Tuesday, April 20, 2010

Not All Women's Groups Are Harmful

After reading Hilary Mason's Stop talking, start coding article, I was thinking about writing a reaction. It slipped my mind until Terri Oda wrote her piece Women in computing groups considered harmful? She gave me the opportunity to reflect on what our Women in Science and Engineering group has been up to at Carleton University (CU-WISE), and why I think not all such groups are all that bad.

This is the part of Hilary's article that stuck out for me:
Many groups have popped up that support women in technology, like Girls in Tech, She’s Geeky, and many others (enumerated in Digiphile’s thoughtful post Why Including women matters for the future of technology and society). More often than not, these groups are the canned food drives of the women in technology movement. They make you feel better, they might do a little good, but they offer no fundamental change to the system that created the problem in the first place.
I remember sitting down with our Dean of Engineering to discuss CU-WISE budget opportunities. Dean Goubran has been incredibly supportive of CU-WISE, but said one important thing: he wanted to see less talking and more doing. He figures people have been sitting at conferences talking about the problem for at least 20 years, but the problem remains. What is it that CU-WISE will do to actually change things?

This was a tough question, especially for a group that had only been around for about a year. How would we know that we've made a real difference?

It's still hard to know for sure, but I think we're getting better and better at this all the time. For example, when I do outreach with groups of girls, I ask them hard questions about why there aren't many women in computing, and why it matters. Their insightful answers impress me and give me hope for their generation.

But we can't really measure how effective our outreach is. At least not yet. I do think it's important to get out there and make the image of computer science and engineering a more positive one, but we won't really know if our attempt is working until much later.

I do often hear a number of women attending CU-WISE events express their gratitude for the existence of such a group. I'm not sure we've saved anyone from dropping out or changing majors, but it does seem like a possibility. But this, too, is hard to measure (I suppose we should attempt to do a proper survey someday!).

So I look to the events we've been holding this past semester. Not a single one revolved around a discussion of the issues women in science and engineering face. The first was an opportunity to put our knowledge of technology to use in the IBM Extreme Blue Case Study Competition. Next, we gave our members the tools they needed to succeed in industry, teaching them skills like networking and negotiation at WISE Steps to Success. We had a smaller presentation on entrepreneurship and then held the very successful Celebration of Women in Science and Engineering.

That last one, the Celebration, had a lot to do with doing. Most talks were women simply presenting their research. As Hilary put it, we got together to "do science."

Based on all this, I think CU-WISE has done a good job of disproving Terri's theory to an extent:
Theory: The more time we spend on women in computing initiatives, the less time we have to actually get stuff done.
Naturally, some of our time must be taken away from doing if we are on the CU-WISE executive team, so we must be careful to balance the commitments properly. (This is something I didn't quite get right during my Masters but have improved thus far in my PhD.) I'm hoping that our group does a good job of encouraging our members to do the most doing possible.

And if we succeed, I think we will be far from harmful.


Haz said...

I think regardless of the type of group, they run the risk of pigeon-holing (excuse the topical pun) themselves into the stereo-typical harmful category. Likely reason is because so many groups started as "just talkers" that there's a prior bias to expecting that as the default behaviour.

I know personally that it can sometimes feel exhausting being on the outside watching flood of info. To some extent it feels like all that goes on is preaching to the converted, and I would much rather see posts about cool research ideas rather than gender issues. But on the other hand, if discrimination still happens (just cause I can't detect it in my office doesn't mean it's not happening elsewhere) then something needs to induce the change. I'm just not convinced that anyone's stumbled on the right answer just yet.

Kate said...

Interesting topic Gail!

The other thing I was thinking is that every time we do group activities where guys are invited, every time we give talks (like the research talks at your celebration or the talks being given at try/catch), every time we have a clubs day booths where guys come to talk to us, we are setting an example for them and (hopefully) getting them on board. That's a more hidden 'doing' that is hard to measure, but I think really makes a difference.

When I had first taken over the presidency of WICS and gave a talk to my 'intro to grad school' class about GHC, I was laughed out of the room by the guys in the class as well as the prof. However, after a few semesters of activism in the department and making WICS highly visible, the next time I went to give a talk to the class, I was not only taken seriously but several guys stood up and openly expressed their support and good wishes for our work. I think little things like that are harder to notice but can go a long ways towards culture change.

Paulina said...

