- Email: [email protected]
The impact of organizational climate on safety climate and individual behavior
Safety Science 34 (2000) 99±109
www.elsevier.com/locate/ssci
The impact of organizational climate on safety climate and individual behavior A. Neal a,*, M.A. Grin b, P.M. Hart c a
ARC Key Centre for Human Factors and Applied Cognitive Psychology, and School of Psychology, The University of Queensland, St. Lucia 4072, Australia b School of Management, Queensland University of Technology, Brisbane 4000, Australia c Department of Psychology, University of Melbourne, Melbourne 3010, Australia
Abstract Relatively little previous research has investigated the meechanisms by which safety climate aects safety behavior. The current study examined the eects of general organizational climate on safety climate and safety performance. As expected, general organizational climate exerted a signi®cant impact on safety climate, and safety climate in turn was related to selfreports of compliance with safety regulations and procedures as well as participation in safety-related activities within the workplace. The eect of general organizational climate on safety performance was mediated by safety climate, while the eect of safety climate on safety performance was partially mediated by safety knowledge and motivation. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Organizational climate; Safety climate; Safety performance
1. Introduction In recent years, there has been a shift in emphasis within the safety literature, away from individual level factors that might be responsible for accidents and incidents, such as error or non-compliance with safety procedures, towards organizational factors, such as safety climate (eg. Reason, 1990). A key assumption of much of this literature is that the relationship between safety climate and system safety is at least partially mediated by individual safety behavior. However, surprisingly little is known about the mechanisms by which safety climate in¯uences safety behaviors of individuals in organizations. Furthermore, relatively little is known about the * Corresponding author. Tel.: +61-7-3365-6372; fax: +61-7-3365-4466. E-mail address: [email protected] (A. Neal). 0925-7535/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0925-7535(00)00008-4
100
A. Neal et al. / Safety Science 34 (2000) 99±109
factors that in¯uence safety climate. The current paper examines the impact of general organizational climate on safety climate, and the impact of safety climate on the knowledge, motivation and performance of individuals in organizations. 2. Organizational climate and safety climate Organizational climate is a multidimensional construct that encompasses a wide range of individual evaluations of the work environment (James and James, 1989). These evaluations may refer to general dimensions of the environment such as leadership, roles, and communication (James and McIntyre, 1996) or to speci®c dimensions such as the climate for safety or the climate for customer service. General perceptions of the organizational context can in¯uence interactions among individuals (Grin and Mathieu, 1997), attitudes toward organizational rewards (Grin et al., 1995; Grin, 1996) and aective responses to the work environment (Michela et al., 1995; Hart et al., 1996a). Perceptions of the general organizational climate develop as individuals attribute meaning to their organizational context based on the signi®cance of the environment for individual values (James et al., 1990). Organizational climate, therefore, is thought to exert a strong impact on individual motivation to achieve work outcomes (Brown and Leigh, 1996). General organizational climate has also been found to in¯uence knowledge and skills by increasing participation in activities such as training (Morrison et al., 1997). Safety climate is a speci®c form of organizational climate, which describes individual perceptions of the value of safety in the work environment. A range of factors has been identi®ed as being important components of safety climate. These factors include: management values (e.g. management concern for employee well-being), management and organizational practices (e.g. adequacy of training, provision of safety equipment, quality of safety management systems), communication, and employee involvement in workplace health and safety. A range of studies have demonstrated that these factors predict safety-related outcomes, such as accidents and incidents (e.g., Zohar, 1980; Brown and Holmes, 1986; Dedobbeleer and Beland, 1991; DeJoy, 1994; Niskanen, 1994; Hofmann and Stetzer, 1996). No studies have investigated links between general organizational climate and speci®c safety climate. It is proposed that general organizational climate provides a context in which speci®c evaluations of the importance of safety are made. This means that the general organizational climate should predict speci®c safety climate (see Hypothesis 1). For example, if employees perceive that there is open communication in the organization, then they may also perceive that communication about safety is valued in the organization. Similarly, if employees perceive that the organization is supportive of their general welfare and well-being, they will be more likely to perceive that the organization values the safety of employees. These perceptions of the speci®c safety climate of the organization, in turn, are hypothesized to in¯uence safety behavior. The following section describes a theory of safety performance that can be used to derive predictions regarding the impact of safety climate on safety behavior.
A. Neal et al. / Safety Science 34 (2000) 99±109
101
3. Safety performance Neal and Grin (1997) proposed a model of safety performance based on current theories of job performance (Borman and Motowidlo, 1993; Campbell et al., 1993). This model distinguishes between performance components, determinants of performance, and performance antecedents. The components of performance represent the major dimensions of task-relevant behaviors involved in a given job. The model incorporates two dimensions of safety performance: compliance and participation. Safety compliance involves adhering to safety procedures and carrying out work in a safe manner. Safety participation involves helping coworkers, promoting the safety program within the workplace, demonstrating initiative, and putting eort into improving safety in the workplace. The determinants of safety performance represent the factors directly responsible for individual dierences in compliance and participation. Campbell et al. (1993) argue that there are only three determinants of individual dierence in performance: knowledge, skill, and motivation. Several authors have criticized this assumption, arguing that there may be other determinants of performance. Hesketh and Neal (1999), for example, argue that situational factors can sometimes produce invidivual dierences in performance. However, available evidence does suggest that knowledge, skill and motivation are important determinants of individual dierences in performance across a wide range of contexts. Safety behavior, therefore, should be determined by knowledge and skills necessary for particular behaviors and by the motivation of individuals to perform the behaviors. Furthermore, knowledge, skill, and motivation should have dierential eects on the dierent components of performance. It is expected that safety knowledge and skill should have a stronger relationship with compliance than with participation (see Hypotheses 2 and 3), while safety motivation should have a stronger relationship with participation than with compliance (see Hypotheses 4 and 5). An individual must understand how to perform work safely and have the skill to be able to do it in order to comply with safety procedures. However, safety knowledge and skill are likely to be less important for participatory activities, since these activities require more generic forms of knowledge and skill (termed ``tacit knowledge'' by Wagner and Sternberg, 1985). Motivation is likely to be more important for participation than for compliance, because participatory activities are frequently voluntary, whereas compliance is generally mandated. Antecedents of performance include individual-level factors such as ability, experience and personality, as well as group and organizational factors, such as leadership, group norms, and climate. General organizational climate and safety climate, therefore, are classi®ed as antecedents of safety performance. A key assumption of the Campbell et al. (1993) model is that the determinants of performance (knowledge, skill and motivation) must mediate the relationship between the antecedents and components of performance. The theory, therefore, predicts that knowledge, skill, motivation should mediate the relationship between safety climate and safety performance (see Hypotheses 6 and 7). Moreover, this suggests that sefety climate and safety behaviors mediate the relationship between general organizational climate and safety perofmance (see Hypothesis 8).
