I started looking into the state of women in the IT industry, open source and PHP for purely selfish reasons. I wanted to satisfy my own curiosity so I decided to do some research. What I found was not only interesting, it shed some light into where we stand, how we got here, why we should even care in the first place, and what we can potentially do about it.
I did want to add a caveat before I go further; much of the data and research I discuss below makes sweeping generalizations about men, women, and our interactions with each other. This is not to say that we all fit these generalizations, or fall into these behavioral patterns... we are not robots, after all, and of course there will be those of you that will say "oh, I never felt that way," or "this never happened to me." Please don't think I mean to offend or offer up stereotypes; I'm only aggregating the data and putting it out there.
A Personal Experience
I have been extremely fortunate in my web development career, in that I have not experienced true gender discrimination firsthand. I know there are countless women out there that have; I think I've just been really lucky. That being said, I'd like to share with you one of the reasons why this topic piques my interest.
My first PHP conference was ZendCon 2005. By that time, I had been working with PHP for a few years, I was an active member of PHPBuilder.com, and had co-authored my first PHP book. I was sent to the conference by PHPBuilder, in fact, because I had been doing weekly news summaries for them, and they wanted a write-up of the conference. I felt a bit nervous going in, only because I don't travel much and it was the first time leaving my husband and 15 month old baby. Plus, I'd not been to a PHP conference before, so I really wasn't sure what to expect. Even still, I walked into the opening keynote a little nervous, but very excited and optimistic.
I was shocked to realize that I was one out of a handful of women there, and I was even more surprised by my own reaction. I immediately felt apprehensive and self-conscious, even to the point of being almost immobile. I think I barely spoke to anyone, and tried meekly to fade into the scenery with those around me. I felt like running out of that room, and I probably would have, if I wouldn't have been so mortified to do so. Thankfully by the end of the conference, I'd been befriended by a few core devs (thanks Wez, Andrei and Marcus!) and a few other PHP devs (which I've unfortunately lost contact with since), and I felt a little more comfortable. Even though I made some new friends, I was still completely out of my comfort zone, and I was quite thankful to get the hell out of there and get home.
Truth be told, I felt the same way at my first few local user group meetings that same year. As the only female in a room of roughly 20 people, I felt incredibly self-conscious, out of place, and like everyone was watching me (which, of course, they weren't). Afraid to open my mouth for fear of looking like an idiot, I think I said maybe two words (which is really funny if you know me personally, because you usually can't shut me up).
Logically, I should have felt right at home. I'm a relatively self-confident individual, I felt comfortable with the language, and I'd been an active member of the PHPBuilder community for years. I'd not had bad experiences with my fellow co-workers or other PHP developers. So what was it about this face-to-face interaction that made me so uncomfortable? Why did I care that I was overwhelmingly in the minority? Interestingly, I felt much the same way Sara J. Chipps did in her blog post about the matter.
Even though I ended up sticking around the PHP community and some of those feelings of self-consciousness have since faded away, I still think about those first meetings I attended. I know many women don't get that far to make it to a conference or user group meeting. For one reason or another, they choose to skip the community bonding, leave an open source project after contributing, or maybe even leave the industry altogether. In fact, we're leaving in droves. I couldn't help but wonder to myself what those reasons were, so I embarked on my own little research project.
IT Job Market is Growing
Initially, I was curious to see if the IT industry was growing and how much potential opportunity was out there. Even with the tenuous state of the economy, I don't think we could be in a better position with regard to job opportunities.
The IT industry is one of the fastest growing industries. According to the Monthly Labor Review (MLR 2004), between 2002-2012, we can expect an incredible growth rate in the U.S. (and presumably other countries) in the following categories:
Computer Software Engineers +45.5%
Computer Systems Analysts +39.4%
Network Systems and Data Communications Analysts +57.0%
Database Admins +44.2%
Network and Computer Systems Admins +37.4%
All other computer specialists +36.5%
We are definitely in the right place, and clearly there is plenty of room for all of us.
How Skewed are the Gender Numbers?
Obviously, women are in the minority in this industry, but just how skewed are the numbers? And what are the trends? First, let's look at the education system, which is one indicator of the state of gender in IT.
Women are less prevalent in IT than they were in 1990, and are getting less prevalent every year.
The Computer Research Association reports that in 1993, women received 17% of the computer science/computer engineering degrees, and that number fell to 12% in 2007. It is interesting to note that when we compare these numbers with degrees in science and engineering discplines as a whole, women have steadily received roughly 50% of these degrees since 2000 (CRA, 2007). Global efforts to bring more women into the fields of science and engineering have evidently been paying off. So what is it about our specific field of interest that makes it so unattractive to women, and more importantly, is making it increasingly
unattractive to women?
