Ann. Occup. Hyg., 2016, Vol. 60, No.

5, 537–550
doi:10.1093/annhyg/mew020
Advance Access publication 19 April 2016

REVIEW

Defining and Measuring Safety Climate:

A Review of the Construction Industry
Literature
Natalie V. Schwatka1*, Steven Hecker2 and Linda M. Goldenhar3

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

1.Department of Environmental and Occupational Health, Center for Health, Work and Environment, Colorado School of Public Health,
University of Colorado, Anschutz Medical Campus, 13001 E. 17th Pl., 3rd Floor, Mail Stop B119 HSC, Aurora, CO 80045, USA;
2.Labor Education and Research Center, University of Oregon, 1289 University of Oregon, Eugene, OR 97403, USA;
3.CPWR—The Center for Construction Research and Training, 8484 Georgia Ave., Suite 1000, Silver Spring, MD 20910, USA
*Author to whom correspondence should be addressed. Tel: +1-303-724-4607; e-mail: Natalie.schwatka@ucdenver.edu
Submitted 23 October 2015; revised 15 March 2016; revised version accepted 16 March 2016.

A B ST R A CT
Safety climate measurements can be used to proactively assess an organization’s effectiveness in iden-
tifying and remediating work-related hazards, thereby reducing or preventing work-related ill health
and injury. This review article focuses on construction-specific articles that developed and/or meas-
ured safety climate, assessed safety climate’s relationship with other safety and health performance
indicators, and/or used safety climate measures to evaluate interventions targeting one or more
indicators of safety climate. Fifty-six articles met our inclusion criteria, 80% of which were published
after 2008. Our findings demonstrate that researchers commonly defined safety climate as perception
based, but the object of those perceptions varies widely. Within the wide range of indicators used to
measure safety climate, safety policies, procedures, and practices were the most common, followed
by general management commitment to safety. The most frequently used indicators should and do
reflect that the prevention of work-related ill health and injury depends on both organizational and
employee actions. Safety climate scores were commonly compared between groups (e.g. management
and workers, different trades), and often correlated with subjective measures of safety behavior rather
than measures of ill health or objective safety and health outcomes. Despite the observed limitations
of current research, safety climate has been promised as a useful feature of research and practice activi-
ties to prevent work-related ill health and injury. Safety climate survey data can reveal gaps between
management and employee perceptions, or between espoused and enacted policies, and trigger com-
munication and action to narrow those gaps. The validation of safety climate with safety and health
performance data offers the potential for using safety climate measures as a leading indicator of per-
formance. We discuss these findings in relation to the related concept of safety culture and offer sug-
gestions for future research and practice including (i) deriving a common definition of safety climate,
(ii) developing and testing construction-specific indicators of safety climate, and (iii) focusing on
construction-specific issues such as the transient workforce, subcontracting, work organization, and
induction/acculturation processes.

© The Author 2016. Published by Oxford University Press on behalf of the British Occupational Hygiene Society.

• 537
538 • Defining and measuring safety climate

K E Y W O R D S : safety behavior; safety commitment; safety culture; safety indicators; safety

perceptions; safety performance; worker involvement

I N T RO D U CT I O N studies across multiple industries, including construc-

The term safety culture, a subset of organizational tion. Glendon (2008) reviewed all safety culture and
culture, is used to broadly describe the value an climate literature across industries. Only Choudhry
organization places on the safety and health of its (2007b) reviewed safety ‘culture’ (not ‘safety climate’)
workforce through its policies, procedures, and prac- as it pertained specifically to the construction industry.
tices (Guldenmund, 2000). It became an integral part To date, no one has conducted a thorough review
of the occupational safety and health business lexicon of the literature to examine how the safety climate
in the 1980s in response to catastrophic events in the construct has been defined, measured, and used to
nuclear power, offshore oil, and commercial aviation improve safety and health outcomes in construction.
industries (Cox and Flin, 1998). More recently, the This presents a significant gap in the field because con-
healthcare sector has turned to understanding safety struction has a number of characteristics that are less

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

culture as a means to improve patient safety and pre- common in fixed industry that may have a negative
vent medical errors (Sammer et al., 2010). A second impact on safety climate. Among these are a job site in
related construct, ‘safety climate’, has been investigated continual flux, a high degree of subcontracting, a tran-
as an indicator of the overall strength of an organiza- sient workforce, and individual craft cultures. Our pri-
tion’s safety culture. Safety climate measurements can mary goal is to begin filling this gap by gaining a more
be used to proactively assess an organization’s effec- comprehensive understanding of where consensus
tiveness in identifying and remediating work-related about safety climate definition and measurement does
hazards, thereby reducing or preventing work-related and does not exist in order to move both research and
ill health and injury. practice forward.
The term safety climate first appeared in the aca-
demic literature in 1980 when Zohar (1980) meas- M ET H O D S
ured workers’ perceptions of various aspects of job Our starting point was Glendon’s (2008) review
safety in manufacturing organizations with high and of the safety climate/culture literature published
low accident rates. He defined safety climate as the between 1 January 1980 and 31 January 2008, where
‘summary of molar perceptions that employees share he identified 203 articles, 20 (7.4%) of which focused
about their work environment [in relation to safety]’ on the construction industry. Next, using the terms
(p. 96) and found that safety climate was related to ‘safety climate’, ‘safety culture’, and ‘construction’, we
safety audit scores. Since then, others have examined searched Web of Science, PsychInfo, Pubmed, and the
the usefulness and accuracy of measuring safety cli- American Society of Civil Engineering publications
mate and whether such data can be used to understand website (ASCELibrary.org) for all articles published
or predict workplace safety and health performance between 1 February 2008 and 1 March 2014. We also
across numerous industries. While the construction found several safety climate research reports in the
industry is a relative latecomer to the safety climate ‘gray literature’.
discussion, recent work by academics and practition-
ers has applied this construct as one way to understand Inclusion criteria
and improve workplace safety and health (Gillen et al., The article had to be published in English and address
2014). at least one of the following: (i) safety climate survey
Meta-analytic and review articles on safety climate development and/or testing in a construction popu-
have included numerous industries in their reviews; lation; (ii) examination of the relationship between
however, specific attention to the construction safety climate and safety and health performance or
industry has not been given. Christian et al. (2009), other related outcome variables; or (iii) application
Clarke (2006, 2010), and Nahrgang et al. (2011) all of a safety climate survey to measure the effectiveness
conducted meta-analytic reviews of safety climate of an intervention designed to improve one or more
 Defining and measuring safety climate • 539

