Accident Analysis and Prevention 42 (2010) 2082–2093

Contents lists available at ScienceDirect

Accident Analysis and Prevention

journal homepage: www.elsevier.com/locate/aap

Safety management practices and safety behaviour: Assessing the mediating

role of safety knowledge and motivation
M.N. Vinodkumar a,∗ , M. Bhasi b
a
Safety and Fire Engineering, School of Engineering, Cochin University of Science and Technology, Kochi 682022, Kerala, India
b
School of Management Studies, Cochin University of Science and Technology, Kochi 682022, Kerala, India

a r t i c l e i n f o a b s t r a c t

Article history: Safety management practices not only improve working conditions but also positively influence employ-
Received 29 November 2009 ees’ attitudes and behaviours with regard to safety, thereby reducing accidents in workplace. This
Received in revised form 22 June 2010 study measured employees’ perceptions on six safety management practices and self-reported safety
Accepted 27 June 2010
knowledge, safety motivation, safety compliance and safety participation by conducting a survey using
questionnaire among 1566 employees belonging to eight major accident hazard process industrial units
Keywords:
in Kerala, a state in southern part of India. The reliability and unidimesionality of all the scales were
Safety management
found acceptable. Path analysis using AMOS-4 software showed that some of the safety management
Safety performance
Safety behaviour
practices have direct and indirect relations with the safety performance components, namely, safety
Mediators in safety compliance and safety participation. Safety knowledge and safety motivation were found to be the key
Path analysis mediators in explaining these relationships. Safety training was identified as the most important safety
management practice that predicts safety knowledge, safety motivation, safety compliance and safety
participation. These findings provide valuable guidance for researchers and practitioners for identifying
the mechanisms by which they can improve safety of workplace.
© 2010 Elsevier Ltd. All rights reserved.

1. Introduction This study attempts to examine the various safety management

The root cause of a great majority of industrial disasters in the
past can be traced back to the absence of an adequate safety man-
agement system. After examining the events before and after the
Bhopal disaster, Bowander (1987) observed that three types of
errors, human error, technological error and system error occurred
simultaneously to trigger off the incident. He pointed out that safety
team from the parent company headquarters had reported two
years before the incident that safety management practices in the
plant was poor. This raises many questions about the way in which
safety management is handled in developing countries like India.
Most of the authors (Bowander, 1987; Chouhan, 2005; Gupta, 2002)
who studied the Bhopal gas leak accident unanimously agreed
that programmes and policies for managing safety in workplace
in major accident hazard process industries in developing coun-
tries are clearly inadequate and require modification. Even though
numerous studies have been reported in safety management from
various parts of the world, there is not much research evidence from
India where safety management is yet to get the priority it deserves.
∗ Corresponding author. Tel.: +91 484 2331698; fax: +91 484 2577405.
E-mail addresses: mnvinodkumar@cusat.ac.in, mnvinodk@gmail.com
(M.N. Vinodkumar), mbhasi@cusat.ac.in (M. Bhasi).
practices and their influence on safety performance in eight major
accident hazard process industrial units in Kerala, a state in the
southern part of India.
Research in the area of occupational safety became prominent
during the past three decades. Its primary objective is to predict
safety related outcomes such as accidents and injuries in order to
provide valuable guidance for improving safety in organizations.
This requires extensive knowledge, not only about the various
aspects that influence safety but also as to how this influence
occurs. The fact that organizational and social factors do influ-
ence safety performance led to extensive research in the field of
safety culture and safety climate (e.g., Cox and Flin, 1998; Donald
and Canter, 1994; Glendon and Stanton, 2000; Guldenmund, 2000;
Hofmann et al., 1995; Lee and Harrison, 2000; Mearns et al., 2003;
Pidgeon, 1998; Vinodkumar and Bhasi, 2009; Zohar, 1980). Even
though a clear consensus is yet to evolve on the dimensions to be
included in safety culture and safety climate, it is widely accepted
that they are good predictors of safety related outcomes (e.g., acci-
dents and injuries) in both Western and Eastern societies (e.g.,
Cooper and Philips, 2004; Dedobbeleer and Beland, 1991; Griffin
and Neal, 2000; Siu et al., 2004; Zohar, 1980, 2000).
Characterized by the shared perceptions of employees, safety
climate can be seen as an organization’s temporal “state of safety”,
or, a snapshot of the prevailing state of safety in the organization
at a discrete point of time (Cheyne et al., 1998). Some researchers

