Two primary lines of investigation are involved in this study of elementary school teacher effectiveness. The first focuses on individual differences in students and how these affect teacher expectations, attitudes, and behavior, and the process involved in the formation and change of expectations and attitudes. The overwhelming conclusion that the data supports is that the vast majority of teacher perceptions are accurate and based on student behavior. In those rare cases where persistently incorrect and dysfunctional perceptions are formed, the problem can be solved by building a data base to create understanding and awareness of how these problems develop. The second line of investigation focuses on the question of teacher effectiveness, particularly in producing student learning gains. The results of the analyses indicate that relative differences in teacher effectiveness were statistically significant and reaffirmed the feasibility of identifying highly consistent teachers and studying them to seek associations between classroom process variables and student outcomes. In studying the correlates of teaching effectiveness, one basic finding was that, for many variables, teacher behavior optimal for producing student learning gains in low socioeconomic status (SES) schools was different from teacher behavior optimal for producing learning gains in high SES schools. (JMF)