Monday, August 19, 2013

Why Aren't There More Women Programmers

When we feel like we're good at something, that's when we can really learn it, sink our teeth in and love it and do it until we're experts.

It is a great motivator, this feeling of confidence. This belief that we can accomplish what we want to do is called self-efficacy. There are four sources of self-efficacy at a particular task (in order of strength):[1]
  1. doing it
  2. seeing people like me do it
  3. social persuasion
  4. your body
Why aren't there more women programmers? Because many women don't feel like they can do it. Therefore they don't try it, or don't persevere. These sources of self-efficacy explain why women (in aggregate) are less likely to program than men.

1. doing it: If you try something and have success, this is the best source of a feeling of self-efficacy. In my generation, more boys than girls started programming at an early age.

2. seeing people like me do it: If my mom can do it, so can I.
People choose careers by picturing themselves in that job, and the picture is based on people we know. If we can't picture ourselves in a career, we don't consider it. The gender dichotomy is strong in our culture, a big part of whom we consider "like me." A boy can picture himself as a programmer. When a girl looks around, she doesn't see people like her coding, attending conferences, speaking, blogging, contributing to open source. Even women developers: looking up in the hierarchy, they see themselves in management, analysis, QA, BI, maybe DBA. Not system administration. Not architecture.

3. social persuasion: Do my friends approve of it?
Here, we look not at the culture of programming, but the culture of women. When I go to Kindergarten Moms Night and someone asks what I do, "computer programming" usually ends the conversation. I don't fit in. Growing up and being grown up, if you enjoy the company of other women, the social persuasion is against development.

4. your body: Do you get a good feeling in your stomach? I don't know about this one. Is it different for men and women?

Women are just as capable of programming as men, but (in aggregate) they don't feel as capable. If we can change that, then we can change the ratio.

#1 is the most direct route, and many in the community are working on introducing programming to young people, especially young women. Yay for them!

#2 is something else we can change. Increase the visibility of women who are already developers, especially those at the highest level. I want to look up, up on stage or up the decisionmaking structure, and see people like me. Conference organizers who go out of your way to elicit proposals from women, thank you. You're helping.

#3 comes the larger social culture, not the culture of programmers. I can't be a typical mom and a community-involved developer. This is not something the programming community can change. Personally, I see #3 as the most intractable obstacle for women.

As a community, #1 and #2 are the ones we can do something about. And hey, they're at the top of the list! If we keep working at this, we'll reach s critical mass. Once programming is (say) 33% women, then #3 will fix itself. Without intervention, the social pressure and lack of role models combine to attract and retain fewer and fewer women. With work, we can turn that spiral around.

--------------------------

[1] At PyCon Canada, Mel Chua gave a great talk on learning. This list is hers.

34 comments:

  1. A good look into Maslow's Hierarchy of Needs and further insight into the situation. Thanks for posting!

    ReplyDelete
  2. It is funny that you see #3 as the biggest problem as it seem to be the less convincing to me.

    Being a man, and starting programming at early age, I can assert that #1, #2 are rather true, but I can assure you that liking programing instead of usual boy activities (like sport) is not celebrated. Any nerd can confirm you the story. Adults will consider you like a pain in the ass, asking too much questions, and for other kid you are just weird.

    You may want to read that : http://www.usc.edu/~douglast/202/lecture23/manifesto.html

    Depression is a major issue in the hacker community, and needless to say that Aaron Swartz, Julian Assange, Bradley Manning, Edward Snowden or Gottfrid Svartholm cases do nothing to help.

    I you think that behavior is somehow approved for bots, you are badly deluded.

    And femnists, by promoting figures like Anita Sarkeesian or Adria Richards rather than badass women programmer like Barbara Liskov, Grace Hopper or Christina Ann Coffin (for the video game world) only makes thing worse for #1 and #2.

    ReplyDelete
  3. deadal, you are not necessarily typical for the programming world. Consider that many people who work in programming were only mildly interested in computers before they went on to study. Most of my co-students, in a typical comp sci mostly-male education, were like that. Most did not fit the typical "nerd" pattern, nor do I think it should be a requirement as you seem to imply.

    You talk about depression, but it's a bit vague and I don't see what this has to do with anything.

