PREDICTING GRADES IN DIFFERENT TYPES OF COLLEGE COURSES

Abstract

The ability of high school grades (high school GPA) and SAT® scores to predict cumulative grades in different types of college courses was evaluated in a sample of 26 colleges. Each college contributed data from three cohorts of entering freshmen, and each cohort was followed for at least four years. Colleges were separated into four levels by average SAT scores. Grade point averages for four categories of courses (English; science, math, and engineering [S/M/E]; social science; and education) were computed, and analyses were run separately for gender within race/ethnicity classifications. Correlations of the combined predictors with course grades over four or more years, corrected for range restriction, ranged from.45 for education courses to 0.64 for S/M/E courses. The SAT increment, that is, the increase in the multiple correlations when SAT scores are added to high school grades, ranged from 0.03 in education courses to 0.08 in S/M/E courses. Because these seemingly small numbers are frequently misinterpreted, an additional analysis showed how the percentage of students succeeding at a high level (cumulative GPA of 3.5 or higher) increases as SAT scores increase for students with similar high school grades. For example, for students with a high school GPA of 3.7 or higher in colleges where the mean combined SAT score is below 1200, only 2 percent of the students at the lowest SAT level (800 or lower combined score) were highly successful in social science courses. At the highest SAT level (1410-1600), 77 percent were highly successful.