0001-4575/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aap.2010.06.021
 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
believe that safety climate is a unidimensional latent variable (Neal
et al., 2000), while others have claimed that it is multi-dimensional
(e.g., Cooper and Philips, 2004; Dedobbeleer and Beland, 1998;
Mohammed, 2002; O’Toole, 2002; Zohar, 1980; Zohar and Luria,
2005), although they do not agree on the number of factors that
constitute it (Cavazza and Serpe, 2009). But one thing accepted by
all is that safety management practices play a vital role in forming
the safety climate in an organization. Safety climate therefore can
be considered as a result of the interactions of factors such as safety
management practices, behavioural and attitudinal factors of man-
agers and workmen, work and general discipline in the organization
and risk perception at work.
Recently, there has been growing interest in research looking at
mediators in the organizational climate, safety climate and safety
performance relationships (e.g., Neal et al., 2000; Siu et al., 2004;
Huang et al., 2006; Wu et al., 2008; Cavazza and Serpe, 2009). These
studies are based on the assumption that human beings strive to
cohere with their environments. For instance, one would expect
that if an organization has a supportive climate for safety, then
individuals would seek to be productive organizational members
by transferring the knowledge and skill that they learned from the
various safety management practices implemented by the manage-
ment to their jobs.
Neal et al. (2000) took safety climate as a single factor con-
taining management values, communication, training, and safety
systems and studied the mediating role of safety knowledge and
motivation on the relationship between safety climate and safety
behaviour. Seo (2005) operationalized perceived safety climate
as management commitment, supervisor support, co-worker sup-
port, employee participation, and competence level in the study
looking for mediators in safety climate safety performance rela-
tionship. Huang et al. (2006) included management commitment
to safety, return to work practices, post-injury administration and
safety training as the constituents of safety climate while ana-
lyzing the mediating role of safety control on the relationship
between safety climate and safety performance. Siu et al. (2004)
took safety attitudes and communication as the factors in safety
climate while assessing the mediating role of psychological strain
in the safety climate safety performance relationship. In another
study, Wu et al. (2008) opined that safety climate constitute of
CEO’s safety commitment, managers’ safety commitment, employ-
ees’ safety commitment, emergency response and perceived risk.
In the study of Cavazza and Serpe (2009), the mediating role of
attitudinal ambivalence of employees towards personal protec-
tive equipments on the relationship between safety climate and
unsafe behaviour was investigated. This study considered company
safety concern, senior managers’ safety concern, work pressure and
supervisors’ attitude towards safety as the dimensions of safety
climate. It is evident from these studies that the choice of safety cli-
mate dimensions can partially be determined by practical interest
(Huang et al., 2006).
Since most of the above studies took safety climate as a single
measured variable or as a latent variable, the individual effects of
the most important constituent of safety climate, namely, employ-
ees’ perception of the safety management practices was never
investigated. There is very few research evidence relating perceived
safety management practices to safety performance, directly or
indirectly. It is evident that there is a gap in the literature in this
area of research. Therefore, the present study intends to investi-
gate the role of safety knowledge and motivation in the relationship
between perceived safety management practices and self-reported
safety behaviour. This study is based on the assumption that the
level of safety management practices as perceived by employees
form only a part of safety climate. A safety management system
reflects the organization’s commitment to safety, and it has an
important influence on employees’ perceptions about the impor-
2083
tance of safety in the company. Since safety management practices
can be considered as indicators of safety culture of upper manage-
ment and favourable safety management practices are expected to
result in improved safety climate, the mechanism with which they
influence safety performance is worth investigating.
1.1. Safety management practices
Safety management relates to the actual practices, roles and
functions associated with remaining safe (Kirwan, 1998). It is
usually regarded as a sub-system of the total organizational
management and is carried out via the organization’s safety man-
agement system with the help of various safety management
practices. Safety management systems are mechanisms that are
integrated in the organization (Labodova, 2004) and designed to
control the hazards that can affect workers’ health and safety. Safety
management practices are the policies, strategies, procedures and
activities implemented or followed by the management of an orga-
nization targeting safety of their employees. They are the essential
elements permitting an effective management of safety in firms
and are designed to comply with the existing legislations appli-
cable to the organization. The extent to which these practices are
implemented in an organization will be manifested through vari-
ous actions and programmes of the management and will be clearly
visible to an insider like an employee. Safety management system
(and its practices) can be regarded as an antecedent of the firm’s
safety climate.
In hypothesizing the constituents in safety management, those
safety management practices should be included that could com-
monly be perceived by the employees and those should also have a
pertinent role in shaping the safety environment. There have been
numerous attempts to identify specific safety management prac-
tices that predict safety performance. Cohen (1977), Cohen et al.
(1975), DePasquale and Geller (1999), Griffiths (1985), Harper et
al. (1997), Shafai-Sahrai (1971), Shannon et al. (1996, 1997), and
Smith et al. (1975) revealed that organizations with lower accident
rates were characterized by a few of the following factors: safety
officers held high rank; management showed personal involve-
ment in safety activities; superior training for new employees;
frequent training for existing employees; display of safety posters
for identifying hazards; well defined procedures for promotion
and job placements; daily communication between workers and
supervisors about health and safety; frequent safety inspections;
higher priority for safety in meetings and decisions concerning
work practice; thorough investigation of accidents; more frequent
attendance of senior managers at health and safety meetings and
empowerment of the workforce. Vredenburgh (2002) included
worker participation, safety training, hiring practices, reward sys-
tems, management commitment and communication and feedback
as the safety management practices in the study of hospital envi-
ronment.
In one of the first investigations of safety climate, Zohar (1980)
found that management’s commitment to safety is a major fac-
tor affecting the success of an organization’s safety programmes.
The safety commitment of the management must result in an
observable activity on the part of the management and must be
demonstrated in their behaviour as well as their words (Hofmann
et al., 1995). Employees’ perception will reflect how employees
believe that safety is to be valued in the organization (Griffin and
Neal, 2000). In high risk environments like chemical industries,
management commitment has been repeatedly highlighted (Cox
and Flin, 1998; Flin et al., 1996; Cox and Cheyne, 2000). Following
Vredenburgh (2002), this study also includes management com-
mitment as one of the management practices, and is measured
using items related to management’s priority for safety, corrective
actions, safety manager attending safety meeting, accident/near
2084 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
miss investigation, and providing adequate personal protective
equipments.
A key element in every successful organization, in any success-
ful accident prevention programme and in any occupational safety
and health programme is effective safety training. It improves
behavioural skills, related knowledge and/or attitudes. Safety train-
ing also provides the means for making accidents more predictable.
To improve the level of safety and health for all employees, orga-
nization should institute a systematic, comprehensive safety and
health training programme for new employees, provide a mentor
for these employees and use a buddy system to help orient new
employees in the safety, health and quality systems (Vredenburgh,
2002). The studies of Lee (1998), Ostrom et al. (1993), Tinmannsvik
and Hovden (2003), Cohen et al. (1975), Smith et al. (1975) and
Zohar (1980) have found that those companies with lower accident
rates were characterized by good safety training for employees.
Therefore, safety training is considered as a management practice
and is measured using items related to training for newly recruited
employees, discussion of safety issues in training sessions, training
to meet emergency situations, encouragement to attend training
programmes, and hazard assessment training.
Employee involvement is a behavioural oriented technique
that involves individuals or groups in the upward communication
flow and decision-making processes within the organization. The
amount of participation can range from no participation, where the
supervisor makes all decisions, to full participation, where every-
one connected with, or affected by the decision, is involved. Since
employees close to the work are the best qualified persons to make
suggestions for improvements, they can be consulted before mak-
ing final decisions, especially for those decisions that affect the
employees (Vredenburgh, 2002). This empowerment of workers
provides them with authority, responsibility and accountability
for required decisions and ensures that both employees and man-
agements are involved in setting goals and objectives. It induces
employees to do their best work as individuals and as a team,
while relieving the manager to plan, lead and mentor (Cohen and
Cleveland, 1983). Worker involvement has been reported as a deci-
sive factor in safety management by Lee (1998), Rundmo (1994),
Dedobbeleer and Beland (1991), Shannon et al. (1996) and Cox and
Cheyne (2000). Therefore, workers’ involvement in safety is con-
sidered as a management practice and is measured using items
related to safety committee comprising of workers’ representa-
tives, involvement of workers in safety related decision making,
involvement in identifying safety problems, and consultation with
workers about safety matters.
Communications of various kinds are used to enhance the gen-
eral effectiveness of any motivational effort. The coverage and
impact of communication will be higher in two-way communica-
tion and can lead to changes in behaviour. Regular communication
about safety issues between managements, supervisors and work-
force is an effective management practice to improve safety in
workplace. Cohen (1977), Vredenburgh (2002), Cox and Cheyne
(2000) and Mearns et al. (2003) included communication and feed-
back as a factor in their surveys using questionnaire among various
category of workers and showed that safety performance is influ-
enced by the level of communication in an organization. This study
also accepts safety communication and feedback as a management
practice and is measured using items related to hazard report-
ing system, open door policy for safety issues, communication
about safety goals and targets between managers and workers, and
opportunity to discuss safety issues in meetings.
It is a usual practice in hazardous industrial units such as pro-
cess/chemical plants in Kerala to prepare safety manuals based on
the operations and statutory requirements laid down by the law
enforcing authorities. It is the management to decide whether it has
to be practiced or has to remain in paper only. Well documented
safety rules and procedures and its enforcement by supervisors
and managers can improve safety behaviour of workers. Glendon
and Litherland (2001) reported this as a reliable factor after factor
analyzing the data collected from construction workers. Cox and
Cheyne (2000) and Mearns et al. (2003) included safety rules and
procedures as a factor in their offshore safety studies and showed
that it has significant correlation with accident rates. Based on the
above findings, safety rules and procedures is taken as a safety man-
agement practice and is measured with the help of questionnaire
items related to effective rules and procedures of work to prevent
accidents occurring, adequate safety department, supervisors and
managers enforcing safety rules, and regular safety inspections.
The use of incentives, awards and recognition to motivate
employees to perform safely is an accepted feature of both orga-
nization behaviour management and total quality management
models (Hagan et al., 2001). They can add interest to the hazard
control programme of an organization and enhance self-protection
action on the part of the workforce (Cohen et al., 1979). A well-
designed reward system should be characterized by high level
of visibility in the organization, offering recognition, which can
help modify behaviour (Vredenburgh, 2002). This study also rec-
ognizes safety promotion policies as one of the safety management
practices and is assessed using items related to counting safe
conduct as a positive factor for promotion, rewards and incen-
tives for reporting hazards, creating awareness among workers by
arranging programmes during safety week celebrations, healthy
competition among workers to report unsafe conditions or acts, and
supervisors welcoming and encouraging workers to report safety
matters.
Managements of industrial plants in many developed countries
follow the policy of recruiting new personnel who are predisposed
