The Relationship between College Freshman Class Size and Other Institutional Characteristics and the Accuracy of Freshman Grade Predictions.

The validity of American College Testing (ACT) scores and self-reported high school grades in predicting college freshman grade average is documented, and the accuracy of prediction equations based on these measures is reported for institutions of differing freshman class size, affiliation, degree level, and racial/ethnic composition. Results are based on data collected from a random sample of 205 colleges that participated in the ACT Research Services in 1974-75 and 1976-77, with a separate prediction equation calculated for each college from its 1974-75 data. The predictive validity of ACT scores and high school grades was weakly related to freshman class size of more than 90. Prediction accuracy was moderately related to the institutional characteristics of affiliation, degree level, and racial/ethnic composition. Among the total college group, the accuracy of separate-sex predictions was less strongly related to freshman class size than to other institutional characteristics studied. However, it was more strongly related to freshman class size at private and four-year institutions than other kinds of institutions. Combined-sex equations based on simple random samples of size 50 from the base year data were almost as accurate as equations based on all records from the colleges, suggesting that ACT data could be used to make predictions of acceptable accuracy at colleges with as few as 50 freshmen. (Author/MSE)