Using data from the National Education Longitudinal Study of 1988 and the Educational Longitudinal Study of 2002, I analyze the association that students' extracurricular activities have with two academic outcomes: mathematics achievement test scores and college expectations. I consider changes between cohorts and gender differences within each cohort. While prior achievement and expectations have the strongest associations with the dependent variables, relationships do appear between extracurricular participation and the dependent variables. There is only one cohort difference in the association between activities and math achievement test scores; computer use was positively associated with students' scores in the 1990s, but had no association with students' scores in the 2000s. School-sponsored activities were associated with higher scores for all students, while television-watching and hanging out with friends were negatively associated with them. School activities were positively associated with the college expectations of students in both cohorts, while spending time hanging out with friends was negatively associated with expectations, especially for students in the later cohort. There were a few gender differences in the association between activities and math scores, but no gender differences in the association between activities and college expectations. The findings indicate that many of the same activities are relevant for math achievement and college expectations in the two time periods. I am grateful for the comments of