Progress in Validity Generalization: Commentseplyesponse ultiplicative Validity Generalization Model: Accuracy of Estimates as a Function of Sample Size and Mean, Variance, and Shape of the Distribution of True Validities

An examination is made of the simulation study of Callender and Osburn (1980), which was aimed at determining the relative accuracy of equations used in validity generalization research to estimate the variance of true validities. It is argued that several of Callender and Osburn's statements were incorrect. Callender and Osburn view this criticism as misconceptions and attempt to refute the criticism on empirical grounds. Hunter, Schmidt, and Pearlman, though, maintain the validity of their objections. In an extension of the original work, the accuracy of estimates of the mean, variance, and lower credibility value of true validities produced by the independent multiplicative model and by a modified dependent model that takes the correlation between range restriction and criterion reliability artifacts into account were examined employing simulation methods. Overall, the results supported the applicability of the models by Callender and Osburn, as well as that of Schmidt and Hunter's noninteractive model.