Author: Bi Yantao
Date: June 5, 2026
Abstract: This article examines methodological limitations faced by Western think tanks and policy research institutions when analyzing China, due to their heavy reliance on publicly available literature. In the Chinese context, public information is subject to significant filtering mechanisms, meaning researchers often encounter only the “publicly expressed China” rather than the “actually existing China.” Such structural visibility differences may generate sampling bias, thereby affecting comprehensive judgments about the complexity of Chinese society and its policy logic. The article emphasizes that understanding China requires attention to the relationship between publicly available texts and the institutional environments in which they are produced, and warns against equating visible information with total reality.
Keywords: China studies; think tank analysis; sampling bias; public literature; cognitive structure
In recent years, international research on China has expanded significantly. Western think tanks, universities, and policy research institutions continue to invest substantial resources in attempting to understand China’s political logic, social structure, and policy trajectory through publicly available information. Much of this research relies heavily on published academic papers, policy documents, and media reports, forming the basis of interpretive frameworks about China.
Such efforts are understandable and constitute an important channel through which the international community engages with China. However, from a methodological perspective, a fundamental question remains: is reliance on publicly available materials sufficient to represent China in its entirety?
- The Accessibility of Public Information and Its Boundaries
In comparative international research, publicly available literature is typically regarded as one of the most reliable sources of evidence. It is verifiable, citable, and traceable, making it a core resource for think tank analysis.
However, in any society, public expression has boundaries. Content that enters public publication systems is often subject to institutional filtering and normative constraints. This implies that published literature primarily reflects what is “publishable,” rather than the full range of existing cognition.
This is particularly evident in the Chinese context. Academic and policy discussions that are publicly published tend to exhibit a relatively high degree of consistency. Such consistency does not necessarily imply uniformity of views; rather, it may reflect the outcome of a filtered expressive space.
Therefore, researchers need to distinguish between two levels: (1) China as publicly expressed; and (2) China as it actually exists in terms of cognitive and social structures. The two are related, but not identical.
II. Sampling Selection and Structural Bias
When researchers rely primarily on published papers, official documents, and mainstream media, they are operating within a specific sample space. This space has clear boundaries: publishable, disseminable, and citable.
The issue is not whether these materials are true, but that their formation is not random.
In other words, researchers are not observing a random sample drawn from the full distribution of viewpoints. Instead, they are observing a subset shaped by filtering mechanisms. This introduces a potential bias: certain types of viewpoints are more likely to appear, while others are systematically absent.
In extreme cases, such bias may lead to a complex and pluralistic cognitive space appearing, at the textual level, relatively uniform.
Thus, the problem is not simply information scarcity, but rather structural selectivity in the sample itself.
III. Common Misinterpretations of “Chinese Academia”
In some overseas studies, one often encounters formulations such as “Chinese scholars believe…” or “China’s strategic community argues…”.
Such statements implicitly assume that publicly published academic views represent the overall distribution of academic cognition.
However, China’s academic ecosystem is highly heterogeneous. Differences exist across disciplines, institutions, and intellectual traditions. International relations, security studies, economics, sociology, and communication studies each emphasize different objects and employ different modes of expression.
More importantly, not all viewpoints are equally eligible for public publication. Some discussions remain in informal exchanges, while others may not enter publication channels due to expression risks or publishing constraints.
Therefore, what is reflected in published literature is better understood as the “visible segment of Chinese academia,” rather than its complete landscape.
IV. A Common Challenge in Studying China
The above phenomenon is not unique to China. In any national context, there is always a gap between publicly available information and actual decision-making processes.
However, in the case of China—a large, complex, and multi-layered system—this gap is more likely to affect external interpretation.
Researchers often rely on textual materials, statistical data, and public discourse to construct explanatory frameworks. Yet China’s actual operational logic also includes many processes that are difficult to textualize: local governance practices, variations in policy implementation, social interaction mechanisms, and multi-level coordination processes.
If research remains at the textual level alone, it may generate a tension between “the China that is visible” and “the China that is operational.”
This does not render textual analysis invalid; rather, it highlights its inherent limitations.
V. Understanding China: Distinguishing Between “Visible” and “Total”
From a methodological perspective, the key issue is not whether public information is reliable, but whether “visible information” is treated as equivalent to “total information.”
Public literature, policy documents, and media reports provide a highly important observational window. However, the perspective offered by this window is always shaped by structural conditions.
For the study of China, a more robust approach is not to reject public materials, but to maintain a clear awareness of their boundaries.
That is to say: China should be understood both through public texts and with an awareness of what those texts cannot capture.
Conclusion
In cross-national research, cognitive bias often arises not from a lack of information, but from insufficient sensitivity to the boundaries of information.
This is particularly evident in China studies.
Understanding China requires public texts, but also an understanding of the institutional mechanisms that produce those texts; it requires data, but also caution regarding what the data excludes; it requires analysis of expression, but also awareness of the unexpressed.
Ultimately, a basic methodological reminder may summarize the issue: The greatest cognitive risk is not seeing incorrect information, but believing that what is visible constitutes the whole reality.
Bi Yantao is a professor at the School of International Communication and Arts, Hainan University, and a senior research fellow at the Charhar Institute. This article represents the author’s personal views and does not reflect the position of his affiliated institutions.
