- Email: [email protected]
Curvilinear effects of personality on safety performance: The moderating role of supervisor support
Personality and Individual Differences 122 (2018) 55–61
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
Curvilinear effects of personality on safety performance: The moderating role of supervisor support
MARK
Xiao Yuana,b, Yongjuan Lia,b, Yaoshan Xua,b,⁎, Naixi Huangc a b c
CAS Key Laboratory of Behavioral Science, Institute of Psychology, Chinese Academy of Sciences, 16 LinCui Road, Chaoyang District, Beijing 100101, China Department of Psychology, University of Chinese Academy of Sciences, 19(A) Yuquan Road, Shijingshan District, Beijing 100049, China Research Institute of Nuclear Power Operation of China, 1021 National Road, East Lake High Tech Zone, Wuhan 430223, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Personality Curvilinear relationship Safety performance Supervisor support climate
Personality has long been assumed to be linearly associated with safety performance. Focusing on the benefits and costs of personality traits, this study investigates the curvilinear effects of personality traits (i.e., Conscientiousness, Emotional Stability and Extraversion) on safety performance. Results support our hypothesized curvilinear effects that their relationships are initially positive but become negative as personality trait scores increase to extremely high. Furthermore, drawing on theory of purposeful work behavior, this study proposes and finds that the curvilinear effects of personality traits on safety performance are moderated by teamlevel supervisor support climate, such that the inflection points after which the personality–performance relationship becomes negative are higher under high levels of supervisor supportive environment than those under low levels of such environment. Implications for research and practice, as well as limitations and directions for future research, are provided.
1. Introduction Workplace safety is an increasing concern worldwide. To ensure high levels of safety performance, considerable research has focused on the role of personality (e.g., Beus, Dhanani, & McCord, 2015). Early studies have shown that Extraversion and Neuroticism were weakly positively associated with accidents within the framework of Eysenck's theory (e.g., Hansen, 1988). With the emergence of the “Big Five” personality theory, the positive role of Conscientiousness in predicting safety behavior was observed (Postlethwaite, Robbins, Rickerson, & McKinniss, 2009). Christian, Bradley, Wallace, and Burke (2009) summarized the importance of three personality traits, namely Conscientiousness, Emotional Stability and Extraversion, to safety performance in organizational settings. However, the relatively weak or equivocal correlations remain as issues (e.g., Beus et al., 2015). A promising idea is that personality–safety performance relationship may be nonlinear instead of linear. Growing researches have begun to examine curvilinear personality–performance relationship (e.g., Le et al., 2011). However, minimal attention has been given to safety performance. Although Carter et al. (2014) has provided initial evidence for curvilinear Conscientiousness–safety performance relationship, the theoretical framework for this relationship remains insufficient. In addition, Emotional Stability and Extraversion should be
⁎
further explored (Christian et al., 2009). Therefore, this study aims to explore the curvilinear effects of Conscientiousness, Emotional Stability and Extraversion on safety performance. Moreover, the effects of personality on performance vary across different situations (Barrick, Mount, & Li, 2013). If a curvilinear relationship exists between personality and safety performance, how will such a relationship vary across different situations, and how the negative consequences of extremely high trait scores can be mitigated? To address these issues, the second objective of this study is to develop and test the theoretical argument that team-level supervisor support climate (SSC) (Bliese & Castro, 2000) moderates the curvilinear personality–safety performance relationship. 1.1. Curvilinear effects of personality on safety performance Neal and Griffin (2006) differentiated two types of safety performance behaviors: safety compliance (rule-complying behaviors) and safety participation (voluntary safety-related behaviors). Although personality traits were assumed to facilitate safety behaviors (Beus et al., 2015), increasing evidence indicates that all positive traits have benefits and costs, and the costs would begin to outweigh the benefits at high levels (Grant & Schwartz, 2011). Accordingly, we believe that personality traits are curvilinearly related to safety behaviors.
Corresponding author at: Institute of Psychology, Chinese Academy of Sciences, 16 Lincui Road, Chaoyang District, Beijing 100101, China. E-mail address: [email protected] (Y. Xu).
http://dx.doi.org/10.1016/j.paid.2017.10.005 Received 2 May 2017; Received in revised form 29 September 2017; Accepted 4 October 2017 0191-8869/ © 2017 Published by Elsevier Ltd.
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
may successfully execute safety behaviors given that they can cope with stress effectively and will not seek adventure. Taken together, personality facilitates safety behaviors primarily through motivational processes. However, the costs of personality outweigh the benefits at extremely high levels, thereby resulting in an overall pattern of curvilinearity.
1.1.1. Conscientiousness and safety performance Conscientiousness describes the extent to which individuals behave responsibly and cautiously, follow rules, and pursue high-order achievement goals (i.e., desire to fulfill obligations) (Barrick et al., 2013). Maintaining safety is a fundamental obligation of employees (Beus et al., 2015). Thus, working safely can aid in pursuing achievement goals. Supporting this view, Conscientiousness has been metaanalytically demonstrated to be positively associated with safety motivation and safety behaviors (Neal & Griffin, 2006). However, an extremely high Conscientiousness may reduce safety performance. First, highly conscientious individuals may be self-deceptive (Martocchio & Judge, 1997) and excessively focus on the tasks at hand (Judge & LePine, 2007). Both characteristics hinder safety performance. Second, extremely high Conscientiousness decreases wellbeing (Carter, Guan, Maples, Williamson, & Miller, 2016), which is a key antecedent of occupational injuries (Yuan, Li, & Lin, 2014). Moreover, high Conscientiousness may decrease safety participation because extremely high conscientious individuals consider going beyond their formal responsibilities (e.g., helping coworkers) inappropriate (Le et al., 2011).
Hypothesis 1. Conscientiousness (H1a), Emotional Stability (H1b), and Extraversion (H1c) are curvilinearly related to safety behaviors. 1.2. Moderating effect of SSC This study regards supervisor support as a contextual variable shared by group members (Bliese & Castro, 2000), and proposes that SSC influences curvilinear personality–safety performance relationships drawing upon theory of purposeful work behavior (Barrick et al., 2013). This theory posits that the primary mechanism underlying personality–performance relationship is motivation, and the motivational process will be triggered depending on opportunities to fulfill employees' goals in specific situations (Barrick et al., 2013). We argue that SSC provides cues congruent with the employees' goals, which in turn optimize motivational striving and maximize the benefits of high trait scores on safety performance. Accordingly, the points at which personality–safety performance relationships turn negative (i.e., inflection points) will occur at a higher level under high SSC than those under low SSC. Conscientious employees are motivated by achievement goals and tend to be engaged in work when they know how they are doing (Barrick et al., 2013). Supervisor feedback functions as an important cue regarding the effectiveness of performance to optimize the achievement striving process among highly conscientious employees under high SSC. By contrast, the positive effect of Conscientiousness on safety behavior will be weaker under low SSC. Emotionally stable employees exhibit a high desire for achievement and communion (Barrick et al., 2013). The benefits of Emotional Stability on safety performance can be maximized under high SSC because supervisor support can provide performance feedback and create a harmonious environment. Under low SSC, however, where feedback and opportunities to interact cooperatively with others are few, working safely cannot aid in accomplishing both goals. Thus, the positive effect of Emotional Stability on safety behavior is weaker. Extraverted employees are motivated to work by their desire to get ahead and gain rewards (Barrick et al., 2013). Performance feedback from supervisor can inform employees of their position within the status hierarchy, thereby fulfilling their intention to get ahead (Barrick et al., 2013). Moreover, supervisor support is an important antecedent of safety climate. Therefore, employees will perceive safety to be highly valued and will be rewarded under high SSC (Zohar & Polachek, 2014). Accordingly, the desire to strive for status and rewards of extraverted employees under high SSC can be fulfilled by achieving high safety performance, and the benefits of Extraversion on safety performance can be maximized. In sum, when employees are provided with opportunities to fulfill goals, the benefits of personality on safety performance are maximized, and the inflection points will occur at a higher level under high SSC.
