Date of Award

Spring 1997

Document Type


Degree Name

Doctor of Philosophy (PhD)



Committee Director

Robert M. McIntyre

Committee Member

Terry L. Dickinson

Committee Member

Debra A. Major

Committee Member

Todd A. Baker


The success of the assessment center method in predicting job performance has been one of the most researched efforts in personnel psychology (Thornton, 1992). However, there is little reported evidence showing that assessment center procedures produce scores that serve as valid representations of separate constructs (Klimoski & Brickner, 1987). It is perhaps ironic, then, that despite the success stories, we still do not understand why assessment centers "work," (i.e., predict performance).

This study examined the subtle criterion contamination hypothesis as an explanation to assessment center validity. The subtle criterion contamination hypothesis states that assessment centers predict managerial performance because assessors actually observe and evaluate participants on the basis of their knowledge of those factors needed to get ahead in the company. These factors, it is argued, may not necessarily be related to performance in the assessment center but are instead relevant to the manager's promotability within the organization (Klimoski & Strickland, 1977).

Descriptions of behavior along three job-relevant skill dimensions as well as two job-extraneous cues--ratee physical attractiveness and ratee sex served as the cues in a Brunswik (1955) lens model framework. Twenty-six experienced assessors and 20 supervisors from a county police department rated 32 profiles of fictitious ratees in a 2 (Rater Source) x 2 (Photo Present) x 2 (Dimension 1 Performance) x 2 (Dimension 2 Performance) x 2 (Dimension 3 Performance) mixed factorial design. For each profile, each rater evaluated the ratee's overall performance and the ratee's future promotability in the organization.

Results indicated that extraneous variables did not add significantly to the rating variance accounted for by the dimensions. However, the ratee photograph affected the weight that raters placed on the dimensions when making their evaluations. Further analyses revealed that ratee attractiveness and ratee sex had no impact on rater evaluations of ratee overall performance. However, ratee attractiveness significantly affected rater evaluations of ratee future promotability. Further, assessor decision strategies appeared to match those of their supervisor counterparts. These results suggest that subtle criterion contamination has minimal impact on assessment center validity. However, further research is encouraged to identify other potential extraneous factors that may have an impact on rater judgments.


In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).