Mapping the Chinese Science Citation Database

Loet Leydesdorff University of Amsterdam, Science & Technology Dynamics, Amsterdam School of Communications Research (ASCoR), Kloveniersburgwal 48, 1012 CX Amsterdam, The Netherlands. Email: loet@leydesdorff.net

Jin Bihui Library of the Chinese Academy of Sciences, 33 Beisihuan Xilu, Zhongguancun, Beijing 100080, P. R. China. Email: jinbh@mail.las.ac.cn .

Methods developed for mapping the journal structures contained in aggregated journal- journal citations in the Science Citation Index are applied to the Chinese Science Citation Database of the Chinese Academy of Sciences. This database covers 991 journals, of which only 37 had originally English titles. Using factor- analytical and graph-analytical techniques we show that this data is dually structured. The main structure is the intellectual organization of the journals in journal groups (as in the international SCI), but the university-based journals provide an institutional layer that orients this structure towards practical ends (e.g., agriculture). The Chinese Science Citation Database exhibits the characteristics of “Mode 2” in the production of scientific knowledge more than its western counterparts. The contexts of application lead to correlation (interfactorial complexity) among the components.

Introduction Aggregated journal-journal citation relations have been organized in the Journal Citation Reports of the Science Citation Index on a yearly basis since 1975. As early as 1965, on the basis of an experimental version of this database, Derek de Solla Price noted the pronounced structure of this matrix and he suggested that it would be possible to decompose it for the mapping of scientific specialties and disciplines (Price, 1965; cf. Simon, 1969). The time series allows us additionally to study the dynamics of the sciences (Leydesdorff, 2002).
At the level of journal clusters, the dynamics can be considered as a baseline that is relatively independent of intentional (social or political) agency (Studer & Chubin, 1982; Zsindely et al., 1982; Leydesdorff, 1987). In national contexts, however, one can wonder whether a more direct couplings between institutional interests and journal structures might exist. Perhaps national elites provide an intermediating, but invisible college (Crane, 1972; Mulkay, 1976). For example, the French CNRS subsidizes approximately 225 journals which are only partially covered by the U.S.-based Science Citation Index (De Looze et al., 1996; Legentil, personal communication). However, Sivertsen (2003) found no bias in the coverage of Scandinavian publications by the Science Citation Index (Garfield, 1979, 1990). The Institute of Scientific Information (ISI) admits that the selection system for inclusion in the Science Citation Index may be biased against journals written in non-latin alphabets. Special care has been taken in the past to include Russian and Japanese journals into the database (Garfield, 1979, 1998). Chinese scientific journals, however, have not been systematically evaluated for their inclusion in the ISI database. The expanded version of the SCI at the Internet included only 31 Chinese scientific journals in 2001. Jin et al. (2002) estimated this as 0.73% of the scientific journal titles available in China. The Library of the Chinese Academy of Sciences has developed the Chinese Science Citation Database (CSCD) since 1989 (Jin & Wang, 1999; cf. Liang & Wu, 2001). In this project, the data for the year 2001 was aggregated in a format similar to the Journal Citation Reports of the SCI. We apply graph-analytical and factor- analytical techniques that were developed in previous projects for analyzing the international Science Citation Index to the Chinese dataset (Leydesdorff & Cozzens, 1993; Leydesdorff, 2004a).
Methods and Materials Data We have examined aggregated journal-journal citation data for 991 Chinese journals. Only 37 of these journals have titles originally in English. The other 954 journal titles were translated into English for the purpose of this project. Among these journals 55,774 citation relations are maintained, that is, 5.7% of the 982,081 (= 9912) possible relations. The corresponding figure was 2.6% for the Science Citation Index and 2.8% for the Social Science Citation Index in the year 2001. However, these indices exclude most of the single citation relations (using a threshold). Since the single citation relations amount in the CSCD to 28,454, the Chinese figure corrected for this comparison would be 2.8%. We pursue the analysis below, including these single occurrences in the Chinese citation matrix. Methods The matrix of 991 x 991 cells was first constructed and then saved as an SPSS systems file. This file can, for example, be factor analyzed. We will use both the overall factor analysis and the routines which we previous

…(Full text truncated)…