to displaying safety conscious attitude in their work. This can be
considered as a management practice as it not only supports a
safety conscious worker, but also helps to improve the overall moti-
vational level of the workforce. Turner (1991), Eckhardt (1996) and
Vredenburgh (2002) found that the consideration of safety perfor-
mance in the selection of employees is a significant predictor of
injury rates. Discussions with top management people from sev-
eral industries revealed that this practice is not followed in Indian
industries due to reasons such as high population and high per-
centage of unemployment. Hence, the above management practice
is not considered in this study.
Another important factor that comes in safety management is
the extent to which the management is able to control drugs and
alcohol consumption of employees during working hours (O’Toole,
2002). Even though this cannot be considered as a management
practice, it could have been included as an item in safety rules
and procedures or management commitment. However, this was
ignored in this study purposefully because of the difficulty in
extracting honest response.
The perceptions of employees on the level of the selected six
safety management practices implemented in their organizations
are considered as organizational factors which can influence their
safety performance. Hence, the above six safety management prac-
tices are considered as antecedents of safety performance in this
study.
1.2. Safety performance
Even though traditional measures of safety performance rely
primarily on some form of accident or injury data, safety related
behaviours such as safety compliance and safety participation can
also be considered as components of safety performance. Safety
compliance represents the behaviour of the employees in ways that
increase their personal safety and health. Safety participation rep-
resents the behaviour of employees in ways that increase the safety
 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
Fig. 1. Hypothesized model.
and health of co-workers and that support an organization’s stated
goals and objectives (Hagan et al., 2001).
The model proposed by Neal and Griffin (1997) based on the the-
ories of job performance (Borman and Motowidlo, 1993; Campbell
et al., 1993) distinguishes between antecedents of performance,
determinants of performance and components of performance.
Neal et al. (2000) considered safety climate as antecedent of safety
performance, safety knowledge and safety motivation as deter-
minants of safety performance and safety compliance and safety
participation as components of safety performance. In another
study, Pousette et al. (2008) measured safety motivation and safety
knowledge as two individual attitudes to safety. Self-rated safety
behaviour was measured by three safety behaviour measures. They
were named as structural safety behaviour (concerning participa-
tion on organized safety activities), interactional safety behaviour
(concerning safety activities in the daily work in interaction with
co-workers and management) and personal safety behaviour (mea-
suring behaviour promoting personal protection). Considering the
above studies, the authors included the perceptions of the employ-
ees on the six identified safety management practices as the
antecedents of safety performance in the current study. The deter-
minants of safety performance were measured by safety motivation
and safety knowledge and components of safety performance were
measured by safety compliance and safety participation in this
study.
1.3. The present study
This study was designed based on the findings of Campbell et
al. (1993) that the determinants of performance (knowledge and
motivation) mediate the relationship between antecedents and
components of performance. The first purpose of this study was
to establish the unidimensionality and reliability of the six safety
management practices scales, safety knowledge, safety motivation,
safety compliance and safety participation. The second purpose of
this study was to investigate the direct and mediating effects of
safety knowledge and motivation on the relationship between the
six safety management practices and safety behaviour (safety com-
pliance and safety participation). The hypothesized model of the
study is depicted in Fig. 1.
In major accident hazard chemical units, a minor error (human
or technical) can sometimes trigger on a chemical reaction which
may go out of control and end up in major accident. Good knowl-
edge of the processes, associated dangers and methods to prevent
them are essential for workers in such plants. An effective safety
management system in place will not only be helpful to improve
the employees’ safety knowledge but also to motivate them. Moti-
vation increases the awareness, interest and willingness of the
2085
employees for better safety performance. The safety management
system implemented in an organization comprise of a set of poli-
cies and practices aimed at positively impacting on the employees’
attitudes and behaviours with regard to risk, thereby reducing their
unsafe acts. Its aim is to raise awareness, understanding, motivation
and commitment among workers (Fernandez-Muniz et al., 2007).
From the above arguments, it appears that the safety manage-
ment practices can influence the safety knowledge and motivation
of employees. Based on the above, the following hypotheses are
advanced.
H1 . The six safety management practices (management com-
mitment, safety training, workers’ involvement in safety, safety
communication and feedback, safety rules and procedures, and
safety promotion policies) predict safety knowledge.
H2 . The six safety management practices (management com-
mitment, safety training, workers’ involvement in safety, safety
communication and feedback, safety rules and procedures, and
safety promotion policies) predict safety motivation.
H3 . Safety knowledge predicts safety compliance and safety par-
ticipation.
H4 . Safety motivation predicts safety compliance and safety par-
ticipation.
H5 . Safety knowledge mediates the relationship between the six
safety management practices and safety compliance.
H6 . Safety motivation mediates the relationship between the six
safety management practices and safety compliance.
H7 . Safety knowledge mediates the relationship between the six
safety management practices and safety participation.
H8 . Safety motivation mediates the relationship between the six
safety management practices and safety participation.
2. Method
2.1. Population and sample
India is a country of diversities. People in over 20 states in India
speak different languages and are significantly different from each
other in culture, literacy level, attitude and work style. This study
was conducted in the year 2003 in the Kerala state which is ranked
first in literacy level. Kerala has twelve major chemical factories
with safety infrastructure like separate safety departments. This
study was initially planned targeting all these twelve units. But
in the final stage, four company managements (Govt. of India-1,
Govt. of Kerala-2 and Private-1) denied permission for this study.
2086 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
Table 1
Details of organizations and response rate.
Org Owned by Main products
1 Govt. of India Caprolactam
2 Govt. of India Ammonia, sulphuric acid, phosphoric acid, ammonium phosphate, and ammonium sulphate
3 Govt. of India Sulphuric acid, phosphoric acid and ammonium phosphate
4 Govt. of India Pesticides: benzene hexa chloride, endosulphan, mancozeb and DDT
5 Govt. of Kerala Caustic soda and chlorine
6 Govt. of India Rare earth chlorides
7 Private Sulphuric acid and zinc
8 Govt. of India Petroleum products-refining
Total
Details about the eight units that were studied are presented in
Table 1. The first three units belong to a large fertilizer manufac-
turing company owned by the government of India and the rest are
independent companies. After getting permission from the respec-
tive managements, the questionnaires were distributed personally
to all employees below the supervisory level, engaged in the gen-
eral shift (8 a.m. to 4 p.m.) and the morning shift (6 a.m. to 2 p.m.).
The authors explained the purpose of this study to the participants
and their participation was voluntary. Completed questionnaires
were personally collected from the participants and a total of 1566
completed response sheets were received with a response rate of
72%. The data collection was completed in about eight month’s
time.
The mean age of the sample was 43.22 years and the mean expe-
rience of the respondents was 17.27 years. Among the respondents,
20% belonged to the age group of 20–35 years and 56% belonged
to the age group of 36–50 years. 21% of the respondents had less
than 10 years of experience and 45% had 10–20 years of experience
in the respective companies. 42% of them were found to have high
school level education and only 5% had a degree in some discipline.
The remaining 53% had intermediate level of education. The female
participants constituted only 4% of the sample. Out of the total sam-
ple, 37% reported to have accident experience while working in the
company and only 6% reported that they had accident experience
in the previous year, i.e., in 2002.
The reason behind opting for a large sample like this was that a
smaller sample selection from various departments in each indus-
trial unit was looked upon with apprehension by the employees
since the matter under study is related to statutory requirements
of safety of workers. They feared that if the data collected by the
researcher is given to the management for any reason, top man-
agement will be able to identify each respondent. A first attempt
for a smaller sample selection in the first organization met with
failure, as the participants were reluctant to respond to the ques-
tionnaire survey due to the aforesaid reasons. The actual reason was
identified after discussions with trade union leaders. Thereafter, it
was decided to give the questionnaire to all eligible respondents
present during the dayshift. Since the survey in each unit took a
minimum of three weeks time during which almost all the work-
ers could work during the dayshift by rotation, almost all workers
could participate in the survey.
2.2. Survey instrument
The questionnaire contained 35 questions to measure the per-
ceptions of the employees about the six safety management
practices, 12 questions to measure the participants’ level of safety
knowledge and safety motivation and 12 questions to measure
self-rated safety compliance and safety participation. This was
prepared based on review of related literature and theory and it
contained questions covering areas of management commitment
(9 items), safety training (6 items), workers’ involvement (5 items),
Questionnaire survey details
Given Returned % Response
280 184 65
441 337 73
302 195 66
220 160 73
256 197 73
205 152 76
204 135 69
267 206 76
2181 1566 72
safety communication and feedback (5 items), safety rules and
procedures (5 items), safety promotion policies (5 items), safety
knowledge (6 items), safety motivation (6 items), safety compli-
ance (7 items) and safety participation (5 items). The content and
substance of most of these 59 questions (Table 2) were taken from
previous questionnaires of Cheyne et al. (1998), Cox and Cheyne
(2000), Coyle et al. (1995), Flin et al. (2000), Glendon and Litherland
(2001), Neal et al. (2000), Varonen and Mattila (2000), Vredenburgh
(2002), Williamson et al. (1997), Zohar (1980) and various other
reports. The scales were further developed by including compli-
mentary additions of new items and some items were reworded
and rephrased to suit local working practices and culture. Care
was taken to maintain the conceptual meaning of the scales close
to the operationalizations made by Cheyne et al. (1998) and Neal
et al. (2000). The contents of this draft questionnaire were dis-
cussed with senior safety professionals from industries and senior
professors in management studies to ensure face validity. After
considering each item in detail, necessary changes were made by
simplifying, rewording, removing and replacing some of them. A
pilot survey was conducted on a selected sample of 100 work-
ers from five industrial units to get the feedback about the clarity
of the items. Subsequently, some of the negatively worded items
were changed to positive for simplicity. The reliability and correla-
tions were examined and those items with item-to-total correlation
below 0.4 were dropped from the survey instrument.
This resulted in dropping 3 items from safety compliance and
one each from all the remaining measures, and hence, the final
questionnaire contained 47 items. It was decided to give the ques-
tions in English as well as the local language based on the feedback
from the workers. Each item was measured on a Likert scale. A
Likert scale is a type of response scale often used in question-
naires, and is the most widely used scale in questionnaire survey
based research (Geller et al., 1996; Grote and Kunzler, 2000). In this
research, respondents were asked to give their preference on a 5-
point Likert scale (strongly disagree, disagree, neither disagree nor
agree, agree, and strongly agree) in order to evaluate the respon-
dents’ level of agreement with each item. Most of the items were
phrased positively and a few items negatively so that strong agree-
ment in the former and strong disagreement in the latter resulted
in a higher score in favour of safety for the concerned item.
The four page questionnaire ready for administration consisted
of two parts. Ten demographic questions about the name of the
company, department, designation, qualification, age, sex, number
of years of experience, accident history, number of accidents expe-
rienced in 2002 while working in this company which resulted in
at least two lost working days as per the Indian Factories Act, 1948
and the number of working days lost due to above accidents in
2002 constituted the first part. The 47 statements related to safety
formed the second part. Space was provided beside each statement
to mark the preference in the 5-point Likert scale. To maintain
anonymity of the respondent, information such as name, badge
number or signature were avoided in the questionnaire.
 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093 2087

