This study was designed to examine student achievement in mathematics by means of curriculum-based tests. The results of the 1979 New York State Regents examination in Eleventh Grade Mathematics were obtained from a representative sample of schools throughout the state. The expectations were that there would be substantial differences in achievement within schools and that between-school differences would be similar to those found in previous studies. The data from 64 schools were analyzed in a number of ways. Results for 127 teachers with more than 10 students were summarized. Mean scores, by teacher, were presented for 13 schools which had four or more teachers. In addition, achievement in different school districts was compared. The results showed surprising ranges of differences within schools. A structural model of variables influencing student learning was presented to suggest explanations for these differences. Follow-up research might focus upon specific within-school characteristics hypothesized to be related to learning outcomes. (Author)