[ I was working on a blog post about the Women In Open Source roundtable I ran, and then Brenda Wallace tweeted: “it seems reasonably easy 2 get women involved in opensource documentation, ui design, and even management. Why is it hard 2 get women coding?” Here’s my longer response, mostly with ideas I got from the roundtable. ]
I ran a panel discussion about Women in Open Source at the PostgreSQL Conference East (last weekend). I talked about all the conference events that I’d seen in the last 1-2 years specific to women, and a pair of researchers talked about communication patterns among women on the KDE women’s list. Then we had a 2 hour discussion with the 10 people in attendance.
Three issues that stuck with me from the discussion were:
* Mentoring: Women do not often receive the same type of mentorship that men receive. This has two important consequences: women don’t feel as connected to the community (don’t learn cultural norms, don’t receive favors, don’t get as much praise or reward for work), and women don’t see clear pathways to greater responsibility or prestige (roles are not obvious unless you’re “in the know”, few role models).
* Self-efficacy: Women consistently rate themselves as far less capable than they prove themselves to be. Example: survey of Computer Science undergrads showed that women rated their preparedness at an average of 0%, while men rated their knowledge and prep around 60-70% — even though GRADES proved that the women had just as much ability/knowledge (these 0% folks were getting As and Bs in courses). The UMD researchers said that self-efficacy has been strongly correlated with success in achieving goals.
* Quality of Life: some coding jobs are low prestige, require superhuman hours and aren’t friendly to women (or men) with children. We didn’t have any research to back this up, but there was a lot of speculation that women 25-45 would not be excited to come back to programming after having children because of the life/work balance problems.
Some or all of these issues can probably be linked to the experience of other minority groups.
As far as “what we can do” — I think we need to work environments more humane and accepting of people who have children. There’s a bit of anti-child culture in some high tech circles, and my personal feeling is that this will continue to turn women away.
I also think we need more mentorship! Women need to mentor women, and men need to mentor women. I think training a group of men on how to mentor women would be greatly beneficial — especially if those men-mentors got some kind of seal of approval at the end of their mentorship bootcamp.
There’s always a tension when men mentor women, and maybe if it was a little more organized and some rules were set — like no using the mentorship program as a personal dating service — it might work better than the ad hoc mentorship stuff we have right now.
Less formal than GSOC, but more formal than just contacting people on a mailing list.
I am happy to see that there ARE women out there who are interested in coding and Postgres. Do you know of any groups in the Central Valley of California? Or just another woman? It is quite difficult here to find someone (man or woman) to discuss these things with face-to-face. Thank you for all the work in trying to bring people together.
I can comment on one point regarding “quality of life”.
My line of work is not strictly in coding or in RDMBS but rather in Engineering. I work a with a large group of engineers from various companies. In our peer group, there are certain members that are recognized as begin elite. (These engineers typically choose projects that the rest deem too technically challenging.) I would describe these individuals (both men and women) as not only technically skilled but also obsessive over the work they do. They are SO focused on the work that they do, that everything else in their life is give second place priority. For this reason, the vast majority of these “elite” individuals’ marriages have ended in divorce due to their spouses feeling neglected.
I wouldn’t be so bold to say that the elite among the communities of programmers also share these same qualities as those that I know in my peer group. But if they did, I wouldn’t be surprised. I believe that any of us “mere mortals” that want to contribute to the community can and should be able to. However, I think that it is important to set reasonable expectations for ourselves. If our circumstances limit our ability to support a project that we have interests in, we (and hopefully the others in the project’s peer group) should be happy with the support that we are able to give. If we have patients, gradually over time as our skills will improve or our circumstances change to the point where we may find ourselves in a position to do more.
Christine – I don’t have any database contacts in Sacramento, Redding, Fresno or Bakersfield at this point. Anyone want to speak up? 🙂
You might try http://www.saclug.org/ or http://www.lugod.org/ for starters. Linux Users Groups are in my experience the best place to start when you’re looking for open source community.
Richard – I agree with setting reasonable expectations! Personal boundaries are the key to my happiness at work. However, sometimes, that’s not enough. Again, I have no research to back this up, but my feeling is that policies that clearly state work expectations that are re-enforced by management are a better attractant for people who have significant family commitments — better than vague or non-existent policies.
Pingback: » Silicon Florist’s links arrangement for April 3, 2008 - Silicon Florist
First of all I want to remark that I don’t think that much store should be placed in these sorts of ‘guesses’ about why women are underrepresented in programming. I mean no one would think that they could just toss out that “well I think that the increase in solar radiation is driving global warming” so why should it be any different with this issue. There is a fuckton of research that has been done on this question so, while I do think there is some value in personal experience in these type of questions, unless it is informed by some knowledge of the studies it says more about the person making the observations than it does about the issue.
