One of the Task Force's most important tasks is to recommend activities and research that NIAAA could sponsor to support colleges and universities in their efforts to change the culture of drinking on campus. Backed by the NIH reputation for rigorous scientific research, NIAAA is in a unique position to foster careful studies of underage and excessive college drinking. Results from such efforts would enable campus and community policymakers to speak with greater confidence about the causes and consequences of the problem and its possible solutions. The Task Force urges NIAAA and the Congress to expand funding to support these vital research endeavors on as many campuses as possible.

From the Task Force's perspective, NIAAA should assume primary responsibility for:

• Supporting the research community's efforts to address existing knowledge gaps and alter the culture of drinking on campus;
• Facilitating long-term, campus-community research aimed at preventing hazardous student drinking; and
• Imparting what is known about the patterns of college drinking and the quality of current interventions to encourage college presidents, administrators, and other campus and community leaders to adopt policies and implement strategies based on research.

The Task Force grouped its recommendations for NIAAA by these three functional areas.

(1) Supporting Development of Improved Research Methods

Improved data collection and extrapolation methods will help equip college administrators to assess the dimensions of the problem on their campuses and understand their situation in comparison to others. To support this activity, the Task Force recommends that NIAAA:

• Design and implement one national surveillance and data system for all colleges and universities to establish reliable estimates of the magnitude of the problem; provide mechanisms to track nationwide changes; assist colleges in monitoring their own campuses; and facilitate intercampus research.
• Take the leadership role in working with other relevant agencies and organizations, researchers, and college administrators to support development of a range of state-of-the-art screening and assessment measures for use by colleges as well as researchers.

The Need for Longitudinal Studies

The overwhelming majority of studies on college student drinking assess students at a single point in time. Although these "cross-sectional" snapshots provide useful information concerning the extent that two factors such as heavy drinking and fraternity membership are correlated, they cannot specify the nature of the causal relationship between the two. For example, if heavy drinking is associated with fraternity memberships, it could be because:

• Greek residence life facilitates drinking (i.e., socialization);
• Heavier drinkers differentially affiliate with Greek organizations because of the drinking opportunities they may afford (i.e., selection); or
• Other factors such as personality traits promote both affiliation and drinking.