It’s a well-known fact that boys underachieve. Every statistic tells us so. But ever since writing this post I’ve been suspicious of gender as the root cause for differences in achievement. Yes, girls outperform boys but is this due to fundamental differences in gender? Or is it more to do with expectations, perception and bias? Or is it, perhaps, an illusion? Might differences in performance be due to other, less beguiling causes?
There’s no doubt that boys and girls are biologically different. But, as Gertrude Stein put it, “A difference to be a difference must make a difference.” Do the very obvious biological differences between the sexes actually make a difference to their academic performance?
This debate between two eminent Harvard professors of psychology, Steven Pinker and Elizabeth Spelke on the science of gender and science is fascinating:
But it’s also very long and I imagine few readers will set aside the two hours needed to appreciate their arguments. Very briefly, Pinker argues that the reason that women are under-represented in the physical sciences and mathematics is, at least in part, due to biological differences whereas Spelke argues that these differences make little difference and the under-representation is due to social factors such as biased perceptions and unequal opportunities.
What’s particularly interesting is that both cite much of the same evidence and research to support their arguments. The facts are not really in doubt, the debate comes down to a matter of interpretation.
Apparently for most human characteristics, while the mean scores are identical for men and women, there are more men at the extremes. Take this example of the normal distribution curves of male and female IQ scores.
There is more variability between the data set of all men than in the data set of all women. To put it another way, if there are more male geniuses there are also more male idiots. Might the fact there are more male professors of mathematics and physics be because there are simply more very clever men? Another reason might be due to men and women’s different priorities. Pinker suggests that men are, on average, more likely to prioritise status at the expense of family. Men seem also to have a preference for things over people. If community service is at one end of a career continuum then physics or engineering would be a the opposite end. Not only that, men are statistically more likely to take risks, work longer hours and think mathematically.
Spelke, rightly, points out that the differences Pinker cites are not deficiencies and that there’s a mismatch between reality and expectations. She shows various examples of the effects of gender labelling such as experiments where participants are told that a baby is either male or female, or where participants are told CVs are from male or female applicants and then asked to evaluate either the child’s behaviour or the merits of the CV. She says that where performance is unambiguous, everyone agrees. If a child’s reaction is clear, there is no dispute about it, but where interpretation is called for, female children are more likely to be viewed as fearful, and sweet whereas males are more likely to be labelled angry or strong. In the case of CVs, star candidates will be feted no matter their gender, but in the case of more mediocre candidates, male candidates’ track records are interpreted as more productive and experienced and are far more likely to be offered jobs.
This is covert rather than overt sexism. Parents do not necessarily deliberately treat girls and boys differently, but of course they treat angry or fearful children differently. Employers may genuinely want to bridge the gender gap but it’s more reasonable to prefer more experienced and productive candidates. I’m as guilty of this as anyone as I explored in this post. It’s certainly not the case that I deliberately favour male contributors, but it may well be the case that I judge the contributions of men and women differently. It’s impossible, Spelke says, to judge intrinsic aptitudes while society’s perceptions are so biased.
Both scientists make excellent points, both agree that girls get better grades at school in all subjects and both concur that biased perceptions and unequal opportunities are pernicious, but where Pinker says gender differences make a difference, Spelke says they do not.
As a father of daughters, I worry about all this. Despite the likelihood that they will outperform their male peers at school, statistically, there’s little doubt they’ll struggle more to make their mark on society. If this is, in part, due to real biological differences, does that make matters better or worse? When surveyed, women’s priorities include:
- Being able to work part time for a limited period
- Living close to family and other relatives
- Having a meaningful spiritual life
- Having friendships.
Men’s priorities are:
- Having lots of money
- Inventing or creating something of worth
- A fulfilling, full-time career
- Being successful.
Does it really matter to what extent these priorities are socially constructed? I have no idea, but it’s clear which of these sets of priorities is more likely to help my daughters forge ahead in the world of work. Is it also clear which list is more likely to make them happy? I recognise myself in the second list and feel at least a little ashamed.
