The Use of a GPA Prediction Formula to Better Prepare the Freshman Student-Athlete for College Academic Success: One School's Approach

Abstract

This study anaylzed academic data from 1,100 athletes, including 324 football students, at the University of Missouri, Columbia from 1980 to 1985. The goal was to develop a prediction formula consistening of variables that best predicted first semester fall GPAs of freshmen student-athletes. The predictions could then be used to help advise appropriate academic coursework for the group, to identify and to assist the incoming high risk freshmen become better prepared academically, and to help identify high risk candidates during recruiting. Variables included ACT composite score; all ACT subscores; high school percentile; high school size; race; hometown location; freshman or transfer status; scholarship status; fall UMC semester hours attempted and earned; winter UMC hours attempted and earned. The criterion variable was fall GPA. Winter GPA and year GPA were also included as variables to look at any significant correlations. A multiple regression procedure was then used to generate a prediction formula that best predicted the fall GPA of the freshmen student-athletes. Using the prediction formula student-athletes are grouped into four categories: no or minimal risk, little risk, moderate risk, and high risk. The categories were used for academic advising, the Study Skills Improvement Program, tutoring needs, hours requirements for study hall and tutoring, and identifying those who might benefit from special learning workshops and diagnostic testing. The hope is that coaches will also use this model for recruiting decisions.