Community College Student Degree Completion using Classification and Regression Tree Analysis

Abstract

Community college leaders often look to institutional researchers to provide information to better inform their decision-making process, which guides local policy making and practice. The purpose of this paper is to acquaint community college institutional researchers with a data visualization technique called classification and regression tree (CART) analysis and demonstrates how it may be used to communicate student degree completion findings to internal and external audiences. The CART analysis provides a data-driven pictorial representation of how successful community college students are at earning an associate degree based on demographic, financial and academic variables. Using a statewide dataset, this study found grade point average, being an underrepresented minority, KEES, earned credit ratio, age, enrollment intensity, Federal Pell Grant, and gender were statistically significant variables of associate degree completion. Implications of these findings and the application of this technique are discussed.