For instance, there is plenty of evidence women are less likely to be very good (or very bad) at math: http://www.sciencemag.org/content/321/5888/494.summary http://www.ncbi.nlm.nih.gov/pubmed/21038941

There is also evidence women are more risk averse, making them less likely to work in startups: http://www.pnas.org/content/106/36/15268.full

There is evidence women avoid computing specifically because they are intolerant of geeks: http://web.archive.org/web/20100106021904/http://scicom.ucsc... (The original article was taken down)

This report also pushes the intolerance hypothesis, as well as women being less dedicated, being fearful of text interfaces, and preferring more collectivist environments: http://web.archive.org/web/20091007234852/http://opensource.... (Please go read the report before criticizing me for citing it. It's written by a woman who is allegedly a feminist.)

It's safe to say that none of this is ironclad proof that the only cause of women not being present in computing is natural causes. But the author's claims that "there's no such evidence" is utterly wrong. There is evidence, he just ignores it.

Lastly, the author completely ignores the fact that computing is very diverse. We have whites, all different types of asians, israelis and a smattering of euros (and of all sexual orientations). Our diversity may not be statistically identical to the general US population, but it's nonsensical to claim a lack of diversity in computing.

I can't see any reason why being genuinely good at math is more important to effective programming than being good at chemistry (layers and sequence), biology (complex systems) or languages (building meaning from abstract or incomplete signals). You do not need to be good at late high-school math to understand Big-O.

I suspect people who are good at or passionate about math are far more likely to stroke their ego with premature optimisation.

Also, there's a vast difference between aptitude for maths, and performing well in maths tests. Attention to detail, regard for the education system, interest in the topic all play a part.

    > There is evidence women avoid computing because
    > they are intolerant of geeks

I would strongly suspect that just as math and CS ability are correlated, so would math and chemistry. Biology less so, since biology is mostly just memorizing (at least for the first 3-4 years of college).

Anyway, I cited math data mainly because it's so widely studied. I get so many downvotes when I discuss this topic with data that I'm not even going to attempt to write about stuff I have no readily available data on.

I see a different pattern that goes in mostly the reverse direction. [...] young nerds [...] are therefore undesirable to be around.

Um, that's not the reverse direction. That's the same direction as what I said: women avoid computing because they are intolerant of geeks.

Note that young nerds also often act "bizarrely" towards men and each other. I certainly did before I learned to pass. But the claim of the author of the studies I cited is that women are less tolerant of "bizarre" behavior.

    > I get so many downvotes when I discuss this topic with data

I accept your correction in that we're making the same point about young men. What I should have said is that the way you phrase it makes it seem like it's women that initiate the pattern. I don't think that's so. Although perhaps you could go back further and ask, what made the men neurotic in the first place, was it cruel rejection or bullying by prom queens - no idea. I just found the phrasing backwards.

Fair enough. I thought the link was obvious enough, but you are keeping me honest. I have no data proving this link myself.

However, I just messaged a friend of mine who occasionally dabbles in math-ed stuff. He recommends these (paywalled) papers as starting points:

http://dl.acm.org/citation.cfm?id=1500963&dl=ACM&col...

http://dl.acm.org/citation.cfm?id=1047480

And an older paper "Identification of Computer Programming Aptitude", by Alspaugh (one of the early works).

I'll asked him to give me copies next time I see him in person (I'm no longer an academic, so I can't access university libraries).

  I would strongly suspect that 
  just as math and CS ability are correlated, 
  so would math and chemistry. 

Anecdotal evidence here: I'm pretty good at math, but failed several chemistry exams in high scool. I just didn't "get" it. Maybe I would be able now, because I'm better at thinking abstractly and systematically than when I was 16, but that would be because I'm a programmer today. For me the correlation definitely doesn't hold. Same goes (to a lesser extent) for biology for me.

Yet, somehow I manage to keep the perspective that technology isn't populated by sexist assholes - sure the ones that are there stand out and make up a lot of the things that come to mind when I'm stressed. But really it's a small percentage [even though that small percentage can really get to you].

Try to keep the same perspective about women, the vast majority of us aren't "bitches".

As lots of us on HN trend towards the geekier side of life, social cues sometimes don't come as naturally to us as they do to many others - it takes time and effort to not be socially awkward, but it's 100% worth it.

The next time something like that happens, try reviewing exactly what happened, how things might have been misinterpreted, and how you can learn & grow from it.