Table 2
Questionnaire items.

Management commitment
1. Safety is given high priority by the management.
2. Safety rules and procedures are strictly followed by the management.
3. Corrective action is always taken when the management is told about unsafe practices.
4. In my workplace managers/supervisors do not show interest in the safety of workers.
5. Management considers safety to be equally important as production.
6. Members of the management do not attend safety meetings.a
7. I feel that management is willing to compromise on safety for increasing production.
8. When near-miss accidents are reported, my management acts quickly to solve the problems.
9. My company provides sufficient personal protective equipments for the workers.

Safety training
1. My company gives comprehensive training to the employees in workplace health and safety issues.
2. Newly recruits are trained adequately to learn safety rules and procedures.
3. Safety issues are given high priority in training programmes.
4. I am not adequately trained to respond to emergency situations in my workplace.a
5. Management encourages the workers to attend safety training programmes.
6. Safety training given to me is adequate to enable to me to assess hazards in workplace.

Workers’ involvement
1. Management always welcomes opinion from employees before making final decisions on safety related matters.
2. My company has safety committees consisting of representatives of management and employees.
3. Management promotes employees involvement in safety related matters.
4. Management consults with employees regularly about workplace health and safety issues.
5. Employees do not sincerely participate in identifying safety problems.a

Safety communication and feedback

1. My company doesn’t have a hazard reporting system where employees can communicate hazard information before incidents occur.a
2. Management operates an open door policy on safety issues.
3. There is sufficient opportunity to discuss and deal with safety issues in meetings.
4. The target and goals for safety performance in my organization are not clear to the workers.
5. There is open communications about safety issues in this workplace.

Safety rules and procedures

1. The safety rules and procedures followed in my company are sufficient to prevent incidents occurring.
2. The facilities in the safety department are not adequate to meet the needs of my organization.a
3. My supervisors and managers always try to enforce safe working procedures.
4. Safety inspections are carried out regularly.
5. The safety procedures and practices in this organization are useful and effective.

Safety promotion policies

1. In my company safe conduct is considered as a positive factor for job promotions.
2. In my company employees are rewarded for reporting safety hazards (thanked, cash or other rewards, recognition in news letter, etc.)
3. In my company safety week celebration and other safety promotional activities arranged by the management are very effective in creating safety awareness
among the workers.
4. There exists very healthy competition among the employees to find out and report unsafe condition and acts.
5. Our supervisor becomes very unhappy and angry when employees find out and report unsafe conditions and acts in our section.a

Safety knowledge
1. I know how to perform my job in a safe manner.
2. I know how to use safety equipments and standard work procedures.
3. I know how to maintain or improve workplace health and safety.
4. I know how to reduce the risk of accidents and incidents in the workplace.
5. I know what are the hazards associated with my jobs and the necessary precautions to be taken while doing my job.
6. I don’t know what to do and whom to report if a potential hazard is noticed in my workplace.a

Safety motivation
1. I feel that it is important to maintain safety at all times.
2. I believe that safety at workplace is a very important issue.
3. I feel that it is necessary to put efforts to reduce accidents and incidents at workplace.
4. I believe that safety that can be compromised for increasing production.a
5. I feel that it is important to encourage others to use safe practices.
6. I feel that it is important to promote safety programmes.

Safety compliance
1. I use all necessary safety equipments to do my job.
2. I carry out my work in a safe manner.
3. I follow correct safety rules and procedures while carrying out my job.
4. I ensure the highest levels of safety when I carry out my job.
5. Occasionally due to lack of time, I deviate form correct and safe work procedures.a
6. Occasionally due to over familiarity with the job, I deviate from correct and safe work procedures.a
7. It is not always practical to follow all safety rules and procedures while doing a job.a

Safety participation
1. I help my co-workers when they are working under risky or hazardous conditions.a
2. I always point out to the management if any safety related matters are noticed in my company.
3. I put extra effort to improve the safety of the workplace.
4. I voluntarily carryout tasks or activities that help to improve workplace safety.
5. I encourage my co-workers to work safely.
a
Item removed from final questionnaire.
2088 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093

Table 3
Means, standard deviation and correlations.

Mean S.D. Age Exp qlf MC TR WI CO SR SP KNO MOT COM PAR ACC

Age 43.22 8.71 –

Exp 17.27 8.53 .86
qlf 3.63 1.42 .40 .33
MC 3.37 0.83 −.08 −.07 −.18
TR 3.58 0.84 −.04 −.03 −.14 .77
WI 3.29 0.77 −.05 −.02 −.18 .72 .68
CO 3.28 0.84 −.04 −.02 −.14 .79 .75 .75
SR 3.40 0.90 −.05 −.05 −.12 .84 .75 .70 .74
SP 3.17 0.84 −.06 −.09 −.02 .69 .64 .60 .64 .69
KNO 3.91 0.59 −.00 .02 −.15 .43 .43 .39 .45 .44 .35
MOT 4.45 0.48 −.08 −.04 −.13 .24 .28 .23 .22 .21 .19 .42
COM 3.88 0.70 −.00 .00 −.07 .51 .48 .40 .49 .51 .41 .65 .39
PAR 3.80 0.61 −.02 −.01 −.03 .39 .39 .36 .39 .40 .39 .52 .43 .60
ACC 0.08 0.34 .04 .04 .07 −.08 −.05 −.09 −.07 −.05 −.06 −.09 −.01 −.07 −.06 –

Correlations of 0.06 or higher are significant at p ≤ 0.05 level.

Correlations of 0.07 or higher are significant at p ≤ 0.01 level.
Correlations of 0.08 or higher are significant at p ≤ 0.001 level.
Abbreviations: number of years of experience (exp); qualification (qlf); management commitment (MC); safety training (TR); workers’ involvement (WI); safety communica-
tion and feedback (CO); safety rules and procedures (SR); safety promotion policies (SP); safety knowledge (KNO); safety motivation (MOT); safety compliance (COM); safety
participation (PAR); number of accident experience in the year 2002 (ACC).

2.3. Data analysis 3.1. Unidimensionality analysis

Descriptive statistics and correlations of the studied variables
were first analyzed. Confirmatory factor analysis was used to verify
the unidimensionality and reliability of the six safety management
practices (management commitment, safety training, safety com-
munication, workers’ involvement, safety rules and procedures and
safety promotion policies), two determinants of safety performance
(safety knowledge and safety motivation) and two components
of safety performance (safety compliance and safety participa-
tion).
Structural equation modelling (SEM) techniques were used to
conduct the path analyses to test the hypotheses and the goodness
of fit of the various models. AMOS-4 software was used for CFA and
path analysis. The remaining analyses were done using SPSS-10
software.
Unidimensionality refers to the existence of a single con-
struct/trait underlying a set of measures (Hair et al., 1998). Removal
of items that reduce unidimensionality helps to solve the problems
associated with unidimensionality. An instrument can be fine-
tuned in this manner. Individual items in the model are investigated
to see how closely they represent the same construct. A Compara-
tive Fit Index (CFI) of 0.9 or higher for the model implies that there
is strong evidence of unidimensionality (Byrne, 1994; Issac et al.,
2006). The unidimensionality of the instrument used in the current
study was tested by computing CFI values for all the measures. The
results are shown in Table 4. Most of the CFI values are found to be
above 0.95 satisfying the more stringent criteria put forward by Hu
and Bentler (1995), demonstrating strong unidimensionality for all
the scales.

