A theory of adverse impact: Meta-analytic update of the 3-step model

Adverse Impact;Race;Parenting;Intelligence;Cognitive Ability;Child Development

Cottrell, Jonathan

关键词: Adverse Impact;    Race;    Parenting;    Intelligence;    Cognitive Ability;    Child Development;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/99293/COTTRELL-DISSERTATION-2017.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

In an attempt to better understand the causes of adverse impact, Cottrell, Newman, and Roisman (2015) have recently introduced a theoretical model called the 3-Step Model of Adverse Impact. The 3-Step Model provides a parsimonious developmental explanation for the origin of the Black-White cognitive test score gap, in which race relates to maternal advantage factors, which in turn relate to parenting factors, which in turn relate to cognitive test scores. One potential limitation of Cottrell et al.’s model, which is common in empirical tests of mediation models, is that the parameter estimates are based upon a single primary sample. To extend the applicability of the 3-Step model, I replicate the 3-Step Model of Adverse Impact (i.e., a mediation model) using Hunter and Schmidt (2004) meta-analytic methods, by conducting a research synthesis of the correlations involved in the 3-Step Model. This meta-analytic examination of the 3-Step Model is a rigorous empirical test to see which parts of the model still hold when using meta-analytic average population estimates across a diverse variety of samples. Results suggest that many of the parameter estimates and correlations of the model are of similar size and direction as in the original Cottrell et al. (2015) model, but less of the gap is meta-analytically explained by the covariates in the 3-Step Model.