Abstract

Survival analysis allows investigators to study the timing and duration of a critical event, which is a binary and dichotomous measure. This study used the stratified Cox proportional hazards model, a branch of survival analysis, to establish the relationship between a specific student learning outcome and its relevant explanatory variables. The outcome variable of interest was the timing of experiencing academic difficulty--dismissal, withdrawal, and leave of absence--due to academic reasons. The explanatory variables included demographics, undergraduate GPAs, MCAT scores, medical school academic performances, medical school curriculum tracks, and financial aid loan amounts. The main focus of this study was to measure the effect (relative hazard) of specific explanatory variables on the academic difficulty after adjusting for other explanatory variables. The data analysis indicated that academic difficulty was significantly associated with the MCAT verbal reasoning score, number of sophomore courses failed, and other relevant variables. By identifying the occurrence of critical events along with the explanatory variables, college decision makers could implement intervention strategies to ensure student success.