Validity Studies Based on a Universal Criterion Scale

Abstract

Most institutional validity studies are based on registered students, although the validity of the predictors in the population of all applicants is the ideal. A natural subset of this population that is also of interest to admissions committees consists of registrants together with applicants who are attending competing institutions. Although the latter have obtained criterion scores, the data cannot be pooled easily because of different grade scales. In “The Scale-Linkage Algorithm: Construction of a Universal Criterion Scale for Families of Institutions,” we showed how to construct a universal criterion scale for a family of comparable institutions. Here the results are applied to the direct estimation of validity for various groups of applicants and to the testing of the assumptions underlying the more familiar extrapolation methods. Data from American Schools of Law are used to illustrate the procedures.