example, a longitudinal study design could enable researchers to monitor the progression of
organisational culture through time. Further, multiple data sources at multiple levels would
strengthen the ability to capture the breadth of organisational culture.

4.3

Guldenmunds safety culture framework

Guldenmund (2000) conceptualises safety culture using Cox and Coxs (1991) architecture of
attitudes to safety and Scheins (1992) model of organisational culture as the basis for his

ISCRR Research Report#: 060-1215-R04

Page 11 of 36

framework. This framework has three levels: the core, the middle layer and the outermost layer.
Level 1, the core, is thought to reflect basic underlying assumptions that are unconscious,
unspecified and spread through the entire organisation. The `core is invisible but may be
inferred/derived from observations, espoused values and artefacts. Level 2, the middle layer, is
thought to represent safety climate. This level encompasses espoused values regarding hardware
(safety equipment), software (safety policies and procedures), people (workers, supervisors,
managers, safety committee, union, authorities) and behaviour. Level 3, the outermost layer,
comprises of artefacts which are visible manifestations of safety culture such as inspections,
posters, wearing (or not) of personal protective equipment, and incidents.
In this framework, Guldenmund (2000) views safety climate as a manifestation of safety culture. The
author argues that it is an integrative approach to safety culture and safety climate that recognises
that both organisational and individual attributes influence safety behaviour and safety performance.
Guldenmund (2000, p. 252) theorises that basic assumptions do not necessarily have to be specific
about safety; however, when commitment to safety is shared within the organisation, the value of
safety is firmly held within basic assumptions.
A major criticism of this framework has been its argument that safety climate may be assessed
through attitudes. Key authors in the safety climate literature have identified perceptions rather than
attitudes as the psychological feature of safety climate (Brown & Holmes, 1986; Glennon, 1982;
Niskanen, 1994; Zohar, 1980; Zohar & Luria, 2005). Furthermore, recent research results by
Pousette, Larsson & Trner (2008) lend support to the distinction between individual safety attitudes
and safety climate. Another criticism of the framework is the simple, unidirectional relationship
between assumptions and attitudes, and between attitudes and behaviours (Cooper, 2000). It has
been demonstrated that behaviours influence attitudes and are not just a consequence of attitudes
(Cox & Cheyne, 2000).

4.4

Coopers reciprocal safety culture model

Cooper (1993, 1997, 2000) developed the reciprocal safety culture model based on Banduras
(1978) model of reciprocal determinism (Figure 1), a social cognitive theory. Bandura (1978)
theorises that peoples behaviour and their environments are reciprocal determinants of each other
and that individuals could respond to their environment and correct their behaviour based upon
what they observe around them. Figure 1 depicts the reciprocal mechanism conceptualised by
Bandura (1978).

ISCRR Research Report#: 060-1215-R04

Page 12 of 36

Figure 1. Banduras (1978, p. 345) model of reciprocal determinism


Abbreviations: B: Behaviour. P: Cognitive and other internal events that can affect perceptions and
actions. E: External environment.
Note: Copyright 2000 by American Psychological Association. Reprinted in accordance with the
American Psychological Associations fair use policy.
Reciprocal relationships are also the basis of Coopers safety culture model. The key components
of Coopers safety culture model are:

Person (safety climate): Individual and group beliefs, attitudes, perceptions and values about
safety, i.e., the safety climate of an organisation.
Job: Observable safety-related behaviours, i.e., safety behaviours on the job.
Situation: This may include objective situational factors such as organisation policies,
management systems, operating procedures and communication styles, i.e., the safety
management system in an organisation.

The `job component and the `situation component in the Cooper model are equivalent to the
behaviour component and the `external environment component in the Bandura model,
respectively. According to Cooper (2000, p. 118), all three components interact reciprocally with
each other in a perpetual dynamic interplay. Cooper (2000, p. 121) suggests that it is possible to
quantify safety culture by measuring each of these components independently or in combination.
Coopers reciprocal safety culture model is broadly similar to Gellers Total Safety Culture model
(summarised below), with the term `situation used instead of `environment. These models have
been considered to be holistic in that they incorporate both psychological and environmental factors
(Cui, Fan, Fu & Zhu, 2013). Cooper (2002) asserts that his model is capable of identifying where
best to focus resources and efforts in order to pursue safety goals.
Criticisms of Coopers model have to date centred on practical issues related to behavioural
observations. Firstly, the level of resources required to observe safety-related behaviours may be
substantial and come at the expense of other areas, such as process safety (Anderson 2005).
Secondly, it has been suggested that behavioural observations may replace root cause analysis
leading to the development of quick fix and simplistic solutions. Finally, observers of unsafe

ISCRR Research Report#: 060-1215-R04

Page 13 of 36

behaviours may be reluctant to report those seen violating safety rules, resulting in underreporting
(Hopkins, 2006).

4.5

Choudhry, Fang and Mohameds model of construction safety culture

Some authors have suggested that construction industry specific models of safety culture are
required to address unique construction-sector challenges such as productivity pressure and low bid
practices, project level influences and attributes, trade sub-cultures and employer type (principal
contractor versus subcontractor) (National Occupational Research Agenda (NORA) Construction
Sector Council, 2008).
Choudhry et al. (2007) propose a safety culture model (and definition) specific for the construction
industry based on Coopers reciprocal model of safety culture (Cooper, 1993, 1997, 2000). The key
components of Choudhry et al.s safety culture model are:

Person (safety climate) for which the authors propose a perceptual audit as a measurement
method.
Behaviour (safety behaviour) which the authors suggest may be evaluated by percent safe
behaviours.
Environment/situation (safety management system) which the authors suggest may be
evaluated by project and site safety audits.

A key feature of Choudhry et al.s model which distinguishes it from the Coopers model is the
incorporation of construction project elements, such as construction worksite safety conditions, into
the `environment/situation component. The authors argue that this component should cover specific
conditions of a construction project as well as objective situational factors such as organisation
policies, management systems and operating procedures.
Choudhry et al. (2007), however, have not provided a strong rationale for why a separate safety
culture model is required for the construction industry. After all, site specificity could be considered
an aspect of the `situation component. Some authors, however, have argued that site specificity is
an important consideration given that each construction site is managed by a temporary
`organisational structure encompassing multiple levels of responsibility for safety - project
managers, principal contractors, subcontractors, supervisors and workers (Lingard, Zhang, Harley,
Blismas & Wakefield, 2014). It follows that this temporary structure may pose unique challenges in
terms of safety and that a specific safety culture model incorporating construction project elements
might be helpful.

4.6

Flins model of safety climate and injury outcomes for healthcare industry

Flin (2007) proposes a conceptual model of safety climate that covers both worker and patient
injury, making this model specific to healthcare settings. Flins model describes safety climate as a
multilevel construct encompassing practices of senior managers and workgroup supervisors. Flin
hypothesises that motivation is a product of safety climate and a proximal antecedent of workers
unsafe behaviours. The model depicts errors as a consequence of workers unsafe behaviours and
as a predisposing factor for both worker and patient injury.
Flins conceptualisation of safety climate as a multilevel construct is consistent with Zohar and
Lurias (2005) safety climate model (discussed below). Flins inclusion of individual motivation in the
safety climate-injury pathway reflects previous findings by Neal, Griffin and Hart (2000) which

ISCRR Research Report#: 060-1215-R04

Page 14 of 36

showed that worker safety knowledge and motivation partially mediated the relationship between
safety climate and safety performance. Flins incorporation of human error as a separate component
reflects the healthcare context of the model and recognises that both individual and system factors
are involved in patient and worker injury. Flin (2007) hypothesises that safety climate is similar for
both worker and patient injury. The author, however, acknowledges that different motivating factors
may influence worker behaviours giving rise to patient adverse events as opposed to (or in addition
to) worker injury.

4.7

Gellers Total Safety Culture model

Geller (1994) proposes a Total Safety Culture (TSC) model consisting of three dynamic and
interactive factors: person, environment and behaviour. Changes in one factor will eventually impact
the other two. The author advocates 10 principles as the basis of a TSC, with safety as a priority
and prevention as a major focus. These principles are as follows:

employee driven safety rules and procedures;


a behaviour-based approach;
a focus on safety processes, not outcomes;
a view of behaviour being directed by activators and motivated by consequences;
focus on achieving success, not on avoiding failure;
observation and feedback on work practices;
effective feedback through behaviour-based coaching;
observation and coaching as key activities;
the importance of self-esteem, belonging and empowerment; and
safety as a priority rather than a value.

Geller (1994, p. 19) believes that employee involvement is essential to developing a TSC. The
author suggests that employee participation in safety programs should be voluntary rather than a
bureaucratic process or regulatory policy. Gellers previous work in organisational safety research
(1991, p. 607) theorises that an actively caring factor - employees acting to optimise the safety of
other employees - is an essential part of safety culture. Actively caring for safety is defined as
going out of ones way to alert co-worker(s) about at-risk behaviour or give positive feedback to
others for performing their task safely (Geller, 2001) . Geller (2001) argues that actively caring is a
positive behaviour that would lead to a safer working environment. Frazier, Ludwig, Whitaker and
Roberts (2013) suggest that actively caring should be referred to as peer support for safety and
reports that this has been found to be one of four core constructs that make up safety culture. The
other constructs reported by Frazier et al. (2013) are management concern, personal responsibility
for safety and safety management systems.

