Admission type is a paradoxical predictor of performance in problem-based learning courses

To determine if a difference in problem-based learning (PBL) performance or student attrition exists based on admission type. This retrospective review analyzed admission type and student academic performance in the PBL course series from 2002 to 2009. Student scores were compiled based on admission type: provisional (high school), traditional with degree (bachelor's or higher), and traditional without degree (some college). ANOVA was performed on overall scores amongst the three admission types. Continuous data were analyzed using two-sided unpaired t-tests, and dichotomous data were analyzed using chi-square test. 600 students received at least one score in the third-year (P3) PBL course series. 56% of students (n = 333) were admitted traditional without degree, while the remainder were split between provisional (n = 125) and traditional with degree (n = 142) admission. Provisionally admitted students had significantly higher scores than either of the other groups (p