That post about women in maths
So, hi, I'm going to talk about women in maths, stats, physics, computer science and engineering.
This is the part where people generally go "Oh, of course. The number of women in those subjects is abysmally low. It's due to the way women are socialised to dislike/not be good at/think they aren't any good at maths."
Except. Not quite.
A little story: when I first started my undergrad, I was fully expecting to be almost alone in a sea of guys. After all, everyone knows there are really few women in maths! Then I went to my first lecture and - women. Lots of women. All around me, women.
A few years later I actually went up to the poster that had pictures and names of everyone in our year, counted, and discovered that we were exactly one-third female. (I felt it was half and half, but there's a phenomenon saying that people tend to register something as being equal in gender at one-third women or so, and perceive true equality as a bias in favour. Either that or the guys skipped more lectures.) At the time, I figured we had unusually many women; pretty much every statistic I have found since, however, tells me we were unusually FEW. Apparently most maths undergraduate courses these days are something like 40% female.
Let me repeat that. Most maths undergraduate courses are almost half female.
There are two things to note about this: first, that it definitely ISN'T so when it comes to computer science and engineering. I think maths, stats and physics are relatively even when it comes to female undergraduates - physics maybe slightly higher - and then for the other subjects in this area the numbers go down. AFAIK (I knoooow I need to look up these stats :() numbers of women in engineering undergrad are well below 20%.
Secondly, the number of women mathematicians falls dramatically at higher levels. And I do mean dramatically. An article I read in the i-squared magazine quoted the number of female PhD students as 20% and the number of full professors as 4% - need to do some stats digging to see if this is true, but it sounds about right for my own experience. This is known as the leaky pipeline effect and is actually at work in AFAIK every academic discipline - the higher up the academic career ladder you go, the less women there are - but is unusually strong in maths.
Neither of these things can be explained away by "women are afraid of maths!" or "women are socialised against maths in high school!" or whatever. For the first, there is clearly another effect at play here! For the second, okay, we need to figure out the whys and whencefores of the leaky pipeline in general but also why it disproportionately effects mathematicians.
So I'd dearly love it if we stopped blaming the entirety of lack of women in all STEM fields on the maths issue because if we do we are missing shit. I mean, don't get me wrong, there are definitely ISSUES there but in my experience we focus on these, a lot, and the fact that HELLO MATHS UNDERGRADUATES ARE ALMOST HALF FEMALE, that the issue doesn't seem so much getting women into maths (although it IS getting women into maths-related subjects like computer science!) but keeping them there, gets brushed under the table.
Here come my thinky thoughts on wtf might be going on.
For the first, next to "women don't do maths" we have other stereotypes! Namely, the ones about practical subjects. In my experience, women are expected to be good at certain practical hands-on things - knitting, cooking, all that women's work stuff - but not others, especially stuff to do with mechanics and repairing shit. No, that's for the guys. And that, I think, is what really hits computer science and especially especially engineering. Not only is it a guy subject, it's a sort of practical, hands-on guy subject! Maths, on the other hand, is completely theoretical, and I suspect it might be considered more socially okay for women because of that. (Something I would DEARLY love to know is statistics for applied vs pure maths and theoretical vs experimental physics, but there's confounding factors there - e.g. at my university most of our applied PhD students are from India, Pakistan and China, whereas pure PhD students tend to be from Europe and the US. Since different countries have different gender balances for maths I think it might be hard to compare, and undergraduate degrees are usually just a plain BSc.)
I have no idea whether this might explain the phenomenon, but I think it's definitely something I've observed in my own life - things like my brother learning to repair his own bicycle but my dad always doing mine for me, that sort of thing. And might explain why I never *once* considered doing a degree or even advanced high school course in computer science, despite the fact that I was good at it and enjoyed it (my mother taught my brother and I to program in Perl when I was nine or thereabouts and I'd write little programs for fun, and later on in school when we had a programming class and the assignment was to make a pixel move across the screen I'd have six moving and be working on whether I could make them change patterns upon pressing a certain key before my classmates were even halfway done) and my mother was a fucking IT professional. /o\
As for the leaky pipeline and its disproportionate effect, this is where I am going to start tossing various things about. I'd feel guilty about this but tbh I feel as if a lot of the "why so few women?" stuff is tossing theories about and I at least have the benefit of first-hand observation. (For probably better researched theories, the London Mathematical Society has a statement out about the loss of women at higher levels which is interesting. And, ha, some of those statistics I was looking for.)