I think the theory only applies to 'technician' type of personalities. It does not apply to 'manager' type of personalities. The differences of these personality types are discussed in Michael Gerber's book "The EMyth Revisited" (http://www.e-myth.com/cs/user/print/post/the-three-business-personalities-entrepreneur-manager-technician). I believe if you're a technician at heart - where your focus is on the present and performing the hands-on work - then you might be overwhelmed by organizing, coordination, and talking that is required to achieve success of an organization, like CU-WISE. But if you're a manager at heart - planner, and organizer who turns the vision into action - then that is the way how your 'getting things done' works.
Both types of 'getting things done' are necessary, and the perspective depends on your type of personality.

Haz said...

Great point...I've got very little manager in me ;).

Gail Carmichael said...

One thing that's funny about CU-WISE is that I never would have gotten involved if I wasn't specifically recruited for my experience as President of the undergrad society. I never would have though it was something I needed or wanted, because I did just fine during undergrad even if I was one of the few females.

But once I got into it, I realized that I was really missing out all that time. I missed out on having girlfriends that I could talk to about girly things! Seems so simple, yet it took that long to figure it out. ;)

Haz: I know what you mean about not figuring out the right answer. One way I like to look at it is that the changes in tech culture that WISE groups are trying to instill would benefit everyone, men included. This isn't really much of a focus, though. Probably we should try to make these groups more inclusive (I'm chipping away at the idea that we're women-only slowly but surely!). Viewing ourselves as a fun club or society who also happens to give back to the community by reaching out and encouraging the next generation seems like a better way to approach it.

Gail Carmichael said...

Kate: Just making the few women visible is indeed a useful form of doing. The next generation needs to see that we exist! ;)

Oliver said...

"Probably we should try to make these groups more inclusive (I'm chipping away at the idea that we're women-only slowly but surely!). Viewing ourselves as a fun club or society who also happens to give back to the community by reaching out and encouraging the next generation seems like a better way to approach it."

Any time there's unilateral exclusivity or societally-induced preferential treatment, it seems to generate animosity between the groups (eg. tutsi vs. hutu in rwanda, english vs irish/scottish in britain, nobility vs. commoner everywhere, french vs. english in canada), which I think does exactly the opposite of fostering the kind of mutual trust and respect necessary to have full integration of the genders. As long as the focus for most people remains "us vs. them", then it's like spinning wheels.

There's clearly a demand for WISE-like groups, and to my understanding it's to address disparity rather than just creating an exclusive club/hangout. This is where I think things get a little sticky. If the group is made more inclusive to the point of neutrality, then is it really much different than forming a specialist subset of ccss/etc which aims through contribution of the larger society's actions to discuss issues from both perspectives? If it retains at least the impression of exclusivity, then there's that whole animosity thing from people who feel disadvantaged by not having the same opportunities. CCSS/etc I believe are already inclusive in themselves. It seems that the status quo among people is to be open, helpful, and friendly, regardless of identifiable characteristics. Any kind of exclusion would then seem to be either institutional or naturally driven. Some "thing" is preventing balanced enrollment.


Oliver said...

As Haz put it "But on the other hand, if discrimination still happens (just cause I can't detect it in my office doesn't mean it's not happening elsewhere) then something needs to induce the change." This is precisely the sentiment I see coming from a lot of people. As far as most of the guys seem to be able to tell, the usual suspect issues either aren't happening, or are almost completely invisible. I saw on one forum a rather comprehensive listing, completely categorized, which listed reference to varying accounts of discrimination. However, it still falls far short of the published stats for such offenses. People hear stats, but without a relatively personal connection, there's not much buy-in, and it gets dismissed. In one ear and out the other. Certainly I would expect that most people are rational and level-headed enough to throw support in punishing those guilty of committing such offenses, but they can't genuinely support something which they don't "see". I imagine this is the most immediate thing to address the institutional problems -- raise more public awareness and prosecution of individual cases so that people develop a personal connection to the issue and put a face to it.

The rest is where I think this whole next-generation thing comes in. This works two-fold. First and foremost, money talks. The more interest there is in a program from a highschool level, the more the higher institutions will respond to that demand. The more the demographics get mixed, the more balanced their influence will be in guiding policy. Second, people are impressionable at all ages, but it seems these impressions all depend on this concept of buyin.. willingness to believe what we're shown. In order to make sure that the issues are getting fair and balanced buyin, I think it's important to make sure impressions are set early.

