Longitudinal Predictions of University Grades for Blacks and Whites

Abstract
The purpose of the study was to compare predictions of freshman through senior year grades for black and white students by sex for new freshmen who entered one university in fall 1968 and 1969. Predictors were SAT-Verbal and SAT-Mathematical scores and high school grades using multiple regression equations. Results showed that different patterns of prediction for different race/sex subgroups emerged. For instance, high school grade averages are less useful in predicting grades beyond the freshman year and are a consistently poor predictor for black males. Black males were the least predictable and white females were the most predictable subgroup. The results are discussed in terms of past research and alternative predictors for subgroups. The authors conclude that predicting grades beyond the freshman year is useful and that separate prediction equations for race/sex subgroups should be employed.

This publication has 16 references indexed in Scilit: