Effects of Observers′ Performance Expectations on Behavior Ratings of Work Groups: Memory or Response Bias?

Abstract

The goal of this study was to pinpoint the processes mediating the effects of preobservation performance expectations on behavioral ratings of work groups. Prior to observing the same work group, 59 subjects were given positive or negative information concerning the group′s performance. Immediately after or 1 week later, subjects completed a questionnaire measuring the group′s behavior. As expected, subjects led to believe that the group had performed well (versus poorly) attributed more effective and fewer ineffective behaviors to the group. Importantly, additional analyses revealed that these effects were mediated by a systematic response bias. Specifically, subjects provided with positive (versus negative) performance information adopted a more liberal decision criterion when judging the occurrence of effective work behaviors and a more conservative decision criterion when judging the occurrence of ineffective work behaviors. There was no evidence of a memory bias in the judgments of effective or ineffective work behavior. By pinpointing where in the rating process bias is introduced and precisely how and why performance expectations exert their influence, these results provide direction for future efforts aimed at eliminating this source of bias in behavioral ratings of work groups.