Apples, oranges, and ironing boards: Comparative effect sizes influence lay impressions of test validity

January 14, 2022
Abstract
Validity information (effect sizes) for selection tests can be difficult for people to understand without some additional context. This study examined how manipulation of the immediate context influenced impressions of the validity of a sales ability selection test. We found that lay people (n = 350) were more favorable toward the test when a set of test validity coefficients was presented after inferior validity coefficients for medical/health treatments, versus a control condition where only the set of test validity coefficients was presented (d = .68). Although we predicted this result based on the presumed inferior status accorded to psychological findings compared with medical/health ones, a third condition where participants were first presented inferior absurd or nonsensical validity coefficients was equally as effective as the medical/health condition. We explain our findings as possibly due to an anchoring and adjustment process and/or a stimulus frequency effect caused by the global (rather than local) context.
Go to Source
Author: Marie Childers,
Scott Highhouse,
Margaret E. Brooks