| The college-choice process is complex and affects many high school students, family members, and public policy-makers, as well as institutions of higher education. Despite increasing numbers of high school graduates in many states, there is little evidence of a decline in competition among four-year colleges for students. Intent to increase the number of students, as well as the quality of enrolled students, will likely result in continued or increased marketing and recruitment strategies. With the use of a successful enrollment forecasting tool, like predictive modeling, an institution of higher education can project the size and quality of an incoming class. One of the primary purposes of this study was to develop such a model to determine if potential common factors existed between those first-time freshmen who enrolled and those who did not. The model was developed based on variables collected from admissions applications submitted by prospective students to Hardin-Simmons University (HSU) during a five-year period from 1999--2003. Hardin-Simmons University is a small, private, liberal arts institution in West Central Texas.; The independent variables used in this study were housing status, anticipated area of study, high school rank, ethnicity, denominational preference, standardized test score, term of interest, gender, origin of application, and region of state. The dependent variable was, of course, enrollment in HSU.; Logistic regression was the statistical procedure used to analyze the data in this study. Logistic regression was selected due to the binary nature of the dependent variable (matriculation or enrollment). The independent variables were processed through the final predictive model to determine the level of influence each had on enrollment to Hardin-Simmons University. Seven of the 10 independent variables were found to be statistically significant in the final model.; One of the stated purposes of this study was to identify the attributes or characteristics among first-time freshmen applicants who matriculated at HSU versus those who did not. This objective was accomplished using the independent variables found to be significant in the study through a logistic regression equation. |