indicators of safety climate. Finally, since the terms comparison was across five Nordic countries (Kines
safety climate and safety culture are frequently used et al., 2011). The median study population sample size
interchangeably both in research and practice and the was 281 employees (interquartile range 181–596) and
vast majority of safety climate researchers follow the included primarily white males. Five articles (9%) spe-
Zohar’s (1980) tradition by measuring safety climate cifically studied safety climate with Latino workers.
using worker perception surveys, articles that said
they were measuring safety culture using worker per- Defining safety climate
ception surveys were also included. The most frequent definition of safety climate we
found was that it reflected employee perceptions of
R E S U LTS safety in the workplace (n = 38, 68%). Only 3 of these
Glendon’s 1980–2008 search results (N = 203) com- 38 articles (8%) specifically stated that they reflected
bined with our search from 1 January 2008 to 1 March worker and manager perceptions, and included man-
2014 (N = 753) resulted in a total of 956 safety culture agers as well as employees as survey subjects. Eight of
and/or safety climate-related articles across all indus- the 38 (21%) specified that perceptions were ‘shared’

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

tries published since 1980. Comparison of the pre- among employees. Twenty-two (58%) stated that
and post-2008 searches revealed that there has been a safety climate generally reflects workers’ perceptions
surge in safety culture and/or safety climate research of how safety is valued by the organization, while 13
in recent years. (34%) defined it more specifically as workers’ percep-
Fifty-six articles met our inclusion criteria, 80% tions of workplace safety policies, procedures, and
of which were published since 2008 (Fig. 1 and practices.
Supplementary Table S2, available at Annals of Five of the 56 (9%) studies claimed that safety
Occupational Hygiene online). About 40% (n = 22) climate reflected employee attitudes rather than per-
were conducted in the USA with the remainder car- ceptions. This distinction was noted as important by
ried out in Australia (n = 9, 16%), Hong Kong (n = 7, both Kines et al. (2011) and particularly Pousette et al.
13%), China (n = 4, 7%), Singapore (n = 3, 5%), (2008), whose findings showed that attitudes and
Sweden (n = 2, 4%), and a number of other countries perceptions predict safety outcomes differently, with
that were infrequently represented (n = 9, 16%). Two attitude questions being more susceptible to social
conducted cross-country comparisons: one com- desirability bias.
pared construction safety climate in the UK, Spain, Rather than providing a definition of safety climate,
and Hong Kong (Meliá et al., 2008), while the other some authors took a different approach. For example,

Figure 1 Articles describing safety climate in construction published between 1991 and 2014.
540 • Defining and measuring safety climate

instead of an overarching definition, Cigularov et al. et al. (2011) first developed and tested the Nordic
(2010) described the specific indicators of safety cli- Safety Climate Questionnaire in the construction
mate (e.g. error management climate and safety com- industries of several Nordic countries, and then tested
munication). Ten of the 56 studies (18%) defined it in other industries.
safety climate as an indicator of safety culture. Two
studies (3.5%) defined it as an indicator of organiza- Surveys developed in other industries but adapted
tional climate (Meliá et al., 2008; Kapp, 2012), and five for and used in construction
studies (8.9%) provided no safety climate definition The majority of the studies (n = 37, 66%) used or made
(Shoji and Egawa, 2006; Burt et al., 2008; Kines et al., adaptations to previously developed non-construction
2010; Abbe et al., 2011; Lopez del Puerto et al., 2013). specific surveys. The most common source (n = 10)
Finally, in 5 of the 56 studies reviewed (8.5%) the was the UK’s Health and Safety Executive (HSE)
authors claimed to be studying safety culture, but safety climate questionnaire (Davies et al., 2001) or
defined (and measured) it in a manner more reflective the Climate Survey Tool (CST) (HSE, 1997). The
of safety climate (Molenaar et al., 2002; Fung et al., CST was subsequently renamed the Safety Climate

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

2005; Molenaar et al., 2009; Gilkey et al., 2013; Feng Tool (SCT) and modified (Sugden et al., 2009). The
et al., 2014). SCT was adapted for use on the London Olympic
construction site (Healey and Sugden, 2012).
Measuring safety climate Besides the CST, five researchers have adapted
We found that researchers typically had three over- safety climate surveys from Zohar (2000); four from
arching goals for measuring safety climate: (i) develop Neal et al. (2000); three each from Geller (1990) and
a new safety climate survey instrument or adapt an the National Institute for Occupational Safety and
existing one to reflect unique characteristics of con- Health (NIOSH) (Dejoy et al., 1995); and two from
struction; (ii) compare safety climate scores across Burt et al. (1998). Some researchers utilized other
groups of workers; or (iii) examine the relationship surveys, but due to the low frequency with which they
between safety climate survey findings and safety and were used, they were not included in our review.
health outcome measures (e.g. self-reported safety
behaviors). Each of these is discussed below. Indicators
Some of the studies that created or adapted safety cli-
Develop or adapt a safety climate survey instrument mate surveys focused on identifying the key ‘factors’
Survey sources of the latent safety climate construct. Construction
Fifteen (27%) of the studies developed their own safety practitioners seem to prefer the term indicator to fac-
climate survey instrument. However, the majority (n = 41, tor (which comes from the statistical procedure factor
73%) adapted and used an instrument developed for analysis). Therefore, in an effort to move the field of
construction or a different industry. The most commonly safety climate survey research toward more practical
adapted safety climate surveys are discussed below. applications we will use the term ‘indicator(s)’.
The average number of safety climate indicators
Surveys developed for construction across all the surveys was 4.01 (range = 1–10), and
Only 7% of the 56 articles used safety climate surveys each indicator was measured by an average of 21.92
that had been developed specifically for, or validated questions (range = 1–78). To help make sense of the
previously in, the US construction industry. In an variation, we developed a categorization scheme for
effort to replicate Zohar’s (1980) safety climate factor grouping ‘like’ indicators (see Supplementary Table
model, Dedobbeleer and Beland (1991) developed S1, available at Annals of Occupational Hygiene online).
and tested a survey in the US construction industry; We found that two categories, (i) safety policies/
this same instrument was used in three subsequent US resources/training and (ii) general management com-
studies (Gillen et al., 2002; Arcury et al., 2012; Sparer mitment to safety, were used by over half the articles
et al., 2013). Mohamed (2002) developed and tested a (Table 1). This was followed by supervisor commit-
survey in the Australian construction industry, which ment to safety and general organizational commit-
Teo and Feng (2011) later used in Singapore. Kines ment to safety (37.5 and 35.7%, respectively). Three
 Defining and measuring safety climate • 541

