Is Wikipedia more reliable ...?

Cecil's Columns/Staff Reports
1

Two parts of your reply are open to improvement. First, specialist encyclopedias tend to comprise articles written by identified experts, so that they are certainly more reliable than general encyclopedias such as the Brittanica. As an example, the University of California Press’s recent Encyclopedia of Islands (disclosure: to which I contributed) amounts to a multi-author review of the area, as reliable as any scholarly work.

Second, and more importantly, Wikipedia is quite uneven in its reliability in different areas. I have found it a good source in history, biography and art, for example, but often poor in biology and some other areas. The most likely reason for this is that those with expertise in some areas are more willing than experts in other areas to take Wikipedia seriously and to make the effort to improve it.

My advice to students is that Wikipedia can be used – criticallly and with caution – as a study source, but that it may not be cited, as we don’t know who is contributing what. This is a very unfashionable view, at least in my area of academia, where I see a lot of pretentious turning up of noses as Wikipedia, like affecting to despise airplane food or wine in boxes. I well recall the sharp intake of breath when I first suggested to a class that they read the Wikipedia entry on a particular topic, as they had all been told that the gentleman’s door must remain forever closed to that low-class source.

Christopher K. Starr
University of the West Indies

2

I heard something similar from a lecturer I was minute taking for when I worked in a university.

3

Column in question:
Is Wikipedia more reliable than the Encyclopaedia Britannica?

4

When I was a high school teacher, I told my students more or less the same thing about Wikipedia. Use it as a good place to start your research, check the references in each article, and go from there.

It is usually pretty reliable and you can learn a lot more from wikipedia than most other websites, but don’t ever cite it in a paper. But then again, don’t cite any encyclopaedia in any paper :smiley: Wikipedia is no different.
Also, glad to see Cecil finally tackle this question. I emailed it to him almost a year ago :slight_smile: He didn’t choose mine specifically to respond to, but it’s pretty much the same question.

5

My understanding is that the Encyclopædia Britannica is largely written by experts too. According to Wikipedia (yes irony etc.):

So it’s not as if there’s just a few people who like general knowledge writing the whole thing.

On Wikipedia itself, I think a lot of those who criticize its reliability underestimate how much many contributors love to browse through the recent edits looking for holes to pick.

6

The trouble is that the reference sections are often the worst part of the article. Articles are indeed quite often comparable in quality to what may be found in a “professional” encyclopedia, but the references are typically much worse, because they skew very heavily towards what may be found on the web, with open access. Often, the appropriate references are either not available online at all, or only behind a paywall that very few, if they are not affiliated with a good academic library, can breach. Instead, we get some random guy’s website, or an academic article or book (which Google may let us see snippets of, if we are lucky) that mentions the relevant topic in passing, because the article or book that is actually most relevant cannot be easily accessed by “regular people”.

Yes, but quite often the holes are being picked in order to substitute popular misconceptions for actual (but, perhaps, little known) facts. The popular misconceptions frequently win in edit wars, because “support” for them can be found in multiple tertiary (or quaternary, or worse) sources on line, whereas clear corroboration of the actual facts is (once again) only to be found in sources that are offline or paywalled). (Also, they win because Wikipedia because Wikipedia is controlled by enthusiastic amateurs who have lts of time to devote to it, not by experts who are too busy actually studying things to defend themselves in protracted edit wars.) Experts, or semi-experts do sometimes write for Wikipedia, but their expert opinions are often then picked apart by Wikipedians who give more credence to what they can find in a quick Google search than to what someone who has been studying a topic for years actually knows about it. The way Wikipedia works, ignorance and popular prejudice almost always drives out true expertise and knowledge (if they conflict).

7

There’s a lot of logic to that. Another issue is that much of the information to be found online, including from professionals, is sourced from Wikipedia. I once read an article on something (I don’t remember what) on the BBC News website. Wanting to learn more, I turned to Wikipedia. The Wikipedia entry seemed very similar with some sentences almost exactly the same. It appeared the Wikipedia entry was the original, and I got confirmation from the author of the BBC News article that it was, in fact, mostly a rewrite of the Wikipedia entry. But now all that information has “the largest broadcast news operation in the world” as a reference. Wikipedia is at least aware of this problem, however. However, I have never noticed popular misconception beating true fact for myself there.

