The underlying assumption that the supervision of teachers has posi tive effects on subsequent teacher classroom behavior has not been sub jected to a rigorous empirical test. The findings of this study support the practice of goal-oriented supervision and do not isolate any differential effects of direct and indirect supervisory styles.Yvonne M. Martin is Assistant Professor of Educational Administra tion, Geoffrey B. Isherwood is Associate Professor of Educational Ad ministration, and Socrates Rapagna is Associate Professor of Education al Psychology and Sociology, McGill University.