1.1.2. Emotional Stability and safety performance Emotional Stability describes the extent to which individuals are calm, resilient, and effective when coping with stress (Barrick et al., 2013). Emotionally stable individuals display less negative emotions and are capable of coping with threatening situations, which facilitate safety performance. Moreover, inspired by a desire for achievement and communion, employees with high Emotional Stability are motivated to undertake safety activities given that working safely can aid in the accomplishment of both goals (Beus et al., 2015). However, the weak and inconsistent correlations between Emotional Stability and safety performance (e.g., Beus et al., 2015; Clarke & Robertson, 2005) imply a possible curvilinear relationship. First, individuals with excessively high Emotional Stability are oblivious to signs of threat (Widiger & Mullins-Sweatt, 2009); thus, they may be incapable of reacting to risks in the workplace. Second, extremely highly emotionally stable employees tend to underestimate the risks of unsafe behaviors because they are excessively optimistic (Anderson & Galinsky, 2006). In addition, extremely high Emotional Stability may hinder the ability of employees to connect with others for the characteristic of too calm (Le et al., 2011), thereby decreasing safety participation. 1.1.3. Extraversion and safety performance Extraversion describes the extent to which individuals are sociable, adventurous, ambitious, and reward-seeking (Barrick et al., 2013). The research findings on Extraversion and safety performance are conflicting (e.g., Beus et al., 2015; Christian et al., 2009); therefore, we infer that the relationship between Extraversion and safety performance may be nonlinear. First, the relationship between Extraversion and safety motivation are complex. Extraverts are motivated by a desire to get ahead of others and to strive for rewards (Barrick et al., 2013). In safety-priority industries where incentive mechanisms to promote safety behaviors are established (Zohar & Polachek, 2014), extraverted employees tend to work safely due to the desire to promoted. Consistently with it, Henning et al. (2009) found that Extraversion was positively correlated with safety attitude. However, this effect may diminish at extremely high levels due to high motivational conflicts between productivity and safety (Xu et al., 2014). In addition, according to Eysenck (1967)'s theory, extraverts are chronically under-stimulated, and thus they regularly seek adventure and perform poorly in vigilant task. By contrast, introverts are chronically overstimulated and attempt to avoid stimuli, and thus they perform poorly in tasks under stressful situations (Eysenck, 1967). Being more likely to operate near the optimal arousal level, ambiverts
Hypothesis 2. SSC moderates the curvilinear relationship between personality (i.e., Conscientiousness, Emotional Stability and Extraversion) and safety performance, such that inflection points are higher under high SSC than those under low SSC. 2. Method 2.1. Participants and procedures Participants were 430 male operators of nuclear power plants in 56
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
3. Results
China. They completed the questionnaire including items regarding personality, perceived supervisor support, and shift number, during their break time. Subsequently, direct supervisors rated their subordinates' safety behaviors. Overall, we collected matched data from 67 teams. Each team comprised 3 to 16 members, with an average size of 6.42 employees.
3.1. Confirmatory Factor Analysis (CFA) We conducted CFA to examine the validity of the measures. Hypothesized six-factor model fit well and the fit indices are better than those of the alternative models (details are available as supplementary material), thereby supporting discriminant validity.
2.2. Measures 3.2. Descriptive statistics and correlations
2.2.1. Personality measures and scoring Personality traits were measured via the Chinese version of the NEO five-factor personality inventory (Costa & McCrae, 1992). Participants rated three 12-item scales using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). In this study, testing curvilinearity and interactions would increase type I error (Carter et al., 2014). To reduce this potential problem, we used the factor analytic (FA) scoring method (i.e., scores were obtained by estimating a onefactor principal axis solution and taking the regression-based factor scores).
Descriptive statistics, coefficient alpha values, and intercorrelations among the variables are presented in Table 1. Both Emotional Stability and supervisor support were significantly related to safety performance. Conscientiousness and Extraversion were significantly related to safety compliance. 3.3. Hierarchical Linear Modeling (HLM) result
2.3. Validation of multilevel data structure
The between-team variance (τ00) and within-team variance (σ2) for safety performance were estimated according to the null model result. The ICCs for safety compliance (ICC = 0.496, τ00 = 0.487, σ2 = 0.495) and safety participation (ICC = 0.473, τ00 = 0.492, σ2 = 0.549) indicated large between-team variance, suggesting that between-team variance should be considered using multilevel analysis. All independent variables were standardized to reduce multicollinearity (Le et al., 2011), and the quadratic terms of personality were calculated based on the standardized values. Outcome variables were also standardized to control the size of the effects. Tables 2–4 present results of stepwise HLM analyses examining the effect of three personality traits on safety performance respectively. A significant effect of a negative quadratic term in step 2 would provide support for Hypotheses 1a–c; Hypothesis 2 would be supported if the interaction between personality traits and SSC or between the quadratic terms of personality traits and SSC is significant in step 3 since the inflection point is jointly determined by the coefficients of the linear and quadratic terms of personality traits (Le et al., 2011).
Supervisor support is a team-level variable that requires an aggregation of data collected from individual responses. To justify the aggregation, we calculated the rwg agreement index and the intraclass correlation coefficients (ICCs). The values of rwg ranged from 0.74 to 1.00; ICC (1) and ICC (2) were 0.25 and 0.72, respectively, which exceeded the recommended levels (LeBreton & Senter, 2008). Consequently, we calculated the mean scores of the ratings for supervisor support of the participants in each group as team-level SSC scores.
3.3.1. Conscientiousness and safety performance The quadratic effect of Conscientiousness in step 2 was significantly negative in predicting safety compliance (γ = − 0.044, p < 0.05) and safety participation (γ = − 0.057, p < 0.05) (see Table 2), supporting Hypothesis 1a. The interaction between Conscientiousness and SSC in step 3 was significant in predicting safety compliance (γ = 0.107, p < 0.01) and safety participation (γ = 0.122, p < 0.001). Simple slope analysis (Dawson, 2014) was conducted to compare the
2.2.2. Perceived supervisor support The four-item supervisor support scale (Cheng, Luh, & Guo, 2003) was used to assess perceived supervisor support. Each item was scored using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).
2.2.3. Safety performance Each component of safety behaviors was measured using three items (Yuan et al., 2014). Each item was scored on a five-point Likert scale ranging from 1 (nearly never) to 5 (nearly always). Items about perceived supervisor support and safety performance are available in the Appendix.
Table 1 Means, Standard Deviations, and correlations among variables. Variable 1. Age 2. Job tenure 3. Conscientiousness 4. Emotional Stability 5. Extraversion 6. Conscientiousness (FA) 7. Emotional Stability (FA) 8. Extraversion (FA) 9. Supervisor support 10. Safety compliance 11. Safety participation
M
SD
1
2
3
4
5
6
7
8
9
10
11
29.88 7.10 3.95 3.71 3.63 0.00 0.00 0.00 3.90 4.70 4.21
2.53 2.64 0.40 0.55 0.47 1.00 1.00 1.00 0.55 0.40 0.68
− 0.85⁎⁎⁎ − 0.05 − 0.06 − 0.13⁎⁎ − 0.06 − 0.07 − 0.14⁎⁎ − 0.12⁎ 0.03 0.22⁎⁎⁎
− − 0.05 − 0.08 − 0.12⁎ − 0.05 − 0.09 − 0.12⁎ − 0.13⁎⁎ 0.00 0.20⁎⁎⁎
0.78 0.52⁎⁎⁎ 0.47⁎⁎⁎ 0.96⁎⁎⁎ 0.52⁎⁎⁎ 0.48⁎⁎⁎ 0.27⁎⁎⁎ 0.18⁎⁎⁎ 0.08†
0.82 0.42⁎⁎⁎ 0.43⁎⁎⁎ 0.99⁎⁎⁎ 0.44⁎⁎⁎ 0.20⁎⁎⁎ 0.21⁎⁎⁎ 0.13⁎⁎
0.76 0.49⁎⁎⁎ 0.43⁎⁎⁎ 0.98⁎⁎⁎ 0.31⁎⁎⁎ 0.14⁎⁎ 0.07
− 0.44⁎⁎⁎ 0.51⁎⁎⁎ 0.28⁎⁎⁎ 0.18⁎⁎⁎ 0.09†
− 0.46⁎⁎⁎ 0.21⁎⁎⁎ 0.21⁎⁎⁎ 0.15⁎⁎
− 0.33⁎⁎⁎ 0.14⁎⁎ 0.06
0.85 0.11⁎ 0.10⁎
0.78 0.52⁎⁎⁎
0.87
Note: N = 430 employees in 67 teams. Reliability estimates (coefficient alphas) of scales are reported in italics on the diagonal. † P < 0.1. ⁎ P < 0.05. ⁎⁎ P < 0.01. ⁎⁎⁎ P < 0.001.