Also keep in mind that there are lots of people who _are_ jerks, but gender has nothing to do with it.

Most importantly, don't let it bias you, and keep your chin up!

The point I was trying to make was that a bunch of negative experiences with women when you are young leaves very deep scars and these scars manifest themselves as unconscious biases so much that whenever you see a beautiful woman (who will most probably reject you if you tried!) you are automatically biased against her.My brain does this to me so often even though I think I am logical than most other people.

Now I will make another point that is also controversial.(someone needs to say these things!)

See when you grow your skills interacting with women a point comes when the things you cant change matter much more than the things you can, so much that the situation seems pretty helpless.

So at that point even if I bought really expensive clothes and bought myself a mercedes or say became a really smooth talker, how women respond to me would only marginally improve as compared to the non-possible changes such as changing my race to become white...changing my height to become six feet tall etc.

So no matter what I do a six feet tall white guy or a good looking muscular black guy will always have significantly higher odds of attracting a given woman over me.

The more you approach the limit the more you begin to see this brutal truth of the dating world.The things that matter most are things you are born with and this sometimes causes you develop significantly negative feelings towards women as a whole.

Wrong. You're suffering from confirmation bias.

Guys obsess about money and looks to prove their worth to females, but most of that shit doesn't matter, actually. (It's held against you if you're severely deficient, of course.)

What does? A strong sense of self-identity, confidence, the ability to be both a risk-taker and a stable provider, and, more generally, being an interesting person. These are much harder to work on, so you don't hear as much talk about them. You need to be able to enjoy your life without a girlfriend. You need to have something you love to talk about, and, preferably something you can be good at.

Suck up your pride, ask for help, and start working toward taking responsibility for more of your life.

Everyone wants to be seen as desirable, but it is not an entire gender's fault that the specific women you've approached don't respond to your advances. It may seem counter-intuitive, but the best way I've found to get dates is to treat women like interesting people instead of potential dates, while putting effort into making myself conventionally attractive (working out, getting regular hair cuts, careful grooming, conversational techniques). It's a lot of work, but it's been more effective than deciding the problem is women. Women talk to each other: the friend you make today has a bunch of friends she can set you up with tomorrow. I mean, if the problem is "all women" you might as well give up.

Plus, in the meantime I've met a bunch of interesting people and had interesting conversations I'd otherwise have missed out on because I was only considering women as potential romantic partners. It is possible to have fulfilling intellectual, emotional and social relationships that have nothing to do with sex. If you do care exclusively about sex, I recommend escorts or online hookups.

If you are interested in relationships, those 20% are unlikely to provide them. I think online dating is a much better choice: I know several friends, including married friends, who met on OkCupid.

The problem I have observed is that nerds assume the problem is women. It's like if you're linking in a common library and it's failing to load: what's more likely, that there's a bug in the library that thousands of other people use or that you accidentally linked against the debug version? Someday I'll get around to writing my "How to Hack Dating: an iterative approach to acquiring relationships" book. Lean Startup principles apply surprisingly well.

We are all human beings with weakly rational brains.Only thing wrong here is that it is not socially acceptable to be biased against women!

I get about 5 mails per week pimping some junior job to me that would probably pay 40% less than my current job. I don't reply to these emails because these people haven't done their homework.

The poster above you is saying that maybe that's how attractive women fail...

For women not to be represented in the population that practices software engineering indicates some intrinsic flaw or inability with them --

but for one that is nice there is at least a hundred shallow bitches around

-- But for you to fail to achieve representation in the population which can acquire a mate or find a willing sex partner is ALSO a flaw with women!

Brilliant! Heads: you win; tails: women lose.

"but young nerds in particular tend to act bizarrely towards women, and are therefore undesirable to be around."

and a personal explanation of why I as a young nerd is slowly developing this kind of a bias.

You can't change other people. Focus on the things you can affect, like finding ways to make yourself a more relaxed and relaxing person to talk to.

Some people are trying to give you feedback in other parts of this thread. Much of it is not constructive, but some is. But you seem to be focussed on reiterating your points rather than taking it on.

Work out how you can be more open to feedback. Find someone who will give you honest feedback. Distinguish between people who give good, direct feedback and people who get off by demeaning you.

I assume you created this account as a throwaway because you had an idea what you were saying would be controversial but this stuff is troubling you you and you wanted to talk about it. Good move. Now take the next step.

