Applied Ergonomics 51 (2015) 163e171

Contents lists available at ScienceDirect

Applied Ergonomics journal homepage: www.elsevier.com/locate/apergo

Job hindrances, job resources, and safety performance: The mediating role of job engagement* Zhenyu Yuan a, b, 1, Yongjuan Li a, *, Lois E. Tetrick c, 2 a

Key Laboratory of Behavioral Science, Institute of Psychology, Chinese Academy of Sciences, 4A Datun Rd, Chaoyang Dist, Beijing 100101, China University of Iowa, Department of Management & Organizations, W217 Pappajohn Business Building, Iowa City, IA 52242-1994, USA c George Mason University, Department of Psychology, David King Hall, Room 3066A, 4400 University Drive, MSN 3F5, Fairfax, VA 22030, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 January 2014 Accepted 30 April 2015 Available online 23 May 2015

Job engagement has received widespread attention in organizational research but has rarely been empirically investigated in the context of safety. In the present study, we examined the mediating role of job engagement in the relationships between job characteristics and safety performance using selfreported data collected at a coal mining company in China. Most of our study hypotheses were supported. Job engagement partially mediated the relationships between job resources and safety performance dimensions. Theoretical and practical implications and directions for future research are also discussed. © 2015 Elsevier Ltd and The Ergonomics Society. All rights reserved.

Keywords: Job characteristics Safety performance Job engagement

1. Introduction Organizations are paying close attention to employee engagement, which has been consistently linked to higher levels of job performance (Rich et al., 2010). Scholars have echoed this widespread interest by incorporating engagement into the positive organizational behavior (POB) movement, which advocates more focused research on positive psychological states, traits, and behaviors of employees (Bakker and Schaufeli, 2008; Luthans and Youssef, 2007). Job engagement has been defined as “a positive, fulfilling, work-related state of mind that is characterized by vigor, dedication, and absorption” (Schaufeli et al., 2002, p. 74). Research on engagement has recently been extended to a broad range of employee behaviors, including in-role performance (Rich et al., 2010), extra-role performance (Xanthopoulou et al., 2008), and personal initiative (Hakanen et al., 2008). Specifically, job engagement mediates the relations between job characteristics and a wide array of employee work behaviors (e.g.,

* This work was supported by National Natural Science Foundation of China (Grant Number: 71071149; 71371179) and Chinese Academy of Sciences (Grant Number: KJZD-EW-L04). * Corresponding author. Tel.: þ86 10 64858728; fax: þ86 10 64872070. E-mail addresses: [email protected] (Z. Yuan), [email protected] (Y. Li), [email protected] (L.E. Tetrick). 1 Tel.: þ1 317 372 8318; fax: þ1 317 274 6756. 2 Tel.: þ1 703 993 1372; fax: þ1 703 993 1359.

http://dx.doi.org/10.1016/j.apergo.2015.04.021 0003-6870/© 2015 Elsevier Ltd and The Ergonomics Society. All rights reserved.

Rich et al., 2010). Given that job design theory (Hackman and Oldham, 1976) and the socio-technical systems approach converge on the importance of job characteristics on employee performance (Grant et al., 2011; Holman et al., 2002; Rousseau, 1977), job engagement might be an important crux of employee safety behaviors. Indeed, safety scholars have hinted at job engagement in shaping safety behaviors and outcomes (e.g., Nahrgang et al., 2011). In the ergonomics literature, scholars have also called for integration of macro-ergonomics and safety research (Murphy et al., 2014). Specifically, individual attitudes (e.g., job engagement) are influenced by the interplay between work system design and workers' safety perceptions. In other words, job engagement may be considered as a proxy of employees' reactions to the “match” between the technical and the social systems. As such, although job engagement is an attitudinal-motivational construct from organizational behavior research, it has important theoretical bearings on applied ergonomics literature as well. Further, investigating job engagement might advance our understanding of how the interdependence of technical and social components of work system design can influence safety behavior via individual attitudes. Safety performance, as a separate domain of job performance (Burke et al., 2002), is closely tied to workplace safety and is a critical determinant of safety outcomes (Zohar, 2000). Similar to the distinction between task and contextual performance (Borman and Motowidlo, 1993), safety performance comprises two components, safety compliance and safety participation (Griffin and Neal,

164

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

2000). Parallel to task performance, safety compliance refers to the core safety activities that need to be carried out by individuals to maintain workplace safety, whereas safety participation taps into voluntary behaviors that help to maintain workplace safety, similar to contextual performance (Griffin and Neal, 2000). To extend the job engagement-job performance relationship one step further, engagement may have important implications for safety performance as well. However, the relationship between engagement and safety performance has received inadequate empirical attention in the literature. Some safety research examined engagement in a tangential way. For example, Hansez and Chmiel (2010) used positive occupational states to measure job engagement. Nahrgang et al. (2011) operationalized engagement as safety involvement, participation, and communication in their meta-analytic model. Given that engagement research has already suffered from construct proliferation (Macey and Schneider, 2008), it is crucial to follow the well-established conceptualization of engagement and explore its relationships with important job behaviors. As such, the first goal of the present study is to examine the relationship between job engagement and safety performance following the wellestablished conceptualization of engagement (Schaufeli et al., 2002). Job characteristics have been recognized as critical antecedents to job engagement and subsequent work behaviors (Crawford et al., 2010; Demerouti et al., 2001; Nahrgang et al., 2011). The present study will examine job characteristics as antecedents to job engagement. Specifically, we draw on the Job Demands-Resources model (JD-R; Crawford et al., 2010; Demerouti et al., 2001; Nahrgang et al., 2011) to examine job engagement as a mediator in the relationship between job characteristics and safety performance. Similar to its predecessor in job design theory (Hackman and Oldham, 1976), the JD-R model examines how various job aspects influence employee behavior through individual motivation and well-being. Although this theory was developed and heavily studied in occupational health psychology, it has much bearing on work design and ergonomics, in that it investigates the psychological effects of technical, environmental, and social job characteristics. According to the JD-R model, job demands are the physical, psychological, social, and/or organizational job aspects that require the input of physical and/or psychological effort and thus have physiological and/or psychological costs. Specifically, hindrance demands thwart personal development and goal attainment whereas challenge demands hold the potential of fostering learning and goal achievement, although they are still associated with psychological and/or physiological costs (Cavanaugh et al., 2000). Job resources refer to the physical, psychological, social, and/or organizational characteristics that can a) facilitate the achievement of work goals, b) address the negative impact of job demands, and/ or c) foster personal learning and development. Accumulating evidence suggests that these three types of job characteristics (i.e., hindrance demands, challenge demands, job resources) have distinctive implications for job engagement (Crawford et al., 2010). Within the context of workplace safety, job hindrances were shown to be an important antecedent to safety outcomes (Nahrgang et al., 2011) whereas the energizing role of job challenges did not receive empirical support (Yuan et al., 2014). Based on these considerations, we focus on job hindrances and job resources as antecedents to job engagement in the present study. Specifically, we include two types of job hindrances, job insecurity and role overload, and two types of job resources, coworker support and management commitment to safety. Since our study sample was from the coal mining industry, job insecurity and role overload are particularly salient to blue-collar employees working in this industry. Coworker support and management commitment to safety are also relevant since the importance of safety is usually