102
A. Neal et al. / Safety Science 34 (2000) 99±109
Neal and Grin (1997) tested the model of safety performance using an archival data set. As expected, compliance and participation were found to form separate factors, and individual safety knowledge was found to mediate at least some of the relationships between safety climate and safety performance. Unfortunately, safety motivation was not assessed in the archival data set, and several dimensions of safety climate were found to have direct eects on safety performance. The current study assessed safety climate and performance in a hospital setting, and included measures of both safety knowledge and safety motivation. Measures of skill were not included. The theoretical model is shown in Fig. 1. The following hypotheses were tested in the present study: H1. Organizational climate predicts safety climate H2. Knolwedge predicts both compliance and participation H3. The relationship between knolwedge and compliance is stronger than the relationship between knowledge and participation H4. Motivation predicts both compliance and participation. H5. The relationship between motivation and participation is stronger than the relationship between motivation and compliance. H6. Safety climate predicts both knowledge and motivation. H7. Knowledge and motivation mediate the relationship between safety climate and safety performance. H8. Safety climate, knowledge, and motivation mediate the relationship between organizational climate and safety performance. 4. Materials and methods 4.1. Sample The sample consisted of 525 employees from 32 work groups in a large Australian hospital (response rate 56%). The participants' mean age was 40.1 years (SD 11.0) and 89% were female. 4.2. Measures 4.2.1. Organizational climate Employees' perceptions about seven dierent aspects of their work environment (appraisal and recognition, goal congruency, role clarity, supportive leadership, participative decision making, professional growth, professional interaction) were assessed using 35 items from Hart et al. (1996b)'s Organizational Climate Scale. This scale is based on the components of the School Organizational Health Questionnaire (Hart et al., 2000) that were designed to assess the organizational factors that are common to most organizations. This questionnaire has been used to assess general organizational climate in a range of private and public sector organizations (e.g. Hart et al., 1997), and has been shown to have good discriminant validity from other related constructs such as organizational stressors (Hart et al., 1996a). Employees
A. Neal et al. / Safety Science 34 (2000) 99±109
103
Fig. 1. Hypothesized relationships among constructs.
were asked to rate the extent to which each item (e.g. ``My work objectives are always well de®ned'', ``There is good communication between groups in this workplace'', and ``This workplace has a clearly stated set of goals and objectives'') described their workplace on a ®ve-point scale ranging from ``strongly disagree'' to ``strongly agree''. Con®rmatory factor analyses showed that the seven separate dimensions could be aggregated at a second-order level to provide an overall index of general organizational climate (a=0.94). 4.2.2. Safety climate Sixteen items assessed perceptions of safety within a hospital setting, including management values, communication, training, and safety systems (a=0.93). Example items `Management is concerned for the safety of employees'', and ``There is open communication about safety issues within this workplace''. Employees responded on a ®ve-point scale ranging from ``Strongly Agree'' (1) to ``Strongly Disagree'' (5). 4.2.3. Determinants of safety performance Four items assessed knowledge about safety practices and procedures (a=0.90). Example items are ``I know how to perform my job in a safe manner'', and ``I know how to maintain or improve workplace health and safety''. Four items assessed individual motivation to perform safety-related activities procedures (a=0.93). Example item are ``I believe that workplace health and safety is an important issue'', and ``I feel that it is important to maintain safety at all times''. 4.2.4. Components of safety performance Four items assessed compliance with safety procedures (a=0.94). Example items are ``I use the correct safety procedures for carrying out my job'', and ``I ensure the highest levels of safety when I carry out my job''. Four items assessed the extent to
104
A. Neal et al. / Safety Science 34 (2000) 99±109
Table 1 Correlations among all measures in the study
1. General climate 2. Safety climate 3. Safety knowledge 4. Safety motivation 5. Safety compliance 6. Safety participation
M
SD
1
2
3
4
5
6
3.20 3.75 4.20 4.55 4.45 3.89
0.68 0.76 0.68 0.60 0.65 0.86
0.94 0.52 0.20 0.21 0.23 0.19
0.93 0.52 0.40 0.42 0.47
0.90 0.65 0.68 0.55
0.93 0.75 0.53
0.94 0.54
0.89
which individuals participated in safety-related activities (a=0.89). Example item are ``I voluntarily carry out tasks or activities that help to improve workplace safety'', and ``I help my coworkers when they are working under risky or hazardous conditions''. Means, standard deviations, coecients alpha and correlations among all the scales are reported in Table 1. 5. Results Hypotheses 1±7 were investigated using structural equation modeling (SEM) procedures. Scale scores were used to assess each construct depicted in Fig. 1 and error variances were estimated by multiplying the scale variance by one minus the scale reliability. Goodness of ®t of the various models was judged in terms of Joreskog and Sorbom's (1989) Goodness of Fit Index (GFT), the Non-normed Fit Index (NNFI; Bentler and Bonnett, 1980) and the Comparative Fit Index (CFI; Bentler, 1990). The CFI is recommended as the best approximation of the population value for a single model (Medsker et al., 1994). The NNFI is recommended as an index that takes account of the parsimony of the estimated parameters (Medsker et al., 1994). The results of this analysis are presented in Table 2. First, the hypothetical model depicted in Fig. 1 was estimated. Next, a partially saturated model was estimated with direct paths from organizational climate to compliance and participation. This model did not provide a signi®cantly better ®t than the hypothetical model Table 2 Comparison of alternative modelsa
Hypothesized model Direct paths from organizational climate to performance Direct paths from safety climate to performance Final model
df
w2
GFI
SRMR
CFI
NNFI
6 4 4 5
34.05 31.37 10.58 11.86
0.98 0.98 0.99 0.99
0.02 0.03 0.02 0.02
0.98 0.98 0.99 0.99
0.95 0.92 0.98 0.98
a GFI, Goodnes of Fit Index; SRMR, Square Root Mean Residual; CFI, Comparative Fit Index; NNFI, Non-normed Fit Index.
