http://arstechnica.com/news.ars/post/20061127-8296.html
Snippet
Experts rate Wikipedia's accuracy higher than non-experts

11/27/2006 4:01:25 PM, by Nate Anderson

A new salvo has been fired in the perennial war over Wikipedia's accuracy. Thomas Chesney, a Lecturer in Information Systems at the Nottingham University Business School, published the results of his own Wikipedia study in the most recent edition of the online journal First Monday, and he came up with a surprising conclusion: experts rate the articles more highly than do non-experts.

This less-than-intuitive finding is the conclusion of a study in which Chesney had 55 graduate students and research assistants examine one Wikipedia article apiece. Each participant was randomly placed into one of two groups: group one read articles that were in their field of study, while group two read randomly-assigned articles. Respondents were asked to identify any errors that they found.

I still won't use it in papers, as a cited source.

But I will use it to find citable sources.

And it generally is good enough source for forum debates. Whenever I have looked up medical knowledge, it has been right on. I have even editted entries on it, for the better(even if you don't believe me).

For scientific or technological issues, Wikipedia is the single best source on the Internet to find answers. Everyone knows this, other than Aldarion_Shard.

I use to be an avid critic of Wiki quoters.

I am very much less so, now.

I will even quote it myself.

I still won't use it in papers, as a cited source.

But I will use it to find citable sources.

And it generally is good enough source for forum debates.
Yup, I agree 100%. That's a very good point about using Wikipedia to find sources. (In theory this should be true for any encyclopedia, though I remember finding it rather inconvenient to use a traditional paper-based encyclopedia this way.)

As for the study, I'm not surprised by the results, though they'd be more meaningful if more than 55 people were involved.

There is a clock on my desk with no batteries. The time never changes, because its an electric clock. Exactly twice every single day, this clock is precisely correct. If I check the time exactly twice a day, at these times, this clock is 100% correct in telling me the time.

This clock is wikipedia.

Horray, sometimes it gives me the right answer! Depending on what questions I ask it (e.g. what time is it now [at 3:15] versus what time is it now [at 4:20]), the answer may even be correct the majority of the time.

A reference that sometimes gives you the right answer is fine for things that dont matter much anyway (such as internet debates). In such situations, you'd be better off just stating your argument more clearlywithout the sources. Nine times out of ten, when a internet debater posts a source, what they are really saying is "I suck at stating my argument, so here is a famous expert's name, please accept my argument on this persons authority".

You know what pisses me off the most about wikipedia, though? If it was a book, itd be fine... you use the resources you have, and if your only reference book on Llamas is a bit out of date or just inaccurate, I wont blame you. Its all you had.

But every single person who ever did a search on wikipedia had access to the REAL sources too! We have this thing called the internet. When you search for "stem cell research" on Google, the #1 hit is the NIH website dedicated to this issue: a fantastic source compiled and maintained by experts. But people will cite the FIFTH result instead, wikipedia, EVEN THOUGH THEY HAVE ACCESS TO THE REAL SOURCE!

Wikipedia is the lazy man's resource: a way for the MTV generation to pretend to do research. When it comes to researching a field in which experts exist, you go to the damn experts, or dont bother.

Aldarion, people aren't writing thesis when they consult Wikipedia. It is a great place to get a usually concise summary of the newest information. If you need more than that, you can consult the references, use a regular search engine or use scholar.google.com or a number of other things. But even if you do go to the actual studies, you can't always get anything more than a snippet unless you subscribe to a journal.

Wikipedia is absolutely fine for a lot of things but if your professional reputation or grades are on the line, you'd better follow it up with more research.

I also like the fact that Wikipedia denotes areas of dispute and has discussion areas on each topic and you can see who wrote stuff and what their credentials are. How often in a book do you see someone honestly labeling what they've written as "in dispute"?

The entire idea of having knowlegeable people sharing, for free, about things they're experts in is incredible. There has been nothing like it in all of the human race.

So stop dissin' Wikipedia. I think it is a noble, worthwhile and amazing project.

Horray, sometimes it gives me the right answer!That's true of <a>any</b> source. There's no single book, study, or paper that is guaranteed 100% correct. On the contrary, most single sources on controversial issues tend to be heavily biased towards one author's opinion, whereas the "open source" nature of Wikipedia works against such bias.But every single person who ever did a search on wikipedia had access to the REAL sources too!Hardly. Many of the sources aren't freely available on the Internet, because publishers and authors won't make money that way.

Even if they are, there's no reason to reinvent the wheel and find them when a Wikipedia entry nicely summarizes them.

Wikipedia is an excellent resource if you realize its limitations, and I have no doubt it will continue getting better and better until it's bought out by some mega-corporation.

Wikipedia is the lazy man's resource: a way for the MTV generation to pretend to do research. When it comes to researching a field in which experts exist, you go to the damn experts, or dont bother.

It doesn't aim to replace scientific reports. It fills the same need as encylopaedias, providing accessible information on a variety of topics very quickly.

