Pedagogical knowledge of working math teachers : Validation of the construct representation in the TEDS-M test

The test for pedagogical knowledge that was developed within TEDS-M (Teacher Education and Development Study: Learning to Teach Mathematics) has proven reliable in several studies for measuring future teachers’ knowledge. Across professional education phases and contexts, broad evidence has been presented for valid interpretations of the measures it provides about (future) teachers. Yet, one essential validation step has not been carried out to date: examination of the suitability of the test for quantitative and qualitative inferences about the pedagogical knowledge of in-service mathematics teachers. In the frame of a study within the project TEDS-Validate, 113 mathematics teachers were tested in order to determine whether the test allows for a reliable and differentiated measurement of their pedagogical knowledge. Following the concept of construct representation (Embretson, 1983), the study examined whether the test also requires construct-relevant cognitive processes as modeled by Konig (2009) and Klemenz and Konig (2019). The results confirm that the test works reliably with in-service mathematics teachers. They further underline that the variance in item difficulties can be explained by the complexity of the cognitive processes they require, thus providing first evidence for construct representation and the basis for qualitative interpretations of teachers’ test results. Finally, these qualitative interpretations were validated via variance analyses showing that teachers who apply more complex cognitive processes in the pedagogical knowledge test also show higher situation-specific pedagogical skills