Social construction of knowledge in Wikipedia
First Monday

Social construction of knowledge in Wikipedia by Noriko Hara and Jylisa Doney



Abstract
This paper investigates how knowledge is constructed collaboratively in a crowd-sourced environment. More specifically, the study presented in this paper empirically analyzes online discussions in regard to Wikipedia entries on the Fukushima Nuclear Power Plant Disaster that occurred in March 2011 in Japan. For this study, we examined the encyclopedia articles in both the English and Japanese versions of Wikipedia. The findings indicate similarities and differences between the two language versions. The implications of the study for collaborative knowledge production are also discussed.

Contents

Introduction
Collaborative knowledge production
Case of Fukushima Nuclear Power Plant Disaster in Wikipedia
Methods
Findings and discussion
Conclusions

 


 

Introduction

In the past, “experts” with proper training and resources to conduct research produced scientific knowledge. The produced scientific knowledge was then distributed appropriately to ordinary citizens who would apply the given scientific knowledge into everyday practice. However, this boundary work (Gieryn, 1983) between experts and non-experts has been challenged recently with the prevalence of Web 2.0 applications, such as Wikipedia, Ask.com, and Yahoo!Answer. Some (Qualman, 2011; Surowiecki, 2005) went as far as to claim that we do not need experts because a crowd of amateurs is better than a limited number of experts. Yet, others (e.g., Keen, 2008) are more critical of amateurs being praised for their contributions online. Jasanoff, coming from the area of science and technology studies, argues that scientific knowledge does not exist in a “void”. It is a result of co-production of social, political, and natural influences (Jasanoff, 2004). Due to the rise of social media, more opportunities are available for ordinary citizens to influence scientific knowledge. Faraj, et al. (2011) provided theoretical frameworks to analyze knowledge collaboration (i.e., knowledge sharing and construction) in open online communities, such as Wikipedia, Facebook, and YouTube. They discuss how knowledge collaboration in open online communities profoundly differs from traditional knowledge collaboration within organizational boundaries, and Faraj, et al. call for more research in this area. With these debates in mind, this paper examines how knowledge is co-constructed among ordinary people in a crowd-sourced environment, namely Wikipedia.

Wikipedia is an online encyclopedia with over 280 different language versions, and its English version alone contains over four million articles. The English Wikipedia is the most prominent version and has over 130,000 active users. As of 5 May 2015 the Japanese-language version of Wikipedia has over 965,000 articles (13th) and over 12,000 active users (5th)(see http://meta.wikimedia.org/wiki/List_of_Wikipedias). In addition to the articles being visible to the general public, Wikipedia provides an online discussion space for each article. This function offers a rich socio-technical milieu for examining how knowledge co-construction is negotiated in an open online environment.

 

++++++++++

Collaborative knowledge production

Empirically, there are several researchers who examined collaborative knowledge production on Wikipedia (e.g., DiStaso, 2013; Introne and Drescher, 2013; Kane, et al., 2014; Keegan, et al., 2011; König, 2013; Messner, et al., 2014; Swarts, 2009). For example, Swarts examined coordination to support knowledge construction on Wikipedia articles about clean coal technology using Actor Network Theory. Swarts also analyzed how facts are created and how this act is linked to “a network of facts, references, links, and alliances” that are not easy to undo [1]. He identified two types of “rhetorical moves.” One is called “opening moves” which are actions that editors take to challenge or support the other editors “that supported” [2] a fact. The other is called “closing moves” in which editors attempt to move away from the discussions of a fact and accept the fact as it is. In three years, the clean coal technology article was edited 95 times, and 40 and 60 opening and closing moves were identified, respectively. By analyzing the editing processes on Wikipedia, Swarts shed light on the collaborative nature of creating facts on Wikipedia and claims that the construction of facts in Wikipedia is very similar to the process in which scientific facts are developed.

Keegan, et al. hypothesize that, after catastrophes, Wikipedia may support information dissemination and disaster response and investigated the co-authorship structures related to the Tohoku Earthquake in 2011. Extracting data within one month from the earthquake, they collected 84 articles focusing on the “Tōhoku earthquake and tsunami.” In order to provide accurate information about the number of missing, injured, and casualties several separate template pages detailing this information were created. They concluded that Wikipedia plays numerous roles as a source of cultural memory and as a historigraphical tool.

König (2013) examined the German Wikipedia article about the 11 September 2001 attacks in the U.S. and discussed the issue of democratization of knowledge. He conducted a content analysis of discussion pages and asked the questions of how knowledge is produced in this open environment and whose voices are heard and ignored. König found that, even though Wikipedia discussion pages allow non-experts to create information, usually the information is marginalized if it contradicts majority interpretations. On the German Wikipedia 9/11 discussion page, established authorities created the “official story,” and alternative opinions were labeled “conspiracy theories” and marginalized. While the alternative opinions were mostly sidelined, they were still mentioned within the main article and linked to their own discussion pages in Wikipedia. According to König, this case study shows a type of democratization where non-experts and experts alike can participate to produce knowledge in a novel way. At the same time, König warns that the participatory nature of Wikipedia does not necessarily lead to a more democratized process of knowledge production and further study is needed.

