Study: Wikipedia as accurate as Britannica

Wikipedia is about as good a source of accurate information as Britannica, the venerable standard-bearer of facts about the world around us, according to a study published this week in the journal Nature.

Over the last couple of weeks, Wikipedia, the free, open-access encyclopedia, has taken a great deal of flak in the press for problems related to the credibility of its authors and its general accountability.

In particular, Wikipedia has taken hits for its inclusion, for four months, of an anonymously written article linking former journalist John Seigenthaler to the assassinations of Robert Kennedy and John F. Kennedy. At the same time, the blogosphere was buzzing for several days about podcasting pioneer Adam Curry's being accused of anonymously deleting references to others' seminal work on the technology.

Related story
Growing pains for Wikipedia

After two scandals in one week, Wikipedia's founder decides to make a change to the anyone-can-contribute encyclopedia.

In response to situations like these and others in its history, Wikipedia founder Jimmy Wales has always maintained that the service and its community are built around a self-policing and self-cleaning nature that is supposed to ensure its articles are accurate.

Still, many critics have tried to downplay its role as a source of valid information and have often pointed to the Encyclopedia Britannica as an example of an accurate reference.

For its study, Nature chose articles from both sites in a wide range of topics and sent them to what it called "relevant" field experts for peer review. The experts then compared the competing articles--one from each site on a given topic--side by side, but were not told which article came from which site. Nature got back 42 usable reviews from its field of experts.

In the end, the journal found just eight serious errors, such as general misunderstandings of vital concepts, in the articles. Of those, four came from each site. They did, however, discover a series of factual errors, omissions or misleading statements. All told, Wikipedia had 162 such problems, while Britannica had 123.

That averages out to 2.92 mistakes per article for Britannica and 3.86 for Wikipedia.

"An expert-led investigation carried out by Nature--the first to use peer review to compare Wikipedia and Britannica's coverage of science," the journal wrote, "suggests that such high-profile examples (like the Seigenthaler and Curry situations) are the exception rather than the rule."

And to Wales, while Britannica came out looking a little bit more accurate than Wikipedia, the Nature study was validation of his service's fundamental structure.

"I was very pleased, just to see that (the study) was reasonably favorable," Wales told CNET News.com. "I think it provides, for us, a great counterpoint to the press coverage we've gotten recently, because it puts the focus on the broader quality and not just one article."

He also acknowledged that the error rate for each encyclopedia was not insignificant, and added that he thinks such numbers demonstrate that broad review of encyclopedia articles is needed.

He also said that the results belie the notion that Britannica is infallible.

"I have very great respect for Britannica," Wales said. But "I think there is a general view among a lot of people that it has no errors, like, 'I read it in Britannica, it must be true.' It's good that people see that there are a lot of errors everywhere."

To Britannica officials, however, the Nature results showed that Wikipedia still has a way to go.

"The (Nature) article is saying that Wikipedia has a third more errors" than Britannica, said Jorge Cauz, president of Encyclopedia Britannica.

But Cauz and editor in chief Dale Hoiberg also said they were concerned that Nature had not specified the problems that it had found in Britannica.

"We've asked them a number of questions about the process they used," Hoiberg said. "They said in (their article) that the inaccuracies included errors, omissions and misleading statements. But there's no indication of how many of each. So we're very eager to look at that and explore it because we take it very seriously."

[PCCRomeo]

Wikipedia is the best, hands down

For it's price, you CAN'T beat Wikipedia. I had to do a report about a month ago over the Victorian Period, not the reign of Queen Victoria but the entire period and Wikipedia offered all the information I needed, when Encarta and Britannica only had articles about Queen Vicki.

[Earl Benser]

Wait until you get to high school....

... Wikipedia isn't acceptable as a stable reference source for
serious reports.

[Earl Benser]

Wait until you get to high school....

... Wikipedia isn't acceptable as a stable reference source for
serious reports.

[Redoran]

Wikipedia vs. IRS

Sooo many comments criticizing Wikipedia. I ask you critics. Have you ever posed a tax question to several IRS agents? If you pose one complex tax question to 5 agents, guranteed you will received 5 entirely different answers. Happy tax season!

[Earl Benser]

For about the same reasons....

.... no one knew what they were talking about.

[jerseyrich]

Whoever said Wikipedia was accurate anyway?

It says right up front that ANYONE can edit it, which includes dopes, idiots, and liars. In other words, worth a look, but all taken with a grain of salt.