emphasized via management commitment and supportive behaviors among colleagues. Taken altogether, the overall goal of the present study is to examine the mediating role of engagement in the relationships between job characteristics and safety performance. In doing so, we aim to extend job engagement into workplace safety research. We examine job characteristics that are particularly salient to coal miners in an effort to inform potential managerial interventions. By looking at job characteristics and engagement, we attempt to highlight the importance of the motivational state of individuals in channeling the effect of job design features on safety behaviors. In the following sections, we will delineate the relationships between study variables drawing on the JD-R model. First we will propose the relationships between the two types of job characteristics (job hindrances and job resources) and job engagement. We will then develop our study hypotheses regarding the relationship between engagement and safety performance, followed by the mediating role of engagement in the relationships between job characteristics and safety performance. 1.1. Job hindrances and job engagement Depending on the nature of job characteristics in question, they can be associated with different employee outcomes (Crawford et al., 2010; Demerouti et al., 2001). According to the JD-R model, job hindrances tend to thwart personal growth and goal attainment and trigger negative cognitions and emotions (Crawford et al., 2010). Negative cognitions and emotions are associated with decreased levels of job engagement (Bledow et al., 2011) in that being fully engaged in one's work requires harnessing of oneself into role performance both cognitively and emotionally (Kahn, 1990). Therefore, we propose that individuals faced with job hindrances tend to adopt passive, emotion-focused coping styles characterized by lower levels of job engagement (Crawford et al., 2010). Specifically, job insecurity implies a high degree of uncertainty about one's employment status and can trigger negative outcomes including lowered well-being and negative emotions (Sverke et al., 2002). Existing studies support the negative relationship between job insecurity and job engagement (Bosman et al., 2005; De Cuyper and De Witte, 2005; Mauno et al., 2007). Role overload, the strongest individual-level predictor of injury (Hofmann and Stetzer, 1996; Zohar, 2000), is another job hindrance that implies a conflict between safety and other performance aspects (Zohar, 2002). Hence, role overload may be associated with psychological and physiological costs that detract from job engagement. Hypothesis 1a. Job insecurity will be negatively related to job engagement. Hypothesis 1b. Role overload will be negatively related to job engagement.

1.2. Job resources and job engagement According to the JD-R model, job resources can be either intrinsically motivating by fostering personal learning and growth, or extrinsically motivating by facilitating goal attainment (Demerouti et al., 2001; Schaufeli and Bakker, 2004). Given that engagement is an affective-motivational construct (Salanova et al., 2005; Schaufeli et al., 2002), job resources might thus be an important driver of job engagement due to its intrinsically and/or extrinsically motivational nature (Crawford et al., 2010; Demerouti et al., 2001). In other words, job resources will initiate a motivational process through which individual faced with job resources tend to have elevated levels of job engagement.

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

165

Specifically, coworker support not only facilitates goal attainment but also provides emotional support (Cohen and Wills, 1985). Coworker support has been consistently linked to increased job engagement (Crawford et al., 2010; Schaufeli and Bakker, 2004). Management commitment to safety, which refers to the extent to which managers are involved in and committed to promoting workplace safety, is also an important job resource (Nahrgang et al., 2011). Employees can derive a sense of control over their work environment and can access other resources when they perceive a high level of management commitment to safety (Chowdhury and Endres, 2010). Management commitment to safety can therefore be extrinsically motivating. Management commitment to safety can also be interpreted as a deep concern for workplace safety and employee well-being. Therefore, employees may feel intrinsically motivated to engage in their jobs upon the fulfillment of basic needs such as needs for safety and personal growth.

therefore be better able to attend to opportunities to promote workplace safety and deploy resources to help their coworkers and/ or the organization. A positive state at work also sensitizes individuals to look on recipients of help in a positive light (George, 1991). Past research has consistently established links between positive states and helping behaviors (Carlson et al., 1988) and prosocial behaviors at work (George, 1991). Workers experiencing job engagement are characterized by a positive and fulfilling state of mind at work and are thus more likely to engage in safety participation, a form of prosocial behaviors in the safety domain.

Hypothesis 2a. Coworker support will be positively related to job engagement.

Having specified the relationships between job characteristics and engagement, and the relationship between engagement and safety performance, we further propose that job engagement will mediate the relationships between job characteristics and safety performance. Although the pivotal effects of job characteristics on safety performance have long been recognized (Christian et al., 2009), the mediating role of job engagement has not received adequate scholarly attention. Existing studies did not test the role of engagement per se but used such constructs as positive occupational states (Hansez and Chmiel, 2010) and safety involvement (Nahrgang et al., 2011). Considering the accumulating evidence supportive of job engagement mediating the relationships between job characteristics and other work behaviors (e.g., task performance, Rich et al., 2010), it is reasonable to expect job engagement to act as a mediator in the relationships between job characteristics and safety performance. As a motivational construct, job engagement captures a unique aspect of human agency that responds to contextual influences (i.e., job characteristics) by affecting behaviors (Kahn, 1990; Rich et al., 2010). Research grounded on the JD-R model has consistently documented the mediating role of engagement in the relationships between job characteristics and job performance (Demerouti et al., 2001; Schaufeli and Bakker, 2004). In a similar vein, job characteristics may influence safety performance through job engagement. Past studies suggest that job characteristics can influence safety behaviors directly and indirectly through safety motivation (Neal et al., 2000) and positive occupational states (Hansez and Chmiel, 2010). In light of these findings, we propose that job engagement will partially mediate the relationships between job characteristics and safety performance dimensions.

