So Where Did All the Women Coders Go?

http://www.ministryoftruth.me.uk/2014/10/23/so-where-did-all-the-women-coders-go/

by nagrom • 12 years ago

48 21 12 years ago

ministryoftruth.me.uk

aninteger• 12 years ago

Maybe they didn't go anywhere? I honestly believe the rate has been a steady 15-20% for the past 30-40 years. It's possible women had degrees in computer science but then pursued careers as system analysts, data entry, QA, "web masters" or IT. None of these are pure software development positions, but maybe they were classified as that in the 80's.

m1117• 12 years ago

They apparently went to biology field http://archive.aacu.org/ocww/volume39_1/images/StRoseGraph1....

cliftonk• 12 years ago

Law, medicine and natural sciences are very different than computer science (plus the latter two are graduate degrees). I'd like to see the same graph but with applied mathematics and physics added.

Jare• 12 years ago

This may help a bit (from a previous HN thread)

http://www.randalolson.com/2014/06/14/percentage-of-bachelor...

maxerickson4• 12 years ago

The data for that comparison is here:

http://www.nsf.gov/statistics/degrees/

(I guess that's what NPR used anyway)

The trends show some in the tables here:

http://www.nsf.gov/statistics/databrf/sdb97326.htm

Between 1985-1995, mathematics (all, not just applied) declined but not as much as CS, physics increased moderately.

scoofy2• 12 years ago

Here the articles points to childhood experiences with hardware, commodore 64s, etc. However, i think this is absolutely nonsense. Now, many people go into college with little idea of what they want to do. Perhaps this is a result of the expansion of college from building a unique skilled career path, to simply being expected.

When talking about demographics and college degrees, i think popular culture is certainly relevant. We are talking about high school and college freshmen discovering themselves. Thus, i'd like to point out that 1984 is the same year the film Revenge of the Nerds came out.

Thus, i'll throw out the hypothesis that, since 80's popular culture was a very regressive era in terms of anti-intellectualism, desire to enter STEM probably took a serious hit at the time in general, much more so with women. That is not to say that previous generations were much better, but gone were the days where the space race inspired tons of kids to pursue STEM education regardless of gender.

jlees• 12 years ago

The graph of overall CS enrollment backs up your comment that there was a decline in STEM degrees in the 80s.

Interestingly the next peak of CS degrees hits in 2002-3. While the obvious explanation of the dot-com boom fits, I would say students graduating at this time were considerably affected by the 1980s home computing culture and that early experiences with a C64 or other computer could have been positive, not negative, influences on their choice to study CS.

(The reason I say this being that I graduated with a CS degree in 2003, so am squarely in that camp, and the dot-com boom had no bearing on that decision - by the time I had picked my A-level subjects and applied to university, the dot-com craziness was not in full swing, though I was in the UK. However, growing up in the C64/BBC Micro/ZX Spectrum era was definitely a big factor in my ensuing nerdiness.)

flomo2• 12 years ago

As someone who grew up in the 1980s, I think the pop-culture argument could go both ways, as the nerds got their revenge, and there were also movies such as War Games (1983) and many other portrayals of genius computer hackers. 'It's hip to be square', as the song went.

More importantly, the PC industry crashed in 1984 as did the video game industry. To some extent, it was all perceived as as a pop-culture fad akin to disco and it retreated into the underground. There was also a general cultural shift within the computing world from mainframe systems to UNIX/DOS/C, and I suspect that may have had a greater effect on women's CS enrollment than anyone has identified.

scoofy2• 12 years ago

Look, as is the case with nearly all of pop culture tropes, we could argue the details ad nauseum, but i hardly think that the nerds of the eighties were positive role models. Even if they were, they were far from anywhere close to those of Houston mission control.

In War Games, our protagonist is the one tech savvy kid who "gets it" while all the other technology experts are villains who put the world in peril because they are bumbling fools.

In Revenge, the nerds were still rejects who, against all odds, won a talent show. One of them, in the process, literally stalks and eventually rapes his unrequited love interest, though, for the sake of a happy ending, she ended up "liking it." Talk about a huge marketing disaster for making CS attractive to women.

