Quantifying Structural Transformation in China
The Quantifying Structural Transformation project led by researchers David Meyer and Victor Shih aims to develop more precise measures of changes in social networks and apply them to the larger networks within which the Chinese leadership operates.
The project builds on data collection from IGCC’s Study of Innovation and Technology in China project.
The research team completed two original quantitative data sets, one of all alternate and full Central Committee members up to the 18th Party Congress and one of all provincial standing committee members from 1978 to 2014. These data sets allow researchers to conduct a range of quantitative network analysis on more than 6,000 Chinese officials.
The project was supported by the Army Research Office (ARO) under Award No. W911NF-12-1-0389 through the US Department of Defense's Minerva Research Initiative. Any opinions, findings, and conclusions or recommendations are those of the project participants and do not necessarily reflect views of ARO.
The project has three main goals:
1. Develop New Measures of Change
Leadership changes in the Chinese Communist Party (CCP) have traditionally been quantified rather coarsely, by counting the numbers of new Politburo, or Central Committee (CC), members. But Politburo and CC (alternate) members are ranked, so that more refined measures of change are possible.
2. Apply Measures of Change to Political/Social Networks and Faction Within Them
Maneuvering for rank within the CCP, as in any political system, is largely a behind-the-scenes process. The fraction of CC members within a particular subset of an individual’s social network, his “faction,” however, turns out to be closely correlated with his political fortunes, as measured by his rank within the CCP. The ties within the factions are hard to tease out and must be inferred indirectly in many cases.
Members of a particular faction are also connected by “clientelist” ties, or, more generally, guanxi (关系), social relations of mutual obligation. In the Chinese context, the relevant ties are believed to be a consequence of having been born in the same province, having been educated in the same school, or having common experiences during the revolution or in a work unit.
The full network (guanxi wang, 关系网) has (at least) four modes—individuals, locations, schools, and work units—with individuals linked to each of the others. The inferred network collapses this richer set of relations to a simple set of dyadic factional ties between individuals. The proposed new methods take into account the multi-modal nature of these networks.
3. Refine Measures of Change for Thick Networks
Throughout the project, researchers have been recording individuals and their roles/jobs in a relational database. They have also been recording information about entities other than individuals (for example, research institutes, corporations, and government funding programs) together with a multitude of relationships among them. These relationships are recorded with starting and ending times, creating a dynamic network describing China’s national innovation system (NIS). This collection of entities of multiple types with multiple kinds of (polyadic) relations among them is known as a “thick network.”Developing more measures of relevant changes in these thick networks and applying them to the larger networks within which the Chinese leadership operates, focusing initially on the subnetworks in China’s NIS, will allow analysis of the likely rate of innovation and development within various science and technology sectors in China.
Quantitative Studies of the Chinese Elite
February 6–7, 2015
UC San Diego
Agenda for public session, Day One [pdf]