Content area
Full text
Assessment of standardized ethnic group differences (d) on predictors of job performance has become an important issue for applied psychologists. A number of studies have used an experimental design in which the predictor of interest was administered after an initial screening predictor. We examined the influence of prior selection on a first predictor on observed ds for second predictors in multiple-hurdle selection systems. Results of a Monte Carlo simulation indicate observed ds on the second predictor are underestimated in the presence of prior selection on another predictor. More important, "downward bias" in observed standardized ethnic group difference is substantial (30-70%) when selection ratios are low, standardized ethnic group differences on the screening predictor are high, and when the first and second predictors correlate above .30. Researchers should consider the influence of range restriction in designing studies of ethnic group differences and comparing ds across predictors, particularly when data are collected under a multiple-hurdle design.
The assessment of ethnic group differences on predictors of job performance is an important concern to applied psychologists and human resource managers who are interested in test fairness and workforce diversity issues (e.g., Aguinis, Corona, & Goldberg, 1998, 2000; Schmitt, Clause, & Pulakos, 1996). Studies that examine ethnic group differences typically focus on a given predictor (e.g., a cognitive ability test or an interview) and report a measure of covariance between predictor scores and ethnicity. A common statistic used in this literature is d or the standardized mean difference. The d statistic represents the difference between group means (e.g., White vs. Black) divided by the pooled (sample weighted, within-group) standard deviation of the relevant groups (Hunter & Schmidt, 1990). All else equal, one prefers a predictor with a small value of d because d, along with the selection ratio, largely determine minority hiring rate and adverse impact (Sackett & Ellingson, 1997).
Studies in applied psychology have directly compared ethnic group differences on various predictors via narrative review (e.g., Reilly & Chao, 1982; Reilly & Warech, 1993), and others compared ds metaanalytically (e.g., Huffcutt & Roth, 1998; Schmitt et al., 1996). Recently, a set of studies has created predictor composites and studied their aggregate level of standardized ethnic group differences (Bobko, Roth, & Potosky, 1999; Schmitt, Rogers, Chan, Sheppard, & Jennings,...