A Logistic Regression Analysis of Student Experience Factors for the Enhancement of Developmental Post-Secondary Retention Initiatives

In response to stagnant undergraduate completion rates and growing demands for post-secondary accountability, institutions are actively pursuing effective, broadly applicable methods for promoting student success. One notable scarcity in existing research is found in the tailoring of broad academic interventions to better meet the specific needs of students from known risk populations. The purpose of this correlational study was to investigate possible predictive relationships among three specific pre-matriculation characteristics (gender, ethnicity and secondary school type) and subsequent academic year retention for residential undergraduate students that completed a developmental education course at a private, liberal arts university. A logistic regression was conducted to examine predictive relationships between these factors for the purpose of establishing the need for more data-based intervention strategies. Student archival data of 1,505 residential undergraduate students at a private, liberal arts university was collected from the institution’s database, and included demographic and enrollment data for identifying retention or attrition among students. Analysis revealed that neither the gender nor ethnicity models produced statistically significant predictive relationships in contrast with U.S. national enrollment trends, though two specific ethnicities, African American and Caucasian were significant. Secondary school type showed a significant predictive relationship in favor of private school students in predicting a positive enrollment response to developmental coursework. These results provide insight into the usefulness of traditional risk factors when applied to the intervention process, as well as meaningful data for the population-specific evaluation of the developmental coursework in terms of promoting year-to-year retention