Messner, et al. (2014) examined the entries regarding nutrition such as high fructose corn syrup in Wikipedia and found that Wikipedia articles on nine kinds of nutrition always ranked top three on three major search engines. In addition, they analyzed the reference sources used in these articles. Messner, et al. identified that the use of media reports had declined between 2007 and 2011. Meanwhile, the use of academic articles had increased during the same period. Their findings indicated that ordinary citizens rely more on Wikipedia articles on health related topics, such as nutrition, and that these articles had seen an approximate 25 percent increase in using more reliable academic sources such as the Journal of the American Medical Association.

Even though these studies reveal some aspects of how co-construction of knowledge occurs in a crowd-sourced environment, particularly in Wikipedia, some questions remained unanswered. First, there is a shortage of research on cross-language examination in Wikipedia because the majority of scholars focus on the most visible (English) Wikipedia (Fichman and Hara, 2014). This paper addresses the following research question to fill the lacuna: How does the global (English Wikipedia) differ from the local (Japanese Wikipedia) practices of knowledge co-construction? Second, we still need a better grasp of how users co-produce knowledge in an open online encyclopedia. Building on the studies by König (2013) and Swarts (2009), we identify the types of activities in which users engage. Third, given König and other scholars’ assertions regarding the “democratization of knowledge” in crowd-sourced environments, we question how the roles of “experts” are perceived in both the English and Japanese Wikipedias.

 

++++++++++

Case of Fukushima Nuclear Power Plant Disaster in Wikipedia

For the current study, we chose to examine the Wikipedia articles on the Fukushima Nuclear Power Plant Disaster that occurred in March 2011 in Japan. Although the incident occurred in a particular geographic area, the plant’s proximity to the ocean and concomitant air and tidal dispersal of leaked radiation made the Fukashima disaster an international incident discussed in traditional and social media outlets. This allowed us to compare a local Wikipedia (Japanese) with the global Wikipedia (English). By using the data available in the online discussion space attached to each article (called “Talk” pages) we examined the process in which co-construction of knowledge occurred in this specific online environment in both the English- and Japanese-language versions.

The basic details of the incident are as follows. On 11 March 2011 at 2:46pm local time, a 9.0 magnitude earthquake struck Japan’s northeastern coast (U.S. Geological Survey, 2011). At 3:27pm local time the first tsunami struck the Fukushima Nuclear Plant, with a larger 46 ft. tsunami striking the Fukushima plant at 3:46pm (Wikipedia, 2013a). As a result of this disaster, 15,883 people were killed and 6,145 people were injured (Wikipedia, 2013b). Following the earthquake and tsunami, residents living near the plant were evacuated (Wikipedia, 2013a), as a meltdown of the Fukushima facility had occurred (Ryall, 2011a; Ryall, 2011b). In the days and months after the Fukushima meltdown, high levels of radiation were discovered in the water near the Fukushima reactors (Matsutani, 2011), in the air surrounding Fukushima (Nakamura, 2011), and in vegetables and tap water in Tokyo (CNN Wire Staff, 2011). Radioactive water from this disaster was also dumped into the sea (Nakamura, 2011). Full decommissioning of the Fukushima Daiichi plant is expected to take 40 years (Japan Atomic Industrial Forum (JAIF), 2011).

 

++++++++++

Methods

We first analyzed the general edit statistics on the Fukushima Daiichi Nuclear Disaster page in the English Wikipedia and Japanese Wikipedia. The data for the English page came from the Revision History Statistics page and the Page View Statistics page located under the “View History” tab on the “Fukushima Daiichi Nuclear Disaster” Wikipedia page. The links to these pages led us to a tool called “Article revision statistics — X!’s tools.” This tool allowed us to extract similar data for the Japanese page.

Next, we conducted a content analysis of Wikipedia Talk pages where discussions of each entry occurred between the first post made and 19 April 2013. Content analysis is a way to systematically analyze what types of content appears in texts. Typically, a coding scheme is developed first, and texts are coded accordingly. In this case, after reading through the posts on both the English and Japanese pages, a coding scheme was developed based on previous literature on co-production of knowledge (e.g., Jasanoff, 2004; König, 2013; Swarts, 2009) (see Table 1 for the coding scheme). The coding of the English page was conducted by one of the authors, and the other author separately coded 10 percent of the posts in the English page in order to check inter-coder reliability — this percentage is aligned with previous research (e.g., Hara, et al., 2010). The inter-coder reliability was calculated by Cohen’s Kappa (Multon, 2010), and the results were: 0.766, 0.543, 0.29, 0.419, 0.306, and 0.429 for the seven categories (Editorial, Facts about the Event, Media Report, Background Knowledge, Citations, Resources, Definition respectively) using ReCal (Freelon, 2010). The discussions on Talk pages posted up to 24 September 2013 were analyzed both in English and Japanese pages.

