| Sumario: | Background: Computational thinking (CT) is a cognitive approach intricately linked with core competencies in Science, Technology, Engineering, and Mathematics (STEM). Numerous studies have explored strategies to effectively integrate CT into STEM education and systematically evaluated the multidimensional impact on student learning outcomes. Objectives and Methods: This meta‐analysis synthesises findings from 46 high‐quality empirical studies to verify the effectiveness of integrating CT into STEM on the development of students' cognitive and non‐cognitive skills and to analyse the moderating effects of various potential factors on skill acquisition. Results and Conclusions: The results demonstrate that the integration of CT in STEM education has a significant effect on enhancing students' cognitive skills (Hedges's g = 0.651, 95% CI [0.527, 0.776], p < 0.001), with this effect significantly moderated by the intervention duration and the instructional tools. Additionally, the implementation of CT in STEM positively impacts the development of students' non‐cognitive skills (Hedges's g = 0.538, 95% CI [0.345, 0.730], p < 0.001), with educational level, sample size, intervention duration, and instructional tools identified as important moderator variables influencing the acquisition of non‐cognitive skills. Based on the findings, this study recommends systematic CT interventions lasting 3 months to 1 year, utilising interactive and practice‐oriented instructional tools such as block‐based programming, maker, and unplugged activities. Furthermore, it suggests tailoring instructional strategies to different educational stages. These recommendations deepen the understanding of the role of CT in STEM education, providing practical guidance for educators and policymakers.
|
|---|