Downloads & Free Reading Options - Results

In this project, we systematically review the existing meta-analyses on Artificial Intelligence (AI) in education. The integration of AI into educational contexts is a rapidly evolving field, capturing the interest of researchers, educators, and policymakers due to the advances and challenges it poses (Kasneci et al., 2023). The applications of AI in education are diverse and include, for example, automated feedback on written text (Fleckenstein et al., 2023), chatbots for language, STEM, or general learning (Laun & Wolff, 2025; Wu & Yu, 2024), or large language models such as the Generative Pre-trained Transformer ChatGPT (Deng et al., 2025; Heung & Chiu, 2025; Kasneci et al., 2023; Muhammad Iqbal Amrullah et al., 2024). Despite this being a relatively young field of research, several meta-analyses have been conducted, exploring various aspects of AI in education (e.g., A. Chen et al., 2025; Heung & Chiu, 2025; Hu, 2024; Wu & Yu, 2024; Zhu et al., 2025). However, there exists a need for a comprehensive synthesis of these findings to understand the current landscape and identify areas in need of further investigation (L. Chen et al., 2020). Prior second-order reviews have focused on higher education or on synthesizing systematic reviews (i.e., Bond et al., 2024; Fu et al., 2024) and did not attempt to review current meta-analyses in K-12 education. The present study aims to fill this gap by conducting a systematic review of existing meta-analyses of interventions and experiments on AI in K-12 and higher education. By mapping the existing meta-analytic research, we aim to provide insights into the diverse users, outcomes, domains, moderating variables, and methodological quality of AI in education. We focus on meta-analyses aggregating the results from interventions or experiments with AI (e.g., AI-based interventions, use of AI chatbots) to make causal inferences possible. Additionally, we will review effect sizes of the impact of AI interventions and common moderators in meta-analytic research on AI in education to broadly inform researchers and educators of potentially important variables influencing the effectiveness of AI use in education. Finally, we will assess the methodological quality of the meta-analyses on AI in education. Thus, this study is aimed at providing a comprehensive overview of AI in education, answering the following research questions based on meta-analytic research aggregating interventions and experiments on AI in education: 1.1. Who is using AI in education? 1.2. What are common outcomes of experimental and intervention research on AI in education? 1.3. In which domains (e.g., language, STEM) is AI in education commonly researched? 1.4. How large or small are the effect sizes of meta-analytic research on AI in education 1.5. What are the common moderators in meta-analytic research on AI in education? 1.6. What is the methodological quality of the meta-analytic research on AI in education?