 

Table 1: Coding scheme.
Coding nameDefinition
EditorialComments related to the editorial aspects of the article, such as the length of the article, naming of categories, etc.
Facts about the eventConfirmed facts about the event, such as the number of people injured, the current status of the power plant, radiation level, etc.
Media report about the eventReporting of media coverage about this event, e.g., U.K. Guardian reported ... .
Background knowledge about the disasterSome background knowledge about the nuclear power plant disaster, e.g., how much water is necessary to keep the nuclear core stabilized
CitationsDiscussions about citation sources
ResourcesSharing of any useful resources related to the event (not relevant to the Wikipedia article per se)
DefinitionsDefinitions of terminology related to the accident
VandalismDiscussions about how to deal with malicious users

 

 

++++++++++

Findings and discussion

Overview of Fukushima Daiichi Nuclear Power Plant Disaster in English and Japanese Wikipedia

To begin studying knowledge construction in Wikipedia, we initially focused on the page edits and page views that occurred in a one year period between March 2011 and March 2012. Page edits within the English “Fukushima Daiichi Nuclear Disaster” Wikipedia page experienced a drastic decrease over time. The majority of edits (3,759) occurred in March 2011, when the tsunami and nuclear disaster occurred (see Figure 1). Based on our analysis of the number of edits per month, we concluded that editing peaks occurred in March 2011 (3,759), August 2011 (187), December 2011 (131), and March 2012 (68). Through our analysis of news reports surrounding the Fukushima Nuclear Disaster, we were able to make connections between peak edits of the English Wikipedia page and newly reported “news” events related to the disaster. It appears that the number of Wikipedia edits was negatively correlated with the number of months since the disaster. Therefore, page edits were highest when the event occurred and decreased as the time since the disaster increased. The number of page views between March 2011 and March 2012 did not appear to be related to the number of edits. Page views of the English “Fukushima Daiichi Nuclear Disaster” Wikipedia page remained relatively constant between March 2011 and March 2012, and peaked in November 2011 with a total of 287,888 page views. Page views also peaked in March 2012, a year after the disaster occurred, with 147,581 page views.

 

Edits of Fukushima Daiichi Nuclear Disaster page in English Wikipedia between March 2011 and March 2012
 
Figure 1: Edits of Fukushima Daiichi Nuclear Disaster page in English Wikipedia between March 2011 and March 2012: “Minor” indicates minor edits, and “IPs” indicates edits made by IP users.

 

Then, we compared the general tendencies of edits in the English and Japanese Wikipedias (see Table 2). The total revisions refer to the total number of edits for 2011–2013 — i.e., the data extracted from the Revision History Statics page.

 

Table 2: Statistical comparisons of English and Japanese articles for Fukushima Daiichi Nuclear Disaster.
 English WikipediaJapanese Wikipedia
Total revisions6,812
(average 159.1/article [3])
2,191
(average 58.0/article)
First edit3/12/2011
15:50:08
3/12/2011
13:42:05
Average time between edits0.14 days0.42 days
Number of editors624461
Number of edits made by the top 10% of active users4,499
(66.05%)
1,394
(63.62%)
Number of edits in 20115,749
(84.39%)
1,759
(80.29%)
Number of edits in 2012561
(8.24%)
350
(15.97%)
Number of edits in 2013502
(7.37%)
82
(3.74%)
Date of last edit12 March 201428 January 2014

 

Table 2 indicates that the majority of the edits (over 80 percent) in both language versions were made in 2011, the year in which the incident occurred. This makes sense because the incident attracted more media and popular attention immediately after it occurred. As with any other event, popular attention paid to the incident decreased as time passed, although sporadic media reporting on the incident continues as of May 2014 (e.g., Fackler, 2014).

While the number of active users in the English Wikipedia as a whole is 10 times greater than that of the Japanese Wikipedia, the number of editors listed in the entry for the Fukushima Daiichi Nuclear Disaster in these two language versions is less disparate: 624 and 461 in the English and Japanese pages respectively. The increased participation of Japanese editor makes sense given the fact that the disaster occurred in Japan, and, as such, one would expect it to be of major interest to the Japanese language speaking population. However, the total number of revisions is three times greater on the English page than the Japanese page. To put this in context, the average number of revisions per article was calculated for the entire language version of Wikipedia — 159.1 for English and 57.9 for Japanese respectively. The total number of revisions for this particular article was divided by the average number of revisions per article, for each article. They came out about the same: 45.2 for English and 37.7 for Japanese Wikipedia respectively. This means that both the English and Japanese language versions of the article attracted about the same level of interest, slightly more in the English Wikipedia, from the editors. Furthermore, edits in the English Wikipedia occur three times more frequently than the Japanese. This is not necessarily explained by the ratio of the number of editors in the English and Japanese versions, which is only 1.35. Indeed, even though the English version does not have three times more editors than the Japanese version, it is updated at faster rate, and it seems that this increased frequency is the norm of the English Wikipedia — the average rate at which articles are revised in the English Wikipedia in general is within a few minutes (Viégas, et al., 2007; Geiger and Ribes, 2010), and the fact that the editors come from all over the world, means that editors are at work almost 24 hours a day, just like global virtual teams (see e.g., Ruppel, et al., 2013).

