Self and supervisor ratings of job-performance: Meta-analyses and a process model of rater convergence
Year: 2005
Type: dissertation
Study designs: Systematic Review, Meta-Analysis
Abstract: This dissertation presents meta-analyses on the congruence of self- and supervisor-ratings for job performance (k=104 independent samples). It examines two indicators of rater convergence (self-supervisor correlations and mean difference scores) and reports results for an extensive set of moderator variables. Correlation coefficients are interpreted as a measure of self-rating validity whereas the higher mean level of self-ratings compared to supervisory ratings is considered to indicate leniency in self-ratings. Self-supervisor ratings yielded an overall correlation of r=.22 (k=96; n=22287). Job type (i.e. blue collar vs. white collar; managerial responsibility) and the use of non-judgmental performance dimensions were the main moderators. The analysis also confirmed the notion that self-ratings of performance are lenient (d=.33; k=70; n=29386). For Western samples, self-ratings were consistently lenient under all conditions, but the extent to which ratings were lenient depended on situational variables (e.g. purpose for ratings). The two measures of rater congruence varied independently and were moderated by different sets of variables. A process model of performance rating is presented that integrates the study’s findings.
Author: Heike Heidemeier
Language: English
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3136976516