I was also interested to find data on the number of women actually working
in the computer science industry, as we know that one does not always follow the same career path as designated by their choice in degree. I, for instance, have a degree in Organizational Behavior (which is probably why I find this topic fascinating). I also know many competent self-taught computer gurus who do not have a degree at all. For these reasons, I did some research on employment as well.
Based on data collected by the Bureau of Labor Statistics and the U.S. Census, the shares of females in these industries is dropping with each passing year. In 1990, women made up 34.1% of workers in these professions. In 2000, the percent of women dropped to 30%, and in 2008 that number is estimated to have dropped to 24.1%. Clearly, if we don't find some solutions to remedy this, women in our industry will become an increasingly rare commodity.
So how do women fare in open source and PHP specifically? The numbers for open source in general are bleak. You may be familiar with a recent study that reports women comprise a mere 1.5% of the F/LOSS community (Ghosh, 2005). Unfortunately, there have been no official studies on the number of women in the PHP community, but I can tell you that at the recent US PHP-centric conferences, typically around 10% of the attendees are women. Not too shabby, but there is definite room for improvement.
Why is this Happening?
The thing that interests me most is the "why." There are numerous organizations and individuals out there who are spending multitudes of resources in trying to solve this mystery. I won't pretend to be more knowledgeable about this topic than any of them, but what I can
offer are some common theories. I suspect it is not just one of these reasons, but all of them working in tandem.Reason #1: Being in the minority sucks for everyone.
Why are women so uncomfortable being in the minority, and why can't we manage to get over it? I hear this from male counterparts all the time: "What's the big deal, anyway? You shouldn't feel strange or odd if you're the only woman in the room because it doesn't matter. You're just as smart as we are. Why are women so worried about this?" Why, indeed. Research shows that both men and women are uncomfortable being in the minority, and when we are, we focus on what makes us different from the rest.
I did a bit of research and found that it's not just women who feel uncomfortable in the gender minority, but men do also. Studies have shown that "being in the numerical minority in a mixed-sex team is not a favorable experience for either women or men (Powell & Graves, 2002; LBS, 2007). Men report feeling uncomfortable, somewhat alienated and highly aware of their minority status when faced with being in a female majority group. (Spelman, et al, 1986). As well, numerous studies have shown that when anyone
is in the minority in a group, they are more apt to place more emphasis on and be more self-conscious of the quality or trait that makes them distinctive from the rest of the group (O'Leary, et al, 1985; Kirkham, 1985). Bottom line is, it's human nature to feel out of place when you're in the minority regardless of gender, and if that means you're the only female around, you're going to be even more acutely aware of it. So ladies, do not adjust your set. There is nothing wrong with you. For the gentlemen out there that have never been in the minority, it's likely difficult for you to relate, and you may be tempted to trivialize our feelings or think we're being overly sensitive.Studies have also shown that being in the gender minority can increase your desire to leave the group.
Another interesting tidbit is that those in a gender minority tend to report "lower life satisfaction, more negative moods, and lower commitment to the organization" (LBS, 2007). For volunteer-centric open source groups that offer little reward but self-satisfaction, this can not be good.
It is also interesting to note that when people feel this discomfort in a diverse group, the easiest way for them to alleviate that discomfort is to leave. If this is not possible (as in the case of a job, for instance), they are likely to instead choose to isolate themselves from the group as much as possible (Bayazit, 2003). Conversely, those who experience positive and frequent social
exchanges, and the perception of inclusion are more likely to feel attached to the team and the other
team members, and be less likely to leave the group (Graves, 1997, Prislin, 2005).
One final word on this: if we're already inclined to leave the group, sexist jokes, inappropriate language, and sexual harassment are not great motivators for us to stay. Just sayin'.
Reason #2: Life gets in the way.
It's funny, when the governing group of PHPWomen tries to get together for status meetings, it's a juggling act. There are but a handful of us, and yet we are all incredibly busy people. We see this on the PHPWomen forums as well, women wishing they had more time to contribute to and interact with the group. I can't say that as a gender we are any busier than men, but what I did find is that we might not have as much control over our own free time.An overwhelming majority of employed women are still responsible for the brunt of domestic duties and childcare, resulting in far less free time, and countless other things we could be doing.
