Why this Site?

  • Our Mission:
  • We exist to shine the light of scrutiny into the dark crevices of Wikipedia and its related projects; to examine the corruption there, along with its structural flaws; and to inoculate the unsuspecting public against the torrent of misinformation, defamation, and general nonsense that issues forth from one of the world’s most frequently visited websites, the “encyclopedia that anyone can edit.”
  • How you can participate:
  •  Visit the Wikipediocracy Forum, a candid exchange of views between Wikipedia editors, administrators, critics, proponents, and the general public.
  • 'Like' our Wikipediocracy page on Facebook.
  •  Follow Wikipediocracy on Twitter!

Press Releases

  • Please click here for recent Wikipediocracy press releases.

Single white males: Systemic bias in Wikipedia’s obsessions

By Johnny Au

Wikipedia competes with Google for the title of most popular reference website on the Internet. However, despite striving for a “neutral point of view“, its coverage of topics is subject to systemic bias. In other words, much of Wikipedia focuses on topics that are of particular interest to young white males. In the case of the English Wikipedia, by far the most popular edition of the “free” language encyclopedias, it is not surprising that Anglophone media, locales, personalities, and events dominate it.

What systemic bias?

According to Wikipedia’s project page on Systemic Bias, an “average” editor on the English Wikipedia is a young white Anglophone male who is technically inclined, formally educated from a primarily Christian or secular country located in the Global North, and more likely to work in the service sector if not still enrolled in school.

Most noticeably underrepresented on the English Wikipedia are women. Although women about equally read Wikipedia along with men, given that only 10–15% of Wikipedia editors are female, there exists a substantial gender gap in content creation and curation. Indeed, female Wikipedians who give editing a try are more likely than their male counterparts to leave Wikipedia for good. Topics of greater interest to girls and women tend (on Wikipedia) to garner shorter articles with less detail than articles of interest to boys and men. For better or worse, commonly mentioned topics that may be of more interest to women include women’s rights, women in business and science, fashion, celebrities, home decor, family life, and cosmetics. Women also generally perceive Wikipedia to be of a lower quality than men do.

Given that Wikipedia is a website on the Internet, those without reliable Internet connections simply cannot contribute to it. Those generally include those who lack financial means in the Global North, those who live in the Global South, those with disabilities, and those who are over 60 years of age. Non-white minorities are also less likely to edit Wikipedia, though model minorities (such as British Jews or Japanese-Americans) are just as likely to edit as their white majority counterparts would. Not just that, but the general Internet population would rather gravitate toward easy-to-use sites like Facebook or Twitter than to learn clunky Wiki markup text. Although Wikipedia recently implemented VisualEditor, which is a WYSIWYG editing interface, it is beta software with many reported flaws. With no easily-learned editing tool in good working order, it’s no wonder Wikipedians tend to be more technically inclined than the average Internet user.

Many Anglophone Wikipedians tend to use mainstream English-language media to back up their claims, given that many are either unable to fluently read any foreign language source or uninterested in trying to translate them. Given that, many Anglophone Wikipedians dismiss non-English sources as unreliable, despite them being trustworthy and able to provide a much-needed non-English viewpoint. As a result, topics that pertain to the world outside the Anglosphere tend to be underrepresented on Wikipedia. Yet, the English Wikipedia does have editors from countries where English is a second language, either officially or practically. Examples of such jurisdictions where English is officially used include Hong Kong, India, and Pakistan; and English is additionally spoken commonly as a second language in the Netherlands, Germany, and other western and northern European nations. Unfortunately, given that English is a non-dominant language for these editors, their English may not be as strong as an Anglophone’s, which may discourage them from editing, fearing ridicule and misunderstanding from Anglophones.

Another source of bias is available leisure time. Those with ample free time are more likely to edit Wikipedia, especially given that editing Wikipedia articles is predominantly done as volunteer work. This means that those who work long shifts or lead busy family lifestyles are less represented on Wikipedia.

What exactly is overrepresented?

