A new study published in the Journal of Political Marketing finds that political candidates on TikTok attract more views and engagement when they use short, emotionally charged, performance-focused videos rather than detailed policy discussions. The research argues the app’s design encourages quick entertainment and visual spectacle over traditional campaigning.
Researchers sought to understand how candidates adapt their communication for a platform with a younger, fast-moving audience and to explain why some formats perform better than others in this digital environment.
The work relies on Media Ecology Theory, which posits that communication platforms are not neutral conduits for information but shape the messages that emerge. It also engages the idea of “affective publics,” describing digital communities that cohere around shared emotions rather than logical debate. The study’s authors, Jenny Bronstein and Tal Eitan, are researchers at the Department of Information Science at Bar-Ilan University in Israel.
The investigators analysed every TikTok video posted by the official accounts of Donald Trump and Kamala Harris during the 2024 U.S. presidential campaign, from the summer 2024 account launches until the day after the November election. The final dataset included 276 videos: 57 from Trump and 219 from Harris.
Videos were watched and coded for structural style and content, with attention to features such as direct-to-camera delivery, casual moments, and footage of crowds. Emotional themes were tracked, noting humour, critiques of opponents, or appearances by celebrities, and the results were then compared with engagement metrics such as views, likes, comments and shares.
Trump posted less frequently but kept his clips short, averaging about 18 seconds. Those shorter videos tended to show higher engagement on average per post.
Harris uploaded much more often, and her videos were longer, averaging about 37 seconds. Her longer, issue-focused material tended to produce steadier but lower average interaction levels.
Both campaigns leaned heavily on visual spectacle to convey momentum, with over 80 per cent of videos from each candidate featuring campaign stops, motorcades or rallies. The energetic crowd scenes and shots of candidates stepping off aircraft acted as symbolic displays of reach and power, and for Trump these spectacle-led clips correlated with higher share counts.
When speaking directly to the camera, Trump favoured an assertive, broadcast-style delivery. He focused on economic themes, border control and confidence in victory, and direct-to-camera appeals made up about two-thirds of his posts, a pattern linked to higher plays and likes for his account.
Harris used direct-to-camera appeals less often—roughly a third of her posts—and when she did speak directly, she adopted a more instructional tone. Her topics included human rights, disaster recovery and reproductive rights, and her direct appeals were generally longer and more explanatory than Trump’s.
Around half of both campaigns’ videos featured informal, relatable moments. Trump cultivated a populist image by visiting a McDonald’s and riding in a garbage truck, while Harris shared personal stories about cooking or caring for family members. The researchers found no statistically significant differences in engagement for these casual clips, suggesting such moments alone did not guarantee greater visibility on the platform.
Humour and entertainment proved highly effective for capturing attention, though the use of jokes, memes or parody accounted for fewer than 10 per cent of videos from either candidate. When Trump posted comedic content, it was associated with significantly higher engagement across all metrics.
The study also looked at how candidates attacked one another on TikTok. Harris referenced Trump in about a fifth of her videos and frequently used split-screen formats to juxtapose a clip of her opponent with a rebuttal. In one example, she said, “As president, I will take on corporate price gouging to lower the cost of groceries. Donald Trump can’t figure out what groceries are.”
Trump referenced Harris in about a third of his videos but largely avoided direct engagement with her specific claims, instead relying on sarcasm, visual exaggeration and mockery. In one video blaming his opponent for job losses, he stated, “In the last month, Kamala Harris has destroyed 24,000 jobs, manufacturing jobs, beautiful jobs.”
Associating with celebrities and online influencers was another key strategy. Trump leaned on male figures from sports and podcasting to reinforce a populist, masculine image, while Harris partnered with diverse musicians, actors and media personalities to stress inclusivity and broader cultural relevance. For Trump, clips featuring these figures correlated with higher share counts, functioning as an efficient shorthand for signalling identity.
The researchers acknowledge several limitations. The study focused exclusively on TikTok, so patterns may differ on platforms with different designs, such as Instagram or X. It analysed only two candidates in a single election cycle, and relied on public engagement metrics that reflect reach rather than how audiences felt about the content or whether it influenced voting behaviour. The coding system captured the presence of features rather than the intensity of the emotional appeal.
The study, titled “Campaigning on TikTok: Platform Affordances, Affective Messaging, and Candidate Communication Styles in the 2024 U.S. Presidential Election,” was authored by Jenny Bronstein and Tal Eitan.
