This meta-analysis of 92 studies found that digital tools have a moderate positive effect on secondary school students' learning outcomes in mathematics and science. Providing teachers with training on how to use the digital tools significantly enhanced their effectiveness, while the type of tool (e.g., simulations, intelligent tutoring systems) also showed some influence, though not as strong statistically.

The authors acknowledge funding from Stiftung Mercator, a German foundation that promotes education, among other areas. While this doesn't necessarily indicate a conflict of interest, it is important to consider the foundation's mission and potential influence on research priorities.

This meta-analysis synthesizes a large body of literature on digital tools in STEM education, providing a valuable overview of the existing research. However, methodological limitations like the prevalence of quasi-experimental designs and the substantial unexplained heterogeneity limit the strength of the conclusions. The potential for publication bias and the broad definition of digital tools are also weaknesses. The identified positive effect of teacher training provides practical implications for educators and policymakers. Overall, this is a useful contribution but leaves room for improvement in future research.