57
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
Table 2 Hierarchical linear modeling results for Conscientiousness. Variable
Safety compliance Step 1
Level 1 Age Job tenure C C2 Level 2 Support Support × C Support × C2 σ2 τ00 Pseudo R2
Safety participation Step 2
Step 3
Step 1
0.091(0.069) 0.059 (0.064) 0.115⁎⁎(0.043)
Step 3
0.009(0.059) 0.188⁎⁎(0.067) 0.034(0.041) − 0.044⁎(0.021)
−0.057⁎(0.027)
0.264⁎⁎⁎(0.077)
0.469 0.444 0.062
Step 2
0.206⁎(0.096)
0.466 0.439 0.069
0.107⁎⁎(0.038) 0.004(0.032) 0.430 0.509 0.076
0.531 0.437 0.077
0.525 0.429 0.093
0.122⁎⁎⁎(0.032) 0.009(0.035) 0.469 0.411 0.110
Note: N = 430 employees in 67 teams. The term in brackets is the robust standard error. C = Conscientiousness, C2 = the quadratic term of Conscientiousness. Hereafter, σ2 = within-team variance, τ00 = intercept variance, Pseudo R2 is obtained using R code available in Aguinis, Gottfredson, & Culpepper, 2013. ⁎ P < 0.05 ⁎⁎ P < 0.01 ⁎⁎⁎ P < 0.001
curvilinear effects and the inflection points under low SSC (−1.00 SD) to those under high SSC (+1.00 SD). Although the curvilinear relationships between Conscientiousness and safety performance hold under various SSC levels, the inflection points under low SSC (safety compliance = 0.27 SD; safety participation = − 0.76 SD) were lower than those under high SSC (safety compliance = 2.22 SD; safety participation = 1.29 SD). The plots (following Dawson, 2014) of the quadratic interactions were shown Fig. 1a–b.
relationships were significant (safety compliance: b = −0.112, p < 0.01; safety participation: b = −0.176, p < 0.001) under low SSC, and the inflection points occurred at − 0.16 SD for safety compliance and −0.36 SD for safety participation. By contrast, no curvilinear relationships (safety compliance: b = 0.012, ns.; safety participation: b = 0.022, ns.), but positively linear relationships existed (safety compliance: b = 0.188, p < 0.001; safety participation: b = 0.222, p < 0.001) under high SSC (see Fig. 1c–d).
3.3.2. Emotional Stability and safety performance The quadratic effect of Emotional Stability in step 2 significantly predicted safety compliance (γ = − 0.045, p < 0.05) and safety participation (γ = − 0.052, p < 0.05) (see Table 3), supporting Hypothesis 1b. In Step 3, the interaction between Emotional Stability and SSC significantly predicted safety compliance (γ = 0.112, p < 0.05) and safety participation (γ = 0.174, p < 0.001), and the interaction between quadratic term of Emotional Stability and SSC significantly predicted safety compliance (γ = 0.062, p < 0.05) and safety participation (γ = 0.099, p < 0.001). Simple slope test showed that the curvilinear Emotional Stability–safety performance
3.3.3. Extraversion and safety performance The quadratic effect of Emotional Stability in step 2 significantly predicted safety compliance (γ = − 0.057, p < 0.001) and safety participation (γ = −0.048, p < 0.05) (see Table 4), supporting Hypothesis 1c. The moderating effects of SSC on the quadratic term of Extraversion–safety compliance relationship (γ = 0.038, p < 0.05) and on Extraversion–safety participation relationship (γ = 0.103, p < 0.05) were significant. Specifically, the quadratic term of Extraversion significantly predicted safety performance under low SSC (safety compliance: b = − 0.095, p < 0.001; safety participation: b = − 0.102, p < 0.05); whereas, the curvilinear effects of
Table 3 Hierarchical linear modeling results for Emotional Stability. Safety compliance Step 1 Level 1 Age Job tenure ES ES2 Level 2 Support Support × ES Support × ES2 σ2 τ00 pseudo R2
Safety participation Step 2
Step 3
0.084(0.069) 0.070(0.063) 0.111⁎⁎⁎(0.034)
Step 1
Step 3
0.010(0.059) 0.190⁎⁎(0.067) 0.084⁎(0.041) −0.045⁎(0.023)
− 0.052⁎(0.025)
0.278⁎⁎⁎(0.076)
0.471 0.432 0.071
Step 2
0.202⁎(0.092)
0.468 0.430 0.080
0.112⁎(0.052) 0.062⁎(0.028) 0.455 0.456 0.088
Note: N = 430 employees in 67 teams. The term in brackets is the robust standard error. ES = Emotional Stability, ES2 = the quadratic term of Emotional Stability. ⁎ P < 0.05 ⁎⁎ P < 0.01 ⁎⁎⁎ P < 0.001
58
0.528 0.425 0.093
0.525 0.411 0.111
0.174⁎⁎⁎(0.044) 0.099⁎⁎⁎(0.028) 0.494 0.447 0.149
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
Table 4 Hierarchical linear modeling results for Extraversion. Safety compliance Step 1 Level 1 Age Job tenure E E2 Level 2 Support Support × E Support × E2 σ2 τ00 pseudo R2
Safety participation Step 2
Step 3
Step 1
0.086(0.068) 0.070(0.062) 0.057(0.040)
Step 3
0.010(0.059) 0.189⁎⁎(0.066) 0.040(0.036) −0.057⁎⁎⁎(0.016)
− 0.048⁎(0.024)
0.282⁎⁎⁎(0.078)
0.478 0.448 0.052
Step 2
0.207⁎(0.093)
0.474 0.433 0.072
0.064(0.062) 0.038⁎(0.017) 0.460 0.400 0.073
0.531 0.437 0.077
0.529 0.426 0.092
0.103⁎(0.041) 0.046(0.031) 0.517 0.407 0.091
Note: N = 430 employees in 67 teams. The term in brackets is the robust standard error. E = Extraversion, E2 = the quadratic term of Extraversion. ⁎ P < 0.05 ⁎⁎ P < 0.01 ⁎⁎⁎ P < 0.001
Extraversion on safety compliance (b = −0.020, ns.) and safety participation (b = − 0.010, ns.) were insignificant, and Extraversion marginally significantly predicted safety compliance (b = 0.093, p = 0.08) and significantly predicted safety participation (b = 0.123, p < 0.05) under high SSC (Fig. 1e–1f). Inflection points under low SSC (safety compliance = − 0.19 SD; safety participation = −0.41 SD) were lower than those under high SSC (safety compliance = 2.38 SD; safety participation = 5.98 SD). Taken together, the results for Conscientiousness and Extraversion supported Hypothesis 2. As for Emotional Stability, their relationship is linearly positive under high SSC, and the inflection points are insignificant. These results are consistent with our theoretical reasoning that SSC optimizes the benefits of Emotional Stability, and a high level of Emotional Stability is associated with high safety performance under high SSC.
4.1. Theoretical implications This study expands curvilinear personality–performance relationship to the occupational safety domain and suggests that both extremely low and high personality levels may impair safety performance. In particular, for safety compliance, the optimal level of Conscientiousness (1.33 SD) was higher than those of Emotional Stability (0.99 SD) and Extraversion (0.47 SD), suggesting that the weak and inconsistent linear personality–safety compliance relations in previous research (Christian et al., 2009; Clarke & Robertson, 2005) do not imply that Emotional Stability and Extraversion have no effects on safety compliance; rather, their relationships are curvilinear. For safety participation, the optimal level of Emotional Stability (0.56 SD) was the highest and it was followed by Extraversion (0.39 SD) and Conscientiousness (0.33 SD) because moderated high level of Emotional Stability could enhance employees' desire for communion (Barrick et al., 2013). Furthermore, our results depict a complete picture of personality–safety performance relationship under different SSC levels. Under low SSC, higher personality levels are initially associated with higher safety performance levels. However, after inflection points occur, higher personality levels are associated with lower safety performance levels. Under high SSC, the inflection points occur at a higher level or become insignificant, and higher personality levels are generally associated with higher safety performance levels. These results indicate that SSC optimizes the benefits of personality traits on performance (Barrick et al., 2013), thus the costs of extremely high trait scores can be mitigated under high SSC. These results also suggest that curvilinear personality–performance relationship is highly sensitive to situation, which may have an implication for the elusiveness of their curvilinear relationship (Converse & Oswald, 2014; Uppal, 2017). In our results, both shape (i.e., linear or nonlinear) and inflection points vary considerably across different contexts. Accordingly, possible situational moderators should be considered in future research to determine the accurate curvilinear personality–performance relationship.