Content analysis

The results of the content analysis are shown in Table 3.

 

Table 3: Results of content analysis.
Coding nameEnglish WikipediaJapanese Wikipedia
Editorial224
(51.97%)
40
(71.43%)
Facts about the event72
(16.7%)
3
(5.36%)
Media report about the event55
(12.76%)
2
(3.57%)
Background knowledge about the disaster24
(5.57%)
3
(5.36%)
Citations12
(2.78%)
9
(16.07%)
Resources102
(23.67%)
2
(3.57%)
Definitions7
(1.62%)
2
(3.57%)
Vandalism5
(1.2%)
2
(3.57%)
Others7
(1.62%)
6
(10.71%)
Total number of treads43156

 

As Table 3 indicates, Editorial content was the most actively discussed content (51.97 percent for the English Wikipedia and 71.43 percent for the Japanese Wikipedia respectively). This is not surprising as the discussion of edits is the reason why most people participate in Wikipedia Talk page discussions. However, users in the Japanese Wikipedia were more likely to engage in editorial revision (and discussion of editorial revisions) than users in the English Wikipedia — at least with regard to the Fukushima Daiichi Nuclear Power Plant Disaster. The second most discussed content in the English Wikipedia for this entry was Resources (23.67 percent); however, Citations (16.07 percent) was the second-most debated content in the Japanese Wikipedia. This was rather counterintuitive, and we will discuss this finding in detail later. Another remarkable finding was that three out of six entries in the category of “Others” in the Japanese Wikipedia are related to determining who the experts are. They debated who should be considered “experts” on nuclear power plant accidents. For example, they debated whether or not politicians should not be considered experts after one editor cited a quote from a politician. This reflects the findings by Gauchat (2012) who, by analyzing the General Social Survey data between 1974 and 2010, argued that science has been politicized in the U.S. For instance, public trust in science had been declining among political conservatives — especially when compared to moderates and liberals. In contrast to the Japanese Wikipedia entry on the Fukashima disaster, of the seven entries in the “Others” category in the English Wikipedia, none are related to the issue of experts. Interestingly, two of the seven were related to users thanking editors for their work on the Wikipedia page and for updating information in real time.

English Wikipedia

When we examined the content of this entry in the English Wikipedia, three themes emerged. The first was that the discussions on the Talk page of this entry were highly editorial in nature. The editorial discussions focused on cleaning up the content and text within the Fukushima article. For example, the editors spent significant time debating whether or not to split the article into several separate articles because some felt that the article was becoming too elongated. This finding, i.e., that editorial issues were the most discussed, was expected. As users are editing an encyclopedia article collectively, they need to coordinate and talk about how to construct the article in a coherent manner. For instance, one of the discussions coded under the category of Editorial was a request not to use present tense for the article:

Some parts of the article (especially ‘Explosion’, which I corrected, but may get reverted by someone) were written in the present tense — which as well as being un-encyclopedic, will look stupid by even tomorrow. — Preceding unsigned comment added by Malau (talk • contribs) 11:05, 12 March 2011 (UTC)

The second theme questioned what types of information were reliable sources to cite in the article. The discussions on the Talk page surrounded the question of whether live blogs, media sources, experts in the field, and even nuclear agencies were reliable sources. Since the incident occurred, daily updates from various sources were available. Some major news media, such as the New York Times and Guardian, had set up live blogs that were updated frequently, if not every hour. In this kind of situation, “facts” changed rapidly, and new facts were found on a daily basis. Some editors suggested waiting for reliable sources. For example:

“Can we not simply wait a few moments until details are published in reliable sources, rather than adding minute-by-minute updates sourced to breaking new blogs?” — Pontificalibus 15:03 15 March 2011 (UTC)

“A live blog, even if from a reliable source, is not considered reliable as continuous content.” — Datheisen 21:23 12 March 2011 (UTC)

“Ok, but give one — three days until we have some good overviews/sources published.” — Chris.urs-o 21:27 12 March 2011 (UTC)

“You mean we should continue to add information from unverified blog posts until better sources publish verified information? Sorry, but no.” — Pontificalibus 21:54 12 March 2011 (UTC)

Which sources to trust was a major issue of the debates among the users in Wikipedia, probably more so for the article featuring this incident than other more mundane article topics.

Third, in relation to the issue of reliable sources, the users in the English Wikipedia engaged in discussions editorial privilege, i.e., who should be allowed to edit the article? One of the most frequent topics the users discussed was whether or not IP users should have the same editing privileges as registered users. IP users are those users who do not register on the Wikipedia site. As such, their comments are signed by IP addresses, instead of unique usernames. Since IP users do not have an explicit history of editing Wikipedia and do not care to register as a user of Wikipedia, they argued for and against the editing by IP users.

