It wasn't that long ago that women made up close to 0% of graduates in law and medicine. There were lots of scientific-sounding essentialist arguments made about gender to explain that. Nowadays close to 50% of graduates in medicine and law are women. Remember than when you hear people spouting off about why women aren't suited to programming.
Translation: people who disagreed with me were incorrect on a different topic. Therefore, everyone who disagrees with me on this peripherally related topic is also likely to be incorrect.
Peripherally related is really stretching it. The way I understood chrismealy's argument is this:
Biological essentialist arguments were previously used to explain the disparity of men and women in law and medicine. Now that we know that those arguments were critically flawed and did not reflect reality, we should be very hesitant to support similar arguments being used to explain the lack of women in STEM disciplines.
Something that frustrates me about the field of computer science is that there are a lot of jerks who think that just because they've "mastered" some programming language or know some obscure unix commands, they are gods and you are nothing.
I don't think this is at all unique to CS, or even to technical fields: it's the Dunning-Kruger Effect all over again.
Regarding women in CS, it would go a long way in helping to simply treat them as people. As an IRC op, I find it sickening how often, particularly in online programming communities, people hear the word "female" and start acting like utter idiots. This results in both unwanted negative and positive attention: but furthermore, continually reinforces the idea that a female is some sort of odd, unusual creature deserving of special study. This doesn't merely bother the victims -- it also helps convince them that they're the odd one out, and that they don't belong.
This doesn't have to be blatant sexism -- it can simply be comments like "oh, you're a girl?!" when someone uses a female pronoun. I've seen this happen dozens times in a single channel in response to a single person over the course of a few months.
Some of you may be thinking that this is just a problem for 4chan and Reddit. But the interesting thing that I've noticed is that the incidence of this problem doesn't seem to be reduced much with more professional communities, or in more mature environments.
Rather, the only statistic that seems to be consistently correlated with these kinds of problems is the perceived male/female ratio in a community. The instance the perception gets created that "this community is almost entirely male", problems begin to occur. This seems to be a constant everywhere from programming communities to writing communities to roleplaying communities to fandoms.
If you're male, you've probably been responsible for some of this at some point. Maybe it was just one comment. I mean, what harm could one comment do, right? But a hundred people making "just one comment" stops being a joke quickly -- and instead makes half of the smart aspiring programmers out there feel unwelcome. This is of course self-reinforcing: by making women feel unwelcome, the "almost entirely male" perception is perpetuated.
(An interesting question might be why the reverse problem rarely exists in communities perceived to be almost entirely female.)
Its not the Dunning-Kruger effect, I would call it the osx effect (or alternatively the nosql or ruby effect). It goes something like this: if you can do something in an esoteric way that gets your more respect (both real and imagined) from people in your community, you will do it and lord it over others who do not have this particular knowledge or skill. The respect you get tends to be inversely proportional to the actual value being provided. (ie osx barely gets you any respect nowadays becuase it provides true value for programmers, using nosql at your startup gives you lots of respect even though a sql/mc combo would probably take much less time and effort to actual get running and fine-tuned). This effect generally appears in the following form: i use osx so therefore I am better than Windows users. I know sed so I am better than cli noobs.
The opposite of this is the windows effect (aka the php effect), where you get disrepected for using a particular technology regardless of how talented or productive you actually are with it.
> The opposite of this is the windows effect (aka the php effect), where you get disrepected for using a particular technology regardless of how talented or productive you actually are with it.
Just wanted to add my related $0.02 rant: Everybody bashes PHP for its associative arrays, but when Python finally implemented the same data-structure (ordered dicts I think they are called) a long, long time after PHP first got them, then everybody started congratulated themselves and saying "what a nice idea the Python guys implemented!".
Sorry to go down that road, but the reason people rag on PHP's associative arrays, is that its the only array that PHP has.
A hashmap with mixed keys is hardly the most practical choice for an all purpose primitive data structure.
Python already has proper arrays where you might have some chance of fast sequential access. The ordered dict or whatever is just another tool in the box.
Hmmmm. I feel I may be proving the OPs entire point rather too well with this line of enquiry. I don't think I've ever trolled myself before. How embarrassing.