4. Discussion Our research advances the field by presenting the curvilinear effects of personality on safety performance and the moderating effects of team-level SSC. To rule out the possibility that these curvilinear effects were due to the skewness of safety performance, the skewness values for safety performance were calculated and compared with the recommended threshold of slightly nonnormal ( ± 1.00) (Lei & Lomax, 2005). Results suggested that the skewness for safety participation (− 0.74) was not a serious issue. Although skewness for safety compliance (− 1.13) slightly exceeds the criterion, this distribution is actually normal among safety-priority occupations, which could be found in previous study (e.g., Martínez-Córcoles, Gracia, Tomás, Peiró, & Schöbel, 2013; Xu et al., 2014). To further clarify these curvilinear effects, correlation analysis between two personality levels (below the inflection point vs. above the inflection point) and safety performance under different SSC levels (below the mean vs. above the mean) was conducted. Under low SSC levels, personality–safety performance correlations were positive at low levels of personality and turned negative at high levels of personality; under high SSC levels, personality–safety performance correlations were positive at low levels of personality and were weaker or turned weakly negative at high levels of personality. This indicated that the curvilinear relationships and interactions indeed existed, unlike the ceiling effects that would be created by skewness (details are available as supplementary materials).
4.2. Practical implications Our findings present important practical implications for personnel selection and interventions. First, to screen out accident-prone applicants, personnel selection can benefit by using two cutoff points instead of top-down selection. Second, the results show that high trait scores can be harmful to safety performance under low SSC, whereas high trait scores lead to high safety performance under high SSC, thereby 59
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
Fig. 1. Cross-level moderating effect of SSC on the curvilinear personality–safety performance relationships.
2014). Furthermore, when implementing a personality development program, the optimal level for personality requirement of a particular job position should be identified, rather than simply improving socially desirable personality.
indicating that enhancing SSC can mitigate the negative effects of extremely high trait scores on safety performance. In particular, supervisors should provide employees with trait-relevant cues (e.g., performance feedback, praise for safety performance; Zohar & Polachek, 60
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
4.3. Limitations and future directions
References
This study has several limitations that should be addressed in future research. First, the cross-sectional design limited causal interpretations. Future experimental and longitudinal designs can help in further investigations. Second, only one sample is used, which may limit the generalizability of our findings. Thus, the pattern of results should be evidenced in other organizations. Finally, our study develops theoretical arguments related to the possible mechanism of curvilinear personality–performance relationship and the moderating effects of SSC, thereby suggesting potential mediators (e.g., motivation). However, we did not measure the potential mediators directly. Thus, we suggest further study to precisely support our theory by exploring possible mediating mechanisms.
Aguinis, H., Gottfredson, R. K., & Culpepper, S. A. (2013). Best-practice recommendations for estimating cross-level interaction effects using multilevel modeling. Journal of Management, 39, 1490–1528. Anderson, C., & Galinsky, A. D. (2006). Power, optimism, and risk-taking. European Journal of Social Psychology, 36, 511–536. Barrick, M. R., Mount, M. K., & Li, N. (2013). The theory of purposeful work behavior: The role of personality, higher-order goals, and job characteristics. Academy of Management Review, 38, 132–153. Beus, J. M., Dhanani, L. Y., & McCord, M. A. (2015). A meta-analysis of personality and workplace safety: Addressing unanswered questions. Journal of Applied Psychology, 100, 481–498. Bliese, P. D., & Castro, C. A. (2000). Role clarity, work overload and organizational support: Multilevel evidence of the importance of support. Work and Stress, 14, 65–73. Carter, N. T., Dalal, D. K., Boyce, A. S., O'Connell, M. S., Kung, M. C., & Delgado, K. M. (2014). Uncovering curvilinear relationships between conscientiousness and job performance: How theoretically appropriate measurement makes an empirical difference. Journal of Applied Psychology, 99, 564–586. Carter, N. T., Guan, L., Maples, J. L., Williamson, R. L., & Miller, J. D. (2016). The downsides of extreme conscientiousness for psychological well-being: The role of obsessive compulsive tendencies. Journal of Personality, 84, 510–522. Cheng, Y., Luh, W. M., & Guo, Y. L. (2003). Reliability and validity of the Chinese version of the job content questionnaire in Taiwanese workers. International Journal of Behavioral Medicine, 10, 15–30. 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. Journal of Applied Psychology, 94, 1103–1127. Clarke, S., & Robertson, I. (2005). A meta-analytic review of the Big Five personality factors and accident involvement in occupational and non-occupational settings. Journal of Occupational and Organizational Psychology, 78, 355–376. Converse, P. D., & Oswald, F. L. (2014). Thinking ahead: Assuming linear versus nonlinear personality-criterion relationships in personnel selection. Human Performance, 27, 61–79. Costa, P., & McCrae, R. (1992). Neo PI-R professional manual. Odessa, FL: Psychological Assessment Resources. Dawson, J. F. (2014). Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29, 1–19. Eysenck, H. J. (1967). The biological basis of personality. Springfield, IL: Charles C. Thomas. Grant, A. M., & Schwartz, B. (2011). Too much of a good thing: The challenge and opportunity of the inverted U. Perspectives on Psychological Science, 6, 61–76. Hansen, C. P. (1988). Personality characteristics of the accident involved employee. Journal of Business and Psychology, 2, 346–365. Henning, J. B., Stufft, C. J., Payne, S. C., Bergman, M. E., Mannan, M. S., & Keren, N. (2009). The influence of individual differences on organizational safety attitudes. Safety Science, 47, 337–345. Judge, T. A., & LePine, J. A. (2007). The bright and dark sides of personality: Implications for personnel selection and team contexts. In J. Langan-Fox, C. Cooper, & R. Klimoski (Eds.). Research companion to the dysfunctional workplace: Management challenges and symptoms (pp. 332–355). Cheltenham, England: Elgar. Le, H., Oh, I. S., Robbins, S. B., Ilies, R., Holland, E., & Westrick, P. (2011). Too much of a good thing: Curvilinear relationships between personality traits and job performance. Journal of Applied Psychology, 96, 113–133. LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods, 11, 815–852. Lei, M., & Lomax, R. G. (2005). The effect of varying degrees of nonnormality in structural equation modeling. Structural Equation Modeling, 12, 1–27. Martínez-Córcoles, M., Gracia, F. J., Tomás, I., Peiró, J. M., & Schöbel, M. (2013). Empowering team leadership and safety performance in nuclear power plants: A multilevel approach. Safety Science, 51, 293–301. Martocchio, J. J., & Judge, T. A. (1997). Relationship between conscientiousness and learning in employee training: Mediating influences of self-deception and self-efficacy. Journal of Applied Psychology, 82, 764–773. 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. Journal of Applied Psychology, 91, 946–953. Postlethwaite, B., Robbins, S., Rickerson, J., & McKinniss, T. (2009). The moderation of conscientiousness by cognitive ability when predicting workplace safety behavior. Personality and Individual Differences, 47, 711–716. Uppal, N. (2017). Moderation effects of perceived organisational support on curvilinear relationship between neuroticism and job performance. Personality and Individual Differences, 105, 47–53. Widiger, T. A., & Mullins-Sweatt, S. N. (2009). Five-factor model of personality disorder: A proposal for DSM-V. Annual Review of Clinical Psychology, 5, 197–220. Xu, Y., Li, Y., Wang, G., Yuan, X., Ding, W., & Shen, Z. (2014). Attentional bias toward safety predicts safety behaviors. Accident Analysis & Prevention, 71, 144–153. Yuan, Z., Li, Y., & Lin, J. (2014). Linking challenge and hindrance stress to safety performance: The moderating effect of core self-evaluation. Personality and Individual Differences, 68, 154–159. Zohar, D., & Polachek, T. (2014). Discourse-based intervention for modifying supervisory communication as leverage for safety climate and performance improvement: A randomized field study. Journal of Applied Psychology, 99, 113–124.
5. Conclusion The results demonstrate curvilinear personality–safety performance relationship, and suggest that a supervisor supportive environment mitigates the negative effects of extremely high personality traits on safety performance. These findings provide insight into the nature of personality–safety performance relationships and curvilinear person–situation interaction. This study also suggests that organizations should use the method involving two cutoff points in personnel selection and conduct interventions concerning supervisor support in safety management. Acknowledgements This study was partly supported by National Natural Science Foundation of China (Grant Numbers: 71371179 and 71501177) and Scientific Foundation of Institute of Psychology, Chinese Academy of Sciences (Y5CX143005). And also, we would like to thank the editor and the anonymous reviewers for their helpful comments and insightful feedback. Appendix A. Items of perceived supervisor support and safety performance used in the study Perceived supervisor support 1. 2. 3. 4.
My My My My
supervisor supervisor supervisor supervisor
is concerned about the welfare of those under him. pays attention to what I am saying. is helpful in getting the job done. is successful in getting people to work together.
Safety compliance 1. He uses all the necessary safety equipment to do his job. 2. He uses the correct safety procedures for carrying out his job. 3. He ensures the highest levels of safety when he carries out his job. Safety participation 1. He promotes the safety program within the organization. 2. He puts in extra effort to improve the safety of the workplace. 3. He voluntarily carries out tasks or activities that help to improve workplace safety. Appendix B. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.paid.2017.10.005.