Japanese Wikipedia

We found four similar, but also slightly different themes in the Talk page of this entry in the Japanese Wikipedia. First, the users in the Japanese Wikipedia of this article focused primarily on discussing “how to write the article.” Contrary to our prediction (and unlike the behavior of English users), Japanese users did not spend much time talking about the incident — again, focusing instead on how to write the article. And while this is exactly what users are supposed to do on Talk pages, it was surprising to find that Japanese editors did not deviate despite how pertinent the incident was to the Japanese-speaking population.

Second, the Japanese Wikipedia has a policy that, when editing an article, users need to discuss the changes in the Talk page first before making revisions, particularly when the changes are major. However, the Fukushima Daiichi Nuclear Power Plant Disaster page allowed free-edits due to the urgency of the incident. This finding surprised us given the degree to which Japanese culture values consensus and collectivism through the maintenance of cohesive group relationships (Hofstede, 1983; Shachaf, 2008). As a result, we speculate that the nature of the discussions in the Japanese Talk page for this article are slightly different from Talk pages for other articles.

Third, similar to the English Wikipedia, several heated discussions on “what sources are reliable” were found in the Talk page of this entry. For example, some discussions were based on the fact that quotes in newspaper articles do not identify specific individuals (i.e., How can we trust anonymous quotes coming from supposedly reliable and scientific sources/individuals?). Another issue was that there was a lack of trustworthy sources available immediately following the incident. Still, some editors argued that the same level of quality found in the citations of other Wikipedia articles should be presented, especially after a few months passed after the incident. By then, there would be more dependable sources, and the editors argued that questionable sources should then be eliminated.

Finally, many users how difficult it was to keep up with the flow of information relevant to the disaster. Some users mentioned the inclusion of confusing information. For instance, one user commented:

“Initially, we could not help but have the article be unclear due to the excessive amount of information and text; we just began to clean up the content in August [2011]” — Kasumin777 03:52 12 September 2011 (UTC).

The same author also asked for better quality for the article later in the same month:

“A nuclear power plant accident is a science and technology topic ... please do not produce a lower quality article ” — Kasumin777 11:21 28 September 2011 (UTC).

However, others argued that:

“In my opinion, there is no other better information source available regarding this accident” [because of the Wikipedia article’s extensiveness and timeliness] (no signature).

English vs. Japanese Wikipedia

When comparing the Fukushima Daiichi Nuclear Power Plant Disaster article in the English and Japanese Wikipedias, there are some similarities and differences. In terms of similarities, the discussion on the Talk pages in both the English and Japanese language versions focused on editorial issues. This was expected, as the purpose of Wikipedia in both versions is to collectively produce an online encyclopedia. Both versions also addressed the issue of reliable sources, i.e., what constitutes scientific fact and what information was simply amateur speculation.

As for differences, the Japanese Wikipedia user community stayed focused on writing an encyclopedia article. As stated, we originally speculated that Japanese editors would be more inclined to discuss the nature of accident, assigning the writing of the article a lower priority, and that they would be more inclined to use the Talk pages to share information and resources related the disaster instead of discussing how to edit the article. However, our findings demonstrate the opposite outcome. The contributors to the English Wikipedia for this article were much more involved in discussion of the new information relevant to the accident in the Talk pages, although they did engage in some discussion of the editing of the article. In addition, we noticed that the English Wikipedia community used the Talk pages to share more resources about the accident than the Japanese Wikipedia community — i.e., it was the users on the English Wikipedia who wanted to go beyond mere editing of an encyclopedia entry online and use the Talk pages for a more broadly themed discussion.

There are three possible explanations for this. First is that the Japanese culture tends to be more task-focused and process-oriented (Shachaf, 2008). They were given an environment (Wikipedia Talk page) in which to discuss the editing of the article, and this is what they did. Second, the contributors in the English Wikipedia may have had previous experience with similar pages where users discussed issues other than the editing of the article. Research has shown that, within the English Wikipedia, posts on Talk pages also include discussions on planning and coordination, requests for information, and references to Wikipedia policies (Viégas, et al., 2007). Third, there were more accessible and useful platforms already in use among Japanese users that were used to share resources and discuss the actual incident itself. For example, Twitter was used extensively to share resources for the Fukushima Nuclear Power Plant Disaster (Thomson, et al., 2012).

 

++++++++++

Conclusions

We examined the Talk pages of the Fukushima Daiichi Nuclear Power Plant Disaster articles in the English and Japanese versions of Wikipedia. We identified that although there are similarities, the discussions on the Talk page in the English Wikipedia, which tends to attract a more global audience, differs from the Talk page in the Japanese Wikipedia, which tends to include information contributed by local, i.e., Japanese, users in terms of practices related to knowledge co-production. Jasanoff (2004) claims that one of the research areas in co-production is how universal knowledge exists in diverse cultural settings, and this paper addresses the processes by which “facts” are developed in two different cultural settings.

