Education Fails Boys

There’s a nice bold statement to kick things off. What’s the evidence for this?

Part 1: The Boring Statistics Bit

If you’re short on time or attention span, you might want to skip to the infographic at the end of this section. But if you’re still with me, let’s dive right in…

First, there’s a gender gap (with boys at the disadvantage) of nearly 20% even in primary schools, and of 7% at GCSE level — the exams that all children in England, Wales and Northern Ireland are expected to take, marking the end of mandatory education.

Those hoping to go to university will generally stay on for another two years to take A-levels — these determine which universities will accept you onto their courses. At this stage, girls outperformed boys in every subject except modern languages, taking more subjects and achieving higher grades. Unsurprisingly therefore, more girls go on to fill university places (49% vs 38%). Amongst those actually completing an undergraduate degree, women outnumber men 1.3:1, and are more likely to obtain 1st-class and 2.1 (“upper seconds”) degrees while men are more likely to obtain 2.2 (“lower seconds”) or 3rd-class degrees (yes, we have a slightly eccentric classification system here).

Notable throughout many of these reports is that there’s often an exception for a narrow range of subjects in which boys still have the edge. There’s much hand-wringing about these “STEM” subjects (science, technology, engineering, mathematics — although this is slightly a misnomer, since it doesn’t apply to all areas of science). It’s worth remembering, then, that these figures, showing women far in the lead, include and are thus reduced by, the effects of the gender gap in the STEM fields. People are working hard to close that gap, and I certainly agree this gap should be closed — but when it is, if nothing is done about every other field of study, then the boys are going to be even more severely behind.

(It’s worth noting, though, that while boys definitely make up the majority of those choosing to take STEM subjects — except biology — girls are still ahead in terms of achieving the top grades. More on this in the infographic.)

Perhaps it’s just my country? Let’s look at some figures from the US Department of Education’s National Center for Education Statistics. These are really interesting because they go back to 1869(!) and nicely show the progression over time, as well as predictions for the future based on current trends.

So back in the 19th century I’m sure you’ll be shocked, shocked to read that men out-achieved women in obtaining undergraduate degrees, nearly 6:1. I’m actually a little surprised that as many as 1,378 women had the opportunity to obtain those degrees in the US back in 1869, but hey, go them! Anyway, if we look at the middle of the 20th century, we see that the M:F ratio is still as high as 3.1:1, but over the course of the next two decades, huge strides were made, with the ratio down to 1.8:1 by the start of the 60s, then 1.3:1 by the start of the 70s. Not parity yet, but making progress.

During the 70s, there was a slow but steady balancing-out, until by 1980 it had reached 1.01:1 — just a tiny tiny skew in favour of men. But as we can see, it didn’t stop there: By 1981, the bias was 1.01:1 in favour of women, and continued to grow. Again, slowly at first, taking the entire rest of the decade to hit a 1.1:1 bias in favour of women. But the bias accelerated, until by the present day, it’s more than 1.3:1 in favour of women — in other words, men now have the same status at university, as women did in the late 1960s!

Is it just undergraduate degrees? Nope. At masters degree level, the effect is even more profound. Again, we can see the historical record of inequities against women. Again, we see huge strides made from the 50s onwards, with equality achieved in the 1980s. And again, we see women pull ahead, except here the bias is even sharper: by the time we reach the present day, the most recent actual (non-predictive) figure shows women are outnumbering men by 54%!

What about at PhD level? Here, interestingly, things are a little different. Women didn’t achieve equality until 2007, which is pretty shocking when you think about it. Of course, you would expect some time-lag given that it takes a long time and a lot of education to even reach PhD level, and then the doctorate itself takes a long time to achieve. It still shouldn’t have taken that long though. But… even here, women have just started to edge ahead, and are predicted to continue to do so — by 10% the subsequent academic year, and 26% by the end of the decade.

Oh, and the statistics look similar in Canada, too, in case you were wondering.

So here’s the infographic. The first part, on the left, shows how degrees awarded in the USA have changed over time. The second part, on the right, sums up STEM participation in the UK (click to zoom):

Part 2: Soc Psych Interlude

So, now we’ve established the situation, I’m going to take a little time out to talk about social psychology.

Human beings are weird. I mean, I like this about us — that we’re weird. It’s interesting and fun. We like to think we’re smart and logical, and we sort-of are, but we have millions of years of completely irrational evolution behind us, and it has a habit of creeping up on us unawares.

At a very broad level, the logical parts of our brains are powerful and useful, but are slow and require a lot of resources to keep active. Much like a laptop running off batteries, we send less power to the “expensive” parts if we can, using older, more efficient, and more instinctive parts of our brain instead wherever possible. These are very fast and surprisingly effective, but they do make mistakes — and they work in really, really strange ways.

For a fun example of this, check out Why Susie Sells Seashells by the Seashore, which shows that even something as simple as the initial letters of your name can impact major life decisions. People named Dennis or Denise, for example, are considerably more likely to choose a career as a dentist! Surprisingly subtle cues can influence our behaviour.

What I’m going to be mostly focusing on are two aspects of soc-psych research that have strong implications for learning and exam results: Stereotype threat, and the Pygmalion effect.

Stereotype threat essentially says that if you’re a member of a group, and there’s a negative stereotype about that group that affects a task, you’re more likely to perform badly at that task — especially if you’ve recently been reminded about that stereotype.

Theories vary as to why, but one of the most popular is that part of your mind (which ideally would be working on the task) is distracted by worrying about whether you’re living up to the stereotype or not. It’s been shown, for example, to decrease the amount of “working memory” (the mental equivalent of scrap paper to make temporary notes on) you have available to you.

A cunning way of verifying that the stereotype is preying on the person’s mind is to give participants a series of partial words (eg “_ _ m b”) and ask them to complete them. Filling in “numb” or “lamb” or “limb” would indicate the stereotype is not weighing on their mind, but filling in “dumb”^† would indicate that it is. Both this technique and the stereotype threat effect are covered by Steele and Aronson’s paper.

(† bearing in mind this blog’s stance on ableism, remember that we’re simply uncovering people’s subconscious fears and biases here.)

The interesting thing is, you can create these kinds of effects completely arbitrarily. You can simply tell children that group-of-children A perform better at a test than group B along some randomly-picked line, and the test results will bear this out. Tell a different classroom of children that B perform better than A, and the results are flipped.

This seems to tie in closely with the Pygmalion effect, which shows that pupils’ ability to learn is affected by the beliefs of their teachers, even if the teachers keep those beliefs strictly to themselves. In the original Oak School Experiment in 1968, researchers told teachers that they’d checked the pupils with a new test, and it indicated that certain pupils were late-developers who would suddenly show rapid growth in their mental abilities during the course of the next year.

At the end of the year, those children had, indeed, achieved gains 2.2x higher than their peers. The problem was, the “new test” was entirely fictional — the pupils had been chosen at random! The simple fact that the teachers expected more from them, had become a self-fulfilling prophecy.

Unsurprisingly, the same effect has been shown to work in reverse — kids live down to low expectations, as well as up to high ones. At this point, you might know where I’m headed next…

Part 3: But What About Teh Menz?

Back to our classrooms. We clearly have a problem with boys failing in education, but why?

Some people think “that’s just how boys are”. One of the articles I linked to earlier suggested that “maturity and motivation” play a part, and some people might wonder if “testosterone poisoning” is also involved. These arguments, funnily enough, I find flawed on many levels, including the fact that the statistics apply to people well into their 20s, the fact that testosterone just doesn’t work that way, the fact that it stinks of the kind of sexism that excluded women from education for a long time, and that in Part 2 we saw how these kinds of views have a dangerous tendency to become self-fulfilling prophecy.

Do these stereotype effects apply to our boys?

Bonny Hartley thinks so, as a result of experiments she conducted in schools in Kent. These tests — which contain haunting echoes of the Clark Doll Experiment — involve providing a child with two silhouette drawings, one of a boy and one of a girl, and then a list of statements. For each statement, the child is asked which of the two drawings the statement applies to. The statements don’t have any explicit gender cues like “he” or “she”, so this lets us find out the child’s own gender-biases (if any), without asking leading questions that might influence their answers. It’s quite a nifty technique.

The statements posed were about behaviour, intelligence and success in school, such as “This child is really clever”, or “This child always finishes their work”. More than 200 children of various ages were tested, and the results were:

• At age 4, boys exhibited no gender bias. They seem to pick randomly between the pictures.
• At age 4, girls consistently pick the girl picture as being cleverer, better-performing, more focused and better behaved than boys.
• The boys continue to hold no gender bias for about 2-3 years, but by around the age of 7 or 8 they start to exhibit gender bias… in favour of girls, at their own expense. Their views start to conform to that of the girls, and this continues throughout the older age groups.
• The girls continue to hold the same views throughout.

I’d love to know more about what’s giving those 4-year-old girls such strong anti-male prejudices. I’d like to also draw attention to the fact that the questions didn’t just cover the children’s own views, but also asked the children what adults believed.

The age range of Bonny Hartley’s research covers primary school level. Only one in eight primary school teachers is male. One in four primary schools have no male teachers at all.

If you’re thinking male teachers might favour secondary schools, you’re right — and yet still, across the school system as a whole, only one in four teachers is male. Not only that, but men are fleeing the teaching profession — this imbalance is accelerating dramatically, with only one new male teacher for every 15 new female teachers entering the profession.

I couldn’t find figures for the US showing the gender of existing or new teachers, but the closest useful piece of information was the number of students obtaining degrees in education. I am going to make the assumption that this tracks new teacher arrival relatively closely. Women have been oustripping men at obtaining these degrees for the past couple of decades — by more than 3:1 (Masters or Doctorate level) or 4:1 (undergraduate level).

Is this relevant? It’s hard to say for sure. Some research has been done into this subject, and it seems to be contradictory. Some, for example, finds that male teachers are vital role models. Other reports, comparing test results of pupils whose teachers’ gender either does or doesn’t match their own, seem to show that having a male teacher doesn’t help boys learn better.

The problem there, I think, is that they’re looking only at the context of a single class, rather than the educational system. It’s not a question of boys needing male teachers to learn. It’s a question of the current system essentially comprising a female-dominated workplace, one which Ms Hartley’s research suggests is distinctly misandrist — particularly at the primary-school level where children are learning what school is and how they fit within it. In an employment context, we’d call that a hostile working environment.

My belief is that to improve boys education, it’s not necessary to pair each boy with a male teacher (or each girl with a female teacher). Instead, it’s necessary to change those fundamental attitudes that are telling boys, before they even reach the age of 7, that they are inferior.