8

One issue with having an article written by a single expert is that said expert will have his or her own biases that will probably be passed on to some extent in the article, wheras an article written by a group of knowledgeable amateurs would, one hopes. settle down to a steady state without too much bias.

I tried to think of an example of such a topic and looked up “Raw Meaty Bones”, as over the last couple of years I have tried to obtain a balanced view of this topic. The problem is that (nearly) all the experts I’ve spoken to are either rabid RMB advocates who think that commercial diets are the spawn of satan, or experts in nutrition who rely quite heavily on sponsorship from commercial food companies (this includes several national veterinary associations). Raw feeding - Wikipedia seems to me to be an extremely well balanced article, much more so than any I have read elsewhere.

I’m not sure that the traditional research method of reading papers direct from the journals is all its cracked up to be either - there is a substantial publication bias that, while it does weed out the “crackpot” theories, will tend to throw out the baby with the bathwater in some cases. The peer review system has an innate tendency to choose studies from established scientists working from recognised universities on mainstream theories, rather than the weirdo office clerk who has devised his own theory of the universe in his spare time.

I also read recently that a scary number (>50%) of <i>published</i> scientific papers (might have been one particular branch of science) used dodgy statistical analyses which cast doubt on their findings (really sorry but I don’t have the reference or even a memory of where I read it)

There is also the argument that old style journals charge far too much for their access (the peer review itself is done by volunteers) and should be replaced by something a little more 21st century - something, perhaps, a little more like wikipedia.

9

I have written numerous Wikipedia pieces, and made contributions to others. Everything I entered was accurate… but I’ve found that, when Wikipedia’s staff finally gets around to editing what I’ve written, they change around the phrasing and/or chronology to the point where the piece no longer makes the least bit of sense!

To use one example, I added a number of facts and anecdotes to their entry on the rock group King Crimson. A few weeks later, bits and pieces of everything I added were still there, but they’d been re-written by someone who definitely didn’t speak fluent English. Coherent paragraphs were split up and turned into run-on sentences and re-positioned haphazardly throughout the article!

Hey, it’s THEIR web site. They don’t have to keep anything I write. They’re free to remove or edit anything I give them.

But the competence of the people making those decisions is, uh, highly questionable, at best.

10

Well, in order to notice them you would have to not be a victim of the popular misconception yourself, and, of course, many or even most people are victims of popular misconceptions. That is what “popular” means. Unfortunately, the only people who can (reasonably) reliably identify popular misconceptions are subject experts, but, by the nature of Wikipedia, the editorial power lies not with them but with amateurs.

Having said that, I have found that,in areas where I do have a degree of expert knowledge, Wikipedia quite often (but far from always) seems to do better in this regard than one might expect, but of course, I can’t tell when the topic is one I don’t know that much about, which is, necessarily, the vast majority. The trouble is that the potential for this sort of problem is a structural feature of Wikipedia, that cannot be eliminated without radically changing the way it is organized (which would very likely cause the whole project to fail).

psythe: It is true that experts have their biases, but they are generally aware of those biases, and have a good understanding of the other expert points of view, even where they disagree with them. Academics are also trained to give a fair presentation of opposing points of view, even where they strongly disagree with them. The problem with non-experts and (even worse) enthusiasts for a topic without the proper academic training, is not that they have their own point of view and are biased towards it (as is everybody), but that they often do not know that their biases are biases, that there is another possible way of looking at things, or, if they do know, they may not care.

11

I had one decidedly negative experience with Wikipedia very early on. I’m trying to remember the chain of events, but to tell the truth my memory has faded a bit. So this may not be 100% reliable (like Wiki :stuck_out_tongue: )

One day at work I was reading an article which was related to coal, and found some facts about coal which were clearly incorrect. IIRC there are at least a dozen major errors, and a handful of things which needed more clarification or were misleading by omission. So I spent about an hour and fixed everything, and added citations to academic papers or textbooks to support each change.

The next day I went out to look at the article again - out of vanity, because I was proud I had helped out - and every single one of my changes had been removed. I thought that it just somehow hadn’t “taken”, so I laboriously re-typed everything back in. This time I was smart enough to save the webpage after I was done.

