♟️ Chess in education: what research does and does not show
Many people say "chess makes kids smarter." Scientists have tested this many times. Here is what they found, in plain words, with every study linked.
The short answer for families
Play chess because it is fun, social and a great way to practice careful thinking. Do not expect chess alone to raise math or reading test scores.
Smaller studies often found gains. When researchers ran bigger and more careful studies, the gains got much smaller or went away. The largest trial, with more than 4,000 children in England, found no change in math, reading or science scores. Children in that trial mostly liked the lessons.
The big picture in one drawing
What the studies say, in plain words
🔎 24 studies, put together (2016)
Two researchers combined 24 studies of chess lessons in schools. On average, children who took chess did a bit better in math and reading. But none of the studies had the ideal design, so the authors asked for more careful research.
📏 The better the study, the smaller the gain (2017)
The same team looked again. Studies with a fair comparison group (children doing a different fun activity) showed smaller gains. Their conclusion: skills learned in chess rarely carry over to other subjects.
🎲 Chess vs. checkers vs. Go (2017)
In two experiments in Italy, some children learned chess, some learned another board game, and some had regular school. Math problem-solving results were about the same. The authors said chess should not replace math lessons.
🏫 The biggest trial: 100 schools, 4,009 pupils (England)
Schools were picked at random to get 30 hours of chess lessons or not. One year later, national test scores in math, reading and science were the same. Half the children said they liked the lessons a lot; 8% did not like them.
🧩 It may depend on how chess is taught (2016)
In a study of 931 primary school pupils, children taught chess by chess instructors who taught problem-solving steps did better in math problem-solving than the other groups. Children taught chess by school teachers without those steps did not. Groups were not compared with another game.
🙂 Thinking and feelings (2012)
A study in Spain compared 170 children in chess with 60 children in soccer or basketball for one school year. The chess group showed gains on thinking tests, coping and problem solving. Children chose their activity, so the groups may have been different from the start.
What we can say, and what we cannot
✅ What we can say
- Many children enjoy learning chess in school (England trial: half liked the lessons a lot).
- Some studies found gains in math problem-solving, thinking or coping, especially when chess was taught with clear problem-solving steps.
- Chess is a good place to practice habits like taking your time, checking before acting, and handling a loss.
- Researchers agree that better-designed studies are needed, and they explain how to do them.
⚠️ What we cannot say
- We cannot say chess "makes kids smarter" or raises IQ. Careful reviews found far transfer (skills spreading to unrelated areas) to be small or zero.
- We cannot say chess raises state test scores. The largest randomized trial found no effect on math, reading or science.
- We cannot say chess should replace math class. One research team said so directly.
- We cannot promise any result for one child. Studies report group averages.
How Goal100 uses this
Chess from Zero is an Extra course: it helps you succeed and is never graded. We offer chess because it is fun, it builds friendships, and it gives a calm place to practice careful thinking and good sportsmanship.
We do not claim it will raise grades or test scores. Math, ELA and the other Required courses stay the place to learn those subjects. In Chess from Zero, the "check before you move" habit is named and practiced on purpose, because the 2016 study of 931 pupils suggests that teaching problem-solving steps openly is what may help.
Study detail table
Study design, sample, finding and limits for each source. We list only studies we opened and checked. Full citations are at the bottom.
| Study | Design | Sample | Main finding | Limits | Link |
|---|---|---|---|---|---|
| Sala & Gobet (2016) Educational Research Review 18 | Meta-analysis 24 studies, 40 effect sizes | 2,788 young people in chess groups; 2,433 in control groups | Overall g = 0.338; math g = 0.382; reading g = 0.248. Longer instruction linked to larger effects. | Authors note no included study had an ideal design (pre/post test, random assignment, and both a do-nothing and an active control group). | doi.org/10.1016/j.edurev.2016.02.002 |
| Sala & Gobet (2017a) Current Directions in Psychological Science 26(6) | Review of meta-analyses Chess, music and working memory training | Two meta-analyses (chess, music) plus one on working memory training | "Small to moderate effects," but effect sizes are inversely related to design quality (for example, presence of active control groups). Concludes far transfer "rarely occurs." | Covers three kinds of training, not chess alone. | doi.org/10.1177/0963721417712760 |
| Sala & Gobet (2017b) Learning & Behavior 45(4) | Two experiments Chess vs. active control (checkers; Go) vs. passive control | Italy. Exp. 1: N = 233, grades 3-4 (25 h of chess). Exp. 2: N = 52, grade 4 (15 h) | No statistically significant differences in math problem-solving or metacognition. Authors: chess "should not replace the traditional curriculum in mathematics." | Exp. 2 sample is small; short courses. | PMC5709436 · DOI |
| Jerrim, Macmillan, Micklewright, Sawtell & Wiggins (2016 EEF report; 2018 journal article) Journal of Human Resources 53(4) | Cluster randomized controlled trial 50 schools chess, 50 control; 30-hour curriculum by trained tutors | England. 100 state primary schools, 4,009 Year 5 pupils (age 9-10); outcome data for 3,865 (97%) | One year later, Key Stage 2 tests: math 0.00 SD, reading -0.06 SD, science -0.03 SD (no effect). No different effect by gender or free school meal status. Half of pupils liked lessons "a lot"; 8% did not. | One program and one age group; science outcome based on teacher judgement. | EEF report (ERIC ED581100) · Journal DOI |
