Evidence for Equitability of the Casper Test Structure: Measurement Invariance Across Gender and Ethnicity

• In Event: AERA Virtual Poster Session 4
  In Poster Session: Poster Session 2

Abstract

The Casper test assesses personal and professional characteristics, and is administered during the applicant screening process of many programs for health profession careers. This study extended psychometric evidence for Casper by investigating structural validity (from a confirmatory factor analytic approach [CFA]) and if construct test bias exists (from a measurement invariance [MI] perspective) across gender and ethnicity. The study conducted increasingly restrictive multi-groups CFAs to test for MI of the one-dimensional, 12-scenario Casper model. Results converged across two independent datasets (N = 2650 and 2332): Strict MI was found across gender (female, male) and reference group White with Asian, Black, Hispanic, and Mixed ethnic subgroups. The results contribute to evidence that Casper is a structurally valid and internally equitable test.

Authors

• Carolyn Allison Watters, Altus Assessments

• Gill Sitarenios, Altus Assessments