Sociocognitive Approach to the Validity of Teaching Assessment Instruments

Abstract

The present study investigates the construct validity of Student Evaluation of Teaching (SET) questionnaires using a sociocognitive approach and rater-mediated assessment methods. The study employed many-facet Rasch measurement (MFRM) and a SET questionnaire to evaluate the quality of teaching in medical courses, accounting for the variability resulting from rater severity and item difficulty. The Central Model of Student Perception (CSP) is a theoretical framework employed to analyze socio-cognitive factors influencing student responses. The findings indicate that students possess the capacity to differentiate between the measured dimensions and correctly interpret the questionnaire items. However, the severity of their responses exhibits considerable variability. The study suggests that SET questionnaires can yield valuable information regarding teaching quality, but only if appropriate instrument design, administration protocols, and data collection and analysis methods are used. The CSP, along with advanced analysis methods like MFRM, provides a flexible approach for evaluating and researching teaching quality.