The Effects of Rater Performance and Perspective on Rating Leniency

Abstract

Rater errors such as leniency/severity have detrimental effects on the validity of performance ratings. A number of rater characteristics have been examined to understand why some raters tend to be consistently more lenient than others; however, gaps remain in our understanding of these rater characteristics and their influence on rating leniency. The present study examined the previously unexplored characteristic of rater performance as a predictor of rater leniency/severity. It was hypothesized that rater performance would be negatively associated with rater leniency, such that high performing employees would be more severe in their evaluations of others. Furthermore, it was hypothesized that this relationship would be particularly pronounced when raters and subjects are peers to one another, and when rater and subject are of the same gender. These hypotheses were tested using a large archival data set including multi-source assessment ratings and annual performance ratings for employees in a multinational healthcare organization. The hypotheses were not supported, and in fact a small positive relationship between rater self-ratings of performance and rater leniency was detected. Implications and future directions are discussed.