This study was planned as a program in which anthropologists and educators would work together in an effort to develop a conceptual framework and set of methods for ongoing field research in actual educational settings. The volume entitled "Readings," not yet published, includes articles designed to be read concurrently with each section of this volume. A major objective of the study was the development of training materials to be used in preparing anthropologists and educators to use field research as a means of problem identification and data gathering in education. Three high schools were examined as social units that operate according to an observable code of rules. The high schools were analyzed at three levels: (1) the sturctural level at which the whole school organization is taken as the unit of analysis; (2) the level or role behavior at which generalized patterns of behavior are taken as the unit of analysis; and (3) the level of individual strategies in response to the pattern of social constraints described at the previous two levels. Fifty-three incidents of classroom behavior were examined to help draw conclusions regarding schools as social systems. It is concluded that there will always be an informal network of power relations in the school that are not named in the formal organization chart. Secondly, there is always the possibility that one role will be in conflict with another role thus causing limitations on roles. Finally it is in the institution's interest to formalize emergent roles and incorporate them into the structure of the school. (SK)