3.2. Reliability analysis

3. Results
Table 3 contains the means, standard deviations and inter-
correlations of all the measures. There are significant negative
correlations between the six safety management practices and self-
reported accident data indicating that better safety management
practices reduce accidents.
All the six safety management practice scores have significant
positive correlations with safety knowledge and safety motiva-
tion suggesting support for hypotheses H1 and H2 . Moreover, both
safety knowledge and safety motivation showed significant posi-
tive correlations with safety compliance and safety participation,
suggesting support for hypotheses H3 and H4 .
Unidimensionality alone, though a necessary condition, is not
sufficient by itself to establish the usefulness of a scale. Once uni-
dimensionality is substantiated, its ‘statistical reliability’ should
be assessed before it is subjected to any further validation anal-
ysis (Sureshchander et al., 2001). Even a perfectly unidimensional
(and otherwise construct valid) scale would be rendered futile if
the resultant aggregate score is ascertained basically by measure-
ment error, with the values of the scores broadly fluctuating over
repeated measures (Gerbing and Anderson, 1988).
Reliability is defined as the proportion of observed score vari-
ance that is attributable to true score variance. There are several
methods to establish the reliability of a measuring instrument,

Table 4
Results of confirmatory factor analysis: unidimensionality and reliability coefficients for safety management practices, determinants and components of safety performance.

Sl. No Safety management measures No. of items Comparative Fit Index (CFI) Cronbach’s alpha (˛)

1 Management commitment 8 0.96 0.86

2 Safety training 5 0.99 0.82
3 Worker involvement in safety 4 0.95 0.69
4 Safety communication and feedback 4 0.98 0.70
5 Safety rules and procedures 4 0.99 0.81
6 Safety promotion policies 4 0.94 0.64
7 Safety knowledge 5 0.99 0.77
8 Safety motivation 5 0.97 0.72
9 Safety compliance 4 0.99 0.76
10 Safety participation 4 0.98 0.66
 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093 2089

Fig. 2. Hypothesized model relating safety management practices, determinants of safety performance and components of safety performance.

and the internal consistency method is the most commonly used
method in studies with cross-sectional design. The internal consis-
tency is estimated using a reliability coefficient called Cronbach’s
alpha (˛) (Cronbach, 1951). An ˛ value of 0.70 or above is considered
to be the criterion for demonstrating strong internal consistency
of established scales (Nunnally, 1978). In the case of exploratory
research, ˛ value of 0.60 or above is also considered as significant
(Hair et al., 1998). Table 4 reveals that all of the six safety man-
agement practices, two determinants of safety performance and
two components of safety performance have good reliability sug-
gesting that the survey items were appropriate indicators of their
respective constructs.
3.3. Path analysis
We conducted a path analysis using the AMOS-4 program to
test the hypotheses advanced. The hypothesized model shown in
Fig. 2 was tested first. The present study used a set of different
types of fit measures (Hair et al., 1998; Hoyle, 1995; Kline, 1998)
such as, Chi-square values (2 ), Comparative Fit Index (CFI), the
Bentler–Bonnett Fit Index (NFI), the Tucker–Lewis Fit Index (TLI),
and root mean square error of approximation (RMSEA) to deter-
mine the goodness of fit of a model. The recommended values for
CFI, NFI and TLI are higher than 0.9, and RMSEA value less than
or equal to 0.06 for good model fit (Hu and Bentler, 1999). A non-
significant Chi-square indicates that the hypothetical model fits the
data.
Table 5 shows the fit indices of the hypothesized and modified
models. In Model 1, the indirect relationships of the six safety man-
agement practices on safety performance components were tested.
The results show that safety knowledge and safety motivation
predicted both safety compliance and safety participation. Fur-
thermore, safety training, safety communication and feedback, and
safety rules and procedures have indirect relationships with safety
compliance as well as safety participation, mediated by safety
knowledge. It is found that safety training has indirect relation-

Table 5
Model specification and fit indices.

Model Feature 2 d.f. p CFI NFI TLI RMSEA

1 With all indirect paths 665.1 14 <0.01 0.94 0.94 0.80 0.17
2 MC-KNO path deleted 665.3 15 <0.01 0.94 0.94 0.81 0.16
3 WI-KNO path deleted 665.5 16 <0.01 0.94 0.94 0.83 0.14
4 SP-KNO path deleted 665.8 17 <0.01 0.94 0.94 0.84 0.13
5 SP-MOT path deleted 666.2 18 <0.01 0.94 0.94 0.85 0.12
6 CO-MOT path deleted 667.9 19 <0.01 0.94 0.94 0.85 0.11
7 MC-MOT path deleted 669.4 20 <0.01 0.94 0.94 0.86 0.11
8 SR-MOT path deleted 670.9 21 <0.01 0.94 0.94 0.87 0.10
9 WI-MOT path deleted 673.6 22 <0.01 0.94 0.94 0.87 0.10
10 MC-COM direct path added 490.6 21 <0.01 0.96 0.95 0.90 0.08
11 SP-PAR direct path added 389.9 20 <0.01 0.97 0.96 0.92 0.06
12 WI-PAR direct path added 382.8 19 <0.01 0.97 0.96 0.92 0.06
13 SR-COM direct path added 370.0 18 <0.01 0.97 0.97 0.91 0.05
14 Error terms of KNO and MOT freely estimated 192.6 17 <0.01 0.98 0.98 0.96 0.04
15 Error terms of COM and PAR freely estimated 24.1 16 0.09 0.99 0.99 0.99 0.02