[Chase_The_Bass]

Wikipedia has different levels of security on their articles. Any article you'll see in Encyclopedia Britannica will have the highest security and editing the article has to be approved at several levels. Then you have all the Britannica quality articles that Britannica doesn't have. Only when it gets down to very obscure articles can people easily edit things. These articles are usually so sparse as it is that no one would credibly use it and is there just for the sake of being there.

[JonathanGCohen]

How To Improve Wikipedia

In the post I wrote about Nature's study, I mention that scientific accuracy would be more likely not to suffer from vandalism or inaccuracies because empirical research by definition can't be based on opinion. However, for historical, biographical, cultural (Re:, everything that can be disputed), Wikipedia's accuracy will always remain an issue. They need to emphasize external links, institute a vandalism warning system where readers can alert others to an article's inauthenticity (see my bicameral article rating system idea on Wikipedia if you're interested), and employ editors who can improve readability and grammatical correctness.

-Jonathan
http://philoneist.com

[Soupir]

Earl Benser: Eat Humble Pie

Hey Earl, Earl, Earl. This proves your big stink about Wikipedia as an accurate source exactly what I said it was: the unfounded whimpers of a neophyte. If a respected academic research journal like Nature can level with us about the accuracy of the Wiki versus Britannica, it proves the Sieng scandal was much more the exception than the rule.

Predicting your hardheadedness, you will find another reason to make a stink. But know this: the errors in the Britannica will have to wait until the next printing to be corrected. The Wiki's have probably already been changed.

[Earl Benser]

You really should pay better attention....

.... Wikipedia may well be accurate most of the time. I have no
problem with that, and never did have. Wikipedia's problem is that
what it says today may be correct, but not what it says tomorrow,
and vice versa. It is exactly that instability that makes Wikipedia
inadequate to serve as a serious reference source.

No humble pie required here. You just need to have better criteria
for evaluating dependable accuracy.

[mgrey]

The Whole is Greater Than Its Individual Parts

Wikipedia proves the value of collaboration on a global basis. The output of many minds results in clarity of purpose and innovation. The lesson to be learned is that if collaboration among strangers across the Internet can result in the Wikipedia -- think how collaboration among colleaguescan transform business.

[uncle frank--2008]

Britannica vs Wilipedia

I have Encyclopedia Britannica installed on my PC. A lot of articles are out of date. For example encyclopedia Britannica describes IBM as a hardware manufacturer while Wikipedia describes IBM as a hardware manufacturer and a major consulting firm (IBM Global Services revenues and business consulting revenues are greater than hardware).

[Decoy256]

Missing the Point

Britannica is reliable. Britannica is slow. It sacrifices speed for accuracy on a narrow spectrum of topics.

Wiki is less reliable, but then again, how often do we need to know all the nitty gritty details of a subject? When trying to find out the distance to the Moon, f wiki causes me to be off by 100,000 miles, no big deal, to me.

But here is the big misunderstanding. Wiki is fast and growing. It has a MUCH broader range of topics AND you can directly link to external websites to find out more. If something happens in the news tonight, you can bet that Wiki will have something up tomorrow. It may not be 100% accurate, but I'll sacrifice a little accuracy for information now.

Additionally, you are not going to find an article on "1337 speak" in Britannica. Nor will you get an article "Stargate: SG-1" in Britannica. As far as American culture goes, Wiki is king. Also, there are alot of fringe interests out there that don't get a whole lot of attention by major publishers, but they have entries in Wiki.

All in all, I think that Wiki has done more to raise the "general" knowledge of people than Britannica, simply because it gives your the information you are looking for quick and gives you enough general information so that you can go from there and research the subject further.

[perspective16]

Wikipedia vs Britannica: it's not as simple as it seems.

There is more to this debate than meets the eye. We don't know all those who were responsible for a Wikipedia article, but neither do we know who wrote Britannica articles -- at least not without a lot of research. One example is that the article on ethics in the Britannica has been written by the controversial Professor Peter Singer, the animal liberationist and euthanasia advocate -- an odd source of "objective" scholarship. I have more on this at MercatorNet: http://www.mercatornet.com/index.php?option=com_content&task=view&id=199. Hence, the real problem is whether Wikipedia can handle non-empirical topics, not so much whether the authorization system works well.

[RobertRFreeman]

Valid Reference due to history!

Wikipedia keeps a history of changes, so you can identify a point in time version of an article for verifiable referencing. Just remember to include the date/time with your reference in place of a version number.

Anything any joe says at any time can be a reference in a college level paper. The authority of the reference is the question. Wikipedia tracks the history of an article, so verfiying the overall validity of an article is as easy as browsing the history to view edits. Generally, the more people that have edited an article, the more accurate it will be. You can also easily spot POV or vandalism by content changed. It's too bad EB doesn't offer a history of changes broken down by source!

