Why Do Girls Still Score Lower Than Boys on Maths Tests?

A massive new study challenges the idea of ‘innate’ gender differences — and points instead to something far more malleable

Image licensed from Shutterstock

The Noösphere is an entirely reader-supported publication that brings social sciences research into frequently overlooked topics. If you read it every week and value the labour that goes into it, consider liking or sharing this essay or becoming a paid subscriber! You can also buy me a coffee instead.

---

‘Math class is tough,’ declared a talking Barbie released by Mattel in 1992.

The doll was quickly recalled after widespread criticism that Barbie — seen by many as a role model for young girls — was reinforcing a damaging gender stereotype. Mattel even offered a swap for those who’d already purchased it.

But while that maths-loathing Barbie may have vanished from store shelves and our world, the sentiment she voiced hasn’t. The idea that girls and women are somehow ‘naturally’ less suited to mathematics — and other intellectually demanding pursuits — is still alive and well today, bolstered by armchair theories of female intellectual inferiority from a parade of (allegedly) Great Men of History and Science. (Wandering wombs! Tiny skulls! Underdeveloped brains! Women, clearly, never evolved!)

Unsurprisingly, some argue that if these Great Men were indeed wrong, and since there are no longer any formal barriers preventing girls and women from pursuing education or succeeding in STEM, then gender gaps in these areas shouldn’t exist. But they do — checkmate, feminists.

It must, obviously, be all women’s fault. And our inherently inferior, impossibly small, pink brains. Is it, though? What’s actually behind the persistent gaps in cognitive performance, like maths tests?

---

It’s true that, globally, women still make up less than half — around 35–40% — of STEM graduates and remain underrepresented in STEM careers. It’s also true that girls often score lower than boys on mathematics tests.

According to the OECD’s 2012 PISA (Programme for International Student Assessment) report, boys outperformed girls in maths in 38 out of the 65 participating countries, while girls outperformed boys in just five. A more recent round of PISA assessments, conducted in 2022, still reveals a persistent gender gap — but, again, not everywhere. While boys scored higher in maths than girls in 40 countries, girls outperformed boys in 17 others. In the remaining 24 countries (up from 22 in 2012), there was no statistically significant difference between genders. But that same recent data also shows that girls continue to be much more likely to report feelings of anxiety about maths than boys, even when performing well.

Still, if the gender gap in maths performance were truly the result of inherently different ‘wiring’ between the genders, I doubt we’d see it shift over time, vary by country, or even reverse in some places. Most likely, we also wouldn’t see it pop up only at a certain age.

The latter is one of the key findings from a massive new study published last week in Nature, which analysed the maths and language performance of every first- and second-grade student in France — a total of 2,653,082 children — over four consecutive years. And yes, upon school entry (T1 in the graph below), at the age of six — or nearly six, since in France children begin school in September of the calendar year they turn six — boys’ and girls’ average performance in mathematics was nearly identical.

However, that changed quite quickly. After just four months of schooling (T2), a gender gap in favour of boys began to emerge. Within 12 months — so by the start of the next school year (T3) — the gap had already become significant, with the top 5% of children in maths comprising more than twice as many boys as girls.

Distribution of national percentile ranks in maths among boys and girls, comprising three measurement periods (T1, T2, and T3) for the 2018 cohort.

The same gender gap emerged consistently across all four cohorts studied — students who began school in 2018, 2019, 2020, and 2021 — and across all socioeconomic groups, regions of France, and types of schools. Variables like parental occupation, family composition, and the socioeconomic status of a school’s population had little influence on the outcome, too.

Interestingly, while the gap was already present among boys and girls born in December as they entered their second year of school, it was absent among their peers born only days later, in January, who were just beginning school. It also widened far less during the COVID-19 lockdown, which coincided with T2 and T3 for the 2019 cohort, when French students missed at least 52 days of school.