Table 1. Frequency of safety climate indicators most commonly measured in 56 construction-specific

safety climate articles
Safety climate indicator Number %

General management commitment to safety 30 53.5

Safety policies, resources, and training 28 50.0
Supervisor commitment to safety 21 37.5
General organizational commitment to safety 20 35.7
Co-workers commitment to safety 18 32.1
Safety communication 16 28.5
Worker involvement in safety 13 23.2

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

Risk appraisal and risk taking 9 14.3

other indicators were included in approximately >25% (7%) compared the safety climate scores across work-
of the articles. Additional indicators were reported but ers and management on single sites with multiple
at low frequency so they are not shown in Table 1. contractors and found that management rated safety
Seventeen (30%) of the surveys in the 56 articles were climate significantly higher (Molenaar et al., 2002;
tested for reliability and validity using statistical tests Fung et al., 2005; Gittleman et al., 2010; Gilkey et al.,
such as principal components analysis (PCA) and 2012). A lower perception of climate among workers
confirmatory factor analysis (CFA). compared to managers could suggest that company
safety programs are not operating as intended, thus
Comparing safety climate scores across groups providing lesser levels of protection.
Twenty-eight studies (50%) compared safety climate Other group comparisons were conducted ‘across
scores across worker subgroups to determine if there ethnicities’ (Latino versus non-Latino), ‘construc-
were significant differences in perception of worksite tion trades’, and ‘union status’. Cigularov et al.
safety climate. Seven studies (12.5%) found significant (2013b) found no difference in safety climate per-
differences in perception of worksite safety climate at ceptions between Latino and non-Latino workers or
the ‘work-group level’ (Glendon and Litherland, 2001; among different construction trades (Cigularov et al.,
Lingard et al., 2009; Kines et al., 2010; Lingard et al., 2013a). However, others found that Latino workers
2010; Biggs and Banks, 2012; Lingard et al., 2012; were likely to perceive a significantly poorer safety
Tholen et al., 2013). The study samples varied between climate than non-Latino workers (Sokas et al., 2009;
work groups within a single company and from mul- Gilkey et al., 2013), and some construction trades
tiple companies. At the ‘company level’, three studies perceived significantly poorer safety climate than
(5%) measured safety climate on multiple job sites of others (Abbe et al., 2011; Arcury et al., 2012). Gillen
one general contractor. Two studies found significant et al. (2002) found that union workers reported a
differences at different job sites (Gittleman et al., 2010; significantly more positive safety climate than non-
Fang and Wu, 2013) and one study did not (Chen et al., union workers.
2013). At the ‘job-site level’ (one job site with multiple
companies), five studies (9%) found that safety climate Examining the relationship between safety climate and safety
scores were more similar among workers employed by performance or other variables
the same company than workers employed by other Thirty-seven of the 56 studies (66%) investigated the
companies (Probst et al., 2008; Lingard et al., 2010; relationship(s) between safety climate survey scores
Healey and Sugden 2012; Liao et al., 2013; 2014). At and one or more antecedent, mediating, or outcome
both the ‘worker and management level’, four studies variables.
542 • Defining and measuring safety climate

A number of studies investigated the degree to mean safety climate scores of the project’s construc-
which organizational or personal worker or supervisor tion companies were higher than industry averages in
characteristics act as antecedents to safety climate per- the UK and were significantly negatively correlated
ceptions. Overall, the data suggest that organizations with injury accident rates, but positively associated
that value human relations (i.e. cohesion and morale), with reportable accident rates. Near-miss reports were
or value both human relations and goal attainment not significantly associated with safety climate scores.
(i.e. efficiency and productivity), have a more positive While it is unclear what the exact definitions of these
safety climate compared to organizations that primar- types of injuries are in the author’s study, it seems as
ily value internal processes (i.e. stability, control, for- though ‘reportable accident rates’ reflect injuries that
malization) or a combination of internal processes and rise to the level of federal reporting requirements
goal attainment (Colley et al., 2013). whereas ‘injury accident rates’ reflect injuries that do
The most commonly studied outcome variable not rise to the level of federal reporting requirements.
(n = 14, 25%), self-reported worker safety behavior, These disparate findings may indicate that a positive
was consistently found to be significantly positively safety climate is associated with fewer minor injuries,

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

associated with the safety climate score. Of the five but more severe injuries that rise to the level of federal
articles (9%) that studied the relationship between reporting requirements. This may indicate better injury
safety climate perceptions and self-reported injuries, reporting practices on job sites with a positive safety
four found a significant negative relationship and climate. However, the authors admit that these find-
one reported no relationship. Of the 10 articles that ings are not conclusive as there may have been some
studied safety climate’s relationship to injuries derived measurement error (e.g. reportable accidents include
from contractor records (n = 10, 18%), 9 articles ‘major injuries’ and ‘three-day reportable accidents’)
found a significant negative relationship and 1 article and they had a low frequency of these outcomes.
reported no relationship. All of the studies just described were cross-
Only three studies (5%) demonstrated a null rela- sectional. Only three (5%) were longitudinal and
tionship between safety climate and safety outcomes. they found that safety climate was predictive of self-
Sparer et al. (2013) found no significant relationship reported safety behaviors (Pousette et al., 2008;
between company safety climate scores and a propri- Tholen et al., 2013) and negatively related to injury
etary composite safety outcome variable called the incident rates (Han et al., 2014) over time at the indi-
Construct Secure Safety Assessment Program score vidual level.
(CSAP). CSAP scores are based on a combination of
contractor’s experience modification rating, lost-time Using safety climate survey data to measure intervention
and Occupational Safety and Health Administration effectiveness
(OSHA) recordable rates, number of OSHA citations, Kines et al. (2010) hypothesized that workers’ safety
and the company’s safety management system assessed climate perceptions would improve after their super-
via document analysis. Glendon and Litherland visors participated in a safety communication inter-
(2001) found that safety climate was not correlated vention. They found that only one of the five safety
with project or company safety behavioral observation climate indicators—‘attention to safety’—improved
scores, contradicting earlier findings in non-construc- significantly while the composite safety climate score
tion industry settings (Zohar, 1980); however, a small did not. Sokas et al. (2009) evaluated the degree
sample size (n = 92) may have contributed to the null to which an OSHA 10-h training course improved
findings. Martin and Lewis (2013) compared safety safety climate perceptions, but no improvement was
climate scores between accident and non-accident detected at follow-up.
groups, and did not find any significant differences. In the London Olympic Park study, the Olympic
Healey and Sugden (2012) examined the corre- Delivery Authority, acting as construction man-
lation between safety climate scores and injury acci- ager, established a Health, Safety and Environment
dent rates, reportable accident rates, and near-miss Standard to guide the procurement of designs and
reports across 15 companies at the London Olympic construction, appoint contractors, and administer a
Park construction site. They found that, in general, the safety culture/climate survey process (Bust, 2011).
 Defining and measuring safety climate • 543