Not surprisingly, a recent study shows that employed women are still six times
more likely to be responsible for childcare and domestic duties (LBS, 2007). In other words, our "free time" is precious and limited. In a culture that values long coding hours and participation, perhaps we just don't have the resources to stack up. Those who are able to spend more time actively participating in open source projects after hours are more easily integrated into the community (Lin, 2005; Krieger, 2006). As well, because our free time is so very precious, if our participation in these projects is not appreciated or readily accepted, we have not less than five million other things competing for our attention.It's easy to get left behind in the IT industry.
It has also been theorized that because our industry moves at such a rapid pace, when we take time off to start a family, it is more difficult for us to jump back into the swing of things than in other professions (Kelan, 2007). We've also seen this sentiment echoed through the forums at PHPWomen; women who take a few years off to focus on family have a lot of catching up to do. This can be overwhelming, especially in a culture with a RTFM mentality. But we'll talk about that next.
Reason #3: There not enough Mr. Miyagis in IT and open source.
We know it's true, and research has backed it up. The IT and open source cultures clearly value independent discovery and have little patience for newcomers (Krieger, 2006). We have flaming wars, trolls, and an RTFM mentality that makes asking questions and learning something new a definite challenge. Heaven help you if you make a mistake or write poor code; you'll end up being mocked on the Intertubes by the coders who take your job when you leave, or you'll end up memorialized forever on the Daily WTF. The acronym RTFM does not exist in any other industry but IT.
A quick check shows that we in IT are the only ones who tell each other to RTFM if a question is asked that is deemed to not be worthy of asking (http://www.acronymfinder.com/RTFM.html
). Why don't we have the patience to help those trying to learn new things? Why are mistakes so unforgivable? It's one thing to strive for excellence in any industry. It's completely different to adopt an elitist mentality that excludes newcomers. I can hear it now: "Mr. Miyagi, why am I waxing on?" "RTFM, Daniel-san."
A groundbreaking study in 1977 showed that when someone is in the minority in a group, their performance underwent closer scrutiny, and were more likely to affect the majority of the group's views about the minority as a whole, which lead to extra burdens and performance pressures on the minority member (Kanter, 1977). In other words, if you're a female in a male majority group, your work is more highly scrutinized and you become the representative for the entire gender. (This sentiment was awesomely communicated by XKCD: http://xkcd.com/385/
). In an industry that not only chastises you for asking questions, but also publicly humiliates you if you make a mistake, I can completely understand why newbies would prefer to keep their mouth shut.
It has also been said that replying with a simple RTFM instead of providing helpful assistance actually makes it more difficult for those who are trying to solve their problem. Googling a question and having the first 10 or 20 results be an RTFM-type answer only makes life more difficult for everyone. In addition, many times there is no complete manual for someone to read, or the documentation is unclear or confusing. Telling someone to RTFM in these cases doesn't do much good.
Alternatively, it has been argued that the RTFM mentality is more along the lines of "teaching a man to fish," instead of handing someone the answer. While I agree with this perspective on on level, and I have firsthand experience with pointlessly trying to help people who are not willing to help themselves, I think that the ones who are earnestly looking for help are the ones who suffer. I'd like to believe that these people are in the majority.Reason #4: There's a whole lotta meh.
It's interesting to note that a recent survey showed that only 66% of men and 85% of women felt that having more actively participating females would be better for the whole FLOSS community (Krieger, et al, 2006). Additionally, that same study pointed out (and was reiterated by Kirrily Robert at the OSCON 2009 keynote) that 80% of women had noticed sexism in the community, whereas only 20% of men had. Don't get me wrong, I'm not blaming the men; I know there are plenty of women out there who don't believe that this is an issue. And if a large percentage of the industry doesn't believe this is a problem, then we don't have a very good chance of changing things.
Why Should We Care, Anyway?
I know some people are really tired of talking about this subject. However, I firmly believe it affects us all equally, regardless of gender, and it should matter to us all. Let me tell you why.Reason #1: Diversity makes us all more creative and innovative.
It seems logical that having a diverse group working on a project would lead to a better outcome. I wanted to prove this with research, but the results were not what I expected at all. When the effects of biodemographic diversity (immediately observable differences such as gender, race, and age) and task-related diversity (functional expertise, education & experience) were studied, unlike task-related diversity, there was no relationship between the quality or quantity of tasks performed by the team and biodemographic diversity (Horwitz & Horwitz, 2007). In other words, diversity solely based on biodemographic diversity did not improve the team's ability to accomplish tasks. This has also been studied specifically in the field of software development and the earlier results were supported. A team will not necessarily be more effective simply because it is diverse (Kang, et al, 2006). I had intended to prove that having a diverse group of individuals made the group "better" as a whole, but research can't back me up. That being said, diversity does have an effect on something equally important.Equally-gendered teams are more innovative, and innovation is crucial to the success of open source.
