Instructional coaching actions that predict teacher classroom practices and student achievement

Although there is emerging evidence supporting the efficacy of instructional coaching models used to support teachers in schools, additional research is necessary to investigate the individual contributions of coaching actions for specific approaches. This study investigated the extent to which three coaching actions-modeling, practice, and feedback-provided during data-driven coaching predicted (a) teachers' intervention implementation fidelity, (b) teachers' use of evidence-based classroom strategies, and (c) class-wide student achievement. In addition, the study explored the potential mediating influence of teachers' intervention implementation fidelity on the relationship between coaching actions and teachers' use of evidence-based strategies. Participants included 16 instructional coaches, one from each of 16 high-poverty charter schools, and 133 coached K-12 classroom teachers assigned to 3649 students. Results from multiple regression analyses indicated that the provision of practice opportunities during coaching was a statistically significant predictor, with greater practice opportunities associated with greater teacher intervention implementation fidelity and greater class-wide student English Language Arts (ELA) and math achievement. Feedback during coaching predicted student ELA but not math achievement. Both practice and feedback opportunities predicted better teacher use of evidence-based classroom strategies. In addition, teachers' intervention implementation fidelity mediated the relationship between teacher practice opportunities and teachers' use of evidence-based classroom strategies. Modeling was not found to be a statistically significant predictor. Study limitations and implications for practice and future research are described.