A. Neal et al. / Safety Science 34 (2000) 99±109
105
[w2(2)=2.68, P>0.05]. A model was then estimated with direct paths from safety climate to compliance and participation. This model did provide a signi®cantly better ®t than the hypothetical model [w2(2)=23.47, P<0.001] indicating that the hypothetical model did not contain sucient paths to account for all of the covariation between safety climate and the two safety performance measures. A ®nal model was estimated after inspecting the path coecients obtained from the second alternative model. The path from safety climate to safety participation was statistically signi®cant in this model. The ®nal model, comprising the hypothetical model plus this additional path, was not statistically dierent from the second partially saturated model [w2(2)=1.28, P>0.05]. The standardized path coecients from the ®nal model are presented in Fig. 2. In summary, the results supported key components of the proposed model. Hypothesis 1 predicted that organizational climate would predict safety climate. This hypothesis was supported. Hypotheses 2 and 4 proposed that knowledge and motivation would predict compliance and participation. Both these hypotheses were supported. Hypothesis 3 predicted that the relationship between knowledge and compliance would be stronger than the relationship between knowledge and participation. This hypothesis was supported. Hypothesis 5 predicted that the relationship between motivation and participation would be stronger than the relationship between motivation and compliance. This hypothesis was not supported. In fact, the relationship between motivation and compliance was stronger than the relationship between motivation and participation. Hypothesis 6 predicted that safety climate would in¯uence both knowledge and motivation. This hypothesis was supported. Hypothesis 7 predicted that knowledge and motivation would mediate the relationship between safety climate and safety performance. This hypothesis was partially supported. Contrary to expectations, there was a signi®cant direct path between safety climate and participation. The ®nal hypothesis predicted that safety climate, knowledge, and motivation would mediate the relationship between organizational climate and safety performance. This hypothesis was also supported.
Fig. 2. Final model from structural equation modeling (SEM) analysis.
106
A. Neal et al. / Safety Science 34 (2000) 99±109
6. Discussion This study developed a model to explain the in¯uence of organizational climate and safety climate on individual safety behavior. The model is important because it provides a link between the organizational environment and speci®c individual behaviors related to safety. Notwithstanding the limitations noted below, the results of this study supported the role of safety climate as a predictor of the determinants and components of safety performance. The results also supported safety climate as a mediator of the impact of general organizational climate on safety-related outcomes. The ®ndings that motivation exerted a weaker eect on participation than upon compliance, and that safety climate exerted a direct eect on participation were unexpected. One explanation for these ®ndings may be that the motivation scale did not capture all aspects of motivation relating to safety performance. The motivation items assessed the general value that employees placed on workplace health and safety, and did not assess the value that individuals place on participation in safety activities within the workplace. If the motivation scale did not capture all aspects of motivation relating to participation, then this could explain why the relationship between motivation and participation was weaker than expected, and why there was a direct path from safety climate to participation. A more appropriate measure of motivation might be to ask individuals how much eort they typically allocate to a range of tasks, including safety compliance and participation, as well as other aspects of the job (e.g. production). This is likely also to help overcome the problem of social desirability in¯uencing responses to the motivation items. Responses on the motivation scale were close to ceiling (M=4.55, SD=0.60), which restricted the ability of the motivation scale to predict performance outcomes. 6.1. Implications The results of the study have a number of implications for both research and practice. First, the study has demonstrated that safety climate is related to general organizational climate. Evaluations of safety climate, therefore, appear to be made within the context of the general organizational climate. This ®nding suggests that interventions designed to improve general organizational climate may have a positive impact on safety climate. This ®nding also suggests that interventions speci®cally aimed at improving safety climate (e.g. providing training, emphasizing the importance of safety) will be more eective when they are carried out within the context of a positive general organizational climate. Second, the study provides valuable evidence regarding the relative impact of general and speci®c forms of climate in organizations. Recent work in areas such as service climate (Schneider et al., 1992) as well as safety climate, suggest that it is useful to identify forms of climate that are conceptually congruent with important organizational outcomes. The results suggest that a speci®c climate for safety is more strongly related to safety performance than the general climate of the organization. General organizational climate did not contribute to performance, once the eects of safety climate had been partialled out. The results encourage a focus on
A. Neal et al. / Safety Science 34 (2000) 99±109
107
speci®c forms of organizational climate when speci®c organizational outcomes, such as safety, are of interest. Third, the study illustrates the eects that safety climate has on individual knowledge, motivation, compliance and participation. The data suggest that if improvements in safety climate are to have any impact on safety performance, then they must ®rst produce changes in knowledge or employee motivation. Traditionally, research has emphasized the motivational properties of climate (e.g. Brown and Leigh, 1996). The current ®ndings are important, because they suggest that safety climate can have independent eects on knowledge and motivation, and that both factors are important determinants of safety behaviors. This information can be used in the design of interventions, to ensure that they target both mediators. Interventions aimed solely at improving safety motivation (e.g. bonus and incentive schemes) are unlikely to be as eective as interventions that target both knowledge and motivation. Furthermore, this information can also be used to guide the development of measurement systems to evaluate the eectiveness of safety management practices. Incorporating assessments of knowledge, motivation, compliance, and participation into safety monitoring systems will provide a more complete assessment, not only of the eectiveness of safety practices, but also of their operation. 