(Disclaimer! I am a pure mathematician. It's possible a lot of this doesn't apply to applied maths!)
Here's a thought: although maths usually gets lumped in with the sciences, it isn't really one. Or at least, the way you do it is completely different. The scientific method doesn't apply, for instance, it's deductive rather than inductive, etc. (I had the fun of taking a Philosophy of Science course in undergrad and discovering none of what it talked about applied to maths at all.) In particular, at least in the pure bits PhDs look VERY different. Apparently in most areas of science you have labs and labwork. You work in a team and as a PhD student you do a lot of the gruntwork and little of the actual coming up with experiments etc. (Correct me if I'm wrong about any of this, science people.)
Maths (the pure bits, at least!) isn't like that, at all. Your supervisor sets you a problem, or rather sets you a lot of background reading and then sets you a problem. Then you're off trying to figure it out - just you, the pen, the paper and the wastepaper basket, to paraphrase an old joke. Your supervisor is there for questions and to bounce ideas off and use zer greater experience to guide you away from paths of investigation that will probably be dead ends, but for the most part? It's your problem.
There are a few things that occur to me about this, when it comes to the leaky pipeline.
For one, it can be really really isolating. Supervisors can cut their PhD students loose pretty awfully. Or less dramatically, if a supervisor isn't giving you a lot of training and feedback and help, that's going to have an effect, both on the quality of your PhD and your willingness to go into academia afterwards. And since it's all essentially one-on-one, a supervisor can easily give zer female students less care and attention than the male ones, whether ze means to or not.
For another, female students may be less likely to pipe up when they feel as if they're being neglected, due to various female socialisation issues. When I was looking for places to do my PhD, I was actually specifically looking for a person where I'd not only feel comfortable asking questions of them but also where they'd come and poke ME to tell them what I was doing and where I was stuck on a regular basis - because I'd done two projects and was doing a third and all of them had gone wrong due to me not asking shit of the person whose job it was to make sure I was doing okay.
Finally, I wouldn't be surprised if some supervisors subconsciously assumed their female students were more likely to leave academia than the male (because, statistically, they are) and therefore more lucrative research problems would be wasted on them. And so gave their female students, however unintentionally, problems that they thought were a little less interesting and a little less likely to lead into further research questions than the ones they gave their male students. (No statistics on this one, also I have no idea how you'd even test that.) Or they may be less willing to trust them with riskier but potentially more interesting problems in favour of ones that are 'safe' but not that interesting.
Another issue is the social one. Mathematics is actually a relatively small subject area in some ways, and due to the level of specialised background knowledge required to do research it divides into very many small subfields which often don't talk much to one another because almost all the people in one field can hardly understand a word of what's going on in the neighbouring one (believe me, maths is truly extreme regarding this). I think as a result of this, there is a lot of who-do-you-know going on in the background. "Oh, so xe's a student of so-and-so" type thing. I was very surprised at how chatty and social and we're-all-old-friends the last conference I went to was. It was a very nice, warm atmosphere, but it made me realise how hard it must be to enter if you're not part of the network already. And not only are women usually at a disadvantage in these types of situations, but it does mean that who you know, how well-connected you are, who your PhD supervisor is can play a large role in your future career... and I suspect women are less likely than men to move cross-country or possibly even internationally in order to have the Big Name Supervisor or be in *the* department for your subject area.
The moving, incidentally, is an issue later on in academia as well. From my experience growing up in an academic household, academia tends to take "you should have no life outside the job" to extremes. Academics are often expected to put in massive amounts of time (my father has been working twelve-hour days ever since I can remember), spend a lot of time at conferences, and, oh yes, go to where the jobs are. Which usually entails moving cross-country. A lot. Interesting fact that demonstrates both this and how academic class privilege can work - various people in academia have read the fact that at least one of my parents is an academic off my CV. It's not that I put down my parents' professions, it's that I was born in one place, went to primary school somewhere else and went to secondary school somewhere else again, these places were in two different countries and all either had or were very close to places with a large, well-known university. In fact, even if I leave off where I went to school (hopefully soon doable!) people have figured it out just from my hometown and birthplace.
