The "gender gap in math" is one of the most quoted and most misquoted topics in education. Some headlines say it has closed. Others say it's back. Both can be true depending on which test, which grade, which cohort, and which part of the score distribution you look at.
Below are the most reliable gender gap in math statistics available in 2026. Every figure comes from a peer-reviewed study, so you can cite them with confidence.
Is there a gender gap in math?
On average, how different are girls' and boys' math scores?
Very little, on average.
A meta-analysis of 242 studies (1990–2007) covering 1,286,350 people found an average gender difference of d = 0.05. The variance ratio was 1.08, meaning male and female score spreads were nearly equal.[4]
U.S. state tests in grades 2–11 showed the same pattern of gender similarity.[9]
A newer meta-analysis of 440 studies published 2010–2022, covering more than 15 million participants, found a small but significant male advantage (d = 0.08).[5]
In the same 2026 paper, a separate analysis of PISA data from 2012 to 2022 found an even smaller average gap among 15-year-olds (d ≈ 0.05).[5]
Quotable: "Across hundreds of studies, the average difference between girls' and boys' math scores is about 0.05–0.08 standard deviations, close to zero." [4][5]
Does it depend on the type of math?
Yes. The 2026 meta-analysis found that a male advantage in geometry, broad mathematics and computation tends to emerge around fifth grade and grows through adolescence. By the end of secondary school it reached about d = 0.54 for geometry and d = 0.26 for broad mathematics. Basic numeracy and advanced mathematics showed differences close to zero.[5]
Do U.S. school districts show a math gender gap?
Using state test data for grades 3–8 in nearly 10,000 U.S. school districts (2008–09 to 2015–16), researchers found:[10]
The average district had essentially no gender gap in math.
The average district had a 0.23 standard deviation gap in English language arts (ELA) favoring girls.
Math gaps varied between districts and tended to favor boys more in socioeconomically advantaged districts and in places with larger adult gender gaps in income, education and occupations.
Chart suggestion: horizontal bar chart of the five effect sizes above, with a reference line at d = 0.2 (a conventional "small" effect).
When does the math gender gap start?
The short answer: early in elementary school. The exact starting point depends on the cohort and the test, so it's worth reading the U.S. and French evidence side by side.
Is there a gender gap in kindergarten math?
Usually little or none at school entry, but it varies by cohort.
In the nationally representative U.S. ECLS-K study (1998–99 cohort), there were no mean differences between boys and girls at school entry. Girls then lost more than two-tenths of a standard deviation relative to boys over the first six years of school, about half as large as the Black–White test-score gap that opens over the same ages.[2]
Across two ECLS-K cohorts more than a decade apart (1998–99 and 2010–11), girls and boys began kindergarten with similar math proficiency, but disparities in achievement and confidence developed by Grade 3. In both cohorts the gap appeared first among high achievers and spread through the distribution.[3]
A 2026 study of almost 12 million U.S. students across nine kindergarten cohorts found a slightly different starting point. In earlier, pre-pandemic cohorts, girls entered kindergarten ahead in both math and reading, but their math advantage has diminished in more recent cohorts. In every cohort, boys surpassed girls in math during elementary school, while girls kept a steady lead in reading.[12]
In plain terms: Earlier U.S. cohorts showed little average math difference at kindergarten entry, and some cohorts showed a small head start for girls. Every one of these studies documents a boy-favoring math gap emerging during elementary school. [2][3][12]
How fast does the gap appear?
The largest study to date followed all French first and second graders: 2,653,082 children over four years. Its findings:[1]
Boys and girls had very similar math scores at school entry.
A gap favoring boys was significant after 4 months of schooling.
It reached an effect size of about 0.20 after 1 year.
By comparing children of the same age with different amounts of schooling, the researchers found the gap was more closely tied to time spent in school than to age alone.
The pattern repeated every year and varied only slightly by family, class, school type or socioeconomic level.
What this study does and doesn't show: It shows when the gap appears, very soon after school entry. It does not establish which aspects of schooling or children's wider experiences cause it. Starting school changes many things at once: formal math teaching, but also peers, expectations, feedback and exposure to stereotypes.[1]
In a study of 2.65 million French children, the math gender gap was absent at school entry, significant after four months of first grade, and about 0.2 standard deviations after one year. [1]
Chart suggestion: line chart with "months of schooling" on the x-axis and gap size (SD) on the y-axis, using the French data points at 0, 4 and 12 months.
