Abstract
A work simulation was conducted to test the effects of workload on stress and performance. Social support was also investigated as a moderator variable. Two hypotheses were tested: (a) stress is an intervening variable between workload and performance and (b) social support moderates the workload-stress relation such that workload leads to lower stress when social support is high. For the 1st hypothesis, a path analysis showed an indirect relation between workload and performance with stress as an intervening variable. For the 2nd hypothesis, there was a significant 3-way interaction between workload, social support, and time. The interaction showed that, in the early stages of the experiment, high social support led to higher (rather than lower) stress. This "reverse buffering effect" did not occur during later stages of the experiment. Several alternative explanations are offered regarding the reverse buffering finding, including the possibility that stress leads to social-support seeking behavior. One practical implication of this study is that managers may need to take into consideration the employee workloads when developing performance goals.