Donald Trump speaks during a campaign event in Fairfield, Conn. on Aug. 13. (Michelle McLoughlin/Reuters)

Republican presidential candidate Donald Trump continues to use conspiracy theories to take aim at the news media. In a series of tweets, Trump claimed that the media was covering him dishonestly, that it was publishing lies, and that the media was covering for Hillary Clinton. Such claims are not new; Trump has accused the media of intentional bias since the beginning of his campaign.

Functioning democracies rely on independent news sources to keep voters informed. When news sources lose trust, voters may have no way of getting the information they need to make informed choices. This prompts the question: Do Trump’s constant attacks on the media have an effect on how citizens view the news media?

My co-authors and I attempted to answer this question in a recent study focusing on a conspiracy involving media bias. We wanted to better understand why, when presented with an informational cue that a conspiracy might be taking place, people would choose to adopt or reject the conspiracy theory.

We fielded a national survey in the weeks before and weeks following the 2012 presidential election. In the first wave of the survey, we measured people’s party affiliation as well as how much conspiratorial thinking affected their worldview. Specifically, we asked people how much they agreed or disagreed with statements such as “Big events like wars, the current recession, and the outcomes of elections are controlled by small groups of people who are working in secret against the rest of us.” Overall, conspiratorial thinking is fairly common among members of both parties.

In the survey’s second wave, fielded after the election, we embedded an experiment to examine how conspiratorial thinking and partisanship affected whether respondents would accept a conspiracy theory about media bias. We asked people whether the media was biased in favor of one of the presidential candidates, but we randomly varied whether that bias was attributed to “a conspiracy” or to “poor journalism.”

We found several things. First, 50 percent believed that the media was biased. When asked a follow-up question about what caused the media bias, about 30 percent of our respondents selected “a conspiracy to deceive the public.”

Second, Republicans were the most likely to believe in the media conspiracy, followed by independents and Democrats. In comparison to strong Democrats, strong Republicans were 44 percentage points more likely to believe that the media are biased because of a conspiracy. This is likely because Republican leaders have said for decades that the media is biased. Think, for example, about attacks on the media by Spiro AgnewRichard Nixon or the most recent Bush administration. It is also possible that Republican respondents believed in the media conspiracy as a way of scapegoating or salving their wounds following Mitt Romney’s defeat.

Third, respondents with high levels of conspiratorial thinking were most likely to believe in the media conspiracy. For example, individuals with strong evidence of conspiracy thinking were 28 percentage points more likely to believe in the media conspiracy than those with little evidence of conspiracy thinking.

Fourth, conspiratorial thinking had the strongest impact on Democrats and the weakest on Republicans. Once we accounted for partisanship, there wasn’t much room for conspiratorial thinking to influence Republicans. Democrats, on the other hand, are more trusting of the media, and therefore only Democrats with strong conspiratorial worldviews were likely to see a media conspiracy. In short, Republicans believed in media conspiracy theories either because of elite cues or because Romney lost, and not so much because they had a conspiratorial worldview.

Finally, the experimental treatment — telling our respondents that there was a media conspiracy afoot — did not make them more likely to believe there was a media conspiracy. Interestingly, rather than being easily convinced of a conspiracy, most respondents were not affected by the conspiracy treatment at all. Part of this is because many people were already convinced that the media was involved in a conspiracy, and were therefore not affected by new information. In fact, the only respondents who were affected by the conspiracy cue were independents who were conspiracy-minded.

To be sure, our experiment was more subtle than the attacks that Trump has levied at the media. Nevertheless, it shows the challenges of trying to convince people of a conspiracy. However, the sheer number of people — particularly Republicans — who believed in a media conspiracy after the 2012 election suggests that we will see something similar after 2016. Certainly Trump’s remarks are unlikely to assuage conspiratorial thinking.

Joseph Uscinski is associate professor of political science at the University of Miami College of Arts and Sciences. He is the co-author of “American Conspiracy Theories” (Oxford University Press, 2014).