Background: The seven categories of the Stanford Faculty Development Program (SFDP) represent a framework for planning and assessing medical teaching. Nevertheless, so far there is no specific evaluation tool for large-group lectures that is based on these categories. This paper reports the development and psychometric validation of a short German evaluation tool for large-group lectures in medical education (SETMED-L: 'Student Evaluation of Teaching in MEDical Lectures') based on the SFDP-categories.
Methods: Data were collected at two German medical schools. In Study 1, a full information factor analysis of the new 14-item questionnaire was performed. In Study 2, following cognitive debriefings and adjustments, a confirmatory factor analysis was performed. The model was tested for invariance across medical schools and student gender. Convergent validity was assessed by comparison with results of the FEVOR questionnaire.
Results: Study 1 (n = 922) yielded a three-factor solution with one major (10 items) and two minor factors (2 items each). In Study 2 (n = 2740), this factor structure was confirmed. Scale reliability ranged between α = 0.71 and α = 0.88. Measurement invariance was given across student gender but not across medical schools. Convergent validity in the subsample tested (n = 246) yielded acceptable results.
Conclusion: The SETMED-L showed satisfactory to very good psychometric characteristics. The main advantages are its short yet comprehensive form, the integration of SFDP-categories and its focus on medical education.
Keywords: Evaluation; Lecture; Medical education; Psychometrics; Questionnaire; SETMED-L.