Thoughts on the online encyclopedia’s gender imbalance
By Andreas Kolbe, with contributions from Nathalie Collida
Wikipedia is notorious for having a sizeable gender imbalance. As Guardian feature writer Anne Perkins put it – somewhat cattily – in a recent article, the site is dominated by “young white western males with a slight personality defect”. Most of them are single and childless. Indeed, as we shall shortly see, many are still children themselves.
The 2010 United Nations University survey was the largest survey examining Wikipedia demographics to date. It had a total of 176,192 respondents, of which almost 60,000 identified as current (or former) Wikipedia contributors; the remaining respondents were Wikipedia readers. The survey reported that –
Overall, the average age of the Wikipedians that participated in the survey is 25.22 years. Half of the respondents are younger than 22 years. The most frequent age that can be observed within the respondents is 18 years. Splitting the respondents in four equally large age groups shows that 25% are younger than 18 years old, 25% are between 18 and 22, a further 25% are between 22 and 30 (e.g. half of the respondents are between 18 and 30 years) and the remaining 25% are between 30 and 85 years old. There is a slight age difference between readers and contributors – readers are, on average, 24.79 years old while contributors show an average age of 26.14 years. Finally, female respondents are younger (23.79 years) than male ones (25.69 years). […]
Contributors show a substantially larger share of males than readers. Among respondents only 12.64% of contributors are female.
This gender imbalance has long vexed the Wikimedia Foundation. In January 2011, in response to a New York Times article by Noam Cohen, the Foundation created a mailing list devoted to discussing the issue. Sue Gardner, the Wikimedia Foundation’s Executive Director at the time, set an ambitious and much-publicised goal to increase female participation to 25% by 2015. Yet the results of the very next survey, conducted by the Wikimedia Foundation a few months later in 2011, were anything but encouraging:
Our editing community continues to suffer from a lack of women editors. The survey provided an even starker view of this than previous studies (only 8.5% of editors are women). It is a strategic priority to address this imbalance.
“I didn’t solve it. We didn’t solve it. The Wikimedia Foundation didn’t solve it. The solution won’t come from the Wikimedia Foundation.”
Wikipedia editor surveys: less than 1 in 50 respondents is a mother
Gardner has since left her position as Wikimedia Executive Director, to be succeeded by Lila Tretikov. Media coverage continues to testify to the sometimes dire effects the gender gap has on the curation of the site’s content, the most recent example being a Guardian editorial inviting readers to –
Compare the coverage of female porn stars, where a page that went up first in 2004 has been edited over 3,000 times by more than a hundred volunteers determined to make it as copiously referenced as possible, with that of “Female writers” which has no quality control at all …
It is sometimes argued that women simply have less time to contribute to Wikipedia, due to family commitments. This is a fallacy. Firstly, the United Nations University survey found that only 33.29% of respondents had a partner, and only 14.72% had children. The difference between readers and contributors was negligible here, and the survey report’s analysis indicated only a minor difference in parenthood percentages for male (15.1%) and female (13.7%) respondents. It is patently obvious that girls and women in the age groups that are most strongly represented in Wikipedia’s demographics typically do not yet have families of their own. Their lack of participation is unrelated to their being bogged down by family responsibilities.
Of course, if we accept the data from these two surveys as an accurate reflection of Wikipedia demographics, they also tell us something else: if only 13.7% of female contributors have children, and the percentage of female contributors lies somewhere between 8.5% and 12.64%, then only 1.16%–1.73% of Wikipedia contributors are mothers.
That is less than 1 in 50 (or at best about 1 in 25, assuming the data need to be corrected for sampling bias; see Author’s note below).
“She’s just not that into you.”
Belying the theory that women – especially women with families – do not have enough time to engage online, there are in fact plenty of social media sites where women dominate. The Handbook of Language, Gender, and Sexuality (Wiley, 2014) notes:
Recently, women have come to outnumber men in some social media domains. They use social network sites such as Facebook more often and more actively than men (Brenner 2012), and female users predominate on the microblogging site Twitter, the consumer review site Yelp, and the online pinboard Pinterest. More males, in contrast, frequent music-sharing sites such as last.fm, as well as Reddit, a social news website known for its sometimes misogynistic content (HuffPost Women 2012; Williams 2012); contributors to Wikipedia are also overwhelmingly male (Lam et al. 2011). Moreover, the professional social network site LinkedIn has attracted almost twice as many males as females. LinkedIn representatives claim that this is because men are better at professional networking than women, at least in some industries (Berkow 2011), whereas women have traditionally focused on maintaining relationships (Fallows 2005; cf. Tannen 1990). Women’s greater concerns about privacy and identity disclosure on social network sites (Fogel and Nehmad 2009) may also predispose them to interact with individuals they already know and trust (Muscanell and Guadagno 2012), which Facebook and other social network sites facilitate through features such as “friending.”
