Reducing shrinkage in diversity tradeoff curves for personnel selection: Comparing local validity studies, meta-analysis, and Bayes analysis.
比较局部效度研究、元分析和贝叶斯分析三种方法在减少多样性-效度权衡曲线收缩上的效果，为选拔实践中同时提升工作绩效和多样性提供方法建议。
For reducing adverse impact, the diversity-validity tradeoff curve approach (De Corte et al., 2007) provides sets of selection predictor weights that can often substantially enhance diversity (i.e., increase adverse impact ratio and number of minority job offers), with no loss of job performance in comparison to unit weights (Wee et al., 2014). A key limitation of this diversity-enhancing approach is the tendency for tradeoff curves to shrink, leading to lesser job performance and diversity outcomes upon cross-validation (Song et al., 2017). The current article evaluates and compares tradeoff curve shrinkage (both validity shrinkage and diversity shrinkage) using three types of validity evidence/calibration studies: (a) a local validity study, (b) a meta-analysis (Schmidt & Hunter, 1977), and (c) a Bayes analysis with empirical priors, which is a weighted combination of a local study with a meta-analysis (Newman et al., 2007). Using simulation, we show conditions where each approach performs best, offering recommendations on ideal methods for diversity improvement (reducing shrinkage and maximizing cross-validity) in local selection settings. Results guide selection practitioners in novel methods (integrating the advantages of meta-analysis, Bayes analysis, and Pareto-optimal weighting) to best combine predictors to simultaneously achieve job performance and diversity objectives in local selection settings. (PsycInfo Database Record (c) 2026 APA, all rights reserved).