Objective: To evaluate a widely used intervention to reduce college student alcohol use, we studied student drinking patterns at colleges that employed social norms marketing programs and those that did not. Method: We examined responses of students in the Harvard School of Public Health College Alcohol Study (CAS) 1997, 1999 and 2001 data sets at 37 colleges that employed social norms marketing programs and at 61 that did not. Information about the students' drinking behavior and their familiarity with social norms marketing messages at their schools was analyzed, as were college administrators' reports about the implementation of social norms marketing campaigns. Schools were grouped on the basis of student reports of exposure to programmatic materials. Trend analyses were conducted on seven standard measures of alcohol consumption, including annual and 30-day use, frequency, usual quantity and volume consumed heavy episodic use, and drunkenness. Results: Almost half of the CAS colleges sampled adopted social norms programs. Those that did were more likely to have large enrollments, not to be religiously affiliated and to have high rates of alcohol use. No decreases were noted in any of the seven measures of alcohol use at schools with social norms programs, even when student exposure and length of program existence were considered. Increases in measures of monthly alcohol use and total volume consumed however, were observed at schools employing social norms programs. Conclusions: This study does not provide evidence to support the effectiveness of social norms marketing programs, as currently utilized, in reducing alcohol use among college students.