LOGIT AND MULTILEVEL LOGIT MODELING OF COLLEGE GRADUATION FOR 1984-1985 FRESHMAN STUDENT-ATHLETES

This article describes statistical research on the academic performanc e of student-athletes in college sports programs. We describe several statistical models used in the prediction of academic success defined by college persistence and graduation. Using longitudinal data on the academic performances of about 3,000 student-athletes in NCAA Division I collegiate sports programs, we formulate legit and multilevel legit statistical models for the prediction of graduation rates. These pred iction models are based on academic, demographic, and athletic variabl es, and are used to account for differences in both the students and t he colleges. These results show (1) moderate but significant relations hips between precollege academic characteristics and college graduatio n, (2) small but significant differential validity of prediction betwe en major student-athlete groups, (3) notable college-level variance in the average graduation rate, (4) small but significant within-college relationships between precollege academic characteristics and college graduation, and (5) differences between colleges accounted for by ins titutional graduation rates. We highlight statistical issues about the application of legit and multilevel models and discuss substantive is sues about the current implications of these results.