Among the available selection strategies (e.g., top down selection), sliding bands with minority preference selection was shown to be the most effective at striking a balance between reducing adverse impact with minimal test utility loss. Unfortunately, all previous research into selection strategy effectiveness failed to model job acceptance rates, a variable shown to decrease overall test utility (Murphy, 1986). In this study we compared the utility and adverse impact ratios obtained from strict top down and sliding bands with minority preference selection strategies in which we varied selection ratios, job acceptance rates, and sample sizes. Across all conditions, utility and adverse impact ratios were found to be lower than was demonstrated in previous research which ignored job acceptance rates. Only one of our four hypotheses was supported in this study. We found that differences in adverse impact ratios between top down selection and banding with race preferential selection was reduced when high scoring minority applicants refused offers at a rate higher than high scoring majority applicants. Thus, the benefits that employers expect to see when utilizing the sliding band with minority preference selection strategy are not as great as previously believed.
| Identifer | oai:union.ndltd.org:WKU/oai:digitalcommons.wku.edu:theses-1549 |
| Date | 01 May 2004 |
| Creators | Daniel, Joshua |
| Publisher | TopSCHOLAR® |
| Source Sets | Western Kentucky University Theses |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Masters Theses & Specialist Projects |
Page generated in 0.0017 seconds