The Impact of Online Videos on Chinese Youth's Perceptions of Japan through the Lens of Agenda-Setting Theory

Abstract

This study investigates how online videos shape Chinese youth’s perceptions of Japan, aiming to explore the agenda-setting process of media producers and its impact on audience cognition. Drawing on agenda-setting theory, the research identifies a dual-pathway framework: under specific situational contexts, video producers—shaped by nationality and profit motives—construct media agendas that are disseminated through video content and form. Audience perceptions are then shaped through individual-level frameworks, including gender, education, cultural familiarity, and media engagement habits such as comment interaction and danmaku participation.

Quantitative analysis reveals that politically neutral, logically structured, and longer videos tend to foster more positive national images. Interactive features of online video platform such as comment sections and danmaku also mediate audience trust and critical engagement with content. Policy implications include the need for governments to actively and effectively guide user-generated content in the digital public sphere, promote non-political representations of national identity, and harness participatory media ecologies—such as those of Bilibili—to strengthen agenda-setting capacity.

Hypothesis

Demographic factors such as gender, age, geographic location, and educational institution influence Chinese youths' favorability toward Japan.

Historical events and the trajectory of China–Japan relations significantly shape Chinese youths' perceptions of Japan.

The extent of exposure to Japanese culture, the voluntariness of such engagement, and the favorability toward Japanese culture positively correlate with more favorable perceptions of Japan.

Online videos function as agenda-setting agents, influencing Chinese youths' perceptions of Japan.

Specific characteristics of online videos impact these perceptions:

Videos with lower political content are more likely to foster positive impressions.
Content produced by organized entities exerts a more substantial influence than that by individual creators.
Videos with overt commercial motives may diminish their persuasive effect due to increased audience skepticism.

The degree of trust in different media channels moderates the influence of online videos on perceptions of Japan; higher trust in video platforms enhances this effect.

Usage patterns of video-sharing platforms, including interaction with features like danmaku (bullet comments) and comment sections, contribute to variations in perceptions of Japan among Chinese youths.

Research Design

Descriptive Analysis

Demographic Variables

The survey primarily focused on gender, age, and education level as key demographic variables. Among the respondents, 38.9% were female and 61.1% were male, indicating a slightly unbalanced gender distribution compared to the approximately equal gender ratio in higher education reported by the Chinese Ministry of Education. This imbalance may be due to the fact that some of the universities where the survey was distributed are science and engineering institutions with predominantly male student populations. In addition, most respondents were aged between 18 and 25, and 68% held a bachelor's degree. Therefore, the sample largely reflects the attitudes of individuals within the 18–25 age group who have attained a bachelor's degree or higher.

Cultural Factors

The survey explored respondents’ attitudes toward Japanese culture by asking how much they like it, how frequently they engage with it, and how well they understand it. It also inquired whether their engagement with Japanese culture is positive or negative.

When asked whether they use Bilibili—an online video platform popular among young audiences—most respondents (74.6%) answered “yes.”

Online Videos Featuring Japanese Content

The survey investigated respondents’ perceptions of the profit motivation, content orientation, emotional tone, and duration of online videos featuring Japanese content. Nearly half of the respondents did not perceive a clear profit-driven motive in these videos, while about 28% reported a distinct commercial orientation. When asked about the thematic orientation and emotional tone of the videos, over half of the respondents described them as neutral (60% for content orientation and 50% for emotional tone). These findings suggest that Bilibili—an online video platform popular among young people—shows a relatively mild degree of commercialization. The videos appear to prioritize content quality over monetization, often taking an objective and balanced perspective when presenting Japan.

In addition, respondents indicated that most Japan-related videos were between 10 and 60 minutes in length. This suggests that, in the year the survey was conducted (2021), Bilibili was less influenced by the trend toward short-form video content.

Interaction Habits of Video Viewers

Although the majority of respondents (77%) do not prefer enabling Danmaku—real-time user comments that scroll across the video screen and create an interactive viewing experience—when watching videos on Bilibili, nearly 57% reported frequently using Danmaku to express their opinions. Additionally, 68% of respondents indicated they often use the comment section to engage with video content, while only 33% reported sharing or reposting videos with others.

Data Analysis

Significance Testing

Chinese youth who actively seek out Japanese cultural content report a higher level of familiarity with Japanese culture.

These individuals also tend to have a more favorable overall impression of the Japanese people.

Regression Analysis

Cultural familiarity has a significant positive effect on the dependent variable: overall impression of the Japanese people.

Cultural affinity also has a significant positive effect on the overall impression of the Japanese people.

ANOVA (Analysis of Variance)

The emotional tone and framing effects of Bilibili videos significantly influence Chinese youth’s overall impressions of the Japanese people.

The nationality of the video producer significantly affects Chinese youth’s impressions of the Japanese people.

Longer videos—such as documentaries and dramas—also have a significant impact on these impressions.

Model Construction and Validation

The analytical model is structured as follows:

Independent Variables: Media agenda attributes, including video emotional tone (emotion), content orientation (tropesis), and video length (time).

Control Variables: Frequency of contact with Japanese culture, level of understanding of Japan, overall favorability toward Japan, and whether the respondent actively seeks information about Japanese culture (initiative).

Mediating Variables: Trust in Bilibili as a media platform (Trust), and viewer concentration while watching videos (focus).

