Wikipedians are a bunch of misogynistic brogrammer-types with too much time on their hands and a decidedly male preoccupation with facts and science—right? Wrong. At least, mostly wrong.
The prevailing stereotype about Wikipedia readers, and to a greater extent, Wikipedia editors, came about after a 2008 survey found that a mere 13 percent of Wikipedia users are female. Since the survey—the first of its kind, conducted by the Wikimedia Foundation and the United Nations University— was resurfaced by the New York Times three years ago, the statistic has been quoted and analyzed and shouted from the hilltops again and again. But is it even true?
A group of researchers had their doubts, and recently took a second look at the 2008 survey results. Their study, published last month in the journal PLoS One found there is indeed a gender gap, but it's not as bad as all that.
The researchers point out a number of flaws in the methodology of the 2008 survey. One, the data was culled from an opt-in survey, and the self-selection process tends to produce biased results. For instance, it's possible a smaller percentage of women chose to participate in the survey than the percentage that actually use the site. Or, perhaps those who did participate chose not to reveal their gender.
Two, the response rate was very low, so the sample size was extremely small. About 180,000 people participated, only 0.4 percent of the 45 million unique visitors to Wikipedia at the time.
Three, the people who opted to respond weren't an accurate representation of the community as a whole. Unsurprisingly, they skewed toward Wikipedia editors rather than passive users. In general, researchers found that women, married people, and parents were underrepresented, while immigrants and students were overrepresented.
The study, funded by MIT, Northwestern University, the Berkman Center for Internet and Society, and the Ford Visionary Leadership Fund, then compared the opt-in survey results with a nationally representative phone survey of Americans that had bee conducted by the Pew Research Center a few months later. They analyzed the two datasets together, estimated bias, and corrected for it.
The result? A slightly less extreme gender gap than previously assumed. The researchers found women accounted for 23 percent instead 18 percent of adult Wikipedia users, and 16 percent instead of 13 percent of total users.
That's not a huge difference, but it's a crucial one, if you consider how the "13 percent" figure has proliferated over the last few years.
The statistic prompted the Wikimedia Foundation's executive director to set a goal to get more female contributors on the site—aiming for 25 percent by 2015.
It inspired some creative protest art. Developer Santiago Ortiz crunched data from the site's API to make a visualization of the Wikipedia editors gender gap, and found that the only article with more female editors than male was the page on cloth menstrual pads. (The articles on Teen Mom 2, and Pride and Prejudice the 1940 film were close runners-up.) Not exactly helpful for busting stereotypes.
It also informed the idea that rampant sexism runs wild in Wikipedia’s contributor community. This spring, novelist Amanda Filipacchi reignited the gender gap controversy when she noticed a trend of editors that found the need to point out when an author was a woman author, instead of simply an author. "The intention appears to be to create a list of 'American Novelists' on Wikipedia that is made up almost entirely of men,” she wrote in a column for the New York Times.
Wikipedia’s contributor culture is notorious for being intimidating, cliquey, abrasive, detail-obsessed and tech-nerdy. There are plenty of studies to back up the notion that that would turn women off. On the other hand, to a large extent the gender gap just reflects the society we live in. The OpEd Project found the same general 85-15 ratio of men to women in op-ed pages and reviews, to say nothing of the C-suite or the US Congress.
It's great to strive for equal representation, because sexism stinks and because a diverse community of users should make Wikipedia more accurate. Where the trouble comes in is suggesting the site's to blame for the gender gap. Wikipedia doesn’t select its editors, and users don’t know the gender of other contributors since most are anonymous.
Overhyping a problem doesn't mean it's not a problem, but it's valuable to distinguish fact from frenzy. That's what Wikipedia strives to be all about, after all: facts. This latest study takes a small but welcome step in that direction.