4.8

Griffin and Neals model of safety climate

Griffin and Neals (2000) conceptualisation of safety climate is based on the literature on
organisational climate (James & James, 1989; James & Mclntyre, 1996) and work performance
(Borman & Motowidlo, 1993; Campbell, McCloy, Oppler & Sager, 1993), and emphasises the role of
senior managers in the promotion of safety. In their model, employees perceptions of management
values for safety, and policies and procedures relating to safety are hypothesised to positively
influence safety performance. This relationship, Griffin and Neal (2000) argue, may be mediated by
worker knowledge, skill, and motivation. The concept of safety performance is clarified in their

ISCRR Research Report#: 060-1215-R04

Page 15 of 36

model as including both safety compliance (active practices) and safety participation (proactive
practices).
A major limitation of their model is that the role of supervisors in the promotion of safety is not
clearly delineated from that of management (Flin, 2007). Additionally, the influence of group
practices on safety compliance and safety participation is not addressed in their model (Heritage,
2012). These omissions, on the surface, appear to disregard the multilevel nature of workplaces
and obscure the collective aspect of safety climate. However, Griffin and Neals model of safety
climate has been used in multilevel safety climate studies (Newnam, Griffin & Mason, 2008;
Brondino, Pasini & da Silva, 2013).

4.9

Helmreich and Merritts safety culture model

Helmreich and Merritts (1998) safety culture model appears to be the only model to date that
considers the influence of professional culture and national culture on safety culture. Helmreich and
Merritt (1998) discuss their conceptualisation in the context of aviation and medicine. The model
distinguishes four types of culture and describes the relationships these cultures have with one
another and with safety behaviour (directly or indirectly). The four types of cultures in the model are:

national culture,
professional culture,
organisational culture, and
safety culture.

Helmreich and Merritt (1998) view national culture as the most distal element of the model and the
least amenable-to-change component, and suggest that it has a direct influence on both
organisational culture and safety culture through communication style, leadership practiced, style of
training and nature of training delivery, and reverence for procedures. In the model, professional
culture is theorised to exert both negative and positive influences on safety culture and training.
Professional pride can lead to a strong motivation to do well. Conversely, proud workers might feel
a sense of personal invulnerability in which they might be less aware of personal limitations and less
accepting of training. Finally, organisational culture is depicted to have a direct relationship with
safety culture which in turn influences safe behaviours. Helmreich and Merritt (1998) argue that
organisational culture has an indirect influence on safety culture via training practices and a direct
impact on safety culture via management commitment to safety, and open communication and trust
between management and employees.
In this mostly unidirectional model, organisational culture appears to have no linkage with
organisational climate. In this respect, Helmreich and Merritts conceptualisation differs from that of
Schein (1985) who argue that organisational climate is a manifestation of organisational culture.
Helmreich and Merritt (1998, p. 135) state that organisational climate is distinct from organisational
culture but also confusingly make the point in the same article that organisational culture is the
major determinant of organizational climate.

4.10 Kennedy and Kirwans safety culture model


Kennedy and Kirwan (1998) conceptualise a four-layer model in which organisational culture is the
inner most layer, safety culture and safety climate are the middle layers, and safety management
practices comprise the outermost layer. This conceptualisation suggests that organisational culture
and safety culture are well hidden from view whereas safety management practices are observable

ISCRR Research Report#: 060-1215-R04

Page 16 of 36

and safety climate is close to the surface. The model does not depict the dynamics between layers
which are important for clarifying the construct of safety culture. Additionally, from a graphic
perspective the model appears to be counter-intuitive because organisational culture is enclosed in
safety culture and this may give the impression that organisational culture is a sub-component of
safety culture rather than the other way round.
Kennedy and Kirwan (1998) take the view that safety culture failures will manifest themselves in
safety management failures. They further argue that, since safety culture is intangible and abstract,
it is better to focus on safety management practices when trying to predict and prevent incidents
because these are observable and measurable.

4.11 Reasons safety culture model


Reason (1997, pp. 195-196) describes safety culture as being a combination of five subcultures:

Informed culture, defined as one in which those who manage and operate the system have
current knowledge about the human, technical, organisational and environmental factors that
determine the safety of the system as a whole. An effective safety information system is the
foundation of an informed culture.
Reporting culture, defined as one in which employees are prepared to report critical
incidents, errors and near misses, particularly their own, in a climate of trust and without fear
of reprisals. Those who provide these reports must be assured that confidentiality will be
maintained and that the information submitted will be acted upon.
Just culture, defined as one in which employees understand the delineation between
unacceptable and acceptable behaviours. Those who carry out unacceptable behaviours will
be punished by way of disciplinary action.
Flexible culture, defined as one in which the organisation has the ability to reconfigure itself
in the face of high-risk operations or certain kinds of emergency. This flexibility enables the
organisation to transfer control to task experts in a crisis, regardless of the hierarchical
nature of the organisation.
Learning culture, defined as one in which the organisation has the willingness and ability to
understand and make changes based on safety information provided internally within the
organisation and externally across the organisational interface. Among the key elements of
this subculture - observing, reflecting, creating and acting - acting is the most difficult
element to carry out successfully.

Reason (1997) argues that safety culture is much more than the sum of its parts, noting that the five
subcultures do not encompass issues such as who should be involved in the disciplinary process,
whether or not organisational safety issues should be discussed at regular board meetings and the
annual budget for organisational safety. Reason (1997, p. 220) also points out that a safety culture
is something that is striven for but rarely attained [emphasis added].
Reasons five subcultures (1997) appear to encompass the culture purportedly observed in high
reliability organisations first discussed by Weick (1987). High reliability organisations (HRO),
according to Weick (1987), have a culture that encourages interpretation, a climate of trust and
openness between management and worker, improvisation and unique action. The theory of HRO,
however, has attracted much criticism because it was developed based on a sample of
organisations identified not by HRO status but by convenience and due to the lack of evidence

ISCRR Research Report#: 060-1215-R04

Page 17 of 36

showing HRO are safer than non-HRO (Hopkins, 2014). Further information on the theory of HRO is
provided below.

4.12 Reasons `Swiss Cheese model


Reasons (1997) Swiss Cheese model (Figure 2) is often used to explain the path from workplace
hazards to organisational accidents2. In this model, Swiss cheese slices represent layers of
defence, holes in the cheese slices represent system vulnerabilities and the path from hazards to
accidents is represented by an arrow. System vulnerabilities arise from errors, unsafe acts or
organisational deficiencies, and over time, these compromise the integrity of the safety system.
Catastrophic organisational accidents occur when errors and deficiencies develop in every layer of
defence. Reason theorises that several layers of mutually supporting defence are required to guard
against organisational accidents. These layers, and the safety system as a whole, are shaped by
an organisations safety culture. Defence layers consist of technical, organisational as well as
people-based measures. Examples of defence layers are technical devices, physical barriers,
protective equipment, system design, regulatory rules, training and supervision.

Figure 2. Reasons (2000, p. 768) `Swiss Cheese model


Copyright 2000 by British Medical Journal. Reprinted with permission.
Reason (1997) distinguishes two types of system vulnerabilities: latent conditions and active
failures. Latent conditions - such as poor design, impossible procedures and maintenance failures lie dormant in the system and thus do not immediately trigger an accident. Active failures are unsafe
acts, violations and errors by workers in the front line that can trigger accidents. Latent conditions
may influence the likelihood of a future accident by contributing to or combining with active failures
to produce errors (Reason, 2000). Reason (2000) points out that active failures are difficult to
predict and mitigate whereas latent conditions can be foreseen and remedied before an error, an
unsafe act or a violation occurs.

Reason (1997) differentiates two types of accidents: individual and organisational. Reason (1997, p.1) defines
organisational accidents as catastrophic events that occur within complex modern technologies such as nuclear
power plants, commercial aviation, the petrochemical industry, chemical process plants, marine and rail
transport, banks and stadiums.
2

ISCRR Research Report#: 060-1215-R04

Page 18 of 36

4.13 Theory of high reliability organisation


High reliability organisations (HROs) are commonly defined as those organisations that handle
highly complex technologies yet have fewer organisational accidents than might be expected.
Commonly cited examples of HRO include air traffic control systems, nuclear power plants and
aircraft carriers. Weick and Sutcliffe (2001) describe five characteristics of HROs, as follows:

Preoccupation with failures rather than successes: reports of errors are encouraged, and
signs and symptoms of failure are looked for.
Reluctance to simplify interpretations: alternative views are actively sought to avoid blind
spots.
Sensitivity to operations: deficiencies are detected early and remedial actions taken.
Commitment to resilience: errors do not disable the system; instead they increase
knowledge, create opportunities for learning and bring people with various experience
together to solve problems.
Deference to expertise: a fluid decision-making system is created which allows people with
the most expertise to deal with problems within their area of expertise.