This kind of thing is trickier in places like HN though, because a statement is only showing off to people who already have the knowledge. There are millions(?) of people reading here, so statistically most of them probably wont know some obscure piece of knowledge about the problems of the key hashing function under random access in numerical PHP arrays. It may be showing off to the couple of people actively engaged in the debate, but to everyone else its educational.
That's most of the reason I come here, to be honest. Crumbs of insight casually discarded during the debates of much cleverer people.
Despite my worries that this is turning into offtopic, I'll note that I haven't noticed that this particular PHP feature gets bashed. I've seen the opposite, people remarking that one of the things that is pleasant in PHP is working with associative arrays.
Actually, I've noticed that there's a lot of arrogant assholes in the collegiate Computer Science atmosphere, maybe more so than in other fields.
But it has nothing to do with the gender issue. These "jerks" will be jerks to anyone who isn't at their level, male or female - I even caught myself sounding a little condescending today during a group study session.
I think men who are non-jerks are still socialized in how to interact with jerks. We understand that a certain number of guys are jerks and we have to deal with them anyway, so we learn how to shrug off the behavior or respond in kind, and we have to learn how to tell when someone crosses the line from macho bluster to being actually dangerous.
Women, I think (someone tell me if I’m full of it here) are brought up differently—they are trained to see guys who act this way as threats, and to avoid them, or at least keep their guard up around them. (There are obviously women who act rudely to other women, but it tends to be a different kind of rudeness.)
This is something that can be true, and I feel that a lot of guys don't really understand. I had difficulty once with a coworker who gave me a birthday present even though he shouldn't have known my birthday, found out my address, asked me out every month even though I said no and gave no indication I was interested, and often times invaded my physical space. While this might seem like innocent social awkwardness (often attributed to some subset of cs guys), it certainly doesn't seem that way when it happens to you.
When guys don't respect social boundaries it can be frightening. Don't be so quick to assume this fear is trained- I didn't start to react with fear until a handful of upsetting experiences. It can be justified fear.
I would use a stronger term than “jerk” to describe that kind of behavior; I had in mind the sort of assholish behavior that some men exhibit among other men (e.g., the condescension described a few comments upthread).
A man who treats women as if their main purpose in life is to be his potential girlfriends has a different problem, and I think most men who act like this are not so much socially awkward as socially mistrained.
This is just another anecdotal datapoint, but I've found substantially less arrogance in my CS program than in some of the other bigger majors here (IR, Govt, Econ). I think it helps that some of the best and brightest students spend a vast majority of their time answering questions, debugging code, and filling in others with that obscure unix knowledge.
Nevertheless, we have very very few female majors.
Out of curiosity, what CS program is that? From the major majors you listed, it sounds as thought it might be where I'm headed next year, and I'd love to hear what a current student thinks of it.
It's a well-known mid-atlantic school in the US. If that still fits the profile, feel free to email me(see my profile) and I'd be happy to discuss things more with you.
People who know a little bit ("mastered a language, or know some unix commands") are more likely to be confident in their abilities (and accordingly, arrogant about them) than someone who knows a great deal. That's the Dunning-Kruger Effect.
Dunning-Kruger doesn't imply arrogance, just a mistaken level of confidence in one's own abilities, based on a failure to appreciate the magnitude of one's incompetence.
Consider a person who plays, for example, a multiplayer game against the same three real life friends on a LAN, and concludes from his performance -- domination of his friends -- that he is an astounding player. Given his knowledge about the game and the skill of other players, it makes sense for him to make this conclusion. However, were he to play online, he would realize that his skills are paltry compared to the those of the best online gamers.
Here, Dunning-Kruger is manifested in this gamer's overconfidence based on lack of knowledge, and lack of meta-knowledge (he doesn't know the importance of what he doesn't know). It's based on a genuine mistake in his assessment of his skills, not necessarily arrogance.
The Dunning-Kruger effect doesn't necessarily imply a narcissistic personality (or any other personality traits); overconfidence in ability is something we are all vulnerable to.
"""The unskilled therefore suffer from illusory superiority, rating their ability as above average, much higher than it actually is, while the highly skilled underrate their own abilities, suffering from illusory inferiority."""