That night, almost all my changes were gone. So I looked into it and discovered that someone had been responsible for both edits. So I tried talking on the discussion page, asking why my additions had been removed. The first few responses were very rude and unhelpful - people posted that my additions had been removed because they weren’t cited (which was a lie, and clearly visible in the logs), that the article was “established” and thus shouldn’t be changed by any random dumbass like me, etc. I argued with them and dug up multiple cites to support why my changes were correct - and really folks, the things I was changing were so clearly wrong, it was ludicrous.

And so I changed it back and the same used removed my changes. So I opened a dialogue with the user who kept undoing my edits. Allow me to summarize a week of psychodrama with the person:

  1. They were a 2nd-year engineering student in Germany.
  2. They had not even taken a class on power plants, but one day in class their professor “told them some things” about how “power plants really work.”
  3. They refused to give the name of their professor.
  4. They refused to admit that they had even possibly misunderstood the professor.
  5. All the things I found which were wrong with what they had posted were in fact wrong, despite me being able to have 10 cites for their hearsay cite they had.
  6. Even extremely blatant logical mistakes they made were still “right.”
  7. They refused to even read my cites, claiming that I probably had “faked” them (I faked peer-reviewed technical papers and got them published in the ASME and at EP/Power-Gen? Really? Really?)
  8. The Wiki people on the discussion page claimed that “a professor’s word is worth an awful lot!” When I pointed out that it was not the professor posting, but a student acting on incorrect hearsay, they fell back on how the student was a frequent contributor whereas I was just some newbie. I pointed out that I had been a professor myself, and was called a liar. I was even accused of being a “vandal.” I noticed when I looked at IP addresses of some of the commentators that they were exactly the same as that of the shit in Germany. :rolleyes:
  9. When it came down to it, the last exchange with this little shit in Germany involved them telling me flat out they had more time and more energy to undo my edits, and so I was going to lose.

So yeah, why bother editing it?

Still, many years later on - I think it was last year, in fact - I was reading an article and found they had badly misspelled the name of a famous female fencer. The spelling of her name is not in dispute, and the Olympic Committee website had her name spelled right. So I changed it.

Guess what? A week later, the misspelling was back. So I changed it and added the cite to the Olympic website. The next day, the misspelling was back. And this wasn’t a case where it was a foreign name or a name of questionable spelling, it was along the lines of “Ellen” spelled “EllnE.” Yet my changing of the text was seen as a “challenge” or “insult” to some vandal, and so it stayed misspelled for months. I checked this morning and it’s spelled right now.

Wikipedia needs some sort of vetting process for its editors. I’m sorry, but my opinion and citations on some issues are worth a fucking lot more than some random person who comes by after Googling around, and until there’s some sort of rating for editors, Wiki is always going to be treated with suspicion by me. The vandals and termites have to be reduced, if they can’t be eliminated.

12

Re: Is Wikipedia more reliable than the Encyclopaedia Britannica?

Cecil’s comments on Wikipedia show the typical misunderstanding of someone who’s not had much experience with how wikipedia is put together.

People who write for Britannica are chosen by editors because of their expertise in the field and write the articles for money. People who write for Wikipedia are not experts and are not paid. The motivation for putting material into Wikipedia is the same as that for putting up any sort of web page: to show off supposed expertise; to push a point of view; for personal gain. Altruism rarely figures.

The process for getting material into Wikipedia is based on confrontation. One person puts it in, a second takes it out, the first puts it back in, the second takes it back out, and so on until one gives up and goes away in a huff. Sometimes the two parties reach a compromise of sorts that holds until a third person sticks their oar in. The whole process tends to be a real dogfight. The people most likely to win at this are the cranks, fanatics, and those with a vested interest; experts fall by the wayside.

When reading something in Wikipedia the first thing you always need to ask is, why was it put there? If you read a glowing report about the small Scottish town of Dunreadhin, it was probably written by someone in the local tourist industry. An article dwelling on the dangers of smoking will be written by an anti-smoking fanatic, an article about a famous actress by her fans, an article on a religion by its adherants, an article on some wonder device by the people trying to sell it. And these people guard their articles to stop outsiders from making unwanted changes.

Cecil talks about whether facts in Wikipedia are true or not. But that’s only the start. Wikipedia articles regularly present opinion as fact, regularly present one sided views of contentious issues, regularly present issues in a totally unbalanced way. Conversely, its also possible to find minority views given as much prominence as the accepted consensus. The trivial is made to sound important, the important trivial. It all depends on who wrote it, why they wrote it, and what it took to reach a compromise.

