We present a multiple spells-competing risks model of stopout, dropout, reenrollment, and graduation behavior. We find that students who experience an initial stopout are more likely to experience subsequent stopouts (occurrence dependence) and are less likely to graduate. We also find evidence of the impact of the length of an initial spell on…

Investigated the timing of dropout by college students over a 5-year period. Found that factors that affect student dropout often have effects that change over time. For instance, students who receive financial aid generally have lower dropout rates than non-aided students, but dropout rates vary depending on the amount and timing of student…

This study attempted to track students (n=2,077) who entered the University of Minnesota (Twin Cities campus) in 1986 but had not completed a degree at that institution within eight years of matriculation. The study utilized the Minnesota Higher Education Services Office's statewide database which allowed researchers to determine if, when, and…

Asserting that graduation and stopout are "competing" or correlated events and that they often should be modeled as such, this study demonstrated that factors affecting timely graduation (e.g., financial aid) often have time-varying effects, and that ignoring these time-varying effects can lead to spurious conclusions that may result in…

This study describes the use of event history modeling as a tool for understanding student departure from college. Using data from the National Center for Education Statistics High School and Beyond (HS&B)/Sophomore Cohort longitudinal study a detailed history of individual students' college careers was constructed using a regression-like…