Andras Schubert

Institute for Research Policy Studies, Hungarian Academy of Science, Budapest, Hungary

Scientometrics in its broadest sense covers all quantitative aspects of the study of scientific and scholarly activity. Its primary interest is in the structural and dynamical analysis of sets of authors, publications, citations, etc. The notorious evaluative aspects are just one of the possible fields of application.

Since scientific and scholarly research are unquestionably social activities, their study should be categorized into the social sciences, even if some of its concepts and methods are borrowed from the harder sciences. On the other hand, social sciences, as a form of scientific and scholarly activity, is a legitim target of scientometric studies. As a result, scientometrics offers a unique possibility for self-reflection: the quantitative way for social science to study itself.

Social sciences, as compared to most areas of sciences, has certain handicaps as targets for scientometric studies. Just to mention a few: because of a relatively high percentage of single-authored works, co-authorship networks are less definite; language and cultural barriers inhibit more markedly the formation and use of a common knowledge pool; the number and even the function of references vary rather widely among the different disciplines, etc. These features make the scientometrics of social sciences challenged but by no means disabled. Not only one of the earliest pioneering attempts of sceintometrics concerned with a social science discipline (psychology), but both structural and dynamical, as well as evaluative scientometrics of social sciences are amply represented in the most recent literature.

In this presentation some typical examples of scientometrics analysis and evaluation techniques will be shown on the model of the papers published in a set of social science journals. Analysis of author networks, publication and citation dynamics, as well as macro-, meso- and micro-level evaluative techniques will be illustrated, and some comparisons with earlier results and with similar characterics of science fields will be made.