Dissertation
Dual purposes of government communication in Chinese local governments
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Summer 2024
DOI: 10.25820/etd.007639
Abstract
This dissertation investigates the dynamics of Chinese local government propaganda on social media by analyzing over 1.4 million posts from more than 250 city governments. The study identifies three predominant topics shared by local governments: public goods information, dogmatic propaganda, and social news. It reveals that public goods information, which is inherently relevant and beneficial to individuals' daily lives, not only attracts a larger audience but also enhances the engagement across all types of propaganda content. Survey experiments further confirm that such information significantly boosts public satisfaction with local governments and lends them greater legitimacy. However, the analysis also exposes the low engagement and persuasive impact of dogmatic propaganda, which constitutes more than 40% of the content shared online by these governments. The dissertation argues that this type of propaganda primarily serves the opportunistic purposes of officials aiming to enhance their performance evaluations. This is evidenced by a strategic increase in propaganda output during official evaluations and before the end of tenures, particularly focusing on evaluation-related topics and among officials eligible for promotion. Such intensified efforts are shown to facilitate the career progression of mayors. Further findings indicate that the creation of propaganda accounts and the selection of their topics are heavily influenced by provincial government requirements and the practices of other cities within the same province, which are often the main competitors for mayoral promotions. In contrast, local socio-economic situations have minimal impact on these decisions, suggesting that the primary goal of managing social media propaganda is to satisfy upper government mandates rather than to influence public opinion genuinely. Additionally, this study demonstrates the effectiveness of transformer-based models, such as BERT and GPT, in text clustering, providing superior results compared to probabilistic topic models like Latent Dirichlet Allocation (LDA) and Structural Topic Model (STM). Given the growing prominence and capabilities of Large Language Models, this dissertation encourages social scientists to explore how these technologies can be leveraged for common analytical tasks in political science, including topic modeling. This research not only provides insights into the mechanics of propaganda in modern authoritarian regimes but also highlights the challenges and potential of digital media in political communication.
Details
- Title: Subtitle
- Dual purposes of government communication in Chinese local governments
- Creators
- Jianjun Yu
- Contributors
- Frederick Solt (Advisor)Frederick Boehmke (Committee Member)Bryce Dietrich (Committee Member)Haifeng Huang (Committee Member)Tianguang Meng (Committee Member)Elise Pizzi (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Political Science
- Date degree season
- Summer 2024
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.007639
- Number of pages
- x, 137 pages
- Copyright
- Copyright 2024 Jianjun Yu
- Language
- English
- Date submitted
- 07/16/2024
- Description illustrations
- illustrations, tables, graphs
- Description bibliographic
- Includes bibliographical references (pages 102-110).
- Public Abstract (ETD)
- This dissertation examines how Chinese local governments use social media for propa ganda by analyzing over 1.4 million posts from more than 250 city governments. It identifies three main types of content: information about public services, traditional propaganda praising the government, and general social news. The study finds that posts about public services, like traffic updates or health advice, are more engaging for the public and enhance the government’s image more effectively than traditional propaganda. Despite making up over 40% of the content, traditional propaganda—which often just praises the government—does not attract many read ers. This research suggests that such propaganda is frequently used by officials to look good in their performance reviews, particularly during times when evaluations are due or promotions are considered. Interestingly, the topics and timing of these posts are often dictated more by provin cial government demands or the actions of neighboring cities than by local needs. The findings highlight a gap between the goals of higher government levels, which push for the use of social media to shape public opinion, and local governments, which use these platforms to meet mini mum requirements or boost officials’ careers. Additionally, this study showcases how advanced technologies, like AI language models, can be used to analyze large amounts of text quickly and effectively, offering new tools for understanding political communication. This research provides insight into the complex relationship between government propaganda and public engagement in the digital age.
- Academic Unit
- Political Science
- Record Identifier
- 9984697940802771
Metrics
1 File views/ downloads
3 Record Views