Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Michael F. Burnett


The purpose of this study was to determine if a model existed which significantly increased the researcher's ability to accurately explain whether or not a recruited student will enroll based upon current recruitment strategies and demographic characteristics. The population for this study was defined as all prospective freshmen students who were recruited to attend Louisiana State University in the fall of 1995 and fall 1996. A random sample of the population was drawn from the population of prospective high school graduating seniors on the admissions data base for the years 1994-96. Each recruitment year sample was stratified into three groups of approximately 600 each. The instrument used in this study was a computerized recording form. Thirty-five variables were analyzed for the 1995 recruitment class and 42 variables were reviewed for the 1996 recruitment class. Data was collected by copying the variables of interest from the undergraduate admissions data base onto the established recording form file. Findings revealed that substantively and statistically significant models exist which improved the researcher's ability to accurately explain enrollment status. The variable which had the highest correlation with enrollment was the number of mail pieces sent to each student. Discriminant analysis was used to identify models which explained from 28% to 60% of the variance of the factors affecting student enrollment. In addition, the models correctly classified between 71% and 88% of the cases. Variables which contributed significantly included: the number of mail pieces a student received, whether or not they were awarded a scholarship, the amount of the scholarship, whether or not they received a response to receipt of their ACT score, and whether or not they were contacted by an LSU Ambassador. The researcher recommended refinement of the modeling process and officials at LSU to engage in further study to assist in the explanation and prediction of enrolling students. It was also recommended that admissions professionals in public, comprehensive universities investigate using this modeling process for enrollment planning.