What Wikipedia really needs is a notification system for changes. That way anyone referencing or modifying an article could register for notifications and be immediately notified of any changes.

One more thing, when researching, always look at all sides of any issue. Wikipedia should be used as one of many references, and is great as a starting point. And once your research is completed and if you find the Wiki article lacking, then change it!

[gambit10103]

"What Wikipedia really needs is a notification system for changes. That way anyone referencing or modifying an article could register for notifications and be immediately notified of any changes."

I second this, further discussion is expedited when an individual knows when his entries have been edited.

[drdavid hill]

Wikipedia - can we really trust it in the long-term

But how accurate is the information that Wikipedia supplies? A set of nerds who think that they know everything but where there is no authoritative reference. i.e to a qualified and highly experienced/respected authority (with 50 years on his or her back at least) et al leaves the information totally open to abuse. This is the greatest danger for the world-at-large, accepting what Wikipedia says without question and where if history is rewritten, future generations will just not know right from wrong. A terrible state of affairs for all future generations to come. The same thing will probably equate to the equivalent of Google. I would steer completely away, for the ramification for the future generations to come are grave to say the very least.

Dr. David Hill Chief Executive
World Innovation Foundation Charity
Bern, Switzerland Registration no.CH-035.7.035.277-9 - 11th July 2005, in the Canton of Bern
www.thewif.org.uk

[pawdyest]

testing the reliability of the two sources-wikipedia.org and encyclopedia such as the britannica encyclopedia, is quite simple..

there are lots of issues here, first is the reliability, the latest edition and revision..

[KBawden1]

"This is the greatest danger for the world-at-large, accepting what Wikipedia says without question and where if history is rewritten, future generations will just not know right from wrong. A terrible state of affairs for all future generations to come." - Dr. David Hill

I would argue that the exact future of Wikipedia you speak of already exists within the present media. How many myths and mistruths are perpetuated currently that the general public don't know aren't correct? Hell, even school textbooks have scores of fundamental errors. Certainly this article doesn't suggest Wikipedia is infallible, but neither is Britannica, which is a greatly respected resource. I do however agree that to become complacent and ignore the faults within Wikipedia's system would be foolish. It's the second guessing of these resources that helps keep their articles accurate. In Britannica's case the lack of this that has probably created it's high factual error rate for what should be, and is often considered, an almost errorless publication.

[celindgren]

Wikipedia is at best garbage. The following remark appeared on Wikipedia - "What I'm seeing is a lot of puffed-up titles with little backing or significance, employment at something one step up from a diploma mill, and no sign of real academic notability (or for that matter real-world notability as measured by major media attention). ?David Eppstein (talk) 08:04, 15 December 2008 (UTC)" The 'professor' was referring in part to American Military University. AMU has over 30,000 students and is regionally accredited. When does this rubbish stop?

C-

[paultil]

"Nature got back 42 usable reviews from its field of experts." The Britannica contains 100,000 articles.
42. So we have 42 out of 100,000. What kind of sample is that? This study is not valid enough for the points it tries to make (according to this article).

[wiki-not-accurate]

I would question wikipedia's accuracy. Anyone can edit it, and a biased editor with political pull can get people banned from editing. It is impossible to check all the references throughly, making the possibility of data manipulation large.

PRESS RELEASE - Wikipedia ?Satanic Ritual Abuse? article promotes PEDOPHILIA http://ritualabuse.us/ritualabuse/articles/press-release-wikipedia-%E2%80%9Csatanic-ritual-abuse%E2%80%9D-article-promotes-pedophilia/

The Truth about Satanic Ritual Abuse - Wikipedia rebuttal - A Rebuttal to Wikipedia?s Portrayal of Satanic Ritual Abuse - by Wanda Karriker, PhD http://ritualabuse.us/ritualabuse/articles/the-truth-about-satanic-ritual-abuse/

[GlennDawson]

The point that needs to be made is that for every idiot who will edit an article so that is shows an inaccuracy, whether intentionally or not, is counterbalanced by every other person in the world, not to mention Wikipedia's extremely efficient moderation team. If an inaccuracy is recorded, odds are extremely high that someone, somewhere will either revert it to the previous entry or correct it. As has been previously stated, the level at which articles become easily editable is sufficiently low to negate the effect of an inccorect edit. Even with large cases, such as those cited in the article, are correctly reffered to as "the exception, not the rule". The article should be read, not as an absolute justification of Wikipedia's accuracy, or as a denouncement of Britannica, but as a defusing statement, which will hopefully reduce the amount of extremism on the issue of accuracy and reliability.