Premium
MULTI‐STAGE SELECTION STRATEGIES: A MONTE CARLO INVESTIGATION OF EFFECTS ON PERFORMANCE AND MINORITY HIRING
Author(s) -
SACKETT PAUL R.,
ROTH LAWRENCE
Publication year - 1996
Publication title -
personnel psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.076
H-Index - 142
eISSN - 1744-6570
pISSN - 0031-5826
DOI - 10.1111/j.1744-6570.1996.tb01584.x
Subject(s) - selection (genetic algorithm) , representation (politics) , psychology , monte carlo method , personnel selection , diversity (politics) , statistics , computer science , machine learning , mathematics , law , political science , politics
This study explores alternative selection strategies available when a firm has two valid predictors that differ in the magnitude of subgroup differences. We examine 14 different selection rules (e.g., select on a composite of the two predictors versus screen on the first and then select on the second versus screen on the first and then select on a composite of the two), and document through a Monte Carlo simulation that the various selection rules can produce markedly different consequences in terms of the level of job performance achieved and the level of minority representation achieved. The selection rules examined include the use of within‐group norming, whichwas restricted by the Civil Rights Act of 1991, and the study examines how selection rules that do and do not include within‐group norming fare in terms of the tradeoffs between performance and minority representation. The study shows that the preferred selection strategy will depend on the relative value the firm places on performance and on minority representation, and that the effects of different screen‐then‐select selection strategies vary as a result of the selection ratios at the screening and selection stages, thus precluding simple conclusions about the merits of each selection strategy.