Expert Commentary

Thanksgiving dinner and talking politics: Research suggests they can (and maybe should) mix

Politics and Thanksgiving dinner might sound like a nightmare recipe. But research suggests talking political turkey just might be a good way to ease partisanship in the U.S.

Thanksgiving
(Unsplash / Meelika Marzzarella)

Between shopping, cooking and cleaning up, playing host on Thanksgiving can be more of a stressor than respite for many people in America. When a seldom seen aunt, uncle or cousin adds a dash of Democrat-versus-Republican partisan politics to the big meal, the thought of getting through Thanksgiving might seem downright unbearable.

“Surveys show that most people look at these sort of Thanksgiving political discussions with dread,” says Jennifer Wolak, a political science professor at Michigan State University who studies the psychology of compromise. “Most people say they hope to avoid having political conversations because they’re conflictual, they’re argumentative. And people also carry these really strong stereotypes about the other party.”

But recent research from Wolak and others suggests those dreaded conversations with relatives who seem to come from a distant political planet could be key to unlocking compromise — and civility — in national political discourse.

Personal compromises, political compromises

People who have personal interactions that do not devolve into conflict, in spite of closely held disparate political beliefs, believe politicians should be able to do the same, Wolak says.

Those who regularly interact with people who preferred a 2020 major party presidential candidate opposite their own are more likely to expect elected leaders to seek compromise, Wolak finds in a May 2022 paper in The Forum: A Journal of Applied Research in Contemporary Politics.

Wolak further writes that “support for compromise climbs with the frequency of people’s discussions with those who hold opposing views.”

Those findings are based on an analysis of survey data from 4,067 adults in the U.S., conducted the summer before the 2020 presidential election by the Survey Center on American Life, a nonprofit, nonpartisan project of the American Enterprise Institute, a center-right think tank.

Wolak says social norms can be a powerful motivator for civility. She suggests, for example, that because it is inappropriate to have a screaming match at Thanksgiving dinner, people’s stereotypes become disrupted when they realize it is possible to have a polite conversation with someone who does not align with them politically.

Despite news outlets devoting thousands of words and hours of video to political polarization, “the average American is still really keen on compromise,” Wolak adds.

Research also offers insights on the value of understanding what the other side wants — validation, oftentimes, not confrontation.

Discussing the results of her 2021 paper on heightened emotions and political conversations — co-written with Anand Sokhey and published in American Politics Research — Wolak says “people who are angry about politics want to talk, but they are actually not so confrontational. They actually want to talk to people who will be a sympathetic outlet more than anything else.”

That does not mean Republicans and Democrats seek one another for validation. In fact, the opposite may be true. Wolak and Sokhey write that strong feelings of anger or enthusiasm “are associated with greater discussion with those who will confirm rather than challenge their beliefs.”

Which circles back to Thanksgiving — the kickoff of the winter holidays, when family members of divergent political stripes often have no choice but to exhibit uncommon displays of cordiality.

Despite potentially dreaded Thanksgiving political conversations, research has found most people don’t find those conversations so bad once they happen. In a July 2019 paper published in the journal Cognition, researchers Charles Dorison, Julia Minson and Todd Rogers analyze results from their surveys on political discussions among 2,455 participants in 2017 and find partisans generally “overestimate the negative affect that results from exposure to opposing views.”

Divergent politics, shorter Thanksgivings

The current era of research on partisanship during Thanksgiving begins with a June 2018 paper in Science by Keith Chen and Ryne Rohla.

Using anonymous cell phone data from more than 10 million Americans, Chen and Rohla track people who traveled for Thanksgiving 2016 from a voting precinct that went for one major party presidential candidate to a precinct that went for the other.

They established individuals’ homes based on location data from 1 a.m. and 4 a.m. during the three weeks before Thanksgiving — most people would likely be at home during those hours. To determine how long families spent enjoying (or avoiding) each other, the researchers looked at travel patterns from 1 p.m. to 5 p.m. on Thanksgiving. They find Thanksgiving was typically 30 to 50 minutes shorter at politically divergent dinners.

“Nationwide, 34 million hours of cross-partisan Thanksgiving dinner discourse were lost in 2016 owing to partisan effects,” Chen and Rohla write.

A subsequent study by Jeremy Frimer and Linda Skitka, published in October 2020 in PLoS ONE, uses surveys to explore the potential link between politics and shorter Thanksgivings — and is less conclusive.

The first survey, which Frimer and Skitka conducted in 2018, was among 579 participants recruited from Amazon’s Mechanical Turk, a crowdsourcing platform many researchers use to find people for studies.

They asked participants how long they spent at Thanksgiving dinner and which topics people talked about, including politics, sports and work. They also asked participants whether they approved or disapproved of then-President Donald Trump, and whether the people they spent Thanksgiving with supported the president. Based on the results of the first survey, Frimer and Skitka did not find that politically diverse Thanksgiving dinners were shorter than politically similar Thanksgivings.

Results were largely the same for a 2019 survey they conducted among 1,146 participants recruited from Mechanical Turk. Frimer and Skitka then combined the survey results, with the larger dataset allowing them to “maximize the precision” of their findings, they write. In this “mega-analysis,” Frimer and Skitka find politically diverse Thanksgiving dinners were shorter by 24 minutes, on average, though they did not find a link between talking politics and participants reporting having had an unpleasant Thanksgiving.

