We quantitatively integrated 169 samples (N= 35,265 employees) that have been used to investigate the relationships of the following 7 work‐related stressors with job performance: role ambiguity, role conflict, role overload, job insecurity, work–family conflict, environmental uncertainty, and situational constraints. Overall, we obtained a negative mean correlation between each job performance measure and each stressor included in our analyses. As hypothesized, role ambiguity and situational constraints were most strongly negatively related to performance, relative to the other work‐related stressors. Analysis of moderators revealed that (a) the negative correlation of role overload and performance was higher among managers relative to nonmanagers; (b) publication year moderated the relation of role ambiguity and role overload with performance, although in opposite directions; (c) the correlations obtained for published versus unpublished studies were not significantly different; and (d) using the Rizzo et al. scale of role ambiguity and role conflict decreased the magnitude of the correlations of these stressors with performance, relative to other scales. Theoretical contributions, future research directions, and practical implications are discussed.

We quantitatively integrated 169 samples (N = 35,265 employees) that have been used to investigate the relationships of the following 7 workrelated stressors with job performance: role ambiguity, role conflict, role overload, job insecurity, work-family conflict, environmental uncertainty, and situational constraints. Overall, we obtained a negative mean correlation between each job performance measure and each stressor included in our analyses. As hypothesized, role ambiguity and situational constraints were most strongly negatively related to performance, relative to the other work-related stressors. Analysis of moderators revealed that (a) the negative correlation of role overload and performance was higher among managers relative to nonmanagers; (b) publication year moderated the relation of role ambiguity and role overload with performance, although in opposite directions; (c) the correlations obtained for published versus unpublished studies were not significantly different; and (d) using the Rizzo et al. scale of role ambiguity and role conflict decreased the magnitude of the correlations of these stressors with performance, relative to other scales. Theoretical contributions, future research directions, and practical implications are discussed.Psychosocial stressors at work represent a ubiquitous and multifaceted phenomenon (Lazarus, 1993); several theoretical frameworks predict that they affect employee attitudes and behaviors (Jex & Crossley, 2005). Most past meta-analytical reviews of these relationships focused only on the linkages of role conflict and role ambiguity with job performance, none of

This article combines meta-analysis with structural equation modeling to compare alternative models of the relationships among work stress, psychological mediators, and job performance. Specifically, the authors examined the mediating effects of job satisfaction and propensity to leave and their effect on the relationships between role ambiguity, role conflict, and job performance. The meta-analysis included both published and unpublished studies conducted over a period of 25 years, resulting in 113 independent samples with more than 22,000 individuals. As hypothesized, the structural model that best fit the meta-analytic estimates was the partial mediation model, in which stress is related to job performance both directly and indirectly through job satisfaction and propensity to leave and in which all path coefficients were reliably different from zero. The results are discussed in terms of theoretical contributions and implications for future stressperformance research.

We investigated the extent to which three socio-demographic variables, employee gender, age, and tenure, moderated the meta-correlations of role conflict and role ambiguity with job performance. To test these moderating effects, we quantitatively synthesized 30 independent studies (total N = 7700). No moderating effects were found for gender and tenure. Controlling for employee tenure and gender, we found that employee mean age had a moderating effect on the role ambiguity-performance correlation, with negative correlations tending to decrease with increasing age. Moreover, two significant two-way interactions were found between the moderators. As the percentage of women in the studies and the respondents' mean age simultaneously increased, and as the mean sample tenure and age of the respondents in the studies simultaneously increased, there was a reduction in the negative relationship between role ambiguity and performance. The implications of these results for future research are discussed.

No abstract