This paper reports the results of a management study of the effectiveness of middle schools in a large, urban, midwestern public school system. Results of the study illustrate the conundrum that all educational reforms are not complementary, i.e., they are at cross-purposes because they are designed to obtain different outcomes. The two reform agendas studied were site-based management with decentralization of decision making about instructional and curricular change and an intention of improving student achievement as measured on a state criterion-referenced test. The study included 11 middle schools employing 400 teachers, with 91 instructional aides, serving 6,845 students. Two principal investigators and five research assistants from a local university reviewed documents, interviewed personnel, parents, and students, and visited the schools. Curriculum management audit standards developed by L. Frase, F. English, and W. Poston (1994) were used to assess school effectiveness. A number of problems with middle school functioning were found. A major implication of the school-by-school review was that the critical balance between systemwide and school-based responses needed to obtain system-wide objectives was missing. None of the schools used data from student performance instruments to link back to validated school-wide performance measures, and no school had taken the initiative to develop assessment measures beyond the state-mandated testing program. No systematic procedures were in place to monitor curriculum or instruction, and at-risk students were not doing well even in the high performing schools. Minority students received a proportionally higher percentage of disciplinary measures than their white counterparts, and sever curricular inequities were found among the schools. Recommendations are made for systematic planning, a core curriculum, administrator accountability, and better responses to low student performance. (SLD)