More than 1 in 3 Americans has cut off a friend, family member, coworker or romantic partner because of political differences — and the trend appears to be accelerating, according to a new study published Monday in PNAS Nexus, a peer-reviewed open-access journal affiliated with the National Academy of Sciences.
Psychologists Mertcan Gungor and Peter H. Ditto of the University of California, Irvine, drew on four datasets totaling 3,791 participants, supplemented by data from the American National Election Studies, to examine what they term “political breakups” — the termination of a personal relationship due to political disagreement.
Their most recent survey, conducted through YouGov in April 2025 with a nationally representative sample of 1,000 U.S. adults, found that 37% of Americans reported losing at least one relationship over politics, with most saying they had lost more than one. Among those who experienced a breakup, 62% said it was with a friend, 40% with a family member, 29% with a coworker and 10% with a romantic partner.
Friendships, the researchers write, occupy a uniquely vulnerable middle ground: close enough that political disagreements tend to surface, but without the structural commitments — shared finances, children, decades-long bonds — that tend to hold romantic and family relationships together.
The partisan divide in who initiates these splits was stark. In a separate survey of roughly 950 adults conducted through Prolific the day before the 2024 presidential election, 66% of Democrats who had experienced a political breakup said they were the ones who ended it, compared with just 27% of Republicans. Democrats were also more likely than Republicans to report having lost a relationship over politics at all — 46% to 29% in the YouGov survey, even after controlling for strength of partisanship and demographic differences.
The researchers also found evidence the phenomenon is intensifying. Among participants who recalled political breakups, 96% placed their most significant split in 2016 or later, with spikes in presidential election years. After the 2016 election, roughly 14% of Americans said they had lost relationships because of that race. A survey conducted just five and a half months after the 2024 election put that figure at 18% — surpassing the 2016 figure in roughly half the time, though the researchers cautioned the comparison is imperfect and election timing may affect how people recall past rifts.
Perhaps the most striking finding involves what political breakups appear to do to perceptions of the opposing side. Using a standard 100-point warmth scale, people who had lost relationships over politics rated opposing voters nearly 8 points colder than fellow partisans who had not experienced a breakup — and those who initiated the split showed even sharper hostility toward rank-and-file voters, rather than toward party leaders.
Those who had experienced a breakup were also significantly more likely to hold distorted views of what their opponents actually believe: Democrats with breakups overestimated the share of Republicans who agreed with white nationalists by about 12.6 percentage points more than Democrats without breakups, while Republicans with breakups overestimated the share of Democrats who thought most white people are racist by about 14.6 percentage points more.
The researchers said the causal direction of that hostility remains unclear, since all datasets were cross-sectional snapshots rather than longitudinal tracking of the same individuals over time. They suspect the relationship runs in both directions: hostility leads to breakups, which generate still more hostility, as people lose one of the few remaining windows into why the other side thinks the way it does and may turn instead to exaggerated media portrayals to justify severing the relationship.
“Given the role of exposure to opposing views in building political tolerance, these ’political breakups’ are a troubling sign for the health of a democracy,” the authors wrote. “And given the importance of relationships for well-being, they have implications for the health of citizens as well.”
The researchers noted several limitations. Most data came from online opt-in panels, which tend to attract more politically engaged participants and may inflate breakup estimates. All data was self-reported, raising concerns about recall accuracy. And social desirability may skew reporting differently by party — their supplemental analysis found some evidence Republicans may underreport political breakups, while Democrats may overreport them.
This article was constructed with the assistance of artificial intelligence and published by a member of The Washington Times’ AI News Desk team. The contents of this report are based solely on The Washington Times’ original reporting, wire services, and/or other sources cited within the report. For more information, please read our AI policy or contact Steve Fink, Director of Artificial Intelligence, at sfink@washingtontimes.com
The Washington Times AI Ethics Newsroom Committee can be reached at aispotlight@washingtontimes.com.