Crocco, Cramer, and Meier (2008) argue that the move toward web-based computing has had an equalizing effect on gendered technology use. If equality is defined as equal in principle access, women in the United States have caught up with men. At the same time, the web is becoming increasingly specialized by gender. Although many sites are male-dominated, women today have more choices of online environments than they did in the past, including social media sites in which they can exercise a degree of control over who reads and comments on their contributions. As discussed further below, users of these social media sites tend to be less anonymous than in earlier text-based forums.
This kind of analysis takes us much closer to the real reasons why women simply go elsewhere online. Women engaging online appear to place more importance than men on spending time with people congenial to them, and prefer to avoid people who are not. They also like to form more meaningful personal relationships than men.
There is anecdotal evidence from Wikipedia to support this. One take-away from my visit to Wikimania 2014 was that two people told me, based on their longstanding experience as members of the English Wikipedia’s Arbitration Committee, that women object more strenuously to sockpuppeting than men, and for different reasons: men condemn it because it corrupts the process of arriving at a neutral Wikipedia article, whereas women feel it is a personal breach of trust if the same person uses several identities to talk to them.
Friends? On Wikipedia, they’re known as “meatpuppets”
In general, Wikipedia is quite hostile to social relationships. First of all, anonymity is a paramount value to Wikipedians (as it is to Redditors), and this is not conducive to deeper relationships. In fact, forming relationships is actively frowned upon in some ways on Wikipedia, and a cause for distrust.
Wikipedia has draconian rules against what it calls “canvassing” (telling friends about a discussion that you are engaged in) and “meatpuppetry” (having friends help you edit articles in ways that support your point of view). Of course there are sound reasons for this – there have been many, many cases in Wikipedia’s history of people acting in secret concert in order to bias articles and discussions. But the unfortunate side effect is that it makes people reluctant to be open about their friendships and allegiances on Wikipedia. It makes them susceptible to attacks.
As for avoiding people who aren’t congenial, Wikipedia’s very nature and status as the web’s most prominent reference site make its articles and sociology somewhat comparable to waterholes in the animal world – they attract species of editors with opposing agendas who have to somehow coexist, despite the tensions between them, in order to access the social resource that Wikipedia represents to them. It’s stressful. Writing on any mildly contentious topic in Wikipedia both women and men are practically bound to come up against the very sort of people whom they might most avoid associating with in their private lives. On Facebook or Twitter, it is easy to manage such situations. You block or unfriend them, and not only will you never hear from them again, they will also not be able to keep tabs on what you write.
In short, thanks to its well-established culture of contentiousness, the deck is stacked in many ways against equal gender participation on Wikipedia.
The social dimension
If we look at the various sites mentioned in the above passage from The Handbook of Language, Gender, and Sexuality, it’s worthwhile to note the social component of sites where women predominate.
Women on Twitter, Facebook and Yelp share personal images with friends and engage actively in campaigns for what they feel are worthwhile causes. The growing political influence of Black Twitter, for example, is to a large extent driven by black women (who are doubly underrepresented on mostly-white Wikipedia). Campaigning is something the software of these sites is specifically designed to facilitate, using mechanisms such as hashtags, shares and likes.
The only parallel measures in the Wikipedia world that have had some success in drawing women editors in are women-themed edit-a-thons and “Wikibomb” events that emphasise the communal aspects of contributing (it is surely no coincidence that these are physical meetings of like-minded people acting without the cover of anonymity) and on-wiki projects like the Gender Gap Task Force or WikiProject Feminism.
In these cases the gender gap itself becomes the catalyst that brings women together in a shared social endeavour. But is that enough? It seems unlikely. Needless to say, such initiatives also breed resentment among other editors, and one can find comments from Wikipedians on Reddit such as the following:
If you really wanna see something ‘nasty’, check out the ‘article alerts’ on that page [WikiProject Feminism]. Anytime there is a RfC (Requests for comment during a content dispute, to get wider input) or an AfD (articles for deletion, to discuss whether an article should be deleted) or a CfD (Category for discussion to decide whether a cat should remain or not), they will -always- vote the shit out of the women ones, making sure that they follow a gyrocentric [sic] pov [point of view] on the RfC and to keep the women’s articles. It’s blatant canvassing but nobody will do anything about it.
