Wolfgang Glänzel

A bibliometric study of reference literature in the sciences and social sciences

Abstract In earlier papers the authors focused on differences in the ageing of journal literature in science and the social sciences. It was shown that for several fields and topics bibliometric standard indicators based on journal articles need to be modified in order to provide valid results. In fields where monographs, books or reports are important means of scientific information, standard models of scientific communication are not reflected by journal literature alone. To identify fields where the role of non-serial literature is considerable or critical in terms of bibliometric standard methods, the totality of the bibliographic citations indexed in the 1993 annual cumulation of the SCI and SSCI databases, have been processed. The analysis is based on three indicators, the percentage of references to serials , the mean references age , and the mean reference rate . Applications of these measures at different levels of aggregation (i.e., to journals in selected science and social science fields) lead to the following conclusions. 1. The percentage of references to serials proved to be a sensitive measure to characterise typical differences in the communication behaviour between the sciences and the social sciences. 2. However, there is an overlap zone which includes fields like mathematics, technology oriented science, and some social science areas. 3. In certain social sciences part of the information seems even to be originated in non-scientific sources: references to non-serials do not always represent monographs, pre-prints or reports. Consequently, the model of information transfer from scientific literature to scientific (journal) literature assumed by standard bibliometrics requires substantial revision before valid results can be expected through its application to social science areas.

A bibliometric study on ageing and reception processes of scientific literature

In an exploratory study, the time behaviour of citations to articles of seven journals representing different scientific fields (sociology, psychology, chemistry, general and inter nal medicine, statistics and probability theory) were analysed to establish: (i) differences in ageing and reception speed between social sciences and other science fields, to determine (ii) if there are connections between ageing and reception, and (iii) if deviations are due to fields or individ ual journals. Bibliometric methods and citation-based indi cators were used within a stochastic model. It was found that obsolescence of the social science journals in the set is slower than for the medical and chemistry journals. The behaviour of the mathematical journal is similar to the ones in social sciences. The study suggests that ageing seems to be specific to the field rather than to the individual journal. On the other hand, slow ageing does not necessarily corre spond with slow response. Impact factors based on the usual two years observation period may therefore be dis torted by deviating ageing behaviour.

A NEW METHODOLOGICAL APPROACH TO BIBLIOGRAPHIC COUPLING AND ITS APPLICATION TO THE NATIONAL, REGIONAL AND INSTITUTIONAL LEVEL

In an earlier study the authors have shown that bibliographic coupling techniques can be used to identify ‘hot’ research topics. The methodology is based on appropriate thresholds for both number of related documentsand the strength of bibliographic links. Those papers are calledcore documents that have more than 9 links of at least the strength 0.25 according toSaltons measure, provided they are articles, notes or reviews. This choice resulted in a selection of nearly one per cent of all papers of the above types recorded in the 1992 annual cumulation of the SCI.Core documents proved important nodes in the network of documented science communication.In the present study, the set ofcore documents is analysed by journals, subfields and corporate addresses. The latter analysis is conducted on both national and regional-institutional level. First all countries which have published at least 20 core documents in 1992 are investigated in terms of their research profiles, their international collaboration patterns and their citation impact. Finally, those eight members of the European Union which have published at least 20 core documents in 1992 are analysed in respect of regional and institutional distribution of core documents.

An item-by-item subject classification of papers published in journals covered by the SSCI database using reference analysis

A serious shortcoming of bibliometric studies based on theSocial Sciences Citation Index is the lack of a universally applicable subject classification scheme as individual papers are concerned. Moreover, the selective coverage of more than thousand scientific journals per annum proved to be an insuperable obstacle in the delimitation of social science subject areas. Subject classification of papers on the basis of assigning journals to subject categories (like those found in the various supplements of ISI databases) works well in case of fully covered and highly specialised journals in the social sciences, too, but fails for multidisciplinary and selectively covered journals. This study presents the results of an item-by-item subject classification approach, where assignment is based on the analysis of the subject categories of reference literature.This analysis extends the results of an earlier study by the authors on the possibility of delimiting subfields in the hard and life sciences based on reference analysis. The assignment proved also reliable for a considerable share of literature in the social sciences. Due to the peculiarities of the database this share is lower in the SSCI than that in the SCI. Although an iterated application of the procedure is expected to increase the number of classifiable publications, it is suggested that in the sociated sciences the method should be used in combination with other means of subject assignment.