Of course, women deserve the opportunity to be as successful as men in any field in which they wish to compete. Feminist activist, Gloria Steinem said, “There are really not many jobs that actually require a penis or a vagina, and all other occupations should be open to everyone.” She’s right. But we ought also to remember that fairness is not sameness. Obviously women should be allowed to judge their successes differently, but maybe society ought to reward rather than punish women for these differences?
I had intended to start with this quote from self-proclaimed ‘dissident’ feminist Camille Paglia, but on reflection decided it might work a lot better at the end:
In the theory of gender I began from zero. There is no masculine power or privilege I did not covet. But slowly, step by step, decade by decade, I was forced to acknowledge that even a woman of abnormal will cannot escape her hormonal identity.
Vive la différance!
I think that my concern is with the first sentence – I do not think boys underachieve per se but that there is a problem with a testing system that is fixated with (i) the kinds of tests we have that are headlined as success (generally the ones that take place in examination halls) and also the fixation for having tests at a fixed chronological age.
If we took a random sample of 16 year olds and then asked their teachers to rank them by (a) physical development (b) emotional development (c) social development then we would find (I think) that 16 year old girls would dominate the upper end of this sorting. If we did the same again by 20 we would find a more even mix. So, we could hypothesise here that girls mature cognitively at a faster rate as well and so we are not testing the same things when we have this chronologically fixed test at this time.
I would be interested in there is any data between the examination results of September/October born boys against July/August born girls to see what this data would look like?
I would contest your statement, “Does it really matter if these priorities are socially constructed”. Yes, if there is a pressure of social acceptance that they determine how those who do not fit the norms are treated by that society.
Most tests turn out to be very good predictors of all sorts of things. The phonics screening check (which has been used by the TEDS studies for decades) accurately predicts GCSE results at 16 and income at 42. In the US, SATs accurately predict the likelihood of becoming a professor, patents etc. and apply equally to girls and boys.
There’s no doubt that the age we are when we take tests makes some difference, but these differences narrow as we age. So the differences between a September and August birth make far more difference the younger you are.
And my ‘statement’ was a question. For some reason you omitted the question mark in your quote. Rather than contesting it, you’ve answered it 🙂
I was a researcher on Cambridge University’s ‘Raising Boys’ Achievement’ project. Aspirations and achievement may be influenced by gender but they are equally influenced by culture and class. Merely focussing on gender is too simplistic.
The middle class majority government in this country and indeed the middle class majority of the teaching profession, enforce their definition of success on society. The same is true of America. Percieved underachievement in school does not equal underachievement in life (not just income or status) and perceived achievement in school does not equal achievement and fulfilment in life. Talk to the depressed man with a masters degree doing a manual job, or a female GP with no work life balance. (I can give you their numbers if you’d like)
Interesting review of the two main positions, thanks David.
Are you aware of the work being done by the Institute of Physics on gender balance in schools? This started with investigations into the participation and attainment of girls in physics, but has since developed to encompass whole-school issues towards gender perception. The evidence indicates that this is the area where interventions need to be directed. From working with a number of schools over the last few years, however, it’s clear that a significant problem exists in school culture.
Schools are not incentivised to look at gender issues; staff awareness of unconscious bias within their own classroom practice is often low; use of sexist or gendered comments by students (or staff) is not picked up or sanctioned in the same way as racist or homophobic language; parents and students are not engaged with gender issues; and, of course, careers and options advice, particularly relating to GCSE and post-16 subject choices, often unconsciously perpetuate gender imbalance.
We therefore believe that a significant effort is required by schools to challenge and overcome some deeply-ingrained societal attitudes and perceptions about gender and identity. This goes beyond physics or even STEM, and requires support and engagement across the school (staff and students); making this problem a focus of both ITT and in-service training for teachers and other staff; and needs to be driven forward by governing bodies (including, for example, equalities impact assessment reviews of all internal and external policies). Ultimately, the responsibility for addressing gender issues in schools lies with senior leaders and headteachers, who should not be allowed to accept the status quo unchallenged; seeing such potential talent lost through implicit bias and stereotyping is a national scandal.