The researchers point out that all this evidence strongly suggests the difference in maths scores between boys and girls isn’t due to ‘innate’ ability — or lack thereof — but rather to the environment they encounter once they begin school. As they write in the paper:

Primary school may mark the moment in children’s curriculum when maths-related activities or exercises (…) start to be more clearly identified as belonging to the maths domain, with separate school textbooks and teaching hours for maths subjects. This sudden labelling of maths-related activities as ‘maths’ (…) might give space for gender stereotypes surrounding maths to emerge, to be internalised by children and, eventually, to affect their self-concept and performance.

Primary school teacher attitudes may contribute to this dissemination, if teachers interact differently with boys and girls, transmit their maths anxiety to girls, encourage girls’ efforts at reading more than at maths, or attribute the successful mathematical performance of boys to their greater intellectual power and the successful mathematical performance of girls to their greater diligence.

Previous research does suggest that teachers frequently underestimate girls’ maths abilities, attributing boys’ success to ‘innate’ talent and girls’ to hard work, which can then undermine girls’ confidence in their capacity to learn. Even when girls and boys perform equally well and exhibit the same learning behaviours, teachers rate girls’ mathematical proficiency as lower. One study even found that it’s precisely teachers who believe gender disparities are ‘no longer a problem’ who are most likely to underestimate girls’ abilities. Female teachers who experience maths anxiety themselves tend to negatively impact girls’ maths performance as well — but not boys’.

Still, it’s not only the school environment or teacher attitudes that reinforce these disparities.

It’s the broader cultural context children grow up in, too.

---

Although the process of gender socialisation throws us into a generously seasoned gender marinade — or perhaps it’s more of a gender stew? — practically the moment we enter the world (if not earlier, as some studies suggest), it doesn’t have a dramatic effect right away. As babies and very young children, girls and boys display remarkably similar abilities, including in their grasp of numbers and space. Even the occasionally observed male or female advantages tend to be small and do not consistently appear across time, geography, testing conditions, or before the age of five or six.

But it’s precisely around that age that studies show girls and boys begin to internalise and repeat gender stereotypes, including those about their intellectual abilities. This shift is likely triggered by their first steps into formal education and the stereotypes embedded within it, yes — as highlighted by the French study and many others — but also by the messages they receive from parents, other adults, peers, and the media.

The stereotype that girls ‘just aren’t good at maths’ — or at least not as good as boys — is particularly pervasive, with parents frequently underestimating girls’ maths ability relative to their actual performance. Just like teachers do.

Peer influence, too, plays a substantial role. Another new study, published in Psychological Science, found not only that the majority of students still believe boys are innately superior at math, but also that girls in classrooms where a greater proportion of peers endorsed this stereotype scored lower, on average, on maths exams. They were also less likely to participate in extracurricular maths classes, more likely to report that maths is difficult, and less likely to see it as relevant to their future careers. Meanwhile, boys in those same classrooms performed slightly better.

Other evidence also suggests that these stereotypes are the most impactful — and damaging — precisely when they emphasise biological factors, making them seem more profound and immutable. As psychologist Cordelia Fine notes in Delusions of Gender:

(…) media reports of gender that emphasise biological factors leave us more inclined to agree with gender stereotypes, to self-stereotype ourselves and even for our performance to fall in line with those stereotypes.

Another study found that women who read a scientific journal article claiming that men are better at maths because of innate biological and genetic differences performed worse on a subsequent maths test than women who read an essay attributing men’s stronger performance to greater effort. Even being told this information by the experimenter had the same effect. Women also tend to perform worse on maths tests when they’re reminded of their gender right beforehand — a phenomenon known as ‘stereotype threat.’

By the time women are nearly fully ‘cooked’ and reach adulthood, it can be challenging to shake off the beliefs society has marinated them in for so long. When you’re told, repeatedly, that you’re hardwired to be better or worse at something simply because of your gender, it can shape your sense of self and what you believe you’re capable of. Or not. In other words, these stereotypes essentially become a self-fulfilling prophecy.

