Specifically, the study found that postings about embarrassment on the social media platform Twitter jumped 45 percent in the first year after President Trump took office.
Yes, this may seem like a “duh” study — at least to anybody who spends any time on Twitter. But the findings are intriguing for what they may reveal about the nation’s psychological mood.
“We found that people felt the need to share their emotions related to Donald Trump’s politics, and embarrassment was the emotion that most clearly described what people felt,” said study co-author Sören Krach, a social neuroscientist at the University of Lübeck in Germany, in a released statement.
How the study was done
For the study, Krach and his University of Lübeck colleagues joined forces with researchers at the University of Michigan. They analyzed tens of millions of tweets sent from U.S. Twitter accounts between June 2015, when President Barack Obama was still in office, and December 2017, almost a year after Trump had entered the White House.
First, they searched the tweets for the word “embarrassment” or its variants. They found that during the final months of Obama’s presidency (until the presidential election in November) the number of tweets referring to embarrassment averaged 909 per million per day. They also found that during the first months of the Trump presidency those numbers increased substantially, to an average of 1,319 per million per day — a 45 percent jump.
That doesn’t necessarily mean that Trump’s presidency was the cause of the increase in embarrassment-related tweets, however. Correlation isn’t the same as causation. So the researchers took a deeper dive into the Twitter data. They examined the most prominent spikes in embarrassment-related tweets to see what other words were linked to them.
The first of these spikes was on Oct. 10, 2016. Although Obama was still in office, that was the day after the second presidential debate between Trump and Hillary Clinton. The second spike occurred on March 18, 2017, the day after Trump appeared to refuse to shake hands with German chancellor Angela Merkel at the White House. The third was on May 26, 2017, the day Trump appeared to shove the prime minister of Montenegro out of his way during a photo shoot at the NATO headquarters in Brussels.
From each of those spikes, the researchers randomly selected 40,000 original tweets (no re-tweets) that contained the embarrassment keywords. After eliminating words without any semantic meaning (such as “the,” “have,” and “over”) from those tweets, they made word-clouds of the remaining words to identify the most frequent ones.
“We found support for the notion that the increased expression of embarrassment was related to Trump or events during his presidency,” the researchers write.
The most frequent words associated with embarrassment on the Oct. 10, 2016, spike (the day after the presidential debate) were “debate,” “Trump” and “@realDonaldTrump.” On March 18, 2017 (the day after the Angela Merkel episode), the most frequent words were “Trump,” “@realDonaldTrump” and “@Potus,” followed closely by “country,” “American,” “America” and “Merkel.” Many of those same words were also most frequent on March 18, 2017 (the day of the pushing incident), as well as “NATO” and “world.”
The researchers then broadened the analysis to include the 10 most prominent spikes during Trump’s presidency.
Finally, the researchers examined all the tweets to see if there were any correlations between Trump and words that express emotions other than embarrassment.
“Beyond embarrassment, there is a slight positive correlations between Trump and the words disgust, shame and anger,” said Dar Meshi, one of the study’s co-authors and an assistant professor of advertising and public relations at Michigan State University, in a released statement. “This means that when people tweet these words, there’s a higher probability that those people are tweeting about Trump.”
A particular type of embarrassment
The study’s limitations are obvious. As already noted, the findings are correlational and thus may not necessarily reflect causation. Also, the study used data from only one social media platform. Furthermore, the researchers were unable to verify that all the tweets came from real people rather than from bots.
Still, the findings are robust enough to raise some interesting questions, particularly about how people respond to vicarious embarrassment (what the Germans call fremdscham, which is apparently the opposite of schadenfreude) when it involves their political representatives.
In their paper, the researchers propose two possible reasons for why mentions of embarrassment have increased under the Trump presidency.
One has to do with the fact that Trump, as president, represents all Americans, including those who don’t agree with his politics. The other is that Trump seems to violate norms and etiquettes on purpose.
The combination of those two factors, say the researchers, cause people to not only feel vicarious embarrassment for Trump and his actions — to cringe as they imagine themselves behaving that way in public — but to feel it as a reflection on themselves rather than on the president alone.
Meshi says that he hopes people who are feeling this secondhand embarrassment will take steps to separate themselves form it — for their own psychological health.
“If you have a negative emotion, understanding where it comes from helps to not let it affect you as much,” he said. “People are having these emotions not only because Trump is their representative, but also because he seems to intentionally violate social norms as the leader of their in-group.”
“Hopefully, by understanding this situation, it can help U.S. citizens avoid experiencing these negative emotions over such a prolonged period,” he added.
FMI: Frontiers in Communication is an open-access publication, so you can read the study in full on the journal’s website.