I either gravitate toward women who tend to respond to my advances negatively or have some behavior that puts women off. Therefore, I'm sexist and consider most if not all women inferior to men in ways including ability in math, engineering. Because I can't get them to sleep with me.

...What??

Read that post again very carefully. Either you look for the wrong women, or you do something women dislike. Both are relatively easy to fix.

Go talk it out, either with a friend or a therapist. The longer you wait, the harder it will be.

I think like all other people you are missing my point.Please read my other responses on this thread.

But Yes I am certainly a failure so far by all conventional standards......but that does'nt mean I have the magical ability to get rid of all my biases....dealing with them is going to take time and in case of this particular bias ,perhaps a lot of positive experiences of interacting with women.

The specious arguments you put forth in the thread are not worth addressing.

a) Most people dont have any prejudices. b) I care about the societal acceptability of my thoughts before thinking them c) One's thoughts should be discarded if they can somehow be negatively viewed in job interviews d) I cannot have my own business and therefore not need anyone to give me a job.

That is a misleading characterization of the paper, which aims at establishing an analytical framework rather than offering itself as a 'report,' within an existing analytical framework. For example, rather than suggesting that women are 'fearful' of text interfaces, the author observes that their continuing popularity in the FLOSS community exacerbates past educational disparities: 'Instead of deducting from biological sex difference, the phenomenon suggests a lingering deficiency of women’s IT education and women-unfriendly products and tools.' The writer goes on to posit that many staples of geek culture (eg long coding pushes) act as exclusionary factors for women who may have to juggle coding with other tasks such as child-rearing, and that women's contributions in areas such as documentation or design are seriously undervalued.

Your summary of the paper is so far off base that I find myself wondering if you inadvertently linked to the wrong document.

You are correct, however, that my use of the term "fearful" was wrong. Thanks for the correction. In actuality, the article doesn't explain why women don't teach themselves how to use gcc.

And yes, you are also correct when you point out that one specific reason women are less dedicated and less willing to put the time in is that they have other interests (such as child rearing and housework).

Apart from the minor issue of whether "fearful" was correct, what do you object to about my characterization?

"Debunking Myths about Gender and Mathematics Performance" http://www.ams.org/notices/201201/rtx120100010p.pdf

Now for the anecdotes: in my experience there has been plenty of outright racial and gender discrimination in computing, science, math, and even medicine. I've seen it. I suspect most if not all of you have seen it. It persists into the 21st century.

That needs to continue to change.

It agrees completely with the data I cited. It shows a variance ratio of 1.1-1.2 across many countries (though not all).

Interestingly, it also shows very little correlation between gender equity and gender disparities in math performance. The strongest correlation it shows is that gender equity is positively correlated with math performance disparities! Gender equity seems to increase [1] math performance of both boys and girls, but it increases boy's scores more.

[1] Of course, the article only shows correlation, not causation, but I didn't feel like rephrasing what I wrote.

While I do think the variance ratio in math is significantly different from unity (without speculating as to the cause), it's worth pointing out that while this does explain a lot of the gender imbalance in math, it doesn't come close in computing. Off the top of my head, in math it's something like 35% women, a 1:2 ratio, whereas in tech it's closer to 10%, or 1:9.

To explain a gap that wide based on variance, we'd have to assume that men are way more than 10% more variable in computer ability than women, probably more on the order of 100%, and that seems very unlikely to me (there's not really any data to look at there because CS performance is not as commonly measured as math is).

This was the most glaring example of statistical dishonesty in this paper: their data shows with perfect clarity that there's a > 1.0 male/female variance ratio for almost every country in the set, and I encourage anyone interested in this to look at their graphs and draw your own conclusions (http://imgur.com/39pja). To me, it looks like a typical noisy measurement (the authors note that the variation within a single country was about 20% from test to test, so we should expect a decently wide distribution (well, the authors don't actually admit that - as a first mathematical blunder in a series of many, they claim that 20% variation is very small and means we should expect a tight distribution)) with a mean somewhere between 1.12 and 1.15, a variance of maybe .1 (just about right for a measurement with around 20% variation, no?), and a decent bit of skew. Pretty good jumping off point for some analysis and explanation, I would think...