Hypothesis 2b. Management commitment to safety will be positively related to job engagement.

1.3. Job engagement and safety performance Individuals experiencing job engagement tend to invest their physical, cognitive, and emotional energies into role performance (Kahn; 1990; Rich et al., 2010). Resource allocation to the task at hand is crucial for successful performance (Beal et al., 2005). Workers who are more engaged will therefore perform better. Specifically, job engagement is often associated with positive emotions, which can broaden individuals' resource repertoires (Bakker and Bal, 2010; Bledow et al., 2011; Fredrickson, 2001). Accordingly, workers with greater levels of job engagement have more resources available to invest in their jobs, which further lead to better performance. Workers high on engagement also experience a high level of arousal, which is functional for initiating goaloriented actions (Bakker and Bal, 2010; Bledow et al., 2011; Langelaan et al., 2006). The positive relationship between engagement and performance has received empirical support from studies utilizing the JD-R model (Rich et al., 2010; Nahrgang et al., 2011). In the context of safety performance, safety compliance refers to the core activities that ensure workplace safety, such as adhering to safety procedures (Griffin and Neal, 2000). Safety knowledge, safety skills, and safety motivation are important determinants of safety compliance (Christian et al., 2009; Griffin and Neal, 2000; Neal et al., 2000; Sinclair et al., 2010). As detailed above, job engagement is associated with expanded repertoires of physical, cognitive, and emotional resources and helps to initiate goaloriented behaviors (Kahn, 1990; Bakker and Bal, 2010). Accordingly, workers high on engagement tend to work with greater intensity, pay attention to role responsibilities, and get emotionally connected to their work (Rich et al., 2010). It thus follows that job engagement might be tied to increased safety compliance, which requires the activation and input of such personal resources as safety knowledge and skills. Hypothesis 3a. Job engagement will be positively related to safety compliance. Safety participation taps into voluntary behaviors that are not mandated by the organization but do enhance workplace safety (Griffin and Neal, 2000). People who are experiencing job engagement have a larger resource repertoire (Bakker and Bal, 2010; Bledow et al., 2011; Fredrickson, 2001). They might

Hypothesis 3b. Job engagement will be positively related to safety participation.

1.4. The mediating role of job engagement

Hypothesis 4a. Job engagement will partially mediate the relationships between job characteristics (i.e., job hindrances and job resources) and safety compliance. Hypothesis 4b. Job engagement will partially mediate the relationships between job characteristics (i.e., job hindrances and job resources) and safety participation. The proposed relationships among study variables are presented in Fig. 1. 2. Method 2.1. Participants The participants in this study were employees from a coal mining company in China. In China, workplace safety is a salient issue in the coal mining industry. To illustrate, 1403 accidents occurred in 2010, taking away the lives of 2433 coal miners in China

166

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

Fig. 1. The proposed relationships among study variables.

(State Administration of Work Safety, 2011). As the state and regional regulatory bodies put more emphasis on occupational health and safety issues, coal mining companies are pressured to prevent accidents while also attending to market competition for production. The company where the present study sampled from was state-owned, with relatively higher standards for workplace safety and production as compared to private coal mining companies in China. As part of a larger project aimed to promote employee well-being, a total of 300 survey questionnaires were distributed along with a support letter from the Human Resource manager, and 250 valid ones were returned (response rate ¼ 83.3%). The participants were informed of the anonymity of their responses. The high response rate may have been partly due to support from the Human Resources manager and the anonymous nature of data collection. The explicit focus of the project on employee well-being may have also helped to reduce resistance to participate in the study.3 Demographics check against the organizational record suggests that our study sample was representative of coal mining employees working in the company. Among the participants, 213 were male (85.2%), 20 were female (8.0%), and 17 were unidentified. Participants had an average age of 32.69 years (SD ¼ 7.39). Their mean job tenure was 5.10 years (SD ¼ 5.15). The study sample consisted of both rank-and-file (75.3%) and higherlevel employees (23%). 59.7% of the employees were in a longterm employment contract with the organization whereas 40.3% were recruited from an outsourcing agency. 2.2. Measures 2.2.1. Job hindrances Two types of job hindrances, role overload and job insecurity, were included in this study. Role overload was measured using the Chinese version of the role stressor scale (Li and Zhang, 2009; Peterson et al., 1995). The scale includes five items that were scored on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Sample items include “My workload is too heavy for me” and “I'm taking too much burden at work.” Job insecurity was measured using the job insecurity subscale from the 49-item job content questionnaire (JCQ; Karasek et al., 1998). A Chinese version of the JCQ is available, and it includes six items (Sha et al., 2003). Sample items include “Is your work steady?” and “Is it likely that you'll lose your current job in this company?” The responses were coded such that a higher score refers to a higher level of job hindrances.

3 Although the unusually high response rate might be considered as a study concern, nevertheless it helped reduce the non-response bias. Otherwise, employees might self-select into the sample such that those with low levels of safety performance will withdraw from the study. Therefore, the high response rate can reduce the concern of the non-response bias and its associated negative consequences such as attenuated study relationships (Greco et al., 2014).

2.2.2. Job resources Two types of job resources, coworker support and management commitment to safety, were included in this study. Coworker support was measured using the coworker support subscale from the Chinese version of the 21-item JCQ (Karasek et al., 1998; Yang and Li, 2004). Four items were scored on a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree). Sample items include “My coworkers are friendly” and “My coworkers help me at work.” Management commitment to safety was measured using three items from the safety climate scale validated in the Chinese context, which includes three items (Lin et al., 2008). Sample items include “Management considers safety to have the same importance as production” and “Management is concerned with safety problems in my workplace”. Responses were coded such that a higher score refers to a higher level of job resources. 2.2.3. Job engagement Job engagement was measured using the Chinese version of the Utrecht work engagement scale (Schaufeli and Bakker, 2003; Zhang and Gan, 2005). The items were scored on a 7-point Likert scale ranging from 0 (never) to 6 (always). Sample items of the three dimensions are: “At my work, I feel bursting with energy” (Vigor); “I am enthusiastic about my job” (Dedication); “Time flies when I'm working” (Absorption). The responses were coded such that a higher score refers to greater job engagement. As the three dimensions are highly correlated with each other (from .82 to .84), we computed a composite score by averaging all of the items, as recommended by Schaufeli et al. (2006). This approach avoids the issue of multicollinearity when highly correlated dimensions are entered into a regression equation. 2.2.4. Safety performance Safety performance was measured using the safety behavior scale from Neal and Griffin (2006). A Chinese version is available (Li et al., 2013). Safety compliance and safety participation both include three items that were scored on a 5-point Likert scale ranging from 1 (almost never) to 5 (almost always). A sample item of safety compliance is “I use all the necessary safety equipment to do my job” and that of safety participation is “I promote the safety program within the organization.” The responses were coded such that a higher score refers to better safety performance. We conducted confirmatory factor analysis to help establish the construct validity of the scale. The two-factor model (safety compliance and safety participation) provided a good fit to the data (c2 (8) ¼ 19.93, p < .05, GFI ¼ .98, NFI ¼ .98, TLI ¼ .98, CFI ¼ .99, RMSEA ¼ .08)4 whereas the alternative model in which every items loaded on

