So let's be clear, I am raising the question of participation, not performance. In fact, when you look at a subject such as physics at A-level, the girls get the better results. But about three times as many boys as girls sit the exam. From the common platform of A-level physics girls and boys go into quite different subjects at university. While the girls head mainly for medicine, dentistry, veterinary and biological science, the boys opt for engineering and technology, maths and computing and physics itself.
Should we be concerned? There has been a vocal strand of feminism which has maintained that the imbalance is prima facie evidence of prejudice, and social justice demands that girls and boys should come to realise that there are no masculine or feminine subjects. But even if this is accepted as a worthy objective, how is it to be achieved?
The proportion of female entrants to A-level and university physics degrees has remained remarkably constant over the past two decades, at 23 per cent and 19 per cent respectively. None of the campaigns such as Women into Science and Technology, nor any of the experiments with single-sex classes, seem to have had a discernible impact.
In fact, girls as a proportion of those sitting physics A-level rose from 14 per cent in 1961 to 23 per cent in 1985 with no direct attempt to increase involvement. (It seems to have come about through the increased opportunity provided by comprehensive education.)
If exhortation and demonstration do not work, some would advocate something more drastic. The biennial Gender and Science and Technology Conferences have after nine gatherings come up with a manifesto for Gender-Inclusive Science that advocates "minimising students' capacity to choose against scientific disciplines". In other words, force them to do it. In part, this was the thinking behind the dual-award science GCSE that has been associated with a fall in both boys' and girls' taking physical science A-levels.
It also emerges in Gray's work at Cambridge that the schools with the smallest gender differences tend to be those with the poorer exam results, perhaps because potential is not being fully developed. So it appears that we can contrive a better balance between the sexes in subjects by limiting choices and capping performance. But neither seems inherently desirable.
Maybe we should accept that girls and boys want to do different things. On this view, the only educational concern is to ensure there should be no artificial barriers. But what is an artificial barrier?
It is grudgingly accepted now, even by the most ardent feminists, that there are psychological differences between the sexes. From the earliest years there are small differences, on average, in verbal abilities with the advantage to girls, and in numerical and spatial abilities with the advantage to boys. These could be expected to lead to proportionally more girls opting for English and languages, and boys, maths, engineering and the physical sciences, which seems to be the case worldwide.
But it is also argued that the script that society writes for our behaviour determines subject choice. Hence, so the argument goes, the need to intervene to change unfair prescription. This is plausible.
But while the scripts for girls' and boys' lives have changed massively in recent decades leading to, among other things, the proportion of women at university rising from 40 per cent in 1980 to 54 per cent now, this has occurred mainly through the addition of new subjects such as nursing, design and tourism. The gender balance in the core subjects has been largely untouched. It seems, therefore, that whatever prompts men and women to veer in different directions, it is deep-seated.
But does it matter that the sexes, when given the chance, choose different subjects? The onus is on those who see it as a problem to restate the reasons why and not rely on facile assumptions. Furthermore, they need to show how any changes can enhance opportunity rather than restrict choices. At the risk of sharing Larry Summers' fate, it is a question that has to be asked.
The writer is Director of the Centre for Education and Employment Research at the University of BuckinghamReuse content