The potential of research which uses the data base from the National Longitudinal Study of the High School Class of 1972 (NLS) for affecting federal policy is discussed. Currently, social science research is rarely used as a basis for policy decisions. A reason for this is that social scientists profess interest in policy issues, yet are naive in understanding how policy decisions are formulated and in knowing how to report research so that it is politically relevant. A report prepared by the Rand Corporation sees the NLS data as having potential for four stages of policy formulation: (1) early warning of forthcoming problems, (2) issue exploration, (3) identification of specific problems, and (4) evaluation of the impacts of past policy decisions. The author suggests only two stages: the exposition or confirmation of a problem or need that was previously suspected, and the evaluation of prior policy decisions or solutions that have begun to fall into disrepute. Political sophistication and an advocate/adversary stance may be the research tools of the future for social scientists. (BW)