Comparing the Disciplinary Profiles of National and Regional Research Systems by Extensive and Intensive Measures

In this paper, by modeling national and regional research systems as complex systems, we compare the dynamics of their disciplinary profiles using extensive (size dependent) indicators as well as intensive (size independent) average productivity indicators of scientific production. Our preliminary findings show that the differences between the disciplinary profiles among countries in the world is of the same order of magnitude of the differences among European countries, that in turn, is of the same order of magnitude of the dynamics among regions within a country. While additional research (that is in progress) is needed to confirm these findings, we describe the main advantages (features) of our approach and outline its usefulness to support evidence-based policy making. Conference Topics Methods and techniques; Citation and co-citation analysis; Indicators; Science policy and research assessment; Country-level studies Introduction, scope and structure of this paper The dynamics of national or regional research systems is one of the most important topics in quantitative science and technology research. Interestingly, a lot of studies have analyzed the disciplinary specialization of countries (see e.g. Glanzel, 2000; Glanzel & Schlemmer, 2007; Glanzel et al., 2006, 2008; Hu & Rousseau, 2009; Tian et al., 2008; Wong, 2013; Wong et al., 2012; Yang et al., 2012; Horlings & Van den Besselaar, 2013; Radosevic & Yoruk, 2014) or have investigated the disciplinary specialization of regions within a particular country, or have conducted case studies on individual regions and/or on a few number of selected disciplines (see e.g. Zhu et al., 2009; Glanzel, Tang & Shapira, 2011). Much less studied are the disciplinary profiles of European countries at the regional level. To the best of our knowledge there are not empirical analyses at European level, investigating the evolution of the disciplinary composition (i.e. the 27 Scopus Subject categories) of regions. Moreover, none of the existing studies have analyzed in a comparative way, the range of variability (briefly: the dynamics) of national and regional research systems which is the aim of our paper. We investigate here this dynamics in terms of both extensive measures of scientific production (i.e. total number of scientific publications, citations and so on) and in terms of intensive average scientific productivity (i.e. number of publications per author). In particular, the investigation of the dynamics of intensive measures of scientific production has an important policy relevance. According to the macroeconomic theory, we have growth convergence when smaller (poorer) countries, in terms of output per capita (e.g. GDP per capita), grow faster than larger (richer) countries. In the context of research systems, we can say that there is a convergence if smaller scientific systems, in terms of scientific output per capita, grow faster than larger one. This is an important question, related to the policy