They compared safety climate scores and accident Indicators

injury rates over time for two contractors and found a As with previous safety climate reviews across other
positive trend for safety climate scores and a negative industry sectors, we found that a plethora of general
trend for accident injury rates. versus industry-specific surveys comprising a variety
of indicators have been used to measure safety climate
DISCUSSION in construction. Certain indicators have been more
The purpose of this review was to better understand commonly viewed as comprising the safety climate
how investigators have defined and measured the construct (e.g. management commitment). Flin et al.
safety climate construct and how they have used safety (2000) tentatively proposed the ‘Big Five’ indica-
climate survey scores to predict or explain safety and tors of safety climate. However, there remains great
health outcomes specifically in the construction indus- variation in both assigned labels and the survey items
try. This type of research has grown in recent years designed to measure the indicators. The most popu-
with almost two-thirds of the reviewed studies having lated category in our categorization scheme, ‘safety
been published since 2010. Our findings mirror those policies, resources, and practices’, included numerous

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

from the broader safety climate literature, which sug- indicators related to general perceptions of safety and
gest that while safety climate is a promising indicator health practices as well as perceptions of particular
of work-related safety and health, more precision is components of a safety system (e.g. training).
needed in how it is defined and measured. The follow- We believe that the use of multiple reliable and
ing sections outline gaps identified in the review and validated indicators provides more detailed guidance
present recommendations for improvement. These on which areas need improvement within an organi-
recommendations are not meant to be exhaustive; zation. Some organizations use safety climate indi-
rather, they are intended to provide a starting point for cator surveys as learning opportunity. For example,
moving construction related safety climate research Gittleman et al. (2010) used safety climate indica-
and practice forward. tors related to management commitment as part of a
safety needs assessment after multiple workers died on
Definitions and measurement a construction site in a short period of time. This, in
fact, may be a more effective use of safety climate than
Definitions as a benchmarking tool. Indeed, the process provided
The majority of researchers defined safety climate as the workers with an opportunity to have their voices
employee perceptions of workplace safety. However, heard, and the company’s management team with
only a few specified what the perceptions referred to feedback on how to improve worker safety and health.
or distinguished between safety climate and safety cul- Alternatively, safety climate scores could be used
ture. Guldenmund (2007) found similar ambiguity in as benchmarks, whether within or across organiza-
the broader safety climate literature, noting that many tions; however, this runs some risk as they are largely
researchers define safety climate in an implicit manner based on subjective perceptions. Scores based upon
and that none defined safety climate with a particular five- or seven-point Likert scales may not be stable or
population in mind. comparable over time or across organizations. Other
At a 2013 workshop on construction safety culture approaches to measurement, such as a rubric-based
and climate, 70 US construction industry stakeholders method discussed below, may be more appropriate for
examined the state of research and practice and called benchmarking purposes.
for improved definitions of safety culture, safety cli- Safety climate researchers have suggested that there
mate, and project safety climate (Gillen et al., 2014). may be a core set of indicators that can be used across
Parallel efforts are underway elsewhere (Lingard et al., industries (e.g. management commitment, employee
2014), lending support to the notion that research- involvement) as well as indicators that may not be as
ers and practitioners should make a concerted effort generalizable (Cox and Flin, 1998; Zohar, 2011). For
to use agreed-upon definitions when researching and example, Huang et al. (2013) developed safety climate
working with the safety climate construct in the con- indicators for lone workers using a truck driver sam-
struction industry. ple. Their ‘delivery limits’ and ‘cell phone disapproval’
544 • Defining and measuring safety climate

indicators are very specific to trucking. Gillen et al.’s to evaluate both new or adapted survey instruments
(2014) workshop report presents indicator labels using EFA or CFA. Such analyses may reveal opportu-
selected by practitioners and researchers to best repre- nities for safety climate survey improvement in meas-
sent the construction industry, and the criteria within urement reliability and validity.
each indicator were specific to the construction indus-
try. For example, leadership involvement includes cri- Language and culture
teria such as ‘the foreman sets the safety tone on the It is clear from this review that safety climate research
job site’. Future research needs to consider features is occurring in numerous countries around the world;
that are particular to the construction industry that however, few of the articles addressed linguistic and
affect safety climate and its accurate assessment. cultural issues relevant to safety climate measurement.
The use of written safety climate surveys among low-­
Employee versus management perceptions literacy or non-native speaking workers may not pro-
The level at which safety climate is defined and meas- vide reliable information. Furthermore, the use of safety
ured, bears further attention. Many of the reviewed climate surveys amongst a diverse group of workers

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

studies defined perceptions as worker/employee from multiple cultural origins may lead to misleading
based, but few specifically stated that they could reflect results. For example, in the USA about one-quarter of
both worker and manager perceptions. Additionally, the construction industry workforce was born in a for-
very few studies compared safety climate scores eign country with a large majority (82%) coming from
between management and workers. Unlike workers, Latin American countries (Center for Construction
management’s responses to safety climate questions Research and Training, 2013). Many Latino workers
may reflect a more idealized safety climate. Comparing not only use Spanish as their first language and have
safety climate indicators across occupational levels low literacy rates (Brunette, 2004) but they also bring
could reveal the quality of safety and health commu- cultural values to work, such as ‘machismo’, ‘respeto’
nication within the organization and a divergence and ‘familia’, values that may impact workers’ interpre-
between espoused and enacted policies and proce- tation of USA-based safety climate questions (Menzel
dures. Indeed, Gittleman et al.’s (2010) research in and Gutierrez, 2010). More cross-cultural research is
construction and Huang et al. (2013) research in the needed on the adoption or adaptation of ethnocentric
trucking industry demonstrates the usefulness of safety climate indicators to other cultures as well as
conducting a gaps analysis between management and safety climate survey translation in multiple languages
worker level safety climate perceptions. and for low-literacy workers.

