Three types of subject-specificity. Exploring differences in the measurement of teaching quality across Norwegian mathematics and language arts classrooms

Measuring teaching quality is important to better understand student learning in classrooms. To directly capture teaching quality constructs, subject-specific and subject-generic observation systems have been developed, with benefits and challenges on both sides. However, we show that subject-specificity may address different things and come with different implicit assumptions about what is subject-specific. As a common ground for empirical work, we draw on a framework of measurement equivalence and discuss three nested claims on subject-specificity. Then, we analyze videotaped lessons from 47 mathematics and 49 language arts classrooms which were scored with the Protocol for Language Arts Teaching Observation (PLATO). Results indicate that scores in Instructional Scaffolding show the largest differences between subjects. We argue that subject-specificity is a claim about a theoretical construct, which should be clearly addressed in future research. • We discuss three nested levels of subject-specificity researchers typically assume. • Videotaped lessons from mathematics and language arts classrooms were analyzed. • Scores in Instructional Scaffolding show the largest differences between subjects. • It is argued that subject-specificity is a claim about a construct of interest.