Why this matters for K–3: Because the gap shows up within the first years of formal schooling, kindergarten through 3rd grade is the window where it first becomes visible. Researchers haven't yet pinned down which classroom or social factors drive it. For what early scores do and don't predict, see our article on what kindergarten math scores actually predict about third grade.
Is the gap bigger at the top?
Are boys overrepresented among top math scorers?
Yes, and this is where the gap is most persistent.
In fall of kindergarten, girls made up only 20% of children above the 99th percentile in math (ECLS-K).[3]
On U.S. state tests, the ratio of male to female score variance ranged from 1.11 to 1.21, slightly greater male variability.[9][13]
In Minnesota state test data, there were 1.45 White boys for every White girl above the 95th percentile, and about 2.0 above the 99th percentile.[9][13]
On NAEP (1990–2011), boys were overrepresented by more than 2:1 among top performers in math and science.[11]
Has the top-end gap shrunk over time?
Yes, substantially, though boys are still overrepresented. Among U.S. seventh graders in talent searches scoring 700+ on SAT-Math:
The 2011–2015 update, based on 320,000 seventh graders, found that ratios in the extreme right tail of math ability had shrunk over the previous 20 years, while still favoring boys.[24]
Among gifted U.S. 7th graders scoring 700+ on SAT-Math, the boy-to-girl ratio fell from about 13.5:1 in the early 1980s to about 4:1 by the 1990s and about 2.5:1 in 2011–2015. [14][24]
What about "greater male variability"?
Boys' scores are slightly more spread out, so boys are overrepresented at both the top and the bottom. The size of that difference varies by country, test and era,[9][13][15] and the top-tail ratio has fallen over time,[14][24] which shows it is not fixed. Two cautions for writers:
Variance ratios of ~1.1–1.2 are modest.
Top-tail ratios depend heavily on how hard the test is. Curriculum-aligned state tests with few difficult items show smaller gaps than harder adaptive tests such as those used in ECLS-K.[3][9]
Did the pandemic widen the math gender gap?
What does peer-reviewed research say about COVID and the math gender gap?
A 2026 analysis of test scores from roughly 7,000 U.S. school districts over 15 years found:[6]
Girls' math and reading scores declined more than boys' during the pandemic, shifting gender gaps toward boys in both subjects.
This shift was larger than at any other time in the past 30 years.
The extra decline for girls was bigger in lower-income communities and where the pandemic disrupted daily life most, for example through high job loss, long school closures and business shutdowns.
In high-resource communities, the pandemic had little differential effect on girls and boys. Even so, girls' scores still declined more than boys' in all but the most affluent places.
Girls' test scores fell more than boys' during COVID across roughly 7,000 U.S. districts, the largest shift toward boys in 30 years. [6]
Note for writers: You may have seen post-pandemic figures from news analyses or testing-company research briefs. Those are not peer-reviewed, so we don't cite them here.
How does the U.S. compare internationally?
Do all countries have a math gender gap?
No. The size, and even the direction, of the gap varies from country to country. That variation shows the gap is context-dependent. On its own, though, it doesn't settle how much of the gap is social and how much (if any) is biological.
A meta-analysis of TIMSS and PISA 2003 data (493,495 students aged 14–16 in 69 countries) found that all mean effect sizes were very small (d < 0.15). National gaps, however, ranged from d = −0.42 (girls ahead) to +0.40 (boys ahead).[15]
The strongest predictors of a country's math gender gap in that analysis were gender equity in school enrollment, women's share of research jobs, and women's representation in parliament.[15]
Do more gender-equal countries have smaller math gaps?
The evidence is mixed, and the question is still open.
