(When) do teacher professional development interventions improve student Achievement? A meta-analysis of 128 high-quality studies

Enormous sums of money are spent worldwide on teacher professional development (TPD), seeking to optimize the quality of in-service teachers’ teaching. An important question is whether this pays off in terms of improved student achievement, which we investigated in this meta-analysis of 128 (quasi-)experimental, high-quality studies with 356 effect sizes. Results showed an average effect size of 0.09, 95 % CI [0.07, 0.11], considered a medium-size effect on student achievement (Kraft, 2023). Effects varied greatly, with the 95 % prediction interval ranging from −0.15 to 0.32. We used confirmatory and exploratory multiple meta-regression models to examine the effects of potential moderators. The confirmatory model revealed that grade level and publication status were predictors of the effect, while TPD goal did not explain differences in effects on student achievement. The exploratory model seemed to indicate that the number of learning-theory principles applied in designing the TPD (out of these four: teacher performance standards, teacher self-regulation, teacher coaching and teacher cooperation) positively influenced student outcomes. Interestingly, the number of TPD hours and the type of trainer were not statistically significant predictors of TPD effectiveness. We reflect upon the findings and the state-of-the-art of TPD research and formulate recommendations for its further development.