Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


This study was designed to examine the influence of several potential sources of bias in the rating process. It specifically examined the potential effect of sex of rater and ratee, information about ratee characteristics, information about the job, and two rater personality factors on the way raters use performance information in coming to a final promotion decision. The study itself took place in a laboratory setting designed to simulate an assessment center. Raters were 320 business students at a large university. Pretesting was used to identify two jobs as recognizably "male" and "female" and two patterns of personality characteristics as either "male" or "female". Male and female raters were then asked to judge promotability of ratees who were identified as either male or female, and who had either "male" or "female" characteristics. Candidates were being considered for either "male" or "female" jobs. Rater cognitive complexity, as measured by the Hidden Figures Test, and attitudes toward women, as measured by the Women as Managers Scale, were also examined. Results were examined in a 2 x 2 x 2 x 2 factorial design with two covariates. Variables were male vs. female rater, male vs. female ratee, male vs. female characteristics and male vs. female job. The rater personality characteristics were treated as covariates. Overall results indicated that a contrast effect was operating and that women having masculine characteristics were strongly preferred for all jobs, even over males having the same characteristics. Women with feminine characteristics were strongly rejected, when compared to males with the same characteristics. Results of regression analysis of characteristics judged as most important also supported these findings. Male and female raters shared the biases identified. There was little evidence that raters made use of job information or that the personality characteristics identified affected the rating process. Practical implications of this study are described and suggestions for future research are made.