Together these characteristics create a state of collective mindfulness, a group-level capacity to


discover and manage unexpected events through an interrelated awareness-action process that is
effective in detecting errors and preventing them from being accumulated (Weick, Sutcliffe &
Obstfeld, 1999). For this reason, HROs are sometimes referred to as mindful organisations. The
theory of HRO has two key concepts: organisational redundancy (duplication and overlap) and
ability to reconfigure in times of crisis or high demands. Proponents of the theory of HRO claim that
safety-related incidents can be prevented by good organisational design and management, and that
redundancy3 enhances safety (Downer, 2009).
As discussed above (with regard to Reasons model of safety culture) some authors have
questioned the evidence base of the theory of HRO, arguing that the theory was developed based
on a limited number of case studies (The Health Foundation 2011; Hopkins, 2014). Furthermore,
significant gaps exist in the literature regarding the links between high reliability and safety culture,
whether HRO are safer than non-HRO and how organisations develop into HROs (The Health
Foundation 2011; Hopkins, 2014).

4.14 Turners theory of man-made disaster (1978)


Turners theory of man-made disaster provides a framework for studying the origins of disasters
resulting from human errors or negligence. Turner systematically analysed disasters from a sociotechnical perspective and identified six distinct stages, as follows:

Stage 1: Notionally normal starting point: individuals and groups adhere to culturally
accepted norms to keep risks at an acceptable level.
Stage 2: Incubation period: danger or warning signals and events accumulate but they go
unnoticed or are misunderstood.
Stage 3: Precipitating event: the occurrence of a dramatic event creates a large scale
disruption and challenges beliefs about risks developed during the incubation period.

In the context of engineering practice, redundancy means having several backup elements ready to take over
when an element fails (Downer, 2009).
3

ISCRR Research Report#: 060-1215-R04

Page 19 of 36

Stage 4: Accident onset: system failure occur leading to direct and unanticipated
consequences.
Stage 5: Rescue and salvage: rapid and ad hoc situation analyses are undertaken leading to
a preliminary stage of cultural readjustments.
Stage 6: Full cultural readjustment: a full assessment is carried out resulting in norms being
adjusted and a new level of precautions being established.

Turner believes that safety-related incidents and disasters occur following a long chain of events
which can be traced back to the incubation period. Based on a systematic analysis of 84
accidents, Turner argues that this period is characterised by misperceptions, miscommunication of
risks/hazards and lack of information flow among individuals and groups which result in small errors
being accumulated over time. Turner, therefore, theorises that phenomena commonly viewed as
elements of organisational culture are at the heart of system vulnerability. Turner views the
consequences of accident as encompassing both the physical and sociological impacts.
Sociological impacts include significant disruption or collapse of existing cultural beliefs and norms
about hazards which have been embedded in working practices.
Turners theory has been drawn upon by various authors to support research on the human and
organisational factors in safety-related incidents (Reason, 2001). This theory, however, has also
been criticised for its use of the `chain-of-events notion to explain safety-related incidents and
disasters. This notion implies linear and direct causal relationships between events and does not
appear to account for indirect or non-linear relationships and feedback loops.

4.15 Zohar and Lurias (2005) multilevel model of safety climate


Zohar is an influential theorist in the safety climate literature, with a research record spanning more
than three decades. Zohars (1980, p. 96) conceptualisation of safety climate, based on Schneiders
(1975) work organisational climate, is a summary of molar perceptions that employees share about
their work environments [] these perceptions have a psychological utility in serving as a frame of
reference for guiding appropriate and adaptive task behaviors. In view of the multiple, albeit
conflicting goals organisations may want to pursue - productivity, quality, safety - Zohars
subsequent research positions safety in the context of competing demands and frames safety
climate as employees evaluating the relative priority of safety such that the overall level of safety
climate represents the shared perceptions of the priority of safety compared to other competing
priorities (Zohar 2010, p. 1518). Another key issue identified by Zohar is that climate perceptions
should be about actual safety practices rather than formally declared safety policies and procedures
because only readily observable actions inform employees of behaviours that are likely to be
rewarded and supported (Zohar, 2008).
Zohars most significant contribution to the evolution of the conceptualisation of safety climate, in
collaboration with Luria, is the development of a multi-level model (Zohar, 2000, 20084, 2010; Zohar
& Luria, 2005) redefining safety climate as a multi-level construct and suggesting group-level and
organisation-level analyses using separate measurement scales. Until recently it has been a
In the 2008 paper Zohar hypothesises that a positive safety climate combined with employee ownership of work
(work ownership climate) leads to safety citizenship behaviours (safety behaviour(s) above and beyond what is
required of workers). Both work ownership climate and safety citizenship behaviours are outside the scope of this
snapshot review. The reader who is interested in these two related constructs should consult references such as
Zohar (2008) for work ownership climate; and Didla, Mearns and Flin (2009) for safety citizenship behaviours.
4

ISCRR Research Report#: 060-1215-R04

Page 20 of 36

common practice in safety climate research for supervisor and manager commitment to safety to be
assessed using the same scale (for example Cooper and Phillips, 2004; Cox & Cheyne, 2000).
Meli and Ses (2007, p. 232) argue that, the particular contribution of each of these main sources
of influence [supervisors, managers, co-workers, safety officers] should be measured separately. In
the multi-level model, Zohar (2008, p. 378) argues that supervisors have discretion in their
interpretation and implementation of safety policy, and that individuals, acting as members of an
organisation and a work unit of that organisation, distinguish between global and local emphasis on
safety.
Zohar and Lurias multilevel model of safety climate is supported by empirical data pertaining to
workers perceptions of their supervisor and managers commitment to safety from 412 work-groups
nested in 40 manufacturing companies, measured separately and aggregated to the work unit and
organisational level respectively (Zohar & Luria, 2005). The data show large safety climate
variations within the same company (meaning some managers care more about safety than do
others) and large safety climate variations between companies in the same sector (implying some
companies care more about safety than do others) (Zohar & Luria, 2005). This multi-level model has
been applied in numerous contexts, including manufacturing (for example Zohar & Luria, 2005;
Kapp, 2012), nursing (Leiter, 2011), oil and gas (Heritage, 2012), and construction supply chains
(Lingard, Cooke & Blismas, 2012).
Another significant contribution by Zohar, in collaboration with Luria, has been their
conceptualisation of safety climate metrics (Zohar & Luria, 2004). Zohar and Luria (2004) propose
that safety climate can be measured by two metrics - climate level being the mean score of
aggregated perceptions (high or low) and climate strength being the degree of consensus in an
organisation or workgroups regarding the priority of safety (strong or weak). Zohar and Luria (2004,
p. 330) argue that changes in the [work] environment (i.e., policies, procedures, practices) lead to
corresponding changes in climate level and strength.
Zohar and Lurias (2005) multi-level model on safety climate appears to integrate well with the multilevel theory of organisational behaviour and current work related to multi-level theorising in
organisational research (Hitt, Beamish, Jackson & Mathieu, 2007; Mathieu & Chen, 2011;
Schneider et al., 2013). A major advantage of a multi-level over a single-level theory is that scores
on workers perceptions of their supervisor and managers commitment to safety may be used to
identify the level(s) of concern within the organisation and develop appropriate safety interventions5
directed at this/these level(s). Zohar (2010) recently suggests that the multi-level theory could be
extended to include mid-level management as a source of behaviour-outcome expectancies.
Other authors - such as Heritage (2012); Geddes (2012) and Brondino et al. (2013) - propose the
inclusion of individual-level perceptions of co-workers because they are an important source of
behavioural norms in the workplace (Roy, 2003; Zhou, Fang & Wang, 2008). Two groups of studies
provide support for this view. The first group of studies applied social exchange theory to safety
research to explain workers propensity to care about safety as a means to reciprocate the support
Dyreborg, Nielsen, Kines, Rasmussen, Olsen, Lipscomb ... Spangenberg (2011, p. 4) define a safety intervention as
[a] measure deliberately applied to decrease the frequency or severity of accidental injuries in the work place []
such as a safety campaign within the workplace aimed at changing attitudes, or safety training aimed at changing
behaviour or it can consist of a combination of such components.
5

ISCRR Research Report#: 060-1215-R04

Page 21 of 36

shown to them by their supervisors and co-workers (Oliver, Cheyne, Toms & Cox, 2002;
McGonagle, Walsh, Kath & Morrow, 2014). The second group of studies investigated the link
between co-worker support and individual safety perceptions or safety behaviours. Hayes,
Perander, Smecko and Trask (1998) showed that co-worker safety strongly correlated with
employees compliance with safety behaviours. McCaughey, Halbesleben, Savage, Simons and
McGhan (2013) found that co-worker support positively moderated the relationships between
supervisor safety leadership and safety perceptions, supervisor safety leadership and work unit
safety, and senior management safety leadership and safety perceptions. These relationships are
congruent with those depicted in Zohar and Lurias (2005) multi-level model and support the
inclusion of co-workers influences in the safety climate-injury pathway.
Nevertheless, the incorporation of a co-worker level into Zohar and Lurias (2005) multi-level safety
climate model to reflect the hierarchical nature of workplace structure would necessitate a more
sophisticated approach to safety climate measurement requiring data to be collected at three
different workplace levels, and analysed and interpreted appropriately. In this context, the model
might be less applicable to organisations with a less-hierarchical structure.

