Using an Integrated Persistence Model to Predict College Graduation

Many researchers have explored the relationships between the likelihood of graduating from college and demographic and pre-college factors such as gender, race/ethnicity, high school grade point average (GPA), and standardized test scores. However, additional factors such as a student’s college major, home address, or use of learning support in college have been examined to a far lesser degree. This study seeks to add these factors to an integrative persistence model in order to examine their impact on predicting college graduation in a six-year timeframe. Results indicate that students with in-state home addresses are more likely to graduate within six years than students with out-of-state home addresses, when controlling for other factors. Findings also suggest that graduation rates vary considerably for different majors and for those using learning support such as tutoring and Supplemental Instruction in college. Therefore, these additional factors become important for institutions to consider, particularly as it applies to implementing new programs, expanding programs proven effective, and/or targeting specific populations of students in order to help them persist to timely college graduation.