The study takes advantage of the nontraditional selection process of the Gates Millennium Scholars (GMS) program to test the association between selectivity of 4-year institution attended as well as other noncognitive variables on the college completion rates of a sample of students of color. The results of logistic regression and propensity score matching suggest these students are slightly more likely to graduate from college if they attend a highly selective institution. There is also evidence that other noncognitive variables such as leadership are good predictors of college completion. This suggests that admission offices interested in attracting a more diverse student body might want to consider expanding the traditional admission criteria. (Contains 12 notes and 2 tables.)