61
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
Curvilinear effects of personality on safety performance: The moderating role of supervisor support
MARK
Xiao Yuana,b, Yongjuan Lia,b, Yaoshan Xua,b,⁎, Naixi Huangc a b c
CAS Key Laboratory of Behavioral Science, Institute of Psychology, Chinese Academy of Sciences, 16 LinCui Road, Chaoyang District, Beijing 100101, China Department of Psychology, University of Chinese Academy of Sciences, 19(A) Yuquan Road, Shijingshan District, Beijing 100049, China Research Institute of Nuclear Power Operation of China, 1021 National Road, East Lake High Tech Zone, Wuhan 430223, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Personality Curvilinear relationship Safety performance Supervisor support climate
Personality has long been assumed to be linearly associated with safety performance. Focusing on the benefits and costs of personality traits, this study investigates the curvilinear effects of personality traits (i.e., Conscientiousness, Emotional Stability and Extraversion) on safety performance. Results support our hypothesized curvilinear effects that their relationships are initially positive but become negative as personality trait scores increase to extremely high. Furthermore, drawing on theory of purposeful work behavior, this study proposes and finds that the curvilinear effects of personality traits on safety performance are moderated by teamlevel supervisor support climate, such that the inflection points after which the personality–performance relationship becomes negative are higher under high levels of supervisor supportive environment than those under low levels of such environment. Implications for research and practice, as well as limitations and directions for future research, are provided.
1. Introduction Workplace safety is an increasing concern worldwide. To ensure high levels of safety performance, considerable research has focused on the role of personality (e.g., Beus, Dhanani, & McCord, 2015). Early studies have shown that Extraversion and Neuroticism were weakly positively associated with accidents within the framework of Eysenck's theory (e.g., Hansen, 1988). With the emergence of the “Big Five” personality theory, the positive role of Conscientiousness in predicting safety behavior was observed (Postlethwaite, Robbins, Rickerson, & McKinniss, 2009). Christian, Bradley, Wallace, and Burke (2009) summarized the importance of three personality traits, namely Conscientiousness, Emotional Stability and Extraversion, to safety performance in organizational settings. However, the relatively weak or equivocal correlations remain as issues (e.g., Beus et al., 2015). A promising idea is that personality–safety performance relationship may be nonlinear instead of linear. Growing researches have begun to examine curvilinear personality–performance relationship (e.g., Le et al., 2011). However, minimal attention has been given to safety performance. Although Carter et al. (2014) has provided initial evidence for curvilinear Conscientiousness–safety performance relationship, the theoretical framework for this relationship remains insufficient. In addition, Emotional Stability and Extraversion should be
⁎
further explored (Christian et al., 2009). Therefore, this study aims to explore the curvilinear effects of Conscientiousness, Emotional Stability and Extraversion on safety performance. Moreover, the effects of personality on performance vary across different situations (Barrick, Mount, & Li, 2013). If a curvilinear relationship exists between personality and safety performance, how will such a relationship vary across different situations, and how the negative consequences of extremely high trait scores can be mitigated? To address these issues, the second objective of this study is to develop and test the theoretical argument that team-level supervisor support climate (SSC) (Bliese & Castro, 2000) moderates the curvilinear personality–safety performance relationship. 1.1. Curvilinear effects of personality on safety performance Neal and Griffin (2006) differentiated two types of safety performance behaviors: safety compliance (rule-complying behaviors) and safety participation (voluntary safety-related behaviors). Although personality traits were assumed to facilitate safety behaviors (Beus et al., 2015), increasing evidence indicates that all positive traits have benefits and costs, and the costs would begin to outweigh the benefits at high levels (Grant & Schwartz, 2011). Accordingly, we believe that personality traits are curvilinearly related to safety behaviors.
Corresponding author at: Institute of Psychology, Chinese Academy of Sciences, 16 Lincui Road, Chaoyang District, Beijing 100101, China. E-mail address: [email protected] (Y. Xu).
http://dx.doi.org/10.1016/j.paid.2017.10.005 Received 2 May 2017; Received in revised form 29 September 2017; Accepted 4 October 2017 0191-8869/ © 2017 Published by Elsevier Ltd.
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
may successfully execute safety behaviors given that they can cope with stress effectively and will not seek adventure. Taken together, personality facilitates safety behaviors primarily through motivational processes. However, the costs of personality outweigh the benefits at extremely high levels, thereby resulting in an overall pattern of curvilinearity.
1.1.1. Conscientiousness and safety performance Conscientiousness describes the extent to which individuals behave responsibly and cautiously, follow rules, and pursue high-order achievement goals (i.e., desire to fulfill obligations) (Barrick et al., 2013). Maintaining safety is a fundamental obligation of employees (Beus et al., 2015). Thus, working safely can aid in pursuing achievement goals. Supporting this view, Conscientiousness has been metaanalytically demonstrated to be positively associated with safety motivation and safety behaviors (Neal & Griffin, 2006). However, an extremely high Conscientiousness may reduce safety performance. First, highly conscientious individuals may be self-deceptive (Martocchio & Judge, 1997) and excessively focus on the tasks at hand (Judge & LePine, 2007). Both characteristics hinder safety performance. Second, extremely high Conscientiousness decreases wellbeing (Carter, Guan, Maples, Williamson, & Miller, 2016), which is a key antecedent of occupational injuries (Yuan, Li, & Lin, 2014). Moreover, high Conscientiousness may decrease safety participation because extremely high conscientious individuals consider going beyond their formal responsibilities (e.g., helping coworkers) inappropriate (Le et al., 2011).
Hypothesis 1. Conscientiousness (H1a), Emotional Stability (H1b), and Extraversion (H1c) are curvilinearly related to safety behaviors. 1.2. Moderating effect of SSC This study regards supervisor support as a contextual variable shared by group members (Bliese & Castro, 2000), and proposes that SSC influences curvilinear personality–safety performance relationships drawing upon theory of purposeful work behavior (Barrick et al., 2013). This theory posits that the primary mechanism underlying personality–performance relationship is motivation, and the motivational process will be triggered depending on opportunities to fulfill employees' goals in specific situations (Barrick et al., 2013). We argue that SSC provides cues congruent with the employees' goals, which in turn optimize motivational striving and maximize the benefits of high trait scores on safety performance. Accordingly, the points at which personality–safety performance relationships turn negative (i.e., inflection points) will occur at a higher level under high SSC than those under low SSC. Conscientious employees are motivated by achievement goals and tend to be engaged in work when they know how they are doing (Barrick et al., 2013). Supervisor feedback functions as an important cue regarding the effectiveness of performance to optimize the achievement striving process among highly conscientious employees under high SSC. By contrast, the positive effect of Conscientiousness on safety behavior will be weaker under low SSC. Emotionally stable employees exhibit a high desire for achievement and communion (Barrick et al., 2013). The benefits of Emotional Stability on safety performance can be maximized under high SSC because supervisor support can provide performance feedback and create a harmonious environment. Under low SSC, however, where feedback and opportunities to interact cooperatively with others are few, working safely cannot aid in accomplishing both goals. Thus, the positive effect of Emotional Stability on safety behavior is weaker. Extraverted employees are motivated to work by their desire to get ahead and gain rewards (Barrick et al., 2013). Performance feedback from supervisor can inform employees of their position within the status hierarchy, thereby fulfilling their intention to get ahead (Barrick et al., 2013). Moreover, supervisor support is an important antecedent of safety climate. Therefore, employees will perceive safety to be highly valued and will be rewarded under high SSC (Zohar & Polachek, 2014). Accordingly, the desire to strive for status and rewards of extraverted employees under high SSC can be fulfilled by achieving high safety performance, and the benefits of Extraversion on safety performance can be maximized. In sum, when employees are provided with opportunities to fulfill goals, the benefits of personality on safety performance are maximized, and the inflection points will occur at a higher level under high SSC.
