Effects of digital game-based STEM education on students’ learning achievement: a meta-analysis
Open accessThe DOI matches its Crossref record; 79 of the 83 references it cites were found in a registry or an index; 2 of the 12 sources its claims cite were checked for support.
Abstract
Many researchers have explored the impact of digital games on learning effects in different STEM subjects. The purpose of this meta-analysis is to examine the effect of digital game-based STEM education on the learning achievement of K-12 or higher education students. The analysis results of effect sizes from 33 studies ( N = 3894) published from 2010 to 2020 showed that digital games contributed to a moderate overall effect size (ES = 0.667, 95% CI [0.520–0.814], p < 0.001) when compared with other instructional methods. Furthermore, the study explored multiple moderator variables and their potential impacts on learning outcomes such as control treatment, subject discipline, educational level, game type, gaming platform, and intervention duration. The findings suggest that digital games are a promising pedagogical method in STEM education that effectively improves learning gains. Additionally, the study concludes with three recommendations for future research and practices on digital games in STEM education.
Overview
CoChat Summary
TL;DR
This meta-analysis found that digital game-based STEM education had a moderate, significant positive effect on students’ learning achievement versus other instructional methods. Across 33 studies (36 effect sizes; N = 3,894), the overall effect size was ES = 0.667 (95% CI [0.520–0.814], p < 0.001).
Key findings
- Overall, digital game-based STEM instruction produced a moderate positive effect on learning achievement: ES = 0.667, 95% CI [0.520–0.814], p < 0.001.
- Compared with control treatments, digital game-based instruction outperformed both multimedia (ES = 0.848, p < 0.001) and traditional instruction (ES = 0.558, p < 0.001).
- By subject, effects were positive across science (ES = 0.750, p < 0.001), mathematics (ES = 0.629, p < 0.001), and technology/engineering (ES = 0.367, p = 0.140), with no significant difference among disciplines (Qb = 2.188, p = 0.335).
- By education level, primary education showed the largest effect (ES = 0.835, p < 0.001), compared with secondary (ES = 0.487, p < 0.001) and higher education (ES = 0.492, p < 0.05); the subgroup difference was not significant (Qb = 5.184, p = 0.075).
- By game type, board games had the largest effect (ES = 1.455, p < 0.001), followed by tutorial games (ES = 0.593, p < 0.001) and immersive games (ES = 0.583, p < 0.001); subgroup differences were significant (Qb = 11.126, p < 0.01).
- By platform, computers (ES = 0.625, p < 0.001) and mobile devices (ES = 0.768, p < 0.001) both improved achievement, with no significant difference between them (Qb = 0.752, p = 0.386).
Methods
- Meta-analysis conducted under a PRISMA-guided search of six databases: Web of Science Core Collection, ERIC, ProQuest, Springer, Scopus, and Wiley.
- Included studies were English-language peer-reviewed journal articles published from 2010 to 2020, focused on K-12 or higher education STEM settings, used an intervention with a digital game treatment and a comparison group, and reported sufficient data for effect-size calculation.
- The final sample was 33 studies with 36 effect sizes and 3,894 participants; analyses used standardized mean difference, heterogeneity tests (Q and I2), random-effects models, and moderator analyses for control treatment, subject discipline, education level, game type, gaming platform, and intervention duration.
Limitations
- The paper states that the meta-analysis included only 33 empirical studies and 36 effect sizes, and that relevant studies may have been missed.
- The paper states that it used a random-effects model rather than a fixed-effects model, and that the review was not fully inclusive but aimed to provide a systematic overview; it also notes a paucity of studies on other outcomes such as cognitive skills and affective influences.
- Evident from the text: some moderator subgroups were small, especially board games (only four studies; 11.11%), so those estimates may be less stable.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
2 Valid · 0 Not valid · 6 Not checked · 32 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“First, they did not comprehensively analyze the subject disciplines of STEM, but only analyzed a particular subject, such as science or mathematics.”