| Sala, Aksayli, Tatlidil, Tatsumi, Gondo & Gobet (2019) Collabra: Psychology 5(1) | Second-order meta-analysis Cognitive training of many kinds, chess included | Many meta-analyses combined (chess one of four program types in one model) | After controlling for placebo effects and publication bias, the overall far-transfer effect was zero, regardless of population or program type. | Does not report chess on its own in the abstract. | doi.org/10.1525/collabra.203 |
| Trinchero & Sala (2016) Eurasia J. of Mathematics, Science and Technology Education 12(3) | Experiment, 3 groups Chess by instructors teaching heuristics; chess by school teachers without heuristics; no chess | 931 primary school pupils | The instructor-plus-heuristics group did better in math problem-solving than both other groups; the teacher-led group did not differ from control. | No active control (another game); the abstract does not say how groups were formed. | doi.org/10.12973/eurasia.2016.1255a |
| Sala, Gorini & Pravettoni (2015) SAGE Open | Experiment, classes randomly assigned 10-15 h chess course vs. normal school | Northern Italy. 560 pupils, grades 3-5, 31 classes, 8 schools | Chess group improved more in math problem-solving (PISA-type items) over about 3 months. | No active control; short time between tests. | doi.org/10.1177/2158244015596050 |
| Kazemi, Yektayar & Abad (2012) Procedia - Social and Behavioral Sciences 32 | Chess group vs. control 6 months of chess | 86 students in chess group, 94 in control (all male); several school levels | Chess group scored higher on metacognitive ability and math problem-solving. | No active control; the abstract does not say if both groups were tested before the course. | doi.org/10.1016/j.sbspro.2012.01.056 |
| Aciego, García & Betancort (2012) The Spanish Journal of Psychology 15(2) | Quasi-experiment Chess vs. soccer or basketball, one school year | Spain. Ages 6-16; 170 in chess, 60 in soccer/basketball | Chess group improved more on WISC-R cognitive scores, coping and problem solving; positive socio-affective results reported. | Children chose their activity (no random assignment); the authors note results varied with the profile of children who chose chess. | doi.org/10.5209/rev_SJOP.2012.v15.n2.38866 |
How to read "g" and "SD": both are effect sizes in standard deviations. 0 means no difference between groups. Positive numbers mean the chess group scored higher.
Teacher talking points
If a parent asks "Will chess raise my child's math grade?"
"The biggest, most careful study found no change in test scores. We offer chess because children enjoy it and it is a good place to practice careful thinking and good sportsmanship."
If an administrator asks for evidence
Share this page. Point to the England trial (no test-score effect) and to the 2016 study of 931 pupils (problem-solving steps taught openly may matter). Present chess as enrichment, not as a math intervention.
If you want to try it in class
Name the thinking steps out loud (for example, "Check, capture, threat before you move") and connect them to other problems. That matches what the more positive studies did.
Sources
All sources opened and checked Nov 8, 2026. We only list studies we could open and confirm.
- Sala, G., & Gobet, F. (2016). Do the benefits of chess instruction transfer to academic and cognitive skills? A meta-analysis. Educational Research Review, 18, 46-57. https://doi.org/10.1016/j.edurev.2016.02.002 (open copy: University of Liverpool repository). Checked Nov 8, 2026.
- Sala, G., & Gobet, F. (2017). Does far transfer exist? Negative evidence from chess, music, and working memory training. Current Directions in Psychological Science, 26(6), 515-520. https://doi.org/10.1177/0963721417712760 (open copy: LSE Research Online). Checked Nov 8, 2026.
- Sala, G., & Gobet, F. (2017). Does chess instruction improve mathematical problem-solving ability? Two experimental studies with an active control group. Learning & Behavior, 45(4), 414-421. https://doi.org/10.3758/s13420-017-0280-3 (free full text: PubMed Central PMC5709436). Checked Nov 8, 2026.
- Jerrim, J., Macmillan, L., Micklewright, J., Sawtell, M., & Wiggins, M. (2016). Chess in Schools: Evaluation Report and Executive Summary. Education Endowment Foundation. https://eric.ed.gov/?id=ED581100. Checked Nov 8, 2026.
- Jerrim, J., Macmillan, L., Micklewright, J., Sawtell, M., & Wiggins, M. (2018). Does teaching children how to play cognitively demanding games improve their educational attainment? Evidence from a randomized controlled trial of chess instruction in England. Journal of Human Resources, 53(4), 993-1021. https://doi.org/10.3368/jhr.53.4.0516.7952R (working paper: UCL Discovery). Checked Nov 8, 2026.
- Sala, G., Aksayli, N. D., Tatlidil, K. S., Tatsumi, T., Gondo, Y., & Gobet, F. (2019). Near and far transfer in cognitive training: A second-order meta-analysis. Collabra: Psychology, 5(1), 18. https://doi.org/10.1525/collabra.203. Checked Nov 8, 2026.
- Trinchero, R., & Sala, G. (2016). Chess training and mathematical problem-solving: The role of teaching heuristics in transfer of learning. Eurasia Journal of Mathematics, Science and Technology Education, 12(3), 655-668. https://doi.org/10.12973/eurasia.2016.1255a. Checked Nov 8, 2026.
- Sala, G., Gorini, A., & Pravettoni, G. (2015). Mathematical problem-solving abilities and chess: An experimental study on young pupils. SAGE Open, 1-9. https://doi.org/10.1177/2158244015596050. Checked Nov 8, 2026.
- Kazemi, F., Yektayar, M., & Abad, A. M. B. (2012). Investigation the impact of chess play on developing meta-cognitive ability and math problem-solving power of students at different levels of education. Procedia - Social and Behavioral Sciences, 32, 372-379. https://doi.org/10.1016/j.sbspro.2012.01.056. Checked Nov 8, 2026.
- Aciego, R., García, L., & Betancort, M. (2012). The benefits of chess for the intellectual and social-emotional enrichment in schoolchildren. The Spanish Journal of Psychology, 15(2), 551-559. https://doi.org/10.5209/rev_SJOP.2012.v15.n2.38866. Checked Nov 8, 2026.
Found a mistake? Tell us on the Corrections page.