What research has shown however, is that diversity (most notably gender diversity) plays an important role in creativity and innovation (LBS, 2007). The more equally gendered your team is, the more innovative they will be. In an industry that strives for breaking new ground, this should be a paramount concern.Reason #2: Right now, the future's not so bright.
Every year, we are losing more and more females to the industry. What does this mean for our daughters and granddaughters? I would love to know that my daughter can choose to enter this field without fear of her being sexually discriminated against, or that she will have to undergo the scrutiny and the performance pressures of females today. Frankly, I want her to be able to start her career and succeed brilliantly without dealing with the bullshit. If we all work to improve the conditions for our collective daughters, we can stop this snowball from growing.
What can we do about it?
Again, there are countless individuals working hard to offer up solutions, but here are a few I think hold the most promise. What I think is most interesting is that these are pretty simple solutions we can implement, that could have tremendous impacts on the state of gender in our industry.Solution #1: Diligently welcome women and newcomers into the industry.
Studies have shown that increasing the number of female participants improves their learning environment, as does the presence of mentors and female role models (Faulkner, 2006, Lagesen, 2007). By increasing the "critical mass" of women, the minority effects are diminished, and women are more likely to stay. This helps alleviate what is known as the 'leaky pipe' effect (women leaving the industry.) We have a mentorship program through PHPWomen (that is incidentally open to all) that has helped corroborate these studies, so I'd like to see more of this type of thing in the Open Source (and really, IT industry as a whole) communities.
In her keynote at OSCON this year, Kirrily Robert mentioned some open source projects that were mostly women. One of these projects was Dreamwidth. Dreamwidth has an exceptional diversity policy for their community (available at http://www.dreamwidth.org/legal/diversity.bml). Here's a quick excerpt from their official stance on diversity:
We welcome people of any gender identity or expression, race, ethnicity, size, nationality, sexual orientation, ability level, religion, culture, subculture, and political opinion. We welcome activists, artists, bloggers, crafters, dilettantes, musicians, photographers, readers, writers, ordinary people, extraordinary people, and everyone in between. We welcome people who want to change the world, people who want to keep in touch with friends, people who want to make great art, and people who just need a break after work. We welcome fans, geeks, nerds, and pixel-stained technopeasant wretches. We welcome Internet beginners who aren't sure what any of those terms refer to.
If that doesn't make you feel welcome, I don't know what would. I was thrilled to learn that the Python group has an official mailing list on this very subject, and is modeling its own statements inspired by the policies of Dreamwidth (available at http://mail.python.org/mailman/listinfo/diversity). I would love to see something like this for the PHP community as well.
Another example is the Women in Open Source (WIOS) track in the Southern California Linux Expo (http://scale7x.socallinuxexpo.org/conference-info/scale-7x-women-in-open-source), which is a fantastic way to increase visibility and keep the discussion going in the open source communities. I would love to see more of this.Solution #2: Focus not only on attracting women into the industry, but on retention.
By openly and actively encouraging participation, interaction and acceptance, we have a better chance of stopping that leaky pipe, as discussed earlier. We try to do this through the PHPWomen forums, but I think this is a neverending initiative that could be taken on by all of us.
Corporations who offer flexibility for working men and women can also greatly help this issue. By making it easier for people to juggle home and family life with work life, you are increasing the probability that they will stick around.
In open source projects, producing code is traditionally seen as being more valuable than producing software (Krieger, 2006). If women are unable to actively commit code changes because they have other commitments keeping them busy, we should encourage them to participate in other ways (such as documentation, which may be less time-sensitive), and value those equally.Solution #3: Leave the comfort zone.
In order for there to be change, we will have to put forth effort and step outside our comfort zones. For women who are new to the field, that may mean joining an open source project, or attending a user group meeting, or even showing up at a conference. Maybe it means not being afraid to ask a question or making the effort to befriend someone else in the group.
For women who are established members of a community, maybe this means making a concerted effort to engage the other females around you. Maybe it means to openly talk about personal discrimination or situations, so that others are more aware of what really goes on. Maybe it means to not get discouraged and keep committing code. Maybe it means give a presentation at a conference or user group meeting, and encourage others to do so.