...which means that an academic with a family needs to drag them with zem, which often means uprooting zer spouse on a regular basis and therefore messing up with zer spouse's chances at career advancement. Does this sound all too feasible for the average woman?
Or...
Thinking back, I've always wanted to be an academic. And in my undergraduate, I was so dead-set on doing a PhD you could not have changed my course with a crowbar. All the same, when the end of my undergraduate was nearing, I felt that I had to make a decision. Namely, a decision whether I wanted to have a family or do an academic career, because I did not think I could do both. Even if I'd been heterosexual, you can't very well do a part-time tenure track so I'd have had to find a husband who would've been willing to do most of the taking care of the kids. And with my orientation? Academic single mother ahoy! Forget about it.
I decided on career. But it was a DECISION, I felt as if I was knowingly giving up the chance at a family, and nothing's really happened since then to make me think having kids is any more feasible. I can easily, so easily, see other women being unwilling to give up the chance of kids.
Do we honestly think most men so much as THINK about this, let alone feel as if it's an either/or decision?
There's one last thing I want to bring up which isn't really directly leaky-pipeline related but is to do with differing status of women and men in mathematics.
In my experience, there's a sort of... allowable eccentricity, for male mathematicians. I sort of go :? at a lot of the maths-related stereotypes because they're... right, and wrong? At the same time? And the eccentric mathematician is one of them. Most mathematicians do NOT fit the stereotype of the completely eccentric wears same clothes for six months looks at his shoes when talking to you head not in the clouds but in orbit mathematician so popular in common culture. But... I find a lot of them have a certain degree of eccentricity. Might be a bit worse at picking up cues, a bit disliking of the standard social activity, a bit odd in their clothing choices, a bit blunt, might not actually be anything corresponding to your usual stereotypes but just a bit different. (Usually, in a gathering of people, I feel awkward and socially clumsy and out of place. In a gathering of mathematicians, I generally feel pretty normal.) And that's seen as okay, and as normal. And then you DO have some who are really, truly eccentric, and... the thing is that in the maths community that's not necessarily seen as bad, so much. Because we have all these famous mathematicians who incidentally were REALLY DAMN ODD. Perelman, to pick a recent one. Erdős. Also stories of mentally ill mathematicians - Cantor, Gödel. Popular culture buys into eccentricity as a sign of genius - mathematicians do too, to a certain extent, because there have been people like this and there are people like this (and there are also some regular mathematicians who are seriously eccentric) and this is part of mathematical culture, somehow. (Incidentally, I'm feeling more and more out of my depth here as a, what, just-beyond-first-year PhD student so any other maths people out there can feel free to chime in.)
The thing is? I DON'T think this is the case for women in maths. I think women in maths are held to... not *quite* the same standards as women elsewhere, but there are expectations in terms of social skills and stuff like that that clash with the eccentricity. Which is to say, I can wear jeans and T-shirts as much as I want and ignore make-up completely (in fact, I've heard a number of negative things about women-who-wear-makeup from fellow mathematicians, so I suspect that for women not appearing to care about your appearance can be a status boost to an extent), but if I start acting really odd? If I start interrupting people and talking about nothing but my subject and wearing flip-flops everywhere and whatnot? I don't think that would be seen as good or even neutral, even where it would be if I were male. I've never seen a female mathematician NEARLY as eccentric as a number of male ones I've met. I think female mathematicians are barred from being the Eccentric Genius and their degree of allowable eccentricity is also lower and more restricted in some ways than it is for the guys.
To tie this into the leaky pipeline I also suspect that as a result of this, female mathematicians might get landed with duties involving communication and social stuff that the male mathematicians aren't as good at or expect them to be better at, and those sorts of duties tend to actually be negatives when it comes to career advancement (which is all about research).
Okay! Meandering over! A lot of the last bit of this is speculation, the main reason for this post is that I really want to have people QUIT acting as if women being frightened away from maths in high school or wtfever is the only issue here. (Or even women being *gag* biologically worse at maths on average or your pet sexist unsupported theory, because how does THAT explain these findings?) And I feel as if a lot of people 'on my side' are brushing the statistics on the number of female maths undergrads under the carpet because it doesn't suit their argument and I want them to knock it off.