1.1.2. Emotional Stability and safety performance Emotional Stability describes the extent to which individuals are calm, resilient, and effective when coping with stress (Barrick et al., 2013). Emotionally stable individuals display less negative emotions and are capable of coping with threatening situations, which facilitate safety performance. Moreover, inspired by a desire for achievement and communion, employees with high Emotional Stability are motivated to undertake safety activities given that working safely can aid in the accomplishment of both goals (Beus et al., 2015). However, the weak and inconsistent correlations between Emotional Stability and safety performance (e.g., Beus et al., 2015; Clarke & Robertson, 2005) imply a possible curvilinear relationship. First, individuals with excessively high Emotional Stability are oblivious to signs of threat (Widiger & Mullins-Sweatt, 2009); thus, they may be incapable of reacting to risks in the workplace. Second, extremely highly emotionally stable employees tend to underestimate the risks of unsafe behaviors because they are excessively optimistic (Anderson & Galinsky, 2006). In addition, extremely high Emotional Stability may hinder the ability of employees to connect with others for the characteristic of too calm (Le et al., 2011), thereby decreasing safety participation. 1.1.3. Extraversion and safety performance Extraversion describes the extent to which individuals are sociable, adventurous, ambitious, and reward-seeking (Barrick et al., 2013). The research findings on Extraversion and safety performance are conflicting (e.g., Beus et al., 2015; Christian et al., 2009); therefore, we infer that the relationship between Extraversion and safety performance may be nonlinear. First, the relationship between Extraversion and safety motivation are complex. Extraverts are motivated by a desire to get ahead of others and to strive for rewards (Barrick et al., 2013). In safety-priority industries where incentive mechanisms to promote safety behaviors are established (Zohar & Polachek, 2014), extraverted employees tend to work safely due to the desire to promoted. Consistently with it, Henning et al. (2009) found that Extraversion was positively correlated with safety attitude. However, this effect may diminish at extremely high levels due to high motivational conflicts between productivity and safety (Xu et al., 2014). In addition, according to Eysenck (1967)'s theory, extraverts are chronically under-stimulated, and thus they regularly seek adventure and perform poorly in vigilant task. By contrast, introverts are chronically overstimulated and attempt to avoid stimuli, and thus they perform poorly in tasks under stressful situations (Eysenck, 1967). Being more likely to operate near the optimal arousal level, ambiverts
Hypothesis 2. SSC moderates the curvilinear relationship between personality (i.e., Conscientiousness, Emotional Stability and Extraversion) and safety performance, such that inflection points are higher under high SSC than those under low SSC. 2. Method 2.1. Participants and procedures Participants were 430 male operators of nuclear power plants in 56
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
3. Results
China. They completed the questionnaire including items regarding personality, perceived supervisor support, and shift number, during their break time. Subsequently, direct supervisors rated their subordinates' safety behaviors. Overall, we collected matched data from 67 teams. Each team comprised 3 to 16 members, with an average size of 6.42 employees.
3.1. Confirmatory Factor Analysis (CFA) We conducted CFA to examine the validity of the measures. Hypothesized six-factor model fit well and the fit indices are better than those of the alternative models (details are available as supplementary material), thereby supporting discriminant validity.
2.2. Measures 3.2. Descriptive statistics and correlations
2.2.1. Personality measures and scoring Personality traits were measured via the Chinese version of the NEO five-factor personality inventory (Costa & McCrae, 1992). Participants rated three 12-item scales using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). In this study, testing curvilinearity and interactions would increase type I error (Carter et al., 2014). To reduce this potential problem, we used the factor analytic (FA) scoring method (i.e., scores were obtained by estimating a onefactor principal axis solution and taking the regression-based factor scores).
Descriptive statistics, coefficient alpha values, and intercorrelations among the variables are presented in Table 1. Both Emotional Stability and supervisor support were significantly related to safety performance. Conscientiousness and Extraversion were significantly related to safety compliance. 3.3. Hierarchical Linear Modeling (HLM) result
2.3. Validation of multilevel data structure
The between-team variance (τ00) and within-team variance (σ2) for safety performance were estimated according to the null model result. The ICCs for safety compliance (ICC = 0.496, τ00 = 0.487, σ2 = 0.495) and safety participation (ICC = 0.473, τ00 = 0.492, σ2 = 0.549) indicated large between-team variance, suggesting that between-team variance should be considered using multilevel analysis. All independent variables were standardized to reduce multicollinearity (Le et al., 2011), and the quadratic terms of personality were calculated based on the standardized values. Outcome variables were also standardized to control the size of the effects. Tables 2–4 present results of stepwise HLM analyses examining the effect of three personality traits on safety performance respectively. A significant effect of a negative quadratic term in step 2 would provide support for Hypotheses 1a–c; Hypothesis 2 would be supported if the interaction between personality traits and SSC or between the quadratic terms of personality traits and SSC is significant in step 3 since the inflection point is jointly determined by the coefficients of the linear and quadratic terms of personality traits (Le et al., 2011).
Supervisor support is a team-level variable that requires an aggregation of data collected from individual responses. To justify the aggregation, we calculated the rwg agreement index and the intraclass correlation coefficients (ICCs). The values of rwg ranged from 0.74 to 1.00; ICC (1) and ICC (2) were 0.25 and 0.72, respectively, which exceeded the recommended levels (LeBreton & Senter, 2008). Consequently, we calculated the mean scores of the ratings for supervisor support of the participants in each group as team-level SSC scores.
3.3.1. Conscientiousness and safety performance The quadratic effect of Conscientiousness in step 2 was significantly negative in predicting safety compliance (γ = − 0.044, p < 0.05) and safety participation (γ = − 0.057, p < 0.05) (see Table 2), supporting Hypothesis 1a. The interaction between Conscientiousness and SSC in step 3 was significant in predicting safety compliance (γ = 0.107, p < 0.01) and safety participation (γ = 0.122, p < 0.001). Simple slope analysis (Dawson, 2014) was conducted to compare the
2.2.2. Perceived supervisor support The four-item supervisor support scale (Cheng, Luh, & Guo, 2003) was used to assess perceived supervisor support. Each item was scored using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).
2.2.3. Safety performance Each component of safety behaviors was measured using three items (Yuan et al., 2014). Each item was scored on a five-point Likert scale ranging from 1 (nearly never) to 5 (nearly always). Items about perceived supervisor support and safety performance are available in the Appendix.
Table 1 Means, Standard Deviations, and correlations among variables. Variable 1. Age 2. Job tenure 3. Conscientiousness 4. Emotional Stability 5. Extraversion 6. Conscientiousness (FA) 7. Emotional Stability (FA) 8. Extraversion (FA) 9. Supervisor support 10. Safety compliance 11. Safety participation
M
SD
1
2
3
4
5
6
7
8
9
10
11
29.88 7.10 3.95 3.71 3.63 0.00 0.00 0.00 3.90 4.70 4.21
2.53 2.64 0.40 0.55 0.47 1.00 1.00 1.00 0.55 0.40 0.68
− 0.85⁎⁎⁎ − 0.05 − 0.06 − 0.13⁎⁎ − 0.06 − 0.07 − 0.14⁎⁎ − 0.12⁎ 0.03 0.22⁎⁎⁎
− − 0.05 − 0.08 − 0.12⁎ − 0.05 − 0.09 − 0.12⁎ − 0.13⁎⁎ 0.00 0.20⁎⁎⁎
0.78 0.52⁎⁎⁎ 0.47⁎⁎⁎ 0.96⁎⁎⁎ 0.52⁎⁎⁎ 0.48⁎⁎⁎ 0.27⁎⁎⁎ 0.18⁎⁎⁎ 0.08†
0.82 0.42⁎⁎⁎ 0.43⁎⁎⁎ 0.99⁎⁎⁎ 0.44⁎⁎⁎ 0.20⁎⁎⁎ 0.21⁎⁎⁎ 0.13⁎⁎
0.76 0.49⁎⁎⁎ 0.43⁎⁎⁎ 0.98⁎⁎⁎ 0.31⁎⁎⁎ 0.14⁎⁎ 0.07
− 0.44⁎⁎⁎ 0.51⁎⁎⁎ 0.28⁎⁎⁎ 0.18⁎⁎⁎ 0.09†
− 0.46⁎⁎⁎ 0.21⁎⁎⁎ 0.21⁎⁎⁎ 0.15⁎⁎
− 0.33⁎⁎⁎ 0.14⁎⁎ 0.06
0.85 0.11⁎ 0.10⁎
0.78 0.52⁎⁎⁎
0.87
Note: N = 430 employees in 67 teams. Reliability estimates (coefficient alphas) of scales are reported in italics on the diagonal. † P < 0.1. ⁎ P < 0.05. ⁎⁎ P < 0.01. ⁎⁎⁎ P < 0.001.