6.2. Limitations of the study The use of self-report measures to assess all dimensions of the safety climate model is a clear limitation of the current study. Estimates of the relationships amongst the measures may, therefore, be confounded by common method variance. Objective measurement of safety behavior and safety outcomes, using a new sample, is necessary to validate the impact of safety perceptions on workplace outcomes. A further limitation of the study was the cross-sectional measurement. It was not possible to test the causal relationships proposed in the theoretical framework, and longitudinal assessment would provide further validation of speci®c relationships. Moreover, the use of structural equation analyses, in itself, cannot provide evidence about causation. The use of structural equation analyses, however, enabled us to demonstrate discriminant validity among the measures used in this study, and enabled us to test a series of hypotheses that were consistent with a causal theory. A ®nal limitation was the absence of measures of safety skill. The inclusion of safety skill would provide a more comprehensive test of the theoretical model, and may account for the unexpected relationship that was found between safety climate and safety participation. These limitations are being addressed in research currently underway. 7. Conclusion Notwithstanding the limitations of this study, the results provide strong empirical support for the proposed theoretical model. The study demonstrates that general organizational climate can in¯uence perceptions of safety climate, and that these
108
A. Neal et al. / Safety Science 34 (2000) 99±109
perceptions of safety climate in¯uence safety performance through their eects on knowledge and motivation. These ®ndings provide valuable guidance for researchers and practitioners trying to identify the mechanisms by which they can improve safety in the workplace. Acknowledgements This research was supported by an Australian Research Council Large Grant (Grant No. A79601474). References Bentler, P.M., 1990. Comparative ®t indexes in structural models. Psychological Bulletin 107, 238±246. Bentler, P.M., Bonnett, D.G., 1980. Signi®cance tests and goodness of ®t in the analysis of covariance structures. Psychological Bulletin 88, 588±606. Borman, W.C., Motowidlo, S.J., 1993. Expanding the criterion domain to include elements of contextual performance. In: Schmitt, N., Borman, W.C. and Associates, Personnel Selection in Organizations, Jossey-Bass, San Francisco, pp. 71±98. Brown, R.L., Holmes, H., 1986. The use of a factor-analytic procedure for assessing the validity of an employee safety climate model. Accident Analysis and Prevention 18, 455±470. Brown, S.P., Leigh, T.W., 1996. A new look at psychological climate and its relationship to job involvement, eort, and performance. Journal of Applied Psychology 81, 358±368. Campbell, J.P., McCloy, R.A., Oppler, S.H., Sager, C.E., 1993. A theory of performance. In: Schmitt, J., Borman, W.C. Associates. Personnel Selection in Organizations. Jossey-Bass, San Francisco, CA, pp. 35±69. Dedobbeleer, N., Beland, F., 1991. A safety climate measure for construction sites. Journal of Safety Research 22, 97±103. DeJoy, D.M., 1994. Managing safety in the workplace: An attribution theory analysis and model. Journal of Safety Research 25, 3±17. Grin, M.A., 1996. Extra-role Behaviors in Organisations: The Mediation of Work and Individual Factors by Organisational Commitment. Paper presented at the 11th Annual Conference of the Society for Industrial and Organisational Psychology, San Diego, CA. Grin, M.A., Mathieu, J.E., 1997. Modeling organizational processes across hierarchical levels: climate, leadership, and group process in work groups. Journal of Organizational Behavior 18, 731±744. Grin, M.A., Tesluk, P.E., Jacobs, R.R., 1995. The eect of bargaining cycles on work-related attitudes: evidence for threat-rigidity eects. Academy of Management Journal 38, 1709±1724. Hart, P.M., Wearing, A.J., Grin, M.A., 1996a. Integrating Personal and Organisational Factors into a Coherent Model of Occupational Well-being. Paper presented at the 21st Annual Meeting of the Academy of Management, Cincinnati, OH. Hart, P.M., Grin, M.A., Wearing, A.J., Cooper, C.L., 1996b. QPASS: Manual for the Queensland Public Agency Sta Survey. Public Sector Management Commission, Brisbane. Hart, P.M., Wearing, A.J. Leipens, I., Grin, M.A., 1997. Linking Organizational Climate to Employee Well-Being and Performance: A Covariance Structure Approach. Paper presented at the 12th Annual Conference of the Society for Industrial and Organisational Psychology, St Louis, MO. Hart, P.M., Wearing, A.J., Conn, M., Carter, N.L., Dingle, R., 2000. Development of the School Organisational Health Questionnaire: A measure for assessing teacher morale and school organisational climate. British Journal of Educational Psychology (in press). Hesketh, B., Neal, A. (1999). Technology and performance. In: Ilgen, D, Pukalos, E., (Eds.) The Changing Nature of Work Performance: Implications for Stang, Motivation, and Development. Jossey Bass, San Francisco, CA.
A. Neal et al. / Safety Science 34 (2000) 99±109
109
Hofmann, D.A., Stetzer, A., 1996. A cross-level investigation of factors in¯uencing unsafe behaviors and accidents Personnel. Psychology 49, 307±339. James, L.A., James, L.R., 1989. Integrating work environment perceptions: Explorations into the measurement of meaning. Journal of Applied Psychology 74, 739±751. James, L.R., McIntyre, M.D., 1996. Perceptions of organizational climate. In: Murphy, K. (Ed.), Individual Dierences and Behavior in Organizations. Jossey-Bass, San Francisco, CA, pp. 416±450. James, L.R., James, L.A., Ashe, D.K., 1990. The meaning of organizations: the role of cognition and values. In: Schneider, B. (Ed.), Organizational Climate and Culture. Jossey-Bass, San Francisco, pp. 40±84. Joreskog, K.G., Sorbom, D., 1989. LISREL VII: User's Reference and Guide. Scienti®c Software, Mooresville, IN. Medsker, G.J., Williams, L.J., Holahan, P.J., 1994. A review of current practices for evaluating causal models in organizational behavior and human resources management research. Journal of Management 20, 439±464. Michela, J.L., Lukaszewski, M.P., Allegrante, J.P., 1995. Organizational climate and work stress: a general framework applied to inner-city schoolteachers. In: Sauter, S.L., Murphy, S.L.R. (Eds.), Organizational Risk Factors for Job Stress. American Psychological Association, Washington DC, pp. 61±80. Morrison, D.L., Upton, D.M., Cordery, J., 1997. Organizational Climate and Skill Utilization. Paper presented to the 12th Annual Conference of the Society for Industrial and Organizational Psychology, St. Louis, MI. Neal, A. & Grin, M.A., 1997. Perceptions of Safety at Work: Developing a Model to Link Organizational Safety Climate and Individual Behavior. Paper presented to the 12th Annual Conference of the Society for Industrial and Organizational Psychology, St. Louis, MO. Niskanen, T., 1994. Assessing the safety environment in the work organization of road maintenance jobs. Accident Analysis and Prevention 26, 27±39. Reason, J. Human Error. Cambridge University Press, Cambridge. Schneider, B., Wheeler, J.K., Cox, J.F., 1992. A passion for service: using content analysis to explicate service climate themes. Journal of Applied Psychology 77, 705±716. Wagner, R.K., Sternberg, R.J., 1985. Practical intelligence in real-world pursuits: The role of tacit knowledge. Journal of Personality and Social Psychology 49, 436±458. Zohar, D., 1980. Safety climate in industrial organizations: theoretical and applied implications. Journal of Applied Psychology 12, 78±85.
