Publication Date

2015

Document Type

Dissertation

Committee Members

David LaHuis, Ph.D. (Advisor); Nathan Bowling, Ph.D. (Committee Member); Valerie Shalin, Ph.D. (Committee Member); Gary Burns, Ph.D. (Committee Member)

Degree Name

Doctor of Philosophy (PhD)

Abstract

Researchers and practitioners look to validity coefficients as indicators of the relationship between predictors and a measure of performance, typically as rated by a supervisor. Previous research suggests that subjective ratings of performance are problematic as they contain variance unrelated to subordinate performance. Though previous research suggested variance in supervisor ratings was random and unsystematic, recent literature suggests distortions may be deliberate and goal-directed. A potential implication of systematic variance in job performance ratings is a possible effect on validity coefficients. In the present study, I used multilevel modeling to determine if supervisor level characteristics influence the average ratings as well as the validity of job performance predictors for entry-level manufacturing incumbents. Results indicate that supervisor intelligence and level of adaptability had a significant relationship with the average performance rating they assigned. Though average ratings were not influenced, as the number of subordinates managed increased the validity of predictors Positive Attitude, Process Monitoring, and Responsibility decreased; and as the length of time managing increased, the validity of the predictor Attention to Detail increased significantly.

Page Count

61

Department or Program

Department of Psychology

Year Degree Awarded

2016


Share

COinS