Interactive Variables: Whether Danmaku is enabled (bullet_screen), frequency of Danmaku use (b_s_fre), frequency of posting comments (comment_fre), and video sharing habits (sharing_habits).

Other Variables: Gender and education level.

Dependent Variable: Audience agenda, represented in this study by the respondent’s overall impression of Japan (impression).

Multicollinearity was evaluated, with all VIF (Variance Inflation Factor) values below 2.5 and an average VIF of 1.4, indicating no significant multicollinearity concerns.

To improve the robustness of the model estimates, heteroscedasticity was tested. Since heteroscedasticity was present, robust standard errors were applied to adjust the regression results.

The table below presents the regression outcomes from Model 5, which includes a full combination of the variables. After controlling for cultural exposure, familiarity, favorability, proactive engagement, Danmaku-related interactions, commenting and sharing behaviors, as well as respondents’ education and gender, the model demonstrates a stable fit. All included variables show statistically significant effects.

Mediation Analysis

A mediation analysis of user trust in Bilibili as a media platform (Trust) indicates that trust has a significantly positive effect on impressions of Japan.

Experimental Method

Due to participant recruitment being limited to Renmin University of China, the three experiments conducted in this study faced limitations in achieving gender balance and had relatively small sample sizes. Nevertheless, the results still offer meaningful insights and provide partial evidence for how a single video can influence viewers’ attitudes toward Japan.

Experiment 1: Danmaku On vs. Danmaku Off (Random Assignment)

The presence or absence of Danmaku (real-time comments) and the level of political content in a video significantly influenced viewers’ attitudes toward Japan. Notably, whether Danmaku was enabled had a particularly strong effect. This highlights the unique influence of Bilibili, a platform known for its interactive Danmaku feature, in shaping viewers’ perceptions of Japan.

Experiment 2: Low-Politics vs. High-Politics Video (Random Assignment)

Participants who watched less politically charged videos—typically focusing on Japanese culture and everyday life—demonstrated increased understanding and more favorable impressions of Japanese culture and people. This, in turn, led to a general improvement in their overall attitude toward Japan.

Experiment 3: User-Generated vs. Organization-Produced Video (Random Assignment)

By contrast, the type of video producer (individual vs. organizational) did not have a significant effect on participants’ attitudes toward Japan.

Conclusion

This study outlines the following agenda-setting pathway:

Within a specific situational frame, video producers deliberately construct media agendas and present them in the form of online videos on Bilibili. Users receive and internally process these agendas based on their knowledge and emotional background, ultimately forming audience agendas—in this case, their impression of Japan. Importantly, users are not passive recipients; through interactions such as “danmaku” (on-screen comments) and comment sections, they may influence each other’s perceptions.

First, video producers shape media agendas based on their attributes and intentions. Survey results show that users focus more on video content than on whether the producer is an individual or an organization, indicating that individuals now rival organizations in agenda-setting influence. However, videos with overt commercial intentions tend to elicit negative responses.

When the producer is Japanese, the resulting impression of Japan is more favorable, suggesting that "self-narration" by nationals effectively promotes a positive national image.

Both the emotional orientation and content framing of a video significantly affect viewers’ impressions of Japan. Low-politics videos (e.g., about daily life or culture) are more likely to be trusted than high-politics ones, especially when viewers are already familiar with the content. Videos that are logically structured and well-argued also enhance viewer trust.

Video length matters: longer videos such as documentaries or dramas tend to provide a more nuanced and favorable impression. However, if too long, they risk viewer fatigue and reduced information retention.

In the individual-level frame, factors like gender show little influence, while age and education have modest effects. Frequency of exposure to Japanese culture and willingness to engage actively have significant impacts—greater familiarity and a proactive attitude result in more favorable impressions. Some respondents display a cognitive separation between the Japanese nation and people, often expressing emotional distance from the state.

Patriotic education, upbringing, and family background also play roles but are secondary. As young people grow older and gain broader perspectives, they tend to reassess and refine their impressions.

Media habits further shape audience agendas. Interaction through danmu and comment sections affects users’ trust in media content and, in turn, their impression of Japan. Danmu and comments promote critical thinking and expose users to diverse perspectives, which shape audience agendas more dynamically.

Bilibili’s unique ecosystem of danmu and open comment sections increases users’ media trust and enhances agenda-setting effectiveness. Both experimental and statistical data confirm that such interactivity serves as a mediator between media exposure and audience impression.

This study demonstrates that online videos—as an emerging media format—play a significant role in shaping audience agendas, particularly national image perceptions. Given the low barriers to content creation, public participation in agenda-setting has increased. Policymakers should therefore pay attention not only to how agendas are initially set but also how they are re-framed in the public domain. Strategic guidance is needed to prevent disordered public discourse.

The findings also offer practical insights for shaping a country’s international image. Non-political content—such as scenic landscapes, cultural practices, and cuisine—tends to foster more favorable impressions among foreign audiences. Videos that are logically sound, sufficiently detailed, and moderately long are more effective in winning viewer trust.

Finally, Bilibili’s interactive ecosystem significantly strengthens the agenda-setting impact of its videos. By leveraging similar mechanisms—such as comment-driven engagement—media platforms and policymakers can enhance the efficacy and credibility of their messaging strategies.