Athlete gender visibility, emotional narratives, and audience engagement in CCTV News’s Olympic short-video coverage on Douyin: an empirical analysis

Short-form video platforms have become important channels for major sports-event coverage, but empirical evidence remains limited regarding how athletes are presented, how emotions are organized, and how these features relate to audience engagement in specific institutional accounts. This study analyzes 219 short-video posts published by the official China Central Television (CCTV) News account on Douyin during the 2024 Paris Olympic Games. Content analysis, t-tests, and regression analysis were used to examine mention-based athlete gender visibility, emotional narratives, and three platform-displayed engagement indicators: likes, comments, and shares.

The results show no statistically significant difference between female and male athletes in mention frequency, indicating relative balance only at the level of being named. Athlete mention frequency was not significantly associated with likes, comments, or shares. Emotional narratives were not directly associated with higher engagement, but posts using emotional narratives contained significantly more athlete mentions.

Emotional narrative also positively moderated the relationship between mention frequency and shares, although not the relationships with likes or comments. These findings suggest that, within this single-account and single-platform case, mention frequency is a potential narrative resource rather than an engagement outcome in itself, and its association with content diffusion depends partly on emotional framing. The conclusions should not be generalized to all Chinese mainstream media or interpreted as evidence of complete gender equality in athlete portrayal.

These authors contributed equally: Gengxin Yi, Feng Gao. School of Physical Education, Guangdong Medical University, Dongguan, China The authors declare no competing interests. All data were drawn from publicly accessible content on Douyin.

The study did not recruit, intervene with, or experiment on human participants, and it did not collect personally identifying or private sensitive information. Ethical approval was therefore not required. The analysis was limited to aggregated coding and statistical analysis of publicly visible content and engagement indicators, and no identifiable personal information is reported.

During manuscript preparation, the authors used OpenAI’s ChatGPT (GPT-5.5) for English translation and language editing. The tool was used only to improve clarity and consistency, not for study conception, research design, data collection, content coding, statistical analysis, generation of results, or formation of conclusions. The authors reviewed and take responsibility for the manuscript.