Wikipedia and its ‘undue weight’ policy

Like most people, I find Wikipedia to be quite useful. It has grown on me over time and I have donated to their fund drives as I think it serves a valuable function and is also a great experiment in open-source knowledge and the wisdom of the crowds. I have found that on topics that I know something about, it has often been quite good. I generally tend to use it for quick and dirty searches of information that I think has a good chance of being correct, such as names and dates and places of events, or for things where I am looking for an overview and not too concerned with a high level of detailed accuracy. I have even on occasion made edits to entries that I thought were incorrect or incomplete and those have remained.

But we know that Wikipedia is not entirely a free-for-all. There is some quality control that goes on behind the scenes and I was curious as to how it works. A recent article in The Chronicle of Higher Education sheds some light on how disagreements are resolved. This account of a scholar’s frustration in trying (and failing) to correct the entry about the 1886 Chicago Haymarket riot and trial, a topic that he had studied for 10 years, shows the difficulties you can encounter when you try to challenge conventional wisdom or entrenched myths within the pages of Wikipedia.

The problem is that important new scholarly work tends to challenge the established consensus and can take quite a long time to become the new consensus. Wikipedia is basically meant to reflect current consensus views on a topic. It is not a scholarly journal and does not work on the peer review model of publication. Hence when it comes to scholarly disputes, journals will always be slightly ahead of the times while Wikipedia will always be slightly behind the times. That is because of their basic models of operation.

Hence one can sympathize with both parties in this case.


  1. Pen says

    I have done a bit of work on Wikipedia in similar circumstances and have not had any bad experiences. I have had good luck using the ‘Principle of Least Astonishment’ and staying aware of any background emotional or political issues that may be motivating other writers or readers. It does sound from the article as if this scholar is a beginner with Wikipedia and has let himself be too easily intimidated. The people he is interacting with are actually the ones who are a bit out of line. They do not ‘own’ the page and seem to misusing Wikipedia’s community rules. But it often happens that pages are maintained by people who care deeply about the subject on social and political levels, and find change in the data threatening. It’s sad, but ‘real’ encyclopedias have their biases too. This scholar should indeed go and read the Wikipedia documentation so that he knows the various recourses for conflict resolution. He hasn’t used them all yet by any means.

  2. 'Tis Himself, OM says

    I have written two wikipedia articles on economics and edited a couple more. Once one of my edits was challenged. I noted that (a) I gave a source (and I would be able to give several more) and (b) the challenger’s source did not say what he claimed it said.

    If there’s a question on a wiki edit, discuss it on the talk page. Sooner or later a consensus will be reached.

  3. Tony says

    I’ve long been curious why many people think Wikipedia is unreliable. Like you Mano, I’ve found that those entries I’m knowledgeable on are frequently accurate. Do you, or anyone else, know why it’s considered unreliable by many?

  4. Mano Singham says

    I admit that I too was more skeptical of Wikipedia at the beginning, mainly because the idea of entries by anonymous people that could be changed overnight did not inspire much confidence. In the early days, there were a few highly publicized cases of hoax entries about high profile people. But as time went by, I realized that it was not such a free-for-all as I had thought and that there were people taking steps to try and ensure that the information is reliable. But I think that for some, that early image of a wild and woolly frontier where people can write anything still remains.

  5. Jared A says

    Oh, wikipedia. I find it to be pretty good as an encyclopedia, too. I think Prof. Messer-Kruse was being a bit naughty, even if he is “right”. Still, I get where he’s coming from.

    Here’s a hilarious, non-illustrative anecdote:

    Once I was using wikipedia to help me sort of cheat on a crossword puzzle. The clue was about a circus that was founded in, I think, 1886.

    I found a circus that fit the clue, but the wikipedia article said it was founded in “188”. Clearly a typo. I checked the history and a previous edit had accidentally partially deleted 1886 in a larger edit, meaning I couldn’t just revert. So I fixed it to 1886. A few hours later I got a very condescending note thanking me for my contribution but since I didn’t cite any sources they were returning it to the previous form. It made me laugh quite a bit. I don’t recall what circus it was, but I hope someone fixed it so some unfortunate middle school student doesn’t write about a Roman circus that survived into modern times.