I think if you "master a language", you don't get lumped in with those who know "a little bit". I think you're just talking about arrogant people in general who think that because they know X, they can berate those who don't know X.
There's some good general observations in here. I remember tutoring a student who was really good (I'm sure that by now he's a better programmer than me) but who felt inadequate (enough so to get a tutor) because other people in the class were bragging about how they did certain tasks in just a few minutes while in reality spent hours on it... or just turned in crap. And it was that same kind of macho sentiment that the author talks about... except it intimidates a lot of guys as well.
Also, most of the people who are jerks about it, don't know just how much they don't know (because that one obscure Linux command is all they need to master to be CS gods, right) and they turn into that guy you fire because his code has no sense of industry best practices (seriously, why unit test like a mere mortal if you're a CS god who knows obscure Linux commands.... FIRED).
So take heart. You're probably a better programmer than you give yourself credit for (whether you're a guy or a girl), because everybody else is a worse programmer than they tell others they are.
"One of the challenges for me while I was at Google was to speak up when I didn't understand something, as I often assumed it was common technical knowledge and that people would pass judgment."
Of course, it is generally good advice in most workplaces to speak up and ask questions to be sure what the colleagues are talking about. But, yes, male or female, many new employees lose out on learning opportunities by not asking questions when they know they don't understand, and perhaps even more from not asking questions when they suppose they do understand. What I've learned by becoming a lot older than the author of the interesting submitted blog post is that I still have plenty to learn, and sometimes I can make a good impression by being willing to ask a possibly dumb question[1] to be sure I know what my colleagues have in mind.
[1] The best teacher I ever had was a teacher who said, of course, "The only dumb question is the question you don't ask," and he really meant it. He tried his utmost to make sure any student who asked him a question got a thoughtful response that implicitly acknowledged that it's important for learners to check and recheck their understanding.
I've been surprised how often a software engineer will describe something technical and arcane, and everyone in the room assumes it must be comprehensible to everyone else. After a few such incidents, I decided it was my responsibility to ask questions until I understood what was being described to my satisfaction, and to hell with trying to pretend I'm smarter than I am.
The result was surprising: I appeared smarter. When you say, "Excuse me, I don't understand this," and everyone else is thinking that, too, and especially if it turns out to be a bad presentation or a fundamentally incoherent concept . . . well, you look perceptive.
And if you make sure you understand things? You look pretty darn smart later on.
>The result was surprising: I appeared smarter. When you say, "Excuse me, I don't understand this," and everyone else is thinking that, too, and especially if it turns out to be a bad presentation or a fundamentally incoherent concept . . . well, you look perceptive.
This is massively massively important point. It goes beyond looking perceptive, it actually makes you look supremely self confident and powerful within a group.
Whether the point is incoherent or not doesn't really matter, when you ask the question, suddenly you're the only one in the room thats not afraid of the conversation, and everyone instantly knows it.
Yes, you look smarter because it's as if you're saying "Hey, I'm good at this stuff, and I still don't understand it, so there is something that is not obvious here". Most people don't ask questions because they aren't confident enough in their own capabilities. They feel that the thing they don't understand should have been obvious, and don't want to admit thet they don't know. It's another example of the Imposter Syndrome...
On the other hand, I have often noticed that once one person in a group starts asking questions, it opens the flood gates and everyone else starts to get involved. I have a theory that once one person asks a question, it signals everyone else in the group about the skill level of at least one other person in the group, and if they can understand the question and the answer, they now feel that they are at the same level as everyone else - they no longer feel like an imposter.
I'm pretty sure this will get downvoted and deleted in 2 seconds, but i'm honest to god not trying to troll, just state some of my personal experiences.
I'm currently working at a pretty small company with women coworkers, and most of them are at least as smart as some of the men i've worked with and one is way smarter than most people i've met. Having said that, in my experience both in college and at work (been working about 6 years now), on average the girls were less capable and smart at CS and programming. From my experience it seemed A LOT like they just didn't commit at nearly the same level as the guys. For most men who are programmers it's almost a way of life - I have never met a woman for whom the same could be said. The one woman I know who is vastly better at her job than anyone I've met, even she is simply a extremely dedicated professional for whom it is just a job, not truly a passion.