The study has its own limitations. For example, we chose a highly visible and controversial scientific topic, i.e., a nuclear power plant disaster. The online behaviors that we observed may be different in more established scientific topics. In addition, a coding scheme for the content analysis could have been developed to include additional categories. In fact, a revamped coding scheme is being implemented for our next project. Third, we relied on visible online behavioral data for the analysis. It is possible that it would have been more enriching to collect additional data, such as interviews; however, this was beyond the scope of the current project.

Despite these limitations, we believe that this study can shed light on the nature of collaborative knowledge construction in an online crowd-sourced environment, especially in terms of scientific knowledge. One of the important issues that the users mentioned was that it was difficult to contribute knowledgeable information relevant to scientific topics, like a nuclear power plant disaster. With the prevalence of social media, ordinary citizens tend to rely more on social and collaborative media to find “authoritative” answers to scientific and historical questions. Currently, Wikipedia is perhaps the most popular online source lay people turn to for information. As such media becomes more popular, we are likely to so see what some scholars (e.g., König, 2013) call the “democratization of knowledge”. Whether tools like Wikipedia come to be regarded as truly authoritative sources is yet to be seen. However, there are other successful examples of citizens contributing scientific knowledge, such as Galaxy Zoo [4] (Raddick, et al., 2010; Wiggins and Crowston, 2011), a Web site (http://www.galaxyzoo.org/) which invites volunteers to classify data sets of millions of galaxy images. Given that scientists are sometimes criticized for their tendency to hoard knowledge, it is liberating to see that knowledge production and distribution no longer rest solely in the hands of the scientists. At the same time, knowledge produced by ordinary citizens may not be entirely reliable, despite the best efforts of these amateur investigators. The challenge lies in determining how authoritative knowledge that is co-produced by ordinary citizens can be with regard to scientific knowledge.

In future studies, we plan to examine online co-construction of knowledge with other contentious scientific knowledge, as well as more established scientific knowledge. And although the issue of experts was not discussed in as much depth as we had expected, other domains of scientific knowledge may yield a more explicit discussion of “experts.” We will investigate this research question further in the future. End of article

 

About the authors

Noriko Hara is an associate professor in the Department of Information and Library Science, School of Informatics and Computing at Indiana University Bloomington. Her research in social informatics emphasizes online knowledge sharing, communities of practice, and collective behaviors in mediated environments.
E-mail: nhara [at] indiana [dot] edu

Jylisa Doney is the Social Sciences Librarian at North Dakota State University. She received an MLS degree from Indiana University in 2014, and a BS degree in Sociology from Utah State University in 2012.
E-mail: jylisa [dot] doney [at] ndsu [dot] edu

 

Notes

1. Swarts, 2009, p. 283.

2. Swarts, 2009, p. 284.

3. The number of average revisions per article was calculated based on the total number of edits divided by the number of articles for each language Wikipedia. The statistics were available at: http://meta.wikimedia.org/wiki/List_of_Wikipedias.

4. Galaxy Zoo has been quite successful in terms of attracting citizen scientists — over 25,000 contributors have classified over 50 million galaxies (Wiggins and Crowston, 2011). In fact, the Galaxy Zoo project was so successful that it produced a collection of similar citizen science projects called Zooniverse (Francis, 2013).

 

References

CNN Wire Staff, 2011. “Bottled water scarce after Japan’s tap water shown unsafe for infants” (24 March), at http://edition.cnn.com/2011/WORLD/asiapcf/03/23/japan.nuclear.crisis/index.html, accessed 19 May 2015.

Marcia W. DiStaso, 2013. “Perceptions of Wikipedia by public relations professionals: A comparison of 2012 and 2013 surveys,” Public Relations Journal, volume 7, number 3, pp. 1–23.

Martin Fackler, 2014. “Panicked workers fled Fukushima plant in 2011 despite orders, record shows,” New York Times (20 May), at http://www.nytimes.com/2014/05/21/world/asia/fukushima-workers-fled-plant-after-accident-despite-orders.html, accessed 22 May 2014.

Samer Faraj, Sirkka L. Jarvenpaa, and Ann Majchrzak, 2011. “Knowledge collaboration in online communities,” Organization Science, volume 22, number 5, pp. 1,224–1,239.
doi: http://dx.doi.org/10.1287/orsc.1100.0614, accessed 19 May 2015.

Pnina Fichman and Noriko Hara (editors), 2014. Global Wikipedia: International and cross-cultural issues in online collaboration. Lanham, Md.: Rowman & Littlefield.

Ruth Francis, 2013. “Guest blog — Galaxy Zoo: Exploring the universe from the comfort of your sofa,” BioMed Central blog (24 October), at http://blogs.biomedcentral.com/bmcblog/2013/10/24/galaxy-zoo/, accessed 20 May 2014.

