How Does Information Flow from the World into China? [ 6 min read ]

Insights

  • Leveraging large-scale social media datasets, researchers find that between 24% and 28% of the most globally retweeted posts on Twitter pertaining to China and COVID-19 when the pandemic first broke out flowed into China’s information ecosystem.
  • China’s state-controlled media and commercialized domestic media played the largest role in facilitating the inflow of information. However, Weibo users without any media or government affiliation also played a role in transmitting these viral tweets, which are restricted from entering into China.
  • Among the viral tweets flowing into China, researchers identified 66% as antagonistic toward its government or people.
  • Nearly half of the inflow of antagonistic content into China was facilitated by its own government and state media channels. To explain this, the researchers suggest that at a time when China’s domestic propaganda amplifies the narrative that the “West” seeks to contain China, hostile content that crosses its borders may serve to reinforce the regime rather than threaten it.
  • Despite some transmission of viral tweets from outside its borders, researchers nevertheless conclude that unmanaged inflow of global information was limited.


Source Publication:  Yingdan Lu et al. (2022). How Information Flows from the World to China. The International Journal of Press/Politics. 

Is the connectivity of digital technology powerful enough to overwhelm the ability of strong regimes to regulate the inflow of global information across their borders? Nowhere is the effort to control the transnational flow of digital information more extensive and sustained than in China. In this study, the researchers attempt to answer whether censorship allows China’s government to act as the sole gatekeeper of information or whether digital technology enables unmanaged information to penetrate China’s digital firewall.

The data. From a repository of Twitter posts related to COVID-19, researchers extracted 1.8 million English-language tweets containing one or more China- related keywords between January 21 and April 30, 2020, when COVID-19 first emerged. Researchers divided this sample of 1.8 million tweets into each week and identified the top 10 most retweeted tweets per week related to COVID-19 and China. Their final sample consisted of 150 tweets (10 “viral” tweets for each of the 15 weeks within the sample).

Researchers then turned to a publicly available Weibo-COV dataset, which included all posts made by a sample of over 20 million active Weibo users. Using a deep-learning classifier to refine the process, the researchers ultimately created a final Weibo dataset containing 6.7 million COVID-19- related posts made between January 16 and April 30, 2020.

Using both a semi-automated deep-learning system and in-depth, manual investigative procedures, researchers sought to identify co-occurring content between the English-language Twittersphere and Chinese-language Weibo for each tweet. Researchers further investigated each tweet-Weibo match to ascertain whether it captured the flow of information from the global information ecosystem into China. Researchers define the term “inflow” as discussion among the Chinese public of events, actions, ideas, or opinions originating outside of China’s borders. Lastly, researchers identified the primary channels of information inflow into China and grouped them into four categories: China’s state-controlled media, commercialized domestic media, Weibo users not affiliated with any media outlet or government, and foreign entities that have Weibo accounts.

Level of information inflow into China. The researchers find that between 24% and 28% of the world’s most retweeted posts on Twitter pertaining to COVID-19 and China between January 21 and April 30, 2020, flowed into China’s information ecosystem. The research team cautions, however, that the study focuses on an exceptional period when COVID-19 first emerged and when China was in the global limelight. The study’s estimates, they suggest, likely represent the uppermost end of information inflow that can be captured from Twitter and Weibo rather than the average.

State-controlled and commercialized domestic media drive information inflow. Of the 150 most retweeted tweets about COVID-19 and China, researchers found that 32 tweets constituted inflow into China’s Weibo from outside its borders. Of these 32 viral tweets, the findings suggest that 53% of inflow was facilitated either by China’s state-controlled media or its commercialized domestic media. Drawing on earlier studies, researchers point out that both state-controlled and commercialized media outlets in China play a gatekeeping function for the regime and fulfill the propaganda goals of the Communist Party of China. This study therefore underscores how effectively China’s Party-state is able to control its social media landscape even during the height of COVID-19’s outbreak when China frequently occupied the global media spotlight.

Weibo users and foreign entities facilitate smaller share of information inflow. Although communication channels heavily regulated by the P.R.C. played a dominant role in facilitating the inflow of global communication content, researchers also found that Weibo users without any media or government affiliation consistently transmitted global information into China (12 of 32 viral tweets). Weibo users living outside of China, for instance, were playing a role in information transmission, while in others, Weibo users were sourcing information from foreign media channels such as The New York Times and overseas Chinese media outlets. In limited instances, furthermore, foreign entities also transmitted information into China (3 of 32 tweets) by directly posting to their Weibo accounts.

State media channels facilitate inflow of antagonistic information toward China. Researchers further analyzed what types of information made their way into China. More specifically, they examined whether tweets that were antagonistic toward the Chinese regime or its people were more or less likely to flow into the country. They found that content hostile toward China’s government and people was more likely to flow into the country than other types of content: among the top 150 viral tweets pertaining to China and COVID-19 that constituted the overall sample, they found that 55 tweets (37%) were antagonistic toward China. Among the 32 out of 150 tweets that the authors determined actually entered China, however, a larger share of content (66%) was antagonistic toward China. Perhaps even more striking, the data showed that government and state media accounts transmitted the largest share of such antagonistic content (48%) compared to any other channels. To explain why this might be so, the researchers point out that the Chinese government regularly amplifies the narrative that the “West” seeks to contain China and stymie its rise. Facilitating the inflow of hostile content from outside its borders, they suggest, reaffirms this view, and may even serve to reinforce the regime rather than threaten it.

Limits on unmanaged information inflow. While censorship combined with media control provides substantial scope for China’s government to shape its digital landscape, the evidence also suggests that social media does provide an opportunity for private actors to influence China’s information environment. Nonetheless, unmanaged inflow of global information into China was limited. As the study demonstrates, even during a global pandemic highly salient in China and around the world, a minor portion of viral content pertaining to COVID-19 and China made its way onto Weibo in China.