Reliability and validity Alternative methods

Over a decade ago, both Flin et al. (2000) and A potential complement to quantitative safety cli-
Guldenmund (2000) noted that the majority of safety mate perception surveys is a rubric-based approach
climate surveys had not been assessed for their relia- that uses narrative descriptors covering a spectrum
bility and validity. Our review reveals little progress, at from poor to exemplary for the targeted indicators
least with respect to the construction industry, with a (CWPR, 2014; Gillen et al., 2014). Rubric descrip-
minority of studies having conducted the appropriate tors themselves need to be validated through quali-
statistical analyses. Of the 17 studies (30%) that con- tative methods, followed by a comparison of climate
ducted some form of factor analysis to validate their assessment through perception surveys and the rubric
survey, 7 (41%) used PCA, a technique better suited to method. Researchers at CPWR: The Center for
reducing the number of survey questions rather than Construction Research and Training and Washington
determining the relationships among them (Fabrigar State University are currently validating a new con-
et al., 1999; see Brown, 2011: p. 22). Exploratory fac- struction-specific tool called the Safety Climate
tor analysis (EFA) and CFA are more appropriate for Assessment Tool (S-CAT). It uses rubrics to measure
assessing the way in which the survey items ‘hang eight leading indicators of safety climate contained
together’, thus being a more accurate reflection of the in the published workbook entitled ‘Strengthening
instrument’s reliability. We encourage investigators Jobsite Safety Climate; Eight Worksheets to Help You
 Defining and measuring safety climate • 545

Use and Improve Leading Indicators’ (L. Goldenhar be taken into consideration when discussing safety
and T. Probst, personal communication). Lingard climate. In the manufacturing industry, Beus et al.
et al. (2014) have proposed such a tool to use within (2010) found that job tenure was significantly posi-
the Australian construction industry. tively associated with safety climate strength because
employees had the opportunity to develop shared
Association with other safety- and health-related perceptions. The high turnover that characterizes con-
variables struction employment might be expected to hamper
The dominant focus in the literature on validating development of shared perceptions; however, at least
safety climate measures alongside subjective out- one study compared and found consistency in safety
comes such as worker self-reported safety behaviors climate scores across multiple sites of a single contrac-
is problematic. Self-reported behaviors or injuries and tor, suggesting that an individual company can influ-
illnesses data are easy to collect, but the dominant ence safety climate beyond one site at one particular
focus on this outcome may put undue emphasis on moment in time (Chen et al., 2013). More research
frontline workers’ responsibility for creating the job is needed to determine if and how a strong safety cli-

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

site’s safety climate while ignoring the influence of mate can develop within transient work environments
higher-level management. Indeed, preventing work- where workers travel from job site to job site.
related ill health and injury depends on both organiza-
tional and employee level actions. Researchers should Subcontracting
expand the scope of their safety climate validation On most construction worksites multiple contracting
efforts to include other safety and health performance businesses work together to finish a particular project.
measures (e.g. supervisor safety leadership or organi- The general and subcontracting entities come to the
zational resources devoted to safety and health). project with their own safety values, policies, and pro-
Some studies correlated safety climate scores with cedures. It is plausible that the project-specific safety
objective safety outcomes such as injury/illness rates, climate is, or may be, established and influenced pri-
workers’ compensation claims, or recordable/report- marily by the general contractor (Lingard et al., 2010).
able accidents. Reasons for this likely include (i) that it Research designed to gather and carefully aggregate
is easier to collect ‘self-reported’ injury and illness data, multilevel data is needed to address how multiple
(ii) that it is difficult for researchers to access adminis- contractors on the same job site may influence each
trative injury and illness data, and (iii) uncertainty as other’s safety climate.
to the appropriate time sequence between collecting
climate and follow-up injury and illness data. In addi- Work organization
tion, since injury and illness incidents on any one con- On the job site, the foreman-led work crew is the basic
struction site are rare, it is difficult to achieve enough unit to which the worker belongs. Indeed, most con-
statistical power to detect causality between safety struction work occurs away from the contractor’s main
climate and adverse outcomes. The London Olympic office or shop. Antonsen’s (2009b) study of safety cul-
project is one example where such outcome data were ture among seamen on oil platform supply ships pro-
readily available (Healey and Sugden, 2012). While vides insights into the culture of a relatively isolated
they found significant relationships between safety occupational group and ways in which safety improve-
climate and objective injury outcome data, they also ments must be approached in such groups. He asks
admitted difficulty in finding definitive trends due to a for greater appreciation by researchers of the ‘differ-
low accident rate among the sample of 15 companies. entiation’ perspective on organizational culture which
argues that cultural understandings (and perhaps by
extension climate perceptions) are sometimes shared,
Construction industry-specific issues
but only within subcultural boundaries (Antonsen,
Transience of the industry 2009a). Defining what constitutes the ‘cultural unit’
Because construction job sites are always in flux as thus becomes critical. As in the subcontracting discus-
projects begin, progress, and end, and workers move sion above, administration of questions and analysis
between contractors, the degree of transience should of data at the work-group, project, and organizational
546 • Defining and measuring safety climate

levels are needed to better elucidate the critical points industry may influence perceptions by which safety
for influencing and improving safety culture and cli- climate is often measured.
mate (Lingard et al., 2014).
Relationship to construction safety ‘culture’ research
Induction/acculturation process While this review focuses on safety climate, the related
In many settings workers are trained and acculturated construct of safety culture must also be discussed, par-
via the apprenticeship model. In union environments, ticularly if investigators propose that measuring safety
the ties between craft workers may be especially climate via surveys is actually measuring an organiza-
strong, and union workers may rely on their fellow tion’s underlying safety culture. Various models of con-
members for safety and health protection more than a struction safety culture have been offered, (Choudhry
distant general contractor (Applebaum, 1981). None et al., 2007a,b; Molenaar et al., 2009; Zou, 2011; Fang
of the research we reviewed discussed the role of and Wu, 2013), and they share the view that multiple
unions in the development of their safety climate indi- elements interact to affect safety and health performance
cators. Additional ethnographic research modeled on and safety culture. Thus, measuring safety climate alone

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

the work of Antonsen (2009b), Gherardi et al. (1998), using perception surveys risks missing important ele-
and others would provide a stronger basis for under- ments of the Occupational Health and Safety manage-
standing how induction and acculturation into the ment system, environmental and behavioral factors,

Table 2. Recommendations for advancing safety climate research and practice in construction

