Topic: Performance Appraisals, Personality
Publication: International Journal of Selection and Assessment (SEP 2009)
Article: Rating level and accuracy as a function of rater personality
Author: H.J. Bernardin, C.L. Tyler, & P. Villanova
Reviewed by: Benjamin Granger
Believe it or not, one of the common problems with performance appraisal ratings is that they are often too lenient! This tendency toward generally high ratings is referred to as the leniency bias .
There is substantial evidence that the leniency bias really occurs. But surely this does not occur all the time, right? (We wish it did!). So can we predict which raters are going to be more lenient in their performance ratings? Bernardin, Tyler, and Villanova (2009) suggest that indeed we can.
Specifically, Bernardin and colleagues tested whether personality factors ( agreeableness and conscientiousness) account for rater leniency as well as accuracy. The authors used a sample of 126 undergraduate students who participated in a group assignment and were required to rate each others’ performance following the assignment. Student ratings were then compared to those of a course professor and a doctoral student (i.e., expert ratings).
As expected, rater agreeableness and conscientiousness related to both rater accuracy and leniency. Raters high in agreeableness tended to make higher (more lenient) and less accurate ratings than those lower in agreeableness.
Raters high in conscientiousness, however, made less lenient and more accurate ratings. Overall, the worse ratings (most lenient and least accurate) were made by raters high in agreeableness and low in conscientiousness.
Like much of the past research supporting the importance of conscientiousness, these results suggest that conscientious managers and employees will make more accurate and less lenient performance ratings. This is important since performance ratings play such a pivotal role in many HR decisions (promotion/demotion, pay, termination, etc.). Finally, although agreeable employees are easier to work with than their less agreeable (perhaps contentious) counterparts, they do tend to make less accurate performance ratings.