www.elsevier.com/locate/ssci
The impact of organizational climate on safety climate and individual behavior A. Neal a,*, M.A. Grin b, P.M. Hart c a
ARC Key Centre for Human Factors and Applied Cognitive Psychology, and School of Psychology, The University of Queensland, St. Lucia 4072, Australia b School of Management, Queensland University of Technology, Brisbane 4000, Australia c Department of Psychology, University of Melbourne, Melbourne 3010, Australia
Abstract Relatively little previous research has investigated the meechanisms by which safety climate aects safety behavior. The current study examined the eects of general organizational climate on safety climate and safety performance. As expected, general organizational climate exerted a signi®cant impact on safety climate, and safety climate in turn was related to selfreports of compliance with safety regulations and procedures as well as participation in safety-related activities within the workplace. The eect of general organizational climate on safety performance was mediated by safety climate, while the eect of safety climate on safety performance was partially mediated by safety knowledge and motivation. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Organizational climate; Safety climate; Safety performance
1. Introduction In recent years, there has been a shift in emphasis within the safety literature, away from individual level factors that might be responsible for accidents and incidents, such as error or non-compliance with safety procedures, towards organizational factors, such as safety climate (eg. Reason, 1990). A key assumption of much of this literature is that the relationship between safety climate and system safety is at least partially mediated by individual safety behavior. However, surprisingly little is known about the mechanisms by which safety climate in¯uences safety behaviors of individuals in organizations. Furthermore, relatively little is known about the * Corresponding author. Tel.: +61-7-3365-6372; fax: +61-7-3365-4466. E-mail address: [email protected] (A. Neal). 0925-7535/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0925-7535(00)00008-4
100
A. Neal et al. / Safety Science 34 (2000) 99±109
factors that in¯uence safety climate. The current paper examines the impact of general organizational climate on safety climate, and the impact of safety climate on the knowledge, motivation and performance of individuals in organizations. 2. Organizational climate and safety climate Organizational climate is a multidimensional construct that encompasses a wide range of individual evaluations of the work environment (James and James, 1989). These evaluations may refer to general dimensions of the environment such as leadership, roles, and communication (James and McIntyre, 1996) or to speci®c dimensions such as the climate for safety or the climate for customer service. General perceptions of the organizational context can in¯uence interactions among individuals (Grin and Mathieu, 1997), attitudes toward organizational rewards (Grin et al., 1995; Grin, 1996) and aective responses to the work environment (Michela et al., 1995; Hart et al., 1996a). Perceptions of the general organizational climate develop as individuals attribute meaning to their organizational context based on the signi®cance of the environment for individual values (James et al., 1990). Organizational climate, therefore, is thought to exert a strong impact on individual motivation to achieve work outcomes (Brown and Leigh, 1996). General organizational climate has also been found to in¯uence knowledge and skills by increasing participation in activities such as training (Morrison et al., 1997). Safety climate is a speci®c form of organizational climate, which describes individual perceptions of the value of safety in the work environment. A range of factors has been identi®ed as being important components of safety climate. These factors include: management values (e.g. management concern for employee well-being), management and organizational practices (e.g. adequacy of training, provision of safety equipment, quality of safety management systems), communication, and employee involvement in workplace health and safety. A range of studies have demonstrated that these factors predict safety-related outcomes, such as accidents and incidents (e.g., Zohar, 1980; Brown and Holmes, 1986; Dedobbeleer and Beland, 1991; DeJoy, 1994; Niskanen, 1994; Hofmann and Stetzer, 1996). No studies have investigated links between general organizational climate and speci®c safety climate. It is proposed that general organizational climate provides a context in which speci®c evaluations of the importance of safety are made. This means that the general organizational climate should predict speci®c safety climate (see Hypothesis 1). For example, if employees perceive that there is open communication in the organization, then they may also perceive that communication about safety is valued in the organization. Similarly, if employees perceive that the organization is supportive of their general welfare and well-being, they will be more likely to perceive that the organization values the safety of employees. These perceptions of the speci®c safety climate of the organization, in turn, are hypothesized to in¯uence safety behavior. The following section describes a theory of safety performance that can be used to derive predictions regarding the impact of safety climate on safety behavior.
A. Neal et al. / Safety Science 34 (2000) 99±109
101
3. Safety performance Neal and Grin (1997) proposed a model of safety performance based on current theories of job performance (Borman and Motowidlo, 1993; Campbell et al., 1993). This model distinguishes between performance components, determinants of performance, and performance antecedents. The components of performance represent the major dimensions of task-relevant behaviors involved in a given job. The model incorporates two dimensions of safety performance: compliance and participation. Safety compliance involves adhering to safety procedures and carrying out work in a safe manner. Safety participation involves helping coworkers, promoting the safety program within the workplace, demonstrating initiative, and putting eort into improving safety in the workplace. The determinants of safety performance represent the factors directly responsible for individual dierences in compliance and participation. Campbell et al. (1993) argue that there are only three determinants of individual dierence in performance: knowledge, skill, and motivation. Several authors have criticized this assumption, arguing that there may be other determinants of performance. Hesketh and Neal (1999), for example, argue that situational factors can sometimes produce invidivual dierences in performance. However, available evidence does suggest that knowledge, skill and motivation are important determinants of individual dierences in performance across a wide range of contexts. Safety behavior, therefore, should be determined by knowledge and skills necessary for particular behaviors and by the motivation of individuals to perform the behaviors. Furthermore, knowledge, skill, and motivation should have dierential eects on the dierent components of performance. It is expected that safety knowledge and skill should have a stronger relationship with compliance than with participation (see Hypotheses 2 and 3), while safety motivation should have a stronger relationship with participation than with compliance (see Hypotheses 4 and 5). An individual must understand how to perform work safely and have the skill to be able to do it in order to comply with safety procedures. However, safety knowledge and skill are likely to be less important for participatory activities, since these activities require more generic forms of knowledge and skill (termed ``tacit knowledge'' by Wagner and Sternberg, 1985). Motivation is likely to be more important for participation than for compliance, because participatory activities are frequently voluntary, whereas compliance is generally mandated. Antecedents of performance include individual-level factors such as ability, experience and personality, as well as group and organizational factors, such as leadership, group norms, and climate. General organizational climate and safety climate, therefore, are classi®ed as antecedents of safety performance. A key assumption of the Campbell et al. (1993) model is that the determinants of performance (knowledge, skill and motivation) must mediate the relationship between the antecedents and components of performance. The theory, therefore, predicts that knowledge, skill, motivation should mediate the relationship between safety climate and safety performance (see Hypotheses 6 and 7). Moreover, this suggests that sefety climate and safety behaviors mediate the relationship between general organizational climate and safety perofmance (see Hypothesis 8).
