An Examination of the Robustness of the General Utility Function

Abstract

Two important parameters of the general utility function-SD,, the standard deviation of job performance, and 4/p, predictor performance-were examined in terms of their sensitivity to departures from underlying mathematical assumptions. Integral calculus was used to assess the effects of statistical violations of the assumptions underlying these parameters. The standard deviation of job performance was found to be robust, while the selection ratio was more grossly affected by departures from normality. The combined effects of these two parameters were assessed by simultaneously examining the effects of varying selection standards and varying job performance on the expected per-person increase in standardized job performance. It was found that such effects were nonsymmetrical and depended on base rates of job performance. Implications for the practical implementation of the utility function are discussed.