In a political environment that remains (at best) officially skeptical about the enterprise, representative sample surveys on Chinese politics have nonetheless grown substantially in number in the past two decades: political scientists trained and based outside mainland China conducted a mere two such surveys in the 1980s, but the number increased more than tenfold in the 1990s and continues to rise steadily. By mid-2008, some sixty articles, books, and book chapters drawing from original representative sample surveys on Chinese politics had appeared – including many articles in top-tier journals of political science and area studies alike. This chapter surveys the surveys and their products. It briefly explains the focus here on probability sample surveys and describes the changing regulatory context within which researchers conduct their surveys. Most of all, it evaluates their achievements, with attention to their cumulativeness, contributions to knowledge, and fit in Chinese area studies. The chapter is not a primer on the conduct of survey research on Chinese politics in mainland China. It is instead a status report and reflection on this research, aimed as much (or more) at its consumers (and nonconsumers) as at survey researchers themselves.

From the corpus of English-language monographs and peer-reviewed journal articles authored by political scientists and published in nonmainland sources through mid-2008, I identify studies that exploit original probability sample survey data. In coauthored works, I include studies that meet these criteria so long as at least one of the authors is a political scientist. I have surely missed some relevant surveys in my search – but not, I think, any represented in studies published in major journals of political science or Chinese area studies or organized by any of the roughly half-dozen major players in the enterprise of survey research on Chinese politics. I focus in this chapter on original surveys, not the small literature by political scientists who analyze datasets produced wholly or mainly by others. At the same time, the paucity of studies analyzing existing datasets is a sign of the relative immaturity of survey research on Chinese politics, a topic that merits discussion and is taken up below.

For decades, students of Chinese foreign policy were confronted with a stark dearth of information relating to China’s position in the international arena. In the 1960s, Allen Whiting’s seminal work, China Crosses the Yalu, while drawing from the author’s extensive governmental experience, referenced only a limited pool of official Chinese sources to describe Beijing’s stance on the Korean peninsula. In the 1970s, the main source of information for Samuel Kim’s book, China, the United Nations and World Order, was Chinese votes in the United Nations General Assembly. A few years later, A. Doak Barnett’s short, but influential, The Making of Foreign Policy in China, was also marked by the use of a limited number of sources. In contrast, in the late 1980s, a new generation of scholars began to gain access to a somewhat broader set of data. The best examples of this trend were Tom Christensen’s consideration of newly available documents relating to the Korean War, followed by Iain Johnston, David Shambaugh, and Robert Ross’s utilization of extensive interview data. However, this being the case, the general informational frame for researching Chinese foreign policy has remained relatively static since the early 1990s, with researchers repeatedly making use of the same limited set of sources (a handful of Chinese-language journals dedicated to international politics, official statements, a smattering of 内部 documents, and interviews with a small circle of foreign policy elites). In comparison, it is widely perceived that the study of Chinese foreign policy is now poised to enter a new era. Indications of such an incipient development are purported to be found in the expanding availability of new data. The most prominent of these sources is generally seen to be the Chinese Internet, which appears to contain a treasure trove of new information related to China’s foreign relations. The question then becomes how significant is the Internet to the study of Chinese foreign policy?

Only two decades ago the main information source for China scholars abroad was the Foreign Broadcast Information Service (FBIS), an open intelligence source of the CIA. FBIS collected, translated, and disseminated available news and information from Chinese media sources. During the Cultural Revolution most foreign researchers were not allowed to conduct fieldwork in China; the five-times-weekly FBIS report thus constituted one of the main sources for information about events in the mainland. Since original media sources only became available during the reform era, those interested in Chinese media reporting usually relied on the information selected and translated by FBIS. Aimed at the U.S. intelligence community, FBIS did not draw an unbiased and representative sample of the Chinese news, but, at the time, it constituted one of the best information sources available to the research community abroad.

Today, the situation could not be more different. Since China’s opening up to the outside world, we have the opportunity to spend extensive time in the country reading, listening, and watching the same news as Chinese citizens. Even when we are not in China, we can access a large number of Chinese media sources electronically using the Internet, cable, or satellite dishes. In addition to increased access, we also have the opportunity to store this information. We can mail abroad newspapers, magazines, and DVDs from China, scan documents, download files, and record television and radio programs. Most of this information is stored electronically. Once preserved, the data gathered in countless hours sit on our hard drives, waiting for analysis. Instead of reading the FBIS report, we simply access a folder on a computer and immediately have data available for research. What a wonderful world for doing research on China – or so it would seem.

Descriptive statistics from representative sample surveys conducted in mainland China provide a static picture that is often soon overtaken by the impact of rapid socioeconomic change. The importance of longitudinal data generally, but especially in such a context, cannot be overstated. Survey researchers seem to recognize this: as discussed in Chapter 10, a remarkable number of surveys conducted in mainland China have a longitudinal component. By far the most ambitious of these is the Beijing Area Study (BAS), an ongoing annual representative sample survey of Beijing residents, designed and conducted since 1995 by the Research Center for Contemporary China (RCCC) at Peking University. This chapter begins with an introduction to the underlying vision and goals of the BAS, as conceived in the early 1990s. It then turns to specific issues of questionnaire content, sampling design, and survey implementation. We pay particular attention to the challenges and changes faced over the first decade of the BAS; major changes in sampling were made in 2007, however, and we review these here. For the most part, we do not present survey findings, except to illustrate particular points in the discussion of methods.

Vision and Goals

The BAS focuses mostly on socioeconomic rather than explicitly political issues. Indeed, “politics” does not even appear in the full project title: Beijing Annual Survey of Social and Economic Development (北京社会经济发展年度调查). This is not simply because political topics are more sensitive than social or economic topics, with implications for survey implementation, although this is certainly a serious consideration for a project with a long projected life span as opposed to a one-shot effort. It also has much to do with the context of the early 1990s, when the vision of the BAS initially emerged. After 1992, with a renewed policy emphasis promoted by Deng Xiaoping on the role of the private sector in the economy, the pace of economic and accompanying social change in China increased rapidly. In the new historic period of “reform and opening,” the broad goal of the BAS designers was to capture, with a continually updated dataset, the impact of the major ongoing reforms on the everyday lives of ordinary Chinese.