Debus, Storrer, Stuckenschmidt & Wessler: Wikiganda: How neutral is Wikipedia?

Wikipedia has grown into the World’s largest Encyclopedia. Studies have shown that the quality – in terms of number of errors – of the contents in Wikipedia has successively improved and is meanwhile superior even to established resources like the Encyclopædia Britannica (Giles 2005, Casebourne et al. 2012). From the beginning on, however, the possible impact of hidden agendas and propaganda have been seen as a risk of Wikipedia as a collectively built resource that might compromise the neutral nature of the encyclopedic knowledge (Denning et al. 2005). This risk has been reduced in the meantime, because research shows that the participation in the editing work in Wikipedia has shifted from a situation where the content is mainly provided by a limited number of power users to a wider community of users contributing (Kittur et al. 2007a). At the same time, sophisticated mechanisms have been introduced to prevent misuse of the editing functions and guarantee neutrality as far as possible (Kittur et al. 2007b). Despite these efforts, more recent studies have shown that there are still different kinds of biases in the content of Wikipedia. Reagle and Rhue (2011) have investigated a possible gender bias in Wikipedia and the Encyclopædia Britannica. They compare biographic entries from the two Encyclopedias with external sources with respect to coverage, gender representation, and article length. The authors find that biographies of female persons are more likely to be missing in Wikipedia than in the Encyclopædia Britannica. Greenstein and others (2014) have investigated whether there is a political bias towards Democratic of Republican positions in Encyclopedias using a terminology-based analysis that has been used to detect political bias in common media (Gentzkow and Shapiro 2010). The study shows that indeed the English version of Wikipedia is likely to be more biased towards a Democratic point of view than the Encyclopædia Britannica. The case study in Roessing/Podschuweit 2011 suggests that this bias has impact on electoral campains.

The results of the recent studies show that there is indeed a point in questioning the neutrality of Wikipedia as an information source. This leads us to the question whether and how Wikipedia is used as a medium for political influence taking. We investigate this question by comparing Wikipedia articles describing political actors (parties, politicians, institutions) across different language versions. We are interested in differences in the presentation of the actors in the different language versions and possible differences between these versions. We are particularly interested in different versions that have their origin in the same political space (e.g. the English vs. the Scottish or the Bavarian vs. the Frisian version) to see whether different language communities take a fundamentally different view. We base our analysis on three different aspects of the corresponding Wikipedia articles:

  • We compare the context of the article in terms of links to other Wikipedia articles. We assume that this provides with an idea of the thematic focus of the article
  • We look at the terminology used in different articles. Research in political science has shown that terminology is a suitable indicator for political position taking and thus can help to identify different positions in articles on the same issue.
  • We look at the images used in different articles. Using the Wikimedia library, we can identify image use in different articles to identify similarities in differences in image use across articles.



Jim Giles (2005). Jimmy Wales’ Internet encyclopaedias go head to head. Nature. Volume 438 Number 7070 pp 889-1050. December 2005

Casebourne, C. Davies, M. Fernandes and N. Norman (2012) Assessing the accuracy and quality of Wikipedia entries compared to popular online encyclopedias: A comparative preliminary study across disciplines in English, Spanish and Arabic. Epic, Brighton, UK

Denning, J. Horning, D. Parnas, and Lauren Weinstein (2005). Wikipedia Risks. Communications of the ACM December 2005/Vol. 48, No. 12

Gentzkow. J. M. Shapiro (2010). What drives Media Slant? Evidence from U.S. Daily Newspapers. Econometrica, Vol. 78, No. 1 (January, 2010), 35–71

Greenstein, F. Zhu (2014). Do Experts or Collective Intelligence Write with More Bias? Evidence from Encyclopædia Britannica and Wikipedia. Working Paper, Harvard Business School 15-023 October 10, 2014

Kittur, E. Chi, BA. Pendleton, B. Suh and T. Mytkowicz (2007). Power of the few vs. wisdom of the crowd: Wikipedia and the rise of the bourgeoisie, World wide web 1 (2), 19 2007

Kittur, B. Suh, BA Pendleton and EH Chi (2007), He says, she says: conflict and coordination in Wikipedia, Proceedings of the SIGCHI 2007

Reagle, L. Rhue. (2011). Gender Bias in Wikipedia and Britannica. International Journal of Communication, 5, 2011

Roessing, N. Podschuweit (2011). Wikipedia im Wahlkampf: Politiker, Journalisten und engagierte Wikipedianer. In: Eva Schweitzer & Steffen Albrecht (Hrsg.): Das Internet im Wahlkampf: Analysen zur Bundestagswahl 2009. Wiesbaden, 297-314.

  • © 2016/2017 Bremen University || Faculty of Linguistics and Literary Science