By Ileana Wachtel January 29, 2024
Chinese citizens who rarely voice open criticism of their government reveal stronger negative views when they can answer questions anonymously, according to a new study published in The China Quarterly.
The study by researchers at the USC Dornsife College of Letters, Arts and Sciences shows an enormous drop in citizen support for the Chinese Communist Party (CCP) and government policies when citizens are surveyed using a method that hides their identities and makes them feel more anonymous than a typical survey.
Why it matters: The study authors — Erin Baggott Carter and Brett Carter, both assistant professors of international relations, and doctoral candidate Stephen Schick — believe documenting citizens’ true opinions on these matters is key to understanding the depth and nature of support for the CCP and its government policies.
- Accurately documenting citizens’ opinions is important to disprove recent assertions that autocratic governments, including the CCP, genuinely have popular support.
- The study also highlights the complexities of public opinion under autocratic regimes. Still, even when using online methods that hide a person’s identity, such as “list experiments,” many worry about surveillance by the government.
The big picture: Traditional surveys conducted in China, which directly question respondents, overstate Chinese citizens’ support for the CCP by up to 28.5 percentage points, according to the study.
- When using a method called a “list experiment,” which confers a heightened sense of anonymity by asking respondents how many statements they agree with rather than which ones, researchers showed CCP support hovers between 50% and 70%, and not 90% as reported in traditional surveys.
- The traditional direct-question survey also showed that only 8% of citizens cited fear of repression as a reason for not protesting, while the list experiment survey revealed that about 40% acknowledged fear as a deterrent.
Zoom in: When the researchers looked at regime support across several demographics, the list experiments revealed that ethnic Han respondents, educated respondents, and CCP members were more supportive.
- When asked directly, 94% of respondents said they backed Chinese President Xi Jinping, and 91% said they believed the government works for the people.
- When using the list experiment method, on the other hand, support for Xi and the government dropped by nearly 30 percentage points.
- In list experiment surveys, however, college-educated respondents are between 10 and 20 percentage points more supportive of the CCP than those who completed early middle school.
- This may suggest the CCP’s efforts to reshape educational curricula have succeeded.
- Similarly, in list experiment surveys, CCP members are about 10 percentage points more supportive of the regime than in direct questioning surveys.
- CCP strategies to win over these demographics have been effective.
Go deeper: A surprising finding showed that when asked directly, respondents who are members of China’s predominant and favored ethnic group, the Han, are less inclined to express support for Xi compared to minority respondents.
- This suggests that minority groups tend to conceal their true beliefs more than Han respondents about Xi and the government when asked directly.
- When using the list experiment survey, this result flips, revealing that ethnic Han support for Xi is about 20 percentage points more than minorities.
What they’re saying: “Given recent claims that autocratic governments often enjoy genuine support, it is especially important to get accurate information about what citizens in countries like China feel about their governments,” said Baggott Carter.
The upshot: The study highlights the need for refined survey methods in politically sensitive areas.
- The researchers underscore the need for scholars to stop using direct-question surveys to measure public opinion in China and other repressive environments.
- They also urge a reassessment of presumed broad support for regimes such as the CCP.
- Finally, the researchers recommend scholars reevaluate previous empirical work that overlooks the existence of concealed preferences.
