Assessing Wikipedia's Accuracy In the wake of questions of accuracy, a survey by the science journal Nature finds that science entries in the volunteer-driven, online encyclopedia Wikipedia are "not markedly less accurate" than those found in Encyclopaedia Britannica.
NPR logo

Assessing Wikipedia's Accuracy

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Assessing Wikipedia's Accuracy

Assessing Wikipedia's Accuracy

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


By now it's been established beyond all doubt that John Seigenthaler, the founding editorial director of USA Today, is not a longtime suspect in the assassinations of John and Robert Kennedy. That misinformation was published as a joke in the free online encyclopedia Wikipedia and brought to light two weeks ago by Seigenthaler himself. The incident called into question Wikipedia's credibility and accuracy. Wikipedia relies on volunteers to write and edit its content. But an article published yesterday in the online version of the journal Nature brought Wikipedia some good news. An investigation carried out by Nature found that science entries in Wikipedia are not, quote, "markedly less accurate than those found in Encyclopedia Britannica." We're joined by Mark Peplow, one of the Nature staffers who conducted this survey.

Mr. Peplow, thanks for being with us.

Mr. MARK PEPLOW (Nature): That's OK.

NORRIS: What were you looking at in this investigation, and what did you find?

Mr. PEPLOW: OK. Well, we drummed up 50 different scientific terms that reasonably similar lengths of entries in both Encyclopedia Britannica Online and Wikipedia. And we asked scientific experts in the field to actually do some peer review on these, and specifically we asked them to look for factual errors, critical omissions or unclear, misleading or biased statements. So having agreed what the areas were, we totted them all up and found that on average, if you looked at an Encyclopedia Britannica article, there were maybe about three errors per entry, and with Wikipedia, there were about four per entry; so not substantially less accurate than the sort of gold standard Encyclopedia Britannica.

NORRIS: And when we talk about these errors, how big or how large are we talking about?

Mr. PEPLOW: The vast majority of them are very small. We didn't regard spelling mistakes as errors. However, if someone's name was misspelled, like an important person in the context of the article, just about qualified as an error. Some of them are very simple errors about dates. However, in a few cases, just a handful of cases, there were sort of more significant conceptuaries where it was clear that they didn't really understand what it was that they were describing and they'd sort of missed the whole point of, say, how a field-effect transistor works.

NORRIS: So the upshot is that the error rate for Wikipedia is not that much different than Encyclopedia Britannica. Were you surprised by that?

Mr. PEPLOW: I was. I must admit I expected Wikipedia to have more errors than Encyclopedia Britannica. What I found more surprising was just how many of these sorts of very tiny errors were found across both encyclopedias 'cause you don't really expect them, particularly not in Britannica, which, like I said, is regarded as the gold standard

NORRIS: So is this good news for Wikipedia or bad news for Encyclopedia Britannica?

Mr. PEPLOW: Well, I don't think you should be seeing this bad news necessarily for either of them because, obviously, both strive for quality. But I think it is particularly good news for Wikipedia because it shows that their model of doing things, where anybody can contribute to this freely available encyclopedia--that model is working. It's not producing complete garbage; it's producing something which is--you know, it's in the same ballpark at least as Encyclopedia Britannica.

NORRIS: Mark Peplow, you work for an esteemed scientific journal. Would you recommend consulting these encyclopedias on scientific topics?

Mr. PEPLOW: I think they're both a good place to go for a first look at a topic. I don't think I'd ever regard any encyclopedia as sort of gospel truth on anything, but they're a really good place to go for a sort of first look at any scientific subject to get your head around it, really.

NORRIS: You've also surveyed about a thousand scientists who publish papers in Nature to find out what they knew about Wikipedia. What did you find there?

Mr. PEPLOW: Well, we found that out of these thousand or so Nature authors, more than 70 percent had heard of Wikipedia, which is pretty good given that--I mean, certainly, I hadn't heard of it two years ago. Seventeen percent of those that had heard of it consulted it on a weekly basis. But interestingly, less than 10 percent of those helped to update it. So one of the ways, obviously, with this sort of community-driven encyclopedia--one of the ways to improve it is for more scientists to get involved and just checking out whatever things are closest to their area of expertise, checking if it's right, and if it's not, spending five or 10 minutes just to update it.

NORRIS: Mr. Peplow, thanks so much.

Mr. PEPLOW: Thank you.

NORRIS: Mark Peplow is a reporter with the science journal Nature.

Copyright © 2005 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.