This is the part where people generally go "Oh, of course. The number of women in those subjects is abysmally low. It's due to the way women are socialised to dislike/not be good at/think they aren't any good at maths."
Except. Not quite.
A little story: when I first started my undergrad, I was fully expecting to be almost alone in a sea of guys. After all, everyone knows there are really few women in maths! Then I went to my first lecture and - women. Lots of women. All around me, women.
A few years later I actually went up to the poster that had pictures and names of everyone in our year, counted, and discovered that we were exactly one-third female. (I felt it was half and half, but there's a phenomenon saying that people tend to register something as being equal in gender at one-third women or so, and perceive true equality as a bias in favour. Either that or the guys skipped more lectures.) At the time, I figured we had unusually many women; pretty much every statistic I have found since, however, tells me we were unusually FEW. Apparently most maths undergraduate courses these days are something like 40% female.
Let me repeat that. Most maths undergraduate courses are almost half female.
There are two things to note about this: first, that it definitely ISN'T so when it comes to computer science and engineering. I think maths, stats and physics are relatively even when it comes to female undergraduates - physics maybe slightly higher - and then for the other subjects in this area the numbers go down. AFAIK (I knoooow I need to look up these stats :() numbers of women in engineering undergrad are well below 20%.
Secondly, the number of women mathematicians falls dramatically at higher levels. And I do mean dramatically. An article I read in the i-squared magazine quoted the number of female PhD students as 20% and the number of full professors as 4% - need to do some stats digging to see if this is true, but it sounds about right for my own experience. This is known as the leaky pipeline effect and is actually at work in AFAIK every academic discipline - the higher up the academic career ladder you go, the less women there are - but is unusually strong in maths.
Neither of these things can be explained away by "women are afraid of maths!" or "women are socialised against maths in high school!" or whatever. For the first, there is clearly another effect at play here! For the second, okay, we need to figure out the whys and whencefores of the leaky pipeline in general but also why it disproportionately effects mathematicians.
So I'd dearly love it if we stopped blaming the entirety of lack of women in all STEM fields on the maths issue because if we do we are missing shit. I mean, don't get me wrong, there are definitely ISSUES there but in my experience we focus on these, a lot, and the fact that HELLO MATHS UNDERGRADUATES ARE ALMOST HALF FEMALE, that the issue doesn't seem so much getting women into maths (although it IS getting women into maths-related subjects like computer science!) but keeping them there, gets brushed under the table.
Here come my thinky thoughts on wtf might be going on.
For the first, next to "women don't do maths" we have other stereotypes! Namely, the ones about practical subjects. In my experience, women are expected to be good at certain practical hands-on things - knitting, cooking, all that women's work stuff - but not others, especially stuff to do with mechanics and repairing shit. No, that's for the guys. And that, I think, is what really hits computer science and especially especially engineering. Not only is it a guy subject, it's a sort of practical, hands-on guy subject! Maths, on the other hand, is completely theoretical, and I suspect it might be considered more socially okay for women because of that. (Something I would DEARLY love to know is statistics for applied vs pure maths and theoretical vs experimental physics, but there's confounding factors there - e.g. at my university most of our applied PhD students are from India, Pakistan and China, whereas pure PhD students tend to be from Europe and the US. Since different countries have different gender balances for maths I think it might be hard to compare, and undergraduate degrees are usually just a plain BSc.)
I have no idea whether this might explain the phenomenon, but I think it's definitely something I've observed in my own life - things like my brother learning to repair his own bicycle but my dad always doing mine for me, that sort of thing. And might explain why I never *once* considered doing a degree or even advanced high school course in computer science, despite the fact that I was good at it and enjoyed it (my mother taught my brother and I to program in Perl when I was nine or thereabouts and I'd write little programs for fun, and later on in school when we had a programming class and the assignment was to make a pixel move across the screen I'd have six moving and be working on whether I could make them change patterns upon pressing a certain key before my classmates were even halfway done) and my mother was a fucking IT professional. /o\
As for the leaky pipeline and its disproportionate effect, this is where I am going to start tossing various things about. I'd feel guilty about this but tbh I feel as if a lot of the "why so few women?" stuff is tossing theories about and I at least have the benefit of first-hand observation. (For probably better researched theories, the London Mathematical Society has a statement out about the loss of women at higher levels which is interesting. And, ha, some of those statistics I was looking for.)
