Two positives make a negative for gender bias in Stem

Could girls’ superior reading skills really be what’s holding back gender diversity in Stem?

Oxford linguistic philosopher J L Austin was invited to give a guest lecture at Columbia University in New York 25 years ago. Known for his expertise in ordinary language philosophy, he noted how many languages there were in which a double negative makes a positive, but none in which a double positive makes a negative.

To which his sharp-witted contemporary, philosopher Sidney Morgenbesser who happened to be in the audience, sarcastically replied, "Yeah, right."

Notwithstanding the gender of both characters in this anecdote, I was immediately reminded of it upon reading a report published this month in the Proceedings of the National Academy of Sciences.

This wide-reaching study concludes that the “comparative advantage” that girls in high school have in reading compared to boys “is the primary reason why women are outnumbered by men in technical fields”.


Using data from the Programme for International Student Assessment involving 300,000 high school students from 64 countries, the study demonstrates how female students who are good at maths are far more likely than male students to be even better in reading.

Consequently, the difference between 15-year-old students’ maths and reading abilities, which is frequently determined by socialisation processes during much earlier stages in their development, “explains up to 80 per cent of the gender gap in intentions to pursue maths-studies and careers”.

The two lead authors – economists Thomas Breda of the Paris School of Economics and Clotilde Napp of the French National Centre for Scientific Research – found that 68 per cent of the students who were better at maths than at reading were boys, while 32 per cent were girls. Conversely, 68 per cent of girls surveyed were better at reading than maths with 32 per cent of male students proving to be avid readers.

The authors also compared individual students’ math skills with his or her reading skills. Here, the gender divide was even more acute. Across 64 countries, they found 59 per cent of boys surveyed were better at maths than reading, and 74 per cent of girls were better at reading than maths.


This isn’t a prosaic “girls read good, boys crunch numbers” study. It’s already well-established that diversity in any discipline or industry makes for improved research/product design outcomes. The authors dispute any implication that the results indicate girls don’t have what it takes to succeed in technical careers.

It’s not that they can’t do maths but the prominence of other skills and interests in different fields tends to sway future educational and career choices. Female students are less inclined to take more maths-related classes once they reach college age.

An insight that is "pretty impressive", according to NYU developmental psychologist Andrei Cimpian, (not involved in the study). "The more you perceive yourself to be good at 'X' relative to 'Y', the more you're likely to pursue 'X '," he said in an interview with the LA Times.

In any other context, being "good" at one subject but even "better" at another would be considered a good thing. But gender constructs still play their part in the make-up of Stem (science, technology, engineering, and mathematics) – a sector where women comprise just 23 per cent of the total workforce in the United States.

Notwithstanding the introduction of "Scientist Career" Barbie by Mattel in recent years and other gestures that play their part in moving things forward, it appears gender norms start to skew things from early on in a child's development.

But another novel study found the language used – which can so often be the source of the problem in the first place – might also be a solution to increasing female participation in Stem.


It has been shown that if girls as young as four are encouraged to get involved in maths and science without linking it to their identity, they are more likely to engage.

Researchers randomly assigned boys and girls a scientific task using different kinds of language to introduce the activity. Some kids were told they would be “doing science”, while others were told that they would “be a scientist”. Girls told they would be doing science persevered for longer periods than those told to “be scientists”.

Since girls don’t necessarily identify with science and maths, asking them to “be scientists” isn’t the right way to go about it. In order to get them interested in scientific projects the language used must focus on the value of the task itself rather than their role in it.

Whenever confronted with examples where marginalised groups – who may be able to provide solutions to big societal problems – aren’t even allowed to participate, my first thought, admittedly, isn’t sympathy for those excluded. It’s frustration. Frustration at the fact that we could be so much more advanced in Stem research if we could overcome the archaic patriarchal legacies dominating so many fields of expertise.

Despite the unique insight provided by this study, I’m sure it provides little comfort to women to learn that being good at maths and even better at reading somehow leads to gender inequality in Stem. Meritocracy? “Yeah, right.”