57
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
Table 2 Hierarchical linear modeling results for Conscientiousness. Variable
Safety compliance Step 1
Level 1 Age Job tenure C C2 Level 2 Support Support × C Support × C2 σ2 τ00 Pseudo R2
Safety participation Step 2
Step 3
Step 1
0.091(0.069) 0.059 (0.064) 0.115⁎⁎(0.043)
Step 3
0.009(0.059) 0.188⁎⁎(0.067) 0.034(0.041) − 0.044⁎(0.021)
−0.057⁎(0.027)
0.264⁎⁎⁎(0.077)
0.469 0.444 0.062
Step 2
0.206⁎(0.096)
0.466 0.439 0.069
0.107⁎⁎(0.038) 0.004(0.032) 0.430 0.509 0.076
0.531 0.437 0.077
0.525 0.429 0.093
0.122⁎⁎⁎(0.032) 0.009(0.035) 0.469 0.411 0.110
Note: N = 430 employees in 67 teams. The term in brackets is the robust standard error. C = Conscientiousness, C2 = the quadratic term of Conscientiousness. Hereafter, σ2 = within-team variance, τ00 = intercept variance, Pseudo R2 is obtained using R code available in Aguinis, Gottfredson, & Culpepper, 2013. ⁎ P < 0.05 ⁎⁎ P < 0.01 ⁎⁎⁎ P < 0.001
curvilinear effects and the inflection points under low SSC (−1.00 SD) to those under high SSC (+1.00 SD). Although the curvilinear relationships between Conscientiousness and safety performance hold under various SSC levels, the inflection points under low SSC (safety compliance = 0.27 SD; safety participation = − 0.76 SD) were lower than those under high SSC (safety compliance = 2.22 SD; safety participation = 1.29 SD). The plots (following Dawson, 2014) of the quadratic interactions were shown Fig. 1a–b.
relationships were significant (safety compliance: b = −0.112, p < 0.01; safety participation: b = −0.176, p < 0.001) under low SSC, and the inflection points occurred at − 0.16 SD for safety compliance and −0.36 SD for safety participation. By contrast, no curvilinear relationships (safety compliance: b = 0.012, ns.; safety participation: b = 0.022, ns.), but positively linear relationships existed (safety compliance: b = 0.188, p < 0.001; safety participation: b = 0.222, p < 0.001) under high SSC (see Fig. 1c–d).
3.3.2. Emotional Stability and safety performance The quadratic effect of Emotional Stability in step 2 significantly predicted safety compliance (γ = − 0.045, p < 0.05) and safety participation (γ = − 0.052, p < 0.05) (see Table 3), supporting Hypothesis 1b. In Step 3, the interaction between Emotional Stability and SSC significantly predicted safety compliance (γ = 0.112, p < 0.05) and safety participation (γ = 0.174, p < 0.001), and the interaction between quadratic term of Emotional Stability and SSC significantly predicted safety compliance (γ = 0.062, p < 0.05) and safety participation (γ = 0.099, p < 0.001). Simple slope test showed that the curvilinear Emotional Stability–safety performance
3.3.3. Extraversion and safety performance The quadratic effect of Emotional Stability in step 2 significantly predicted safety compliance (γ = − 0.057, p < 0.001) and safety participation (γ = −0.048, p < 0.05) (see Table 4), supporting Hypothesis 1c. The moderating effects of SSC on the quadratic term of Extraversion–safety compliance relationship (γ = 0.038, p < 0.05) and on Extraversion–safety participation relationship (γ = 0.103, p < 0.05) were significant. Specifically, the quadratic term of Extraversion significantly predicted safety performance under low SSC (safety compliance: b = − 0.095, p < 0.001; safety participation: b = − 0.102, p < 0.05); whereas, the curvilinear effects of
Table 3 Hierarchical linear modeling results for Emotional Stability. Safety compliance Step 1 Level 1 Age Job tenure ES ES2 Level 2 Support Support × ES Support × ES2 σ2 τ00 pseudo R2
Safety participation Step 2
Step 3
0.084(0.069) 0.070(0.063) 0.111⁎⁎⁎(0.034)
Step 1
Step 3
0.010(0.059) 0.190⁎⁎(0.067) 0.084⁎(0.041) −0.045⁎(0.023)
− 0.052⁎(0.025)
0.278⁎⁎⁎(0.076)
0.471 0.432 0.071
Step 2
0.202⁎(0.092)
0.468 0.430 0.080
0.112⁎(0.052) 0.062⁎(0.028) 0.455 0.456 0.088
Note: N = 430 employees in 67 teams. The term in brackets is the robust standard error. ES = Emotional Stability, ES2 = the quadratic term of Emotional Stability. ⁎ P < 0.05 ⁎⁎ P < 0.01 ⁎⁎⁎ P < 0.001
58
0.528 0.425 0.093
0.525 0.411 0.111
0.174⁎⁎⁎(0.044) 0.099⁎⁎⁎(0.028) 0.494 0.447 0.149
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
Table 4 Hierarchical linear modeling results for Extraversion. Safety compliance Step 1 Level 1 Age Job tenure E E2 Level 2 Support Support × E Support × E2 σ2 τ00 pseudo R2
Safety participation Step 2
Step 3
Step 1
0.086(0.068) 0.070(0.062) 0.057(0.040)
Step 3
0.010(0.059) 0.189⁎⁎(0.066) 0.040(0.036) −0.057⁎⁎⁎(0.016)
− 0.048⁎(0.024)
0.282⁎⁎⁎(0.078)
0.478 0.448 0.052
Step 2
0.207⁎(0.093)
0.474 0.433 0.072
0.064(0.062) 0.038⁎(0.017) 0.460 0.400 0.073
0.531 0.437 0.077
0.529 0.426 0.092
0.103⁎(0.041) 0.046(0.031) 0.517 0.407 0.091
Note: N = 430 employees in 67 teams. The term in brackets is the robust standard error. E = Extraversion, E2 = the quadratic term of Extraversion. ⁎ P < 0.05 ⁎⁎ P < 0.01 ⁎⁎⁎ P < 0.001
Extraversion on safety compliance (b = −0.020, ns.) and safety participation (b = − 0.010, ns.) were insignificant, and Extraversion marginally significantly predicted safety compliance (b = 0.093, p = 0.08) and significantly predicted safety participation (b = 0.123, p < 0.05) under high SSC (Fig. 1e–1f). Inflection points under low SSC (safety compliance = − 0.19 SD; safety participation = −0.41 SD) were lower than those under high SSC (safety compliance = 2.38 SD; safety participation = 5.98 SD). Taken together, the results for Conscientiousness and Extraversion supported Hypothesis 2. As for Emotional Stability, their relationship is linearly positive under high SSC, and the inflection points are insignificant. These results are consistent with our theoretical reasoning that SSC optimizes the benefits of Emotional Stability, and a high level of Emotional Stability is associated with high safety performance under high SSC.
4.1. Theoretical implications This study expands curvilinear personality–performance relationship to the occupational safety domain and suggests that both extremely low and high personality levels may impair safety performance. In particular, for safety compliance, the optimal level of Conscientiousness (1.33 SD) was higher than those of Emotional Stability (0.99 SD) and Extraversion (0.47 SD), suggesting that the weak and inconsistent linear personality–safety compliance relations in previous research (Christian et al., 2009; Clarke & Robertson, 2005) do not imply that Emotional Stability and Extraversion have no effects on safety compliance; rather, their relationships are curvilinear. For safety participation, the optimal level of Emotional Stability (0.56 SD) was the highest and it was followed by Extraversion (0.39 SD) and Conscientiousness (0.33 SD) because moderated high level of Emotional Stability could enhance employees' desire for communion (Barrick et al., 2013). Furthermore, our results depict a complete picture of personality–safety performance relationship under different SSC levels. Under low SSC, higher personality levels are initially associated with higher safety performance levels. However, after inflection points occur, higher personality levels are associated with lower safety performance levels. Under high SSC, the inflection points occur at a higher level or become insignificant, and higher personality levels are generally associated with higher safety performance levels. These results indicate that SSC optimizes the benefits of personality traits on performance (Barrick et al., 2013), thus the costs of extremely high trait scores can be mitigated under high SSC. These results also suggest that curvilinear personality–performance relationship is highly sensitive to situation, which may have an implication for the elusiveness of their curvilinear relationship (Converse & Oswald, 2014; Uppal, 2017). In our results, both shape (i.e., linear or nonlinear) and inflection points vary considerably across different contexts. Accordingly, possible situational moderators should be considered in future research to determine the accurate curvilinear personality–performance relationship.
4. Discussion Our research advances the field by presenting the curvilinear effects of personality on safety performance and the moderating effects of team-level SSC. To rule out the possibility that these curvilinear effects were due to the skewness of safety performance, the skewness values for safety performance were calculated and compared with the recommended threshold of slightly nonnormal ( ± 1.00) (Lei & Lomax, 2005). Results suggested that the skewness for safety participation (− 0.74) was not a serious issue. Although skewness for safety compliance (− 1.13) slightly exceeds the criterion, this distribution is actually normal among safety-priority occupations, which could be found in previous study (e.g., Martínez-Córcoles, Gracia, Tomás, Peiró, & Schöbel, 2013; Xu et al., 2014). To further clarify these curvilinear effects, correlation analysis between two personality levels (below the inflection point vs. above the inflection point) and safety performance under different SSC levels (below the mean vs. above the mean) was conducted. Under low SSC levels, personality–safety performance correlations were positive at low levels of personality and turned negative at high levels of personality; under high SSC levels, personality–safety performance correlations were positive at low levels of personality and were weaker or turned weakly negative at high levels of personality. This indicated that the curvilinear relationships and interactions indeed existed, unlike the ceiling effects that would be created by skewness (details are available as supplementary materials).