But not in this article. The authors merely point to those graphs and claim that they obviously disprove the greater male variance hypothesis. In other words - they point to a distribution of admittedly noisy measurements that is clearly centered around ~1.13 or so, with almost zero density below 1.0, and claim that it proves that the mean of the distribution is 1.0, (with an implicit "STFU Larry Summers"). When I first read this, I thought they were trolling me, the result is so clearly wrong.

[As an aside, it's worth noting that a variance ratio in the 1.1 to 1.2 range is enough to explain away most, if not all, of the gender imbalance in mathematics, if we make the assumption that the variance ratio is the same throughout all of mathematics education (which is not likely true - IIRC these measurements were all at the 8th grade level)]

Their argument? Because the measured variance ratios are not identical, we should ignore the mean of the distribution. Seriously, that's it. I'm not talking about a statistical calculation showing that the null hypothesis (that the variance ratio is 1.0) should not be rejected, mind you (because such a calculation would not allow us to accept the null hypothesis - a quick look at the graph is enough to be sure of that), or in fact any statistical argument at all. They quite literally claim that any variation in the measured variance means that the entire distribution is meaningless.

They also try to confuse the issue a little bit by pointing out that there's a bit of correlation between the variance ratio and the variance; while interesting and certainly worthy of further explanation (not done in this paper), this is completely and utterly irrelevant to the variance hypothesis, yet they imply that it somehow disproves it as well.

Quite frankly, the referees should have caught this, I can't remember the last time I've seen such bad statistical reasoning in a legitimate math journal. There are glaring issues all over the rest of the paper, too, where they've filtered and re-filtered data many times until they obtain the correlations that they want, where they chop data into bins in suspect ways, etc., but to actually enumerate all these errors in detail would make this a much longer rant.

It's a shame, too, because it appears that some of the article is solid, and it presents some interesting data that definitely warrants more investigation; unfortunately, they went absolutely bananas in several places, drawing completely unfounded conclusions from the data they generated, so I would be hesitant to cite this paper as proof of anything.

Especially the first one: women are better at math in some cultural settings and worse in others, suggesting that the gender gap is culturally determined. In order to prove that these things were biologically determined you would either need to show that they were present at birth, across a large number of different cultural contexts, or had direct, traceable biological causation. In the absences of any of those three, I and most social scientists will assume that gender differences are the product of culture.

I think you've got an American bias there. :-)

I'm joking.

but fowler is using "natural" in a sense more like "unbiased by historical accident" or "what we would get if a bunch of people were suddenly created out of nothing and formed a society from zero".

so there's nothing new here. grumpy, indignant old men (self aware or not) believe the status quo (which only incidentally favours them) to be natural. people like fowler do not.

[edit: i've removed fowler's superlatives to be less unpleasant (sorry - in my defense i have been configuring maven all day...); the grumpy old man characterisation is from the parent post]

He is clearly starting from a conclusion and looking for the evidence and/or complicated logic to satisfy his own conclusion. In discussions of inequalities you will always find evidence to support the 'but it should be equal stance', because

a) Egalitarian assumptions invalidate otherwise valid explanations (which as sherlock holmes/bayes have it means all other explanations are more likely)

b) The world is a big complicated place, with lots of unknowns and ambiguities

c) Positive feedback may have reenforced an inequality, exaggerating it and thus making it hard to explain the magnitude using only simple explanations.

In this argument, he is only smart in the sense that he is taking a socially acceptable/convenient position.

Yes, you are correct. I made an edit to that effect.

If I observe something, it is occurring, correct? Or I could not have observed it.

The question becomes why is it occurring? We can observe "simple" things like trees and rocks and disagree on the reasons happen the way they do.

I'm not trying to argue from ignorance here. Certainly you could ask each and every person and draw some general conclusions. But that kind of approach is very fluffy and you could spin the results to say just about anything you wanted. If you have a master's degree in computer science and decided to be stay-at-home dad, is that because the system is flawed? Perhaps you just prefer being with your kids? Most people don't make those kinds of decisions for any one reason. It's complex.

Like I said, I love diversity. I think my biggest problem is this continuing thing we do where we define diversity as external characteristics. It's the internet age. I could care less what your sexual organs are or skin color is.

EDIT: Just to be sure I am advancing the discussion, the moral question is this: Assume I run a company with 100 employees. Only 20 of them are a member of a sub-group that is 30% in the larger population. Is it the moral thing for me to do to hire another 10 people of this subgroup, even if it means not hiring people of another group that might be better qualified just so the ratios match up?

