The purpose of the School Improvement Model (SIM) Project is to investigate the links between systematically developed teacher performance evaluation, administrator performance evaluation, staff development interventions, and the quality of education as measured by student achievement results. Following the 4th year of the 5-year study, preliminary findings made possible the development of a model of the effects of SIM project activities on student performance, based on results obtained in the Minneapolis Public Schools. Fourth-grade mathematics scores on norm-referenced tests served as measures of achievement, and both the classroom level and the individual student level of improvment were tested, using analysis of covariance. Variables tested included socioeconomic status, race, and attendance. Socioeconomic status and race were found to predict both pretest and posttest scores, but evidence indicated that the SIM process reduces the impact of these variables and does affect achievement in a metropolitan school setting. (PGD)