Earlier studies found smaller gaps in more gender-equal countries. An analysis of PISA 2003 (about 276,000 students in 40 countries) found the math gender gap disappears in more gender-equal countries such as Norway and Sweden.[16]
That link proved sensitive to which countries are included. In several Muslim-majority countries there is little or no math gender gap despite low measured gender equality.[2]
The 2026 meta-analysis found the opposite direction. Across 2010–2022 studies, math gender differences were more pronounced in countries with higher gender equality (measured by the Global Gender Gap Index) and in regions such as North America, Europe, and Central and South America.[5] When the authors controlled for region, this was a tendency rather than a clear-cut result (p = .051). They interpret their findings as supporting a role for both biological and socio-cultural factors, without establishing how much each contributes, and describe the gender-equality relationship as unresolved.[5]
Math anxiety shows a related pattern. Gender-equal countries tend to have lower overall math anxiety but a larger gap between girls' and boys' math anxiety.[17]
Math anxiety and confidence by gender
Are girls more anxious about math than boys?
On average, yes, and the confidence gap is often larger than the achievement gap.
In U.S. ECLS-K data, gender differences in math confidence were larger than gender differences in math interest and achievement in elementary school.[18]
Across countries, boys report more confidence in math than girls even where performance is similar.[15]
Using PISA data, girls report higher math anxiety than boys in most countries, and the gap is larger in more gender-equal, wealthier countries.[17]
Girls generally report lower math confidence than boys, and the gender difference in confidence can be substantially larger than the difference in achievement. [18]
For more detail, see our Math Anxiety Statistics 2026 roundup and our guide to dyscalculia vs. math anxiety in K–3.
Is a teacher's math anxiety linked to girls' outcomes?
One well-known study found an association. In a study of 17 first- and second-grade female teachers:[19]
Teachers' math anxiety was unrelated to students' math achievement at the start of the year.
By the end of the year, higher teacher math anxiety was associated with girls (but not boys) being more likely to endorse the stereotype that "boys are good at math, girls are good at reading." Girls who endorsed it had lower end-of-year math achievement.
The study notes that more than 90% of U.S. early elementary teachers are women.
Caveat: This was a small, correlational study (17 classrooms). Teachers were not randomly assigned to different levels of math anxiety, so it can't show cause and effect. Treat it as an important signal, not a precise estimate.[19]
Parents matter too. See how parents accidentally increase math anxiety in kids.
Stereotypes: how early do they start?
At what age do children believe "math is for boys"?
In a study of 247 U.S. children aged 6–10, the "math is for boys" stereotype was measurable as early as second grade, on both implicit (reaction-time) and self-report measures. Boys identified with math more strongly than girls did.[20]
By age 6, girls are less likely than boys to believe members of their own gender are "really, really smart." Also by age 6, girls start to avoid activities described as being for "really, really smart" children. At age 5, there were no such differences.[7]
Quotable: "By age 6, girls are less likely than boys to see their own gender as 'really, really smart', and they begin avoiding activities labeled as for smart kids." [7]
Do teachers grade or rate girls differently?
Do teachers underrate girls in math?
Studies find teacher bias in both directions, depending on the country and the measure.
U.S. (ECLS-K): Teachers rated girls' math proficiency lower than that of boys with similar test scores and learning behaviors. Girls were rated as equally competent only when they were also seen as working harder and behaving better.[3][21] An analysis of these ratings suggested they may contribute to the early growth of the math gender gap.[21]
Israel: Comparing teacher-graded exams with blindly graded national exams, students whose teacher's grading favored their gender later did better in math and were more likely to enroll in advanced high-school math.[22]
France: Middle school teachers favored girls in grading. A math teacher 1 SD more biased in favor of girls raised girls' probability of choosing a science track by 3.6 percentage points. Without this bias, the gender gap in choosing science would be 12.5% larger in favor of boys.[23]
"Teacher bias" is real but context-specific. In the U.S. evidence it tends to work against girls in teachers' perceptions of math ability. In the French evidence it sometimes works in their favor in grades. Both are linked to girls' later math choices.
The STEM pipeline
How many math and STEM degrees do women earn?
Using historical NSF data summarized by Cheryan et al. (2017), women earned approximately 44% of U.S. mathematics bachelor's degrees, compared with less than 20% in computer science, engineering and physics. (Note: These figures describe the early 2010s, not the current distribution.)[8]
The researchers found that math performance and discrimination influence who enters STEM but do not explain why some STEM fields are much less gender-balanced than others. Masculine field cultures, too little early experience and gaps in self-efficacy explain more.[8]
Data note: These degree shares come from National Science Foundation data from the early-to-mid 2010s, as reported in a peer-reviewed review. More recent federal degree counts exist but are government statistical releases rather than peer-reviewed papers, so they are not cited here. If you need current degree counts, check the latest NSF National Center for Science and Engineering Statistics (NCSES) release directly.