102
A. Neal et al. / Safety Science 34 (2000) 99±109
Neal and Grin (1997) tested the model of safety performance using an archival data set. As expected, compliance and participation were found to form separate factors, and individual safety knowledge was found to mediate at least some of the relationships between safety climate and safety performance. Unfortunately, safety motivation was not assessed in the archival data set, and several dimensions of safety climate were found to have direct eects on safety performance. The current study assessed safety climate and performance in a hospital setting, and included measures of both safety knowledge and safety motivation. Measures of skill were not included. The theoretical model is shown in Fig. 1. The following hypotheses were tested in the present study: H1. Organizational climate predicts safety climate H2. Knolwedge predicts both compliance and participation H3. The relationship between knolwedge and compliance is stronger than the relationship between knowledge and participation H4. Motivation predicts both compliance and participation. H5. The relationship between motivation and participation is stronger than the relationship between motivation and compliance. H6. Safety climate predicts both knowledge and motivation. H7. Knowledge and motivation mediate the relationship between safety climate and safety performance. H8. Safety climate, knowledge, and motivation mediate the relationship between organizational climate and safety performance. 4. Materials and methods 4.1. Sample The sample consisted of 525 employees from 32 work groups in a large Australian hospital (response rate 56%). The participants' mean age was 40.1 years (SD 11.0) and 89% were female. 4.2. Measures 4.2.1. Organizational climate Employees' perceptions about seven dierent aspects of their work environment (appraisal and recognition, goal congruency, role clarity, supportive leadership, participative decision making, professional growth, professional interaction) were assessed using 35 items from Hart et al. (1996b)'s Organizational Climate Scale. This scale is based on the components of the School Organizational Health Questionnaire (Hart et al., 2000) that were designed to assess the organizational factors that are common to most organizations. This questionnaire has been used to assess general organizational climate in a range of private and public sector organizations (e.g. Hart et al., 1997), and has been shown to have good discriminant validity from other related constructs such as organizational stressors (Hart et al., 1996a). Employees
A. Neal et al. / Safety Science 34 (2000) 99±109
103
Fig. 1. Hypothesized relationships among constructs.
were asked to rate the extent to which each item (e.g. ``My work objectives are always well de®ned'', ``There is good communication between groups in this workplace'', and ``This workplace has a clearly stated set of goals and objectives'') described their workplace on a ®ve-point scale ranging from ``strongly disagree'' to ``strongly agree''. Con®rmatory factor analyses showed that the seven separate dimensions could be aggregated at a second-order level to provide an overall index of general organizational climate (a=0.94). 4.2.2. Safety climate Sixteen items assessed perceptions of safety within a hospital setting, including management values, communication, training, and safety systems (a=0.93). Example items `Management is concerned for the safety of employees'', and ``There is open communication about safety issues within this workplace''. Employees responded on a ®ve-point scale ranging from ``Strongly Agree'' (1) to ``Strongly Disagree'' (5). 4.2.3. Determinants of safety performance Four items assessed knowledge about safety practices and procedures (a=0.90). Example items are ``I know how to perform my job in a safe manner'', and ``I know how to maintain or improve workplace health and safety''. Four items assessed individual motivation to perform safety-related activities procedures (a=0.93). Example item are ``I believe that workplace health and safety is an important issue'', and ``I feel that it is important to maintain safety at all times''. 4.2.4. Components of safety performance Four items assessed compliance with safety procedures (a=0.94). Example items are ``I use the correct safety procedures for carrying out my job'', and ``I ensure the highest levels of safety when I carry out my job''. Four items assessed the extent to
104
A. Neal et al. / Safety Science 34 (2000) 99±109
Table 1 Correlations among all measures in the study
1. General climate 2. Safety climate 3. Safety knowledge 4. Safety motivation 5. Safety compliance 6. Safety participation
M
SD
1
2
3
4
5
6
3.20 3.75 4.20 4.55 4.45 3.89
0.68 0.76 0.68 0.60 0.65 0.86
0.94 0.52 0.20 0.21 0.23 0.19
0.93 0.52 0.40 0.42 0.47
0.90 0.65 0.68 0.55
0.93 0.75 0.53
0.94 0.54
0.89
which individuals participated in safety-related activities (a=0.89). Example item are ``I voluntarily carry out tasks or activities that help to improve workplace safety'', and ``I help my coworkers when they are working under risky or hazardous conditions''. Means, standard deviations, coecients alpha and correlations among all the scales are reported in Table 1. 5. Results Hypotheses 1±7 were investigated using structural equation modeling (SEM) procedures. Scale scores were used to assess each construct depicted in Fig. 1 and error variances were estimated by multiplying the scale variance by one minus the scale reliability. Goodness of ®t of the various models was judged in terms of Joreskog and Sorbom's (1989) Goodness of Fit Index (GFT), the Non-normed Fit Index (NNFI; Bentler and Bonnett, 1980) and the Comparative Fit Index (CFI; Bentler, 1990). The CFI is recommended as the best approximation of the population value for a single model (Medsker et al., 1994). The NNFI is recommended as an index that takes account of the parsimony of the estimated parameters (Medsker et al., 1994). The results of this analysis are presented in Table 2. First, the hypothetical model depicted in Fig. 1 was estimated. Next, a partially saturated model was estimated with direct paths from organizational climate to compliance and participation. This model did not provide a signi®cantly better ®t than the hypothetical model Table 2 Comparison of alternative modelsa
Hypothesized model Direct paths from organizational climate to performance Direct paths from safety climate to performance Final model
df
w2
GFI
SRMR
CFI
NNFI
6 4 4 5
34.05 31.37 10.58 11.86
0.98 0.98 0.99 0.99
0.02 0.03 0.02 0.02
0.98 0.98 0.99 0.99
0.95 0.92 0.98 0.98
a GFI, Goodnes of Fit Index; SRMR, Square Root Mean Residual; CFI, Comparative Fit Index; NNFI, Non-normed Fit Index.
