This study explores the effects of an academic seminar on the persistence of first-year college students, including effects on students most at risk of dropping out. A secondary interest was demonstrating the utility of using classification and regression tree analysis to identify relevant predictors of student persistence. The results of the study showed that persistence rates improved for all students who took the seminar and, most importantly, that the persistence rate for higher-risk students showed double the improvement of that of lower-risk students. Results also suggest that changes in student perception of the value of a college education and engagement in classroom discussion are factors for improving the persistence of higher-risk students.