(Disclaimer! I am a pure mathematician. It's possible a lot of this doesn't apply to applied maths!)
Here's a thought: although maths usually gets lumped in with the sciences, it isn't really one. Or at least, the way you do it is completely different. The scientific method doesn't apply, for instance, it's deductive rather than inductive, etc. (I had the fun of taking a Philosophy of Science course in undergrad and discovering none of what it talked about applied to maths at all.) In particular, at least in the pure bits PhDs look VERY different. Apparently in most areas of science you have labs and labwork. You work in a team and as a PhD student you do a lot of the gruntwork and little of the actual coming up with experiments etc. (Correct me if I'm wrong about any of this, science people.)
Maths (the pure bits, at least!) isn't like that, at all. Your supervisor sets you a problem, or rather sets you a lot of background reading and then sets you a problem. Then you're off trying to figure it out - just you, the pen, the paper and the wastepaper basket, to paraphrase an old joke. Your supervisor is there for questions and to bounce ideas off and use zer greater experience to guide you away from paths of investigation that will probably be dead ends, but for the most part? It's your problem.
There are a few things that occur to me about this, when it comes to the leaky pipeline.
For one, it can be really really isolating. Supervisors can cut their PhD students loose pretty awfully. Or less dramatically, if a supervisor isn't giving you a lot of training and feedback and help, that's going to have an effect, both on the quality of your PhD and your willingness to go into academia afterwards. And since it's all essentially one-on-one, a supervisor can easily give zer female students less care and attention than the male ones, whether ze means to or not.
For another, female students may be less likely to pipe up when they feel as if they're being neglected, due to various female socialisation issues. When I was looking for places to do my PhD, I was actually specifically looking for a person where I'd not only feel comfortable asking questions of them but also where they'd come and poke ME to tell them what I was doing and where I was stuck on a regular basis - because I'd done two projects and was doing a third and all of them had gone wrong due to me not asking shit of the person whose job it was to make sure I was doing okay.
Finally, I wouldn't be surprised if some supervisors subconsciously assumed their female students were more likely to leave academia than the male (because, statistically, they are) and therefore more lucrative research problems would be wasted on them. And so gave their female students, however unintentionally, problems that they thought were a little less interesting and a little less likely to lead into further research questions than the ones they gave their male students. (No statistics on this one, also I have no idea how you'd even test that.) Or they may be less willing to trust them with riskier but potentially more interesting problems in favour of ones that are 'safe' but not that interesting.
Another issue is the social one. Mathematics is actually a relatively small subject area in some ways, and due to the level of specialised background knowledge required to do research it divides into very many small subfields which often don't talk much to one another because almost all the people in one field can hardly understand a word of what's going on in the neighbouring one (believe me, maths is truly extreme regarding this). I think as a result of this, there is a lot of who-do-you-know going on in the background. "Oh, so xe's a student of so-and-so" type thing. I was very surprised at how chatty and social and we're-all-old-friends the last conference I went to was. It was a very nice, warm atmosphere, but it made me realise how hard it must be to enter if you're not part of the network already. And not only are women usually at a disadvantage in these types of situations, but it does mean that who you know, how well-connected you are, who your PhD supervisor is can play a large role in your future career... and I suspect women are less likely than men to move cross-country or possibly even internationally in order to have the Big Name Supervisor or be in *the* department for your subject area.
The moving, incidentally, is an issue later on in academia as well. From my experience growing up in an academic household, academia tends to take "you should have no life outside the job" to extremes. Academics are often expected to put in massive amounts of time (my father has been working twelve-hour days ever since I can remember), spend a lot of time at conferences, and, oh yes, go to where the jobs are. Which usually entails moving cross-country. A lot. Interesting fact that demonstrates both this and how academic class privilege can work - various people in academia have read the fact that at least one of my parents is an academic off my CV. It's not that I put down my parents' professions, it's that I was born in one place, went to primary school somewhere else and went to secondary school somewhere else again, these places were in two different countries and all either had or were very close to places with a large, well-known university. In fact, even if I leave off where I went to school (hopefully soon doable!) people have figured it out just from my hometown and birthplace.