Caveats every writer should know
Averages hide the story. The mean gap is small (d ≈ 0.05–0.10). Most of the action is in the top tail, confidence, and course choices.[4][5][3]
The test matters. Curriculum-aligned state tests often show no gap. Harder or adaptive tests (ECLS-K, NAEP, SAT, competitions) show larger gaps.[3][9]
The cohort matters. Some U.S. cohorts start kindergarten with no math gap and some with a small advantage for girls. All show a boy-favoring gap emerging during elementary school.[2][3][12]
"d" units are not grade levels. An effect of 0.2 SD is "small" statistically but meaningful when it opens within one school year.[1]
Timing is not cause. The French study shows when the gap appears, not why. It does not identify which aspects of schooling are responsible.[1]
The international picture is contested. Earlier studies linked gender equality to smaller gaps; the 2026 meta-analysis found a tendency in the opposite direction. Results depend on measures, samples and methods.[5][16]
Data are binary. Nearly all large datasets record gender as boy/girl only. Non-binary and gender-diverse students are invisible in these statistics.
News numbers are not studies. Many widely shared 2025 figures come from news analyses or testing-company reports that are not peer-reviewed.
FAQ
Is there still a gender gap in math in 2026?
Yes. On average it is small, but a boy-favoring gap emerges early in elementary school, is larger among top scorers, and shifted further toward boys during the pandemic.[1][6][12]
When does the math gender gap start?
Within the first years of school. In France it was absent at school entry, significant within four months of first grade and about 0.2 SD after one year.[1] In the U.S., earlier cohorts started kindergarten roughly even (some with a small lead for girls), and boys pulled ahead in math during elementary school.[2][3][12]
Are boys naturally better at math?
The evidence doesn't support a simple "yes." Girls and boys start school with similar math skills, and in France the gap was more closely tied to time in school than to age.[1] National gaps also vary widely, and in some countries girls are ahead.[15] Researchers still debate the relative roles of biological and social factors.[5]
Are girls more anxious about math?
On average, yes. Girls report more math anxiety and lower math confidence, and the gender difference in confidence can be substantially larger than the difference in achievement.[17][18]
Did COVID make the math gender gap worse?
Yes. Girls' scores declined more than boys' across roughly 7,000 U.S. districts, especially in low-income communities.[6]
Why are there more boys at the top of math tests?
Boys' scores are slightly more variable, and the gap at the top appears as early as kindergarten.[3][9] The top-end ratio has also fallen sharply, from about 13.5:1 in the early 1980s to about 2.5:1 in 2011–2015 among gifted 7th graders.[14][24]
What can parents and teachers do?
The studies above describe when and where gaps appear; they don't test specific fixes. Reasonable, research-informed steps include paying attention to the K–3 years, when the gap first shows up;[1] being aware of adults' own math anxiety;[19] and avoiding "natural genius" framing, since children absorb brilliance stereotypes by age 6.[7] These are practical suggestions, not interventions proven to close the gap.