Frimer and Skitka write they are not seeking “the final word” but rather to further explore the question Chen and Rohla raised. They conclude, “Americans appear to be largely successful at putting aside their political differences and enjoying Thanksgiving dinner with relatives and friends with whom they differ.”

In another paper, published in November 2020 in the journal Social Forces, Byungkyu Lee uses Chen and Rohla’s cell phone data and finds families with strong but dissimilar political views had shorter Thanksgiving meals, especially in states where 2016 presidential election results were close.

“These results suggest that the disruptive role of partisanship in shaping social interaction is more salient where politics are anticipated to appear as a ‘hot’ topic at the family dinner table due to the contestation of current election outcomes,” Lee writes.

“You wouldn’t try to convince a die-hard Alabama football fan not to be an Alabama football fan. You know that’s a losing battle. You might be able to convince someone that Alabama’s quarterback didn’t have a great game or that Auburn played well last week. The goal isn’t to convert the Alabama fan. The goal is to have a conversation about football that doesn’t rely on blind loyalty, ad hominem attacks, or false accusations.

What does that conversation look like?”

Kayce Mobley and Sarah Fisher. “Political Scientists in Polite Company: Talking Politics with Family Members,” Journal of Political Science Education.

Cooler heads prevail

Kirsten Cornelson, an assistant professor of economics at the University of Notre Dame, writes in an August 2022 preprint paper that “Thanksgiving causes us to have more political discussions with parents, siblings, and extended family. Outside of Thanksgiving, we are much more likely to have these discussions with spouses and friends.” Preprints are studies published on a public server, such as SSRN, which have not yet been peer reviewed or published in an academic journal.

Across weekly surveys Cornelson conducted in fall 2019 and fall 2021 among 1,368 U.S. and Canadian participants recruited from Mechanical Turk, she finds people who break bread with partisan family members during Thanksgiving tend to more closely align their political views with those partisans after Thanksgiving. That shift lasted about two weeks, with people then reverting to their original leanings.

Cornelson is now running another survey to bolster her data gathered so far.

“It seems like it’s actually very hard to change your opinions,” she says. “But I wanted to see if having that exposure in a more naturalistic setting, in the context of a positive social interaction, if there was actually room for us to change our mind — if we talk to people that we disagree with — and it seems like there is.”

Cornelson also explores a 2000s electoral reform in Canada, which further shows how Thanksgiving gatherings influence politics. Under a 2007 law, federal and provincial governments were ordered to hold elections on a fixed date starting in 2009. Most provinces chose election dates within the weeks before or after Canadian Thanksgiving, which usually falls in early October.

In elections held just after Thanksgiving, Cornelson finds the winning party’s vote share increased by about 10 percentage points, compared with before the electoral reform.

“These results suggest that there is more agreement among the electorate immediately after Thanksgiving, which is consistent with social influences on voting,” she writes.

The surveys don’t capture whether families are having political discussions or whether temporary changes of heart come from socializing in polite company with family members who hold different political beliefs.

Other recent research supports the finding that benefits of such conversations are short lived — indicating, if the goal is to reduce partisanship, that regular interactions among people with dissimilar politics can help. For people who openly talk politics at Thanksgiving but believe those conversations lead nowhere, Cornelson says her research suggests people with opposing views are listening.

“When you’re having these conversations, they can feel really frustrating, but actually you’re being influenced,” she says. “It means you’re influencing your family as well. There is actually some element of productive discussion going on, even when it may not feel like it.”

Further reading

The Promise and Pitfalls of Cross-Partisan Conversations for Reducing Affective Polarization: Evidence from Randomized Experiments
Erik Santoro and David Broockman. Science Advances, June 2022.

The Social Foundations of Public Support for Political Compromise
Jennifer Wolak. The Forum: A Journal of Applied Research in Contemporary Politics, May 2022.

You Can’t Handle the Truth! Conflict Counterparts Over-Estimate Each Other’s Feelings of Self-Threat
Charles Dorison and Julia Minson. Organizational Behavior and Human Decision Processes, May 2022.

We Need to Talk: How Cross-Party Dialogue Reduces Affective PolarizationMatthew Levendusky and Dominik Stecula. Elements in Experimental Political Science, November 2021.

Enraged and Engaged? Emotions as Motives for Discussing PoliticsJennifer Wolak and Anand Edward Sohkey. American Politics Research, September 2021.

Close Relationships in Close Elections
Byungkyu Lee. Social Forces, November 2020.

Are Politically Diverse Thanksgiving Dinners Shorter than Politically Uniform Ones?
Jeremy Frimer and Linda Skitka. PLoS ONE, October 2020.

Selective Exposure Partly Relies on Faulty Affective Forecasts
Charles Dorison, Julia Minson and Todd Rogers. Cognition, July 2019.

Political Scientists in Polite Company: Talking Politics with Family Members
Kayce Mobley and Sarah Fisher. Journal of Political Science Education, November 2018.

The Effect of Partisanship and Political Advertising on Close Family TiesM. Keith Chen and Ryne Rohla. Science, June 2018.

The Rise of Negative Partisanship and the Nationalization of U.S. Elections in the 21st Century
Alan Abramowitz and Steven Webster. Electoral Studies, March 2016.

About The Author