Abstract
This article reports the results of a validation study of Hudson's Teaching Evaluation Form (TEF), a measure of teaching effectiveness based on student evaluations. Analyses of data from two independent pilot tests (involving 1,707 students, Spring/Summer 1994) and from three schools of social work (1,649 students, Spring 1996) showed that internal consistency reliability coefficients were extraordinarily high across the five samples. The results of factor analysis demonstrated the content and construct validity of the TEF. Finally, comparison, through logistic regression, with deans' ratings of teaching effectiveness strongly supported the convergent construct validity of the instrument.