How to evaluate the degree of interdisciplinarity of an institution?
- 710 Downloads
The Stirling index of the set of references of the corpus documents is widely used in the literature on interdisciplinary research and is defined as the integration score of the corpus under study. Such an indicator is relevant at the scale of a research institution, however, there is a gap between the integration scores of individual documents, and a global score computed on the whole set of references. The difference between the global index and the average of individual document indexes carries another relevant information about the corpus: it measures the diversity between the reference profiles of the corpus documents. It is, therefore, named between article index whereas the average of the individual article indexes is called within article index. The statistical properties of these two indexes as well as of the global index are derived from a general approximation method for distributions and lead to statistical tests which can be used to make meaningful comparisons between an institution indexes and benchmark values. The two dimensions of the global index provide a more acute information on the interdisciplinary practices of an institution researchers in a given research domain and is, therefore, likely to contribute to strategic and management issues.
KeywordsInterdisciplinary research Indicators Stirling index decomposition Inertia of a set of weighted points Asymptotic distribution of indicators Statistical tests
Mathematics Subject Classification60F05
The authors are grateful to Ismael Rafols and Tommaso Ciarli for a fruitful discussion on an earlier version of this paper. The issues presented in the “Discussion” section largely result of this interaction. In particular, I. Rafols made us aware of the relevance of analysing interdisciplinarity at the level of a department rather than at the university level. This issue will be considered in future work with different institutions.
- Adams, J., Jackson, L., & Marshall, S. (2007). Bibliometric analysis of interdisciplinary research. Report to the Higher Education Funding Council for England. Leeds: Evidence.Google Scholar
- Garner, J., Porter, A. L., Borrego, M., Tran, E., & Teutonico, R. (2013). Facilitating social and natural science cross-disciplinarity: Assessing the human and social dynamics program. Research Evaluation, 22(2), 134–144.Google Scholar
- Geyer, C. (2001). Stat 5102 Lecture notes. University of Minnesota. Retrieved August 21, 2013, from http://www.stat.umn.edu/geyer/5101/notes/n2.pdf.
- Omey, E., & Van Gulck, S. (2008). Central limit theorems for variances and correlation coefficients. HUB research paper. HUB, Brussels. Retrieved August 14, 2013, from http://www.edwardomey.com/nonsave/CLTforsandr.pdf.