Application Of Event History Modeling For Examining College Student Departure Behavior

The purpose of this study is to provide a better understanding of the longitudinal nature of college student attrition using national data (BPS:90/94) sponsored by the National Center for Education Statistics. Many earlier studies that investigated student departure often failed to incorporate the timing of departure even though one knows that student attrition is a longitudinal process. Given the temporal nature of student departure, it is reasonable to suspect that the direction and magnitude of factors associated with departure behavior may differ over time. Using event history modeling, the study herein examined how the factors that affected student departure changed over time. For example, educational aspirations had stronger effects in the first and second years. Students who received financial aid generally had lower attrition rates than nonaided students. However, the likelihood of departure varied depending on the amount and timing of student financial aid. Additionally, examples of the longitudinal effects of departure risks are presented for students with various characteristics.