Wikipedia’s use in education continues to be controversial. I think that’s a pity, because the Wikipedia contains huge amounts of useful information, and - perhaps even more importantly - using it teaches very important lessons in the analysis of information sources.
In this post I’m going to summarize a few simple methods for analyzing any given Wikipedia page for likely reliability without having any knowledge of the subject matter the page is addressing.
Firstly, Wikipedia has a number of ways of notifying readers directly that an article may have issues with bias, reliability, accuracy or may suffer from various other problems. These problems are highlighted with a banner on the page (or section) which is believed to have issues.
Some examples are show below:
These banners are attempts by the Wikipedia community to alert readers to problems which the community can agree on.
Other problems aren’t highlighted as obviously, so the reader needs to do some work to understand how the article was developed.
To understand this, it’s problem best to address a specific article. Joseph McCarthy was a controversial figure in US politics during the 1950’s. His wikipedia article is a good example of how many wikipedia articles about about controversial topics suffer in comparison to other sources.
Attached to each article in wikipedia are two very useful pages; the history page and the talk page. The history page records the history of that article by showing who made what edit when, while the talk page allows co-authors to communicate about the article.
Looking at the history page, we can see that the Joseph McCarthy article is fairly old - the earliest version was from February 2002. This period of time should have given the article a chance to mature, and so you would expect the rate of editing to slow down over time. Unfortunately, in this case this doesn’t appear to be the case.
| Month |
Number of Edits |
| Jan 2007 |
70 |
| July 2006 |
52 |
| Jan 2006 |
166 |
| July 2005 |
75 |
| Jan 2005 |
18 |
| July 2004 |
16 |
| Jan 2004 |
3 |
| Jul 2003 |
41 |
| Jan 2003 |
3 |
| July 2002 |
0 |
| Feb 2002 |
1 |
It’s indicative of the kind of article this has become that since 2005 it has consistently been in a state of flux as authors with different political persuasions try and put their spin on it.
There are a huge number of other ways to analyze Wikipedia articles, and I could write for hours about them. But this post has already taken me a long time to get to this stage, so instead I’ll point to a Google Tech Talk on patterns in online communities called “Augmenting Social Cognition: From Social Foraging to Social Sensemaking“. In particular, around 28:30 minutes into it theres a great discussion about a statistical model which has been developed to predict Wikipedia articles in conflict with a accuracy of greater that 80%. Wouldn’t it be great to have a browser toolbar which could automatically warn students about the potential for a particular Wikipedia article to be controversial?