4 GFI ¼ Goodness-of-fit index; NFI ¼ Normed fit index, TLI ¼ TuckereLewis index, CFI ¼ comparative fit index, RMSEA ¼ root mean square error of approximation.

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

the same latent variable did not fit well to the data (c2 (9) ¼ 91.50, p < .01, GFI ¼ .88, NFI ¼ .91, IFI ¼ .92, TLI ¼ .86, CFI ¼ .91, RMSEA ¼ .19). Model comparison supported that the two-factor model was superior to the alternative model (c2 (1) ¼ 71.58, p < .001). Therefore, safety compliance and safety participation, although correlated with each other, should be treated as two separate safety performance dimensions.

167

Table 2 Hierarchical regression analysis results predicting job engagement, safety compliance, and safety participation.

Dependent variable: job engagement Contract type Gender Job level Job tenure Job insecurity Role overload Coworker support Management commitment to safety R2 DR2 Dependent variable: safety compliance Contract type Gender Job level Job tenure Job engagement R2 DR2 Dependent variable: safety participation Contract type Gender Job level Job tenure Job engagement R2 DR2

2.2.5. Control variables Given the demographic differences within the study sample, they might obfuscate the relationships among study variables. As such, gender, job tenure, job level, and job contract type were controlled for to rule out the possibility that these variables caused relationships among study variables. 3. Results 3.1. Analytical strategy Hierarchical regression analysis was conducted to test the study hypotheses. Hypotheses 1a, 1b, 2a, and 2b were tested by first entering control variables and then entering job characteristics (i.e., job hindrances and job resources) to predict job engagement. To test hypotheses 3a and 3b, control variables and job engagement were entered to predict safety compliance (3a) and safety participation (3b), respectively. To test hypotheses 4a and 4b, control variables were entered in the first step, followed by job characteristics in the second step and job engagement in the third step to predict safety compliance (4a) and safety participation (4b), respectively.

Step 1

Step 2

.02 .25** .08 .03

.07 .19** .05 .00 .08 .08 .20** .34** .28** .20**

.08**

.00 .14* .16* .04

.01 .03 .13 .03 .43** .23** .17**

.06*

.10 .11 .21** .12

.11 .00 .17* .10 .44** .26** .18**

.09**

Note. N ¼ 250. *p < .05. **p < .01. Standardized beta is reported.

were supported. Job engagement was positively related to safety compliance (see Table 2; b ¼ .43, p < .01) and to safety participation (see Table 2; b ¼ .44, p < .01). We followed the approach recommended by Baron and Kenny (1986) to examine the mediating role of job engagement, followed by bootstrapping (Preacher and Hayes, 2008). Independent variables (i.e., job characteristics) must first be found to be significantly related to the putative mediator (i.e., job engagement). This requirement was partially met because we found that coworker support and management commitment to safety were significantly related to job engagement (see Hypotheses 2a and 2b) while job insecurity and role overload were not (see Hypotheses 1a and 1b). To move on to the next step, the mediator should be significantly related to the dependent variables (i.e., safety compliance and safety participation), and this was supported when we tested Hypotheses 3a and 3b. For the last step, job characteristics should be significantly related to safety performance dimensions and the significant relationships should become nonsignificant (full mediation) or at least become smaller in magnitude (partial mediation), and the

3.2. Descriptive statistics The means, standard deviations, alpha levels, and correlations of all study variables are presented in Table 1. All of the significant relationships among the variables were in the expected directions. 3.3. Hypothesis testing Hypotheses 1a and 1b, which proposed negative relationships between job insecurity and job engagement and between role overload and job engagement, were not supported (see Table 2; job insecurity, b ¼ .08, ns; role overload, b ¼ .08, ns). Hypotheses 2a and 2b, which proposed positive relationships between coworker support and job engagement and between management commitment to safety and job engagement, were supported (see Table 2; coworker support, b ¼ .20, p < .01; management commitment to safety, b ¼ .34, p < .01). Hypotheses 3a and 3b, which proposed that job engagement would be positively related to safety performance dimensions,

Table 1 Reliabilities, means, SDs, and intercorrelations among the study variables. Variables

M

SD

1

2

3

4

5

6

7

8

1. 2. 3. 4. 5. 6. 7. 8.

5.10 2.44 3.35 2.93 3.63 3.37 3.60 3.54

5.15 .55 .84 .50 .87 1.09 1.05 .84

e .06 .08 .01 .12 .10 .11 .18**

.66 .31** .14* .09 .00 .30** .04

.84 .05 .01 .09 .13* .12

.73 .27** .27** .27** .32**

.73 .40** .48** .33**

.92 .47** .49**

.92 .70**

.82

Job tenure Job insecurity Role overload Coworker Support Management commitment to safety Job engagement Safety compliance Safety participation

Note. N ¼ 250. M ¼ mean; SD ¼ standard deviation. *p < .05. **p < .01. Cronbach's a values on diagonal, where applicable.