4.2. Practical implications Our findings present important practical implications for personnel selection and interventions. First, to screen out accident-prone applicants, personnel selection can benefit by using two cutoff points instead of top-down selection. Second, the results show that high trait scores can be harmful to safety performance under low SSC, whereas high trait scores lead to high safety performance under high SSC, thereby 59
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
Fig. 1. Cross-level moderating effect of SSC on the curvilinear personality–safety performance relationships.
2014). Furthermore, when implementing a personality development program, the optimal level for personality requirement of a particular job position should be identified, rather than simply improving socially desirable personality.
indicating that enhancing SSC can mitigate the negative effects of extremely high trait scores on safety performance. In particular, supervisors should provide employees with trait-relevant cues (e.g., performance feedback, praise for safety performance; Zohar & Polachek, 60
Personality and Individual Differences 122 (2018) 55–61
X. Yuan et al.
4.3. Limitations and future directions
References
This study has several limitations that should be addressed in future research. First, the cross-sectional design limited causal interpretations. Future experimental and longitudinal designs can help in further investigations. Second, only one sample is used, which may limit the generalizability of our findings. Thus, the pattern of results should be evidenced in other organizations. Finally, our study develops theoretical arguments related to the possible mechanism of curvilinear personality–performance relationship and the moderating effects of SSC, thereby suggesting potential mediators (e.g., motivation). However, we did not measure the potential mediators directly. Thus, we suggest further study to precisely support our theory by exploring possible mediating mechanisms.
Aguinis, H., Gottfredson, R. K., & Culpepper, S. A. (2013). Best-practice recommendations for estimating cross-level interaction effects using multilevel modeling. Journal of Management, 39, 1490–1528. Anderson, C., & Galinsky, A. D. (2006). Power, optimism, and risk-taking. European Journal of Social Psychology, 36, 511–536. Barrick, M. R., Mount, M. K., & Li, N. (2013). The theory of purposeful work behavior: The role of personality, higher-order goals, and job characteristics. Academy of Management Review, 38, 132–153. Beus, J. M., Dhanani, L. Y., & McCord, M. A. (2015). A meta-analysis of personality and workplace safety: Addressing unanswered questions. Journal of Applied Psychology, 100, 481–498. Bliese, P. D., & Castro, C. A. (2000). Role clarity, work overload and organizational support: Multilevel evidence of the importance of support. Work and Stress, 14, 65–73. Carter, N. T., Dalal, D. K., Boyce, A. S., O'Connell, M. S., Kung, M. C., & Delgado, K. M. (2014). Uncovering curvilinear relationships between conscientiousness and job performance: How theoretically appropriate measurement makes an empirical difference. Journal of Applied Psychology, 99, 564–586. Carter, N. T., Guan, L., Maples, J. L., Williamson, R. L., & Miller, J. D. (2016). The downsides of extreme conscientiousness for psychological well-being: The role of obsessive compulsive tendencies. Journal of Personality, 84, 510–522. Cheng, Y., Luh, W. M., & Guo, Y. L. (2003). Reliability and validity of the Chinese version of the job content questionnaire in Taiwanese workers. International Journal of Behavioral Medicine, 10, 15–30. 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. Journal of Applied Psychology, 94, 1103–1127. Clarke, S., & Robertson, I. (2005). A meta-analytic review of the Big Five personality factors and accident involvement in occupational and non-occupational settings. Journal of Occupational and Organizational Psychology, 78, 355–376. Converse, P. D., & Oswald, F. L. (2014). Thinking ahead: Assuming linear versus nonlinear personality-criterion relationships in personnel selection. Human Performance, 27, 61–79. Costa, P., & McCrae, R. (1992). Neo PI-R professional manual. Odessa, FL: Psychological Assessment Resources. Dawson, J. F. (2014). Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29, 1–19. Eysenck, H. J. (1967). The biological basis of personality. Springfield, IL: Charles C. Thomas. Grant, A. M., & Schwartz, B. (2011). Too much of a good thing: The challenge and opportunity of the inverted U. Perspectives on Psychological Science, 6, 61–76. Hansen, C. P. (1988). Personality characteristics of the accident involved employee. Journal of Business and Psychology, 2, 346–365. Henning, J. B., Stufft, C. J., Payne, S. C., Bergman, M. E., Mannan, M. S., & Keren, N. (2009). The influence of individual differences on organizational safety attitudes. Safety Science, 47, 337–345. Judge, T. A., & LePine, J. A. (2007). The bright and dark sides of personality: Implications for personnel selection and team contexts. In J. Langan-Fox, C. Cooper, & R. Klimoski (Eds.). Research companion to the dysfunctional workplace: Management challenges and symptoms (pp. 332–355). Cheltenham, England: Elgar. Le, H., Oh, I. S., Robbins, S. B., Ilies, R., Holland, E., & Westrick, P. (2011). Too much of a good thing: Curvilinear relationships between personality traits and job performance. Journal of Applied Psychology, 96, 113–133. LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods, 11, 815–852. Lei, M., & Lomax, R. G. (2005). The effect of varying degrees of nonnormality in structural equation modeling. Structural Equation Modeling, 12, 1–27. Martínez-Córcoles, M., Gracia, F. J., Tomás, I., Peiró, J. M., & Schöbel, M. (2013). Empowering team leadership and safety performance in nuclear power plants: A multilevel approach. Safety Science, 51, 293–301. Martocchio, J. J., & Judge, T. A. (1997). Relationship between conscientiousness and learning in employee training: Mediating influences of self-deception and self-efficacy. Journal of Applied Psychology, 82, 764–773. 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. Journal of Applied Psychology, 91, 946–953. Postlethwaite, B., Robbins, S., Rickerson, J., & McKinniss, T. (2009). The moderation of conscientiousness by cognitive ability when predicting workplace safety behavior. Personality and Individual Differences, 47, 711–716. Uppal, N. (2017). Moderation effects of perceived organisational support on curvilinear relationship between neuroticism and job performance. Personality and Individual Differences, 105, 47–53. Widiger, T. A., & Mullins-Sweatt, S. N. (2009). Five-factor model of personality disorder: A proposal for DSM-V. Annual Review of Clinical Psychology, 5, 197–220. Xu, Y., Li, Y., Wang, G., Yuan, X., Ding, W., & Shen, Z. (2014). Attentional bias toward safety predicts safety behaviors. Accident Analysis & Prevention, 71, 144–153. Yuan, Z., Li, Y., & Lin, J. (2014). Linking challenge and hindrance stress to safety performance: The moderating effect of core self-evaluation. Personality and Individual Differences, 68, 154–159. Zohar, D., & Polachek, T. (2014). Discourse-based intervention for modifying supervisory communication as leverage for safety climate and performance improvement: A randomized field study. Journal of Applied Psychology, 99, 113–124.
5. Conclusion The results demonstrate curvilinear personality–safety performance relationship, and suggest that a supervisor supportive environment mitigates the negative effects of extremely high personality traits on safety performance. These findings provide insight into the nature of personality–safety performance relationships and curvilinear person–situation interaction. This study also suggests that organizations should use the method involving two cutoff points in personnel selection and conduct interventions concerning supervisor support in safety management. Acknowledgements This study was partly supported by National Natural Science Foundation of China (Grant Numbers: 71371179 and 71501177) and Scientific Foundation of Institute of Psychology, Chinese Academy of Sciences (Y5CX143005). And also, we would like to thank the editor and the anonymous reviewers for their helpful comments and insightful feedback. Appendix A. Items of perceived supervisor support and safety performance used in the study Perceived supervisor support 1. 2. 3. 4.
My My My My
supervisor supervisor supervisor supervisor
is concerned about the welfare of those under him. pays attention to what I am saying. is helpful in getting the job done. is successful in getting people to work together.
Safety compliance 1. He uses all the necessary safety equipment to do his job. 2. He uses the correct safety procedures for carrying out his job. 3. He ensures the highest levels of safety when he carries out his job. Safety participation 1. He promotes the safety program within the organization. 2. He puts in extra effort to improve the safety of the workplace. 3. He voluntarily carries out tasks or activities that help to improve workplace safety. Appendix B. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.paid.2017.10.005.
61