The English Wikipedia has much more coverage on topics such as speculative fiction, animation, video games, politicians (especially American and British), celebrities, and roads, highways, railroads, and related infrastructure (such as train stations) in the Anglosphere (especially North America and the United Kingdom). Cruft describes something otherwise minor that is written in great detail. There is no shortage of cruft on Wikipedia. For example, more has been written about The Simpsons (right down to many articles on individual episodes being awarded “Good Article” status) than about all of John Milton’s works (including Paradise Lost) combined. Not just that, the English Wikipedia has much greater coverage of the works and their adaptations of either English fantasy writer Tolkien or Martin than on all of the Four Great Classical Novels of China and their adaptations combined. Apparently, recent Anglophone media is of greater interest than pre-20th century media, especially non-English media. Even general interest articles often have an “In popular culture” section, which unsurprisingly, tends to feature many examples from Anglophone media. No doubt you will likely encounter a random reference to The Simpsons, Family Guy, or other popular American animated sitcoms where you least expect, and those references can be mere passing mentions in those respective television programs. Wikipedia articles about historical figures might have a reference to a game in the Civilization series, for example, such as the barely-important fact that Hiawatha of the Iroquois appears in Civilization V. Image
Apparently of greater interest than Paradise Lost
Sun Wukong from Journey to the West, one of the Four Great Classical Novels of China; a certain American television show featuring a cartoonishly yellow-skinned family is more popular than this in the great encyclopedia of knowledge
Zhuge Liang from Romance of the Three Kingdoms, another one of the Four Great Classical Novels of China; once again, The Simpsons is more popular than this in Wikipedia

A large number of featured articles (the highest honor bestowed on content) on Wikipedia discuss freeways in North America. For example, here are the lists of featured articles on Canadian highways and American highways. Of course, those who don’t use North American highways would have little interest in these Wikipedia articles. Likewise, there are many articles on various train stations, including run-of-the-mill abandoned stations, as well as LRT stations that are nothing more than a roof on a concrete platform.

Highway 401 in Ontario is a featured article.

No doubt coverage of deaths in the Global North rises to greater interest among Anglophone Wikipedians. There is much more coverage about Al Qaeda attacks in the United States, in the United Kingdom, in Spain, and in France, taking over 3,000 lives in those four countries, than there is about the Darfur conflict in Sudan, which had over 400,000 casualties.

Controversial figures and issues in the Anglosphere, even those who are otherwise minor, tend to have much greater coverage than some major non-controversial non-Anglophone figures and issues. For example, much has been written about the Gamergate controversy, as compared to laws regarding pachinko parlours in Japan; the former is of little interest to people outside the Anglosphere (including Japan, the country that produces the second-most video game titles in the world), while the latter is of little interest to people within the Anglosphere.

Recentism is also a major biasing factor in Wikipedia. If it happened on Facebook, it’s more important than something that happened somewhere in the tropics. Much more has been written, for example, about the Ice Bucket Challenge than about the former Prime Minister of Rwanda, Bernard Mazuka.

Even within the Anglosphere, there is systemic bias. The article alcohol advertising lacks examples from Canada, while on the flip side, the article on Parclo (partial cloverleaf interchange) emphasizes Canadian examples.

Thus, topics of great interest to young white Anglophone males have come to dominate the English Wikipedia.

So what?

Unlike intentional vandalism, which is often blatant and thus might be corrected quickly, even with a bot such as ClueBot NG, systemic bias can be very subtle and remain undetected—or not cared about—for years on end. Humans are naturally biased and generally write about what they know. Thus, it is difficult for most Wikipedians to write to address a neutral point of view. Systemic bias is very damaging to Wikipedia in the long term as a result, especially given that the English edition is the most popular version and is more likely to use the perspective of an average Anglophone. Systemic bias of the sort described wouldn’t be a concern if only single white English-speaking males with ample personal time comprised the globe’s population. However, we know that the world’s population is only fractionally made up of young white Anglophone males.

