Content Styles and User Engagement on Social Media: An Empirical Study Based on Marvel and Legendary Accounts

Abstract

This empirical study investigates the impact of content styles on user engagement across social media platforms, specifically analyzing the official accounts of Marvel and Legendary Entertainment on Weibo, Douyin, and Instagram. By employing quantitative research methods—namely content analysis, descriptive statistics, analysis of variance (ANOVA), and mediation analysis—the study examines 60 social media posts, each categorized into one of four content styles: informational, entertainment, emotional, and interactive. The results reveal notable differences in engagement performance: interactive content generates significantly more comments, while emotional content leads to higher shares. Furthermore, emotional word density plays a mediating role in the relationship between emotional content and user sharing behavior, suggesting that emotionally charged language enhances content virality. These findings support the framework of Uses and Gratifications Theory and Emotional Contagion Theory, confirming that users are more likely to engage with content that meets their emotional and participatory needs. This study offers valuable practical implications for media and entertainment brands aiming to optimize their social media strategies through tailored content design and emotional resonance.