Persuasion or Insulting? Unpacking Discursive Strategies of Gender Debate in Everyday Feminism in China

Title of the Paper

Persuasion or Insulting? Unpacking Discursive Strategies of Gender Debate in Everyday Feminism in China

Paper Information

• Subject Area: Gender issues, social media, feminist online discussions
• Keywords: Gender debate, everyday feminism, linguistic strategies, social media, feminist advocacy, constructive design

Research Background and Issues

• What problems or challenges did the authors identify?
  • Everyday feminism in China has gradually become a critical topic on social media, but it has also sparked intense gender debates.
  • Gender debates often involve insulting or confrontational language, which may escalate conflicts, induce fear, and reduce public well-being.
  • Intense discussions on social media lack systematic research, particularly on how users employ linguistic strategies to express opinions and influence others.
• Why is this issue important?
  • Gender debates are an essential component of advancing social gender equality, but aggressive and uncivil communication may undermine feminist advocacy efforts.
  • Understanding the linguistic strategies behind these debates can support more constructive discussions and provide insights for technological design.
• Research Motivation and Related Work
  • The authors reviewed prior studies on feminist theories and design, as well as the background of gender discussions on social media, highlighting that existing research focuses more on one-way attacks rather than two-way interactions.
  • This study aims to fill the research gap in user-generated linguistic strategies and their effects.

Solutions

• What methods or solutions did the authors propose?
  • A mixed-method approach combining qualitative coding and quantitative regression analysis to systematically analyze linguistic strategies in gender debates.
  • Proposed a classification system for user-generated debate strategies and evaluated their impact on user engagement and responses.
• What is innovative about this solution?
  • The authors derived strategies from user-generated content rather than existing theoretical frameworks, making them more applicable to real-world scenarios.
  • They comprehensively explored the relationship between uncivil and constructive interactions in gender debates and analyzed how strategies influence user behavior and attitudes.
• What are the implementation steps and key technologies used?
  1. Data Collection and Preprocessing:
     • Collected 38,636 gender-related discussion posts and 187,539 comments from one of China's largest social media platforms, Weibo.
     • Filtered data based on keywords and further refined relevant content using the BERT model.
  2. Strategy Classification and Analysis:
     • Employed open coding to extract 15 linguistic strategies from 1,928 random samples, categorizing them into five major groups.
  3. Correlation Analysis:
     • Used negative binomial regression analysis to examine the impact of linguistic strategies on user engagement (e.g., likes, comments, shares) and user responses (e.g., uncivil, constructive, supportive, or opposing stances).

Research Findings

• What specific findings were obtained?
  1. Strategy Classification:
     • Identified five major categories of linguistic strategies: derogatory strategies (e.g., sarcasm, insulting labels), gender differentiation strategies (e.g., role reversal), intensification strategies (e.g., overgeneralization), mitigation strategies (e.g., suggestions), and cognitive guidance strategies (e.g., empathy elicitation and gender education).
  2. Strategy Impact:
     • Certain strategies showed significant positive effects, such as role reversal, which promoted user sharing behavior (increased shares by approximately 7.27 times).
     • Other strategies, such as sarcasm and overgeneralization, attracted opposing opinions and could exacerbate conflicts.
  3. Gender Differences:
     • Female users tended to use strategies emphasizing gender differences, such as "gender exclusion," while male users were more likely to adopt suggestion-based or problem-revealing strategies.
  4. Polarization Trends:
     • The proportion of uncivil content significantly increased after 2018, although constructive interactions still accounted for a notable share.
• What advantages does it have compared to existing solutions?
  • Systematically revealed the diversity of strategies in gender debates rather than focusing solely on one-way attacks or uncivil content analysis.
  • Combined behavioral data (e.g., likes, shares) to analyze strategy effects, providing more quantitative and intuitive insights.
• What were the experimental or evaluation results?
  • Regression analysis confirmed the positive impact of strategies such as role reversal and evidence provision on user engagement, as well as the tendency of strategies like sarcasm to attract opposing viewpoints.
  • Found that suggestion strategies might cause a "rebound effect" (triggering more opposing opinions), highlighting the complexity of strategy usage.
• Limitations and Future Directions
  1. Data Sampling Issues: The study is based on partial samples, which may affect the comprehensiveness of the results.
  2. Platform Censorship: Posts on Weibo may have been censored, reducing the presence of sensitive or extreme content in the dataset.
  3. Data Authenticity: Gender information of users on social media may not align with reality, impacting conclusions about gender differences.
  4. Insufficient Research on Strategy Interactions: Did not deeply explore how strategies in posts influence strategies in comments.

Conclusion

This study, through data analysis of everyday feminism gender debates on Chinese social media, proposed a comprehensive classification method for user-generated strategies for the first time and explored their multidimensional impact on user engagement and responses. The research provides a new perspective on how social media can facilitate constructive discussions on gender issues, while also highlighting the complexity and challenges of everyday feminism in the digital age.