Field Notes

source

Some cold facts about math and gender

StatusStableUpdatedStands on2 sourcesSupportsnotes

Ken LaCorte’s May 2025 episode of Elephants in Rooms reviews the evidence on gender differences in mathematics performance, covering US standardized tests, international comparisons, and competing nurture and nature explanations.

The timestamped transcript was saved 2026-07-24.

Evidence presented

On the SAT math section, boys averaged 530 and girls 512 in 2022, an 18-point gap that has shrunk from roughly 60 points in 1985. No recorded year shows girls outscoring boys on SAT math, but the gap is small (roughly two questions on a 58-question exam).

The male advantage is most pronounced at the extremes. Of students in the top 10% of SAT math scores, 61% are male, and boys are more than twice as likely to earn a perfect 800. Boys are also overrepresented in the bottom 10% (56%). LaCorte calls this the “males at the tails” phenomenon: boys show greater variability in math performance and appear more at both ends of the distribution curve.

On the AP math exams, the top-end gap is also shrinking: from 52 girls per 100 boys at the highest score in 1995 to 69 per 100 by 2019.

In the classroom, the pattern reverses. Girls outperform boys in math grades, hold higher GPAs overall, and complete advanced math classes at equal or higher rates. The National Assessment of Educational Progress (“Nation’s Report Card”) finds almost no gender differences in fundamental math skills.

Internationally, PISA data across 80 countries shows boys scoring significantly higher in math in 40 countries, girls higher in 17, and no meaningful difference in the rest. Scandinavian countries show no meaningful gap; Jordan, Qatar, Bahrain, Malaysia, and Thailand show girls outperforming boys, sometimes by larger margins than anything seen in the US. Italy shows the largest gap favoring boys.

Explanations covered

Nurture (social): stereotype threat, math anxiety, self-confidence differences (76% of US high-school boys call themselves confident in math class versus 58% of girls), and attributional style (boys blame external causes for difficulty, girls blame themselves). Countries using collaborative teaching methods (Norway, Sweden) narrow gaps; countries emphasizing individual competition (Italy, Japan) widen them.

Nature (biological): the greater male variability hypothesis, attributed to males having a single X chromosome without the balancing effect of a second, is cited as explaining the wider male spread. Men outperform women on mental rotation tests by about 35% on average, relevant to geometry, physics, and engineering. Testosterone’s role is noted as unsettled.

Preference: in gender-equal societies like Scandinavia, women choose STEM fields at lower rates despite encouragement, and the participation gap is sometimes wider than in less equal countries. Women within STEM lean toward biology and medicine; men cluster in physics and computer science.

Limitations

LaCorte’s video is a journalistic synthesis, not a systematic review. His script sources are linked in a Google Doc rather than presented with explicit methodology. The video selects and narrates studies conversationally; it should be treated as a survey of the landscape, not as primary evidence for any specific claim. The presenter’s framing emphasizes nuance but follows a structure that moves from nurture to nature explanations.

  • Greater male variability — the biological hypothesis invoked to explain the male-tail pattern in math scores
  • What if boys were never the problem — covers the broader boys’ education gap, including the same tail-distribution pattern observed in GPAs and standardized tests
  • The Disposable Male — treats the male-tail pattern as an evolutionary expectation, not a statistical anomaly, connecting it to reproductive strategy

Built on 2 sources (2 external).

Working out connections…