Deen G. Freelon, 2010. “ReCal: Intercoder reliability calculation as a Web service,” International Journal of Internet Science, volume 5, number 1, pp. 20–33, at http://www.ijis.net/ijis5_1/ijis5_1_freelon_pre.html, accessed 19 May 2015.

Gordon Gauchat, 2012. “The politicization of science in the public sphere: A study of public trust in science in the United States, 1974–2010,” American Sociological Review, volume 77, number 2, pp. 167–187.
doi: http://dx.doi.org/10.1177/0003122412438225, accessed 19 May 2015.

R. Stuart Geiger and David Ribes, 2010. “The work of sustaining order in Wikipedia: The banning of a vandal,” CSCW ’10: Proceedings of the 2010 ACM conference on Computer supported cooperative work, pp. 117–126.
doi: http://dx.doi.org/10.1145/1718918.1718941, accessed 19 May 2015.

Thomas F. Gieryn, 1983. “Boundary-work and the demarcation of science from non-science: Strains and interests in professional ideologies of scientists,” American Sociological Review, volume 48, number 6, pp. 781–795.
doi: http://dx.doi.org/10.2307/2095325, accessed 19 May 2015.

Noriko Hara, Pnina Shachaf, and Khe Foon Hew, 2010. “Cross-cultural analysis of the Wikipedia community,” Journal of the American Society for Information Science and Technology, volume 61, number 10, pp. 2,097–2,108.
doi: http://dx.doi.org/10.1002/asi.21373, accessed 19 May 2015.

Geert Hofstede, 1983. “The cultural relativity of organizational practices and theories,” Journal of International Business Studies, volume 14, number 2, pp. 75–89, and at http://www.jstor.org/stable/222593, accessed 1 June 2014.
doi: http://dx.doi.org/10.1057/palgrave.jibs.8490867, accessed 19 May 2015.

Joshua E. Introne and Marcus Drescher, 2013. “Analyzing the flow of knowledge in computer mediated teams,” CSCW ’13: Proceedings of the 2013 Conference on Computer Supported Cooperative Work, pp. 341–356.
doi: http://dx.doi.org/10.1145/2441776.2441816, accessed 19 May 2015.

Japan Atomic Industrial Forum (JAIF), 2011. “Earthquake report,” number 291 (December), at http://www.jaif.or.jp/english/news_images/pdf/ENGNEWS01_1323921629P.pdf, accessed 1 June 2014.

Sheila Jasanoff (editor), 2004. States of knowledge: The co-production of science and social order. New York: Routledge.

Gerald C. Kane, Jeremiah Johnson, and Ann Majchrzak, 2014. “Emergent life cycle: The tension between knowledge change and knowledge retention in open online co-production communities,” Management Science, volume 60, number 12, pp. 3,026–3,048.
doi: http://dx.doi.org/10.1287/mnsc.2013.1855, accessed 19 May 2015.

Brian Keegan, Darren Gergle, and Noshir Contractor, 2011. “Hot off the wiki: Dynamics, practices, and structures in Wikipedia’s coverage of the Tōhoku catastrophies,” WikiSym ’11: Proceedings of the 7th International Symposium on Wikis and Open Collaboration, pp. 105–113.
doi: http://dx.doi.org/10.1145/2038558.2038577, accessed 19 May 2015.

Andrew Keen, 2008. The cult of the amateur: How blogs, MySpace, YouTube, and the rest of today’s user-generated media are destroying our economy, our culture, and our values. New York: Doubleday.

René König, 2013. “Between lay participation and elite knowledge representation,” Information, Communication & Society, volume 16, number 2, pp. 160–177.
doi: http://dx.doi.org/10.1080/1369118X.2012.734319, accessed 19 May 2015.

Minoru Matsutani, 2011. “Plutonium traces point to core leak: Heavy fuel rod damage feared; trenches fill up with toxic water,” Japan Times (30 March), at http://www.japantimes.co.jp/news/2011/03/30/news/plutonium-traces-point-to-core-leak/#.UcMX6nftuKI, accessed 1 June 2014.

Marcus Messner, Marcia W. DiStaso, Yan Jin, Shana Meganck, Scott Sherman, and Sally Norton, 2014. “Influencing public opinion from corn syrup to obesity: A longitudinal analysis of the references for nutritional entries on Wikipedia,” First Monday, volume 19, number 11, at http://firstmonday.org/article/view/4823/4162, accessed 19 May 2015.
doi: http://dx.doi.org/10.5210/fm.v19i11.4823, accessed 19 May 2015.

Karen D. Multon, 2010. “Interrater reliability,” In: Neil J. Salkind (editor). Encyclopedia of research design. Thousand Oaks, Calif.: Sage, pp. 627–629.
doi: http://dx.doi.org/10.4135/9781412961288.n194, accessed 19 May 2015.

David Nakamura, 2011. “Japan utility dumps radioactive water into Pacific to ease storage woes,” Washington Post (5 April), at http://www.washingtonpost.com/world/japan-dumps-contaminated-water-into-pacific-leaks-to-go-on-for-months/2011/04/04/AFf1pLbC_story.html, accessed 1 June 2014.

