aceModerate evidence
More
Review of 10 studies finds AI adds to flipped classes' test gains
Study title: Artificial Intelligence-Supported Flipped Learning and Academic Achievement: A Systematic Review and Meta-Analysis with Implications for Sustainable, Quality Education
- Evidence type
- Systematic review or meta-analysis
- Sample
- n = 937
- Sources
- 1 sourcePeer-reviewed
- Published
- 2 min read
- Truvace record
- TRV-2026-2764cite or check it
Study title: Artificial Intelligence-Supported Flipped Learning and Academic Achievement: A Systematic Review and Meta-Analysis with Implications for Sustainable, Quality Education
A systematic review and meta-analysis pooled ten experimental and quasi-experimental studies with 937 learners to test what AI adds to flipped learning. Compared with flipped classes that did not use AI, AI-supported flipped classes produced a pooled effect of g = 0.67 on academic achievement (95% CI 0.32 to 1.02).
Read the study abstract
This study meta-analyzed the incremental contribution of artificial intelligence (AI) to flipped learning on students’ academic achievement, relative to flipped learning without AI. Following a protocol registered on the Open Science Framework (OSF), systematic searches were conducted in Web of Science and Scopus, supplemented by an ERIC search, with coverage through 25 July 2026; eligible studies used experimental or quasi-experimental designs comparing an AI-supported flipped condition with a matched non-AI fl…
The gain
A meta-analysis of ten experimental studies (937 learners) found students in AI-supported flipped classes scored higher on achievement than those in flipped classes without AI (Hedges' g = 0.67, 95% CI 0.32 to 1.02).
In brief
The comparison matters because the control group was itself flipped learning, so the estimate approximates the contribution of the AI component alone, not the combined effect of AI and the flipped model.
The evidence is weak in places. Results varied widely between studies, and a 95% prediction interval of −0.10 to 1.43 means a new setting could see a large effect or a near-zero one. Overall certainty, rated by GRADE, was low.
Main points
- Ten studies met the strict criteria: two randomized controlled trials and eight quasi-experimental studies, with 937 learners in total.
- Each study compared an AI-supported flipped condition with a matched flipped condition without AI, on an academic-achievement outcome.
- The pooled effect was g = 0.67 (95% CI 0.32 to 1.02, p = 0.002) in a random-effects model with the Knapp–Hartung adjustment.
- Heterogeneity was substantial (I2 = 64.6%), and the prediction interval ran from −0.10 to 1.43.
- Egger's test was significant (p = 0.01), consistent with a small-study effect; excluding studies with fewer than 15 participants per arm cut the estimate to g = 0.53.
- GRADE rated overall certainty of the evidence as low.
What the review found when it combined the studies
These are the review’s own pooled results, copied from its abstract. Truvace did not recalculate them.
Student academic achievement with AI-supported flipped learning compared with flipped learning without AI
Hedges' g 0.67 (95% CI 0.32 to 1.02)
937 people · search to 25 July 2026
The review’s own words
Using Hedges’ g in a random-effects model (REML with the Knapp–Hartung adjustment), the pooled effect was positive, moderate and statistically significant: g = 0.67, 95% CI [0.32, 1.02], t(9) = 4.36, p = 0.002.
Code checks every number above against the review’s text before it is shown. The short labels are an AI model’s reading of that text, so the review’s own sentence sits under each one. A pooled figure is only as strong as the studies behind it. How Truvace uses pooled results
The full story
The review followed a protocol registered on the Open Science Framework. Searches covered Web of Science and Scopus, supplemented by ERIC, through 25 July 2026, and only studies with a matched non-AI flipped comparator were eligible.
One further study used conventional, non-flipped instruction as its comparator. It was excluded from the primary analysis and kept only as a sensitivity check.
The authors conclude that AI can add value to flipped learning, but that the benefit depends on implementation quality and context rather than being an inherent property of the technology. They say the findings should be read with caution given a small, heterogeneous evidence base partly influenced by small studies.
Sources
1 on file- Sustainabilitydoi:10.3390/su18178715
Talk it over
Who gets this benefit, and who was left out of the study or story?
Review of 10 studies finds AI adds to flipped classes' test gainshttps://truvace.com/item/artificial-intelligence-supported-flipped-learning-and-academic-achievement-a-sy
Who gets this benefit, and who was left out of the study or story?
The gain A meta-analysis of ten experimental studies (937 learners) found students in AI-supported flipped classes scored higher on achievement than those in flipped classes without AI (Hedges' g = 0.67, 95% CI 0.32 to 1.02).
Answer in the threadHow do we know? What can this kind of evidence show, and what can’t it?
The evidence This rests on a systematic review or meta-analysis with a sample of 937, from Sustainability.
Answer in the threadWhat does it leave open? What else would you want to know before acting on it?
The record’s own caveat The evidence base is small (ten studies) and varied, with a significant small-study effect; the estimate fell to g = 0.53 when the smallest studies were excluded, and overall certainty was rated low.
Answer in the threadHave you seen AI used this way where you study, work or live? Did it go the same way?
Answer in the thread
The questions are the same on every record. The details under them are taken from this record as written.
What readers say