    Your last comment is completely off base though. Anita Sarkeesian is not well-known because of "feminists" (who?) promoting her. She is well known because her minor kickstarter project got an unbelievably disproportionately negative, clearly misogynistic response. That's not "feminists" fault, it's rather a good illustration of just how inherently sexist our society still is.

    ReplyDelete
  4. deadal, you have a good point, and @chris__martin made it too, on twitter.

    When we were kids, all programmers were isolated from their peers.

    Do you still feel persecuted, now that you're an adult?

    If you want a community with lots of men who are also programmers, Meetup.com will have several if you're in a major city. Or go online to any programming forum.

    If you want a community where being a programmer and being a woman are both normal, well... people have created these intentionally. They don't exist otherwise.



    ReplyDelete
  5. There's something I missed about #3, the social isolation from other women that comes with programming.

    In the programming community, and in geek culture, there are personality traits that help me fit in: having opinions and defending them, resilience to criticism, taking and giving teasing. Standing out is a positive characteristic, as is individualism. It's perfectly acceptable to say, "That's your opinion. I'm going to do it my way."

    None of these characteristics go over well in a community of women. Women in our culture promote consensus. They aren't confrontational. They spend far more time agreeing with each other than disagreeing. Women encourage each other, and say "look what we did" more than "look what I did."

    I'm not saying one is better than the other. Only that they're different. That one person can't fit in well in both communities. The social pressures coming from women's culture contrast with programming communities, where arguments are seen as constructive, and having a thick skin is a virtue. Consensus-building skills won't get you far in an online forum where ideas are built through passionate disagreement.

    So even if the "you do what? oh." doesn't bother a woman, what we do and become to fit in with our (gender) peers is in opposition to the pressures in programming culture.

    ReplyDelete
    Replies
    1. All the more reason women are critical to have in our community. When there is imbalance in the way we solve problems, it is more difficult to evolve the way we make software. There is a serious technical limitation to any one way of thinking and we continually face the same problems of technical debt and misalignment with business over and over again. Perhaps if we had more nurturing and collaboration, we could spend more time creating than fixing.

      Delete
  6. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. Boy. I'm glad I don't work where you do.

      Delete
  7. Regarding #3, "computer programming" can end the conversation with almost anyone. Unless you're in the field, you don't likely know enough to converse from there. It's also so generic and broad -- what do you really do?

    I traveled, and was surprised at the positive and inquisitive response I'd get to "I make videogames". Ten or fifteen years ago it was a conversation ender because it came across like you play videogames and your some wasterel. In recent years, almost everyone has had some brush with games and can converse about them.

    Your addendum to #3 in the comments here... I can see more of what you mean. The "geek" personality doesn't mesh well with most non-geek. It really seems #2 and #3 are working in a mutually recursive fashion!

    ReplyDelete
    Replies
    1. Thanks for this!

      You're right, #2 and #3 reinforce each other. If we can change #2, then #3 will slowly change too.

      and then, just as geek culture can benefit from women, more women could benefit from some of the individualism in geek culture.

      Delete
    2. More recursion! Yes! The generations currently of working-age might not see much change, unfortunately, but I think geek is chic for the younger generation isn't it? Regardless of gender? I probably have a biased view since most young girls I know of are daughters of nerds. Or from projects on the Raspberry Pi. Biased sample pools, certainly!

      Delete
  8. Thanks for sharing!

    I'm going to add some observations according to what I saw at my university, in Mexico. There are very few female students in majors related to IT (Electronic Engineering, Computer Science) but this disproportion seems to exist only in these fields. Other majors in the school of engineering are about 50% men, 50% women, or 60%-40% at worst, including highly technical, difficult careers, like biomedical engineering.

    I found that fields like bio-engineering, industrial engineering, biomedical engineering, even mechatronics or mechanical engineering where more balanced in the women-to-men ratio. Interestingly, many really brilliant women in those careers (specially those related to biology in some way), but most of them seemed to have little or no interest for computers. Even some of the few women who studied electronic engineering with me preferred to focus their studies on telecommunications, despite being more than capable in software development or embedded systems, but they just are not that interested.

