By offering information, counseling, and tutoring, college outreach programs attempt to smooth the path between high school and higher education for at-risk students. But do these program work? This paper uses longitudinal data from the Education Longitudinal Study to construct two quasi-experiments to assess the effectiveness of college outreach. The first compares outreach program participants with a propensity score matched sample of program non-participants to measure the effects of targeted college outreach programs. The second assesses the effects of school-wide college outreach programs by comparing students in school-wide outreach high schools with students in a matched sample of high schools that offer no formal outreach. The results suggest that targeted outreach programs do little to change the educational experiences of participating students. However, there is limited evidence to suggest that school-wide outreach programs may have modest "spillover" effects, improving the educational outcomes of relatively unengaged students at participating schools. (Contains 5 notes, 7 tables, and 2 figures.)