Combining predictors to achieve optimal trade-offs between selection quality and adverse impact.

Abstract

The authors propose a procedure to determine (a) predictor composites that result in a Pareto-optimal trade-off between the often competing goals in personnel selection of quality and adverse impact and (b) the relative importance of the quality and impact objectives that correspond to each of these trade-offs. They also investigated whether the obtained Pareto-optimal composites continue to perform well under variability of the selection parameters that characterize the intended selection decision. The results of this investigation indicate that this is indeed the case. The authors suggest that the procedure be used as one of a number of potential strategies for addressing the quality-adverse impact problem in settings where estimates of the selection parameters (e.g., validity estimates, predictor intercorrelations, subgroup mean differences on the predictors and criteria) are available from either a local validation study or meta-analytic research.

Personality and Cognitive Ability as Predictors of Effective Performance at Work

“…In a series of papers, DeCorte and colleagues (DeCorte et al 2007(DeCorte et al , 2010(DeCorte et al , 2011 provided an analytic technique that optimizes the goals prescribed by an organization. The various aspects that are entered into the optimization procedure include the predictor subset and its characteristics, the selection ratio or selection rule, the staging and sequencing of the selection procedure, and the weighting of the predictors.…”

Computing Pareto‐optimal Predictor Composites for Complex Selection Decisions

“…Also, it is assumed that the job performance and the predictor scores follow a multinormal distribution with the same variance/covariance matrix but a different mean structure in the applicant subpopulations. However, it is noted that essentially the same data and identical distributional assumptions underlie all previous related work on the estimation of the effects of predictor weighing on the balance between selection quality and diversity in both the simple (De Corte et al, ) and the complex selection decision context (Scholarios et al, ). Also, certain aspects of the distributional assumption can be relaxed in that the present method can be adapted to handle situations where the predictor variance/covariance matrix varies across the different applicant subgroups.…”

Practical Impact of Predictor Reliability for Personnel Selection Decisions

“…If cognitive ability is the most valid predictor of job performance, as is often found, then being less selective on this measure may reduce adverse impact but also will compromise validity (this is more likely to be the case to the extent that job performance is saturated with cognitive ability). For example, De Corte, Lievens, and Sackett (2007) demonstrated that noncognitive ability measures had to be weighted substantially heavier relative to cognitive ability measures before satisfying the 4/5 th s rule. Consequently, expected standardized criterion performance of the selected applicants decreased from .78…”