I'm not saying this means guys get the right to be assholes, I'm just saying I don't think this is entirely because there are so few girls or because of stereotypes, i think there is more to it than that, there is a statistically discernible difference in skill - from my limited personal experience. Please proceed with the downvote into oblivion :).
People who make computers their life are doing life wrong.
There's no doubt that computers serve as a solitary outlet for many lonely, ostracized, and anxious people. The fact that computers become the only good thing in their life is not a positive and it's usually not a choice. It's something thrust on them by their psychological circumstance.
The real tragedy is that there is an entire cohort of people doing computer programming because they suffer from anxiety disorders and programming is capable of affording them a solitary way of making money as well as giving them something to escape into.
Anyway these people are not really helping themselves or the industry. They do sub-standard work, they suffer from depression and anxiety, they find collaboration difficult and uncomfortable and they accept being underpaid due to low self-esteem. The low self-esteem causes them to lash out against others when they have a chance, like insulting people who don't know something.
It would be better for everyone if these people got therapy and came to enjoy the other things that life can provide. Warm and fulfilling relationships being the big one.
Life is about a lot more than just working. If this applies to surgeons and Presidents and astronauts, it should apply to computer scientists as well. Never judge someone negatively for having a life.
You could say also that 'people who make mathematics/science/art their life are doing life wrong'. Almost all the great mathematicians, physicists, etc are doing it wrong? Maybe.
At least this is how they get into the field. Programming was the life for me until I met my wife and had 2 children. With a family, things are different. But our topic is mostly about how people get into the field. True and early passion for programming, and social-acceptance ignorance is more rare (although existing) amongst women. In my opinion this is the real difference between women and men and not that women are not intelligent enough.
There is a masochist bent to many CS programs, where students are basically told that they need to "sink or swim", then get pushed into the water. I saw more than a few people just hit a wall and give up.
I think that in general, women tend to thrive in different circumstances than men. A big part of the way that CS is presented to students is that it is a "lone wolf" endeavor, and students are expected to just figure stuff out. That appeals to the hardcore, anti-social nerds and feeds the male ego. It's not appealing to women or most men.
The IT team that I run now is probably 60% female. They rock, and only one of them has a degree in CS. It's a real shame that in a field where top companies claim to be short on talent, keyword filtering is virtually eliminating half of the potential workforce.
One of the great things about Stanford CS program is that it tries very hard not to have the "sink or swim" mentality, especially for the first 2 classes. For those classes, you have a large lecture, but then you have 10-person sections once a week, 6 hours a night where you can come and have section leaders help you with assignments, and "interactive grading" with your section leader where you spend 10-15 min going over your assignment and hearing what you did well and where you can improve.
This has been hugely successful; on the order of half of undergrads take the first class, and gender balance is pretty good in them.
Strange as it sounds, lots of incoming freshman have never had to figure out anything before. Their "education" has consisted of nothing but regurgitating standardized lessons for standardized tests.
These students need some time to learn how to think. That doesn't mean they can't, just that they don't know how yet.
That is incredibly sad. Surely if you care about CS, or medicine or law for that matter, you would find out as much as possible before-hand?
Figuring stuff out should be the number one priority of education. I think it should be a policy. Even if that wouldn't change much by itself, beauracracies being what they are, a few more people would take it to heart.
I'm not speaking of academics so much as tinkering, for which computers are the ultimate toy.
When thinking about the geek or IT fields, I expect that people should like figuring stuff out, the typical nerd/engineer mindset. That's not everybody in CS programs I know, but my inner geek wants it to be.
This is only part of the story. As far as I can see it, the biggest difference between men and women is that there are less women who don't give a shit about 'social acceptance'.
The OP's post is mostly about schools, social issues, etc...
Since I have seen a computer as a 12 year old guy, I did not give a shit about other people's opinion on my programming skills, I did not learn programming in school, etc... I just needed to program. My mind did not rest until I did not give it a programming poblem to solve. When I created a little game, I felt that I created a new world from scratch.