A. Neal et al. / Safety Science 34 (2000) 99±109
105
[w2(2)=2.68, P>0.05]. A model was then estimated with direct paths from safety climate to compliance and participation. This model did provide a signi®cantly better ®t than the hypothetical model [w2(2)=23.47, P<0.001] indicating that the hypothetical model did not contain sucient paths to account for all of the covariation between safety climate and the two safety performance measures. A ®nal model was estimated after inspecting the path coecients obtained from the second alternative model. The path from safety climate to safety participation was statistically signi®cant in this model. The ®nal model, comprising the hypothetical model plus this additional path, was not statistically dierent from the second partially saturated model [w2(2)=1.28, P>0.05]. The standardized path coecients from the ®nal model are presented in Fig. 2. In summary, the results supported key components of the proposed model. Hypothesis 1 predicted that organizational climate would predict safety climate. This hypothesis was supported. Hypotheses 2 and 4 proposed that knowledge and motivation would predict compliance and participation. Both these hypotheses were supported. Hypothesis 3 predicted that the relationship between knowledge and compliance would be stronger than the relationship between knowledge and participation. This hypothesis was supported. Hypothesis 5 predicted that the relationship between motivation and participation would be stronger than the relationship between motivation and compliance. This hypothesis was not supported. In fact, the relationship between motivation and compliance was stronger than the relationship between motivation and participation. Hypothesis 6 predicted that safety climate would in¯uence both knowledge and motivation. This hypothesis was supported. Hypothesis 7 predicted that knowledge and motivation would mediate the relationship between safety climate and safety performance. This hypothesis was partially supported. Contrary to expectations, there was a signi®cant direct path between safety climate and participation. The ®nal hypothesis predicted that safety climate, knowledge, and motivation would mediate the relationship between organizational climate and safety performance. This hypothesis was also supported.
Fig. 2. Final model from structural equation modeling (SEM) analysis.
106
A. Neal et al. / Safety Science 34 (2000) 99±109
6. Discussion This study developed a model to explain the in¯uence of organizational climate and safety climate on individual safety behavior. The model is important because it provides a link between the organizational environment and speci®c individual behaviors related to safety. Notwithstanding the limitations noted below, the results of this study supported the role of safety climate as a predictor of the determinants and components of safety performance. The results also supported safety climate as a mediator of the impact of general organizational climate on safety-related outcomes. The ®ndings that motivation exerted a weaker eect on participation than upon compliance, and that safety climate exerted a direct eect on participation were unexpected. One explanation for these ®ndings may be that the motivation scale did not capture all aspects of motivation relating to safety performance. The motivation items assessed the general value that employees placed on workplace health and safety, and did not assess the value that individuals place on participation in safety activities within the workplace. If the motivation scale did not capture all aspects of motivation relating to participation, then this could explain why the relationship between motivation and participation was weaker than expected, and why there was a direct path from safety climate to participation. A more appropriate measure of motivation might be to ask individuals how much eort they typically allocate to a range of tasks, including safety compliance and participation, as well as other aspects of the job (e.g. production). This is likely also to help overcome the problem of social desirability in¯uencing responses to the motivation items. Responses on the motivation scale were close to ceiling (M=4.55, SD=0.60), which restricted the ability of the motivation scale to predict performance outcomes. 6.1. Implications The results of the study have a number of implications for both research and practice. First, the study has demonstrated that safety climate is related to general organizational climate. Evaluations of safety climate, therefore, appear to be made within the context of the general organizational climate. This ®nding suggests that interventions designed to improve general organizational climate may have a positive impact on safety climate. This ®nding also suggests that interventions speci®cally aimed at improving safety climate (e.g. providing training, emphasizing the importance of safety) will be more eective when they are carried out within the context of a positive general organizational climate. Second, the study provides valuable evidence regarding the relative impact of general and speci®c forms of climate in organizations. Recent work in areas such as service climate (Schneider et al., 1992) as well as safety climate, suggest that it is useful to identify forms of climate that are conceptually congruent with important organizational outcomes. The results suggest that a speci®c climate for safety is more strongly related to safety performance than the general climate of the organization. General organizational climate did not contribute to performance, once the eects of safety climate had been partialled out. The results encourage a focus on
A. Neal et al. / Safety Science 34 (2000) 99±109
107
speci®c forms of organizational climate when speci®c organizational outcomes, such as safety, are of interest. Third, the study illustrates the eects that safety climate has on individual knowledge, motivation, compliance and participation. The data suggest that if improvements in safety climate are to have any impact on safety performance, then they must ®rst produce changes in knowledge or employee motivation. Traditionally, research has emphasized the motivational properties of climate (e.g. Brown and Leigh, 1996). The current ®ndings are important, because they suggest that safety climate can have independent eects on knowledge and motivation, and that both factors are important determinants of safety behaviors. This information can be used in the design of interventions, to ensure that they target both mediators. Interventions aimed solely at improving safety motivation (e.g. bonus and incentive schemes) are unlikely to be as eective as interventions that target both knowledge and motivation. Furthermore, this information can also be used to guide the development of measurement systems to evaluate the eectiveness of safety management practices. Incorporating assessments of knowledge, motivation, compliance, and participation into safety monitoring systems will provide a more complete assessment, not only of the eectiveness of safety practices, but also of their operation. 