    That's what I think that the main issue here is point #3. Is something cultural. Still, programming is a real field since what? 50 years at most? It became mainstream AFTER the first feminist movements, while physics, biology or chemistry are hundreds of years old, when women were not allowed to study or had to overcome a lot of obstacles. Yet these fields don't seem to be as "male-centric" as programming (although I don't have the numbers to support this statement).

    So, why is programming (or seems to be) so different when it comes to gender?

    BTW, sorry if my english is ugly or I use some terminology incorrectly.

    ReplyDelete
    Replies
    1. Thank you for your observations! This is very interesting, and your English is great. Why is it that women avoid computers?

      Perhaps there is a fear - that's the whole self-efficacy thing.

      Someone pointed out that video games are a way of getting comfortable with the computer. Video games are unfriendly to women: male protagonists, oversexualized female characters. Maybe as girls play more, and social interaction is so computer-based these days, maybe it will get better.

      Delete
    2. That's not really all that true.
      Many games have female protagonists, even more games lets you simply choose your gender.
      Do a Google image search for any game that strikes your fancy and you will find male characters are in the same situation as female characters in games.
      A search for God of War fills my screen with images of naked, bulky male.
      No One Lives Forever fills my screen with a classy, but very deadly lady, no more sexualised than the aforementioned, less so in fact.

      Women tend towards the group more so than men, it's why you generally have a groomsman and bridesmaids at a wedding.
      When men first started conquering the digital world, it was the outcasts that did it, not the jocks. They ventured into it because of strong individualism.
      This accelerated men into that world while the women "stayed behind", more mindful of tradition and not breaking with the status quo (that nerds are lame).

      Computers are still considered lame, Facebook, Twitter, Gran Turismo and The Sims aren't considered an extension of the computer, it's just there. How the silicone and metals make stuff work is "bleh". We see that in the youngest generations, where a kid can find years old dirt on someone, but can't navigate the Control Panel and doesn't know what "Run" or a command line is.
      This is my take on why still today few women enter computer science fields, as you say there is no positive social aspect to it.

      Delete
  9. IIRC, more women were studying computer science in the past before the advent of the personal computer, but then the number of men going into the program exploded. So why did the number of men increase but not the number of women? I think that may be a real hint how we can increase the number of women in STEM fields.

    ReplyDelete
  10. Hi! Thank you for posting an article with your thoughts on why there aren't more women developers. I would like to add that although the community is very active and most devs like to dedicate a lot more time to development community and culture than your average mom, I also want to say that being a developer works really well for moms -- there is a ton of flexibility in schedule and the ability to work from home, part time or consult with short term contracts is huge.

    I think a major disadvantage of a women coming into development that isn't often discussed is the dominant male energy present in the community. Just look at any letters from a recruiter:

    Random (fake) example:
    "There is a ping pong table, beer in the fridge and xbox in the office... Women strongly encouraged to apply."

    Seriously, that may be interesting for a minority of women, including myself when I was younger (and perhaps that's why I was drawn to and able to become a successful developer), but in reality, that is a major turn off for women.

    I also think the idea that women are encouraged to apply without creating a space that appeals to women is putting the burden on the women and not really doing any work -- they have been encouraged so we've done our part in creating balance in our community...

    Maybe the development community could take some action of changing the environment to balance the masculine energy so women find it more appealing than a frat house.

    With an imbalance as great as this one, action needs to be taken to make a difference. Simply saying we need more women is not enough.

    ReplyDelete
    Replies
    1. Yes, there is a dominant masculine energy, you're right. When hiring managers think, "What will draw candidates to our company?" the candidate they picture in their heads is white, male, geeky, and likes beer.

      The "everybody loves beer" meme is another one that limits our community. It comes up at conferences.

      And I agree on programming as a great mom career! There are jobs that require a lot of hours and being on-call, and those aren't. And there are jobs (even good ones! like mine) that are 40 hrs/wk with flexibility to work from home and good benefits.

      Delete
  11. "Women are just as capable of programming as men" - how do you know this is true?

    ReplyDelete
    Replies
    1. It seems the most logical a priori position in the absence of concrete evidence to the contrary.

      Delete
    2. "The world is not flat"

      "How do you know that's true?"

      "All evidence pointing otherwise"

      Delete
    3. Not sure if the parent comment meant it this way, but I know there have been studies into the differences between the way men and women's brains work. Could a biological disposition could also be a factor (in addition to the other ones discussed) in the imbalance?

