Access Type
Open Access Dissertation
Date of Award
January 2024
Degree Type
Dissertation
Degree Name
Ph.D.
Department
Psychology
First Advisor
Andrew Speer
Abstract
Profile matching is an assessment scoring method that quantifies the level of fit with an ideal profile. The method is sometimes used to score personality trait data in employment selection contexts. Some initial research has suggested that profile matching may be a valid method of scoring these data, however, little is known about how the criterion-related validity and standardized mean demographic differences that are produced by profile matching compare to other methods of scoring personality data. Additionally, little is known about the trait characteristics and validation conditions that affect what scoring method may be most appropriate. The present research utilized simulated conditions as well as archival manager data to address this gap by investigating the effect of calibration sample size, mean trait validity, demographic trait differences, demographic proportions in the ideal profile of profile matching, and nonlinear trait-criterion relationships on the criterion-related validity and standardized mean demographic differences produced by profile matching and other trait scoring methods (i.e., job analysis, unit-weighting, linear regression, and quadratic regression). Results suggest that profile matching tends to produce the lowest validity estimates of the scoring methods while regression tends to produce the highest validity estimates when calibration sample sizes are high and job analysis weighting is likely to produce the highest validity estimates when calibration samples are low. When there are trait-criterion relationships that are nonlinear and have an underlying linear effect, quadratic regression tends to produce the highest validity estimates. Under these conditions, profile matching produced the lowest validity estimates of the methods investigated. Results suggested that profile matching tends to produce lower standardized mean demographic differences than the other methods investigated and the standardized mean demographic differences of profile matching depend on representation within the ideal profile. However, these differences have minimal practical significance when demographic trait differences are low. Implications for practice are outlined.
Recommended Citation
Stahl, Wyatt, "Profile Matching: Identifying Conditions That Impact The Appropriateness Of Personality Scoring Methods" (2024). Wayne State University Dissertations. 3980.
https://digitalcommons.wayne.edu/oa_dissertations/3980