Your work with the highschool students is probably the best direction to changing things. Despite "most" people not knowing what they want to do at that age, innately people have a feeling for what they like and don't like, and these impressions start out early. As early as elementary school when/where boys and girls invariably self-segregate by interest, peer pressure, etc, we're instantly polarized, stake claim to things, and come to build a twisted understanding of the world based on a relatively lawless schoolyard/classroom culture. What I hear repeatedly from people who's friends avoided science/tech is basically "there's no interest". Showing people on the fence that tech is an option is a great start, but the gender segregation I think is rooted at the elementary level. If we can ensure mixed-gender group exposure to fun tech projects geared to appeal relatively equally, either homogenously or by speciality, at an early age, we can get not just the exposure to girls+tech, but also the reintegration of the genders before segregation takes hold, which should encourage trust and respect.. because working together is favoured by working amicably.

Gah.. so long.. My bad. Less however just seems.. insufficient to make the points effectively. x_x

Haz said...

"I saw on one forum a rather comprehensive listing, completely categorized, which listed reference to varying accounts of discrimination."

- Reference plz?

So what are the indicators that things are working? In Sept, I'll be sharing an office with 3 gals and 2 other guys, share my (female) advisor with 2 gals and 1 other guy, and it seems most of the social comp sci events I attend have at least a 50% split. So the weighting doesn't seem to be issue locally. I don't notice or think of any of this unless another blog post comes up addressing it. Are there other subtle signs that can't be read from raw stats?

Oliver said...

I don't keep browser history, so this took a fair bit of aimlessly wandering to find again. :P


It's not much, but it's certainly the best example I've come accross documenting real accounts of such issues. Like you say, we don't notice or think of it except when a blog addresses it, but at that point the address is more often than not a general reference to the issue at large and not specific references. As far as workplaces go, my team at my last job was split 15/11 F/M, and all 3 levels of management that we had any interaction with were all female, and it was 50/50 in the gov 6 years ago. As you say, weighting doesn't seem to be the issue locally.

Random speculation.. if the supply chain in hs/ug is split x/y, then it's reasonable to expect the workforce would also be split that way. My own observations would suggest that that split has an x+a/y-b bias in larger orgs (NT/IBM/MS/GOO/gov), so logically that leaves smaller orgs to have x-a/y+b. 1) If we want to adjust the disparity, start with the supply chain. 2) If we take that y >> x, then the few large orgs that exist constitute a minority representation of business, and the many small orgs constitute a majority. A majority which, from the x-a/y+b factor has a more pronounced disparity, leading to the reasonable conclusion that cs in business is biased towards exclusivity. Tools like that data visualization thing (demo'd at TED and now implemented by google's publicdata project) would be handy if stats were available for workforce breakdown and company size.

Indicators that things are working... could I guess be simplified down to comparable representation. I see no overt reason for there to be any bias on ability, so that leaves interest. The curious question is.. what's to be done if after all the effort, there's still a lack of balanced interest? It's curious, but I actually see (personal observation only) huge disparity of gender in a number of fields... emergency services, military, farming, etc... but there doesn't appear to be any bias or exclusionism preventing entry, so it's very curious why CS stands out so much in this regard.

Haz said...

The list is a little more general than I thought -- I assumed it was a more pointed analysis of the issues women face in Comp Sci.

Good point on the analysis -- does the world bank data set have enough info?

Oliver said...

World Bank data set seems fairly lacking in this regard, from what I've been able to find so far. It's actually curious that world bank collects all these different stats. National census bureaus and revenue agencies would likely be the best bet. They'll have employment records for each person for each company. Fairly simple at that point to compile the data.

Angelica said...

I think that there are two ways to attack the problem:

(1) Top down: role models like Fran Allen, Marissa Mayer, Helen Greiner, awesome female profs...

(2) Bottom up: grassroots movements like Women in CS clubs, workshops, outreach to high schools

Both are valid, but if I think if we want to attack the problem on a more fundamental level, we should be doing more of (1) than (2). Grassroots projects are useful for the short-term, but seem more of a stop-gap measure.

Think of the impact it would have if we had a female Bjarne Stroustrup!

Gail Carmichael said...

This discussion is really god. I wish I was able to contribute but have been travelling.

Oliver said...

Everything on the internet lives forever. Never too late to contribute! ^_^

Haz said...

"This discussion is really god."
- Gail

Tsk, tsk. What would Freud say? ;)

Gail Carmichael said...

Haha - Freud's going to be busy with the increasing popularity of the iPhone. ;) I will try to remember to write when I get home.

Shamoon said...

Your blog is really Excellent. it inspires the reader who has that great desire to lead a better and happier life. Thanks for sharing this information and hope to read more from you..

Post a Comment

Comments are moderated - please be patient while I approve yours.