Despite converging theoretical arguments regarding non-linear relationships between workload and employee attitudes (i.e. job satisfaction) and mental well-being outcomes, prior empirical support for these curvilinear effects has been mixed. In this study we offer and test two potential explanations that may help to reconcile this discrepancy. First, existing workload scales do not assess the full range of workload, thereby making it difficult to detect curvilinear relationships. Second, outcomes typically examined are too distal and there are different mediators (i.e. boredom and frustration) that explain effects at the low and high ends of the workload continuum, respectively, which also serves to obscure curvilinear effects. We examined these possibilities in two North American samples (N = 499 and 493) that employed different designs (i.e. cross-sectional versus multi-wave surveys). Overall, we find support for our hypotheses; ability to detect curvilinear effects is enhanced when using too much/too little rating scales that capture the entire workload continuum. Furthermore, boredom mediated the impact of low workload on outcomes, whereas frustration mediated the impact of high workload on outcomes. Therefore, this study helps clarify why prior studies may have inconsistently observed non-linear relationships between workload and outcomes. We discuss the implications for both researchers and practitioners.
- too much/too little