4.16 Summary remarks


Theories and models used to provide the conceptual foundations for the development of safety
culture and safety climate measures come from diverse academic disciplines and this diversity
explains the considerable variability in which safety researchers and practitioners can frame, define
and study safety culture and safety climate. To date, none of these theories or models has been
universally accepted as clearly articulating the construct domain of safety culture and safety climate
and no single model or theory may be applicable to all types of organisations (Guldenmund, 2000).
Huang, Zohar, Robertson, Garabet, Lee and Murphy (2013) argue that not all workplaces are
`organisations with traditional work environments in which workers interact with one another and
with their supervisors under the same roof. Indeed, some workplaces consist of a very small
number of individuals or are just individuals working independently (lone workers). Huang et al.
(2013) note that while non-traditional work environments have become increasingly more prevalent,
our conceptualisation of safety culture and safety climate is still limited to traditional `organisations
and the dearth of research on safety culture and safety climate in non-traditional work settings6
should be addressed.
Nevertheless, it is generally accepted now that safety culture is a subset of organisational culture
which is unconscious and invisible whereas safety climate is a snap shot or manifestation of safety
culture that is observable, temporal in nature, subject to change and close to the surface
(Guldenmund, 2000; Flin, Mearns, O'Connor & Bryden, 2000).

4.17

Implications for measurement

The theories and models discussed above propose somewhat different hypothetical causal
pathways for linking safety culture and/or safety climate to safety outcomes, with each theory or
model identifying a slightly different set of antecedents, moderators and mediators. Thus, these
theories and models draw attention to slightly different aspects of the same constructs and provide
different frameworks for safety culture and safety climate measurement and intervention. It is these

Non-traditional work settings include lone working, remote/mobile working and teleworking.

ISCRR Research Report#: 060-1215-R04

Page 22 of 36

differences in the conceptualisation of safety culture and safety climate that, in part, account for the
proliferation of safety culture and safety climate measures.
A common feature across the different theories and models discussed above is that they support
the conceptualisation of safety culture and safety climate as multidimensional constructs. According
to Law, Wong and Mobley (1998) a multidimensional construct consists of a number of interrelated
attributes or dimensions [] conceptualised under an overall abstraction and it is theoretically
meaningful and parsimonious to use this overall abstract as a representation of the dimensions. A
key challenge for safety culture and safety climate measurement, therefore, is the identification of a
common set of attributes or dimensions that are applicable across industries (Cox & Flin, 1998; Flin,
2007) . Some authors have also advocated for the identification of industry-specific attributes or
dimensions, arguing that once such attributes or dimensions become known it would be possible to
develop and test hypotheses regarding the process of safety climate emergence (Choudhry et al.
2007; Zohar, 2010).
Finally, the theories and models discussed above suggest that safety culture and safety climate are
likely to be influenced by interactional processes at multiple levels. Thus, it is important to identify
the key drivers of safety culture and safety climate at each level and investigate cross-level
influences. Recent work by Zohar (Zohar, 2008, 2010; Zohar & Luria, 2005) suggests that safety
climate is a multi-level construct which should be measured at both group and organisational levels
using separate measurement scales. As discussed above, further enhancements to this model have
been suggested, including the incorporation of co-workers influences as the most proximal level of
the safety climate-injury pathway (Geddes, 2012; Brondino et al., 2013). Geddes (2012, p. 66) asks,
since safety climate reflects collective assessments of safety conditions in the workplace, should
our operationalisation not reflect the practices of employees at all work-levels, not just those of
managers and supervisors? [emphasis added].

5. Approaches used to measure safety culture


Although scholars disagree on how to define safety culture, they appear to be in consensus that this
construct is difficult to assess precisely (Guldenmund, 2000; Nielsen, 2014; Patankar & Sabin,
2010). Recalling Schein's three-layer model of organisational culture (Schein, 1985), it is possible to
explain why this is the case. The innermost layer of this model represents the core component of
culture which is invisible, unconscious but may be inferred from espoused values (the middle layer)
and artefacts (the outermost layer). The middle layer of the model represents stated beliefs and
values which may be assessed quantitatively, albeit imperfectly since these internal states are
difficult to observe directly (Schein, 1985). The outermost layer of the model is artefacts of culture
which are visible and easily measured. However, Schein (1984, p. 3) argues that data on artefacts
are easy to obtain but hard to interpret without digging below the surface to uncover why" a group
behaves the way it does. Schein (1984, p. 5) further cautions, To stay at the level of artefacts or
values is to deal with the manifestations of culture, but not with the cultural essence.
Digging below the surface to answer the `why question, according to Schein (1984, 1991), may be
accomplished by a multi-level, multisource design - studying organisational socialisation of new
members; analysing an organisation's history and incident response; and analysing the
assumptions and values of the founder(s) and leader(s) of this organisation. In-depth-interviews
may provide one of the key types of data for this assessment, which when combined with

ISCRR Research Report#: 060-1215-R04

Page 23 of 36

quantitative data on artefacts (data triangulation), may reveal the cultural essence of organisations
(Schneider et al., 2013).
Several scholars have suggested that a multi-level, multisource design is suitable for studying
safety culture (Glendon & Stanton, 2000; Guldemund, 2000; Haukelid, 2008; Nielsen, 2014; Oltedal,
2011; Zohar, 2010). In reviewing this literature two main approaches have been identified:

a model-based approach in which the framework for the measurement is a specific


conceptual model of safety culture, and
a pragmatic7 approach in which a combination of conceptual models and theories is used to
provide the conceptual foundation for the measurement.

Each of these approaches is described below.

5.1

Measuring safety culture using a specific conceptual model or framework

This approach involves the use of a specific theory or model to provide a framework for safety
culture measurement. Among the models and theories summarised in Section 4, Scheins model of
organisational culture (Schein, 1985) and Coopers reciprocal safety culture model (1993, 1997,
2000) have been used for this purpose. Other models and theories that have not been discussed in
Section 4 but have been used for this purpose include the balanced scorecard (BSC) model for
managers developed by Kaplan and Norton (1991), the European Foundation for Quality
Management (EFQM) model (Nabitz, Klazinga & Walburg, 2000) and safety culture maturity models
(SCMM) (for instance Lawrie, Parker & Hudson, 2006; Hudson, 2007).
Nielsen (2014) chose Scheins conceptualisation of organisational culture to model safety culture
and this approach was operationalised as follows:

artefacts were assessed through behavioural indicators, structural conditions, documents


and safety climate,
espoused values were determined through attitudes toward safety and structural conditions,
and
basic assumptions were assessed based on an analysis of the artefacts and espoused
values.

Cooper (1993, 1997, 2000) also followed the multi-level, multisource approach but developed his
own model - the reciprocal safety culture model based on Banduras (1978) model of reciprocal
determinism (Figure 2). The key components of the reciprocal safety culture model are: person
(safety climate), job (safety behaviours on the job) and situation (safety management system).
Cooper (2000) suggests that it is possible to quantify safety culture by measuring each of these
components independently or in combination. The attitudes and perceptions of individuals within an
organisation may be assessed using safety climate questionnaires, group interviews and focus
groups. Safety-related behaviours may be evaluated by checklists, risk identification tools and
composite outcome measures. Situational factors may be assessed using audits of safety
management systems, weekly inspections and environmental surveys. Cooper (2000) further
For another view on safety culture measurement approaches, the reader is referred to the paper by Guldenmund
(2010, p.1467) which describes three approaches to studying safety culture:
academic: through the use of field research or ethnography;
analytical through the use of survey questionnaires;
pragmatic: through experience and expert judgment.
7