References
Martinot, P., Colnet, B., Breda, T., Sultan, J., Touitou, L., Huguet, P., Spelke, E., Dehaene-Lambertz, G., Bressoux, P., & Dehaene, S. (2025). Rapid emergence of a maths gender gap in first grade. Nature, 643(8073), 1020–1029. Open access (PMC) · DOI
Fryer, R. G., & Levitt, S. D. (2010). An empirical analysis of the gender gap in mathematics. American Economic Journal: Applied Economics, 2(2), 210–240. DOI
Cimpian, J. R., Lubienski, S. T., Timmer, J. D., Makowski, M. B., & Miller, E. K. (2016). Have gender gaps in math closed? Achievement, teacher perceptions, and learning behaviors across two ECLS-K cohorts. AERA Open, 2(4). Open access
Lindberg, S. M., Hyde, J. S., Petersen, J. L., & Linn, M. C. (2010). New trends in gender and mathematics performance: A meta-analysis. Psychological Bulletin, 136(6), 1123–1135. DOI
Caviola, S., Giofrè, D., Toffalini, E., Esposito, L., Tonizzi, I., & Geary, D. C. (2026). Sex differences in mathematics: A meta-analytic review across content, grades, and geographic regions. Educational Psychology Review, 38, Article 58. Open access · DOI
Reardon, S. F., Richardson, S., & Wilson, S. (2026). Gender differences in the educational effects of the COVID-19 pandemic. The ANNALS of the American Academy of Political and Social Science, 716(1), 69–90. Free access
Bian, L., Leslie, S.-J., & Cimpian, A. (2017). Gender stereotypes about intellectual ability emerge early and influence children's interests. Science, 355(6323), 389–391. DOI
Cheryan, S., Ziegler, S. A., Montoya, A. K., & Jiang, L. (2017). Why are some STEM fields more gender balanced than others? Psychological Bulletin, 143(1), 1–35. DOI
Hyde, J. S., Lindberg, S. M., Linn, M. C., Ellis, A. B., & Williams, C. C. (2008). Gender similarities characterize math performance. Science, 321(5888), 494–495. DOI
Reardon, S. F., Fahle, E. M., Kalogrides, D., Podolsky, A., & Zárate, R. C. (2019). Gender achievement gaps in U.S. school districts. American Educational Research Journal, 56(6), 2474–2508. DOI
Reilly, D., Neumann, D. L., & Andrews, G. (2015). Sex differences in mathematics and science achievement: A meta-analysis of National Assessment of Educational Progress assessments. Journal of Educational Psychology, 107(3), 645–662. DOI
Kuhfeld, M., & Burchinal, M. (2026). Gender gaps in early grades: Questioning whether schools disadvantage young boys. Educational Researcher (advance online publication, 25 June 2026). DOI
Hyde, J. S., & Mertz, J. E. (2009). Gender, culture, and mathematics performance. Proceedings of the National Academy of Sciences, 106(22), 8801–8807. Open access
Wai, J., Cacchio, M., Putallaz, M., & Makel, M. C. (2010). Sex differences in the right tail of cognitive abilities: A 30 year examination. Intelligence, 38(4), 412–423. DOI
Else-Quest, N. M., Hyde, J. S., & Linn, M. C. (2010). Cross-national patterns of gender differences in mathematics: A meta-analysis. Psychological Bulletin, 136(1), 103–127. Free PDF (APA) · DOI
Guiso, L., Monte, F., Sapienza, P., & Zingales, L. (2008). Culture, gender, and math. Science, 320(5880), 1164–1165. Author-hosted PDF · DOI
Stoet, G., Bailey, D. H., Moore, A. M., & Geary, D. C. (2016). Countries with higher levels of gender equality show larger national sex differences in mathematics anxiety and relatively lower parental mathematics valuation for girls. PLoS ONE, 11(4), e0153857. Open access
Ganley, C. M., & Lubienski, S. T. (2016). Mathematics confidence, interest, and performance: Examining gender patterns and reciprocal relations. Learning and Individual Differences, 47, 182–193. Publisher page
Beilock, S. L., Gunderson, E. A., Ramirez, G., & Levine, S. C. (2010). Female teachers' math anxiety affects girls' math achievement. Proceedings of the National Academy of Sciences, 107(5), 1860–1863. Open access (PMC)
Cvencek, D., Meltzoff, A. N., & Greenwald, A. G. (2011). Math–gender stereotypes in elementary school children. Child Development, 82(3), 766–779. Institutional copy (UW I-LABS) · DOI
Robinson-Cimpian, J. P., Lubienski, S. T., Ganley, C. M., & Copur-Gencturk, Y. (2014). Teachers' perceptions of students' mathematics proficiency may exacerbate early gender gaps in achievement. Developmental Psychology, 50(4), 1262–1281. DOI
Lavy, V., & Sand, E. (2018). On the origins of gender gaps in human capital: Short- and long-term consequences of teachers' biases. Journal of Public Economics, 167, 263–279. Open access (accepted manuscript, Warwick) · DOI
Terrier, C. (2020). Boys lag behind: How teachers' gender biases affect student achievement. Economics of Education Review, 77, 101981. DOI
Makel, M. C., Wai, J., Peairs, K., & Putallaz, M. (2016). Sex differences in the right tail of cognitive abilities: An update and cross cultural extension. Intelligence, 59, 8–15. DOI
