Strategies of Chinese State Media on Twitter

ABSTRACT

How do state-controlled broadcasters reach foreign publics to engage in public diplomacy in the era of social media? Previous research suggests that features unique to social media, such as the ability to engage in two-way communication with audiences, provide state-controlled broadcasters new opportunities for online public diplomacy. In this paper, we examine what strategies were used by four Chinese state-controlled media outlets on Twitter to reach foreign publics as the Chinese Communist Party worked to expand its public diplomacy and international media outreach efforts. We find that all outlets increased the volume and diversity of content while none engaged in interactive, two-way communication with audiences, and none appeared to artificially inflate their follower count. One outlet, China Global Television Network, made outsized gains in followership, and it differs from the other Chinese outlets in that it was rebranded, it disseminated a relatively lower share of government-mandated narratives pertaining to China, and the tone of its reporting was more negative. These results show that during a period when Chinese state-controlled broadcasters gained followers on Twitter, outlets made limited use of features unique to social media and instead primarily used social media as a broadcast channel.

KEYWORDS:

• State-media
• china
• social media
• twitter
• public diplomacy

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Open scholarship

This article has earned the Center for Open Science badges for Open Data and Open Materials through Open Practices Disclosure. The data and materials are openly accessible at https://doi.org/10.7910/DVN/EVVQGV.

Disclosure statement

J.P. received direct research funding from Meta in 2017 and received an honorarium from Meta for serving on the News Integrity Expert Circle from 2020-2021; J.P. has attended an event where food was paid for by a social media company.

Supplemental Material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/10584609.2023.2233911

Notes

1. Note that this is not the first time the CCP has sought to reach global audiences. This is merely the latest iteration of China’s efforts to go global (Jirik, 2009). We use the term “Going Global” in line with others’ usage of the term to reference this 2017 effort (c.f. Thussu et al. (2017)).

2. There is debate over the distinction between these concepts, especially since the term propaganda is used differently across disciplines – ranging from any content produced by state-controlled outlets to only content that persuades through deception or coercion (Bakir et al., 2019; H. Huang, 2015; Macdonald, 2006; L. J. Martin, 1971). However, what is common between public diplomacy and propaganda is the idea of intervening in other countries’ information environments for geopolitical gain.

3. Their Twitter handles are CGTNOfficial, XHNews, PDChina, and ChinaDaily.

4. See http://www.gov.cn/xinwen/2016-03/17/content_5054992.htm 拓展海外传播网络，丰富传播渠道和手段。打造旗舰媒体，推进合作传播，加强与国际大型传媒集团的合资合作，发挥各类信息网络设施的文化传播作用。

5. We focus on English-language content to maximize the geographic scope of the analysis. Focusing on English-language content allows us to look at how Chinese state media reaches audiences in Africa, Asia, Australia, Europe, North America, and South America.

6. We chose three million as the threshold because this was the size of RT’s following, which we used as a proxy for global reach.

7. Twitter handles are AJEnglish and RT com. Al-Jazeera has two Twitter accounts: AJEnglish with 8.1 million followers and AJENews with 2.1 million followers. We analyze AJEnglish in this paper, but the analysis is consistent if we use AJENews.

8. Twitter handles are BBCBreaking and cnnbrk.

9. What evidence we have suggests that followers of BBC and CNN are global, not domestic. Based on users who allow geolocation on Twitter, the country with the highest number of followers of BBCBreaking is India, followed by the United States, then Nigeria, UK, Pakistan, and South Africa. Overall, among users allowing geo-location, 18% of CNN followers are US-based (the remaining 82% are based elsewhere in the world), and only 11% of BBC followers are UK-based.

10. From https://www.fmprc.gov.cn/web/fyrbt_673021/jzhsl_673025/202105/t20210511_9171262.shtml ; in Chinese: “面对由BBC、CNN等媒体占据主导地位的全球媒体格局，中国正在创造一个全球新闻媒体的替代选择…不应只有CNN和BBC.”

11. See http://cpc.people.com.cn/n/2013/0821/c64094–22636876.html.

12. This strategy has undergone some revisions in the intervening years. For example, some categories were further clarified in 2018; see http://china.chinadaily.com.cn/2018-12/07/content_37371741.htm.

13. For original text of speech, see http://www.xinhuanet.com/politics/2013-12/31/c_118788013.htm.

14. See http://china.chinadaily.com.cn/2018-12/07/content_37371741.htm.

15. See http://www.china.org.cn/chinese/2014-01/16/content_31213800.htm.

16. See Section 41 http://www.gov.cn/jrzg/2013-11/15/content_2528179.htm. 提高文化开放水平。坚持政府主导、企业主体、市场运作、社会参与，扩大对外文化交流，加强国际传播能力和对外话语体系建设，推动中华文化走向世界。理顺内宣外宣体制，支持重点媒体面向国内国际发展。培育外向型文化企业，支持文化企业到境外开拓市场。鼓励社会组织、中资机构等参与孔子学院和海外文化中心建设，承担人文交流项目。

17. See http://www.gov.cn/zhuanti/2014-01/31/content_2596913.htm.

18. See http://www.gov.cn/xinwen/2016-03/17/content_5054992.htm.htm 拓展海外传播网络，丰富传播渠道和手段。打造旗舰媒体，推进合作传播，加强与国际大型传媒集团的合资合作，发挥各类信息网络设施的文化传播作用。

19. See socialblade.com. The earliest available data differs across the four outlets because Social Blade began its data collection at different times (see Section S1 of the SI). However, the same methodology was used for all accounts and over time to collect follower numbers. When we examine audience growth over time by looking at the metadata of accounts’ list of followers, which is sorted according to date, the same pattern holds. This method is less precise than a daily snapshot of follower count, which is what Social Blade collects, but the consistency of results gives us reassurance about the validity of Social Blade’s methodology.

