Acceso usuarios
Noruega
Reino Unido
Jyväskylä, Finlandia
In a cluster randomized control trial, we investigated the effects of video-based teacher coaching (INTERACT) on the quality of teacher–student interactions (TSIs) and student engagement among upper secondary school students (n = 1,341). Ninety-eight teachers from 12 schools were randomly assigned to the intervention group (n = 49) or the practice-as-usual group. Teachers in the intervention group were assigned to a coach (10 coaches in total) and participated in seven coaching cycles throughout the 2022–2023 school year. Multilevel intent-to-treat (ITT) analysis indicated that the intervention had small effects, with statistically significant improvements in the three TSI domains: emotional support, classroom organization, and instructional support. However, the results indicated no significant effect or discernible impact on student emotional, behavioral, or cognitive engagement. Multilevel complier average causal effect (CACE) estimation using a combined dosage and fidelity compliance marker increased the effect sizes for the TSI domains to medium to large, and small to medium effects were documented for emotional and behavioral engagement according to Cohen’s standards. For this type of intervention, these effects are considered large. For observed student engagement, the ITT analysis indicated a small intervention effect, and the CACE analysis indicated a medium effect; however, none of these effects were statistically significant. Using ITT and CACE analyses, we document both the effects of offering teacher coaching versus practice-as-usual independent of the quality of implementation, and the effects for classrooms in which the teacher implemented the intervention with high quality. Practical implications for policy and practice, as well as implications for future research, are discussed.
