Top-down decomposition of the Journal Citation Report of the Social Science Citation Index: Graph- and factor-analytical approaches

摘要

The aggregated journal-journal citation matrix of the Journal Citation Report 2001 of the Social Science Citation Index is analyzed as a single domain in terms of both its eigenvectors and the bi-connected components contained in it. The traditional disciplines (e.g., economics, psychology, or political science) can be retrieved using both methods. These main disciplines do interact marginally. The space between them is occupied by a large number of small clusters of journals indicating specialties that gravitate among the major disciplines. These specialties operate in a mode different from that of the disciplines. For example, the impact factors are low on average and the developments remain volatile. Factor analysis enables us to study how the smaller bi-connected components are related to the larger ones. Factor analysis also highlights methodological differences among groups which may be theoretically connected in a single bi-component.