Wikipedia:Wikipedia Signpost/2011-02-07/Gender gap

New York Times article sparks extended discussion of Wikipedia's "gender gap"
Concerns about the small proportion of women editing Wikipedia have been voiced for a long time, e.g. by senior Wikimedia figures including Sue Gardner and Jimmy Wales in recent interviews. However, The New York Times front-page story on January 31 (see last week's "In the news") brought an enormous amount of additional attention to the topic. This attention included further international media coverage (some summarized by Gardner on her personal blog), and a renewed discussion among Wikipedians. Much of the latter discussion took place on the newly opened "Gendergap" mailing list, which is hoped to "become a space where Wikipedians and non-Wikipedians can share research and information and tactics for making Wikipedia more attractive to women editors" (Gardner). The Gendergap mailing list discussion reached almost 200 postings within less than a week. The issue was also highlighted in a posting on the Foundation's official blog.

On the Gendergap list, Sue Gardner recommended a discussion on Metafilter on the topic ("Wikipedia, Snips & Snails, Sugar & Spice?"). Jessamyn West, who works as a full-time community manager at Metafilter and had written one of the eight commentaries invited by the NYT after its initial article, described on the list why the Metafilter community had a more balanced gender ratio – around 60/40 male/female: I credit this both to some aggressive moderation in what is otherwise a lightly moderated site [we delete rape jokes and I'll take the heat when people flip out about censorship] some cultivation of female members and some visible norm-setting among all the moderators for how we want the community to run. Sue Gardner concluded that "the lesson for Wikimedia [from the Metafilter example], is that if the community makes something a priority, and continually reinforces it, then culture change can be achieved. I find this heartening because I think the people at Metafilter are fairly similar to the people at Wikimedia".

Gardner outlined a "kind of 'theory of the problem'", starting by saying the reason why the gender gap should be considered a problem for Wikipedia at all: "We want women to contribute to Wikipedia because we want Wikipedia to contain the sum of all human knowledge, not just the stuff that men know." She then went on to list the reasons for the gap, one being that "for many reasons that there's no point articulating because they're outside our control, women tend to be less tech-centric than men, and they tend to see technology as less "fun", something to be addressed by usability efforts. Secondly, that "women tend to have less free time than men, and they tend to spend their free time less in solitary pursuits". Adding to this, she observed a "social/cultural barricade [that] is essentially: women (tend to) dislike fighty cultures more than men".

Annie Lin and Sage Ross from the Foundation's Public Policy Initiative pointed out "that the Wikipedia Campus Ambassador program is currently about 55% male and 45% female – a gender ratio that we are quite proud of" and that "a lot of our Campus Ambassadors are people who totally new to the community". (Gardner, too, had remarked in earlier interviews that women seemed more ready to volunteer for such activities when classes were asked on campus.)

Another issue, criticized as "hardcoded discrimination", is that in languages where there are different female and male forms of the word "user", such as German ("Benutzerin" vs. "Benutzer"), a user page on the corresponding Wikipedia will appear to denote a female Wikipedian as male (such as in de:Benutzer:Example), and standard messages inviting people to create a user account might appear to address newbies as male too. Sue Gardner called this "awful" and "a key piece of information that is important and new (at least to me)."

Empirical basis: The UNU-MERIT study
The estimate that only 12.64% of Wikipedians are female, which formed the bases of much of the debate – having been quoted in some form in nearly all the recent media coverage, as well as in various WMF interviews in past months – comes from the 2010 UNU-MERIT study. In a posting on "Floatingsheep" (a group blog by researchers from the University of Kentucky and the University of Oxford) the authors wondered "if this figure accurately reflects the Wikipedia community", asking about possible sample bias ("for example, Russia and Russian speakers are the largest language and country groups represented in the survey even though the Russian section of Wikipedia is only the 8th largest linguistic group"), and further possible selection bias: "There were three times as many male respondents as female respondents. Does this accurately reflect the makeup of the Wikipedia audience? Given the unexpected results for language and country, it is not clear if there might be gender bias as well". Indeed, a different estimate of the Wikipedia audience by Quantcast (quoted by Jezebel writer Anna North in her contribution to the NYT debate, "The antisocial factor") gives vastly different numbers for Wikipedia's readers: 52 percent men, 48 percent women.

In addition, the Floatingsheep post noted a lack of information about the methodology on wikipediasurvey.org (which may be somewhat mitigated by the slides from a Wikimania 2009 talk about the then ongoing study). Wikipedia researcher Joseph Reagle (who is currently working on the topic of free culture and sexism) also noted that it was "as most surveys" subject to selection bias, but quoted an earlier, smaller survey which had given an even lower percentage: 7.3%.

Assuming its validity, the report of the UNU-MERIT study includes several further insights into the gender gap besides the much-quoted number, including: The overall share of unregistered users among female Wikipedians [i.e. visitors to Wikipedia sites] is significantly higher than the respective share within male Wikipedians (52% vs. 35%). ... This gender difference is not surprising, and is probably explained by female Wikipedians being more protective of their privacy than male Wikipedians, and thus less likely to register. ...

The share of ex-contributors within 10–17 years old female Wikipedians is 3.1% and exceeds thus by far the respective share of this group in the [corresponding] male age cohort (2.2%). ...

... female contributors to Wikipedia appear less specialised in thematic fields than their male colleagues, and while their degree of specialisation increases with age, this increase is is less for male Wikipedia contributors. ...

In accordance with overall patterns of education and labour markets in many economies, the share of scientists among male contributors is about three times larger than the share of scientists among female contributors. Another gender specific is that female contributors tend to focus on philosophy, religion (belief systems) and social sciences at a young age while male contributors focus on these thematic fields in the oldest age cohort. ...

The reasons why women in this age cohort [32 and older] spend more than 2 hours more per week than men have to be further analysed, but the fact that women at this age are less often full-time employed, often stay at home in order to care for children, and often work as freelancers may play a role here.

The report noted that the gender gap does not only show in edits, but extends to financial contributions: "Men are obviously more willing to donate money to Wikipedia than women, as they show considerably higher shares of donors in all age cohorts".