Research map / Video / Video-language understanding

Audience Engagement with Political Messaging on YouTube Shorts (Student Abstract)

AAAI 2025 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 0 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Compares audience engagement, comment toxicity, and polarization on political YouTube Shorts versus long-form videos from 100 US political channels (38,838 videos).

Problem

How audiences engage with political messaging on short-form YouTube Shorts compared with long-form videos, in terms of engagement, toxicity, and polarization.

Approach

Uses YouTube's Shorts launch as a natural experiment: pairs each Short with the temporally closest long-form video from the same channel, then contrasts comment-level engagement, Perspective API toxicity, and a machine-translation-based polarization score.

Key results

• Shorts receive considerably more likes than Longs for both liberal and conservative channels, with no significant difference in views, and shorter comments with more emoji usage on Shorts.
• On a 5K comment sample per group, Perspective API (threshold 0.7) shows Shorts discussions are more toxic than Longs, with conservative Shorts the most toxic.
• Polarization score (higher means less polarized) is 71.4% for Shorts versus 80.1% for Longs, indicating Shorts are more polarized.

Measured effects reported in the paper

Surprising findings

• Shorts show no significant difference in views despite receiving more likes.

Limitations

• Relies on a single toxicity classifier (Perspective API) at a fixed threshold on a 5K sample per group.
• Short-to-long mapping by upload time is approximate; a few Shorts map to the same Long video.
• Only 100 channels with documented leanings from one channel list are studied.

Experimental setup

Datasets: YouTube Shorts and long-form videos from 100 US political channels (Wu and Resnick 2021 channel list) YouTube user comments collected via YouTube API

Metrics: likes views comment length in tokens emoji usage fraction of toxic comments polarization score

Builds on: KhudaBukhsh et al. 2021 machine-translation polarization framework

Closest papers by meaning

• Infer Induced Sentiment of Comment Response to Video: A New Task, Dataset and Baseline (NeurIPS 2024 · similarity 0.80)
• Exploiting Fine-Grained Skip Behaviors for Micro-Video Recommendation (AAAI 2025 · similarity 0.79)
• When Neutral Summaries Are Not That Neutral: Quantifying Political Neutrality in LLM-Generated News Summaries (Student Abstract) (AAAI 2025 · similarity 0.78)
• Measuring And Improving Persuasiveness Of Large Language Models (ICLR 2025 · similarity 0.78)
• Quantifying Political Polarization through the Lens of Machine Translation and Vicarious Offense (AAAI 2024 · similarity 0.78)
• SMTPD: A New Benchmark for Temporal Prediction of Social Media Popularity (CVPR 2025 · similarity 0.77)