If you’re interested, you can find out more about the work of the IOP on these matters at the link below, and we are holding a conference in October called ‘Opening Doors’ to launch a new report and recommendations for schools.
All the best
David
http://www.iop.org/education/teacher/support/girls_physics/page_41593.html
No, I wasn’t aware of the IoP work on gender. Thanks for the link. I am aware of various school based initiatives such as the rather patronisingly dubbed ‘skirting science’. I’ve no idea what kind of impact these have though.
I’d agree that senior leaders are responsible for addressing issues of inequality – this is tricky in a profession in which the majority of teachers are female but the majority of school leaders are male.
Of course up until the late 1980’s much research (psychological, sociological and educational) and government intervention was focused on why girls underperformed, compared to boys, at all levels (and most subjects) after the age of 12. The assigned title for one of my first essays at university, back then.
Well, quite. As most of the grosser, overt sexism has been hacked back, women have been more able than ever before to achieve something like their potential. Interestingly, Pinker notes that in many academic fields women are very well represented on PhDs and tenure track. In the US, 2001, 65% of the PhDs earned in education were by women, 54% in social science, and 47% in life sciences. But only 26% in physical sciences and 17% in engineering. He cites this as evidence that the women who achieve at the highest academic level prefer poeple over ‘things’.
I was taught about gender stereotyping during PGCE and think it’s very important for teachers. It’s a great shame this has been lost. I fully agree with David C above that whole school culture needs addressing – via PD and through TeachMeets and through Twitter. Delighted to be part of @WomenEd raising and supporting these issues.
Hi David, you ask a philosophical question around which priorities will help your daughters “forge ahead in the world of work” and which make it more likely they’ll be happy. When I was a child my mum nodded to sexism when she said; “You’ll need to work harder than any man for the same results.” And so I did. And I saw what she meant. She was right. She called me to be smarter than other people’s prejudices, to dispel this issue by being an example. Yet I still lost my way because I was trying to be an example within a world economy dominated by the masculine values… your second list. At the point of being recommended anxiety pills “to make me normal again” I removed myself from that world and am currently dedicating my time to figuring out who I want to serve and creating my own space to serve those people, in my own way. It’s unbelievable how many adults I’m meeting now who carry the same story; men and women who’ve left their jobs at the point of breakdown are now flourishing with their own ‘feminine value’ businesses. Here’s an example of how it’s done: http://peopleoverprofit.com/4-slow-difficult-steps-to-building-a-million-dollar-company/
I haven’t had chance to watch the video yet, but it sounds like it’s exploring some similar terrain as Baumeister who wrote ‘Is there anything good about men’. This is a good introductory speech-http://denisdutton.com/baumeister.htm
His view is that men tend to dominate both the ‘top’ and ‘bottom’ of society and explores some possible reasons why. Definitely an interesting read.
I’d be interested to know how those IQ figures were obtained. Setting aside the fact that it is a contested concept, just what percentage of a population actually takes an IQ test? What was the sample size, the demographic etc?
To answer MaggieD’s question, the norm groups for IQ tests are generally the most carefully constructed, balanced, and representative groups used in any research. More importantly, different ways of constructing norm groups, by different researchers, in different countries, produce similar results. I am not sure that anything can be regarded as a “fact” in social science, but the claim that males are more variable than females in measures of cognitive function seems to come pretty close.
Thank you 🙂
[…] 7. Do gender differences make a difference? 18th July […]
I’m reading ‘What if everything you knew about education was wrong?’ and having finished the first chapter was inspired to write on one the examples about the impact of gender in schools.
Now, this may be me displaying one of the strategies to counter cognitive dissonance, but I had some questions about the Mr Garvey example. I teach in a subject where, at my school last year 73% of girls achieved a C or above and 55% of boys did so at GCSE. So I have a motivation to understand these differences; and address them if, one, they can be addressed and, two, they should be addressed.
