Mental simulations to facilitate teacher learning of ambitious mathematics instruction in coaching interactions

Abstract

Abstract Background Many studies have shown that ambitious, “student centered” approaches to STEM instruction benefit K-12 student learning. However, relatively little research has systematically investigated the learning processes that support teachers to skillfully enact these challenging pedagogies. In this study, we used a mixed-methods, case-comparison design to examine one kind of teacher learning routine, Mental Simulations for Teacher Reflection (MSTR), for advancing robust teacher learning in the context of one mathematics-focused instructional coaching intervention. Specifically, this study draws from a large, state-wide representative dataset to select carefully matched, contrasting cases to analyze the quality of coach–teacher conversations for teachers who showed very similar baseline instructional quality but then large differences in levels of improvement. We began by qualitatively coding detailed transcripts from selected coach–teacher pairs as they reflected on lesson artifacts (i.e., lesson plans, student work, and coach observations) using MSTR as an analytical lens. Next, quantitative analyses were conducted to determine the extent to which mental simulations characterized significant differences in the conversations of high- vs. low-instructional growth pairs. Lastly, additional qualitative analyses explored finer-grain distinctions in the quality of mental simulation talk in high- vs. low-growth pairs. Results Quantitative analyses showed high-growth pairs were significantly more likely to engage in mental simulation talk compared to their low-growth counterparts. Moreover, the high-growth pairs were much more likely to initiate (i.e., raise an instructional ambiguity or problem for discussion) as well as complete (i.e., generate and weigh alternative instructional strategies) a MSTR routine. Qualitative analyses further revealed that engaging teachers’ in-depth pedagogical reasoning to connect specific teaching moves to conceptual learning goals in mental simulations was a key distinction of the high-growth coaches. Conclusions These findings indicate MSTR captured meaningful variation in coaching quality in this context. Notably, all coaches discussed the same instructional topics with teachers (i.e., teaching–learning goals and dimensions) and engaged in the same training that did not explicitly include MSTR, suggesting the possibility that MSTR captured a more implicit process of effective coaches. This study thus offers insight into the ‘black box’ of teacher learning and how it can be supported in similar professional learning contexts.