The real question is why this kind of enthusiasm and this kind of social-proof ignorance is more rare (although existing) amongst women.
It was a thought provoking blog. When I submitted to HN, I expected it would end up like most of my submissions and simply add a point of karma to someone who had already posted the same thing ages ago.
However, it appears that the blog author re-arranged how she recorded her blog entries, giving this one a different address.
Interesting that the conversation this time feels quite a bit different than the last time.
You wouldn't get the impression looking at it now, but it actually was pretty similar. HilbertSpace's post came a bit later, and overwhelmed the entire conversation. You can see that some of the replies to it are from as much as 10 days later.
I'm going to get down voted for this but the main reason CS doesn't attract women is that it is not a regulated profession. Competition is high and you have to stay on top of your game regardless of your diploma. It's the same reason there are few women who are entrepreneurs.
Women tend to be more attracted to stable jobs with a well defined path to success such as medicine or teaching.
When I studied, there still were computer rooms. Not sure if they still exist today. I had to ask for help in the computer room a couple of times. Like when I accidentally launched Vim and did not know how to escape from it.
A few months ago Dr. Christine Alvarado gave a GREAT talk at Google about how the CS profs at Harvey Mudd restructured their introductory CS course to help attract women. I'm not usually that interested in this issue, but found the talk really interesting and engaging. Highly recommended.
From observing my own (non USA) CS class, some 20 years ago, (where we had about 30% female), and having visited the campus on occasion since then, I think I see the following in action.
Firstly - there aren't that many courses at college that start with students who are well versed in their chosen area of study. The school of medicine for example doesn't take in practicing doctors - architecture doesn't take practicing architects and so on.
But often a large portion of a computer science class are already proficient programmers, and a sizable number have probably already made money programming in one form or another. In other words they choose this major not because they want to learn, but because they're confident they can _already_ do the work.[ Aside: there is another group exactly like this - the sportsmen - which makes the jocks and the nerds pretty much the same thing - but that's a post for another day.]
So I hypothesize there's a direct correlation between the numbers of those doing programming at school (formally or informally), and those doing it at college level. During school years male and female behaviour is vastly different (programming is very often an outlet for "loners" with poor social skills - which let's be honest are mostly guys.)
Here's the thing though. I grew up in a non-US school where the labeling of people was not encouraged. Also it was easier to foster your own self-identity without belittling the identity of others. Sure we had beefy physical types, and scrawny nerdy types, but at the same time the school went out of its way to validate each person at whatever they did well. Excellence, of any sort, was recognized. We had sporty girls, and nerdy girls and super-bright girls, and (to be fair) some dumb girls - but "the computer club" wasn't some sort of social dead-end.
So to college - where the class was split about half-half of those that had programmed before and those that hadn't. And as I say, about 30% female. So the ones-that-could would spend a lot of time teaching and helping the ones that couldn't. If some of those I helped happened to be female, well that wasn't my fault was it? And if some of the ones doing the helping were bright diligent girls, who helped me apply my somewhat erratic studying habits to actually passing exams - well, it's a win-win.
Of course we had some jerks - and sad to say the raw "jerk quotient" seems to be climbing. But they're jerks to everyone really, and frankly you'll encounter lots of jerks, from both sexes, all the way through life.
My advice to ladies doing CS - find the nice, smart, guys in the class. They're there. Sure they may need some social help - but you can do that. And I bet if you make a little effort you'll find they're _really_ keen for your friendship.
To the guys (and I'm assuming that's most of you reading this post) apply just a small amount of that alleged brain-power you have to consider how you would feel in a class that's 90% female. Now take just a few moments to go be nice to someone. Invite them into your study group. Offer some assistance when they're struggling. Just be _friendly_. They may not yet know the difference between a++ and ++a, but they will. In the meantime you'd be amazed at what they can do for you.
It's a smart man who thinks just a few years ahead of everyone else.
I believe that, as you say, the core of the issue is that some students are well versed at the start, and some have no knowledge whatsoever. As a student that was well versed, I can say with absolute certainty that I found it very frustrating to be paying thousands to learn the difference between = and ==, and the function of if statements. If you are well versed, at least the first two years of computer science curriculum are painfully easy, and the only interesting thing to do is see how quickly you can complete assignments.
