A simplified method of computing the effectiveness of tests in selection

Abstract

The Taylor-Russell method of predicting the number of successful workers, for various validities and selection ratios, is compared with a direct method in which the number successful is predicted from the data used to obtain the validity coefficient. Four sets of data are presented in which the predicted percent successful is compared with the percent actually successful, as determined from follow-up groups. "While the average errors of the Taylor-Russell method were… slightly less than the average errors of the direct method, the differences were small." (PsycInfo Database Record (c) 2025 APA, all rights reserved)