Biases in the recording of performance-related information: The effects of initial impression and centrality of the appraisal task

Abstract

Two variables, initial impression and rating task centrality, were hypothesized to bias the recording of performance-related behavioral incidents used by raters during appraisal. Eighty students viewed two short videotapes of a single lecturer delivering lecture segments and collected behavioral incidents (both good and bad) exhibited by the lecturer using the behavior diary format. Students were randomly assigned to a single experimental condition where both initial performance of the lecturer (good vs poor performance on the first lecture seen) and importance of the behavior collection task (high vs low rating task centrality) were manipulated. The hypothesized effects of initial impression and rating task centrality were only partially demonstrated. In particular, initial impression produced a significant contrast effect in the subsequent recording of behavioral incidents. There was no significant difference in the number of incidents collected under varying conditions of rating task centrality. However, a significant rating task centrality × initial impression interaction showed that while under low rating task centrality subjects in the negative impression condition collected more incidents than subjects in the positive condition, and the opposite effect was found for subjects in the high rating task centrality condition.