168

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

proposed mediator should still be a significant predictor. To test for the mediating role of engagement, we first entered control variables, then entered job characteristics and finally entered job engagement to predict safety compliance (4a) and safety participation (4b), respectively. When predicting safety compliance (see Table 3), coworker support (b ¼ .21, p < .01) and management commitment to safety (b ¼ .39, p < .01) were positively related to safety compliance before job engagement was entered. Their standardized regression coefficients became smaller in magnitude after job engagement was entered (coworker support, b ¼ .15, p < .05; management commitment to safety, b ¼ .29, p < .01), and job engagement was found to be significantly related to safety compliance (b ¼ .28, p < .01). Job engagement explained an additional 6% of the variance in safety compliance. Bootstrapping results also confirmed the mediating role of engagement such that the 95% confidence interval (CI) for the indirect effects of engagement did not include zero for coworker support (indirect effect: .12; 95% CI: [.04, .23]) and management commitment to safety (indirect effect: .12; 95% CI: [.06, .19]). For hypothesis 4b (see Table 3), coworker support (b ¼ .21, p < .01) and management commitment to safety (b ¼ .25, p < .01) were positively related to safety participation before job engagement was entered. Their standardized regression coefficients became smaller in magnitude after job engagement was entered (coworker support, b ¼ .14, p < .05; management commitment to safety, b ¼ .14, p < .05), and job engagement was found to be significantly related to safety participation (b ¼ .34, p < .01). Job engagement explained an additional 8% of the variance in safety participation. The 95% confidence interval for the indirect effects of engagement based on bootstrapping did not include zero for coworker support (indirect effect: .12; 95% CI: [.05, .21]) and management commitment to safety (indirect effect: .11; 95% CI: [.06, .18]). Taken together, these results suggest that hypotheses 4a and 4b, which proposed the partial mediating role of engagement between job characteristics and safety performance dimensions, were partially supported. Job engagement partially mediated the relationships between job resources (i.e., coworker support and management commitment to safety) and safety performance dimensions (i.e., safety compliance and safety participation). 4. Discussion Although job engagement has received extensive attention from researchers and practitioners alike, its effect on safety performance has yet to be empirically examined. The goal of the present study was to examine the mediating role of engagement in the relationships between two types of job characteristics (i.e., job hindrances and job resources) and safety performance dimensions (i.e., safety compliance and safety participation) by drawing on the JD-R model (Demerouti et al., 2001). We found that job engagement partially mediated the relationships between job resources and safety performance dimensions. The present study contributes to safety research by highlighting the importance of job engagement in channeling the effect of job resources on individual safety behaviors and piecing together the roles of job design (i.e., job characteristics) and person variable (i.e., job engagement) in determining safety performance. First, we found support for the mediating role of engagement in the relationships between job resources (i.e., coworker support and management commitment to safety) and safety performance dimensions (i.e., safety compliance and safety participation). Consistent with the JD-R model (Demerouti et al., 2001), our study findings echo past studies that documented the mediating effects of engagement between job characteristics and employee work behaviors including task performance (Rich et al., 2010), contextual performance (Xanthopoulou et al., 2008), and personal initiatives

(Hakanen et al., 2008). Considering that job engagement, to our knowledge, has rarely been examined in safety research (Hansez and Chmiel, 2010; Nahrgang et al., 2011), our study helps to fill the void by supporting the mediating role of job engagement as conceptualized by Schaufeli et al. (2002) in the context of safety. Focusing on job engagement e an affective-motivational concept e helps illuminates the mechanism through which the social and technical aspects of the job (i.e., job characteristics) influence individual safety behaviors. While past studies on general job performance supported the full mediating effect of job engagement (Demerouti et al., 2001; Crawford et al., 2010), we found job engagement to be a partial mediator. This is consistent with safety studies that demonstrated the direct effects of certain job resources on safety behavior in the presence of mediators such as safety motivation (Neal et al., 2000) and positive occupational states (Hansez and Chmiel, 2010). The job resources included in the present study (coworker support and management commitment to safety) might initiate other psychological processes in addition to job engagement. Perceived coworker support and management commitment to safety might influence safety performance through social exchange as employees feel obliged to reciprocate the support they receive by complying with safety procedures and helping other coworkers with safety issues (Blau, 1964; Hofmann and Morgeson, 1999). This unexamined mechanism might account for the direct link between job resources and safety performance. Nevertheless, the present study was able to reveal job engagement as an important psychological process through which job design features could influence safety performance. Second, we replicated previous studies by showing that job resources were positively related to job engagement, as proposed in the motivational process of the JD-R model (Demerouti et al., 2001). In addition to the well-established link between coworker support and job engagement (e.g., Crawford et al., 2010; Schaufeli and Bakker, 2004), we were able to establish the relationship between management commitment to safety, a crucial job resource in safety research (Hansez and Chmiel, 2010; Zohar, 1980), and job engagement. Management commitment to safety and the broader concept of safety climate have frequently been studied in safety research but empirical evidence of their relationships with job engagement was limited (Hansez and Chmiel, 2010; Nahrgang et al., 2011). Moreover, the effect of management commitment to safety on engagement appeared stronger than that of coworker support on engagement, which provides further support for the usefulness of examining safety-related job resources in predicting job engagement. In contrast to the significant effects of job resources on job engagement, we found no significant relationships between job hindrances (i.e., job insecurity and role overload) and job engagement. However, in the final regression equation that predicted safety compliance with control variables, job characteristics, and job engagement (see DV: safety compliance step 3 in Table 3), job insecurity was found to be negatively related to safety compliance, consistent with past studies (Probst, 2004; Probst and Brubaker, 2001). As such, our finding lends support to the detrimental effect of job insecurity as a hindrance stressor on safety compliance (Clarke, 2012). The positive relationship between role overload and safety participation (see DV: safety participation step 3 in Table 3) was unexpected, as existing evidence suggests that role overload is the strongest individual-level predictor of injury (Hofmann and Stetzer, 1996; Zohar, 2000). A tentative explanation involves the cross-sectional design of the present study. Safety participation might well be a part of the perceived role expectation of employees working in safety-critical industries. To meet the role expectation, individuals who engage in more safety participation might perceive