1 Create and use a shared, construction-specific definition of safety climatea

• Based on worker and management perceptions
• Contractor-specific and project-specific definitions
2 Develop construction-specific measures of safety climatea
• Consider relevance to construction work
• Use agreed-upon indicators
• Use appropriate question referents (e.g. supervisor, job site, general contractor, etc.)
• Use rigorous survey research methods
• Consider various evaluation tools—Likert scales or rubrics
3 Test construction-specific indicators of safety climate for reliability and validitya
• Use rigorous survey analytic methods
• Test relationship with leading and lagging indicators of safety and health
Objective outcomes
• Injury and illness records (e.g. OSHA logs or workers’ compensation claims)
• Safety and health performance records (e.g. site safety audit scores)
Subjective outcomes
• Company or site focused (e.g. company and site safety leadership)
• Employee focused (e.g. near misses)
• Use prospective study designs
a
Construction-specific issues to consider for each recommendation: (i) transience of the industry, (ii) subcontracting, (iii) work organization, and (iv)
induction/acculturation process.
 Defining and measuring safety climate • 547

and an organization’s underlying safety and health- et al. (2009), Edelson et al. (2009), Fang et al. (2006),
related values and beliefs. Some have argued that only Hoffmeister et al. (2013), Hon et al. (2013), Hon et al. (2014),
triangulated methods (e.g. safety climate surveys, key Jorgensen et al. (2007), Niskanen (1994), Shen et al. (2014),
Siu et al. (2004), Sunindijo and Zou (2013), Zhou et al.
informant interviews, job-site observations, etc.) can
(2008), Zhou et al. (2011).
fully investigate these deeper cultural features (Glendon *Abbe OO, Harvey CM, Ikuma LH et al. (2011) Modeling the
and Stanton, 2000). While safety climate surveys are relationship between occupational stressors, psychosocial/
the least expensive and least labor-intensive means of physical symptoms and injuries in the construction indus-
collecting perception data, alone they may not provide try. Int J Ind Ergon; 41: 106–17.
a comprehensive understanding of which indicators Antonsen S. (2009a) Safety culture and the issue of power. Saf
need to be targeted and which interventions are most Sci; 47: 183–91.
Antonsen S. (2009b) Safety culture: theory, method and improve-
likely to ultimately improve safety outcomes.
ment. Burlington, VT: Ashgate.
Applebaum HA. (1981) Royal blue: the culture of construction
Construction safety climate research and practice workers. Spindler G, Spindler L, editors. New York, NY:
recommendations Holt, Rinehart and Winston.

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

Based on the aforementioned discussion, we offer a *Arcury TA, Mills T, Marín AJ et al. (2012) Work safety climate
number of recommendations for advancing construc- and safety practices among immigrant Latino residential
tion safety climate research and practice in Table 2. construction workers. Am J Ind Med; 55: 736–45.
Beus JM, Bergman ME, Payne SC. (2010) The influence of
CO N C LU S I O N S organizational tenure on safety climate strength: a first look.
The safety climate construct has captured the attention Accid Anal Prev; 42: 1431–7.
*Biggs SE, Banks TD. (2012) A comparison of safety climate and
of many in the construction industry. While we found
safety outcomes between construction and resource func-
some inconsistencies in definitions and indicators, the tions in a large case study organization. Occupational Safety
results generally show us that safety climate may be in Transport Conference. Gold Coast, Australia, pp. 1–6.
a useful leading indicator of construction safety and Sugden C, Marshall M, Binch S et al. (2009) The development
health performance. We urge researchers and practi- of HSL’s safety climate tool—a revision of the health and
tioners to continue studying and refining the safety cli- safety climate survey tool. In Bust P, editor. International
mate construct using the recommendations provided, conference on Contemporary Ergonomics. Boca Raton,
FL: Taylor & Francis, pp. 245–52.
so we can continue to advance our understanding
Brown TA. (2011) Confirmatory factor analysis for applied
of the best methods to develop and sustain a strong research. New York, NY: The Guildford Press.
safety climate and thereby improve safety and health Brunette MJ. (2004) Construction safety research in the
outcomes for construction workers. United States: targeting the Hispanic workforce. Inj Prev;
10: 244–8.
S U P P L E M E N TA RY DATA Burt CDB, Gladstone KL, Grieve KR. (1998) Development of
Supplementary data can be found at http://annhyg. the Considerate and Responsible Employee (CARE) scale.
Work Stress; 12: 362–9.
oxfordjournals.org/.
*Burt CDB, Sepie B, McFadden G. (2008) The development of
a considerate and responsible safety attitude in work teams.
A C K N O W L E D G E M E N TS Saf Sci; 46: 79–91.
This project was funded by cooperative agreement Bust P. (2011) Learning legacy: lessons learned from the London
number U60-OH009762 to the CPWR—The Center 2012 Games construction project. Available at learninglegacy.
for Construction Research and Training from the independent.gov.uk. Accessed 1 February 2016.
National Institute for Occupational Safety and Health Center for Construction Research and Training. (2013) The
(NIOSH). Its contents are solely the responsibility of construction chart book: the U.S. construction industry and its
workers. 5th edn. Silver Spring, MD: CPWR.
the authors and do not necessarily represent the offi-
*Chen Q, Jin R. (2013) Multilevel safety culture and climate
cial view of NIOSH. survey for assessing new safety program. J Constr Eng
Manag; 139: 805–17.
REFERENCES *Chen Q, Jin R, Soboyejo A. (2013) Understanding a contrac-
*References with an asterisk before the last name of the first author tor’s regional variations in safety performance. J Constr Eng
were included in the review. Chen and Jin (2013), Choudhry Manag; 139: 641–53.
548 • Defining and measuring safety climate

*Choudhry RM, Fang D, Lingard H. (2009) Measuring safety *Fang D, Wu H. (2013) Development of a Safety Culture
climate of a construction company. J Constr Eng Manag; Interaction (SCI) model for construction projects. Saf Sci;
135: 890–9. 57(C): 138–49.
Choudhry RM, Fang D, Mohamed S. (2007a) The nature of safety *Fang DP, Chen Y, Wong L. (2006) Safety climate in construc-
culture: a survey of the state-of-the-art. Saf Sci; 45: 993–1012. tion industry: a case study in Hong Kong. J Constr Eng
Choudhry RM, Fang D, Mohamed S. (2007b) Developing Manag; 132: 573–84.
a model of construction safety culture. J Manage Eng; 23: *Feng Y, Teo EAL, Ling FYY et al. (2014) Exploring the inter-
207–12. active effects of safety investments, safety culture and pro-
Christian MS, Bradley JC, Wallace JC et al. (2009) Workplace ject hazard on safety performance: an empirical analysis. Int
safety: a meta-analysis of the roles of person and situation J Proj Manag; 32: 932–43.
factors. J Appl Psychol; 94: 1103–27. Flin R, Mearns K, O’Connor P et al. (2000) Measuring safety
*Cigularov KP, Adams S, Gittleman JL et al. (2013a) climate: identifying the common features. Saf Sci; 34:
Measurement equivalence and mean comparisons of a 177–92.
safety climate measure across construction trades. Accid *Fung IWH, Tam CM, Tung KCF et al. (2005) Safety cultural
Anal Prev; 51: 68–77. divergences among management, supervisory and worker
Cigularov KP, Chen PY, Rosecrance J. (2010) The effects of groups in Hong Kong construction industry. Int J Proj

