Predicting the academic performance of freshmen at the University of Wisconsin-La Crosse

Abstract

This study investigated the possibility of predicting the academic performance of first semester freshmen through the use of selected academic and nonacademic variables obtained from admissions information. The population studied was the 1987-88 fall semester freshman class at the University of Wisconsin-La Crosse. Student data was obtained from the Administrative Computing Center, and analyzed using Analysis of Variance and Multiple Regression. Through the test of the null hypothesis, it was found that high school class rank was the best overall predictor of first semester academic performance. There was also a significant relationship between first semester GPAs and the students' choice of housing while attending UWL. ACT subtest and composite scores did show a significant relationship when analyzed individually with first semester GPA, however, when included as a part of the Multiple Regression equation with all other variables, the ACT scores were not significant. Other variables used in the study, major indecision, hometown population, parents' level of income, and sex differences, were found to be insignificant when studying their relationship to first semester GPAs.