However, the same could be said about the social forces dragging boys down in language-based skills. While girls are subtly — and not-so-subtly — steered away from maths, reading is still often seen as not ‘boyish’ enough of an activity. Unsurprisingly, the recent French study found a gender gap in language as well, this time favouring girls. But unlike the maths gap, the language gap was already present before school began — and didn’t change much during schooling. So, if school isn’t the culprit here, what is? I’ve actually written more about this in a recent essay. And as I noted there, recent studies show that baby boys are far less likely to be read to daily than baby girls. Parents are also less likely to read to their sons than to their daughters and are less likely to encourage them to do it themselves. The result? Boys read less — especially fiction — choose simpler books, and tend to struggle more with reading comprehension and language skills than girls.

But let’s not forget that reading proficiency is also the foundation for learning across the board — and a major building block for empathy. So when boys fall behind in reading, it affects their performance not just in school, but in life and relationships, too.

The confident claims about ‘hardwired’ gender differences in ability or interest — which, today, are also propped up by studies exaggerating or misinterpreting sex differences in human neuroanatomy — clearly hurt us all. Not just as children, but well into adulthood.

And yet, they persist.

---

Before the digital age, computers were actually humans. And mostly female humans. That’s because throughout the 19th century and well into the 20th, computing and calculating were considered tedious, menial, and simple enough work ‘even’ for a woman to handle. In the early 1940s, a member of the Applied Mathematics Panel even coined the term ‘kilogirl’ to describe ‘roughly a thousand hours of computing labour.’

Still, the work wasn’t nearly as simple as people thought. As historian Mar Hicks points out:

In a lot of cases, the women doing these computation jobs actually had to have pretty advanced math skills and math training, especially if they were doing very complex calculations.

It was also these ‘female computers’ who, while underpaid and underrecognised, catalogued thousands of stars, supported engineers during both world wars, developed new computational methods, helped get humankind to the moon (thanks in large part to Black mathematician Katherine Johnson) and, essentially, laid the groundwork for the digital age. But when computing became male-dominated in the early 1990s, and its prestige began to rise, the women who once pioneered it were swept into history’s dustbin, their contributions minimised or forgotten.

That’s the thing about patriarchal, status-quo-serving myths. Once they take root, they don’t just survive contradiction — they absorb it, twist it, and bend reality to fit whatever version is most convenient. Just like what happened in the case of computing. When it was mostly done by women, it was framed as easy as pie. But when men took it over, suddenly it became brilliant, complex, and worthy of big bucks.

It’s also no wonder why the stereotypes of women’s inferior intellectual capacities still exist when the many women who disproved them continue to be mostly absent from our collective memory. Unfortunately, even women who continue to make a difference today don’t always get the recognition they deserve either — they remain underrepresented in science awards and prizes, with their contributions still less likely to be recognised across most fields and at nearly all career stages.

And yes, it does matter that we point all this out. Multiple studies have shown that exposure to counter-stereotypical information can be a powerful antidote to gendered assumptions. Like our cognitive abilities, these stereotypical beliefs are — thankfully — malleable. They can change. But only if we’re willing to challenge them. This could mean, for example, exposing children to diverse role models they can relate to, helping girls cope with maths anxiety or build confidence in STEM, and encouraging boys — and their parents — to embrace frequent reading.

The truth is, gender gaps in intellectual or language abilities don’t persist because of ‘innate’ differences between girls and boys.

They largely persist because of gender stereotypes that insist those differences are real, along with all the nudges and cues and rewards — or penalties — quietly steering girls and boys, then women and men, down two separate paths while pretending it’s just ‘nature’ at work.

---

The 1992 Barbie doll wasn’t entirely wrong — maths class can be tough. But it gets a whole lot tougher when you, or the people around you, believe you’re being held back by some immutable trait you can’t do anything about.

Girls shouldn’t be expected to just power through a system stacked against them — they deserve a system that stops confusing cultural bias for scientific fact, and that actively works to eliminate such inequities.

---