Jump to content


From Meta, a Wikimedia project coordination wiki
This page is part of the Proceedings of Wikimania 2005, Frankfurt, Germany.




Editing notes:

  • Short, summary

The Information Behavior of Wikipedia Users[edit]

  • Author(s):'' {{{...}}}
  • License: Phoebe Ayers
  • Slides: Phoebe Ayers
  • Video: {{{radio}}}
  • 'Note:' {{{slides}}}

About the slides: {{{agenda}}}

About license


<include>[[Category:Wikimani templates{{#blocked:}}]]</include>


Why do people who are not contributors use the Wikipedia? What questions does the Wikipedia help answer, and what information needs does it meet? What do users think of the site and the quality of the information that they find there? What implications does the Wikipedia as a source have for traditional library services? This paper discusses possible answers to these questions, using data from two short research studies, one involving the Wikipedia Reference Desk and one surveying users of the Wikipedia, as well a theoretical framework from library and information science. Findings include the fact that Wikipedia users tend to use the site for non-essential information needs, but also trust the information that they find on the site. It is possible that Wikipedia provides the framework for a new theoretical model of why people trust information they find. The paper concludes with suggestions for further research.


This brief presentation explores the information behavior of Wikipedia users -- people who are interested in the Wikipedia as a source of information, rather than as contributors to that information. This is important because after all, a reference work does little good if it´s not used. Furthermore, without knowing who uses the Wikipedia, those who do contribute to it are writing for an invisible audience. It seems self evident that many, many people do use Wikipedia for information, simply because of the number of hits the site gets. So who does use the site, what do they use it for, and what do they think of it?

I can tell you right now that I won´t answer those questions, but I do have some ideas for answering them.

This presentation is called the information behavior of Wikipedia users, and information behavior, briefly, refers in library and information science to the study of how people use and interact with information. We can hypothesize a number of reasons why someone might use Wikipedia, including the facts that it's easily available, appears in search engine results, or that it covers a large scope of information -- but it is also possible that people use the site because they believe that it's an accurate and authoritative source of information. Given that this is the case, I am interested in three things:

  • what people use the information from Wikipedia for (eg school, work, etc.),
  • whether they think the information they find there is accurate (which may not be equivalent to valuable),
  • and if so, why they think the information they find on Wikipedia is accurate.

As part of my work in the Library and Information Science master's program at the University of Washington, I did two brief, informal studies to look at these questions. The first anonymously surveyed askers and answerers of questions on the Wikipedia Reference Desk, with a survey posted on the Reference Desk. The Wikipedia Reference Desk (WRD) is a part of the community section on the English-language Wikipedia(1), where anyone can ask or answer reference questions. The second study involved in-depth interviews with people I personally knew who self-describe as Wikipedia users, but who rarely contribute to the site. I asked questions to try and get at my areas of interest.(2)

Briefly, the results of these studies were that people who use the WRD tend to ask questions that aren´t terribly important to them -- what we can call incidental or casual information needs. However, they do tend to trust the answers they receive there, even though they´re coming from unvetted strangers -- not a new concept to Wikipedians, but definately strange to traditional librarians. (In fact, this kind of information behavior may be rooted in other online communities, such as USENET, and thus any theories developed about users of the Wikipedia may be applicable to online communities generally.)

Similarly, the users of the Wikipedia that I interviewed typically went to the site for things that they were simply curious about. However, although they all said that they would verify the information they found with other sources if it was for an important need, e.g. work, in general they did trust and find accurate the information from the Wikipedia that they found. It´s important to note that I only interviewed people who were internet-savvy and who were familiar with how Wikipedia works, but it was particularly interesting that the back and forth editing process that is a hallmark of Wikipedia was considered both a pro and a con by the users I interviewed -- a pro because inaccuracies could be found by peer review, but a con because of edit wars, users inserting their own opinion into articles, and the like.

Basically, I view these two projects as pilot studies that will help to refine a larger, later study. I think that the consequences of knowing why users are using the site and what they think of it is tremendously important not just to the Wikipedia but also to my field of library and information science -- for libraries because we work hard to cater to what people want, and thus knowing what people turn to the internet for could be extremely valuable, and to information science because the Wikipedia represents such a new kind of source and potentially a new kind of information behavior.

Nonetheless, it is possible that the behavior of Wikipedia users can be fit into older, more established information science theories, particularly those by Patrick Wilson on the creation of authority, thus both expanding these theories and building a theoretical framework for Wikipedia behavior. Wilson states briefly that there are several factors in determining a text's authority and whether a reader will trust it. These factors include publisher authority, author authority, and whether information fits with the prior experience of the reader (3). Since Wikipedia until recently didn't have publisher authority, and as we all know there is no distinct authorship to articles, it is thus remarkable if readers do trust the information they find on Wikipedia, suggesting that a new type of authority matrix is being generated.

In the future more detailed studies, either expanded qualitative surveys or anonymous quantitative studies, should help answer these questions. One difficulty is finding an appropriate methodology to survey users without violating their privacy or disrupting their experience using the Wikipedia. In short, I´m left with a lot of questions, and there´s much work left to be done.

1. And other language Wikipedias; however, I dealt exclusively with the English version.

2. This study was done with Michelle Batchelor, now of UNLV Libraries; bchelle .at. gmail.com

3. Wilson, P. (1983). "Cognitive Authority" in Second-Hand Knowledge: an Inquiry into Cognitive Authority. Westport, Conn.: Greenwood Press.