patrick.net

@terriDeaner

Just one of the thousands of editors out there (all volunteer).

Much bigger than others...
From Wikipedia:

"Wikipedia currently has 3,560,966 articles in total in the English version alone. (This article count is also available on the main page.) the Encyclopædia Britannica had about 85,000 articles with 55 million words in total, and Microsoft's Encarta had about 63,000 articles and 40 million words in total."

Bigger isn't always better, but a fair comparison for accuracy would be with head-to-head articles instead of taking some of Wikipedia's fringe articles.

That was done in 2005 in a study commissioned by Nature magazine:

http://www.nature.com/nature/journal/v438/n7070/full/438900a.html

Be sure to read the rebuttal by Britannica, and the subsequent responses by Nature.

They concluded that Wiki came pretty close to Encyclopedia Britannica, based on 42 articles compared between the two sources. Some key problems with the analysis:

1. Huge bias towards type II error (failure to find a difference between the two encyclopedias even though one exists) due to serious undersampling: they compared only 42 articles out of a possible 85,000 (from your estimate), which is 0.05% of all possible sample comparisons. This was not brought up in the Britannica rebuttal, but is a big problem in my opinion.

2. Nature never double-checked the factual content of their reviews, Nature claims in one of their rebuttals that they blindly gave Wiki and Britannica articles to reviewers, so any errors should balance out. Not true if their small set of reviewers is biased for or against one of the sources. For example, consider if all of the reviewers were unknowingly also Wiki editors, or if their research had been used by others to write the Wiki articles. In this extreme case, there would be a strong systematic bias against Britannica.

3. Nature and Britannica are, at heart, both publishing companies. Nature sells magazines based on controversy (Britannica = Wiki), Britannica sells encyclopedias based on quality (Britannica > Wiki). Both have incentive to spin the study results in their favor, so who should you believe?(hint... always consult the primary references)

By the way, I didn't think my examples were from fringe articles. 'Fraggle Rock' was a internationally popular show from the 1980's, and 'Lion' was a gold star article that you posted.

@terriDeaner

Thanks for the study, this is a good one! Some comparisons have just taken any article at random from each which I felt isn't fair.
(My definition of "fringe article" might be too loose. My definition was just any articles that wouldn't even be on Encyclopedia Britannica. I should have written that more clearly to avoid confusion. Certainly "Lion" isn't a fringe article. I wonder if "Fraggle Rock" is...Lol!)

You're welcome. I checked Britannica online, and although there was no separate entry for 'Fraggle Rock' there was an entry for its creator, Jim Henson.

At any rate, part of why I do like Wikipedia is because there is at least a little bit of information on just about any subject. The devil is always in the details though.

Study: Wikipedia as accurate as Britannica
http://news.cnet.com/2100-1038_3-5997332.html

Though you can't beat Wiki's convenience if you're sitting at a computer.

I still want to get my daughter a set of physical Encyclopedias. I recall hours spent leafing through them... I understand Britannica is stopping the print version?

>I still want to get my daughter a set of physical Encyclopedias.

I agree.

>I understand Britannica is stopping the print version?

Not sure...could be.

kentm says

Study: Wikipedia as accurate as Britannica

http://news.cnet.com/2100-1038_3-5997332.html
Though you can’t beat Wiki’s convenience if you’re sitting at a computer.
I still want to get my daughter a set of physical Encyclopedias. I recall hours spent leafing through them… I understand Britannica is stopping the print version?

Note that this article references the material from the commissioned Nature article cited above. My point again: read past the headline, and look at the source material.

By the way, I LOVE print encyclopedias.

Tenouncetrout says

Revised history, at its best.

history is always written by the winners.

At least in the zoology section, some editor will slap tags on the article in a day or two (if not.within hours) if something is not referenced or is even slightly inaccurate…

Any other errors found?

American in Japan says

Any other errors found?

Last time I looked, the letter 'Q' was replaced with the number 4, but it didn't really affect the readability that much so I don't think anyone has complained yet...

In this case be careful of using only Wikipedia:

http://www.kevinmd.com/blog/2008/11/dont-use-wikipedia-for-drug-information.html

C'mon now AiJ... a little slow on the draw here...

Ironcially, wikipedia is self-aware:

http://en.wikipedia.org/wiki/Reliability_of_Wikipedia

there's a section on "Comparative studies" that you'll probably find interesting

Thanks for the link!

Wikipedia is good for anything non-controversial. When you get into something where people strong disagree, the wikipedia admins (usually high school/college kids with too much time on their hands) dominate the articles and systematically control who's edit stays and whos doesn't.

The articles I read / edit are dominated by graduate students, but I agree some came push their views a bit (at least for formatting).

There are some stubborn Administrators on Wikipedia that make their opinion known.

American in Japan says

There are some stubborn Administrators on Wikipedia that make their opinion known.

The whole thing is sold as some egalitarian open process. In reality, Wikipedia has a ridiculous number of rules, is highly highly regulated, and the whole process is political and dictated by a small group of people. Their marketing is awesome!

Still there is lots of great information on animals, languages, astronomy, chemistry, math and country infomation. I agree that a small group of people with lots of time on their hands have undue influence. I am cautious when reading the articles on companies.

Did anyone else find any inaccurate articles?

Wikipedia whitewashes history. You should never use any Wikipedia article that deals directly or indirectly with people, money, politics, religion, culture, companies, products, or history.

Also, don't quote Wikipedia. It makes you look like an idiot. Only idiots and the intellectually lazy resort to encyclopedias. Remember when you were in elementary school and the teacher told you not to use the encyclopedia and go to the library instead? Only super-idiots trust encyclopedias without peer review. The fact that the dumbest 80% of America uses Wikipedia, doesn't make you look better for quoting it.

