Mapping Diversity of Publication Patterns in the Social Sciences and Humanities: An Approach Making Use of Fuzzy Cluster Analysis

Abstract Purpose To present a method for systematically mapping diversity of publication patterns at the author level in the social sciences and humanities in terms of publication type, publication language and co-authorship. Design/methodology/approach In a follow-up to the hard partitioning clustering by Verleysen and Weeren in 2016, we now propose the complementary use of fuzzy cluster analysis, making use of a membership coefficient to study gradual differences between publication styles among authors within a scholarly discipline. The analysis of the probability density function of the membership coefficient allows to assess the distribution of publication styles within and between disciplines. Findings As an illustration we analyze 1,828 productive authors affiliated in Flanders, Belgium. Whereas a hard partitioning previously identified two broad publication styles, an international one vs. a domestic one, fuzzy analysis now shows gradual differences among authors. Internal diversity also varies across disciplines and can be explained by researchers’ specialization and dissemination strategies. Research limitations The dataset used is limited to one country for the years 2000–2011; a cognitive classification of authors may yield a different result from the affiliation-based classification used here. Practical implications Our method is applicable to other bibliometric and research evaluation contexts, especially for the social sciences and humanities in non-Anglophone countries. Originality/value The method proposed is a novel application of cluster analysis to the field of bibliometrics. Applied to publication patterns at the author level in the social sciences and humanities, for the first time it systematically documents intra-disciplinary diversity.

[1]  Anton J. Nederhof,et al.  Bibliometric monitoring of research performance in the Social Sciences and the Humanities: A Review , 2006, Scientometrics.

[2]  Olle Persson,et al.  Bibliometric analysis of two subdomains in philosophy: free will and sorites , 2015, Scientometrics.

[3]  Wolfgang Härdle,et al.  Applied Multivariate Statistical Analysis: third edition , 2006 .

[4]  D. Hicks.  The Four Literatures of Social Science , 2004 .

[5]  Frederik T. Verleysen,et al.  Clustering by publication patterns of senior authors in the social sciences and humanities , 2016, J. Informetrics.

[6]  J. Moody.  The Structure of a Social Science Collaboration Network: Disciplinary Cohesion from 1963 to 1999 , 2004 .

[7]  Anton J. Nederhof.  A bibliometric study of productivity and impact of modern language and literature research , 2011 .

[8]  Hanna-Mari Puuska,et al.  Scholarly Publishing Patterns in Finland - A comparison of disciplinary groups , 2014 .

[9]  R. Whitley.  The Intellectual and Social Organization of the Sciences (Second Edition: with new introductory chapter entitled 'Science Transformed? The Changing Nature of Knowledge Production at the End of the Twentieth Century') , 1984 .

[10]  Pei-Shan Chi.  Changing publication and citation patterns in political science in Germany , 2015, Scientometrics.

[11]  Eric H. J. Spruyt,et al.  Changing publication patterns in the Social Sciences and Humanities, 2000–2009 , 2012, Scientometrics.

[12]  Yang Lin,et al.  Fragmentation of the Intellectual Structure of Political Communication Study: Some Empirical Evidence , 2004, Scientometrics.

[13]  Raf Vanderstraeten,et al.  Scientific Communication: Sociology Journals and Publication Practices , 2010 .

[14]  Svein Kyvik,et al.  Changing trends in publishing behaviour among university faculty, 1980-2000 , 2003, Scientometrics.

[15]  Ali S. Hadi,et al.  Finding Groups in Data: An Introduction to Chster Analysis , 1991 .