A statistical approach to infer causal effects of the Chinese publication growth on bibliometric impact measures

Leiden Repository

A statistical approach to infer causal effects of the Chinese publication growth on bibliometric impact measures

Type: Article in monograph or in proceedings
Title: A statistical approach to infer causal effects of the Chinese publication growth on bibliometric impact measures
Author: Stahlschmidt S.Hinze S.
Journal Title: STI 2018 Conference Proceedings
Start Page: 553
End Page: 561
Publisher: Centre for Science and Technology Studies (CWTS)
Issue Date: 2018-09-11
Keywords: Scientometrics
Abstract: The unprecedented growth of Chinese publications, understood as a macro level intervention on the global bibliometric universe, is analysed on its causal effects on the bibliometric measurement process. By their size and particular bibliometric characteristics these publications might not be considered as marginal, but impose a structural change upon all countries contributing to the global science system. A causal analysis is conducted based upon the statistical approach of treatment effects models. In detail a counterfactual bibliometric world without Chinese publications is artificially constructed and contrasted with the actual equivalent. We observe that the Chinese publications cause an increase in the expected citation counts as they included above average numbers of references. At the same time the citations of Chinese publications are especially focused upon scientifically leading, as well as Asian countries, resulting in an increase in their measured impact. But as any identification of a causal effect via statistical means arises as an interpretation, we subsequently discuss the limitation of our approach namely the imperfect knowledge on the counterfactual state and any implications arising for further causal evaluations, i.e. the Chinese publication growth illustrates only one of several structural effects concurrently driving the evolution of the bibliometric universe.
Handle: http://hdl.handle.net/1887/65244
 

Files in this item

Description Size View
application/pdf STI2018_paper_104.pdf 1.392Mb View/Open

This item appears in the following Collection(s)