With Houston, the term "rocket science" became synonymous with extreme intelligence. The entire country was lining up to buy those "heroes" beers. Astronaut was something akin to Fireman for children hopes and dreams. There has never been a CS moonshot for all to see, as the last war hero we know of was Alan Turing, who's accomplishments remained classified long after his public humiliation and death.

This is all to say, for better or worse, that CS, STEM, and intellectualism in general, were not lionized. Indeed the opposite was the case.

jameshart2• 12 years ago

Would be interesting to see the equivalent analysis of enrollment in CS degrees in China and India. I work with dozens of female coders; 90% of them are from China, India or Russia.

tomp2• 12 years ago

I imagine that being a programmer is a much more lucrative job in China and India than being a lawyer/doctor, compared to the US. Programmers are always present on the global market, while lawyers and doctors are inherently local. I might be wrong though, this is pure speculation/no data.

jameshart2• 12 years ago

That's... true, I guess. What impact do you think it would have on gender ratios?

hajile2• 12 years ago

My answer is that women were never there to begin with and the number of women has always been fairly constant.

My mother started her degree in the 70's because CS was a relatively new field and seemed interesting (let's not forget the "computers will solve the world's problems" marketing of the time). She took lots of programming courses and discovered that she was good at programming (and sometimes did a bit of coding at work), but preferred administering the machines and doing keypunching (and later other forms of data entry) where programming is minimal. From what she has told me, this was fairly common for the women she knew (not a matter of sexism nor of ability -- only of preference).

As the 80s progressed, computer jobs shifted. Data entry no longer required a CS degree, so although the number of women doing that job increased, they were relegated to other non-CS degrees (and weren't really interested in CS to start with).

As a more general idea, I think former Harvard president Larry Summers was potentially correct when he said that studies have repeatedly shown that men and women are different in many ways and some of these (IQ for example) have a huge impact on STEM (he was then forced to resign -- not because anyone disputed the numerous studies in this area, but because it is politically incorrect).

For example, men outnumber women in the top and bottom of IQ spectrum (a couple standard deviations either way) note: the AVERAGE is very similar. There are good evolutionary reasons for this. Dumb women were a greater liability to children than dumb men and the tradeoff of some really smart visionaries (and dumb muscle if you would) worked well for human advancement (but was only possible in men).

As a quick aside, though IQ does not accurately reflect overall intelligence, it does reflect many of the most important the parts of intelligence needed in STEM fields.

At the top of the IQ spectrum (the place where the most successful STEM people tend to be grouped), men outnumber women 2-4 to 1 (depending on the study). The total number of STEM jobs in the USA is 4-5% of the population. If most of those people are in the highest IQ quartile, then we would expect men to greatly outnumber women with women accounting for 20-30% of the people "smart enough" to perform the job. This is very closely reflected by reality and also explains why programs to "create more STEM talent" are largely doomed to fail because you can't train people to be more intelligent. Programs to recruit the best and brightest in impoverished and under-educated communities (where IQ scores represent lack of education instead of lack of intelligence) would reap results, but those don't seem to attract the attention of the powers that be.

Another point of interest is that a higher Asperger's quotient (AQ) seems to correlate to programmers in particular (with most programmers being much higher on the autistic spectrum than the average person). Similar to high IQ ranges, Autism spectrum disorders (ASD) favor men 4 to 1 over women.

A final note about sexism: Sexism exists in every job (and goes both directions to one degree or another). The degree to which it exists in CS is dependent on several interesting factors (for example, that AQ factor which means that it may just be bad social skills rather than actual sexism), but does not necessarily correlate to a significant difference in women and technology (that is, if differences such as IQ are a limiting factor, removing all sexism against women would still result in more men in STEM).