Perhaps Computer Science should be treated differently because of this issue. If every student took a 1-credit skills evaluation and placement course, they could be assessed and properly placed so the students that already have some knowledge can move ahead and start learning right away, while the remaining students can take the introductory courses at an appropriate pace.
...assessed and properly placed so the students that already have some knowledge can move ahead and start learning right away, while the remaining students can take the introductory courses at an appropriate pace.
And enrollment would drop because half the class skips years 1 and 2. Not to mention costs would rise, because instead of paying for a 250 student lecture on "=" vs "==", you'd be paying for a 50 student lecture on compilers. Good for the students, but bad for the college. Probably will never happen, except at a few very good schools.
We had the same issue with the first couple semesters - there were a bunch of us that could breeze through the assignments in a couple hours, which left us most of two weeks till the next one.
To combat this we assigned ourselves numerous programming challenges - mostly code for code's sake, but effectively sponging techniques off each other. We effectively competed with each other to see who could do what better.
We also went overboard with the assignments a bit - we'd take the basic assignment and "add features" to out-do each other. Often we'd have to hand in a "normal" solution though for marking - the marker didn't appreciate a 20 page printout for a problem that should be half a page of code.
And of course there was also some time taken to help others - to effectively give them "extra assignments" to get their code up to scratch.
We had access to equipment which we certainly couldn't afford ourselves (although it's laughingly primitive by today's standards) so there was never any shortage of creative ideas.
To make things a little more interesting we also signed up for other classes in the science faculty outside of "only" computer science. I did classes in Oceanography and Astronomy, Physics, Applied Maths and so on. The courses were a lot of fun, but the exams were a pain.
I went to Drexel University. There was a required introductory course, with a little bit of JavaScript programming, that weeded out many people that would immediately change majors. After that, anyone could skip the 2-3 C++ programming courses by taking placement exams, as you suggest.
Did they charge for the placement exams? The school I attended had credit by exam, but it cost the same, or nearly the same, as actually taking the course.
What was the placement exam like? I think to really be fair an exam like that shouldn't be so much a C++ exam, as it should be a programming exam. In other words, C++ syntax isn't as important as understanding what polymorphism is. It should be acceptable to answer programming questions on an exam like that in pseudo code.
The tests were free, if you passed you simply had permission to take the higher level courses without those prerequisites. It's been a long time so I might be mis-remembering, but I remember there being some short answer questions, I think C++ was used in questions where you were asked to look at code, but pseudocode was acceptable for answering the programming questions.
Part of the problem as I see it is that male dominance in CS extends all the way to the top. At a particular university I have in mind (but will not name for legal reasons) nearly all of the lecturers were male. Most women who were staff in the department were secretarial (in fact, all of the administration staff were female), and the male staff were every bit as bad, if not worse, than your stereotypical male IRC user. We're talking about ignorance and disrespect running so deep that the way some girls (undergrad and postgrad) have been treated by staff is tantamount to sexual harassment, so I find it... amusing when we're talking about the behaviour of students, or social fixes, when to me it feels like the whole damn system is stuck in the 50s.
I might be oblivious (I'm male), but I've found that mostly among fairly old CS professors, some of whom are indeed not very progressive on gender issues, and not even very good at hiding it. I haven't noticed a lot of trouble from male professors in their 20s, 30s or 40s, though. Most of them seem genuinely concerned that the percentage of students in their courses is so low, and interested in whether there's anything they can do to improve the environment. In many cases the gender balance is noticeably lower than when they were CS students, since female enrollments have dropped considerably over the past 20 years.