As former Wikimedia Foundation chairperson Michael Snow and Wikipedia co-founder Jimmy Wales said in an open letter:

“How can we build on our success to overcome the challenges that lie ahead? Less than a fifth of the world’s population has access to the Internet. While hundreds of thousands of volunteers have contributed to Wikimedia projects today, they are not fully representative of the diversity of the world. Many choices lie ahead as we work to build a world wide movement to create and share free knowledge.”

What can be done about systemic bias?

Fortunately, there are some ways that might counteract systemic bias. It is said that when you walk in another person’s shoes, you gain the perspective of that person. Therefore, it is important for Wikipedians to actively learn about the perspectives held and issues faced by non-Anglophones—especially women, the elderly, racial minorities, the economically disadvantaged, and the disabled. Understand your own biases, and don’t attack those who are underrepresented on Wikipedia. They too are trying to help improve the encyclopedia, even if it may not be obvious to you, and if they disagree with you. You can learn a foreign language or at least be more open to considering non-English media sources as references in Wikipedia. Even foreign language editions of Wikipedia should be consulted for different perspectives on universal subjects. You can also join the Countering Systemic Bias Wikiproject. Better yet, you can recruit those who are underrepresented on Wikipedia to join in, if you believe that more volunteer labor added to this system will benefit humankind.

Only by countering systemic bias can we make Wikipedia a truly free encyclopedia, free to all, regardless of one’s background and interests.



Image credits: Wikimedia, “Simpsons Family Picture” by 20th Century Fox Film Corp. ~ Licensed under Fair use via Wikipedia

17 comments to Single white males: Systemic bias in Wikipedia’s obsessions

  • Radiant Orchid

    “It is said that when you walk in another person’s shoes, you gain the perspective of that person.” I would like to give my full support to this idea – where do I send my shoes?

  • The fault lies in ourselves, not our stars. Travel outside your borders and your expectations. Listen. Take in what is being said, and study upon it.

    Or, just ban and block.

  • Observer

    You fail to acknowledge the Wikipedias written in other languages: http://meta.wikimedia.org/wiki/List_of_Wikipedias. Going by the active users numbers some of the “top” Wikipedias have clearly just inflated their article counts and probably don’t have much real content, but the Spanish, Portuguese, Arabic, Persian, Hindi, Chinese, Vietnamese, etc. Wikipedias all have hundreds of thousands real articles and cover a major chunk of the third world. Perhaps your post is more a reflection on how prone people are to project themselves on others.

    • Hersch

      The existence of Wikipedias in other languages does little to mitigate the bias problem in the English Wikipedia.

  • For a Wikipedia “Believer”, Mr. Au did a very fine job with this blog post. I hope we see more.

  • no

    What you are crying about can be more easily summed up by WP:IDONTLIKEIT.

    • WP:ABC verbiage is senseless most of the time, used as a cudgel the rest of the time to stifle dissent, and roundly ridiculed outside Wikipedia.
      Don’t bother to use it here.

      • Teratornis

        Is your use of a WP: abbreviation to deride the use of WP: abbreviations an instance of intentional or unintentional humor? Either way it was funny, but sad at the same time to evade no’s argument. Clearly the author does not like systemic bias on Wikipedia, but fails to explain how bad the problem is compared to other problems the Wikipedia editor faces (such as figuring out how to write something that doesn’t get deleted, finding reliable sources for obscure topics, grappling with inefficient editing of reference templates, negotiating content disputes with people who reason fallaciously, etc.). Given the stupendous success of Wikipedia, it’s hard to see Wikipedia’s unequal success at soliciting labor donations from different demographic groups as constituting any sort of threat to Wikipedia itself, at least in the foreseeable future. Wikipedia is not going away, no matter how excessive the highway articles become.

        If some demographic groups have less interest in building a free encyclopedia catering to their interests, what can some other group do about their lack of generosity? Imagine a world in which young white males had no interest in donating their labor to Wikipedia – what could, say, elderly Pacific Islander women do about that? Wikipedia has never placed the burden of recruitment on editors. Some people look at the site and just take it upon themselves to contribute, doing the hard work to learn the rules and chip through all the technical and cultural obstacles we throw in their path. We’ve never had to “sell” people on the idea of contributing, and we have no idea of how to do it. Thus even if systemic bias constitutes some sort of a problem, which ranks significantly among our long list of real problems, at best we can only complain about it (i.e., say I don’t like it). If anybody knew how to solve it, they would have.

