That was part of Hill & Shaw's assertion (see endnote in the original article). I'm not sure I'm entirely convinced.

I guess this page wouldn't be complete without a reference to WP:Clubhouse? An Exploration of Wikipedia’s Gender Imbalance, a 2011 paper which contains a lot of interesting data related to all of this.

Anonymous editor surveys asking the question: Are you male or female?

Heh. The large gender imbalance has been reported by the Wikimedia Foundation for years. The survey-based estimates (sources are the UNU survey and a WMF survey) are not corrected in the footnote (which is about which statistics to use to estimate the percentage of mothers). The gender imbalance of Reddit is cited to Huffpost, it's 72% male (which is less male-dominated than Wikipedia), and the most extreme of all the major social media sites listed there. There are multiple citations for effect on content, including New York Times, Atlantic and a recent Guardian editorial. 1 in 50 relates to survey respondents, not contributors (which some have claimed may have a *slightly* higher proportion, based on sampling bias). For participation dropping after age 20 see UNU survey (linked). WMF efforts to address the gender gap are well publicised, Sue Gardner talked about it to the press until she was blue in the face. Women aged 18-34 in Facebook and Pinterest: sources linked. The surveys were commissioned by the Wikimedia Foundation itself, and comparison to social media is relevant in relation to the argument that women have no time to be online. Relevance of anonymity on women's participation per quoted text from Wiley Handbook. User interface impact is a hypothesis, based on recent discussions on Wikipedia's Gender Task Force page.

If your post is representative of Wikipedians' ability to read sources, Wikipedia is not destined for greatness.

It's exactly the other way round, according to this Huffpost piece. There are 99 million more monthly female visitors to the various social media sites (included in the analysis were Facebook, Yelp, Twitter, Myspace, Pinterest, YouTube, LinkedIn, Flickr, last.fm, deviantART, Google+, Digg, Bebo, Reddit and many others). Even in gaming, the largest demographic are now adult women.

I do think paid crowdsourcing is the future. All the talk about "sharing" is hypocritical spin, given that Google and other scrapers are using Wikipedia content to make money from ads, while unpaid volunteers do all the work. See Wall Street's internet darlings require an endless supply of idiots – Sharing Economy? Mug Economy, more like.

In terms of social development, the internet currently compares to the darkest age of the industrial revolution. So, more power to you.

Many academics have said the same thing to me. No desire, and not enough time, to argue endlessly with nincompoops. There is currently an initiative underway, focused on medical articles, to get funding for experts to peer-review Wikipedia articles. Once an article is up to scratch, there would then be a permanent link to the peer-reviewed version displayed on the article page. This might be a more promising approach, and it could scale to other topic areas as well. Experts would (1) be paid, (2) have the guarantee that their work will have some permanence, (3) derive a degree of kudos from their having been appointed to do this work. Funding would, in this case, come from charities interested in making reliable medical info available online. Currently, for example, there is a Wikipedian-in-Residence at Cancer Research UK, who is working with CRUK experts on Wikipedia's articles on cancer. The position is funded by the Wellcome Trust.

Quite. I think the whole discussion about what turns women off once they're there addresses only the smaller half of the problem. The main question is, why don't women come to begin with.

If you take things on people's say-so, you end up with this. Reliability is bad as it is – looking at an article, you can never be sure, without checking the references, whether it is a bunch of nonsense or a well-researched, accurate article. But if you allow everyone – well-intentioned, knowledgeable people like yourself as well as pranksters and hoaxers – to add stuff without citations, the site would quickly be corrupted altogether. No one can tell if you are sincere or making stuff up out of whole cloth.

Kozierok's First Law: "The apparent accuracy of a Wikipedia article is inversely proportional to the depth of the reader's knowledge of the topic."

Yup. And of course, you are looking at the English Wikipedia, which is the most-developed language version of Wikipedia. Yet Wikipedia claims to be available in over 280 languages, when in many of them, coverage is really, really rudimentary. See e.g. the "Mind the zombies" slide from a recent Wikimania presentation – basically, only 125 language versions of Wikipedia have more than 5 editors. The others are, to all intents and purposes, dead.

Note also that even English Wikipedia contributor numbers (as opposed to reader numbers, which are immense) are really quite small. (Someone else has pointed this out above.) If you look at this table, you'll see that there are only about 3,000 regular editors in the English Wikipedia, i.e. people who make more than 100 edits a month (i.e. about three a day). That number has shrunk considerably over the past few years, from a March 2007 high of 4785. At the same time, of course, the number of articles continues to increase constantly (now at 4.6 million). There are fewer contributors, and more articles to be watched over.

So Wikipedia has many articles that it does not have the (wo)manpower to curate adequately. In the early days, of course, everyone thought that "eventually" all these articles that someone started would become little masterpieces, but it's becoming clear that this will not happen. Little-watched biographies in particular are a problem, as the only people interested in them are usually the subjects and/or people who hate them for some reason, so they turn either into puff-pieces or hatchet jobs, with no one really noticing (there are well over half a million articles that no one has on their watchlist). Yet they are the top search hit when someone Googles the name online.

This reminds me of Newslines.org, a news-based crowdsourcing project that overlaps to a certain extent with Wikipedia, with the difference that they *do* pay their contributors. They report that their gender split is reversed: they have more women contributors than men, and also have more contributors from ethnic minorities than Wikipedia (in fact, their two leading contributors are black women).

The comparison is relevant in response to the argument that women simply don't have time to spend online and edit Wikipedia. They clearly do have time to spend online, but are spending it elsewhere.

Yes indeed. Well done for pointing it out. While Wikipedia is a top-ten website, the vast majority of people using it never edit it. With men, it's a minuscule proportion, and with women an even more minuscule proportion. Still, the disparity has an effect on the content.

If you don't get the "surely", imagine an encyclopedia where 90% of the writers were women. Do you think it would be as good as an encyclopedia could be?

Incidentally, since you mention it, gays are well represented on Wikipedia, I think. African-Americans on the other hand are poorly represented, and you can tell from some of the content in related topic areas. The hair straightener hoax described here for example probably wouldn't have succeeded if there weren't a dearth of Black editors.

We're not talking about any occupation or hobby, but about the web's primary reference site. No one has a problem if football or knitting forums have an unequal gender balance, but Wikipedia's coverage ends up lopsided if one half of humanity is barely there.