I've been pretty lucky with my uni, for the most part the profs and students are all very level-headed and don't care about gender much. There are maybe two somewhat-douchey but bearable guys and one prof who's always saying weird, stereotypical things about women during lectures, but mostly the guys just want to talk shop. Between 2nd and 3rd year I switched from wearing mostly jeans & sneakers to wearing mostly skirts & dresses & pretty flats and the way I was treated didn't change noticeably. The most atmospheric douchiness was during 1st year. For several months there was a man sexually assaulting dozens of women on and around campus, and I remember some idiots who kept talking about how it was sooo sexist that female students were being offered free rape whistles and self-defense classes as a result, and how the guy just had needs so you couldn't blame him, and how rape was just so much fun. It wasn't straight-up harassment, I guess, because they weren't talking to me, but being alone with them late at night in the computer labs made me feel very unsafe. It was a frequent and rather unavoidable occurrence and I can see how a woman considering CS as one of multiple possible majors would have been seriously put off. I was pretty put off and CS was the only major I'd even considered since I was 13; that year I worried a lot about what I would do if the guys in the working world were like that, too. (Never saw them in the second year labs, guess they spent too much time chatting and not enough time doing their homework, haha.)
Even with a decent school environment it's nice to have some female profs around as role models. I have male role models too, of course, but it's so awesome to get to see hard proof that hey, that awesome woman I want to grow up to be is not only attainable but standing RIGHT THERE in front of the classroom. Two in particular seem so fulfilled, well-rounded, and passionate about their fields that it makes me know that I made the right choice. I think if all my lecturers were male, or if the few female profs I had were the grudgingly-teach-from-the-textbook variety (of which I've had in both genders), I would feel a lot less secure. I could honestly see my alternate-universe self switching over to math or econ if the comp sci department at alternate-universe uni was as bad as the school you're talking about. Comp sci is great but it's not worth crushing your soul for, especially when there are so many other fields these days where good programming skills can give you an edge.
"He once told me that even though the females are fairly quiet, and the boys in the class showed off a lot, when it came down to projects and exams, the female average was often higher."
I think that this is actually one of the reasons that so few girls are in Computer Science. My girlfriend is a CS major and she very often feels inadequate compared to all of the other super exceptional girls in the program. It isn't because she's inadequate either. Her grades are fine. The issue comes when every one of the other girls is at the top of the class. She's not being judged against the whole class like the guys are, but she's being judged against just the super exceptional girls.
I can only imagine that lots of girls feel this way in their first intro class and give up.
One of my issues is that, although my grades on projects, exams, assignments, etc., are well above average, I tend to zoom in on the mistakes that I make and beat myself up for them. Weird kludges, or silly bugs, etc. So I don't brag about projects that have flaws, which is all of the nontrivial/interesting ones, because I feel like I'll get judged just based on those, and anything I did right won't count. I think this is partly a gender thing and partly a parenting thing. (HS Me: Look, I got 98% in Japanese! Dad: Where's the other 2%!? Me: sadface.)
Guys brag about their projects a lot, and what they say sounds really good, so I often assume their projects are better than mine and that they're better programmers. (Even though I probably got a better grade, and when I actually get a chance to dip down and see their source code it's usually not so hot either, at least by the standards I measure myself by.) End result, even though I might be 'exceptional' grades-wise, I have this mentality that I'm just good at school, not good at programming (compared to peers). Extremely stupid, but if you really really don't want to believe in yourself you can always rationalize it somehow. Your GF is above-average just for being there, since most female students (or potential female students) get scared off, and she's still getting it done.
"Victory belongs to the most persevering." - Napoleon Bonaparte
The article mentioned the HS experience, where other girls gave up on the CS course. That was telling to me - how many people drop Anything in HS? I never knew anyone who dropped a course - unless for severe illness etc.
Is it a female thing, to drop a course if it gets hard? That may be the largest factor in the statistics by far.
I dropped a few courses (I'm a guy) in my GCSE's and later an A-level once I came to the realization that there was no value in taking more than the necessary number of courses. Why get 7 A's and B's plus 3 C's and D's when you can simply get 7 A's and B's? More free time, fewer exams, less homework, no downsides...pure win all around.
I think this says something sad about the governing college. I don't know what it's like for a "general CS" student at my University, but I'm fortunate to be part of a community that is VERY inter and intrasupportive of each other, especially when someone is behind or deficient in knowledge. This really doesn't seem to suffer at all across gender lines as well.
But then again, this community recruits based on business skills and computer programming skills. I don't know that sexism and arrogance would come off well in an interview.
http://www.census.gov/compendia/statab/2011/tables/11s0300.p...