    • Vigilant

      If this were wikipedia, we’d ban you for posting anonymously to avoid scrutiny and then spend a bunch of time looking for your main account to ban that.

      Here, you just get ridiculed for being a prat.

  • metasonix

    Better yet, try WP:GOTOALIBRARYDAMMIT.

  • Eagle

    Mr. Au offers an interesting analysis. The forum thread that he created on overrepresented content topics was far more revealing than the few examples cited in his blog post. There are two different problems with overrrepresented content: 1) that energy is diverted from created adequate content on topics which most people agree should be represented in an encyclopedia and 2) that amateur editors have taken the overrepresented content to such an extreme that it cannot be adequately currated by Wikipedia’s crowdsourcing techniques. Wikipedia is guilty of both failings. More importantly, Wikipedia’s power elite actively ban or chase away editors who seek to fix this serious problem.

    • Teratornis

      Of the two problems you list for overrepresented content, the first seems suspect in two ways. Firstly, for “energy” to have been “diverted” from some other use, it must in some sense be available for that other use. But given that Wikipedia relies on donated labor, it can only get the labor that volunteers are willing to donate. If someone has a particular interest in, say, highways, that’s what the person will write about. The person does not have “energy” to write about something of no interest to him or her. That is, there is no “energy”, only interests, and Wikipedia has no control over what interests people. To get people to do things they find uninteresting generally requires paying them, since everyone is interested in money. Secondly, if there are “topics which most people agree should be represented in an encyclopedia” then “most people” are free to donate their time to write about them. But again, since Wikipedia relies on donated labor, it doesn’t matter what “most people” want in an encyclopedia. It only matters what most people who are inclined to donate their labor want. Lower coverage of other topics does not so much reflect “systemic bias” by Wikipedia, but rather systemic bias by people who know and care about those topics against sharing their knowledge for free – and being willing to overcome the many unavoidable barriers against sharing knowledge on Wikipedia (such as the huge investment in learning Wikipedia’s complex rules for content).

      Thus instead of complaining about the generosity of young white males who donate their labor to Wikipedia, let’s complain about the stinginess of all the other groups of people who donate less. But since only a tiny minority of even the most represented group donates labor to Wikipedia, one could argue that the interests of young white males are inadequately covered here, compared to the depth of coverage we could have.

  • Ritchie333

    Everyone should be aware of the systemic bias on Wikipedia, especially those who edit Wikipedia a lot. Long-term editors and admins can get so caught up in what they do that they gain a false sense of self-importance, not realising that they may never read important criticism and feedback. Most casual readers do not bother to comment that a crap and biased article is crap and biased, they just vote with their feet and go elsewhere.

  • Vejvančický

    “… their English may not be as strong as an Anglophone’s, which may discourage them from editing, fearing ridicule and misunderstanding from Anglophones.” – Czech is my first language. 2015 is my eighth year on en:wiki, I’m an administrator and all that blah blah. I still feel that for some people (no matter how good or bad people) I’m not a partner to discussion, just because of my “Czenglish”. I understand, to some extent. It is quite possible that I would treat with despect and distrust someone who tries to write encyclopedia(!) in Czech without perfect knowledge of that language.

  • g-host

    Notability is a huge source of bias. It’s a subjective measure that gets voted on by the largely white male editors, and it compounds all the other problems.

  • anon

    OH YOU GOTTA BE FUCKING KIDDING ME! The Simpsons is one of the longest running television series of all time, of course it would have a lot of coverage.

  • Anon

    There’s also the problem of bias against red links, and their ‘likelihood’ of being created – their removal compounds the issues of bias in Wikipedia. It seems that (white male) notability has to be proved even before a red link can be created in an article.