This is true before we even account for women simply making the life choice to NOT be stuck in a tiny cubicle working excessive hours and weekends due to a salaried job while dealing with a pointy-haired boss who has the personality of an agitated wolverine. My mother chose to step out of the rat race and spend time with her children as we grew up (and my father picked up the slack -- despite it's effect on our relationship). When we were gone and she decided to go back to work, she needed to go back to college again because CS changes a lot in a very short time. I suspect that many other women face similar choices and make similar decisions (or simply opt to never try because those intelligent enough to do the job are more likely to make that life choice before entering college).

maxerickson4• 12 years ago

This story links Planet Money, which leaned on NSF statistics, I guess the ones published here:

http://www.nsf.gov/statistics/nsf13327/content.cfm?pub_id=42...

If you grab table 33, you ostensibly have the number of computer science degrees awarded between 1966 and 2010. If you grab the xls, hey, you can add them up. Between 1966 and 1985, there were ~195,000 bachelors degrees awarded in computer science fields (exactly what fields would depend on the NSF methodology). About 61,000 went to women.

If you look here:

http://articles.chicagotribune.com/1987-08-02/news/870226065...

You have an article talking about keypunching requiring minimal training, with about 324,000 data entry keyers working in the U.S. in 1984.

I have trouble believing the narrative that the women earning CS degrees in the 70s were, as a group, ignorant of the opportunity to just go directly to data entry (which was obviously there, there were more people doing data entry than had university CS degrees).

hajile2• 12 years ago

The specific concern I was addressing in the first part of my post was the drop in female CS majors that occurred in the mid 1980s. I was only addressing the difference in graduates between then and now (not all female CS graduates by any means) and only what I believe to be the largest factor (as there were certainly other economic and social factors that affected the situation).

Considering 1985 and 2007 (1985 having the largest amount of female graduates in math and CS), we see women going from 38.8% to 25.4% of the labor force. All we need to show is that the difference in female math and CS graduates in 1985 were primarily mainframe operators and/or keypunch operators (although the number of graduates in math could affect the CS numbers). In addition, I nowhere stated that women of that period were ignorant of the requirements of keypunching.

My mother's earnings as a mainframe operator and keypunch operator with a degree in CS far outstripped the earning potential of someone with just a keypunching certificate. If others were aware of this difference, then getting the degree would be worth the extra work (keeping in mind that a degree also carries the possibility for promotion). This is not much different from women getting bachelors degrees only to work at secretarial positions today or pursuing a masters in Nursing when an associates degree suffices for all non-management jobs.

The second part of my post was focused on the overall numbers (that remaining 25% of women who have always worked in CS) and a potential reason for the relative stability of that number over time (presuming the peak and decline in the 1970-80s can be accounted for independently).

waterlesscloud• 12 years ago

The NSF study has an appendix showing the degrees which are lumped under "Computer Science".

One such degree is indeed "Data Processing, Data Processing Technician".

Unfortunately the breakdown of exact degrees isn't available, but given this information it's entirely possible that a shift away from such fields plays a factor in the number of women with "Computer Science" degrees.

It's not at all an unreasonable theory to consider it may be a factor.

http://www.nsf.gov/statistics/nsf13327/content.cfm?pub_id=42...

orionblastar• 12 years ago

In order to learn how to program, you have to have an interest in learning how to program.

When I grew up in the 1980's girls and women didn't have an interest in learning how to program as much as boys or men had. When I went to UM Rolla in 1986 there were very few women in my computer science classes, and for some reason they were most likely to drop out of college before freshman year or change to a different university or different major.

I asked and I was told that the classes were designed for males to learn, but not for females to learn. That because of this women and girls had a harder time to learn math and science because the books were written by men and not women and thus were sexist. That I am only a programmer because I have this male privilege?

I really don't know what to say about that, I always thought males and females had the same potential to learn, and never thought of any book I read or class I took as sexist towards women and girls.

But now more and more women are taking arts degrees instead of science and majoring in women's studies, politics, communications, and other areas.

Just because they don't study computer science does not mean they aren't smart, just smart in other areas. They never developed an interest in learning how to code and I don't know why.

So Where Did All the Women Coders Go?

Related Stories