20. For details on ITSA, see Appendix A.1.4.

21. See https://www.independent.co.uk/tech/twitter-fake-followers-lost-delete-accounts-culla8444236.html, https://www.washingtonpost.com/technology/2018/07/06/twitter-is-sweeping-out-fakeaccounts-like-never-before-putting-user-growth-risk/.

22. There is limited geographic information about CGTN followers. Based on analysis of accounts that do provide geographic info, there is a larger share of followers from regions such as South Asia, Africa, South America, and the Middle East than from North America or Europe.

23. See https://botometer.osome.iu.edu/.

24. The 95% threshold is also recommended by the developers of the Botometer method.

25. See https://twitter.com/CGTNOfficial/status/963880542650368001.

26. A total of six non-Chinese language CCTV channels were rebranded as CGTN.

27. See https://www.cgtn.com/about-us and http://www.xinhuanet.com/politics/201612/31/c1120226953.htm.

28. See https://www.thepaper.cn/newsDetail_forward_1641624 forward 1,641,624. One year later, in March 2018, CGTN merged with China National Radio and China Radio International to form the media conglomerate China Media Group (http://politics.people.com.cn/n1/2018/0322/c1001–29883390.html). This change, which also affected China National Radio and China Radio International, means that while none of the other three outlets in our analysis rebranded, CGTN was not the only Chinese state broadcaster to undergo rebranding.

29. Not all CCTV social media accounts were rebranded as CGTN; some CCTV accounts remained under the CCTV name.

30. See: http://web.archive.org/web/20190216145101/https://twitter.com/CGTNOfficial

31. See: http://web.archive.org/web/20221026071029/https://twitter.com/CGTNOfficial ³³Prior to 2017, the CGTN Twitter account belonged to that of CCTV-9/CCTV News.

32. Although CGTN was announced on December 31st 2016, it was not registered for operations until March 2017 and did not begin actively producing social media content under the CGTN brand until July 2017.

33. We use word2vec to train our own word embeddings with dimension size 100 and window size 5 using all available tweets from each outlet.

34. Preprocessing includes removing unreadable characters, standardizing different stylized references to the same country, removing stop words, and lemmatization. Unreadable characters are removed and all tokens are converted to lowercase regardless of whether the aforementioned custom preprocessing is performed.

35. Results remain the same when we use other sentiment scoring methods such as sentimentr (Rinker, 2019).

36. For reference, international broadcasters RT and Al-Jazeera cover a high but smaller number of countries, an average of 106 countries per month; BBC and CNN cover a smaller number of countries, 32 and 47 countries per month during the period of interest, respectively. Figure S6 in Appendix A.2.4 shows the the trend-line of the number of countries mentioned by both Chinese and non-Chinese outlets.

37. The baseline sentiment for Al-Jazaeera, RT, BBC, and CNN ranged from −0.1 to −0.2 on a −1 to 1 scale. See Figure S7 and Table S8 in the Appendix.

38. In 2018, Xinhua is said to have used Twitter ads to boost Xinhua content portraying Hong Kong protesters as violent (https://techcrunch.com/2019/08/19/twitter-is-blocked-in-china-but-its-state-newsagency-is-buying-promoted-tweets-to-portray-hong-kong-protestors-as-violent/). On Facebook, Xinhua and CGTN are among the largest buyers of ads (Molter & DiResta, 2020; Tambe & Friedman, 2022). Using the Facebook Ad Library,⁴¹, which has data going back to May 7, 2018, we found that between May 7, 2018 and May 7, 2019, Xinhua ran approximately 230 ads, including ads in all content categories, targeting all global regions (@XinhuaNewsAgency on Facebook and @chinaxinhuanews on Instagram). In the same time period, with the same inclusion criteria, CGTN ran approximately 160 ads (@ChinaGlobaTVNetwork on Facebook, @cgtn on Instagram); People’s Daily ran three ads (@PeoplesDaily on Facebook and @peoples daily on Instagram); and China Daily ran 10 ads (@chinadaily on Facebook and @chinadailynews on Instagram).

39. For example, Al Jazeera’s popularity is associated with its ability to report breaking news and access information that other outlets could not (Seib, 2010).

Additional information

Funding

The work was supported by the John S. and James L. Knight Foundation National Science Foundation [1831481].

Notes on contributors

Yingjie Fan

Yingjie Fan is a Ph.D. Candidate in the Department of Politics at Princeton University.

Jennifer Pan

Jennifer Pan is a Professor of Communication and Senior Fellow at the Freeman Spogli Institute at Stanford University.

Jaymee Sheng

Jaymee Sheng is a data scientist and applied AI researcher. Her research focuses on responsible application of natural language processing and AI in healthcare and public policy.