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

169

greater levels of role overload. However, this explanation awaits examination in future research. We were also unable to find support for the mediating role of engagement in the relationships between job hindrances and engagement. Future research might want to revisit this question by using a more reliable job insecurity measure (alpha ¼ .66 in the present study) and a more rigorous design (e.g., longitudinal research). The current study, although explicitly grounded on a prominent job characteristics theory (JD-R), also has important implications for applied ergonomics research. Historically, job characteristics approach has complemented socio-technical theory in influencing job design research (Holman et al., 2002). Our study, by applying a more nuanced theoretical lens in the job characteristics literature, adds to our understanding of job design in a safety-critical industry by showing how specific job features may impact individual attitudes and behavior. Further, we answer the recent call for integration of ergonomics and safety research by examining how individual attitudes and behaviors are shaped by their perceptions of work design system and safety-related practices (Murphy et al., 2014). Specifically, job hindrances such as role overload may signal a misfit between the technical system and the social system and thus adversely influence work engagement and safety performance. The positive effects of job resources, on the contrary, suggest that social interactions (i.e., coworker support) and perceptions of safety practices (i.e., management commitment to safety) can jointly lead to higher levels of engagement and safety behaviors. The effects of job hindrances and job resources have implications for the work system design by highlighting the differential impact of various job characteristics.

to call for longitudinal designs to better examine the causal relations among study variables. The second limitation deals with the way that study variables were measured. The reliance on self-report measures might raise concern about common method variance. To assess this concern, Harman's single-factor analysis was conducted to examine the potential influence of common method variance (Podsakoff et al., 2003). The first factor accounted for 38% of the total variance, suggesting that common method bias should not be a great threat to our study findings. Further, it is important to note that the variables of interest in the present study all deal with personal experiences and behaviors into which self-report measures can provide useful insights (Spector, 1994). Nevertheless, we call for future research to collect data from multiple sources to minimize the concern of common method bias. Specifically, some objective job characteristics such as occupational hazards (e.g., Ford and Tetrick, 2011) and safety outcomes such as microaccidents (e.g., Zohar, 2000) could be obtained from the organization. A third limitation is that management commitment to safety, the core component of safety climate, was used as an individuallevel variable in the present study. Although this variable has long been examined at a unit level (Zohar, 2000; Zohar and Luria, 2004), it is not rare to measure it at the individual level (Hansez and Chmiel, 2010). The intraclass coefficient was extremely low (.05), suggesting that grouping effects were marginal in our data. Future studies might nevertheless examine management commitment to safety and the broader concept, safety climate, as a unitlevel job resource to examine possible cross-level effects. This could yield valuable insights into the contextual influence of safety climate as an important job resource. The final limitation of the present study involves the generalizability of our study findings. The participants were from a coal mining company operating in China. Moreover, the response rate (83.33%) was much higher than the norm. This specific setting might have influenced our study findings and limited its generalizability to other contexts. We included job characteristics that were salient to our study sample (i.e., blue-collar coal mining workers). However, these job aspects could be less important for other occupations. When conducting a study in another safetyrelevant occupation, researchers might want to closely examine the specific set of job characteristics relevant for the job incumbents. In the present study we relied on the JD-R theory (Demerouti et al., 2001) to propose the mediating role of job engagement in the relations between job aspects and safety performance. Although recent meta-analytic evidence suggests the JDR model is applicable in the Chinese context as well (Yuan and Jia, 2014), we encourage future researchers to consider possible contextual variables that might influence the applicability of the model. We recommend that researchers extend the present study by including other theoretically important variables. For example, burnout, a construct that is closely related to engagement, could also be investigated to provide a full-scale empirical test of the JD-R model (Demerouti et al., 2001; Li et al., 2013; Nahrgang et al., 2011). Moreover, future research could look into other safety-critical industries in light of the potentially limited generalizability of our research findings.

5. Limitations and future directions

6. Practical implications

The present study is not without its limitations and the research findings should be viewed with caution. To begin with, the crosssectional design of the present study limited causation interpretations. As mentioned earlier, there is a possibility that safety participation could lead to perceived role overload. We would like

Our findings have several practical implications. First, safetycritical companies and safety consulting companies are advised to recognize the importance of job engagement in the safety domain. Although the importance of job engagement in the general job performance domain has been widely recognized, safety

Table 3 Hierarchical regression analysis results testing the mediating role of job engagement.

Dependent variable: safety compliance Contract type Gender Job level Job tenure Job insecurity Role overload Coworker support Management commitment to safety Job engagement R2 DR 2 Dependent variable: safety participation Contract type Gender Job level Job tenure Job insecurity Role overload Coworker support Management commitment to safety Job engagement R2 DR 2

Step 1

Step 2

Step 3

.00 .14* .16* .04

.02 .05 .12 .01 .23** .01 .21** .39**

.06*

.32** .27**

.04 .01 .10 .01 .25** .01 .15* .29** .28** .38** .06**

.10 .11 .21** .12

.03 .06 .18* .07 .06 .12 .21** .25**

.09**

.25** .16**

.06 .00 .16* .07 .03 .14* .14* .14* .34** .33** .08**

Note. N ¼ 250. *p < .05. **p < .01. Standardized beta is reported.

170

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171

researchers and practitioners have been slow to embrace job engagement, a construct that grew out of positive psychology, in safety research and practice (Crawford et al., 2010; Macey and Schneider, 2008; Nahrgang et al., 2011). Job engagement, which captures human agency in the workplace, is a promising way to enhance individual safety performance. Managers are encouraged to calibrate their managerial practices and foster employee engagement in an attempt to promote employee safety performance. Safety consulting firms could include job engagement as another key avenue to leverage individual safety performance. A good example is the JD-R-based online tool that has been commercially implemented in the Netherlands (Schaufeli and Dijkstra, 2010). As part of the JD-R monitor, valid scales are pooled together to measure job demands, job resources, and job engagement, along with other potentially important variables. The online tool will in turn generate feedback on the individual level, unit level, and organization level. This information serves as the input for a systematic process of organizational change e starting from problem identification, the design of the JD-R monitor, internal communication, survey and feedback, to intervention, evaluation, and the next round of problem identification. Organizations in safety-critical industries can easily tailor this monitor to identify and address the specific set of job characteristics that impedes job engagement and safety behaviors. The job characteristics included in the present study point to another practical implication. Management are advised to take deliberate steps to foster a work environment characterized by coworker support and managerial support for safety since these two factors can influence job engagement and safety performance. Managers should set the tone for workplace safety by being individually involved in safety matters. Coworker support should also be encouraged within the organization. In terms of job hindrances, organizations could reconsider their work design and organizational policy to reduce job hindrances such as job insecurity. 7. Conclusions The present study provides empirical support for the importance of job engagement in safety research. Our findings suggest that coworker support and management commitment to safety can act as job resources that positively influence safety performance both directly and indirectly through job engagement. Job insecurity was negatively related to safety compliance, whereas role overload was positively related to safety participation. The mediating role of job engagement holds promise as a leverage to enhance safety performance and improve workplace safety. References Bakker, A.B., Bal, M.P., 2010. Weekly work engagement and performance: a study among starting teachers. J. Occup. Organ. Psychol. 83, 189e206. Bakker, A.B., Schaufeli, W.B., 2008. Positive organizational behavior: engaged employees in flourishing organizations. J. Organ. Behav. 29, 147e154. Baron, R.M., Kenny, D.A., 1986. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J. Personal. Soc. Psychol. 51, 1173e1182. Beal, D.J., Weiss, H.M., Barros, E., MacDermid, S.M., 2005. An episodic process model of affective influences on performance. J. Appl. Psychol. 90, 1054e1068. Blau, P., 1964. Exchange and Power in Social Life. Wiley, New York. Bledow, R., Schmitt, A., Frese, M., Kühnel, J., 2011. The affective shift model of work engagement. J. Appl. Psychol. 96, 1246e1257. 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, pp. 71e98. Bosman, J., Rothmann, S., Buitendach, J.H., 2005. Job insecurity, burnout and work engagement: the impact of positive and negative affectivity. SA J. Ind. Psychol. 31, 48e56. Burke, M.J., Sarpy, S.A., Tesluk, P.E., Smith-Crowe, K., 2002. General safety performance: a test of a grounded theoretical model. Pers. Psychol. 55, 429e457.