      So, not to say that women who program are any less capable than men who program, but could there be a natural tendency for fewer women to have an interest in programming in the first place?

      Even if it is true, I doubt it would have as much of an impact as the other factors discussed in the article.

      Delete
    4. I don't believe there are any studies into the programming capabilities of women versus men, so I suppose we're left with evidence and logic to make the determination.

      The evidence is clear that there are a lot more men than women programmers. But that probably implies interest more than ability. (Although especially in this field, interest goes a long way in helping one gain ability.)

      From my experience with women in the field, I don't see any evidence that women aren't as capable. Even if it were the case, I think it could easily be explained by less opportunity to improve, rather than innate ability.

      And I guess that brings up the real question. How much of programming is innate ability versus learned skills? I'd posit that it's almost 100% learned skills, and I believe there are studies showing that practice is more important than innate ability in almost every endeavor. And once again, the more interested you are, the more time you're going to spend learning.

      Delete
    5. Computer programming is strongly related to math and science. In regards to math and science ability, the evidence does not point at all to innate gender differences.
      http://thesocietypages.org/socimages/2013/03/07/the-truth-about-gender-and-math/

      And even somethings that are commonly accepted about brains working differently between genders, like spatial ability, may not be innate.
      http://www.pnas.org/content/108/36/14786

      Over the last 40 years % of women employed has increased, and % of women in STEM has increased at a slower rate, but still increased. Computer programming followed this pattern for the first 20 years, and in the last 20 years decreased steadily and is very inconsistent even with other technical fields. Innate gender differences wouldn't account for this change over time.
      http://www.theatlantic.com/business/archive/2013/09/the-brogrammer-effect-women-are-a-small-and-shrinking-share-of-computer-workers/279611/

      In Malaysia, women make up 50-60% of computer science majors. Innate gender differences wouldn't account for this gap reversal across borders either.
      http://gender.stanford.edu/news/2011/malaysian-women-redefine-gender-roles-technology

      Innate gender difference for computing is just a red herring.

      Delete
  12. Be obliged your material thanks for sharing....imeshlab appreciating your material....

    ReplyDelete
  13. I just stumbled upon your blog and wanted to say that I have really enjoyed reading your blog posts. Any way I'll be subscribing to your feed and I hope you post again soon.A fantastic presentation. Very open and informative.You have beautifully presented your thought in this blog post. Home Based Business for Women

    ReplyDelete
  14. Depression is a major issue in the hacker community, and needless to say that Aaron Swartz, Julian Assange, Bradley Manning, Edward Snowden. free netflix account

    ReplyDelete
  15. Depression is a major issue in the hacker community, and needless to say that Aaron Swartz, Julian Assange, Bradley Manning, Edward. professional house cleaning

    ReplyDelete
  16. We are existing because we are not dead; though our daily way of life is packed with the emotional jumble of concepts, feelings, and applications, at another stage we have a low-level careful system that is always looking out for existential threats. http://www.swissphysio.co.uk

    ReplyDelete
  17. We are existing because we are not dead; though our daily way of life is packed with the emotional jumble of concepts, feelings, and applications. more learn

    ReplyDelete
  18. Moreover to monitoring moment-to-moment threats such as an beginning car or a decrease banister, our threat verifying starts to intuit a distant but progressively approaching dark thinking — the approaching end, the biggest boundary. stumbleupon

    ReplyDelete
  19. Born out of loss of lifestyle, designed out of deconstructed individual continues to be, a dreadful whole of wonderful parts. Watch Once Upon a Time Online

    ReplyDelete
  20. As the international provide of copper is constantly on the tense up, the market for illegal copper will likely improve,” notices the evaluation. christmasstockingsetsy

    ReplyDelete
  21. “If you deliver an SMS using a phone company’s system, let it be of value and then you will generate regard and the group will agree to your services since they are the significant funders of the price range,” he says. my link

    ReplyDelete
  22. woah! I’m really enjoying the template/theme of this website. It’s simple, yet effective. A lot of times it’s very hard to get that perfect balance between superb usability and visual appeal. I must say you’ve done a very good job with this. http://moviesfreedom.com/

    ReplyDelete