Note: 2 : Chi-square, d.f.: degrees of freedom, p: probability level, CFI: Comparative Fit Index, NFI: Bentler–Bonnett Normed Fit Index, TLI: Tucker–Lewis Fit Index, RMSEA:
root mean square error of approximation, MC: management commitment, TR: safety training, WI: workers’ involvement in safety, CO: safety communication and feedback,
SR: safety rules and procedures, SP: safety promotion policies, KNO: safety knowledge, MOT: safety motivation, COM: safety compliance, PAR: safety participation.
2090 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
Fig. 3. Final model with standardized path coefficients.
ships with safety compliance and safety participation, mediated by
safety motivation. However, the remaining path coefficients were
non-significant. Moreover, both TLI and RMSEA values were below
the acceptable limits indicating that the model does not fit the data
adequately.
Model modification was done using model trimming procedure
by deleting the paths with non-significant path coefficients, one at
a time. That path with the lowest value of critical ratio (CR) was
taken as the first candidate for deletion. Critical ratio (CR) is the
ratio between the parameter value and the standard error of the
parameter, and is greater than 1.96 if p < 0.05. Model 2 was obtained
by deleting the non-significant path relating management commit-
ment and safety knowledge. The fit indices are reported in Table 5.
Following this procedure, we reached at Model 9 which reported
all significant path coefficients. While testing each model, the Chi-
square difference test was used (Kline, 1998). A non-significant
value of this statistic suggests that the overall fit of the two models
is comparable (Kline, 1998), and the principle of parsimony then
suggests that the model with the highest degree of freedom is the
better choice (Nachtigall et al., 2003). The fit indices indicate that
the model does not fit the data adequately (2(22) = 673.6, p < 0.01,
CFI = 0.94, NFI = 0.94, TLI = 0.87, RMSEA = 0.10). Even though a sig-
nificant Chi-square tends to be observed with large sample size
(Hu and Bentler, 1995), the TLI and RMSEA values are not within
the acceptable limits. Therefore, Model 9 is not accepted.
The table of modification indices in the model output showed
that the values associated with the regression paths flowing from
management commitment to safety compliance and also from
safety promotion policies to safety participation are high. Since
these direct relationships are logical and meaningful, the model
was re-specified in the next two steps with the above paths spec-
ified as free parameters to obtain Model 11. Further review of
modification indices suggested direct path from workers’ involve-
ment in safety to safety participation and again from safety rules
and procedures to safety compliance. This resulted in Model 13
which reported acceptable model fit indices except a significant
Chi-square (2(18) = 370, p < 0.01, CFI = 0.97, NFI = 0.97, TLI = 0.91,
RMSEA = 0.05).
Further review of modification indices showed that correlated
error between safety knowledge and safety motivation and also
between safety compliance and safety participation are excep-
tionally high. Unquestionably, the specification of correlated error
terms for the purpose of achieving a better fitting model is not an
acceptable practice. But, this can also be done if such specifica-
tion is supported by a strong substantive and/or empirical rationale
(Joseberg, 1993). Considering the mentioned correlations logical
based on the above argument, the model was subsequently re-
specified with these two parameters freely estimated to reach at
the final model (Model 15) depicted in Fig. 3. The results given
in Table 5 show that the Chi-square has become non-significant
and all other fit indices have improved considerably (2(16) = 24,
p = 0.09, CFI = 0.99, NFI = 0.99, TLI = 0.99, RMSEA = 0.02) indicating
good model fit. The normed Chi-square (2 /d.f.) value reached 1.5
which is within the recommended range of 1.0 and 2.0.
The finalized model (Fig. 3) reveals that the squared multiple
correlations (R2 ) for the endogenous variables in the model are
0.22 for safety knowledge, 0.17 for safety motivation, 0.51 for safety
compliance and 0.37 for safety participation. The model shows that
out of the six safety management practices, only safety training,
safety communication and feedback and safety rules and proce-
dures predict safety knowledge. Hence, hypothesis H1 is partially
supported. Similarly, only safety training predict safety motivation.
Hence, hypothesis H2 is partially supported. It is seen that safety
knowledge predicts both safety compliance and safety participa-
tion supporting hypothesis H3 . Hypothesis H4 also is supported
because safety motivation predicts both the safety performance
components. Since there is no significant direct path between
safety training and safety compliance, we then conclude that safety
knowledge is a mediator between safety training and safety com-
pliance. Similarly, the absence of a significant direct path from
safety communication and feedback to safety compliance indicates
that safety knowledge mediates the relationship between safety
 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
communication and feedback and safety compliance. However, a
significant direct path from safety rules and procedures to safety
compliance indicates that safety knowledge partially mediates the
relationship between safety rules and procedures and safety com-
pliance. Overall, hypothesis H5 receives only partial support.
Hypothesis H6 predicted that safety motivation mediates the
relationship between the six safety management practices and
safety compliance. But, Fig. 3 reveals that safety compliance is pre-
dicted by safety motivation and safety motivation is predicted by
only one safety management practice, namely, safety training. Since
there is no significant direct path connecting safety training and
safety compliance, we can conclude that safety motivation is medi-
ator between safety training and safety compliance. Hence, hypoth-
esis H6 receives only partial support. With regard to hypothesis H7 ,
since there are no significant direct paths connecting safety train-
ing, safety communication and feedback and safety rules and pro-
cedures to safety participation, we can conclude that safety knowl-
edge mediates the relationship between these three safety manage-
ment practices and safety participation. Since the other three safety
management practices, namely, management commitment, work-
ers’ involvement in safety and safety promotion policies do not
predict safety knowledge, we conclude that H7 receives only partial
support. Due to the absence of a direct significant path from safety
training to safety participation, we can conclude that safety moti-
vation mediates the relationship between safety training and safety
participation. Since no other safety management practices predict
safety motivation, hypothesis H8 receives only partial support.
Contrary to expectations, this study revealed some significant
direct paths between certain safety management practices and the
safety performance components. Management commitment was
found predict safety compliance directly without influencing safety
knowledge or safety motivation. However, safety rules and pro-
cedures showed a direct path to safety compliance along with an
indirect path through safety knowledge, revealing partial media-
tion. Workers’ involvement in safety and safety promotion policies
showed significant direct relationships with safety participation
indicating the predictive capacity.
4. Discussion
The first purpose of this study was to establish the unidimen-
sionality and reliability of the six safety management practices, two
determinants and two components of safety performance. These
have been successfully demonstrated initially so that further anal-
yses of the data can be carried out.
Another purpose of the study was to test the direct and medi-
ating role of safety knowledge and safety motivation on the
relationship between the six safety management practices and
safety performance components. The results of the path analysis
provided partial support of the hypothesized model. It is evident
from Fig. 3 that the path coefficients are significant and the overall
model has an acceptable fit to the data. As there are very few studies
relating to employee safety in Indian societies, this study provides
sound insight into safety management practices and safety perfor-
mance.
Out of the six safety management practices, only safety train-
ing, safety communication and feedback and safety rules and
procedures were found to predict safety knowledge. These three
safety management practices which contribute towards transfer-
ring information regarding the methods of carrying out their job in
the healthiest and safest way possible is expected to improve the
safety knowledge of the employees. This finding is in line with the
observations of Burke et al. (2002), Stout et al. (1997) and Smith-
Crowe et al. (2003). Even though all the six safety management
practices were found to have significant positive correlation with
2091
safety knowledge, the remaining three of them failed to predict
safety knowledge. It can be argued that employees’ perception of
management commitment give an overall picture derived from
the totality of the employees’ assessment about the interest of the
management in the safety and health of employees as manifested
in various activities and initiatives of the management towards
safety. Hence, this may not necessarily predict safety knowledge of
employees. The perception of employees on the level to which the
management encourages or welcomes employees’ involvement in
safety related decision making and activities, and the safety promo-
tional policies of the management did not predict safety knowledge.
This is an important finding because involving employees in safety
matters and encouraging and offering rewards for safe working is
a traditional way of managing safety at work and may produce the
desired results without considerably affecting the workers’ safety
knowledge.
Another finding of this study that only safety training predicted
safety motivation, was quite revealing. Safety promotion policies
were expected to motivate employees with rewards and incentives
for safe working. It appears that the items in safety promotion poli-
cies scale could not capture what it was intended to. The item “In
my company safe conduct is considered as a positive factor for job
promotions” might have received erratic responses because such a