6.2. Limitations of the study The use of self-report measures to assess all dimensions of the safety climate model is a clear limitation of the current study. Estimates of the relationships amongst the measures may, therefore, be confounded by common method variance. Objective measurement of safety behavior and safety outcomes, using a new sample, is necessary to validate the impact of safety perceptions on workplace outcomes. A further limitation of the study was the cross-sectional measurement. It was not possible to test the causal relationships proposed in the theoretical framework, and longitudinal assessment would provide further validation of speci®c relationships. Moreover, the use of structural equation analyses, in itself, cannot provide evidence about causation. The use of structural equation analyses, however, enabled us to demonstrate discriminant validity among the measures used in this study, and enabled us to test a series of hypotheses that were consistent with a causal theory. A ®nal limitation was the absence of measures of safety skill. The inclusion of safety skill would provide a more comprehensive test of the theoretical model, and may account for the unexpected relationship that was found between safety climate and safety participation. These limitations are being addressed in research currently underway. 7. Conclusion Notwithstanding the limitations of this study, the results provide strong empirical support for the proposed theoretical model. The study demonstrates that general organizational climate can in¯uence perceptions of safety climate, and that these
108
A. Neal et al. / Safety Science 34 (2000) 99±109
perceptions of safety climate in¯uence safety performance through their eects on knowledge and motivation. These ®ndings provide valuable guidance for researchers and practitioners trying to identify the mechanisms by which they can improve safety in the workplace. Acknowledgements This research was supported by an Australian Research Council Large Grant (Grant No. A79601474). References Bentler, P.M., 1990. Comparative ®t indexes in structural models. Psychological Bulletin 107, 238±246. Bentler, P.M., Bonnett, D.G., 1980. Signi®cance tests and goodness of ®t in the analysis of covariance structures. Psychological Bulletin 88, 588±606. Borman, W.C., Motowidlo, S.J., 1993. Expanding the criterion domain to include elements of contextual performance. In: Schmitt, N., Borman, W.C. and Associates, Personnel Selection in Organizations, Jossey-Bass, San Francisco, pp. 71±98. Brown, R.L., Holmes, H., 1986. The use of a factor-analytic procedure for assessing the validity of an employee safety climate model. Accident Analysis and Prevention 18, 455±470. Brown, S.P., Leigh, T.W., 1996. A new look at psychological climate and its relationship to job involvement, eort, and performance. Journal of Applied Psychology 81, 358±368. Campbell, J.P., McCloy, R.A., Oppler, S.H., Sager, C.E., 1993. A theory of performance. In: Schmitt, J., Borman, W.C. Associates. Personnel Selection in Organizations. Jossey-Bass, San Francisco, CA, pp. 35±69. Dedobbeleer, N., Beland, F., 1991. A safety climate measure for construction sites. Journal of Safety Research 22, 97±103. DeJoy, D.M., 1994. Managing safety in the workplace: An attribution theory analysis and model. Journal of Safety Research 25, 3±17. Grin, M.A., 1996. Extra-role Behaviors in Organisations: The Mediation of Work and Individual Factors by Organisational Commitment. Paper presented at the 11th Annual Conference of the Society for Industrial and Organisational Psychology, San Diego, CA. Grin, M.A., Mathieu, J.E., 1997. Modeling organizational processes across hierarchical levels: climate, leadership, and group process in work groups. Journal of Organizational Behavior 18, 731±744. Grin, M.A., Tesluk, P.E., Jacobs, R.R., 1995. The eect of bargaining cycles on work-related attitudes: evidence for threat-rigidity eects. Academy of Management Journal 38, 1709±1724. Hart, P.M., Wearing, A.J., Grin, M.A., 1996a. Integrating Personal and Organisational Factors into a Coherent Model of Occupational Well-being. Paper presented at the 21st Annual Meeting of the Academy of Management, Cincinnati, OH. Hart, P.M., Grin, M.A., Wearing, A.J., Cooper, C.L., 1996b. QPASS: Manual for the Queensland Public Agency Sta Survey. Public Sector Management Commission, Brisbane. Hart, P.M., Wearing, A.J. Leipens, I., Grin, M.A., 1997. Linking Organizational Climate to Employee Well-Being and Performance: A Covariance Structure Approach. Paper presented at the 12th Annual Conference of the Society for Industrial and Organisational Psychology, St Louis, MO. Hart, P.M., Wearing, A.J., Conn, M., Carter, N.L., Dingle, R., 2000. Development of the School Organisational Health Questionnaire: A measure for assessing teacher morale and school organisational climate. British Journal of Educational Psychology (in press). Hesketh, B., Neal, A. (1999). Technology and performance. In: Ilgen, D, Pukalos, E., (Eds.) The Changing Nature of Work Performance: Implications for Stang, Motivation, and Development. Jossey Bass, San Francisco, CA.
A. Neal et al. / Safety Science 34 (2000) 99±109
109
Hofmann, D.A., Stetzer, A., 1996. A cross-level investigation of factors in¯uencing unsafe behaviors and accidents Personnel. Psychology 49, 307±339. James, L.A., James, L.R., 1989. Integrating work environment perceptions: Explorations into the measurement of meaning. Journal of Applied Psychology 74, 739±751. James, L.R., McIntyre, M.D., 1996. Perceptions of organizational climate. In: Murphy, K. (Ed.), Individual Dierences and Behavior in Organizations. Jossey-Bass, San Francisco, CA, pp. 416±450. James, L.R., James, L.A., Ashe, D.K., 1990. The meaning of organizations: the role of cognition and values. In: Schneider, B. (Ed.), Organizational Climate and Culture. Jossey-Bass, San Francisco, pp. 40±84. Joreskog, K.G., Sorbom, D., 1989. LISREL VII: User's Reference and Guide. Scienti®c Software, Mooresville, IN. Medsker, G.J., Williams, L.J., Holahan, P.J., 1994. A review of current practices for evaluating causal models in organizational behavior and human resources management research. Journal of Management 20, 439±464. Michela, J.L., Lukaszewski, M.P., Allegrante, J.P., 1995. Organizational climate and work stress: a general framework applied to inner-city schoolteachers. In: Sauter, S.L., Murphy, S.L.R. (Eds.), Organizational Risk Factors for Job Stress. American Psychological Association, Washington DC, pp. 61±80. Morrison, D.L., Upton, D.M., Cordery, J., 1997. Organizational Climate and Skill Utilization. Paper presented to the 12th Annual Conference of the Society for Industrial and Organizational Psychology, St. Louis, MI. Neal, A. & Grin, M.A., 1997. Perceptions of Safety at Work: Developing a Model to Link Organizational Safety Climate and Individual Behavior. Paper presented to the 12th Annual Conference of the Society for Industrial and Organizational Psychology, St. Louis, MO. Niskanen, T., 1994. Assessing the safety environment in the work organization of road maintenance jobs. Accident Analysis and Prevention 26, 27±39. Reason, J. Human Error. Cambridge University Press, Cambridge. Schneider, B., Wheeler, J.K., Cox, J.F., 1992. A passion for service: using content analysis to explicate service climate themes. Journal of Applied Psychology 77, 705±716. Wagner, R.K., Sternberg, R.J., 1985. Practical intelligence in real-world pursuits: The role of tacit knowledge. Journal of Personality and Social Psychology 49, 436±458. Zohar, D., 1980. Safety climate in industrial organizations: theoretical and applied implications. Journal of Applied Psychology 12, 78±85.