President Trump meets with Russian President Vladimir Putin during a bilateral meeting on the sidelines of the G-20 summit in Osaka, Japan, on June 28. (Susan Walsh/AP)

President Trump’s time in office has been a roller coaster of startling tweets, controversies and scandals. The news media has reported on numerous difficulties, including possible violations of the emoluments clause, self-dealing, sexual assault allegations, improperly awarded security clearances, indictments and resignations. Nevertheless, he has had an unusually stable approval rating.

Through all this, Trump can seem like Teflon: impervious to scandal. Democrats already dislike him, Republican partisans remain loyal, and Americans’ minds are hard to change. This dynamic can trouble those who believe public opinion should respond to new information. If scandalous news no longer affects voters’ opinions of politicians, politicians will be less likely to care if they are involved in a scandal.

And if intense, negative, and substantial news coverage about a politician cannot change opinions, that may limit the news media’s ability to serve as a critical watchdog against government misdeeds.

Presidential scandals at a time when news consumption is changing dramatically

The presidency invests immense power in one celebrity at the center of a national drama, making presidential scandal an attractive subject for coverage — even for news outlets sympathetic to the president.

However, news consumption is changing. Currently, Americans distrust the media along partisan lines. Many selectively read sources with which they already agree. An incredible number of sources offer information (and misinformation) about national politics; half of Americans get their news from Facebook or other social media sites; and local news outlets are disappearing precipitously. A polarized and nationalized politics is the result: Americans are hearing more and more about the president and Congress and liking them less and less.

Given a fractured news media and a polarized public, Trump’s nearly immovable approval may not seem surprising. But we still wanted to see whether scandals can influence voters’ opinions, or if partisanship is too strong for scandals to influence opinions.

How we studied scandal

Researching the effects of scandals is not easy. We define scandals as charges of misbehavior that get sustained attention from the media, making it less effective to randomly offer respondents a single scandal story, compare their responses with a control group, and predict their responses to a real-world scandal. Media consumption, particularly exposure to real-world media content, is nearly impossible to measure well. Further, on any given day of the Trump administration, there is too much notable news to be able to judge with any certainty whether a particular scandal did or did not change public opinion. Our approach allowed us to isolate the effects of scandal coverage over one week.

We used Amazon’s Mechanical Turk to enroll 1,187 Americans in our study, paying them to use our “news portal” (a news aggregator website we designed with frequently updated headlines from Google News) as their primary news source for a week, with bonus payments for high-volume users. We also collected data on users’ partisanship, education, age and other demographic variables. We randomly assigned respondents to see different sets of stories and surveyed them before and afterward about their opinions on politics and the news. While we could not be sure about respondents’ other news use, we measured their use of the portal and have no reason to believe that their use of other news or interpersonal discussion varied systematically across the randomly assigned conditions.

The news was totally dominated by the Trump-Russia scandal that week. Our study went into the field June 12, 2017. President Trump had fired FBI Director James B. Comey on May 9; Robert S. Mueller III was appointed special counsel shortly thereafter; and Comey testified before Congress on June 8.

Taking advantage of that scandal-filled news cycle, we randomly assigned some of our subjects to see almost no Trump-Russia stories that week, while others randomly received an unaltered amount of stories about the scandal. This design let us isolate the effects of Trump-Russia news on opinions over one week.

Yes, scandal coverage did affect our respondents but only Republicans

We found that only Republicans were significantly influenced by the scandal coverage or lack thereof. Those who saw comparatively more Trump-Russia stories rated his job performance 7.6 percent lower than Republicans who did not read those stories, and rated their positive emotions toward him (such as pride, enthusiasm, and hope) 10.9 percent lower than those kept in the dark. Democrats had non-statistically significant reactions. Republicans did not change their attitudes toward the media, and our results did not change based on whether they clicked on the stories.

In other words, simply changing the balance of scandal headlines that they saw was enough to change Republicans’ attitudes toward Trump. Exposure to sustained coverage of a Trump scandal had detectable, negative effects strong enough to overcome Republicans’ partisanship.

This hardly means that Republicans will abandon Trump any time soon. His approval with Republicans is steady and typical of an average GOP president. As with political ads, intense media events like scandals probably have effects that decay quickly, as people revert to their usual partisanship.

What do our results mean for 2020?

Our study was conducted two years ago. Republicans’ attitudes toward Trump may be more settled today, and more difficult to change with new information. New information will command more media and public attention than old information, and many of Trump’s scandals are old news at this point. Our results also point out the importance of selective exposure: If Republican-leaning outlets such as Fox News are covering Trump’s scandals less extensively, their viewers may not receive scandal-infused news that, as we show, could change their opinions.

But it is simply not true that Trump is not hurt by his scandals or that Republicans never change their opinion of him. Trump pays a price when a scandal attracts intense media attention — particularly among those who are supposed to be most loyal to him.

Though resilient, Trump is not Teflon.

Don’t miss anything! Sign up to get TMC’s smart analysis in your inbox, three days a week.

Joshua P. Darr (@joshuadarr) is an assistant professor of political communication in the Manship School of Mass Communication and department of political science at Louisiana State University.

Nathan P. Kalmoe (@NathanKalmoe) is an assistant professor of political communication in the Manship School of Mass Communication and department of political science at Louisiana State University.

Kathleen Searles (@kesearles) is an assistant professor of political communication in the department of political science and Manship School of Mass Communication at Louisiana State University.

Mingxiao Sui (@mingxiaosui) is an assistant professor of media and communication at Ferrum College.

Raymond J. Pingree (@pingree) is an associate professor of political communication in the Manship School of Mass Communication.

Brian K. Watson (@BrianKWatson) received his PhD from the Manship School of Mass Communication in 2019, and is a research analyst at Crosswind Media & P.R.

Kirill Bryanov (@KBryanov) is a PhD candidate in the Manship School of Mass Communication.

Martina Santia (@Martina_Santia) is a PhD candidate in the Manship School of Mass Communication.