If the answer to that question is "yes", then I have two follow-up questions. One, how many kinds of subgroups do I need to track? One-armed people? People who have beards? Who gets to decide what subgroups are special or not? Two, is it moral for a voter to make somebody do something because they personally find that it has a moral outcome? If I wanted people to be nicer in the world, could I pass a law that required all of them to give 10% to charity? Does something I feel morally outraged about automatically mean I should go mucking around with somebody else's freedom? If so, where's the stopping point?

As for Jim Crow laws, please note that I am not saying that society shouldn't evolve. I'd argue that some degree of legislating morality is necessary for society to move along -- even though I find it most distasteful. But there should be time limits on these things. That's why I bring up the internet. I really don't need grandpa's generation telling me how I should think. This is something each generation needs to settle for itself.

Even if you don't care about those things, it feels shallow and dangerous to pretend to live in a world where nobody cares about those things.

All rhetoric should be looked at critically, and there's nothing wrong with being extra critical of rhetoric that's historically dubious.

What's your initial response to the phrase "separate but equal"?

Yes there is. It biases our thinking.

Here is a non-controversial example of this: "Millikan measured the charge on an electron...got an answer which we now know not to be quite right...It's interesting to look at the history of measurements of the charge of an electron, after Millikan. If you plot them as a function of time, you find that one is a little bit bigger than Millikan's, and the next one's a little bit bigger than that, and the next one's a little bit bigger than that, until finally they settle down to a number which is higher.

Why didn't they discover the new number was higher right away? It's a thing that scientists are ashamed of - this history - because it's apparent that people did things like this: When they got a number that was too high above Millikan's, they thought something must be wrong - and they would look for and find a reason why something might be wrong. When they got a number close to Millikan's value they didn't look so hard. And so they eliminated the numbers that were too far off, and did other things like that..."

(The narrative comes from Richard Feynman.)

If we look on some hypothesis extra critically, and others only reasonably critically, we will bias our views towards the beliefs which we apply less scrutiny to.

(See also this discussion, where I defend a paper claiming to prove the existence of ESP: http://news.ycombinator.com/item?id=2068699 )

I'm going to stand by the notion that extraordinary claims require extraordinary evidence, even if that may create a small bias toward conventional thinking.

In Bayesian terms, it sounds like you are merely stating that you assign a very small prior to the "natural reasons" hypothesis. I.e., you are claiming that before looking at the evidence, you believe "natural reasons" has a 1 in 1 million (or some similarly large number) chance of being true. Is this correct?

These discussions aren't theoretical; to pretend they are is actively harmful.

Further, it's not even a correct use of the Appeal to Consequences fallacy. If the natural causes hypothesis is true, it does not logically follow that anyone should be oppressed.

It does, however, logically follow from the "natural causes" hypothesis that you can't have both equal rights and equal proportions (i.e., either you discriminate against men, or you have fewer women).

Do you think having more women in our profession would actually be a bad thing?

Fowler's point is that lopsided ratios existed many times in the past, and people argued that they were natural, but they turned out not to be, as was proven when the barriers were torn down and the ratios equalized. So in that sense, he's arguing that ratio observations do not count as "evidence" in this context.

However, what he's overlooking is the fact that in most cases, those imbalanced ratios in the past were buttressed by direct barriers to entry - women and minorities were usually actively excluded, and then when the active discrimination stopped, things started to equalize, often very rapidly (the "Jackie Robinson" effect).

We have seen this happen in most fields (tech has just about the worst gender imbalance in any field apart from nursing). And now we're at a point where explicit barriers to entry are all but absent. So as time goes on, the observation that the ratios are not approaching 1:1 in tech suggests more and more strongly that there's something else going on, and that the methods used to create parity in other fields (fighting -isms, mainly) are not addressing the root cause of inequality in tech.

That's not to say that this is evidence that there's not some other sort of more subtle discrimination going on; that's certainly possible, and I'm open to evidence that something in tech is somehow worse than in other fields (I've never seen any numbers to suggest this, and when looking at data like salary gaps, tech appears to be more equal than most fields). But it does mean that we should be wary about drawing the same causal conclusions that we drew 60 years ago when the situation was very different...

No. That tells you there's a difference, but absolutely nothing about the causes. You may not know of an artificial causal effect that inhibits greater participation in software by women, but not knowing about it doesn't mean that it doesn't exist (unless you have falsified all available hypotheses, naturally).