TikTok “edits” can reshape how voters see politicians, study suggests

A new study published in Social Media + Society finds that highly stylized TikTok videos, known as “edits,” can alter how people perceive political figures. Specifically, videos portraying politicians as physically attractive or “badass” increased ratings of their attractiveness and, in some cases, improved their favorability among viewers. These effects were particularly strong for Donald Trump.

The researchers set out to understand the political implications of a fast-growing but underexplored form of online media: TikTok “edits.” These are short videos, created by splicing together clips of well-known public figures, usually set to dramatic music and enhanced with filters and visual effects. Unlike traditional political messaging, which emphasizes policy positions or personal credibility, these videos aim to evoke emotion and aesthetic appeal.

“TikTok users are constantly producing new types of media that social scientists don’t have much context for understanding. I came across TikTok ‘edits’ in a video essay by journalist Jules Terpak, and figured that this was potentially important for how politics operates on video-based platforms,” said study author Kevin Munger, an assistant professor and the Chair of Computational Social Science at the European University Institute in Florence and author of The YouTube Apparatus.

The researchers conducted a large-scale online experiment in June 2024, just before Joe Biden withdrew from the presidential race. They recruited 2,807 participants through Prolific, a platform commonly used for behavioral and social science research. After screening for attentiveness, the final sample included 2,303 U.S. adults whose demographic composition matched the national population in terms of age, gender, and race.

Participants were randomly assigned to either a control group or a treatment group. The control group watched three non-political TikTok edits, featuring popular celebrities like Ana de Armas and Cristiano Ronaldo. The treatment group, by contrast, viewed three political edits: one featuring Robert F. Kennedy Jr., one featuring Donald Trump, and one featuring Joe Biden. These political edits fell into two genres that emerged during the researchers’ review of popular TikTok content: “thirst trap” edits, which presented the politician as physically attractive and charming, and “badass” edits, which portrayed them as powerful and intimidating.

Free daily newsletter

Each participant in the treatment group saw one Trump edit, one Biden edit, and the same RFK edit (a shirtless gym video). By randomizing which edit of Trump and Biden each participant viewed, the researchers could isolate the effects of the edit type (thirst trap or badass) on perceptions of attractiveness and overall favorability.

The videos were selected from TikTok’s most popular “edit” content using search terms like “Biden edit” and “Trump edit.” The team chose the highest-performing examples of each genre based on view counts and subjective quality. Though this approach introduced variation in video length and production quality, it allowed the experiment to reflect the actual content circulating online.

After viewing the videos, participants rated the politicians on two measures: physical attractiveness and general favorability, the latter assessed using a standard 100-point “feeling thermometer” scale. The researchers also collected demographic information, including political affiliation, age, gender, and social media use.

The researchers found that all four political edits caused statistically significant increases in perceived attractiveness. For Biden, the thirst trap edit had a particularly strong effect, boosting attractiveness ratings by about 0.35 points on a five-point scale—roughly 250% more than the increase from the badass edit. For Trump, both edits had similar effects, increasing attractiveness by about 0.18 to 0.20 points.

However, when it came to favorability, the differences were more pronounced. Both of Trump’s edits led to meaningful increases in his overall favorability, with the badass edit producing a larger effect (3.1 points) than the thirst trap (2.2 points). In contrast, neither of Biden’s edits had any significant effect on how participants evaluated him overall. In fact, the point estimate for the Biden badass edit was slightly negative, though not statistically significant.

This difference in impact between Trump and Biden prompted the researchers to consider a possible explanation: video quality. Although both sets of edits were selected from among the most popular available on TikTok, the Trump edits were judged to be more compelling, both in terms of production value and viewer engagement. Since the study was not designed to control for quality, it’s difficult to separate the effect of video content from the way it was produced and presented. The researchers argue that this mirrors the real world, where content quality is uneven and high-quality videos are more likely to go viral.

Interestingly, the effects of the edits were not strongly influenced by the viewers’ political leanings, gender, age, or TikTok use. While partisanship did shape baseline ratings—Republicans rated Trump as more attractive and Democrats rated Biden more favorably—the changes caused by the videos were fairly consistent across groups. This suggests that even viewers who do not align politically with a candidate might be swayed by a compelling video edit.

“Politics on short-form video isn’t really about facticity — it would be meaningless to call these ‘edits’ either true or false,” Munger told PsyPost. “We’re used to thinking about ‘truth’ or ‘misinformation’ on social media, but this isn’t the only factor for understanding what consuming this content does to political opinions. The point is portray a given politician in an aesthetically appealing way — and this makes the physical appearance of the politician more important. ”

The authors caution that their research is exploratory and based on a specific set of videos that may not generalize to all forms of short-form political content. Future studies could improve control by producing edits in-house, varying only the specific elements of interest such as filters, music, or pacing. They also highlight the need to better understand the creators of these videos—many of whom appear to be hobbyists more interested in making eye-catching content than in promoting a political agenda.

“This is a first effort at studying the phenomenon of TikTok ‘edits,'” Munger said. “I used an online convenience sample to recruit my subjects. Also, the set of politicians being studied included only white men; it’s very possible that the results would be different if there had been non-white politicians or women who were running for office when I ran the study.”

“I’d like to better understand the incentives of the people making these edits. They mostly seemed to be hobbyists rather than political activists, just trying to produce the coolest videos they could. What happens when political campaigns begin to understand the power of these edits and try to make them themselves? Is this the beginning of a ‘post-influencer’ era in short-form video, where content is remixed and edited rather than created from scratch?”

The study, “Thirst Traps and Quick Cuts: The Effects of TikTok ‘Edits’ on Evaluations of Politicians,” was authored by Kevin Munger and Valerie Li.