Do real research instead. It's not much harder and you get far better results. Check out my previous rants on Wikicrapia for more details.

Dan8267,

Good point. I write and edit articles on Wikipedia, so I am particularly concerned with referencing. I missed the "Wikicrapia" post...I'll try to find it.

FortWayne says

Tenouncetrout says

Revised history, at its best.

history is always written by the winners.

Quality Auto Repair Since 1979

You mean whiners?

I like wikipedia, and use it regularly. But not because I think it's better than doing extensive research on my own.

For my usual purposes, it's useful and accurate.

Wiki works for me. But for school projects, I always direct my children to my 1911 set of Brittanica. If the subject existed then, it's in there and there are obscure little factoids that get the kids an "a."

Thanks EllieMae!

Dan8267 says

Do real research instead. It's not much harder and you get far better results.

That is limited too, unless you have a way to get scholarly research articles without begin forced to pay through the nose or having to visit a university library.

New Renter says

That is limited too, unless you have a way to get scholarly research articles without begin forced to pay through the nose or having to visit a university library.

Better to have little or even no information than to get misinformation. I'll take nothing over the deliberate misinformation on Wikipedia any day.

Still, Google is pretty darn good at finding articles from reputable publications that you can read for free.

@Dan8267

Which Wikipedia articles specifically have you found to be the worst? Just curious in a "Patrick" sort of way...

American in Japan says

"to realize that America's mania for home-buying is out of all proportion to sober reality, one needs to look no further than the current subprime lending mess... As interest rates—and mortgage payments—have started to climb, many of these new owners are having difficulty making ends meet... Those borrowers are much worse off than before they bought."

You should hear what people were saying in 2005.. its just the east side of some city,, wont impact the the ubber rich west side or the Fortress... then the end of 2008 hit and many places also fell like dominoes..

Dan8267 says

Still, Google is pretty darn good at finding articles from reputable publications that you can read for free.

not anymore ...

Google Kills Its Own "Timeline" Feature
www.readwriteweb.com/.../google_kills_its_own_tim...Share

Jon Mitchell

by Jon Mitchell - in 13,379 Google+ circles - More by Jon Mitchell
Nov 11, 2011 –

As Google works to emphasize up-to-the-minute search results, it has also quietly killed off a search feature that helped users search for content from the past. As users in the Google search help forum have noticed, the Timeline feature for Web search has disappeared. It helped filter search results for specific timeframes.

I wanted to ask this one again...Any problems or biased articles?

I believe everything written on paper or the internet but especially Patnet.

Wikipedia tends to be manipulated by certain industries, in ways that make it similar to commercial news:

Newsweek: "Why Almost Everything You Hear About Medicine Is Wrong"

Vanity Fair: "Deadly Medicine"

Part of that results from the reliance on commercial and industry publications, so the marketing bias of those sources gets carried into Wikipedia articles, and part of it results from outright manipulation by public relations firms. Wikipedia tries to stop public relations firms excessively manipulating articles, on a case by case basis, but there are always more.

Also, regarding American culture and politics, there is a recurring partisan battle between factions approximating the major political parties. In Wikipedia as in life, the Republicans imagine themselves morally superior and crusade to save everyone's souls, while the Democrats imagine themselves intellectually superior and dismiss any disagreement as ignorance. (In fairness to the Democrats, many of the self-styled "conservative" editors are really ignorant, and the same pattern can be observed on PatNet; not all conservatives are stupid, but stupid people are disproportionately likely to call themselves "conservative", and that brings down the average.) The result is articles can get pushed one way or another, some articles become battlegrounds while other related articles get ignored and can be outdated or plain wrong.

I do like Wikipedia for pop culture though, and it's a good place to look for a variety of source links on a topic. Also Wikisource. I would never quote from Wikipedia, because any fool can write anything in there and then quote himself a minute later. Like any online forum, including PatNet, it's more useful if you check what the actual linked sources say.

Thanks. Wikipedia has very good articles on chemistry, astronomy and for animals / plants. Also many good articles for university teams in basketball and football.

Wiki gets updated way too fast it seems like.
I'll often hear about someone famous dying, and Google them and the Wiki page will already be updated.
Although reports will state that details were not known or released at that time.

I've seen biased opinions when it came to politics or national events. But you can never get away from that, whoever writes will always be biased.

@Fortwayne,

You could be right. Could you give me a few specific articles?

"This kind of feedback loop—wherein an error that appears on Wikipedia then trickles to sources that Wikipedia considers authoritative, which are in turn used as evidence for the original falsehood—is a documented phenomenon. There’s even a Wikipedia article describing it. Some of the most well-known examples involve Wikipedia entries for famous people, such as when users edited the article on the British actor Sacha Baron Cohen to say he had worked at Goldman Sachs. When a Wikipedia editor tried to remove the apocryphal detail, it took some convincing. Because it had since appeared in several articles on Cohen in the British press, the burden was on Wikipedians to disprove the myth.

“As a long-time Wikipedia editor, it frustrates me when journalists don’t fact check Wikipedia and end up reproducing errors, because Wikipedia can only work the way it does if we have reliable sources to cite,” Stuart Geiger, a Ph.D. student at the University of California, Berkeley, School of Information wrote in an e-mail."

marcus says

I like wikipedia, and use it regularly. But not because I think it's better than doing extensive research on my own.

For my usual purposes, it's useful and accurate.

That is a very accurate statement about wikipedia. For the most part, it's convenient and useful. However, if you quote statements from it, people that do their own research will recognize where you got the information.

theoakman says

Wikipedia is good for anything non-controversial.

Fully agree with this statement. There are a ton of editors that are of tremendous value to Wikipedia, but then you have some members that want to spin an agenda m