Reverse-engineering censorship in China: Randomized experimentation and participant observation

Science

Volumn 345, Issue 6199, 2014, Pages

  King, Gary ^a   Pan, Jennifer ^a   Roberts, Margaret E ^b  

^a HARVARD UNIVERSITY   (United States)

^b UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; CHINA; CONTROLLED CLINICAL TRIAL; CONTROLLED STUDY; CROWDING (AREA); GOVERNMENT REGULATION; HUMAN; INTERNET; INTERPERSONAL COMMUNICATION; LEGAL ASPECT; RANDOMIZED CONTROLLED TRIAL; SOCIAL MEDIA;

BLOGGING; CHINA; COMMUNICATION; CROWDING; GOVERNMENT REGULATION; HUMANS; SOCIAL MEDIA;

EID: 84906707964     PISSN: 00368075     EISSN: 10959203     Source Type: Journal    
DOI: 10.1126/science.1251722     Document Type: Article

References (49)

1. Freedom House, "Freedom of the press, 2012"; www.freedomhouse.org.
2. G. King, J. Pan, M. E. Roberts, How censorship in China allows government criticism but silences collective expression. Am. Pol. Sci. Rev. 107, 326 -343 (2013). doi: 10.1017/S0003055413000014
3. S. L. Shirk, China: Fragile Superpower: How China's Internal Politics Could Derail Its Peaceful Rise (Oxford Univ. Press, New York, 2007).
4. S. L. Shirk, Changing Media, Changing China (Oxford Univ. Press, New York, 2011).
5. M. K. Whyte, Myth of the Social Volcano: Perceptions of Inequality and Distributive Injustice in Contemporary China (Stanford Univ. Press, Stanford, CA, 2010).
6. L. Zhang, A. Nathan, P. Link, O. Schell, The Tiananmen Papers (Public Affairs, New York, 2002).
7. A. Esarey, Q. Xiao, Political expression in the Chinese blogosphere: Below the radar. Asian Surv. 48, 752-772 (2008). doi: 10.1525/AS.2008.48.5.752
8. R. MacKinnon, Consent of the Networked: The Worldwide Struggle For Internet Freedom (Basic Books, New York, 2012).
9. P. Marolt, Grassroots agency in a civil sphere? Rethinking internet control in China. In Online Society in China: Creating, Celebrating, and Instrumentalising the Online Carnival, D. Herold, P. Marolt, Eds. (Routledge, New York, 2011), pp. 53-68.
10. M. Dimitrov, The resilient authoritarians. Curr. Hist. 107, 24-29 (2008).
11. P. L. Lorentzen, Regularizing rioting: Permitting public protest in an authoritarian regime. Q. J. Pol. Sci. 8, 127-158 (2013). doi: 10.1561/100.00012051
12. P. L. Lorentzen, China's strategic censorship. Am. J. Pol. Sci. 58, 402-414 (2014). doi: 10.1111/ajps.12065
13. X. Chen, Social Protest and Contentious Authoritarianism in China (Cambridge Univ. Press, New York, 2012).
14. E. Malesky, P. Schuler, Nodding or needling: Analyzing delegate responsiveness in an authoritarian parliament. Am. Pol. Sci. Rev. 104, 482-502 (2010). doi: 10.1017/S0003055410000250
15. G. Distelhorst, Y. Hou, Ingroup bias in official behavior: A national field experiment in China. Q. J. Pol. Sci. 9, 203-230 (2014). doi: 10.1561/100.00013110
16. D. Bamman, B. O' Connor, N. Smith, Censorship and deletion practices in Chinese social media. First Monday 17 (no. 3) (March 2012). doi: 10.5210/fm.v17i3.3943
17. T. Zhu, D. Phipps, A. Pridgen, J. Crandall, D. Wallach, The velocity of censorship: High-fidelity detection of microblog post deletions. In 22nd USENIX Security Symposium (Washington, DC, 14 to 16 August 2013); www.usenix.org/ conference/usenixsecurity13/technical-sessions/paper/zhu.
18. R. MacKinnon, China's censorship 2.0: How companies censor bloggers. First Monday 14 (no. 2) (February 2009). doi: 10.5210/fm.v14i2.2378
19. See supplementary materials on Science Online.
20. We also added our own ethics rules, not required by the IRB, which dictate that we avoid, wherever possible, influencing or disturbing the system we are studying (19). The similarity to the Prime Directive in Star Trek notwithstanding, this seems like the appropriate stance for scientists attempting to understand the world, as distinct from advocates trying to change it, and in any event is more likely to yield reliable inferences.
21. In the process of setting up the site, they recommended that we hire two or three censors for every 50,000 users. That enables us to back out an estimate of the total number of censors hired within firms at between 50,000 and 75,000, not counting censors within government, 50 Cent Party members, or the Internet police.
22. See (40, 41) for numbers of registered users, which are substantial even if we account for automated sites created by marketing firms (42).
23. K. Imai, G. King, E. Stuart, Misunderstandings between experimentalists and observationalists about causal inference. J. R. Stat. Soc. Ser. A 171, 481-502 (2008). doi: 10.1111/j.1467-985X.2007.00527.x
24. All posts were made to mainland China accounts. Some were submitted from outside China, when feasible, and many from within China. Recent work has noted that overseas accounts are subject to less stringent censorship regulations than mainland accounts (43). This issue does not affect our work because all accounts created and used were mainland China accounts. Users could control account location when creating the account by specifying a location in China, by entering a local mobile number, or by creating the account from a local IP address. We used all of these methods.
25. For each of our three rounds, we wrote 200 posts on non- collective action events (split equally between proand anti-government) and 200 posts on collective action events or edge cases (again split equally between proand anti-government). Thus, 600 posts submitted relate to non- collective action events, and 600 relate to collective action events or edge cases. We have in total four collective events and two edge cases, and so 400 posts focused on collective action events and 200 on edge cases.
26. J. Crandall et al., Chat program censorship and surveillance in China: Tracking TOM-Skype and Sina UC. First Monday 18 (no. 7) (July 2013). doi: 10.5210/fm.v18i7.4628
27. J. Fallows, The connection has been reset. Atlantic (March 2008); www.theatlantic.com/magazine/archive/2008/03/-the-connection-has-been-reset/ 306650.
28. K. Imai, L. Keele, D. Tingley, T. Yamamoto, Unpacking the black box of causality: Learning about causal mechanisms from experimental and observational studies. Am. Pol. Sci. Rev. 105, 765-789 (2011). doi: 10.1017/S0003055411000414
29. J. Pearl, Direct and indirect effects. In Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (Morgan Kaufmann, San Francisco, 2001), pp. 411-420; https://dslpitt.org/uai/papers/01/p411-pearl.pdf.
30. O. Blanchard, A. Shleifer, "Federalism with and without political centralization: China versus Russia" (National Bureau of Economic Research, Cambridge, MA, 2000); www.nber.org/papers/w7616.
31. S. Heilmann, E. Perry, Mao's Invisible Hand: The Political Foundations of Adaptive Governance in China (Harvard University Asia Center, Cambridge, MA, 2011).
32. Y. Qian, G. Roland, Federalism and the soft budget constraint. Am. Econ. Rev. 88, 1143-1162 (1998).
33. Y. Qian, B. R. Weingast, Federalism as a commitment to perserving market incentives. J. Econ. Perspect. 11, 83-92 (1997). doi: 10.1257/jep.11.4.83
34. G. Egorov, K. Sonin, Dictators and their viziers: Endogenizing the loyalty-competence trade-off. J. Eur. Econ. Assoc. 9, 903-930 (2011). doi: 10.1111/j.1542-4774.2011.01033.x
35. G. Salton, Automatic Text Processing: The Transformation, Analysis, and Retrieval of Information by Computer (Addison-Wesley, Reading, MA, 1988).
36. D. Kelleher, S. Luz, Automatic hypertext keyphrase detection. In Proceedings of the 19th International Joint Conference on Artificial Intelligence (Erlbaum, Hillsdale, NJ, 2005), pp. 1608-1609.
37. For examples, see (44).
38. For news coverage of the protests, see (45-48).
39. D. Hopkins, G. King, A method of automated nonparametric content analysis for social science. Am. J. Pol. Sci. 54, 229-247 (2010). doi: 10.1111/j.1540-5907.2009.00428.x
40. K. Hong, China's Twitter-like Sina Weibo service now has over 50 million active users per day. The Next Web, 13 August 2013; http://tnw.co/1fdNFPS.
41. S. Millward, Tencent Weibo, the 'other weibo' that nobody cares about, reaches 540 million users. Tech in Asia, 22 January 2013. http://bit.ly/1byjSNW.
42. K. W. Fu, M. Chau, Reality check for the Chinese microblog space: A random sampling approach. PLOS ONE 8, e58356 (2013). doi: 10.1371/journal.pone. 0058356; pmid: 23520502
43. J. Q. Ng, Weibo keyword un-blocking is not a victory against censorship. Tea Leaf Nation, 21 June 2013; http://bit.ly/1kfqNBC.
44. P. Barefoot, "Principal, get a room with me, spare the schoolchildren!" China Smack, 31 May 2013; http://j.mp/19yuv7E.
45. Al-Jazeera, Democracy push as Hong Kong marks handover. 1 July 2013; http://j.mp/145Jvpp.
46. S. Lee, K. Wong, Hong Kong protests to underscore Leung's record-low appeal. Bloomberg BusinessWeek, 28 June 2013; http://j.mp/13r3v7v.
47. J. Ngo, July 1 protest is Hong Kong's taste of democracy. South China Morning Post, 30 June 2013; http://j.mp/15PcwBt.
48. C. Yung, Annual Hong Kong protest to focus ire on leader. Wall Street Journal, 28 June 2013; http://j.mp/13FJB3w.
49. G. King, J. Pan, M. E. Roberts, Replication Data for: Reverse Engineering Chinese Censorship: Randomized Experimentation and Participant Observation (2014). doi: 10.7910/DVN/26212