In particular, while the family effect may be a small effect (one of my female friends may take a less hardcore programming job/take time off to have a family in the near future) it can’t explain why women are so underrepresented early in school. The bit about being self-effacing probably has some truth to it as well but (not going to bother to look up the cite now) studies have found some troubling differences in how girls and boys relate to computes at VERY VERY young ages (like we are talking under 3rd grade maybe kindergarten). That is boys are (on a statistical level) more likely to be motivated to want to take electronics ‘apart’ (literally or figuratively) and figure out what makes it go.
However you want to explain this effect it makes me quite pessimistic that there is a practical complete ‘solution’ to this problem (though the later factors that magnify these relatively early observered behavioral differences may possible to reduce). But this begs the question of why it’s even important to solve this problem. Society should strive to make sure that no one is locked out of doing what they want because of their gender but I would argue that we already have such a system as far as CS goes (not to deny that some discrimination still occurs but it is now the exception). If it turns out that statistically women are more likely to prefer different careers than men so what? I mean it sucks for guys like me who like to date techie type girls but as a social problem I don’t see the issue so long as it occurs because of choice and ability to perform in certain courses rather than discrimination and bias. Note that choosing to stay home and raise children (rather than giving birth and leaving the child with the husband) is a choice and presumably reflects a preference by the couple making the choice. Now jobs shouldn’t impose any arbitrary demands on their employees whether it is coming in at 9am rather than 11am or not allowing part time work when it is just as effective. However, sometimes there are valid business reasons to want people to work long hours or come in at 9am and if it turns out that statistically women are more likely to choose family over programming as a result why doesn’t this just show they had a better (overall) option?
Anyway just some rambling thoughts on the issue. I realize much of it doesn’t directly bear on what you said but I just wanted to put down some thoughts.
(Reposting since I want this to be under my ‘nickname’ and you are set up to use my fullname. I would appreciate if you would delete the above comment and this remark…I don’t care if people figure out who I am but my blog comments shouldn’t come up first on a google for my name).
—-
First of all I want to remark that I don’t think that much store should be placed in these sorts of ‘guesses’ about why women are underrepresented in programming. I mean no one would think that they could just toss out that “well I think that the increase in solar radiation is driving global warming” so why should it be any different with this issue. There is a fuckton of research that has been done on this question so, while I do think there is some value in personal experience in these type of questions, unless it is informed by some knowledge of the studies it says more about the person making the observations than it does about the issue.
In particular, while the family effect may be a small effect (one of my female friends may take a less hardcore programming job/take time off to have a family in the near future) it can’t explain why women are so underrepresented early in school. The bit about being self-effacing probably has some truth to it as well but (not going to bother to look up the cite now) studies have found some troubling differences in how girls and boys relate to computes at VERY VERY young ages (like we are talking under 3rd grade maybe kindergarten). That is boys are (on a statistical level) more likely to be motivated to want to take electronics ‘apart’ (literally or figuratively) and figure out what makes it go.
However you want to explain this effect it makes me quite pessimistic that there is a practical complete ‘solution’ to this problem (though the later factors that magnify these relatively early observered behavioral differences may possible to reduce). But this begs the question of why it’s even important to solve this problem. Society should strive to make sure that no one is locked out of doing what they want because of their gender but I would argue that we already have such a system as far as CS goes (not to deny that some discrimination still occurs but it is now the exception). If it turns out that statistically women are more likely to prefer different careers than men so what? I mean it sucks for guys like me who like to date techie type girls but as a social problem I don’t see the issue so long as it occurs because of choice and ability to perform in certain courses rather than discrimination and bias. Note that choosing to stay home and raise children (rather than giving birth and leaving the child with the husband) is a choice and presumably reflects a preference by the couple making the choice. Now jobs shouldn’t impose any arbitrary demands on their employees whether it is coming in at 9am rather than 11am or not allowing part time work when it is just as effective. However, sometimes there are valid business reasons to want people to work long hours or come in at 9am and if it turns out that statistically women are more likely to choose family over programming as a result why doesn’t this just show they had a better (overall) option?
Anyway just some rambling thoughts on the issue. I realize much of it doesn’t directly bear on what you said but I just wanted to put down some thoughts.
Just to clarify what I’m saying is that it’s a mistake (which no one here has made) to start with the idea that we should have equal percentages of men, women, tall people short people, whatever in any field.
We should start with the idea that any apparent unfairness in any field or the preparation for the field be eliminated. So we should go around rooting out any practice which seems to be unfair to any group or unreasonably discourages them. What statistical correlations happen in the end results seem kinda besides the point (unless they point us to something that strikes us as directly unfair).
Anyway I apologize for taking this further off topic with this comment. The question of why it is that women do less CS is an interesting sociological one in it’s own right and I don’t want to divert the conversation from it to some less interesting social debate but I just wanted to make sure I didn’t want to give a misleading impression of what I was saying earlier.