Effects of Fact-Checking: How Fake News Coverage Can Influence Campaigning

• In Event: Fact-Checking and Misinformation Correction

Abstract

Does publicly covering fake news affect political campaigns? A large body of research has been dedicated to detecting fake news and the creation of fact-checking systems to debunk fake news on politics. Several studies have also been conducted on the willingness of citizens to accept fact-checks of misleading claims from parties and candidates in campaigns. However, far less is known about the effects of fact-checks on the campaigns themselves. As fact-checking becomes more prominent it is worth to examine how parties and candidates respond to coverage on fake news in their campaigns. Do they accept and welcome fact-checks? Does it reduce the amount of fake news in an ongoing campaign when the parties and candidates are aware of being fact-checked? Can fact-checking even have a deterrent effect on campaigns regarding their decisions whether to take liberties with the truth in their communication or not?

To fill this gap, in this paper we look at the example of the 2020 New Zealand election during which the Facebook campaigns of all parties and candidates have been closely monitored and fact-checked by a team of electoral researchers. Their results on the use of fake news have been widely published in the media during the campaign and journalists have also confronted the parties with these results. We argue that New Zealand is a most-likely case to find effects of fake news coverage on party and candidate behavior. New Zealand has a small media market, where fake news coverage can get a lot of media attention. Political elites must fear serious reputational damages when being called out for the spread of fake news compared to larger markets supplying an abundancy of data and analysis. We conduct at mixed-method study by combining qualitative interviews, quantitative text analysis of the parties’ Facebook posts, and pre-election survey data.

First, with our qualitative interviews show that social media campaigners of the parties in fact closely followed fake news coverage in the media during the campaign; in general, considered fact-checking an important tool in keeping election campaigns honest and fair and acknowledged its importance. Campaigners also showed a tendency to defend and justify themselves when personally being called out for particular incidents of misinformation they have been responsible for. Second, drawing on data from a quantitative text analysis of over 3,000 scraped Facebook posts in the last four weeks of the campaign, we show that the amount of fake news also declined over the course of the campaign once fact-checking results had been published and widely discussed in the media. To examine whether fact-checking actually had an effect on the campaigns and can be considered the cause behind this over-time decline in fake news, we utilize a regression discontinuity design (RDD) which allows us to estimate the effect of fact-checking on the spread of fake news and elicit the causal effect of the fact-checking intervention. We find that the treatment of fact-checking indeed had an influence on the campaigns. Third, we use survey data to show that the parties and the candidates are also well-advised to cut back on fake news when fact checkers are active during a campaign, as fake news coverage can affect public opinion. Our results are important, as they show that fact-checking in smaller countries with manageable media markets can have a corrective effect on the campaigns. Importantly, they also raise the ethical question of political scientists intervening in ongoing election campaigns.

Authors

• Mona Krewel, Victoria University of Wellington

• Julius Lagodny