A CONCEPTUAL MODEL FOR STUDYING THE OPERATION OF SOCIAL POWER IN SCHOOL SYSTEMS WAS DEVELOPED ON THE ASSUMPTION THAT A SCHOOL SYSTEM MUST BE UNDERSTOOD IN TERMS OF ITS SUPPORTING ENVIRONMENT. TWELVE DIRECTIONAL HYPOTHESES WERE TESTED TO DETERMINE CORRELATIONS BETWEEN FOUR TYPES OF COMMUNITY POWER STRUCTURES (DOMINATED, FACTIONAL, PLURALISTIC, AND INERT), FOUR KINDS OF SCHOOL BOARDS (DOMINATED, FACTIONAL, STATUS CONGRUENT, AND SANCTIONING), AND FOUR KINDS OF SUPERINTENDENT'S ROLES (FUNCTIONARY, POLITICAL STRATEGIST, PROFESSIONAL ADVISOR, AND DECISIONMAKER). FINDINGS GENERALLY SUPPORTED THE STUDY'S MAJOR THESES OF A POSITIVE CORRELATION BETWEEN SIMILAR TYPES OF COMMUNITIES AND SCHOOL BOARDS AND A DIVERSE SET OF RELATIONSHIPS BETWEEN KINDS OF SUPERINTENDENT'S ROLES, WHEN CORRELATED WITH TYPES OF COMMUNITIES AND SCHOOL BOARDS. DATA FOR CHI SQUARE ANALYSIS WERE OBTAINED BY MEANS OF FOCUSED INTERVIEWS AND TAPE RECORDED RESEARCH CONDUCTED IN 51 COMMUNITIES IN THE NORTHEAST AND MIDWEST REPRESENTING 11 LARGE URBAN COMMUNITIES (ABOVE 25,000 POPULATION), 14 SMALL URBAN (2500-24,999), 10 SUBURBAN, AND 16 RURAL (UNDER 2500). TWO SPECIFIC RECOMMENDATIONS WERE FORMULATED FROM THE STUDY TO REDUCE THE SCHOOL SUPERINTENDENT'S VULNERABILITY TO SHORT TERM DEMANDS--(1) THE SCHOOL SUPERINTENDENT SHOULD BE GIVEN AT LEAST A THREE-YEAR CONTRACT RENEWABLE ANNUALLY, AND (2) STATE DEPARTMENTS OF EDUCATION SHOULD ASSUME RESPONSIBILITY FOR THE MANAGEMENT OF SCHOOLS AT THE LOCAL LEVEL. (JK)