The Scientometric Measurement of Interdisciplinarity and Diversity in the Research Portfolios of Chinese Universities
Lin Zhang,Loet Leydesdorff
DOI: https://doi.org/10.2478/jdis-2021-0027
2021-01-01
SSRN Electronic Journal
Abstract:Purpose Interdisciplinarity is a hot topic in science and technology policy. However, the concept of interdisciplinarity is both abstract and complex, and therefore difficult to measure using a single indicator. A variety of metrics for measuring the diversity and interdisciplinarity of articles, journals, and fields have been proposed in the literature. In this article, we ask whether institutions can be ranked in terms of their (inter-)disciplinary diversity. Design/methodology/approach We developed a software application (interd_vb.exe) that outputs the values of relevant diversity indicators for any document set or network structure. The software is made available, free to the public, online. The indicators it considers include the advanced diversity indicators Rao-Stirling ( RS ) diversity and DIV* , as well as standard measures of diversity, such as the Gini coefficient, Shannon entropy, and the Simpson Index. As an empirical demonstration of how the application works, we compared the research portfolios of 42 “Double First-Class” Chinese universities across Web of Science Subject Categories (WCs). Findings The empirical results suggest that DIV* provides results that are more in line with one's intuitive impressions than RS , particularly when the results are based on sample-dependent disparity measures. Furthermore, the scores for diversity are more consistent when based on a global disparity matrix than on a local map. Research limitations “Interdisciplinarity” can be operationalized as bibliographic coupling among (sets of) documents with references to disciplines. At the institutional level, however, diversity may also indicate comprehensiveness. Unlike impact (e.g. citation), diversity and interdisciplinarity are context-specific and therefore provide a second dimension to the evaluation. Policy or practical implications Operationalization and quantification make it necessary for analysts to make their choices and options clear. Although the equations used to calculate diversity are often mathematically transparent, the specification in terms of computer code helps the analyst to further precision in decisions. Although diversity is not necessarily a goal of universities, a high diversity score may inform potential policies concerning interdisciplinarity at the university level. Originality/value This article introduces a non-commercial online application to the public domain that allows researchers and policy analysts to measure “diversity” and “interdisciplinarity” using the various indicators as encompassing as possible for any document set or network structure (e.g. a network of co-authors). Insofar as we know, such a professional computing tool for evaluating data sets using diversity indicators has not yet been made available online.