  6. jamessweet says

    A few years ago I spent a lot of time “behind the scenes” at Wikipedia, working on resolving disputes, cleaning up articles, etc. One of the things you learn is to be able to “smell” how accurate* an article is likely to be. You can tell when an article has been a recent battleground, and often can get a good feel for how well that battle has been resolved — even without checking the Talk page. (And when in doubt, of course, please always check the Talk page, where you can often get a lot of background on any controversies)

    You can also “smell” pages that have not yet been battled over, but which were written with a clear bias. There’s a certain “Rah rah team!” aesthetic that will have been purged from any article that has gotten serious attention. Those articles are rarer than you might think, since there are factions within Wikipedia working very hard to clear all of that drek out.

    In short, if you know how to “sniff” the articles, Wikipedia can be an extremely reliable source. Truth be told, I cannot think of a single source than I find more trustworthy across such a broad range of topics. News media? HAH! Turning to a scholarly journal or to a library leaves one with the task of having to vet which information is good and which isn’t, a task which has at least been nominally done for you with Wikipedia. There are specialized sources that I trust for certain topics more than Wikipedia, perhaps… but as a broad source of general knowledge, there is simply nothing that beats it. If I’m wrong, name something.

    * I use the term “accuracy” here with the caveat that is well expressed in Mano’s penultimate paragraph. The purpose of Wikipedia is not to present the truth per se; it is to present the current mainstream consensus. Note that this is not the same as “tell both sides” journalism; Wikipedia is generally unambiguous in its dismissal of alternative medicine, for example. However, Wikipedia does not exist to peer-review new research, or to synthesize new conclusions; it exists to present the current mainstream consensus. There is quite a bit of value in this approach, but of course it is an important caveat to be aware of.

  7. jamessweet says

    Heh. That’s an unusual case, somebody was being sloppy. There are semi-automated tools that some people use to police large amounts of recent edits at the same time. Somebody spotted a date change with no citation, and unthinkingly hit a button that — with a single press — reverted the edit with an explanatory comment, and generated an automated message to your page. The person didn’t bother to look for even an instant to sanity-check whether the reversion made sense. Definitely sloppy.

    If you had written the person back, they surely would have fixed it. But I understand that’s a pain in the butt. That’s actually a problem that Wikipedia is very much struggling with: Edits from new users tend to be viewed with suspicion, especially when they may not yet know how to appropriately comment their revisions, etc., so that it communicates what they changed and why in a way that Wikipedians will understand; but on the flip-side, the easier you can make it for a new user to make an edit and see it remain, the more people you can hook. It’s a very fine line to walk, and frankly there is not really a best solution. Countless people have been turned off to Wikipedia by having their first edit inappropriately reverted; but there are countless more whose first edit was erroneous or even malicious. And even if you could perfectly sort the wheat from the chaff, what about the person whose first edit is made in good faith but is erroneous? They will be turned off from making further edits, but at the same time you can’t just let their factually incorrect information stand, can you? It really is an impossible problem.

  8. Henry Gale says

    Sometimes, reading the Talk page of an article is more interesting than the article itself.

    That said, I suggested to my nephew that when doing research on some topic Wikipedia is an excellent starting point. However, Wikipedia should never be the last source consulted.

  9. P Smith says

    Tony –

    Do you mean aside from the Essjay scandal, the Jimmy Wales/Rachel Marsden scandal, the misappropriation of money by Jimmy Wales scandal, the…. There have been so many it’s hard to keep track. And that’s without mention of things where wikipedia wasn’t directly at fault, where its open-edit system allowed corporations, political or religious groups to alter and falsify content.

    As to MS’s point about consensus, the consensus among most Americans is that Saddam Hussein was involved with 9/11, Iran has nuclear weapons and that “god” exists. “Consensus” is a horrible way to decide things, it can lead to a mob mentality, an echo chamber. If edits don’t agree with the “consensus”, they get removed, regardless of their veracity. Wikipedia’s “experts” and moderators are often no more objective than creationists who “peer review” one another.

    I’ll stick with my Penguin Concise Encyclopedia (a 1000 page paperback) and recently purchased Encyclopedia Britannica CD, thank you very much. I trust their content because they’ve been properly vetted. I don’t trust wikipedia’s content.


  10. mental reservation says

    The only articles on Wikipedia which I realize as being clearly biased are fringe woo or woo-related topics (complete with strawman pseudo-criticism that gets refuted on the spot). As every minute a new woo is being invented, the pracitioners will be the first ones to write about it, and it takes some time until the article is “battled over”.

Leave a Reply

Your email address will not be published. Required fields are marked *