Recent studies have shown that in heterogeneous populations differential validities are often found in the population subgroups. In the light of the increasing societal emphasis on higher education, the valid prediction of academic success for all racial groups has become a necessity. This study is concerned with the differential validity for black and white students of academic predictors currently used at the University of Maryland. The University uses a Predictive Index Equation, which is a multiple regression equation involving high school grades, SAT-Verbal and SAT-Math scores. The Maryland grade-point average at the end of the freshman year is used as the criterion. In this study high school grades, SAT-V and SAT-M were examined separately as well as combined in multiple regression equations. The samples were categorized by race and sex. The results indicated that the Predictive Index worked as well for black students as it did for whites. For the sample studied, SAT scores were correlated with grades about as highly for whites as for blacks, although high school grades were not a valid predictor for black males. In addition to describing the study, the report examines other studies concerned with this topic. (AF)