The reliability of an encyclopedia is not just a matter of how many errors it makes. It also depends on the objectivity of the editors, the quality of the material, the balance in presentation, and similar things. Britannica is edited; it scores highly in such things. Wikipedia is not edited; when you get past the showpiece articles, it just doesn’t make the grade.

“Many Wikipedia articles are now wisely prefaced with warnings about dubious aspects of what lies below.” These aren’t editorial warnings. Each such warning is the result of a quarrel over the material being presented, the loser’s departing shot before bowing out of the fight.

13

You mean experts like Margaret Murray, whose rubbish article in the 1929 Britannica stayed around for forty years, eventually leading to a whole fake religion built around it, and which the Britannica has been apologizing for ever since? Those experts?

14

They may have gotten paid, but most writers for Britannica did not write for money; they wrote to get their name in the Encyclopedia Britannica or to get knowledge about their field out to the world.

They may or may not be experts. Of course, frequently an expert is not the person you want writing an article; Wikipedia is full of mathematical articles written by experts and accessible only to experts.

I suppose your point here is to push your point of view? The only way altruism rarely figures is if you define it narrowly and then think poorly of your fellow man.

Joseph Mccabe Lies Of Britannica » Internet Infidels is an old article on how the Catholic Church censored the 14th edition of the Encyclopedia Britannica. Biased, yes, but his claims about how articles were changed are verifiable. Have you ever read The Myth of the Britannica, by Harvey Einbinder? All about how carelessly the Britannica was written, and how much bias there was. A haigography of Calvin Coolidge was written for EB shortly after his death and hung around for decades, for example.

When reading anything, you need to ask why was it put there. The Encyclopedia Britannica is no different.

I doubt it; more likely an enthusiastic local. In any case, do you really think that Encyclopedia Britannica sent someone out to every small Scottish town? More likely, they grabbed what information they had, consisting of government statistical studies and the output of the local tourist industry, reformatted it encylopedically, and printed it.

It does not depend on the objectivity of the editors; the proof of the pudding is in the eating. Nor does it depend on the quality of the material, whatever that means. Grammar is not reliability. It depends on the reliablity of the encyclopedia.

For a more funny example then the ones quoted above, the article on stage magic in the Britannica once said that Harry Houdini was the greatest living magician. The article was written by Harry Houdini.

15

:confused: What do you mean “Wikipedia’s staff”? The people making those decisions were probably other random editors who did the same thing you did.

16

Not so. At least, not an intended feature. In light of this:

It in fact seems deliberately misleading.

17

Seem’s I didn’t make myself clear. :slight_smile:

People writing for EB are chosen by editors for their expertise and write their articles in return for money. That’s how encyclopaedias work. It means that their work is being checked by someone who has hired them and who is looking for accuracy, balance and so forth. The motivation is not to ‘win’, it’s to write something acceptable to the editor. It’s that chain of responsibility that gives you reliability. The system isn’t perfect but at least it’s a system.

I did not mean to imply that EB is perfect. No reference work ever is. But I’d trust EB. I know too much about how Wikipedia works to ever trust it.

18

Not intended, but that’s how it works in practice. I’ve had a lot of experience with the system over a number of years. Certainly there are rules, there are guidelines, there are processes for dealing with problems, but they don’t work. In the end persistence wins out. If you keep coming back and changing it eventually the other guys leaves. Unless you get two people who know the system and just won’t back down, because then it comes down to who’s got the most friends.

19

Again, many of them didn’t write for money. Getting paid was an incidental compared to the honor of writing for EB.

I.e. to ‘win’. You consistently describe Wikipedian motivations in negative terms, while describing EB motivations in positive terms.

It’s that chain of responsibility that put “Harry Houdini is teh best magacian evar!!!” into print and censored anything bad the Popes did.

That’s illogical. How much you know about system does not corrolate to how much trust you should put into it. In fact, when comparing disparate systems, I’d ignore the system and look solely at the output, since that’s what matters.

20

Somehow this all brings to mind the first statement in the Book of Bokonnon:
“All of the true things I am about to tell you are shameless lies.” :smiley:
(for you poor unfortunates unfamiliar with the Book of Bokonnon, see Cat’s Cradle by Kurt Vonnegut)