The Role of Evaluation in Eliminating Social Loafing

Comparability of performances within four-person groups was manipulated in a brainstorming task. Crossed with this manipulation of evaluation potential, participants' outputs either were individually identifiable or were pooled. Replicating previous social-loafing research (Latane, Williams, & Harkins, 1979), when outputs were identifiable, participants generated more uses than when their outputs were pooled. However, this difference emerged only when participants believed that their individual outputs could be evaluated through comparison with their co-workers' performances. When participants believed that their individual outputs were not comparable and thus could not be evaluated, there was no difference in the number of uses generated by participants whose outputs were identifiable and those whose outputs were pooled. These data suggest that to eliminate social loafing participants must feel not only that their outputs are individually identifiable as suggested by Williams, Harkins, and Latane (1981), but also that these outputs can be evaluated through comparison with the outputs of their co-workers.