Sunday, November 11, 2007

Research about Wikipedia


I heard a story on public radio about a new research study just released about Wikipedia, and that study led me to a couple others. Here is a summary of them:

  • An ongoing study at the University of Minnesota found that only 1/10 of one percent of Wikipedia editors accounted for 44% of the "content value" of the site (meaning content that ends up being permanent).
  • The same study found that 42% of errors on Wikipedia were caught before anyone saw them (they have anti-vandalism bots) and that your overall chance of encountering an error of any type on a Wikipedia page is .0037%
  • A study by MIT showed that the median of the survival time of all types of misinformation on Wikipedia was 90 minutes... but that there are "outlier" articles that survive longer in error without detection.
  • Finally, a study by researchers at Dartmouth showed that anonymous one-time wikipedia contributors had the best quality of contributions overall. The contributions of "power" contributors increased in quality over time. Sociologists are interested in what motivates anonymous users to contribute - what they "get out of it".
What does it mean that so few Wikipedia users are editors, and that even fewer editors are major contributors? Does the low error rate reassure you, or do you think that it is not low enough given Wikipedia's reach?

No comments: