Prior selection causes biased estimates of standardized ethnic group differences: Simulation and analysis

Assessment of standardized ethnic group differences (d) on predictors of jo b performance has become an important issue for applied psychologists. A nu mber 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 A for se cond predictors in multiple-hurdle selection systems. Results of a Monte Ca rlo simulation indicate observed A on the second predictor are underestimat ed in the presence of prior selection on another predictor. More important, "downward bias" in observed standardized ethnic group difference is substa ntial (30-70%) when selection ratios are low, standardized ethnic group dif ferences 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 c omparing ds across predictors, particularly when data are collected under a multiple-hurdle design.