UNIVERSITIES and workplaces are often targets for improving the number of women working across the “Stem” disciplines — science, technology, engineering and mathematics.

But new research suggests that children’s earliest school days are the beginning of gender disparity in mathematics ability, thus raising the question of how far these problems can be solved in isolation.   

It is a long-established and obvious fact that women are underrepresented in the Stem disciplines. It manifests in several different ways: across the Stem subjects, fewer women are admitted to university, employed in jobs (whether in academia or industry), or receive awards and grants for their research.

Many different initiatives have been proposed to mitigate the stark gender disparity in Stem workplaces, from “awareness raising” talks to diversity, equality and inclusion committees, to charter marks funded by grants from the European Union (such as the Athena SWAN award). 

It’s certainly true that changes can and should be made on a workplace level.

In the early ’90s, Nancy Hopkins was a professor in developmental biology at the Massachussetts Institute of Technology (MIT). She was no stranger to sexism in her department or her field, having struggled for years to get recognition and support for her research, but it was only when she was denied extra laboratory space for her experimental work on zebrafish that she decided to do something about it.

To prove that there was a gender disparity in the amount of laboratory space allocated within the department, she got a tape measure and measured her entire laboratory, as well as those of her male colleagues. 

The tape measure confirmed what she already knew: not only did she have less space than her fellow tenured male colleagues, but also less than the average for male junior professors. She had also been recently removed from lecturing a course she had co-developed in favour of a male colleague.

Armed with this evidence, she wrote a letter to the MIT president about the gender discrimination she’d observed, and was quickly joined by almost all the other tenured women faculty in the School of Science. At the time, there were 15, compared to 202 men. 

Quickly, their demands grew to encompass salary disparity, maternity leave, combating isolation, and the lack of women on committees and leadership roles within the institution. Hopkins and her colleagues’ efforts eventually led to action: MIT, among other things, gave out back pay to compensate them for years of salary inequality, and built day care on campus. 

Despite these material gains, MIT remains a place with a sizable gender disparity in promotions and hiring. And problems of sexism like those experienced by Hopkins still persist. At Scripps Institute of Oceanography in La Jolla, California, a 2023 report revealed that its women scientists have on average half as much research space as their male colleagues. Not only that, but less storage space and even smaller offices. 

The underrepresentation of women in Stem is stark at these academic institutions. Yet a study published in the journal Nature last month reveals troubling evidence that the “gender gap” could begin far earlier, at least in mathematics: almost as soon as children start school. Led by PhD student Pauline Martinot, researchers assessed the mathematical capabilities of over three million French children in their first two years of school (five to seven years old), in order to assess when gender disparity in mathematical ability becomes obvious.

The researchers note that in infancy and early childhood, “boys and girls exhibit similar core knowledge of number and space.” Therefore, the eventual gender disparity that develops is thought to reflect an internalisation of “the sociocultural stereotype that girls are bad at maths” — either on the part of the child or those around them.

The study suggests that this internalisation begins very early on and happens very rapidly: boys and girls performed similarly in maths exercises upon entering school, but after just four months, a gender gap in favour of boys became highly statistically significant. These findings were replicated each year over four years and occurred independently of the child’s age upon starting school, strongly indicating that schooling is the driver: the disparity is social rather than innate. 

Perhaps a reason for this, the researchers suggest, is that primary school is the first time that maths-related activities are clearly identified as “maths” as a separate subject from other exercises.

This labelling may be the trigger for gender stereotypes surrounding maths to emerge from both teachers and families, and end up internalised by children, thus affecting their performance. The simple axiom that “boys and girls are different” will reinforce any small gender disparity that emerges. 

The research also found that a larger gender gap emerges for children from families with higher socio-economic status, which has been corroborated elsewhere.

The authors propose that this could be a consequence of such families investing more in formal education at the onset of schooling. Ironically, whether through helping children with homework or paying for extra tutoring, these families likely provide more opportunities for gender stereotypes to be transmitted to their children and reinforced. 

There is, by now, abundant evidence that from the highest level of Stem research to the very earliest school days, women and girls are repeatedly subject to sexist attitudes that actively reduce their own capabilities and abilities to think, research and problem solve.

Changes that target the workplace and the adult as an atomised individual are battling against insidious forces of oppression that have a pervasive influence years before.

What is to be done? The study presents some strategies for intervention, but these are largely underwhelming. They include “improving teacher training,” “providing girls with ways to cope with competitive stress” or “implementing self-affirmation tasks to protect girls from stereotype threats.”

One issue the authors don't engage with is whether improving gender disparities on an educational level is even possible without wide-scale societal transformation, a transformation away from a system based on patriarchal oppression.