Gender-Related Differential Validity and Differential Prediction in Interest Inventories

Autor:	Katja Pässler, Andrea Beinicke, Benedikt Hell
Rok vydání:	2013
Předmět:	Organizational Behavior and Human Resource Management Association (psychology) Psychology Gender related Differential item functioning Social psychology General Psychology Applied Psychology Differential (mathematics) Developmental psychology
Zdroj:	Journal of Career Assessment. 22:138-152
ISSN:	1552-4590 1069-0727
DOI:	10.1177/1069072713492934
Popis:	Following the suggestions of National Institute of Education and American Psychological Association Standards, this article addresses the issue of differential validity and differential prediction in the vocational interest domain as one major concern of test fairness. In order to investigate potential prediction bias in vocational interest measures, we first compare gender-specific validity coefficients for the prediction of person–environment fit and satisfaction to test for differential validity, and second, we examine gender differences in the slopes and intercepts of the regression model predicting person–environment fit to test for differential prediction. Results show evidence of differential validity and some indications of differential prediction in a standard Holland Interest Inventory. However, removing items showing large gender-specific differential item functioning, that is, controlling for measurement bias, slightly reduced prediction bias. Practical implications are discussed and further research objectives are suggested.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::3f3aa5445a982d57f8fecc0ca6fb30b3 https://doi.org/10.1177/1069072713492934 Zobrazit plný text záznamu