Friday, February 22, 2008

Men, women, and the Internet

Sean Carroll and the New York Times look at the gender gap in the context of computer science in general and the Internet in particular.

It seems that we are looking at pretty much the same numbers. Nevertheless, the conclusion that the likes of Sean Carroll arrive at seems somewhat incompatible with a fully healthy brain.

The Internet as we know it has only been around for 15 years or so. Everyone started from zero. There have been no historical differences that would systematically lead men and women to do different things with the Internet. Most people use the Internet when they are alone and they can really do whatever they want (and can). The maximum harm that biased parents and other people can do is to discourage one sex from computers in general: but such a hypothetical pressure couldn't influence the internal composition of the activities on the Internet.

Nevertheless, multiple studies have demonstrated that men and women use the Internet very differently. Even coherent and (mostly) friendly multi-sexual communities like ours demonstrate a measurable difference between the interests of men and women.

If you want a more scientific example, two years ago, Pew Internet and American Life published their report:

Men are from Google, women are from Yahoo
Men prefer the functionality of the Internet: the net is an office, a library, a playground. Women use the Internet as a tool for communication. They use the e-mail much more than men do (also negatively: only 6% of women, but 12% of online men, don't use e-mail) and they also participate in the online social networks more frequently than men do.

Men have a noticeably higher variation in the interests, a usual feature of men as an ensemble, and use the Internet for many reasons - news, politics, sports, do-it-yourself, software, music downloading, rating of products, online learning. Note that there was nothing here about "nurturing relationships." ;-)

So the data that Sean Carroll refers to are just another massive confirmation of the indisputable and biologically rooted cognitive differences between the sexes, pointing to the very same type of correlation that we have known from other contexts before the Internet became important: women are socially oriented while men are technically and functionally oriented.

And what about Sean's statement that women shouldn't be doing anything that is related to computers? That's really silly: a typical straw man. The information technologies, operating systems, and essential software and services have simply been simplified - mostly by boys and men - sufficiently so that everyone, men and women, can use it for whatever activity they find attractive and important. Those technologies that haven't been simplified - such as Linux - remain overwhelmingly male. It can't be otherwise.

We have been hearing bizarre conspiracy theories about the oppression and discrimination - and the underlying equivalence - of men and women for decades. Billions of dollars have been invested into various programs to attract girls to computer science. And the result? Of course, the percentage of women in computer science continues to drop:

Wiki: Women and IT
Wiki: Decline of women in CS in Canada
Boston Globe: Gender gap in CS growing
The percentage was decreasing in the last decade and the percentage also decreases as one moves towards more selective levels of computer science education and the industry.

It should be expected that the percentage of women in certain very technically oriented portions of the information technology will continue to decrease because things and their functional relationships are becoming increasingly complex and more standard deviations above the average are needed than ever before: and men have an edge here. The people who expect the women to catch up with men in Linux or drivers or XML standards are living in a virtual reality.

The attempts to make this composition 50:50 is a typical example of a cargo cult science. Superficially, they are trying to pretend that they follow the scientific method. However, something must be wrong because the airplanes don't land and the women are not closing the gender gap in computer science. What's missing? Well, they don't eliminate hypotheses that have already been disproven, something that is arguably important for the scientific method.

Freedom, democracy, and human rights are about equal opportunities. But equal opportunities do not imply equal outcomes, for dozens of objective (and subjective) reasons. Some people still dream about a 50:50 composition in technical fields etc. Dreaming is OK. However, the same people tend to waste a lot of money and accuse others from crimes and nastiness just because the 50:50 distribution is not respected in reality. But this fact is not a fault of particular people; it is a "fault" of the laws of Nature.

Please don't try to dictate Nature how She should behave especially because She is female and such a dictate is thus a form of sexism. ;-)

And that's the memo.

Bonus (Asymptotia)

Clifford Johnson just published a feminist cartoon. Well, both teachers at the picture are pretty much right, either accurately or statistically. But a third teacher who would say that all the boys (or boys relatively to girls) suck in math would be wrong because men have have discovered or invented 95+ percent of mathematics known to our civilization.

Moreover, I think that the two pictures, when studied more carefully, make it clear that this theory simply cannot explain a lack of women in CS or mathematics: the pictures falsify the explanation involving discrimination only. If a person who can't integrate x-squared is told that he or she sucks in mathematics, it has virtually the same - stimulating or discouraging - effects as if he or she is told that his or her gender sucks in math. How could there be a substantial difference here?

If a hypothetical talented person gets discouraged by such comments, both of them will have the same impact on the individual person. It doesn't really matter what the rest of a randomly selected community (other women and girls) do. The important thing for the student is whether he or she himself or herself can do the integrals. From this viewpoint, the information contains on both pictures is identical. He or she cannot.

You can't do it because you are a girl? Because you are an Albanian? Because you are the dumbest among your siblings? Because you have had a brain injury? Who cares? The result for you is the same. All of the justifications matter and neither of them is absolute.

And frankly speaking, I agree that the math skills of someone who integrates x-squared as pi simply suck! ;-) Whether such people are discouraged from mathematics immediately or later cannot have any substantial impact on the gender composition of the world of mathematics (or computer science). Moreover, the true heroes of exact sciences usually didn't give much room to others to tell them that their abilities sucked.

I am convinced that Clifford and others would be capable to analyze this system properly if they wanted and see that their hypothesis doesn't hold much water. However, they don't want to do it. Dogmas are more important than the truth.

No comments:

Post a Comment