ISCRR Research Report#: 060-1215-R04

Page 24 of 36

proposes that the reciprocal safety culture model may be used for benchmarking purposes,
comparing safety culture in different work areas in the same organisation or comparing safety
culture across organisations within an industry group. Cooper (2002) asserts that this model is
capable of identifying where best to focus resources and efforts in order to pursue safety goals.
Coopers model has been applied in the measurement of the safety culture of a number of
organisations (Cooper, 2002; Cooper and Phillips, 2004). A safety climate questionnaire has been
developed and validated (Cooper, 2002; Cooper and Phillips, 2004). This questionnaire, however, is
a proprietary tool (D. Cooper, personal communication, 12 August 2014) and therefore was not
included in the review of safety culture and safety climate measures (see Vu & De Cieri, 2015).
Among the measures included in the review by Vu and De Cieri (2015), one safety climate measure
has been based on the Cooper model: the Organizational Safety Climate Scale by Pessemier and
England (2012).
As discussed in Section 1, Coopers reciprocal safety culture model is broadly similar to Gellers
(1994) TSC model, with the term `situation used instead of `environment. Geller and colleagues
(1996) developed and validated a safety culture survey based on the TSC model; however, that
survey is a proprietary tool and hence was not included in the review by Vu and De Cieri (2015) (S.
Geller, personal communication, 26 July 2014). Among the measures included in that review, none
has been based on Gellers TSC model.
Another model that has been adapted from its original purpose and used as the basis for developing
tools purported to measure safety culture is the balanced scorecard (BSC) model for managers
developed by Kaplan and Norton (1991). This model has been developed to examine business
performance and improve management process from four perspectives: financial, customer, internal
business, and innovation and learning. Mohamed (2003) used this model to develop a scorecard for
benchmarking construction safety culture. Mohamed (2003) argues that an organisations safety
performance goals can be monitored and benchmarked using the same model with the four
perspectives slightly modified to reflect occupational health and safety (OHS) in the construction
context: management, operational, customer and learning. For each of these areas the organisation
sets safety-related goals and identifies corresponding performance measures to monitor goal
attainment.
Mohamed (2003) tested this scorecard in 62 organisations in the construction industry and reported
statistically significant positive correlations between (i) the extent of management involvement to
improve safety and hazard identification ability; (ii) the extent of management involvement to
improve safety and the number of safety reviews; (iii) the higher number of safety initiatives and the
lower number of safety incidents due to a lack of consideration for safety matters in planning; and
(iv) customer satisfaction rating and savings due to accident reduction. While these findings indicate
the potential utility of the scorecard, there may be potential limitations to its applicability to safety.
This scorecard as adapted by Mohamed (2003) does not appear to adequately cover employee
involvement, a key dimension of safety culture. For example, the scorecard does not have an item
for employee participation in safety-related decision making. Another limitation of the scorecard is
that there may be a contradiction between the `balanced aspect, which might suggest a trade-off
between productivity and safety, and the priority of safety that is central to safety management.

ISCRR Research Report#: 060-1215-R04

Page 25 of 36

Another method of assessing safety culture is to consider safety within the context of a quality
management model. For instance, Mariscal, Garca, Herrero and Otero (2012) developed a 28-page
questionnaire to assess five dimensions (28 attributes) of safety culture using the EFQM model. The
questionnaire was pilot tested in a nuclear power plant to identify `evidence of safety culture using
qualitative methods (facilitated working groups, document analysis) involving approximate 10% of
plant staff from all hierarchy levels. This approach appears to be engaging, comprehensive and
capable of exposing underlying values and assumptions regarding safety at the plant. Nevertheless,
in view of the limited testing conducted to date, this questionnaire would require further testing to
confirm its validity, reliability and utility.
Finally, several measures have been developed to assess an organisations level of maturity to
guide the development of its safety culture (Bergersen, 2003; Filho, Andrade & Marinho, 2010;
International Atomic Energy Agency, 2002; Fleming, 2000; Hudson, 2001; Hudson, 2007; Joint
Planning and Development Office Safety Working Group, 2008; Lawrie et al., 2006). Each level of
maturity is characterised by a specific set of indicators based on the behavioural and cultural
elements of an organisations safety culture. This category of measures is commonly referred to as
a safety culture maturity model (SCMM). It has been suggested that organisations may convert
information obtained from administering a safety culture checklist or questionnaire to a maturity
level using a SCMM and thus may be able to set safety goals and targets for improvement (Heese,
2012).
The `label of maturity levels varies slightly between measures. For example, Fleming (2000)
defines five maturity levels through which organisations can progress sequentially: emerging,
managing, involving, cooperating and continually improving. Fleming (2000) argues that at the
highest level of maturity (`continually improving), organisations have a consistent approach to
safety and remain vigilant about safety complacency. Hudson (2001, 2007) also defines five
maturity levels but labels them as: pathological, reactive, calculative, proactive and generative.
According to Hudson (2001, 2007) a generative culture entails the integration of safety into every
aspect of operations, collective learning, systems thinking and continuous safety improvement. A
comparison of the similarities and differences between the levels in the two SCMMs is outside the
scope of this review. It should be noted that both measures are proprietary products.
The theoretical models providing the basis for the development of SCMMs include Westrums model
of organisational typologies (Westrum, 2004) and Reasons model of safety culture (1997).
Flemings SCMM (2000) was based on the maturity model concept used in the field of software
engineering as a framework to improve software development processes. It thus appears that
SCMMs are theoretically grounded. Little evaluation, however, has been conducted regarding what
constitutes a good or a bad safety culture (Lawrie et al., 2006).
In summary, the model-based approach to measuring safety culture has the advantage of being
guided by theoretically grounded models, enabling a defined set of potentially relevant variables to
be identified and considered in the development of measures. Another advantage of this approach
is that it helps focus the measurement process. Many of the safety culture models discussed in
Section 1 have been used as the sole basis for a number of safety culture measures. Management
models have also been employed to develop safety culture measures. These measures, however,
have not been widely tested and may require further validation.

ISCRR Research Report#: 060-1215-R04

Page 26 of 36

5.2

Measuring safety culture using a pragmatic approach

The majority of measures designed to purportedly measure safety culture have not been developed
by formally developing or specifying a measurement model (Flin et al., 2000; Heritage, 2012).
Instead, several researchers/practitioners have employed a pragmatic approach in the development
of questions or statements for inclusion in their measures by:

reviewing the safety literature, and/or


conducting focus groups, and/or
interviewing academic and non-academic subject matter experts, and/or
adapting previously published measures.

Once a bank of suitable questions or statements has been selected for inclusion in the measures, a
typical procedure is for the measures to undergo pilot testing with a small number of participants
after which they might be revised before being further validated with a larger number of participants
in the same or different setting(s). Measures that have been developed using this pragmatic
approach include checklists/prompt lists (for example Piers, Montijn & Balk, 2009) and
questionnaires (for example Ostrom, Wilhelmsen & Kaplana, 1993; Harvey, Erdos, Bolam, Cox,
Kennedy & Gregory, 2002; Hvold, 2005; Hay, 2010; Lawrie et al., 2006).
The pragmatic approach allows researchers/practitioners to design their measures to meet specific
measurement needs (Patankar et al., 2010). However, since the conceptual domain or
measurement model is not specifically defined, measures in this category sometimes incorporate
dimension(s) which may not be universally accepted as being related to safety culture. For instance,
Lee (1998), Lee and Harrison (2000) and Harvey et al. (2002) included job satisfaction as a
dimension in their safety culture questionnaires. Harvey et al. (2002, p. 22) claim that job
satisfaction is related to safety culture because it is to some extent a function of management style
and commitment. It would appear, however, that while job satisfaction may be related to
management commitment to safety, it might be independent of it and is influenced by a multitude of
factors, including the nature of the job, individual factors and other organisational factors (Wilkin,
2013; Ziegler, Hagen & Diehl, 2012; Tietjen & Myers, 1998). Beus, Payne, Bergman and Arthur
(2010) refers to the inclusion of unrelated construct(s) in a safety culture (or safety climate) measure
as content contamination.
The correspondence between the conceptualisation of safety culture, and its definition and
measurement is an important consideration for researchers and organisations as well as OHS
regulators. It would therefore be essential to consider this as a criterion for selecting safety culture
measures, particularly for measures developed using a pragmatic approach. Operationalising this
criterion, however, might not be straight forward because a consensus on safety culture definitions
does not exist and a common set of safety culture dimensions is not currently available (Vu & De
Cieri, 2014).
Nevertheless, what is currently agreed by scholars regarding the key conceptual elements of safety
culture may be used to facilitate this operationalising process. Scholars appear to be consistent in
theorising that safety culture is a subset of organisational culture which is unconscious and invisible
whereas safety climate is a snap shot or manifestation of safety culture that is observable and close
to the surface (Helmreich & Merritt, 1998; Glendon & Stanton, 2000; Guldenmund, 2000).
Additionally, it is broadly agreed that safety culture has three aspects that complement each other:
organisational, individual and collective (Cooper, 1993, 1997, 2000; Geller, 1994). The most

ISCRR Research Report#: 060-1215-R04

Page 27 of 36

commonly discussed elements of the organisational aspect of safety culture appear to be


management practices and structures, and communication (e.g., Gadd & Collins, 2002; Wiegmann,
Zhang, von Thaden, Sharma & Gibbons, 2004). Scholars also appear to be in agreement that the
individual (person) aspect of safety culture includes elements such as individual behaviour,
knowledge, skills and ability while the collective aspect includes elements such as group dynamics,
and collective practices and vision (e.g., Cooper, 1993, 1997, 2000; Patankar & Sabin, 2010;
Reiman & Rollenhagen, 2014).
Another point to note is that measures in this category mostly measure safety climate - artefacts
and espoused values regarding safety, the outermost layer and the middle layer of safety culture,
respectively. Basic assumptions; the innermost layer of safety culture which is unconscious,
unspecified and invisible; may be inferred from espoused values and artefacts but the inference
may be incomplete and may misrepresent the rich, complex nature of safety culture (Mearns, 2013).
Schein (1999, p. 11) argues that this type of measures cast the theoretical net too narrowly []
thereby limit the domain of inquiry. Thus, the label `safety culture that has been given to many of
these measures is perhaps a misnomer or otherwise a consequence of (i) authors being unwilling to
make a distinction between safety culture and safety climate (e.g., Blewett, 2011; Hopkins, 2006) or
(ii) the two terms being used interchangeably (e.g., Minerals Council of Australia, 1999; Wu, Lin &
Shiau, 2010). Qualitative methods, such as observation or in-depth interview, might be useful to
uncover basic assumptions regarding safety culture and could be used in conjunction with the
administration of a survey questionnaire (Antonsen, 2009; Guldenmund, 2000, 2010). In other
words, a mixed methods research design drawing qualitative and quantitative data from multiple
sources and at multiple levels would be recommended.

