The major problem faced by those who would evaluate teaching has been the difficulty of defining teaching effectiveness or establishing criteria for measurement. Traditionally, evaluations have focused on teacher performance in the classroom or on teacher traits as the criteria of effectiveness. However, previous studies have indicated that learning can occur in the presence of different teacher traits ("warm" or "cold", "authoritarian" or "permissive") and as a result of various methods (lecture, group discussion, or no teaching activity at all). Hence, the literature has promoted evaluation based on a "teaching-learning" paradigm which requires that teaching be defined as the activity which causes learning. A previous study, conducted by the author and based on questionnaires sent to 933 full-time faculty and administrators in nine randomly selected Illinois Public Community Colleges, indicated that though teacher performance was ranked as the number one basis for evaluation, 59% rated student learning as the best single indicator of teaching effectiveness. However, these same respondents also indicated that they believed that there was a connection between student learning and teacher behavior and that effective teachers did things differently in the classroom than non-effective teachers. (MB)