The fact Mr Garvey identified pupils’ attendance and Key Stage 2 English (and Teacher Attendance, but I’m not going to focus on that) results as the most significant factors affecting a students’ GCSE results does not seem to me to leave the hypothesis that gender has an impact falsified, since we can still interpret why a student’s KS2 English and their attendance is lower as having some possible link to gender. Now, my expertise in statistics is none, so I wonder if this is just basic misunderstanding on my part – the fact that gender was identified as a non-significant factor in Mr Garvey’s data analysis might count out this possible interpretation. However, the question that remains with me is why is it that more boys than girls are doing worse at KS2 English and have lower attendance (if that’s what Mr Garvey’s analysis suggest) over repeated years?
Somewhere else in the first chapter the book cites some research on behaviour in school as a key factor in doing well. But this led to confusion on my part. If, behaviour in school is a/the key factor and if the ‘distribution of boisterous pupils among two genders is much the same’ why is there a repeated disparity in the final figures between girls and boys average GCSE results? Surely, if there was a ‘much the same’ distribution of behaviour between genders and that was a very decisive factor, we’d expect roughly equal average achievement between genders (or at least, no pattern of lower average male achievement over recent years). In short, what becomes the explanation/s for the repeated unequal average achievement between genders in recent years if it has nothing whatsoever to do with gender?
I’d appreciate any help with those questions; and in the spirit of the book, I’m happy to be wrong-headed with my questions, (well, part of me is – the bit that wants the truth, not the bit that wants to be right). Really enjoying the book.
I also thought about Mr Garvey and the emotional responses of him and his bosses. His bosses probably felt they were trying to do all they could to help students and that Mr Garvey was being supercilious – a little professor. Mr Garvey probably felt self-righteous and attuned to the truth against the unthinking idiots on high. I wondered how best for them all to get through that emotional dissonance so that they could work together on doing what they all wanted (or at least the better angels of their nature wanted) – improving the outcomes of students as best they could, based on the best available understanding of what would be effective to that end.
Would you like me to pass on your query to the ‘real’ Mr Garvery?
Yes please, that would be great.
My god, it was a little scary seeing how the list of men’s priorities agree with my own (particularly 2 and 4).
The problem I have with Pinker’s argument is: how much of the observed differences are due to decades or centuries of gender bias? E.g. Is the figure of 26% studying for PhDs in the physical sciences due to either explicit or implicit gender bias driving those women away? Not to mention that academic career structures are basically tailor-made to suit men rather than women, with there being a huge lack of job security and money during the ages of 25-40.
There is still a shocking amount of outright sexism in the physical sciences, never mind implicit sexism.
It may well be true that men prefer physical sciences, but I have difficulty believing that the “equilibrium” ratio should be 74:26.
The IOP report points out some of the issues girls’ cite as reasons for not continuing with physics post-16 e.g.:
1) Unclear career options.
2) Belief that you need to be a genius to do physics.
3) A focus on equations/maths rather than ideas.
All of those can be easily fixed or debunked, with a little effort. Bear in mind, 49% of schools in the UK didn’t send a single girl on to A-level physics when the report was written (compared to around 15% for boys, if I remember correctly), a shocking figure. We can do a lot better than that.
On balance, I really think the differences in physical sciences participation is likely due to lingering explicit/implicit sexism, and failures to address girls’ concerns when it comes to studying physical sciences, rather than innate gender differences.
[…] Growing up, girls usually have average scores, skills and retain that through adulthood also. Meanwhile boys are known to underachieve in childhood/teenagehood, but later scatter more into “low” or “high” skill areas. Getting […]
[…] Growing up, girls usually have average scores, skills and retain that through adulthood also. Meanwhile, boys are known to underachieve in childhood/teenagehood, but later scatter more into “low” or “high” skill areas. Getting […]