6. Approaches used to measure safety climate


As noted in Section 2, many measures purported to evaluate safety culture might be viewed instead
as safety climate measures. Approaches used to measure safety climate are similar to those used
to measure safety culture, viz. two approaches are commonly used: a model-based measurement
approach and a pragmatic approach. Each of these approaches is discussed below.

6.1

Measuring safety climate based on specific conceptual models

This approach entails using a specific conceptual model pertinent to safety climate to develop safety
climate measures. Zohar and Lurias multi-level safety climate model exemplifies such a model and
is probably the most advanced theoretical model on safety climate to date (Zohar, 2010; Zohar &
Luria, 2005). In Zohar and Lurias model, safety climate is conceptualised at both organisational and
group levels and is operationalised as practices of senior managers and workgroup supervisors,
respectively. Zohar and Luria (2005) have developed and validated a tool for each of these levels.
These tools have been further validated in various settings (for instance Huang et al., 2013;
Navarro, Lern, Toms & Silla, 2013). Further enhancements to this model have been suggested,
including the incorporation of co-workers influences as the most proximal level of the safety
climate-injury pathway (Brondino et al., 2013; Geddes, 2012; Silva & Pasini, 2012). A specific tool
for co-worker safety climate has been developed and validated (Geddes, 2012).
In contrast to Zohar and Lurias (2005) multi-level model, Griffin and Neals model of safety climate
(2000) focuses on the role of senior managers. In their model, employees perceptions of
management values for safety, and policies and procedures relating to safety are hypothesised to
positively influence safety performance. The authors developed and validated a safety climate scale
ISCRR Research Report#: 060-1215-R04

Page 28 of 36

as well as scales matching other constructs in this model. This set of scales has been further
validated (see for example Wallace, 2005; Colley, Lincoln and Neal, 2013).
Zohar and Lurias safety climate scales (Zohar & Luria, 2005) and those developed and validated by
Griffin and Neal (2000) have been used in combination with scales developed by other scholars.
This combination approach might be viewed as an attempt to capture all the relevant dimensions of
safety climate. This pragmatic approach is discussed in Section 6.2 below.

6.2

Measuring safety climate based on a pragmatic approach

Similar to the pragmatic approach to measuring safety culture, a pragmatic approach to measuring
safety climate involves using a combination of elements from different theoretical models to develop
safety climate questionnaires. For example, Brondino et al. (2013) developed a safety climate
measure by combining elements of the multi-level safety climate model by Zohar and Luria (2005)
with those in the model of safety climate by Griffin and Neal (2000). Brondino et al. (2013) also
added to their questionnaire items from Meli and Ses (1998, 2007), therefore tailoring the
questionnaire to the study setting.
This is a convenient approach to developing safety climate measures; however, since the
conceptual domain is not specifically defined, measures in this category sometimes incorporate
dimension(s) which may not be universally accepted as being related to safety climate. For
example, fatalism and optimism were included in a safety climate scale developed and tested by
Williamson, Feyer, Cairns and Biancotti (1997); risk perceptions were included in the Offshore
Safety Questionnaire developed and tested by Mearns, Flin, Gordon and Fleming (1998). Flin et al.
(2000) argue that while personality dispositions may have a direct effect on safety behaviours, they
do not predict safety outcomes. As for the inclusion of perceptions of risks in safety climate scales,
the authors view is that some workers continue to take risks despite having fairly accurate
perceptions of the risks they face. This observation would suggest that this inclusion is problematic,
particularly when it is unclear whether perceptions of risks are included as a dimension of safety
climate or as factors with the potential to impact on safety climate. Griffin and Neal (2000) argue
that safety climate only refers to perceptions of workplace attributes specifically about safety and
should not include perceptions of risks.
Another controversy associated with the pragmatic approach is the inclusion of attitudinal items in
safety climate measures which are considered by some scholars to relate to safety culture (see
Guldenmund's (2000) model of safety culture in Section 4.3). Safety climate measures belonging in
this category include the Offshore Safety Questionnaire developed and tested by Mearns et al.
(1998) and the safety attitudes survey developed and tested by Lee (1998). The combination of
perceptual and attitudinal items in a safety climate survey itself is not an issue, it is the lack of clarity
regarding the conceptualisation of safety climate, safety culture and related constructs that is
potentially problematic. It thus appears that the measurement of safety climate should begin with a
clear understanding of what this construct is about, how it differs from safety culture and how it
relates to other constructs and to safety outcomes (Hinkin, 1995; MacKenzie et al., 2011).

7. Conclusion
Safety culture and safety climate have been conceptualised by scholars working in diverse
disciplines and this has led to the development of divergent theories and models, with each theory
or model identifying a slightly different set of potentially relevant attributes or dimensions for

ISCRR Research Report#: 060-1215-R04

Page 29 of 36

measurement. It is therefore important that the measurement of safety culture or safety climate
should begin with a clear understanding of what the construct is, how it differs from its alternative
construct, what should be included in the construct domain and how the construct relates to safety
outcomes. Mixed methods research designs collecting qualitative as well as quantitative data from
multiple sources, at multiple levels and at different time points could advance understanding of both
constructs.
Two main approaches have been used in the measurement of these two alternative constructs: (i) a
model-based approach in which the framework for measurement is a specific conceptual model or
theory of safety culture, safety climate and/or related construct(s), and (ii) a pragmatic approach in
which elements of different conceptual models and theories of safety culture, and/or safety climate
and/or related construct(s) are used to inform the development of a measure. The conceptual
domain used to inform the development of a measure is an important consideration in the choice of
a safety culture or safety climate measure because this domain would likely influence the nature
and scope of measurement. Organisations interested in safety culture or safety climate
measurement should make sure that this conceptual domain is consistent with their definition of the
construct of interest.

ISCRR Research Report#: 060-1215-R04

Page 30 of 36

References
Anderson, M. (2005). Behavioural safety and major accident hazards: Magic bullet or shot in the
dark? Process Safety and Environmental Protection, 83(2), 109-116.
Bandura, A. (1978). The self system in reciprocal determinism. American Psychologist, 33(4), 344358.
Bergersen, C. E. B. (2003). Tools to be used in survey and improve safety culture in the European
railway industry. (Diploma), Norwegian University of Science and Technology, Trondheim.
Beus, J., Payne, S., Bergman, M., & Arthur, W. (2010). Safety climate and injuries: An examination
of theoretical and empirical relationships. Journal of Applied Psychology, 95, 713.
Blewett, V. (2011). Clarifying culture. Canberra: Safe Work Australia.
Blewett, V., & Flower, J. (2011). Best practice in OHS culture Final draft report: University of South
Australia.
Borman, W. C., & Motowidlo, S. J. (1993). Expanding the criterion domain to include elements of
contextual performance. In N. Schmitt & A. W. C. Borman (Eds.), Personnel selection in
organizations (pp. 71-98). San Francisco: Jossey-Bass.
Brondino, M., Pasini, M., & da Silva, S. (2013). Development and validation of an Integrated
Organizational Safety Climate Questionnaire with multilevel confirmatory factor analysis.
Quality & Quantity, 47(4), 2191-2223.
Brondino, M., Silva, S. A., & Pasini, M. (2012). Multilevel approach to organizational and group
safety climate and safety performance: Co-workers as the missing link. Safety Science,
50(9), 1847-1856.
Brown, R. L., & Holmes, H. (1986). The use of a factor-analytic procedure for assessing the validity
of an employee safety climate model. Accident Analysis & Prevention, 18(6), 455-470.
Campbell, J. P., McCloy, R. A., Oppler, S. H., & Sager, C. E. (1993). A theory of performance. In N.
Schmitt & A. W. C. Borman (Eds.), Personnel selection in organizations (pp. 35-70). San
Francisco: Jossey-Bass.
Cheyne, A., Cox, S., Oliver, A., & Toms, J. M. (1998). Modelling safety climate in the prediction of
levels of safety activity. Work & Stress, 12(3), 255-271.
Choudhry, R. M., Fang, D., & Mohamed, S. (2007). The nature of safety culture: A survey of the
state-of-the-art. Safety Science, 45(10), 993-1012.
Colley, S. K., Lincolne, J., & Neal, A. (2013). An examination of the relationship amongst profiles of
perceived organizational values, safety climate and safety outcomes. Safety Science, 51(1),
69-76.
Cooper, D. (1993). Goal-setting for safety. The Safety and Health Practitioner(November), 32-37.
Cooper, D. (1997). Evidence from safety culture that risk perception is culturally determined.
International Journal of Project and Business Risk Management, 1(2), 185-202.
Cooper, D. (2000). Towards a model of safety culture. Safety Science, 36(2), 111-136.
Cooper, D. (2002). Safety culture: A model for understanding and quantifying a difficult concept.
Professional Safety(June), 30-36.
Cooper, D., & Phillips, R. A. (2004). Exploratory analysis of the safety climate and safety behavior
relationship. Journal of Safety Research, 35(5), 497-512.
Cox, S., & Cox, T. (1991). The structure of employee attitudes to safety: A European example. Work
& Stress, 5(2), 93-106.
Cox, S., & Flin, R. (1998). Safety culture: Philosopher's stone or man of straw? Work & Stress,
12(3), 189-201.
Cox, S. J., & Cheyne, A. J. T. (2000). Assessing safety culture in offshore environments. Safety
Science, 34(13), 111-129.
Cui, L., Fan, D., Fu, G., & Zhu, C. J. (2013). An integrative model of organizational safety behavior.
Journal of Safety Research, 45, 37-46.
Davis, R. M., & Pless, B. (2001). BMJ bans accidents : Accidents are not unpredictable. BMJ :
British Medical Journal, 322(7298), 1320-1321.
Didla, S., Mearns, K., & Flin, R. (2009). Safety citizenship behaviour: A proactive approach to risk
management. Journal of Risk Research, 12(3-4), 475-483.