...which means that an academic with a family needs to drag them with zem, which often means uprooting zer spouse on a regular basis and therefore messing up with zer spouse's chances at career advancement. Does this sound all too feasible for the average woman?
Or...
Thinking back, I've always wanted to be an academic. And in my undergraduate, I was so dead-set on doing a PhD you could not have changed my course with a crowbar. All the same, when the end of my undergraduate was nearing, I felt that I had to make a decision. Namely, a decision whether I wanted to have a family or do an academic career, because I did not think I could do both. Even if I'd been heterosexual, you can't very well do a part-time tenure track so I'd have had to find a husband who would've been willing to do most of the taking care of the kids. And with my orientation? Academic single mother ahoy! Forget about it.
I decided on career. But it was a DECISION, I felt as if I was knowingly giving up the chance at a family, and nothing's really happened since then to make me think having kids is any more feasible. I can easily, so easily, see other women being unwilling to give up the chance of kids.
Do we honestly think most men so much as THINK about this, let alone feel as if it's an either/or decision?
There's one last thing I want to bring up which isn't really directly leaky-pipeline related but is to do with differing status of women and men in mathematics.
In my experience, there's a sort of... allowable eccentricity, for male mathematicians. I sort of go :? at a lot of the maths-related stereotypes because they're... right, and wrong? At the same time? And the eccentric mathematician is one of them. Most mathematicians do NOT fit the stereotype of the completely eccentric wears same clothes for six months looks at his shoes when talking to you head not in the clouds but in orbit mathematician so popular in common culture. But... I find a lot of them have a certain degree of eccentricity. Might be a bit worse at picking up cues, a bit disliking of the standard social activity, a bit odd in their clothing choices, a bit blunt, might not actually be anything corresponding to your usual stereotypes but just a bit different. (Usually, in a gathering of people, I feel awkward and socially clumsy and out of place. In a gathering of mathematicians, I generally feel pretty normal.) And that's seen as okay, and as normal. And then you DO have some who are really, truly eccentric, and... the thing is that in the maths community that's not necessarily seen as bad, so much. Because we have all these famous mathematicians who incidentally were REALLY DAMN ODD. Perelman, to pick a recent one. Erdős. Also stories of mentally ill mathematicians - Cantor, Gödel. Popular culture buys into eccentricity as a sign of genius - mathematicians do too, to a certain extent, because there have been people like this and there are people like this (and there are also some regular mathematicians who are seriously eccentric) and this is part of mathematical culture, somehow. (Incidentally, I'm feeling more and more out of my depth here as a, what, just-beyond-first-year PhD student so any other maths people out there can feel free to chime in.)
The thing is? I DON'T think this is the case for women in maths. I think women in maths are held to... not *quite* the same standards as women elsewhere, but there are expectations in terms of social skills and stuff like that that clash with the eccentricity. Which is to say, I can wear jeans and T-shirts as much as I want and ignore make-up completely (in fact, I've heard a number of negative things about women-who-wear-makeup from fellow mathematicians, so I suspect that for women not appearing to care about your appearance can be a status boost to an extent), but if I start acting really odd? If I start interrupting people and talking about nothing but my subject and wearing flip-flops everywhere and whatnot? I don't think that would be seen as good or even neutral, even where it would be if I were male. I've never seen a female mathematician NEARLY as eccentric as a number of male ones I've met. I think female mathematicians are barred from being the Eccentric Genius and their degree of allowable eccentricity is also lower and more restricted in some ways than it is for the guys.
To tie this into the leaky pipeline I also suspect that as a result of this, female mathematicians might get landed with duties involving communication and social stuff that the male mathematicians aren't as good at or expect them to be better at, and those sorts of duties tend to actually be negatives when it comes to career advancement (which is all about research).
Okay! Meandering over! A lot of the last bit of this is speculation, the main reason for this post is that I really want to have people QUIT acting as if women being frightened away from maths in high school or wtfever is the only issue here. (Or even women being *gag* biologically worse at maths on average or your pet sexist unsupported theory, because how does THAT explain these findings?) And I feel as if a lot of people 'on my side' are brushing the statistics on the number of female maths undergrads under the carpet because it doesn't suit their argument and I want them to knock it off.