Why Wikipedia Articles Vary So Much In Quality

Hugh Pickens writes "A new study shows that the patterns of collaboration among Wikipedia contributors directly affect the quality of an article. 'These collaboration patterns either help increase quality or are detrimental to data quality,' says Sudha Ram at the University of Arizona. Wikipedia has an internal quality rating system for entries, with featured articles at the top, followed by A, B, and C-level entries. Ram and graduate student Jun Liu randomly collected 400 articles at each quality level. 'We used data mining techniques and identified various patterns of collaboration based on the provenance or, more specifically, who does what to Wikipedia articles,' says Ram. The researchers identified seven specific roles that Wikipedia contributors play (PDF starting on page 175): Casual Contributor, Starter, Cleaner, Copy Editor, Content Justifier, Watchdog, and All-round Editor. Starters, for example, create sentences but seldom engage in other actions. Content justifiers create sentences and justify them with resources and links. The all-round contributors perform many different functions. 'We then clustered the articles based on these roles and examined the collaboration patterns within each cluster to see what kind of quality resulted,' says Ram. 'We found that all-round contributors dominated the best-quality entries. In the entries with the lowest quality, starters and casual contributors dominated.'"

Re:Because different people write the entries?

[bunratty]

Why not take a few seconds to tag the article as needing cleanup [wikipedia.org], then letting the lazyweb do the work? Remember, it takes all kinds of editors to write good articles. You can focus on the work that's easiest for you.

back in the day

[circletimessquare]

there was a book called the cathedral and the bazaar

http://en.wikipedia.org/wiki/The_Cathedral_and_the_Bazaar [wikipedia.org]

it delineates the difference between bottom up and top down organization, specifically in regards to software development models like linux versus gnu

obviously, this overlaps thematically with wikipedia in that wikipedia was once a bazaar, and is now becoming a cathedral

regardless of which model is better for wikipedia, the pluses and minuses of the cathedral versus the bazaar models of software development should be instructive for what exactly wikipedia is winning, and losing, in its trade off between bazaar and cathedral

Roles

[lyinhart]

In the five years that I've been a Wikipedia editor, I've played most of these roles, but right now I'm definitely a watchdog. I primarily revert vandalism. It's a good way to stay out of edits wars. At this point, most of the stuff on Wikipedia is way too messy to clean up and/or improve. I'd rather clean up Cowboys Stadium on any given Sunday in the Fall than clean up content on Wikipedia. As for deletionists? They deal with the administrative (sigh) aspect of Wikipedia, while this study seems to be driven mainly on the content itself.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Wikipedia's Editors

[gsslay]

Why is it that editors think deleting articles somehow makes it better?

Because ;

- if the quality of Wikipedia is measured by averaging the quality of all its articles, deleting the crap raises the quality of Wikipedia.
- crap inevitably attracts more crap. If the crap articles weren't deleted they would multiply.
- crap pages, written by people who mistake Wikipedia for a free web-host for their fan site, give Wikipedia a bad name.
- if you can't find the good articles for stumbling over the crap, you're likely to stop looking and go some place else.

If crap pages weren't deleted Wikipedia would drown under them. Regardless of infinite disk space, or unlimited bandwidth. Wikipedia is essentially a database. If you fill a database with too much garbage it becomes useless, no matter how much data of true value in in there also. The noise to signal ratio becomes unbearable.

Re:My experience with WikiPedia

[bunratty]

Why not seek dispute resolution [wikipedia.org] in these cases?

Then add the citation

[tepples]

Equally nonsensical are the seemingly random insertion of [citation needed] tags on things that are matters of public record.

In that case, you can help Wikipedia by removing the {{citation needed}} and replacing it with <ref>name of the relevant public record</ref>.

Re:Then add the citation

[tepples]

Because verifiability is the core content policy of Wikipedia. If you don't like it, contribute to another web site such as Everything 2.

Re:Missing role: deleters

[crossmr]

Actually the policies themselves say that blogs and other self-published sources are never good enough unless they happen to be written by the subject of the article. Even then, they're not used for notability, but they can be used as reliable sources. The only time self-published sources can be used for anything not about the subject is in the case when its written by a recognized expert in the field. Even then, its reliable, but its usefulness in establishing notability is questionable. The threshold for inclusion for the vast majority of subjects is pretty simple:
Find yourself 1 (preferably two) articles by reliable sources that are independent of the subject, that give the subject significant coverage and aren't simply trivial mentions. Which includes things like a 2 sentence entry in a top X list, or trivial name drops like "This product is a lot like products X, Y, and Z but we find it to be much better" (and there is no further mention of X, Y, or Z in it). That's it.
This requirement satisfies two things:
1) Notability
2) giving you some information to at least form a basis of an objective article. If all you are doing is writing it based off their website or including things written on random blogs you have serious neutrality issues and really can't write a useful article based off of those things and keep it encyclopedic.