The big drawback is that any single entry at any time might lack editorial control or accuracy. It still compares well to my printed enclopaedia as it is free, it has been updated with events post-1977, it has many more entries on a wider range of topics, and it of course has many links. Another asset of my paper encyclopaedia, that wikipaedia will probably never possess, is its laminated cover that makes it an excellent place to put a mug of tea.

Aldarion, you're missing the point. The Wikipedia entry directs you to the NIH page that you mentioned, as well as dozens of other resources. Having many resources is better than having one resource, right?

And by comparing it to a clock that is correct twice a day, you are greatly overstating the level of inaccuracy of Wikipedia. I've checked entries related to areas I am knowledgeable in, and the entries are generally accurate (though admittedly these areas are scientific or technical, which tend to be Wikipedia's strongest areas as well).

I think it is a noble, worthwhile and amazing project.
And I think it is a fundamentally flawed concept.

I actually dont disagree with you guys about the result: most of the time, the information found in wikipedia is correct. No, I disagree with it in PRINCIPLE: the entire notion of a wiki is flawed. Entries should be editable ONLY by experts.

Using wikipedia is like being assigned to answer a set of questions about American History, and copying all the answers off your friend. In the end you get the same answer, and both involved reading it one place and then writing it down on your paper, so you might ask: what is the harm?

The harm is that HOW you get the answer matters as much as WHAT answer you get. Using wikipedia makes people lazy. Nobody bothers to read the real sources because they are too inaccessible (i.e. they may require actual though and knowledge of the ubject). Instead, they read the dumbed-dpwn wiki version.

Wikipedia is an excellent resource if you realize its limitations
I agree completely. I *like* wikipedia. There is no better source for unimportant topics like fictional series or comic books or internet fads.

We just disagree about its limitations. I think it is entirely inappropriate to ever use it for anything scientific.

No, I disagree with it in PRINCIPLE: the entire notion of a wiki is flawed. Entries should be editable ONLY by experts.On scientific and technological topics, it generally <b>IS</b> edited by "experts," meaning people who have special knowledge in the field.

The range of people editing Wikipedia have, on the whole, far more expertise than the staff of any encyclopedia.Using wikipedia is like being assigned to answer a set of questions about American History, and copying all the answers off your friend.Your friend only copied it from a history book. And the history book only copied it from other books. And so on.

Just because Wikipedia entries use facts from other sites doesn't mean it's inaccurate or a blind cut and paste.Using wikipedia makes people lazy. Nobody bothers to read the real sources because they are too inaccessible (i.e. they may require actual though and knowledge of the ubject). Instead, they read the dumbed-dpwn wiki version.Try reading a typical scientific article with about 50 references. Much of the article will draw from those references, with just a few new ideas. Do you have to read all 50 cited articles to avoid being "lazy"? And what happens if you do, and each of those cites 50 others? Do you have to understand every scientific fact in the field just to get a simple answer?

Entries should be editable ONLY by experts.


People keep using the term "expert" like it means something special in this context. What is an expert? Who defines who is an expert? Other experts?

There's nothing magical about an "expert" that makes them right. Believe me, I've deposed enough of them.....

And I think it is a fundamentally flawed concept.

I actually dont disagree with you guys about the result: most of the time, the information found in wikipedia is correct. No, I disagree with it in PRINCIPLE: the entire notion of a wiki is flawed. Entries should be editable ONLY by experts.

Define expert.

Using wikipedia is like being assigned to answer a set of questions about American History, and copying all the answers off your friend. In the end you get the same answer, and both involved reading it one place and then writing it down on your paper, so you might ask: what is the harm?

The same analogy can be made when you answer that same set of questions by opening up Rutherford's Concise Summarization of the Hullaballoo behind the Uprising of those Damned Colonials.

The harm is that HOW you get the answer matters as much as WHAT answer you get. Using wikipedia makes people lazy. Nobody bothers to read the real sources because they are too inaccessible (i.e. they may require actual though and knowledge of the ubject). Instead, they read the dumbed-dpwn wiki version.

I would venture to suggest that the wiki version isn't that dumbed down and provided more in depth answers than you are likely to find in an encylopedia. Answers which are updated continually, rather than the relatively static nature of texts. How you get the knowledge matters not, if the knowledge if accurate.


I agree completely. I *like* wikipedia. There is no better source for unimportant topics like fictional series or comic books or internet fads.

We just disagree about its limitations. I think it is entirely inappropriate to ever use it for anything scientific.

I think you can find out a better understanding of any given scientific topic you can find on wiki, than you could get if you spent a full day looking up texts on the issue. Hell, make it a week, since it would take you a few days just to find sufficiently related articles and books which are of an appropriate level for your interest.

Entries should be editable ONLY by experts.


People keep using the term "expert" like it means something special in this context. What is an expert? Who defines who is an expert? Other experts?

There's nothing magical about an "expert" that makes them right. Believe me, I've deposed enough of them.....


aybe he'd be satisfied if every wiki topic was ended with "to a reasonable degree of <insert scientific field> certainty"? =P