An examination of psychometric bias due to retesting on cognitive ability tests in selection settings

• Main Authors: LIEVENS, Filip, REEVE, Charlie L., HEGGESTAD, Eric D.
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2007
• Subjects: Retesting; cognitive ability tests; measurement bias; psychometric bias; Human Resources Management; Organizational Behavior and Theory
• Online Access: https://ink.library.smu.edu.sg/lkcsb_research/5693; https://ink.library.smu.edu.sg/context/lkcsb_research/article/6692/viewcontent/retestbias.pdf
• Institution: Singapore Management University

Using a latent variable approach, the authors examined whether retesting on a cognitive ability measure resulted in measurement and predictive bias. A sample of 941 candidates completed a cognitive ability 14 test in a high-stakes context. Results of both the within-group between-occasions comparison and the between-groups within-occasion comparison indicated that no measurement bias existed during the initial testing but that retesting induced both measurement and predictive bias. Specifically, the results suggest that the factor underlying the retest scores was less saturated with g and more associated with memory than the latent factor underlying initial test scores and that these changes eliminated the test's criterion-related validity. This study's implications for retesting theory, practice, and research are discussed.