Carlson, M., Charlin, V., Miller, N., 1988. Positive mood and helping behavior: a test of six hypotheses. J. Personal. Soc. Psychol. 55, 211e229. Cavanaugh, M.A., Boswell, W.R., Roehling, M.V., Boudreau, J.W., 2000. An empirical examination of self-reported work stress among U.S. managers. J. Appl. Psychol. 85, 65e74. Christian, M.S., Bradley, J.C., Wallace, J.C., Burke, M.J., 2009. Workplace safety: a meta-analysis of the roles of person and situation factors. J. Appl. Psychol. 94, 1103e1127. Chowdhury, S.K., Endres, M.L., 2010. Customer involvement in service operation and its impact on job stress and work place injury. Acad. Manag. J. 53, 182e198. Clarke, S., 2012. The effect of challenge and hindrance stressors on safety behavior and safety outcomes: a meta-analysis. J. Occup. Health Psychol. 17, 387e397. Cohen, S., Wills, T.A., 1985. Stress, social support, and the buffering hypothesis. Psychol. Bull. 98, 310e357. Crawford, E.R., LePine, J.A., Rich, B.L., 2010. Linking job demands and resources to employee engagement and burnout: a theoretical extension and meta-analytic test. J. Appl. Psychol. 95, 834e848. Demerouti, E., Bakker, A.B., Nachreiner, F., Schaufeli, W.B., 2001. The job demandsresources model of burnout. J. Appl. Psychol. 86, 499e512. De Cuyper, N., De Witte, H., 2005. Job insecurity: mediator or moderator of the relationship between type of contract and various outcomes? SA J. Ind. Psychol. 31, 79e86. Ford, M.T., Tetrick, L.E., 2011. Relations among occupational hazards, attitudes, and safety performance. J. Occup. Health Psychol. 16, 48e66. Fredrickson, B.L., 2001. The role of positive emotions in positive psychology: the broaden-and-build theory of positive emotions. Am. Psychol. 56, 218e226. George, J.M., 1991. State or trait: effects of positive mood on prosocial behaviors at work. J. Appl. Psychol. 76, 299e307. Grant, A.M., Fried, Y., Huillerat, T., 2011. Work matters: job design in classic and contemporary perspectives. In: Zedeck, S. (Ed.), APA Handbook of Industrial and Organizational Psychology, Building and Developing the Organization, vol. 1. American Psychological Association, Washington DC, pp. 417e453. Greco, L.M., O'Boyle, E.H., Walter, S.L., 2014. Absence of malice: a meta-analysis of nonresponse bias in counterproductive work behavior research. J. Appl. Psychol. 100, 75e97. Griffin, M.A., Neal, A., 2000. Perceptions of safety at work: a framework for linking safety climate to safety performance, knowledge, and motivation. J. Occup. Health Psychol. 5, 347e358. Hackman, J.R., Oldham, G.R., 1976. Motivation through the design of work: test of a theory. Organ. Behav. Hum. Perform. 16, 250e279. Hakanen, J.J., Perhoniemi, R., Toppinen-Tanner, 2008. Positive gain spirals at work: from job resources to work engagement, personal initiative and work-unit innovativeness. J. Vocat. Behav. 73, 78e91. Hansez, I., Chmiel, N., 2010. Safety behavior: job demands, job resources, and perceived management commitment to safety. J. Occup. Health Psychol. 15, 267e278. Hofmann, D.A., Morgeson, F.P., 1999. Safety-related behavior as a social exchange: the role of perceived organizational support and leader-member exchange. J. Appl. Psychol. 84, 286e296. Hofmann, D.A., Stetzer, A., 1996. A cross-level investigation of factors influencing unsafe behaviors and accidents. Pers. Psychol. 49, 307e339. Holman, D., Clegg, C., Waterson, P., 2002. Navigating the territory of job design. Appl. Ergon. 33, 197e205. Kahn, W.A., 1990. Psychological conditions of personal engagement and disengagement at work. Acad. Manag. J. 33, 692e724. Karasek, R.A., Brisson, C., Kawakami, N., Houtman, I., Bongers, P., Amick, B., 1998. The job content questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J. Occup. Health Psychol. 3, 322e355. Langelaan, S., Bakker, A.B., van Doorman, L.J.P., Schaufeli, W.B., 2006. Burnout and work engagement: do individual differences make a difference? Personal. Individ. Differ. 40, 521e532. Li, F., Jiang, L., Yao, X., Li, Y., 2013. Job demands, job resources and safety outcomes: the roles of emotional exhaustion and safety compliance. Accid. Anal. Prev. 51, 243e251. Li, C., Zhang, Y., 2009. The effects of role stressors on physical health and mental health among Chinese teachers. Psychol. Dev. Educ. 1, 114e119. Lin, S.-H., Tang, W.-J., Miao, J.-Y., Wang, Z.-M., Wang, P.-X., 2008. Safety climate measurement at workplace in China: a validity and reliability assessment. Saf. Sci. 46, 1037e1046. Luthans, F., Youssef, C.M., 2007. Emerging positive organizational behavior. J. Manag. 33, 321e349. Macey, W.H., Schneider, B., 2008. The meaning of employee engagement. Ind. Organ. Psychol. 1, 3e30. Mauno, S., Kinnunen, U., Ruokolainen, M., 2007. Job demands and resources as antecedents of work engagement: a longitudinal study. J. Vocat. Behav. 70, 149e171. Murphy, L.A., Robertson, M.M., Carayon, P., 2014. The next generation of macroergonomics: integrating safety climate. Accid. Anal. Prev. 68, 16e24. Nahrgang, J.D., Morgeson, F.P., Hofmann, D.A., 2011. Safety at work: a meta-analytic investigation of the link between job demands, job resources, burnout, engagement, and safety outcomes. J. Appl. Psychol. 96, 71e94. Neal, A., Griffin, M.A., 2006. A study of the lagged relationships among safety climate, safety motivation, safety behavior, and accidents at the individual and group levels. J. Appl. Psychol. 91, 946e953.

