Lably
Skip to content

Ideological bias on Wikipedia

  • 15 min read
  • 4 chapters

Related on Lably

Through Bias, Wikipedia and Policy

Lab · Class 11–12Behavioural Finance — Common BiasesLab · Class 7–11Athenian DemocracyLab · Class 12Inflation & Economic PolicyLab · Class 8Mughal Empire

The English Wikipedia has an internal policy which states that articles must be written from a neutral point of view, which has the goal of representing fairly, proportionately, and without bias, the significant points of view that have been verifiably published by reliable sources on a topic. Collectively, findings show that Wikipedia articles edited by large numbers of editors with opposing ideological views are at least as neutral as other similar sources, but articles with smaller edit volumes by fewer—or more ideologically homogeneous—contributors are more likely to reflect the editorial bias of those contributing.

State of research

A 2012 study by Shane Greenstein and Feng Zhu of the Harvard Business School examined a sample of 28,382 articles related to U.S. politics as of January 2011, measuring their degree of bias on a "slant index" based on a method developed by Matthew Gentzkow and Jesse M. Shapiro in 2010, to measure bias in newspaper media. This slant index purports to measure an ideological lean toward either the Democratic or Republican parties, based on key phrases within the text such as "war in Iraq", "civil rights", "trade deficit", "economic growth", "illegal immigration" and "border security". Each phrase is assigned a slant index based on how often it is used by Democratic or Republican members of U.S. Congress. This lean rating is assigned to a Wikipedia contribution that includes the same key phrase. The authors concluded that older Wikipedia articles were mostly biased to the left, although recent articles are more neutral. They suggest that articles did not change their bias significantly due to revision, but rather that over time newer articles with contrasting viewpoints played a role in rebalancing the average perspectives among the entries.

In a subsequent 2014 study, the same researchers compared about 4,000 Wikipedia articles related to U.S. politics with the corresponding articles in Encyclopædia Britannica using similar methods as their 2010 study to measure "slant" (Democratic vs. Republican) and to quantify the degree of bias. The studies found that Wikipedia articles with fewer revisions were more slanted towards Democratic views than are Britannica articles and had more overall levels of bias, particularly those focusing on civil rights, corporations, and government. Entries about immigration trended toward the Republican party.

A 2014 study conducted on ten different versions of Wikipedia revealed that disputes among editors on the subject of politics (including politicians, political parties, political movements, and ideologies) accounted for approximately 25% of disputes.

A 2015 study of English Wikipedia examined the removal of positive or negative information in biographies of U.S. senators, though not in terms of ideological bias. The researchers introduced positive and negative content, sourced from reliable references, into the biographical entries of U.S. senators. Their findings revealed that negative content was more likely to be removed and were removed at a faster rate compared to positive content. The researchers concluded that a significant editorial bias exists in Wikipedia entries in favor of current U.S. senators of either major party. However, when a similar test was conducted on the Wikipedia pages of recently retired and deceased senators, the same discrepancy in the removal of positive and negative content was not observed. This suggests that the bias identified is specific to the pages of active politicians and does not indicate a systemic issue within Wikipedia. The authors concluded that information generated through collaborative projects such as Wikipedia may be susceptible to an editorial bias that favors politically active individuals.

In a 2017 report from Harvard's Berkman Klein Center for Internet & Society, Wikipedia was identified as a center-right website in the study's "candidate valence" scale, which used the political orientation of X users who shared content from the website as a measure of the website's political lean throughout the 2016 U.S. elections. Wikipedia was one of only three websites identified as center-right by the report, the other two websites being RealClearPolitics and National Review.

A 2023 study compared articles on controversial topics across multiple community-managed wikis to test whether the policy orientation of a collaborative wiki project would produce a slant in the content, by selecting the crowd of contributors. The findings showed that the content of wikis with explicit ideological biases, such as RationalWiki and Conservapedia, is more unbalanced than that of wikis (such as Wikipedia) or encyclopedias (such as Encyclopædia Britannica) advocating neutrality. Wikipedia's content had no significantly bigger slant than that of Britannica, while both RationalWiki and Conservapedia were "more loaded with moral content".

A 2023 study from the American Political Science Review found that changes in interpretations of wiki policy and the departure of certain editors over time led Wikipedia to take firmer stances on contested issues, resulting in an increase in the site's perceived reliability and "producing a type of content that is distinctly anti-pseudoscience and anti-conspiracy theories, and which has the perception of a liberal bent in U.S. politics".

A 2024 paper published by the Manhattan Institute, a conservative think tank, concluded that Wikipedia articles tended to associate greater negative sentiment with public figures and organizations on the right than those on the left.

User collaboration

A 2012 study focusing on edit wars suggested that consensus can often be reached within a reasonable timeframe, even in controversial articles. The conflicts that tend to prolong these edit wars are driven by the influx of new users. Most edit wars were carried out by a small number of users frequently in conflicts, despite their low overall productivity. In these debates, resolution was often reached not based on the merits of the arguments but rather due to external intervention, exhaustion, or the numerical dominance of one group.

A study conducted in 2013 focused on users who had declared their support for either the U.S. Democratic or Republican parties. These users tended to contribute more frequently with voices aligning with their own political orientation. However, they were not inclined to avoid collaboration with political opponents or prefer working exclusively with allies. The authors proposed that the shared identity of being a Wikipedian might outweigh potentially divisive aspects of personal identity, such as political affiliation. This finding distinguishes Wikipedia from other social platforms, such as Twitter and blogs, where users often exhibit strong polarization by predominantly interacting with users who share similar political orientations. In contrast, Wikipedia can be characterized as a platform where users display a higher degree of interaction across political orientations, akin to forums and similar platforms.

A 2018 study found that a majority of editors on the French Wikipedia had a propensity to share equally in a dictator game. This was correlated with their involvement on Wikipedia (as measured by the time spent and attachment).

Drawing from experimental research findings, Holtz et al. (2018) proposed a model of knowledge production in Wikipedia, employing the concept of "productive friction". This posits that a certain level of interpretative conflict is necessary for the collective process to generate knowledge. The model draws an analogy to the socio-cognitive conflict model used in psychology to elucidate individual learning. According to this hypothesis, if the tensions or friction within a group are too low, the potential for knowledge construction is limited since the existing knowledge is deemed sufficient. Conversely, if the friction within a community of contributors becomes excessively high, it can lead to the dismissal of respective ideas or even the division of the group, similar to how an individual may struggle to adapt and learn when confronted with an overwhelming amount of novelty.

A 2019 study conducted among American users of the English version highlighted significant political orientation bias among users contributing to political topics, finding that the more edits made to an entry, the more balanced the political orientation of the contributing pool became. The study indicated that the quality of articles, as recognized by the Wikipedia community, improves as the diversity of political orientation among contributors increases. Politically polarized group members generally produced better articles compared to groups of highly aligned users or moderates. Positive effects of polarization were observed not only in articles related to politics but also in those concerning social issues and even science. Politically polarized groups engage in frequent disagreements, stimulating focused debates that result in higher quality, more robust, and comprehensive edits. However, these findings are subject to limitations. The contributors who participated may suffer a self-selection bias, which can influence outcomes.

Sign in to keep reading

3 more chapters are free with a Lably account.

More in Computing and communication

ISBNUnique numeric book identifier since 1970X (social network)American social networking serviceWayback MachineDigital archive by the Internet ArchivePython (programming language)General-purpose programming languageMacOSOperating system for Apple computersUnicodeCharacter encoding standard