policy may not be practical in the state of Kerala where trade union
activities are comparatively high. Workers’ involvement in safety
was also expected to motivate the employees but the study failed
to establish that relationship.
Workers’ involvement in safety and safety promotion policies
were found to have reliability coefficients lower than 0.7. This
might have influenced their relationship with safety knowledge
and safety motivation.
Analyzing the relationship between safety management prac-
tices and safety performance, it is found that management
commitment and safety rules and procedures predicted safety
compliance directly whereas safety training and safety commu-
nication and feedback predicted safety compliance indirectly. It is
also evident that workers’ involvement in safety and safety promo-
tion policies predicted safety participation directly whereas safety
training, safety communication and feedback, and safety rules and
procedures predicted safety participation indirectly. These find-
ings provide evidence to show that by focusing on implementation
of safety management practices, organizations are most certain to
reap benefits in terms of safety performance. A possible explanation
for this argument is that employees may perceive implementa-
tion of safety management practices as indicators of organizations’
commitment towards employees’ safety. Earlier work by Mearns
and Hope (2005) in the offshore sector also indicated that greater
organizational investment in health and safety related activities
led to worker loyalty in terms of compliance with safety rules and
regulations.
The lower reliability coefficient of safety participation (˛ = 0.66)
might have affected the predictive capacity of the safety manage-
ment practices. There is a possibility that the items could not extract
what they were expected to. Since the population in this study is
new to such investigations, it is possible that they took extra caution
in responding to the questions and finally erred in the process.
Testing the mediation effects, safety knowledge emerged as a
distinct and clear mediator in the relationship between three major
safety management practices and the two components of safety
performance. Analyzing the mediating role further, it was found
that safety motivation mediates the relationship between safety
training and the two components of safety performance. Some of
the behavioural characteristics of workers like reluctance to com-
ply with safety rules, a kind of non-participation in safety related
matters and certain degree of complacency with regard to safety
can be attributed to ignorance about the various processes, associ-
2092 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
ated hazards and their consequences, ignorance about the correct
usage of personal protective equipments, etc. This study suggests
that to get positive results in terms of safety compliance and par-
ticipation, safety management practices like safety training, safety
communication and enforcement of safety rules and procedures
must be designed to produce changes in safety knowledge and
motivation of the workers. When Neal et al. (2000) reported the
mediating role of safety knowledge and safety motivation on the
relationship between safety climate (sum of perceptions of man-
agement values, communication, training and safety systems) and
safety performance components, this study has gone on to identify
the relative importance of various safety management practices.
The present study investigated the relationship between six
safety management practices and safety performance. The results
suggest that some safety management practices can have indepen-
dent effects on knowledge and motivation, and that both factors
are important determinants of safety performance. This informa-
tion can be used in the design of interventions with emphasis
on the most influencing safety management practices to ensure
that they target both the mediators, namely safety knowledge and
motivation, to produce best results. Interventions aimed solely at
improving any one are unlikely to be as effective as interventions
that target both the mediators. This research can also help in the
development of measurement systems to evaluate the effective-
ness of safety management practices. Incorporating assessments of
safety knowledge, safety motivation, safety compliance and safety
participation along with the safety management practices into the
safety monitoring systems will provide a more comprehensive
assessment, in terms of the effectiveness of the safety management
practices and their operation as well.
The direct influence of management commitment on safety
compliance can be considered as a result of individual wisdom
of the employees, earned from the overall interest shown by the
managements towards the safety of their employees, to protect
themselves from accidents. The direct relationship between work-
ers’ involvement in safety and safety participation appears like a
natural reaction from the employees’ side for empowering and
involving them in safeguarding themselves. Similarly, the direct
influence of safety promotion policies on safety participation may
be the result of encouragement and rewards and also of the human
desire to be accepted and valued in a group.
The most important finding of this study is the role of safety
training in safety management. It is found to predict safety knowl-
edge, safety motivation, safety compliance and safety participation.
Managements need to give the highest level of priority to safety
training convincing the employees about the need for safety perfor-
mance. Safety training may be designed to impart good knowledge
about the various processes, associated hazards and the safety mea-
sures to be taken by the employees in case of emergencies. Such
safety training programmes may be conducted regularly and the
participation may be made compulsory. To motivate the employ-
ees, some of them should be given opportunities to present case
studies and the others must actively participate in discussions. Reg-
ular evaluation of safety knowledge, level of safety motivation and
safety skills also must be made an integral part of safety training
programmes.
This study gains importance since it was conducted in an
unstudied population in India, which is a developing country. The
findings being meaningful indicate that employees’ awareness and
priority for safety in India are reasonably good, opening avenues
for more research in the field of safety.
4.1. Limitations of the study
This research has the limitations that are inherent in studies
using perceptions and self-reported response data. The unit of anal-
ysis was limited to individual level instead of organizational level
due to the absence of sufficient number of participating organiza-
tions. Since this study was a cross-sectional one carried out during
a short period of time, a longitudinal study to validate the findings
is in progress. Safety skill could have been included as a third deter-
minant of safety performance but was omitted due to the difficulty
in its measurement. Safety skill sometimes might have accounted
for the direct relationships found between management commit-
ment and safety compliance and also safety promotion policies and
safety participation.
5. Conclusions
The results of this study provide strong empirical support for
the theoretical model that antecedents, determinants and com-
ponents of safety performance are closely associated. The study
demonstrated the validity and reliability of the six perceived safety
management practices, two determinants and two components
of safety performance. The predictive capacity of the six safety
management practices on safety knowledge, safety motivation,
safety compliance and safety participation were demonstrated. The
study also demonstrated that the perceptions of safety manage-
ment practices influence safety performance through their effects
on safety knowledge and safety motivation which act as mediators.
Direct effects of four safety management practices on components
of safety performance were also identified. The results of this study
also highlight the need of safety training of the workforce. These
findings provide valuable guidance for researchers and practition-
ers for identifying the mechanisms by which they can improve
safety of workplace.
References
Borman, W.C., Motowidlo, S.J., 1993. Expanding the criterion domain to include
elements of contextual performance. In: Schmitt, N., Borman, W.C., Associates
(Eds.), Personnel Selection in Organizations. Jossey-Bass, San Francisco, CA, pp.
71–98.
Bowander, B., 1987. The Bhopal accident. Technological Forecasting and Social
Change 32 (2), 169–182.
Burke, M.J., Sarpy, S.A., Tesluk, P.E., Smith-Crowe, K., 2002. General safety per-
formance: a test of a grounded theoretical model. Personnel Psychology 55,
429–457.
Byrne, B.M., 1994. Structural Equation Modeling with EQS and EQS/Windows—Basic
Concepts, Application and Programming. Sage Publications, USA.
Campbell, J.P., McCloy, R.A., Oppler, S.H., Sager, C.E., 1993. A theory of performance.
In: Schmitt, N., Borman, W.C., Associates (Eds.), Personnel Selection in Organi-
zations. Jossey-Bass, San Francisco, CA, pp. 71–98.
Cavazza, N., Serpe, A., 2009. Effects of safety climate on safety norm violations:
exploring the mediating role of attitudinal ambivalence toward personal pro-
tective equipment. Journal of Safety Research 40, 277–283.
Cheyne, A., Cox, S., Oliver, A., Tomas, J., 1998. Modelling safety climate in the pre-
diction of levels of safety activity. Work and Stress 12, 255–271.
Chouhan, T.R., 2005. The unfolding of Bhopal disaster. Journal of Loss Prevention in
the Process Industries 18 (4–6), 205–208.
Cohen, A., 1977. Factors in successful safety programs. Journal of Safety Research 9,
168–178.
Cohen, H.H., Cleveland, R.J., 1983. Safety program practices in record-holding plants.
Professional Safety 28, 26–33.
Cohen, A., Smith, M., Anger, W., 1979. Self-protective measures against workplace
hazards. Journal of Safety Research (11), 121–131.
Cohen, A., Smith, M., Cohen, H.H., 1975. Safety Program Practices in High Versus
Low Accident Rate Companies. HEW Publication No. (NIOSH) 75-185. National
Institute of Occupational Health and Safety, Cincinnati, OH.
Cooper, M.D., Philips, R.A., 2004. Exploratory analysis of the safety climate and safety
behaviour relationship. Journal of Safety Research 35 (5), 497–512.
Cox, S.J., Cheyne, A.J.T., 2000. Assessing safety culture in offshore environments.
Safety Science 34, 111–129.