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

error management climate and safety communication on Manag; 23: 504–12.
safety: a multi-level study. Accid Anal Prev; 42: 1498–506. Geller ES. (1990) Overview of the safety performance solu-
*Cigularov KP, Lancaster PG, Chen PY et al. (2013b) tions, Inc. safety culture survey. Blacksburg, VA: Safety
Measurement equivalence of a safety climate measure Performance Solutions, Inc.
among Hispanic and White Non-Hispanic construction Gherardi S, Nicolini D, Odella F. (1998) What do you mean by
workers. Saf Sci; 54(C): 58–68. safety? Conflicting perspectives on accident causation and
Clarke S. (2006) The relationship between safety climate and safety management in a construction firm. J Conting Crisis
safety performance: a meta-analytic review. J Occup Health Manag; 6: 202–13.
Psychol; 11: 315–27. *Gilkey DP, del Puerto CL, Keefe T et al. (2012) Comparative
Clarke S. (2010) An integrative model of safety climate: link- analysis of safety culture perceptions among homesafe man-
ing psychological climate and work attitudes to individual agers and workers in residential construction. J Constr Eng
safety outcomes using meta-analysis. J Occup Organ Psychol; Manag; 138: 1044–52.
83: 553–78. *Gilkey DP, del Puerto CL, Rosecrance J et al. (2013)
*Colley SK, Lincolne J, Neal A. (2013) An examination of the Comparative analysis of safety culture and risk perceptions
relationship amongst profiles of perceived organizational among Latino and non-Latino workers in the construction
values, safety climate and safety outcomes. Saf Sci; 51: industry. J Saf Health Environ Res; 9: 94–104.
69–76. *Gillen M, Baltz D, Gassel M et al. (2002) Perceived safety climate,
Cox S, Flin R. (1998) Safety culture: philosopher’s stone or job demands, and co-workers’ support among union and non-
man of straw? Work Stress; 12: 189–201. union injured construction workers. J Saf Res; 33: 33–51.
CWPR. (2014) Strengthening jobsite safety climate: eight Gillen M, Goldenhar LM, Hecker S et al. (2014) Safety culture
worksheets to help you use and improve leading indicators. and climate in construction: bridging the gap between research
cited 15 March 2015. Available at http://www.cpwr.com/ and practice. Silver Spring, MD: CPWR, pp. 1–60.
sites/default/files/CPWR%20Activity%20Sheets%20 *Gittleman JL, Gardner PC, Haile E et al. (2010) Case Study
Booklet%20081514.pdf. Accessed 1 February 2016. City Center and Cosmopolitan Construction Projects, Las
Davies F, Spencer R, Dooley K et al. (2001) Summary guide to Vegas, Nevada: lessons learned from the use of multiple
safety climate tools. Norwich, UK: HSE Books. sources and mixed methods in a safety needs assessment. J
*Dedobbeleer N, Beland F. (1991) A safety climate measure for Saf Res; 41: 263–81.
construction sites. J Saf Res; 22: 97–103. Glendon AI. (2008) Safety culture: snapshot of a developing
Dejoy DM, Murphy LR, Gershon RM. (1995) The influence concept. J Occup Health Saf; 24: 179–89.
of employee, job/task, and organizational factors on adher- *Glendon AI, Litherland DK. (2001) Safety climate factors,
ence to universal precautions among nurses. Int J Ind Ergon; group differences and safety behavior in road construction.
16: 43–55. Saf Sci; 39: 157–88.
*Edelson J, Neitzel R, Meischke H et al. (2009) Predictors of Glendon AI, Stanton NA. (2000) Perspectives on safety cul-
hearing protection use in construction workers. Ann Occup ture. Saf Sci; 34: 193–214.
Hyg; 53: 605–15. Guldenmund FW. (2000) The nature of safety culture: a review
Fabrigar LR, Wegener DT, MacCallum RC et al. (1999) of theory and practice. Saf Sci; 34: 215–57.
Evaluating the use of exploratory factor analysis in psycho- Guldenmund FW. (2007) The use of questionnaires in safety
logical research. Psychol Methods; 4: 272–99. culture research—an evaluation. Saf Sci; 45: 723–43.
 Defining and measuring safety climate • 549

*Han S, Saba F, Lee S et al. (2014) Toward an understanding of wp-content/uploads/2014/08/Health-Safety-Culture-

the impact of production pressure on safety performance in August-2014.pdf.
construction operations. Accid Anal Prev; 68: 106–16. *Lopez del Puerto C, Clevenger CM, Boremann K et al. (2013)
*Healey N, Sugden C. (2012) Safety culture on the olympic park. Exploratory study to identify perceptions of safety and risk
Cited 11 October 2014. Available at http://www.hse.gov. among residential Latino construction workers as distinct
uk/research/rrpdf/rr942.pdf. Accessed 11 October 2014. from commercial and heavy civil construction workers. J
*Hoffmeister K, Gibbons AM, Johnson SK et al. (2013) The Constr Eng Manag; 140: 04013048-1–7.
differential effects of transformational leadership facets on *Martin H, Lewis TM. (2013) Pinpointing safety leadership
employee safety. Saf Sci; 62(C): 68–78. factors for safe construction sites in Trinidad and Tobago. J
*Hon CKH, Chan APC, Yam MCH. (2014) Relationships Constr Eng Manag; 140: 04013046-01–04013046-10.
between safety climate and safety performance of build- *Meliá JL, Mearns K, Silva SA et al. (2008) Safety climate
ing repair, maintenance, minor alteration, and addition responses and the perceived risk of accidents in the con-
(RMAA) works. Saf Sci; 65(C): 10–9. struction industry. Saf Sci; 46: 949–58.
*Hon CKH, Chan APC, Yam MCH. (2013) Determining Menzel NN, Gutierrez AP. (2010) Latino worker perceptions
safety climate factors in the repair, maintenance, minor of construction risks. Am J Ind Med; 53: 179–87.
alteration, and addition sector of Hong Kong. J Constr Eng *Mohamed S. (2002) Safety climate in construction site envi-