Looks that matter
Quite apart from anonymity and the patterns of social interaction, the sites where men are most dominant – Wikipedia and Reddit – are on the whole very dry and text-based. Male-dominated sites like Reddit and Wikipedia demonstrate that men don’t seem to care much about visual aesthetics if there is function. The sites where women predominate look quite different from Wikipedia. Pinterest is full of gorgeous, nourishing images uploaded by contributors. So is Yelp. A large part of social activity on Twitter and Facebook revolves around sharing images with friends.
The Wikipedia interface is increasingly coming to be seen as old-fashioned and visually unattractive. This has reached the point where third-party provider WikiWand has been able to raise $600,000 from an investor to create a new site that presents Wikipedia content with a more pleasing interface, a development that seems to be viewed with some concern by the Wikimedia Foundation. Jan-Bart de Vreede, chairman of the Wikimedia Foundation board, recently said,
Truth is, we are at a crossroads, and unfortunately have been for quite some time. Blaming each other for being there does not make much sense, as it would probably result in us spending more time at that crossroads. If you want me to take part of the blame, I will. Other internet projects (not limiting ourselves to websites) are passing us by left and right, and none of them have the non-profit goals that we have. In fact, some of them, with more commercial propositions, are actively undermining us.
When asked which sites he meant, he pointed to Quora, Facebook and – WikiWand. He later attempted to downplay his remark, but it stands to reason that if readers flock to the more attractive interface of third-party mirror sites like WikiWand, this will not just affect wikipedia.org’s pageviews and Alexa ranking – it will also sharply reduce the number of people who will see the Foundation’s fundraising banners, and will turn Wikipedia into a mere consumable with an even smaller reader-editor conversion rate (WikiWand pages do have an “Edit” hyperlink that takes users to wikipedia.org, but it is far less prominent and quite hard to find).
Modernising the Wikipedia interface
The efforts the Wikimedia Foundation has made to date to modernise the Wikipedia interface have been hampered by poor implementation and caused a major crisis in relations between the Foundation’s paid staff and the community of unpaid contributors. Last year’s VisualEditor was full of serious bugs, and unable to handle the precise situations where a WYSIWYG editor might have been most useful. It was received so poorly that a volunteer administrator simply switched it off, relegating it to an opt-in feature that now seems to see little use. The Foundation’s most recent new software feature, the Media Viewer, was so strongly opposed in the two most important language versions of Wikipedia (English and German) that Wikimedia Foundation Deputy Director Erik Möller was first hauled in front of the English Wikipedia’s arbitration committee (for threatening a volunteer administrator) and then had to introduce a new “superprotect” feature to prevent volunteer administrators on the German Wikipedia from simply switching that new software feature off, too (see our previous coverage). As a result, the community is in uproar, and a petition has been launched on-wiki and on change.org asking the Wikimedia Foundation to undo its move. At the time of writing, it had garnered well over 600 signatures, most of them from long-standing and committed Wikipedians.
Judging by her comments to date, Executive Director Lila Tretikov intends to stand firm, and would rather lose part of the volunteer community than back down on this issue. The problem is that large parts of the unpaid volunteer community consider the work the Foundation employees have delivered to date to be inept, and historical evidence is on their side. Tretikov will have to improve the performance of her software engineering and product development departments considerably to get the community behind her again. She is new to the job, and while she may well have the background and stamina to pull it off (she was previously Vice-President of Engineering at SugarCRM, Inc.), this will not be achieved without pain. There is little doubt that a more visually enticing and professional interface would go some way towards making Wikipedia a less unappealing site to women.
But for Wikipedia to actually become a platform fully embraced by women, it would have to change its culture in fundamental ways, reducing its emphasis on anonymity and providing more opportunities for meaningful companionship and satisfying social relationships between its contributors. Failing that, women will simply continue to vote with their feet, and find their enjoyment and altruistic fulfilment elsewhere.
Ms Marielle Volz, a Wikimedia Foundation intern, has advised me that Shaw & Hill (2013) have argued that women and parents are less likely to participate in surveys; on this basis, they have re-interpreted the UNU survey results and estimated that true figures for female contributors and parents would be closer to 16.1% and 25.3% respectively. This would place the percentage of mothers among Wikipedia contributors at around 4.1%, i.e. roughly one 1 in 25. It should be noted that Hill, one of the authors of the study, declared that he was an unpaid member of the advisory board of the Wikimedia Foundation. I am indebted to Ms Volz for bringing the Shaw & Hill paper to my attention, and for pointing me to the gender breakdown of parenthood in the UNU analysis that is now reflected in the text above. Our discussion, which took place on the Wikimedia Gendergap mailing list, can be found here.
Image credits: Flickr/mikeedesign, Wikimedia Commons/Walton LaVonda, Wikimedia Commons/mistermundo/derivative work by Saibo ~ licensed under Creative Commons Attribution 2.0 Generic (image 2 public domain)