ISCRR Research Report#: 060-1215-R04

Page 31 of 36

Downer, J. (2009). When failure is an option: Redundancy, reliability and regulation in complex
technical systems. London: Centre for Analysis of Risk and Regulation, London School of
Economics and Political Science.
Dyreborg, J., Nielsen, K., Kines, P., Rasmussen, K., Olsen, O., Limpscomb, H. J., . . .
Spangenberg, S. (2011). Safety interventions for the prevention of accidents in the
workplace. Title registration for a review protocol. The Campbell Collaboration.
Filho, A. P. G., Andrade, J. C. S., & Marinho, M. M. d. O. (2010). A safety culture maturity model for
petrochemical companies in Brazil. Safety Science, 48(5), 615-624.
Fleming, M. (2000). Safety culture maturity model Offshore Technology Report 2000/049. London:
The Keil Centre.
Flin, R. (2007). Measuring safety culture in healthcare: A case for accurate diagnosis. Safety
Science, 45(6), 653-667.
Flin, R., Mearns, K., O'Connor, P., & Bryden, R. (2000). Measuring safety climate: Identifying the
common features. Safety Science, 34(13), 177-192.
Frazier, C. B., Ludwig, T. D., Whitaker, B., & Roberts, D. S. (2013). A hierarchical factor analysis of
a safety culture survey. Journal of Safety Research, 45(0), 15-28.
Geddes, F. R. (2012). Managers, mates and the role of social exchange : a multilevel model of
safety climate and proactive safety behaviour. (Doctoral dissertation), Curtin University,
Perth.
Geller, E. S. (1991). If only more would actively care. Journal of Applied Behavior Analysis, 24(4),
607-612.
Geller, E. S. (1994). Ten Principles for achieving a total safety culture. Safety Management, 39(9),
18-24.
Geller, E. S. (2001). Actively caring for occupational safety: Extending the performance
management paradigm. In C. M. Johnson, W. K. Redmon & T. C. Mawhinney (Eds.),
Handbook of Organizational Performance: Behavior Analysis and Management (pp. 303
326). Binghamton, New York Haworth Press Inc.
Geller, E. S., Roberts, D. S., & Gilmore, M. R. (1996). Predicting propensity to actively care for
occupational safety. Journal of Safety Research, 27(1), 1-8.
Glendon, A. I., & Stanton, N. A. (2000). Perspectives on safety culture. Safety Science, 34(13),
193-214.
Glennon, D. E. (1982). Safety climate in organisations. Ergonomics and occupational health. Paper
presented at the 19th Annual Conference of the Ergonomics society of Australia and New
Zealand.
Griffin, M. A., & Neal, A. (2000). Perceptions of safety at work: A framework for linking safety climate
to safety performance, knowledge, and motivation. J Occup Health Psychol, 5(3), 347-358.
Guldenmund, F. W. (2000). The nature of safety culture: A review of theory and research. Safety
Science, 34(1-3), 215-257.
Harvey, J., Erdos, G., Bolam, H., Cox, M. A. A., Kennedy, J. N. P., & Gregory, D. T. (2002). An
analysis of safety culture attitudes in a highly regulated environment. Work & Stress, 16(1),
18-36.
Haukelid, K. (2008). Theories of (safety) culture revisited - An anthropological approach. Safety
Science, 46(3), 413-426.
Hvold, J. I. (2005). Safety-culture in a Norwegian shipping company. Journal of Safety Research,
36(5), 441-458.
Hay, D. (2010). Safety Culture Assessment tool. An overview. New Zealand: Workplace Press.
Hayes, B. E., Perander, J., Smecko, T., & Trask, J. (1998). Measuring perceptions of workplace
safety: Development and validation of the Work Safety Scale. Journal of Safety Research,
29(3), 145-161.
Heese, M. (2012). Got the results, now what do you do? Safety culture transformation from theory
into practice. Aviation Psychology and Applied Human Factors, 2(1), 25-33.
Helmreich, R. L., & Merritt, A. C. (1998). Error management: A cultural universal in aviation and
medicine. In R. L. Helmreich & A. C. Merritt (Eds.), Error management in aviation and
medicine. Brookfield, Vermont: Ashgate.

ISCRR Research Report#: 060-1215-R04

Page 32 of 36

Heritage, K. M. (2012). Lagged relationships between a multilevel model of safety climate and
employee safety outcomes. (Doctoral dissertation), Curtin University, Perth.
Hinkin, T. R. (1995). A review of scale development practices in the study of organizations. Journal
of Management, 21(5), 967-988.
Hopkins, A. (2006). Studying organisational cultures and their effects on safety. Safety Science,
44(10), 875-889.
Hopkins, A. (2014). Issues in safety science. Safety Science, 67(August), 6-14.
Huang, Y.-H., Zohar, D., Robertson, M. M., Garabet, A., Lee, J., & Murphy, L. A. (2013a).
Development and validation of safety climate scales for lone workers using truck drivers as
exemplar. Transportation Research Part F, 17(February), 5-19.
Hudson, P. (2007). Implementing a safety culture in a major multi-national. Safety Science, 45(6),
697-722.
International Atomic Energy Agency (IAEA). (2002). Safety culture in nuclear installations. Guidance
for use in the enhancement of safety culture. Vienna, Austria: IAEA.
James, L., & James, L. (1989). Integrating work environment perceptions: Explorations into the
measurement of meaning. Journal of Applied Psychology, 74(5), 739-751.
James, L. R., & McIntyre, M. D. (1996). Perceptions of organizational climate. In K. Murphy (Ed.),
Individual differences and behavior in organizations. San Francisco: Jossey-Bass.
Joint Planning and Development Office Safety Working Group. (2008). Safety culture improvement
resource guide. Washington: Next Generation Air Transportation System.
Kaplan, R. S., & Norton, D. P. (1991). The balanced scorecard - Measures that drive performance.
Harvard Business Review(January-February), 71-79.
Kapp, E. A. (2012). The influence of supervisor leadership practices and perceived group safety
climate on employee safety performance. Safety Science, 50(4), 11191124.
Kennedy, R., & Kirwan, B. (1998). Development of a hazard and operability-based method for
identifying safety management vulnerabilities in high risk systems. Safety Science, 30(3),
249-274.
Law, K. S., Wong, C.-S., & Mobley, W. M. (1998). Toward a taxonomy of multidimensional
constructs. Academy of Management Review, 23(4), 741-755.
Lawrie, M., Parker, D., & Hudson, P. (2006). Investigating employee perceptions of a framework of
safety culture maturity. Safety Science, 44(3), 259-276.
Lee, T. (1998). Assessment of safety culture at a nuclear reprocessing plant. Work & Stress, 12(3),
217-237.
Lee, T., & Harrison, K. (2000). Assessing safety culture in nuclear power stations. Safety Science,
34(13), 61-97.
Leiter, M. (2011). Developing a safety climate: Shared assumptions and interventions. Richmond,
British Columbia: WorkSafe BC.
Lingard, H., Cooke, T., & Blismas, N. (2012). Do perceptions of supervisors safety responses
mediate the relationship between perceptions of the organizational safety climate and
incident rates in the construction supply chain? Journal of Construction Engineering and
Management, 138(2), 234-241.
Lingard, H., Zhang, R., Harley, J., Blismas, N., & Wakefield, R. (2014). Health and safety culture.
Melbourne: Centre for Construction Work Health and Safety.
MacKenzie, S. B., Podsakoff, P. M., & Podsakoff, N. P. (2011). Construct measurement and
validation procedures in MIS and behavioral research: Integrating new and existing
techniques. Management Information Systems Quarterly, 35(2), 293-334.
Mariscal, M. A., Herrero, S. G., & Toca Otero, A. (2012). Assessing safety culture in the Spanish
nuclear industry through the use of working groups. Safety Science, 50(5), 1237-1246.
Mathieu, J. E., & Chen, G. (2011). The etiology of the multilevel paradigm in management research.
Journal of Management, 37(2), 610-641.
McCaughey, D., Halbesleben, J. R., Savage, G. T., Simons, T., & McGhan, G. E. (2013). Safety
leadership: Extending workplace safety climate best practices across health care
workforces. Adv Health Care Manag, 14, 189-217.