Cox, S., Flin, R., 1998. Safety culture: philosopher’s stone or man of straw? Work and
Stress 12 (3), 189–201.
Coyle, I., Sleeman, S., Adams, D., 1995. Safety climate. Journal of Safety Research 22,
247–254.
Cronbach, L.J., 1951. Coefficient alpha and the internal structures of tests. Psychome-
trika 16, 297–334.
Dedobbeleer, N., Beland, F., 1991. A safety climate measure for construction sites.
Journal of Safety Research 22, 97–103.
 M.N. Vinodkumar, M. Bhasi / Accident Analysis and Prevention 42 (2010) 2082–2093
Dedobbeleer, N., Beland, F., 1998. Is risk perception one of the dimensions of safety
climate? In: Freyer, A., Williamson, A. (Eds.), Occupational Injury: Risk Preven-
tion and Intervention. Taylor and Francis, London, pp. 73–81.
DePasquale, J.P., Geller, E., 1999. Critical success factors for behaviour based safety:
a study of twenty industry-wide applications. Journal of Safety Research 30,
237–249.
Donald, I., Canter, D., 1994. Employees attitudes and safety in the chemical industry.
Journal of Loss Prevention in the Process Industries 7, 203–208.
Eckhardt, R., 1996. Practitioner’s influence on safety culture. Professional Safety 7,
23–25.
Fernandez-Muniz, M., Montres-Peon, J.M., Vazquez-Ordas, C.J., 2007. Safety manage-
ment system: development and validation of a multidimensional scale. Journal
of Loss Prevention in the Process Industries 20, 52–68.
Flin, R., Mearns, K., Fleming, M., Gordon, R., 1996. Risk Perceptions and Safety in the
Offshore Oil and Gas Industry (OTII 94454). HSE Books, Suffolk.
Flin, R., Mearns, K., O’Connor, P., Bryden, R., 2000. Measuring safety climate: identi-
fying the common features. Safety Science 34, 177–193.
Geller, E.S., Roberts, D.S., Gilmore, M.R., 1996. Predicting propensity to actively care
for occupational safety. Journal of Safety Research 27, 1–8.
Gerbing, D.W., Anderson, J.C., 1988. An updated paradigm for scale development
incorporating unidimensionality and its measurement. Journal of Marketing
Research 25, 186–192.
Glendon, A.I., Litherland, D.K., 2001. Safety climate factors, group differences and
safety behavior in road construction. Safety Science 39, 157–188.
Glendon, A.I., Stanton, N.A., 2000. Perspectives on safety culture. Safety Science 34,
193–214.
Griffin, M.A., Neal, A., 2000. Perceptions of safety at work: a framework for linking
safety climate to safety performance, knowledge, and motivation. Journal of
Occupational Health and Psychology 5, 347–358.
Griffiths, D.K., 1985. Safety attitudes of management. Ergonomics 28, 61–67.
Grote, G., Kunzler, C., 2000. Diagnosis of safety culture in safety management audits.
Safety Science 34, 131–150.
Guldenmund, F.W., 2000. The nature of safety culture: a review of theory and
research. Safety Science 34, 215–257.
Gupta, J.P., 2002. The Bhopal gas tragedy: could it have happened in a developed
country? Journal of Loss Prevention in the Process Industries 15 (1), 1–4.
Hagan, P.E., Montgomery, J.F., O’Reilly, J.T., 2001. Accident Prevention Manual for
Business and Industry, 12th ed. NSC, Illinois, USA.
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis.
Prentice-Hall International, New Jersey, USA.
Harper, A.C., Cordery, J.L., de Klerk, N.H., Sevastos, P., Geelhoed, E., Gunson, C.,
Robinson, L., Sutherland, M., Osborn, D., Colquhoun, J., 1997. Curtin industrial
safety trial: managerial behavior and program effectiveness. Safety Science 24,
173–179.
Hofmann, D., Jacobs, R., Landy, F., 1995. High reliability process industries: individ-
ual, micro, and macro organizational influences on safety performance. Journal
of Safety Research 26, 131–149.
Hoyle, R.H., 1995. The structural equation modeling approach: basic concepts and
fundamental issues. In: Hoyle, R.H. (Ed.), Structural Equation Modeling: Con-
cepts, Issues, and Application. Sage Publications, Inc., Thousand Oaks, CA, pp.
1–55.
Hu, L.T., Bentler, P.M., 1995. Evaluating model fit. In: Hoyle, R.H. (Ed.), Structural
Equation Modeling. Concepts, Issues, and Applications. Sage, London, pp. 76–99.
Hu, L.T., Bentler, P.M., 1999. Cutoff criteria for fit indexes in covariance structure
analysis: conventional criteria versus alternatives. Structural Equation Modeling
6 (1), 1–55.
Huang, Y.H., Ho, M., Smith, G.S., Chen, P.Y., 2006. Safety climate and self-reported
injury: assessing the mediating role of employee safety control. Accident Anal-
ysis and Prevention 38, 425–433.
Issac, G., Rajendran, C., Anantharaman, R.N., 2006. An instrument for measurement
of customer perception of quality management in the software industry: an
empirical study in India. Software Quality Journal 14, 291–308.
Joseberg, K.G., 1993. Testing structural equation models. In: Bollen, K.A., Long, J.S.
(Eds.), Testing Structural Equation Models. Sage Publications Inc., Newbery Park,
CA, pp. 294–316.
Kirwan, B., 1998. Safety management assessment and task analysis—a missing link?
In: Hale, A., Baram, M. (Eds.), Safety Management: The Challenge of Change.
Elsevier, Oxford.
Kline, R.B., 1998. Principles and Practices of Structural Equation Modeling. The Guil-
ford Press, London.
Labodova, A., 2004. Implementing integrated management systems using a risk
analysis based approach. Journal of Cleaner Production 12, 571–580.
Lee, T., 1998. Assessment of safety culture at a nuclear reprocessing plant. Work and
Stress 12, 217–237.
Lee, T., Harrison, K., 2000. Assessing safety culture in nuclear power stations. Safety
Science 34, 61–97.
Mearns, K., Hope, L., 2005. Heal and Well Being in the Offshore Environment: The
Management of Personal Health. Research Report 305. HSE Books, Norwich.
2093
Mearns, K., Whitaker, S.M., Flin, R., 2003. Safety climate, safety management practice
and safety performance in offshore environments. Safety Science 41, 641–680.
Mohammed, S., 2002. Safety climate in construction site environment. Journal of
Construction Engineering and Management 128, 375–384.
Nachtigall, C.K., Ulf: Funke, Friedrich, Steyer, Rolf., 2003. Why should we use
SEM? Pros and cons of structural equation modeling. Methods of Psychological
Research Online 8 (2), 1–22.
Neal, A., Griffin, M.A., 1997. Perceptions of safety at work: developing a model to
link organizational safety climate and individual behaviour. In: Paper Presented
to the 12th Annual Conference of the Society for Industrial and Organizational
Psychology, St. Louis, MO.
Neal, A., Griffin, M.A., Hart, P.M., 2000. The impact of organizational climate on safety
climate and individual behavior. Safety Science 34, 99–109.
Nunnally, J.M., 1978. Psychometric Theory, 2nd ed. McGraw-Hill, New York.
Ostrom, L., Wilhelmsen, C., Kaplan, B., 1993. Assessing safety culture. Nuclear Safety
34 (2), 163–172.
O’Toole, M., 2002. The relationship between employees’ perceptions of safety and
organizational culture. Journal of Safety Research 33, 231–243.
Pidgeon, N., 1998. Safety culture: key theoretical issues. Work and Stress 12 (3),
202–216.
Pousette, A., Larsson, S., Torner, M., 2008. Safety climate cross-validation, strength
and prediction of safety behaviour. Safety Science 46, 398–404.
Rundmo, T., 1994. Association between safety and contingency measures and occu-
pational accidents on offshore petroleum platforms. Scandinavian Journal of
Work Environment and Health 20, 128–131.
Seo, D.C., 2005. An explicative model of unsafe work behaviour. Safety Science 43,
187–211.
Shafai-Sahrai, Y., 1971. An Inquiry into Factors that Might Explain Differences in
Occupational Accident Experience of Similar Size Firms in the Same Industry.
Division of Research, Graduate School of Business Administration, Michigan
State University, East Lansing, MI.
Shannon, H.S., Walters, V., Lewchuk, W., Richardson, J., Moran, L.A., Haines, T., Verma,
D., 1996. Workplace organizational correlates of lost-time accident rates in man-
ufacturing. American Journal of Industrial Medicine 29, 258–268.
Shannon, H., Mayr, J., Haines, T., 1997. Overview of the relationship between orga-
nizational and workplace factors and injury rates. Safety Science 26, 201–217.
Siu, O., Phillips, D.R., Leung, T., 2004. Safety climate and safety performance among
construction workers in Hong Kong: the role of psychological strains as media-
tors. Accident Analysis and Prevention 36 (3), 359–366.
Smith, M.J., Cohen, H.H., Cohen, A., Cleveland, R.J., 1975. On-site observations of
safety practices in plants with differential safety performance. In: National
Safety Congress Transactions, vol. 12, National Safety Council, Chicago.
Smith-Crowe, K., Burke, M.J., Landis, R.S., 2003. Organizational climate as a
moderator of safety knowledge-safety performance relationships. Journal of
Organizational Behavior 24, 861–876.
Stout, R.J., Salas, E., Kraiger, K., 1997. The role of trainee knowledge structures in
aviation team environments. International Journal of Aviation Psychology 7,
235–250.
Sureshchander, G.S., Rajendran, C., Anantharaman, R.N., 2001. A holistic model for
total quality service. International Journal of Service Industry Management 12,
378–412.
Tinmannsvik, R.K., Hovden, J., 2003. Safety diagnosis criteria—development and test-
ing. Safety Science 41, 575–590.
Turner, B.A., 1991. The development of a safety culture. Chemistry and Industry 4,
241–243.
Varonen, U., Mattila, M., 2000. The safety climate and its relationship to safety
practices, safety of work environment and occupational accidents in eight wood-
processing companies. Accident Analysis and Prevention 32, 761–769.
Vinodkumar, M.N., Bhasi, M., 2009. Safety climate factors and its relationship with
accidents and personal attributes in the chemical industry. Safety Science 47,
659–667.
Vredenburgh, A.G., 2002. Organizational safety—which management practices are
most effective in reducing employee injury rates? Journal of Safety Research 33,
259–276.
Williamson, A.M., Feyer, A., Cairns, D., Biancotti, D., 1997. The development of a
measure to safety climate: the role of safety perceptions and attitudes. Safety
Science 25, 15–27.
Wu, T.C., Chen, C.H., Li, C.C., 2008. A correlation among safety leadership, safety
climate and safety performance. Journal of Loss Prevention in the Process Indus-
tries 21, 307–318.
Zohar, D., 1980. Safety climate in industrial organizations: theoretical and applied
implications. Journal of Applied Psychology 65, 96–102.
Zohar, D., 2000. A group-level model of safety climate: testing the effect of group
climate on micro accidents in manufacturing jobs. Journal of Applied Psychology
85, 587–596.
Zohar, D., Luria, G., 2005. A multilevel model of safety climate: cross-level rela-
tionships between organization and group-level climates. Journal of Applied
Psychology 90, 616–628.