What the sources say
- Partly supportedcites Riopel et al., 2019 sourceExperimental — accuracy not yet measured
The overview ('Previous Meta-Analyses on Serious Games'; 'Moderators') does support the gist that past syntheses left the natural-science/STEM disciplines insufficiently examined: the main limitation is 'the lack of sufficient examination of the effect of serious games in the domain of natural sciences', with broad reviews (Clark et al. 2015; Sitzmann & Ely 2011; Vogel et al. 2006) pooling several knowledge domains 'as a single whole, thus precluding a precise conclusion with regard to the domain of science'. But the claim's second half — that they 'only analyzed a particular subject such as science or mathematics' — is not what our copy says of those meta-analyses; it states the opposite, e.g. Vogel et al. 'did not focus on a particular subject area, and included studies from various knowledge domains'. Our copy does note subject-specific metasyntheses (Rutten et al. 2012; Young et al. 2012; Brinson et al. 2015; Ma & Nickerson 2006), yet frames the gap as science (and its internal disciplines), never as a comprehensive STEM analysis. (LLM, conf=0.72)
- Not checked — no full text of the sourcecites Tokac et al., 2019 source
- Not checked — no full text of the sourcecites Tsai & Tsai, 2020 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“However, these studies have reached no uniform conclusions regarding the effects of digital games on students' STEM learning gains.”
What the sources say
- Supportedcites Hung et al. (2014) source
The Discussion and conclusions section states that the positive achievement result 'is not obvious, since some previous studies have reported unfavorable results or negative effects of DGBL on students' learning performance (Charsky and Ressler 2011; Pierfy 1977; Randel et al. 1992)', i.e. the paper explicitly treats prior evidence on games' effects on learning performance as non-uniform. Alongside this, the Background and motivation and Literature review sections marshal the opposite line of findings (educational computer games promoting motivation and achievement: Burguillo 2010; Liu and Chu 2010; Kuo 2007; Sung and Hwang 2013; Hwang et al. 2013c in natural science; Wang and Chen 2010 in programming), so the mixed-result picture spans STEM subjects rather than mathematics alone. The cited sentence is a broad background characterization, and our copy clearly treats the topic that way. (LLM, conf=0.80)
- Not checked — no full text of the sourcecites Neimeyer (2006)
- Not matchedcites Giannakos, 2013; Khan et al., 2017
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The effect size of digital game-based instruction compared to multimedia is 0.848 (p < 0.001).”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Regarding education levels, results in Table 2 suggest no significant difference in the effect sizes for the different education levels (Qb = 5.184, p = 0.075).”
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Moreover, the results indicate no significant differences among different intervention durations (Qb = 5.070, p = 0.280)”
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
17 more claims
Create a free account to see all 40 claims and to verify the 1 reference still open.
- The quote only partly states this claim. 3
- It may state two separate claims as one. 9
- Probably not a substantial claim. 6
Citations
Cites
Statistical Power Analysis for the Behavioral Sciences
Muller, K. (1989) · Technometrics · doi:10.1080/00401706.1989.10488618 · Cited by 99,167
Measuring inconsistency in meta-analyses
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003) · BMJ · doi:10.1136/bmj.327.7414.557 · Cited by 63,139 · cited 1×
“The I 2 test supplements the Q -test, where 0%–25% indicates that heterogeneity is considered low, 25%–75% indicates moderate heterogeneity, and 75%– 100% indicates substantial heterogeneity (Higgins et al., 2003).”
Bias in meta-analysis detected by a simple, graphical test
Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997) · BMJ · doi:10.1136/bmj.315.7109.629 · Cited by 56,722 · cited 2×
“Publication bias from multiple sources may affect the results of meta-analysis studies (Egger et al., 1997).”
BackgroundPRISMA-P Group, Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P., & Stewart, L. A. (2015) · Systematic Reviews · doi:10.1186/2046-4053-4-1 · Cited by 26,223 · cited 1×
“Studies were searched for using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) method (Moher et al., 2010, 2015).”
MethodOperating characteristics of a rank correlation test for publication bias
Begg, C. B., & Mazumdar, M. (1994) · Biometrics · doi:10.2307/2533446 · Cited by 16,955 · cited 1×
“In the Begg and Mazumdar rank correlation (Kendall’s Tau with a continuity correction), the results ( Z = 1.457 < 1.96, p = 0.145 > 0.05) indicate that there is insignificant publication bias (Begg & Mazumdar, 1994).”
ResultInfluentialBorenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009) · doi:10.1002/9780470743386 · Cited by 14,225 · cited 1×
“When only positive study results are published, it leads to publication bias (Borenstein et al., 2009).”
BackgroundPreferred reporting items for systematic reviews and meta-analyses: The PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2010) · International Journal of Surgery · doi:10.1016/j.ijsu.2010.02.007 · Cited by 12,156 · cited 1×
“Studies were searched for using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) method (Moher et al., 2010, 2015).”
Method(2000) · Cited by 9,198
The file drawer problem and tolerance for null results.
Rosenthal, R. (1979) · Psychological Bulletin · doi:10.1037/0033-2909.86.3.638 · Cited by 9,068 · cited 1×
“The classic fail-safe N test is a procedure to evaluate whether publication bias can be ignored (Rosenthal, 1979).”