ISCRR Research Report#: 060-1215-R04

Page 33 of 36

McGonagle, A. K., Walsh, B. M., Kath, L. M., & Morrow, S. L. (2014). Civility norms, safety climate,
and safety outcomes: A preliminary investigation. J Occup Health Psychol, 19(4), 437-452.
Mearns, K., Flin, R., Gordon, R., & Fleming, M. (1998). Measuring safety climate on offshore
installations. Work & Stress, 12(3), 238-254.
Meli, J. L., & Ses, A. (1999). La medida del clima de seguridad y salud laboral. Anales de
Psicologa, 15(2), 269-289.
Meli, J. L., & Ses, A. (2007). Supervisors safety response: A multisample confirmatory factor
analysis. Psicothema, 19(2), 231-238.
Mohamed, S. (2003). Scorecard approach to benchmarking organizational safety culture in
construction. Journal of Construction Engineering and Management, 129(1), 80-88.
Nabitz, U., Klazinga, N., & Walburg, J. (2000). The EFQM excellence model: European and Dutch
experiences with the EFQM approach in health care. European Foundation for Quality
Management. International Journal of Quality Health Care, 12(3), 191-201.
National Occupational Research Agenda (NORA) Construction Sector Council. (2008). National
Construction Agenda: National Institute for Occupational Safety and Health
Navarro, M. F. L., Lern, F. J. G., Toms, I., & Silla, J. M. P. (2013). Validation of the group nuclear
safety climate questionnaire. Journal of Safety Research, 46(September), 21-30.
Neal, A., Griffin, M. A., & Hart, P. M. (2000). The impact of organizational climate on safety climate
and individual behavior. Safety Science, 34(13), 99-109.
Newnam, S., Griffin, M. A., & Mason, C. (2008). Safety in work vehicles: A multilevel study linking
safety values and individual predictors to work-related driving crashes. Journal of Applied
Psychology, 93(3), 632-644.
Nielsen, K. (2014). Improving safety culture through the health and safety organization: A case
study. Journal of Safety Research, 48, 7-17.
Niskanen, T. (1994). Safety climate in the road administration. Safety Science, 17(4), 237-255.
Oliver, A., Cheyne, A., Toms, J. M., & Cox, S. (2002). The effects of organizational and individual
factors on occupational accidents. Journal of Occupational and Organizational Psychology,
75(4), 473-488.
Oltedal, H. A. (2011). Safety culture and safety management within the Norwegian-controlled
shipping industry. (Doctoral dissertation), University of Stavanger.
Ostrom, L., Wilhelmsen, C., & Kaplana, B. (1993). Assessing safety culture. Nuclear Safety, 34(2),
163-172.
Patankar, M., & Sabin, E. (2010). The safety culture perspective. In E. Salas & D. Maurino (Eds.),
Human factors in aviation (pp. 95-122). San Diego, California: Elsevier Inc.
Pessemier, W. L., & England, R. E. (2012). Safety culture in the US fire service: an empirical
definition. International Journal of Emergency Services, 1(1), 10-28.
Piers, M., Montijn, C., & Balk, A. (2009). Safety culture framework for the ECAST SMS-WG: Dutch
National Aerospace Laboratory.
Pousette, A., Larsson, S., & Trner, M. (2008). Safety climate cross-validation, strength and
prediction of safety behaviour. Safety Science, 46(3), 398-404.
Reason, J. (1997). Managing the risks of organizational accidents Hampshire, England: Ashgate
Publishing Limited.
Reason, J. (2000). Human error: Models and management. British Medical Journal, 320(7237), 768770.
Reason, J. (2001). Score your safety culture. Flight Safety Australia Magazine.
Reiman, T., & Rollenhagen, C. (2014). Does the concept of safety culture help or hinder systems
thinking in safety? Accid Anal Prev, 68, 5-15.
Roy, M. (2003). Self-directed workteams and safety: a winning combination? Safety Science, 41(4),
359-376.
Schein, E. H. (1984). Coming to a new awareness of organizational culture. Sloan Management
Review, 25(2), 3-16.
Schein, E. H. (1985). Organizational culture and leadership. San Francisco: Jossey-Bass
Publishers.

ISCRR Research Report#: 060-1215-R04

Page 34 of 36

Schein, E. H. (1991). Legitimating clinical research in the study of organizational culture.


Cambridge, Massachusetts: MIT Sloan School of Management.
Schein, E. H. (1992). Organizational culture and leadership (Second Ed.). San Francisco: John
Wiley & Sons
Schein, E. H. (1999). Sense and nonsense about culture and climate. In N. Ashkanasy, C.
Wilderom & M. Peterson (Eds.), Handbook of organizational culture and climate (pp. xxiii
xxx). Thousand Oaks: Sage Publications.
Schneider, B. (1975). Organizational climates: an essay. Personnel Psychology, 28(4), 447-479.
Schneider, B., Ehrhart, M. G., & Macey, W. H. (2013). Organizational Climate and Culture. Annual
Review of Psychology, 64(1), 361-388.
The Health Foundation. (2011). Measuring safety culture. London: The Health Foundation.
Tietjen, M. A., & Myers, R. M. (1998). Motivation and job satisfaction. Management Decision, 36(4),
226-231.
Turner, B. A. (1978). Man-made disasters. London: Wykeham.
Vu, T., & De Cieri, H. (2014). Safety culture and safety climate definitions suitable for a regulator: A
systematic literature review. Melbourne, Australia: Institute for Safety, Compensation and
Recovery Research (ISCRR), Monash University, Research Report #0414-060-R2C.
Vu, T., & De Cieri, H. (2015). A review and evaluation of safety culture and safety climate
measurement tools. Melbourne, Australia: Institute for Safety, Compensation and Recovery
Research (ISCRR), Monash University, Research Report 060-1215-R03.
Weick, K. E. (1987). Organizational culture as a source of high reliabilit:y. California Management
Review, 24(2), 112-127.
Weick, K. E., & Sutcliffe, K. M. (2001). Managing the unexpected: Assuring high performance in an
age of complexity (First ed.). San Francisco: JosseyBass.
Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (1999). Organizing for high reliability: processes of
collective mindfulness. In R. S. S. a. B. M. Staw (Ed.), Research in Organizational Behavior
(pp. 81123). Stanford: Jai Press.
Westrum, R. (2004). A typology of organisational cultures. Quality and Safety in Health Care,
13(suppl 2), ii22-ii27.
Wilkin, C. L. (2013). I can't get no job satisfaction: Meta-analysis comparing permanent and
contingent workers. Journal of Organizational Behavior, 34(1), 47-64.
Williamson, A. M., Feyer, A.-M., Cairns, D., & Biancotti, D. (1997). The development of a measure
of safety climate: The role of safety perceptions and attitudes. Safety Science, 25(13), 1527.
Zhou, Q., Fang, D., & Wang, X. (2008). A method to identify strategies for the improvement of
human safety behavior by considering safety climate and personal experience. Safety
Science, 46(10), 1406-1419.
Ziegler, R., Hagen, B., & Diehl, M. (2012). Relationship between job satisfaction and job
performance: Job ambivalence as a moderator. Journal of Applied Social Psychology, 42(8),
2019-2040.
Zohar, D. (1980). Safety climate in industrial organizations: Theoretical and applied implications.
Journal of Applied Psychology, 65(1), 96-102.
Zohar, D. (2000). A group-level model of safety climate: testing the effect of group climate on
microaccidents in manufacturing jobs. Journal of Applied Psychology, 85(4), 587-596.
Zohar, D. (2008). Safety climate and beyond: A multi-level multi-climate framework. Safety Science,
46(3), 376-387.
Zohar, D. (2010). Thirty years of safety climate research: Reflections and future directions. Accident
Analysis and Prevention, 42(5), 1517-1522.
Zohar, D., & Luria, G. (2004). Climate as a social-cognitive construction of supervisory safety
practices: Scripts as proxy of behavior patterns. Journal of Applied Psychology, 89(2), 322333.
Zohar, D., & Luria, G. (2005). A multi-level model of safety climate: cross-level relationships
between organization and group-level climates. Journal of Applied Psychology, 90(4), 616628.

ISCRR Research Report#: 060-1215-R04

Page 35 of 36

ISCRR Research Report#: 060-1215-R04

Page 36 of 36