For men who have never been in the minority, maybe it means take the plunge and put yourself in those shoes. Experience what it's like to feel out of place and singled out. If you're male, maybe you could befriend a female, or at the very least try and empathize with her situation.
For all of us, regardless of gender, we could try resisting the RTFM urge and simply offer practical help, or a link without the snarky comments.Solution #4: Take it local.
Women are busy, as has been discussed earlier. Based on current research, a local outreach to women at the user group level may be one solution to involving them and engaging them in the community, as it will be less time prohibitive and easier for them to attend (
I would love to see OSS user groups adopt a welcoming policy as mentioned before, and actively encourage women to not only attend, but also participate in group activities.
Summing it Up, and thanks PHP!
I hope I have at the very least given us all some food for thought and a little insight as to how we got to where we are. I hope that we can all see how this issue will only get worse if we don't do something about it, and that even small initiatives can have a great impact.
I've said this before, and I'll say it again. On a personal note, I have been thrilled with the support we have received for PHPWomen by our male counterparts. These guys get it. They get that it is not just our problem. They get that even though they can't relate 100% to our situation, they are willing to validate our concerns and work with us to find a solution. They get that bringing more women into the PHP world just helps us all. And they get that we're not about excluding anyone, but about including everyone. A heartfelt thank you to them, and I think other open source communities could learn from their example.
Bayazit, Mahmut & Mannix, Elizabeth, Should I Stay or Should I Go? Predicting Team Members' Intent to Remain on the Team, Small Group Research, June 2003, 34, 270-323
Computing Research Association, CRA Taulbee Trends: Female Students & Faculty (http://www.cra.org/info/taulbee/women.html)
Faulkner, W.; Sørensen, K.; Gansmø, H.; Rommes, E.; Pitt, L.; Lagesen Berg, V.; McKeogh, C.; Preston, P.; Williams, R.; Stewart, J., Strategies of Inclusion: Gender and the Information Society, Gender Technology and Development, 2007; 11: 157-177
Graves, Laura M. & Elsass, Priscilla M., Sex and sex dissimilarity effects in ongoing teams, Human Relations, 2005, 58, 191-221
Ghosh, R. A.; Glott, R.; Krieger, B.; Robles, G. 2002. Free/Libre and OpenSoftware: Survey and Study. Part IV: Survey of Developers. Maastricht: Institute of Infonomics /Merit.
Horwitz, Sujin, & Horwitz, Irwin B., The Effects of Team Diversity on Team Outcomes: a Meta-Analytic Review of Team Demography, Journal of Management, December 2007, 33, 987-1001
Kang, Hye-Ryun; Yang, Hee-Dong; Rowley, Chris, Factors in Team Effectiveness: Cognitive and Demographic Similarities of Software Development Team Members, Human Relations, 2006, 59, 1681-1710
Kanter, Rosabeth, Men and Women of the Corporation, 1977
Kelan, Elisabeth, 'I Don’t Know Why' – Accounting for the Scarcity of Women in ICT Work, Women’s Studies International Forum, September, 2007 30, 6, 499-511
Kirkham, Kate, Managing Diversity in Organizations, 1985
Krieger, Bernhard, and Leach, James, Free/Libre Open Source Software: Policy Support; Gender: Integrated Report of Findings, March 2006
Lagesen, Vivian A., The Strength of Numbers: Strategies to Include Women into Computer Science, Social Studies of Science, 2007, Vol. 37, No. 1, 67-92
Lin, Yuwei, Inclusion, diversity, and gender equality: Gender dimensions of the Free/Libre Open Source Software Development, 2005
London Business School Research Team, Innovative Potential:Men and Women in Teams, Findings, 2007
Monthly Labor Review, Employment Projections for 2012; Concepts and Contexts, February 2004
O'Leary, Virginia E.; Unger, Rhonda K; Wallston, Barbara S., Women, Gender, and Social Psychology, March 1985
Powell, Gary N. and Graves, Laura M., Women and Men in Management, November 2002
Prislin, Radmila & Christensen, P., The Effects of Social Change Within a Group on Membership Preferences: To Leave or Not to Leave?, Personality and Social Psychology Bulletin, 2005 Vol. 31, No. 5, 595-609
Sørensen, K., Gender and Inclusion Policies for the Information Society, August 2004
Spelman, Duncan; Crary, Marcy; Wethersby, Rita; Bocialetti, Gene, Men Students in "Women in Management" Courses: Learnings & Dilemmas, Journal of Management Education, 1986, 10, 89-99