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016

Manag; 139: 519–28. ronments. J Constr Eng Manag; 128: 375–84.
HSE. (1997) Health and safety climate survey tool—information *Molenaar K, Brown H, Caile S et al. (2002) Corporate culture:
pack (MISC097). UK: HSE Books. a study of firms with outstanding construction safety. Prof
Huang YH, Zohar D, Robertson MM et al. (2013) Development Saf; July: 18–27.
and validation of safety climate scales for mobile remote *Molenaar K, Park J, Washington S. (2009) Framework for meas-
workers using utility/electrical workers as exemplar. Accid uring corporate safety culture and its impact on construction
Anal Prev; 59: 76–86. safety performance. J Constr Eng Manag; 135: 488–96.
*Jorgensen E, Sokas RK, Nickels L et al. (2007) An English/ Nahrgang JD, Morgeson FP, Hofmann DA. (2011) Safety at
Spanish safety climate scale for construction workers. Am J work: a meta-analytic investigation of the link between job
Ind Med; 50: 438–42. demands, job resources, burnout, engagement, and safety
*Kapp EA. (2012) The influence of supervisor leadership prac- outcomes. J Appl Psychol; 96: 71–94.
tices and perceived group safety climate on employee safety Neal A, Griffin MA, Hart PM. (2000) The impact of organiza-
performance. Saf Sci; 50: 1119–24. tional climate on safety and individual behaviour. Saf Sci;
*Kines P, Andersen LP, Spangenberg S et al. (2010) Improving 34: 99–109.
construction site safety through leader-based verbal safety *Niskanen T. (1994) Safety climate in road administration. Saf
communication. J Saf Res; 41: 399–406. Sci; 17: 237–55.
*Kines P, Lappalainen J, Mikkelsen KL et al. (2011) Nordic Safety *Pousette A, Larsson S, Törner M. (2008) Safety climate cross-
Climate Questionnaire (NOSACQ-50): a new tool for diag- validation, strength and prediction of safety behaviour. Saf
nosing occupational safety climate. Int J Ind Ergon; 41: 634–46. Sci; 46: 398–404.
*Liao P-C, Lei G, Fang D et al. (2014) The relationship between Probst TM, Brubaker TL, Barsotti A. (2008) Organizational
communication and construction safety climate in China. J injury rate underreporting: the moderating effect of organi-
Civ Eng; 18: 887–97. zational safety climate. J Appl Psychol; 93: 1147–54.
*Liao P-C, Lei G, Xue J et al. (2013) Influence of person-organ- Sammer CE, Lykens K, Singh KP et al. (2010) What is patient
izational fit on construction safety climate. J Manag Eng; 31: safety culture? A review of the literature. J Nurs Scholarsh;
04014049. 42: 156–65.
*Lingard HC, Cooke T, Blismas N. (2009) Group-level safety cli- *Shen Y, Tuuli MM, Xia B et al. (2014) Toward a model for
mate in the Australian construction industry: within-group forming psychological safety climate in construction pro-
homogeneity and between-group differences in road con- ject management. Int J Proj Manag; 33: 223–35.
struction and maintenance. Constr Manag Econ; 27: 419–32. Shoji T, Egawa Y. (2006) Structure of safety climates and its
*Lingard HC, Cooke T, Blismas N. (2010) Safety climate in effects on workers’ attitudes and work safety at Japanese
conditions of construction subcontracting: a multi-level construction work sites. J UOEH; 28: 29–43.
analysis. Constr Manag Econ; 28: 813–25. *Siu O-L, Phillips DR, Leung T-W. (2004) Safety climate and
*Lingard HC, Cooke T, Blismas N. (2012) Do perceptions safety performance among construction workers in Hong
of supervisors’ safety responses mediate the relationship Kong. Accid Anal Prev; 36: 359–66.
between perceptions of the organizational safety climate *Sokas RK, Emile J, Nickels L et al. (2009) An intervention effec-
and incident rates in the construction supply Chain? J tiveness study of hazard awareness training in the construction
Constr Eng Manag; 138: 234–41. building trades. Public Health Rep; 124 (Suppl. 1): 160–8.
Lingard H, Zhang R, Harley J et al. (2014) Wakefield. Health *Sparer EH, Murphy LA, Taylor KM et al. (2013) Correlation
and safety culture. Melbourne, Victoria, Australia: RMIT between safety climate and contractor safety assessment
University. Available at http://www.constructors.com.au/ programs in construction. Am J Ind Med; 56: 1463–72.
550 • Defining and measuring safety climate

*Sunindijo RY, Zou PXW. (2013) Conceptualizing safety man- considering safety climate and personal experience. Saf Sci;
agement in construction projects. J Constr Eng Manag; 139: 46: 1406–19.
1144–53. Zohar D. (1980) Safety climate in industrial organizations:
*Teo EAL, Feng Y. (2011) The indirect effect of safety invest- theoretical and applied implications. J Appl Psychol; 65:
ment on safety performance for building projects. Archit Sci 96–102.
Rev; 54: 65–80. Zohar D. (2000) A group-level model of safety climate: testing
*Tholen SL, Pousette A, Törner M. (2013) Causal relations the effect of group climate on microaccidents in manufac-
between psychosocial conditions, safety climate and safety turing jobs. J Appl Psychol; 85: 587–96.
behaviour—a multi-level investigation. Saf Sci; 55: 62–9. Zohar D. (2011) Safety climate: conceptual and measure-
*Zhou Q, Dongping F, Mohamed S. (2011) Safety climate ment issues. In Quick JC, Tetrick LE, editors. Handbook
improvement: case study in a Chinese construction com- of occupational health psychology. Chapter 8. 2nd edn.
pany. J Constr Eng Manag; 137: 86–95. Washington, DC: APA. pp. 141–64.
*Zhou Q, Fang D, Wang X. (2008) A method to identify strat- *Zou PXW. (2011) Fostering a strong construction safety cul-
egies for the improvement of human safety behavior by ture. Leadersh Manag Eng; 11: 11–22.

Downloaded from http://annhyg.oxfordjournals.org/ by guest on June 24, 2016