Are women worse programmers than men?

Published at: 2010-04-23 18:34:18 UTC in A Foo walks into a Bar... blog. Summary >>

As a student of a university with 90% male alumni, and with regular attempts of women to abuse their "girl power", I participate in "female programmers" flamewars quite often.

In these flamewars I usually try to stick to "logical" side, trying to analyze things without any bias to political correctness or male sexism. However, this position is not appreciated by both women (they disregard all but that "we are all equal" stuff) and men (half of them thinks it's too offensive, and the other half thinks that I'm too soft). But now I have the place without moderation, where I can tell everything I want.

FAQ

I'll start with a small FAQ, which explains the position I stand for in this post.

• Do you think that women are worse programmers than men?
  Yes.
• Does it mean that we should give preference to male humans when hiring?
  No.
• Why is there a debate about whether women are good at programming in the first place?
  Because answers to previous two questions are different.
• I know a famous programmer/computer scientist/developer Y, and she's a woman!
  So? I don't say that there's no good female developers, I just say that there's fewer of them.

So I'll try to elaborate these questions.

Are women worse developers than men?

There are slightly more women than men among humans. However, there's just a small amount of women in software development. I observe this in the place I work, and I hear a lot from others that they observe the same at their workplaces. If you're a software developer, take a look who surrounds you: how much of them are women? Not much I guess, much less than 50%.

I consider this experimental data. Sociology? Psychiatry? Who cares?! I'm partly a physicist, and experimental data are superior: if there's a smaller amount of female programmers in our more or less free world, then it means that they are just worse at it. Women either don't have a skill to be a programmer (and it's obvious that to enter the profession you should have a higher amount of skill) or don't have enough motivation. I don't really know what's happening there, but the result is obvious: male are more confident in becoming programmers. It's a fact.

Moreover, we usually observe that things are so bad, that for every female programmer (either "good" or "bad") there's usually at least one "good" male programmer in the office.

On this basis, it's not wrong to say that women, in general, are worse programmers than men, since a few of them bear the skills and will big enough to choose it as a career.

Let's formulate it in more mathematical manner. It means that if you compare a random woman and a random man you met in the street, a male will have better average result than a female subject. Or, if we discard probability distribution cruft, that P(W and G) < P(M and G), where W and M are events of male and female becoming programmers, and G is the event of someone becoming a good programmer (P(x) being the probability). The above statement is true, just because P(W and G) < P(W) is true by definition, and P(W) < P(M and G) is true by observation.

Why can't we then think that a woman we're going to hire is worse than a man we're going to hire?

That's simple. The above statement means that a random woman is worse than a random man when it comes to programming. But when you're making a hiring decision (or when you just evaluate your colleague), it's no longer a "random" woman.

Now it's a woman that explicitly chose one of her (possible) careers as software developer. So she already became a programmer; good or bad--we're yet to decide. And now her skills have to be evaluated in the same manner as men's, since there's no more bias.

What's the probability of her being a good programmer? It's a conditional probability: P(G | W), i.e. "someone is a good programmer given that it is a woman that became a programmer". The probability of a male being good is similar: P(G | M). So now we should compare P(G and W)/P(W) and P(G and M)/P(M). We shown above that P(W and G) < P(M and G), but it's also true that P(W) < P(M). So maths is of no help of us in determining which is bigger: P(G | W) or P(G | M)?

I don't have the answer to this question. In my causal observations I didn't see any evidence that female programmers are worse or better than male programmers. And this post doesn't try to solve this contradiction.

What it shows is that even since a random woman is a worse programmer than a random man, this does not mean that women who chose programming as their career are worse programmers than male developers. This really depends on context.

So, programmer girls object to the improper treatment, programmer males think that girls are worse programmers, non-programmer girls just don't care, and all of them are right--in a proper context. But it's understandable why it's incorrect to make a professional assumption based on gender in the office. It's just based on nothing, and observations made in previous section are no longer true when it comes to programmer women.

Why is there a debate about whether women are good at programming in the first place?

Because not many people are familiar with the concept of conditional probability. And because it's very hard to believe both that women are worse than men, and that it's not the basis to consider your colleague Jane a worse developer.

But there's one more thing. I suppose that some women do not tend to avoid discrimination, trying to abuse it when it's in their best interest. And even if they don't try to abuse it, it sometimes looks like they treat men and women differently in the first place.

The most notable example would be a quote from the deleted SO question (you can only view it if you have 10k+ rep there):

I do agree men and women think differently

So if you agree that you think differently, why do you not agree when you're treated differently? And why do some of you not object to that treatment when it's convenient to some of you?

It also seems that some people are like they just want to be discriminated. An extreme case is when a female programmer puts her gender into her nickname or domain name of her homepage... I encountered it in a discussion in the Google Summer of Code 2009 private mailing list (yes, we had a flamewar of this kind there as well). Can't cite the discussion (the list is private), but I think I can share that some people cited she.geek.nz, and the most ardent female custodian of equality had her homepage at... latvialinuxgirl.blogspot.com!

Well, your domain name is sometimes the first thing anyone knows about you in the Internet, and is it really necessary to show the you're a girl in it? And if you think that it is, then why do you think that your gender is important to your identity? And how does it stack with your equality speeches?

These are rhetorical questions. I know the answer, and you do too. To reach equality, everyone should stop demonstrating gender as something that's important to their career and professional identity. For example, my website title aims intergalactic political correctness--you can't know that I'm a human, let alone a male one.

I know a famous programmer/computer scientist/developer Y, and she's a woman!

Cool! I would like to re-cite a deleted StackOverflow answer by richj:

• Ada Lovelace
• Grace Hopper
• Barbara Liskov
• Rebecca Wirfs-Brock
• Barbara Moo

If you know zero of them you most likely know nothing about programming.

Once when we were celebrating an event at work, I made up a toast. I said that women are not welcome only in the most awful things in the world, namely the war. And that I wanted to lift a glass for the women currently there who justify our domain, software development, as something that is worth being.

So, to you, to all of the women cited above!

Back to A Foo walks into a Bar...