Defense Date


Document Type


Degree Name

Doctor of Philosophy


Educational Studies

First Advisor

Dr. Beverly J. Warren


The purpose of this investigation was to determine what factors contribute to student-athlete retention, specifically looking at scholarship support, gender, and sport type (individual or team sport). Eight Colonial Athletic Association (CAA) schools provided data on all student-athletes participating from the 200 1-02 through the 2004-05 academic years. The investigator collected data on site at each institution (University of Delaware, Drexel University, Georgia State University, James Madison University, University of North Carolina-Wilmington, Old Dominion University, Virginia Commonwealth University, and the College of William and Mary) and obtained on each student-athlete including year, sport, gender, amount of scholarship support (both athletic and other), total cost of tuition, and retention status.Using analysis of variance (ANOVA), discriminant function analysis, and binary logistic regression, the data were analyzed in regard what type of relationship the independent variables (scholarship support as a percentage of tuition, gender and sport type) have with the dependent variable, retention. Of the nearly 13,000 observations, 12,027 were retained (92.7%) and 953 were not retained (7.3%). Analyses revealed that scholarship support alone was not significantly related to retention, but that gender and sport type were both significant predictors of retention, with women and individual sport athletes being retained at a higher rate than their male and team sport counterparts. Additionally, the combination of scholarship support, gender and sport type resulted in a predictive model of student-athlete retention.Of the 953 non-retained observations, 609 were male and 344 were female, and 338 competed in an individual sport while 61 5 were team sport athletes. Chi-Square Goodness of Fit tests revealed that there were more males and fewer females who were not retained than would be expected, and that there were more team sport athletes and fewer individual sport athletes not retained than would be expected (Chi-square = 5 1.058, df = 1). The findings suggest that further research be conducted on student-athletes as an individual population in regard to retention.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

June 2008

Included in

Education Commons