Ideological Inclination Embedded in Neutral Expression in Authoritarian Regime

• In Event: Poster Session: Public Policy Preferences and Attitudes toward Democracy & European Politics and Society
  In Poster Session: Poster Session: Public Policy Preferences and Attitudes toward Democracy

Abstract

In the context of a censored public opinion environment, this study explores how individuals express their views on controversial topics. Recent literature has introduced a theoretical framework centered around satirical forms of expression used by citizens in authoritarian regimes, such as memes and jokes. However, the effectiveness of satirical expression often hinges on contextual nuances and the users' comprehension of its underlying meaning. Therefore, employing a comprehensive big data approach, we analyzed over 16,231 responses from contemporary Chinese citizens on Zhihu, a prominent social media platform akin to Chinese Reddit or Quora, across ten diverse topics varying in their levels of controversy, including the sensitive topic of homosexuality. Leveraging semi-supervised machine learning techniques and text emotion analysis, our findings challenge the assumption that controversial topics elicit more extreme criticism; instead, criticism tends to revolve around emotionally neutral text tones. Additionally, we gathered user-related data, including their liked posts and followed individuals, allowing us to conduct a topic analysis that predicts users' potential ideological inclinations. Our findings suggest that individuals with conservative and traditionalist inclinations tend to express emotionally neutral criticism, infused with sympathy for traditional values. On the other hand, those with liberal and modernist perspectives similarly offer emotionally neutral criticism, but often advocate for change. This research will deepen our comprehension of public opinion dynamics within authoritarian regimes, illuminating implicit attitudes subtly woven into nearly neutral expressions within a censored public opinion environment.

Author

• Yang Fan, Hangzhou Normal University