News organizations as fact-checkers: Any potential issue?

Guest Blogger

From the workshop on  “News and Information Disorder in the 2020 US Presidential Election,” 

Jisu Kim and Soojong Kim 

During the presidential election, major news organizations, including USA Today, The New York Times, The Associated Press, and The Washington Post, have verified with fact-checking organizations what politicians said in their speeches and presidential debates. However, unlike fact-checking organizations, which are not strongly connected with a specific group of audiences, several news organizations have been preferred and trusted by different audiences. For example, CNN and Fox News, two major political news sources, are differentially trusted by these audiences. According to Pew, more than 60% of Democrats trust CNN, while more than 60% of Republicans trust Fox News. Only 20% of Republicans got their political and election news from CNN; 60% of Republicans got their political and election news from Fox News in 2019. 

Considering the public’s different preference levels toward individual media organizations based on their political predisposition, how does the public differentially perceive and share news articles that correct misinformation about politicians? If the public perceives a relationship between a news organization and a politician as favorable, are their evaluations regarding the credibility of the news reports that fact-check or correct misinformation about one of the politicians different, depending on the news organization?

Warranting theory can provide an interesting explanation in regard to the above questions. Developed by Joseph B. Walther and Malcolm Parks in 2002, warranting theory posits that information perceived as not being manipulated or controllable by the person to whom it refers is considered more valuable, legitimate, or credible than information that can be modified or have its dissemination controlled by the person described in the information. In addition, some scholars have explained that perceptions of the relationship between the information source and the target to which the information refers can affect evaluations of the information’s credibility. In other words, when the source is perceived as having a favorable relationship with the target to which the information refers, the perceived credibility of the information would be low: the public could assume that the source is motivated to modify the informational content or selectively distribute the information about the target due to its favorable relationship with the target. Thus, this source-target relationship can affect the perceived manipulation likelihood of the information about the target from the source, which can eventually affect perceptions regarding the information’s credibility. 

We examined this theory in the context of the relationship between news organizations and politicians. We investigated how the public evaluates the credibility of fact-checking news articles published by a news organization that is perceived as having a favorable relationship with a political candidate to which the news articles refer. In order to answer this question, we conducted an online experiment. We sought to examine the effect of the perceived favorable source (news organization)-target (politician) relationship on the public’s evaluation with respect to the credibility of news articles that corrected negative misinformation about politicians. 

Specifically, at the beginning of the online experiment, we began by encouraging participants to consider specific source-target relationships. In particular, we showed participants a short passage conveying favorable relationships between President Trump and Fox News and between President-elect Biden and CNN. Afterward, participants were randomly assigned to one of two groups: for one group, participants viewed articles published by sources (i.e., news organizations) that corrected negative information about the target (i.e., politician) they tended to favor. In this condition, the participants’ perception regarding the favorable relationship between the source and the target could be amplified; as a result, the perceived manipulation likelihood would increase. In contrast, in the second group, participants viewed articles published by sources that corrected negative misinformation about the political figure they tended to oppose. Thus, participants in the second group might assume a low likelihood of manipulation by the news sources. The articles’ content was adapted from actual articles published by fact-checking organizations. We examined how the participants’ evaluation of the news articles and political figures might be influenced by their perception of the news organizations’ manipulation likelihood. Participants’ disposition to trust others, knowledge about the misinformation described in the news articles, and political predisposition were controlled for in the analyses. 

For the preliminary finding, we could not find evidence regarding a significant effect of perceived manipulation likelihood of the news organizations (i.e., source) on the news articles’ perceived credibility. Interestingly, the perceived manipulation likelihood of the news organizations influenced participants’ evaluation of the target politicians in this study. Although the effect only existed for Trump, the fact that participants’ evaluation of Biden was not influenced by the news organizations’ perceived manipulation likelihood suggests that certain characteristics of the targets might moderate this effect. In addition, given that we used news articles that corrected negative misinformation about political candidates in the election, the result implies that the perceived favorable relationship between the news organizations and politicians might negatively affect the public’s evaluation of politicians, regardless of the content. 

This approach can be applied to other source-target relationships in online and social media environments. What about the relationship between the platform as a source of information and politicians or other targets to whom social media posts on the platform refer? What happens if the public perceives a politician as having the ability to control the distribution of the social media posts? What happens if the public perceives a politician and a fact-checking organization as being closely connected? These questions need to be addressed in future misinformation research. 

One recent survey shows that the public’s support for social media companies’ fact-checking of politicians in general and President Trump  specifically can differ, depending on their political predisposition. Although it was not our main finding, when we analyzed the data separately among liberals and conservatives, we observed a differential tendency between both groups in terms of evaluating the news articles’ credibility with regard to correcting misinformation about political figures. In a future experiment, we will examine the dynamics among the public’s political predispositions, perceptions regarding the relationships between news organizations and politicians, and politicians to which the fact-checking news articles refer. 

Jisu Kim is a resident fellow at the Information Society Project and an affiliated scholar at the Justice Collaboratory at Yale Law School. Soojong Kim is a postdoctoral fellow at Stanford Center on Philanthropy and Civil Society, jointly affiliated with the program on Democracy and the Internet and the Digital Civil Society Lab.

Cross posted at the Knight Foundation