BackgroundA basic introduction to fixed-effect and random-effects models for meta-analysis
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2010) · Research Synthesis Methods · doi:10.1002/jrsm.12 · Cited by 6,793 · cited 1×
“Thus, the random-effects model was more appropriate (Borenstein et al., 2010), and showed that analysis of the moderator variables was necessary.”
Method
Research works printed but not linked: 1
- Stege, L., Lankveld, G. V., & Spronck, P. (2012). Teaching high school physics with a serious game. International Journal of Computer Science in Sport, 11(1).
Cited by
Timotheou, S., Miliou, O., Dimitriadis, Y., Sobrino, S. V., Giannoutsou, N., Cachia, R., Monés, A. M., & Ioannou, A. (2023) · Education and Information Technologies · doi:10.1007/s10639-022-11431-8 · Cited by 816
“Integration of digital games in teaching was also found as a promising pedagogical practice in STEM education that could lead to increased learning gains (Martinez et al., 2022; Wang et al., 2022).”
BackgroundEffectiveness of digital educational game and game design in STEM learning: a meta-analytic review
Gui, Y., Cai, Z., Yang, Y., Kong, L., Fan, X., & Tai, R. H. (2023) · International Journal of STEM Education · doi:10.1186/s40594-023-00424-9 · Cited by 284
“…examined the effects of game-based science learning (Riopel et al., 2019; Tsai & Tsai, 2020), examined the effectiveness of game-based math learning (Byun & Joung, 2018; Tokac et al., 2019), or examined the effects of digital game-based STEM education on student knowledge gains (Wang et al., 2022).”
BackgroundGame-based learning in computer science education: a scoping literature review
Videnovik, M., Vold, T., Kiønig, L. V., Bogdanova, A. M., & Trajkovik, V. (2023) · International Journal of STEM Education · doi:10.1186/s40594-023-00447-2 · Cited by 261
“When digital game-based learning is implemented, including key game design elements (collaboration, choice, feedback), there is typically a positive impact on student engagement (Serrano, 2019; Wang et al., 2022).”
BackgroundMethodYang, C., Zhang, J., Hu, Y., Yang, X., Chen, M. H., Shan, M., & Li, L. (2024) · International Journal of STEM Education · doi:10.1186/s40594-024-00487-2 · Cited by 176
“Wang et al. (2022) conducted a systematic review of integrating STEM education into K-12 curricula, providing valuable insights into this educational approach.”
BackgroundPatiño, A., Soledad, M., & Fernández, M. B. (2023) · Smart Learning Environments · doi:10.1186/s40561-023-00229-x · Cited by 134
Metadata
- Authors
- Wang, L.-H., Chen, B., Hwang, G.-J., Guan, J.-Q., & Wang, Y.-Q.
- Affiliations
- Zhejiang Normal University (CN)
- National Taiwan University of Science and Technology (TW)
- Corresponding author
- Wang, Liang-Hui
- ORCID
- Wang, Liang-Hui
- Venue
- International Journal of STEM Education
- Published
- 2022-03-17
- Volume
- 9
- Issue
- 1
- ISSN
- 2196-7822 (ISSN-L)
- OpenAlex
- W4220750697
- Cited by
- 316
- References
- 83
- Citations by year
- 2026: 63 · 2025: 82 · 2024: 90 · 2023: 67 · 2022: 11
- FWCI
- 12.61
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- International Journal of STEM Education (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Philosophy and Social Science Planning of Zhejiang Province (19NDJC252YB)
- National Education Science “13th Five-Year” Planning Key Project of Ministry of Education in China (DCA160262)
- Topics
- Social Sciences › Psychology › Developmental and Educational Psychology › Educational Games and Gamification
- Social Sciences › Education › Education and Learning Interventions
- Physical Sciences › Computer Science › Information Systems › Mobile Learning in Education
- Keywords
- game-based learning, video games, STEM education, academic performance, K-12 students, university students, game types, game platforms
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Effects of digital game-based STEM education on students’ learning achievement: a meta-analysis. https://app.cochat.ai/papers/10.1186/s40594-022-00344-0
- The paper
- Wang, L.-H., Chen, B., Hwang, G.-J., Guan, J.-Q., & Wang, Y.-Q. (2022). Effects of digital game-based STEM education on students’ learning achievement: a meta-analysis. International Journal of STEM Education. https://doi.org/10.1186/s40594-022-00344-0
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.