Z. Yuan et al. / Applied Ergonomics 51 (2015) 163e171 Neal, A., Griffin, M.A., Hart, P.M., 2000. The impact of organizational climate on safety climate and individual behavior. Saf. Sci. 34, 99e109. Peterson, M.F., Smith, P.B., Akande, A., Ayestaran, S., Bochner, S., Callan, V., et al., 1995. Role conflict, ambiguity, and overload: a 21-nation study. Acad. Manag. J. 38, 429e452. Podsakoff, P.M., MacKenzie, S.B., Lee, J., Podsakoff, N.P., 2003. Common method biases in behavioral research: a critical review of the literature and recommended remedies. J. Appl. Psychol. 88, 879e903. Preacher, K.J., Hayes, A.F., 2008. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav. Res. Methods 40, 879e891. Probst, T.M., 2004. Safety and insecurity: exploring the moderating effect of organizational safety climate. J. Occup. Health Psychol. 9, 3e10. Probst, T.M., Brubaker, T.L., 2001. The effects of job insecurity on employee safety outcomes: cross-sectional and longitudinal explorations. J. Occup. Health Psychol. 6, 139e159. Rich, B.L., LePine, J.A., Crawford, E.R., 2010. Job engagement: antecedents and effects on job performance. Acad. Manag. J. 53, 617e635. Rousseau, D.M., 1977. Technological differences in job characteristics, employee satisfaction, and motivation: a synthesis of job design research and sociotechnical systems theory. Organ. Behav. Hum. Perform. 19, 18e42.
, J.M., 2005. Linking organizational resources and work Salanova, M., Agut, S., Peiro engagement to employee performance and customer loyalty: the mediation of service climate. J. Appl. Psychol. 90, 1217e1227. Schaufeli, W.B., Bakker, A.B., 2003. Utrecht Work Engagement Scale (UWES), Preliminary Manual (Version1, Novermber 2003). Utrecht University: Occupational Health Psychology Unit. Schaufeli, W.B., Bakker, A.B., 2004. Job demands, job resources, and their relationship with burnout and engagement: a multi-sample study. J. Organ. Behav. 25, 293e315. Schaufeli, W.B., Bakker, A.B., Salanova, M., 2006. The measurement of work engagement with a short questionnaire a cross-national study. Educ. Psychol. Meas. 66, 701e716. Schaufeli, W.B., Dijkstra, P., 2010. Bevlogen aan het werk [Engaged at work]. Thema, Zaltbommel, Netherlands.

171


, V., Bakker, A.B., 2002. The meaSchaufeli, W.B., Salanova, M., Gonz
alez-Roma surement of engagement and burnout: a two sample confirmatory factor analytic approach. J. Happiness Stud. 3, 71e92. Sha, Y., Liu, P., Li, J., Na, D., 2003. The validation of Chinese version of job content questionnaire in health professionals. China Occup. Med. 30, 24e27. Sinclair, R.R., Martin, J.E., Sears, L.E., 2010. Labor unions and safety climate: perceived union safety values and retail employee safety outcomes. Accid. Anal. Prev. 42, 1477e1487. Spector, P.E., 1994. Using self-report questionnaires in OB research: a comment on the use of a controversial method. J. Organ. Behav. 15, 385e392. State Administration of Work Safety, 2011. Workplace Safety Report in 2010. Retrieved from. http://www.chinasafety.gov.cn/. €swall, K., 2002. No security: a meta-analysis and review of Sverke, M., Hellgren, J., Na job insecurity and its consequences. J. Occup. Health Psychol. 7, 242e264. Xanthopoulou, D., Bakker, A.B., Heuven, E., Demerouti, E., Schaufeli, W.B., 2008. Working in the sky: a diary study on work engagement among flight attendants. J. Occup. Health Psychol. 13, 345e356. Yang, W., Li, J., 2004. Measurement of psychosocial factors in work environment: application of two models of occupational stress. Chin. J. Ind. Hyg. Occup. Dis. 22, 422e426. Yuan, Z., Jia, X., 2014, June. A meta-analytic examination of the job demandsresources model in Chinese context. In: Paper Presented at the 1st Human Resource Division International Conference, Beijing, China. Yuan, Z., Li, Y., Lin, J., 2014. Linking challenge and hindrance stress to safety performance: the moderating role of core self-evaluation. Personal. Individ. Differ. 68, 154e159. Zhang, Y., Gan, Y., 2005. The Chinese version of utrecht work engagement scale: an examination of reliability and validity. Chin. J. Clin. Psychol. 13, 268e270. Zohar, D., 1980. Safety climate in industrial organizations: theoretical and applied implications. J. Appl. Psychol. 65, 96e102. Zohar, D., 2000. A group-level model of safety climate: testing the effect of group climate on microaccidents in manufacturing jobs. J. Appl. Psychol. 85, 587e596. Zohar, D., 2002. Modifying supervisory practices to improve subunit safety: a leadership-based intervention model. J. Appl. Psychol. 87, 156e163. Zohar, D., Luria, G., 2004. Climate as a social-cognitive construction of supervision safety practices: scripts as proxy of behavior patterns. J. Appl. Psychol. 89, 322e333.