Erik Qualman, 2011. Socialnomics: How social media transforms the way we live and do business. Hoboken, N.J.: Wiley.

M. Jordan Raddick, Georgia Bracey, Pamela L. Gay, Chris J. Lintott, Phil Murray, Kevin Schawinski, Alexander S. Szalay, and Jan Vandenberg, 2010. “Galaxy Zoo: Exploring the motivations of citizen science volunteers,” Astronomy Education Review, volume 9, number 1, at http://dx.doi.org/10.3847/AER2009036, accessed 1 June 2014.

Cynthia P. Ruppel, Baiyun Gong, and Leslie C. Tworoger, 2013. “Using communication choices as a boundary-management strategy: How choices of communication media affect the work-life balance of teleworkers in a global virtual team,” Journal of Business and Technical Communication, volume 27, number 4, pp. 436–471.
doi: http://dx.doi.org/10.1177/1050651913490941, accessed 19 May 2015.

Julian Ryall, 2011a. “Nuclear meltdown at Fukushima plant,” Telegraph (12 May), at http://www.telegraph.co.uk/news/worldnews/asia/japan/8509502/Nuclear-meltdown-at-Fukushima-plant.html, accessed 28 May 2014.

Julian Ryall, 2011b. “Nuclear fuel has melted through base of Fukushima plant,” Telegraph (9 June), at http://www.telegraph.co.uk/news/8565020/Nuclear-fuel-has-melted-through-base-of-Fukushima-plant.html, accessed 28 May 2014.

Pnina Shachaf, 2008. “Cultural diversity and information and communication technology impacts on global virtual teams: An exploratory study,” Information & Management, volume 45, number 2, pp. 131–142.
doi: http://dx.doi.org/10.1016/j.im.2007.12.003, accessed 1 June 2014.

James Surowiecki, 2005. The wisdom of crowds. New York: Anchor Books.

Jason Swarts, 2009. “The collaborative construction of ‘fact’ on Wikipedia,” SIGDOC ’09: Proceedings of the 27th ACM International Conference on Design of Communication, pp. 281–288.
doi: http://dx.doi.org/10.1145/1621995.1622051, accessed 19 May 2015.

Robert Thomson, Naoya Ito, Hinako Suda, Fangyu Lin, Yafei Liu, Ryo Hayasaka, Ryuzo Isochi, and Zian Wang, 2012. “Trusting tweets: The Fukushima disaster and information source credibility on Twitter,” Proceedings from the 9th International ISCRAM Conference, at http://www.iscramlive.org/ISCRAM2012/proceedings/112.pdf, accessed 1 June 2014.

U.S. Geological Survey, 2011. “Earthquake details,” at http://earthquake.usgs.gov/earthquakes/eqinthenews/2011/usc0001xgp/, accessed 19 May 2015

Fernanda B. Viégas, Martin Wattenberg, Jesse Kriss, and Frank van Ham, 2007. “Talk before you type: Coordination in Wikipedia,” HICSS ’07: Proceedings of the 40th Annual Hawaii International Conference on System Sciences, p. 78, and at http://innovis.cpsc.ucalgary.ca/innovis/uploads/Courses/InformationVisualizationDetails2009/Viegas2007b.pdf, accessed 1 June 2014.
doi: http://dx.doi.org/10.1109/HICSS.2007.511, accessed 19 May 2015.

Andrea Wiggins and Kevin Crowston, 2011. “From conversation to crowdsourcing: A typology of citizen science,” HICSS ’11: Proceedings of the 2011 44th Hawaii International Conference on System Sciences, pp. 1–10, at http://origin-www.computer.org/csdl/proceedings/hicss/2011/4282/00/07-04-02.pdf, accessed 1 June 2014.
doi: http://dx.doi.org/10.1109/HICSS.2011.207, accessed 19 May 2015.

Wikipedia, 2013a. “Timeline of the Fukushima Daichii nuclear disaster” (15 June), at http://en.wikipedia.org/wiki/Timeline_of_the_Fukushima_Daiichi_nuclear_disaster, accessed 1 June 2014.

Wikipedia, 2013b. “2011 Tōhoku earthquake and tsunami” (13 July), at http://en.wikipedia.org/wiki/2011_Tōhoku_earthquake_and_tsunami, accessed 1 June 2014.

 


Editorial history

Received 17 February 2015; revised 13 May 2015; accepted 19 May 2015.


Copyright © 2015, First Monday.
Copyright © 2015, Noriko Hara and Jylisa Doney. All Rights Reserved.

Social construction of knowledge in Wikipedia
by Noriko Hara and Jylisa Doney.
First Monday, Volume 20, Number 6 - 1 June 2015
http://www.firstmonday.dk/ojs/index.php/fm/article/view/5869/4572
doi: http://dx.doi.org/10.5210/fm.v20i6.5869





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.