aceModerate evidence
Published recordEducationGood Space
More
Can Generative AI Feedback Effectively Enhance Learning Outcomes? A Meta-Analysis of 36 Experimental and Quasi-Experimental Studies
- Truvace record
- TRV-2026-2785raw data (JSON)
- Version
- v1Published
- First published
- Evidence type
- Systematic review or meta-analysis
- Sources
- 1 sourcePeer-reviewed
- sha256
- 82685272e570…check it
Although generative artificial intelligence (GenAI) feedback shows promise for educational applications, its actual impact on learning outcomes and the factors influencing its effectiveness remain unclear. This study conducted a systematic review and meta-analysis to evaluate the effectiveness of GenAI feedback and identify key moderating factors. Following the PRISMA 2020 guidelines, we reviewed 36 experimental and quasi-experimental studies published between 2023 and 2025, yielding 72 effect sizes. The results…
The gain
A meta-analysis of 36 experimental and quasi-experimental studies (72 effect sizes, 2023 to 2025) found generative AI feedback had a moderate positive effect on academic achievement (g = 0.61).
What this doesn’t fix
Benefits for metacognitive development are described as promising but less established, and effects on non-cognitive outcomes were modest, so the authors call for further research on those outcomes.
Sources
1 on file- Education Sciencesdoi:10.3390/educsci16060816
Cite this record
v1Truvace Impact Record TRV-2026-2785, v1: “Can Generative AI Feedback Effectively Enhance Learning Outcomes? A Meta-Analysis of 36 Experimental and Quasi-Experimental Studies.” Truvace, 2026-11-12. /record/TRV-2026-2785 (accessed on the day you copy it). sha256 82685272e570eaff…
Change history
1 versionEvery change made to this record since it was first published, and why. Nothing has changed since it went up.
Certified into the record
Check this record hasn’t been altered
How to check it yourself, without trusting this page
Download any version from /api/record/TRV-2026-2785 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.