Exploration of social science´s compactness by cluster analysis

In this paper, we proposed to explore the compactness of social science by cluster analysis. In terms of the aggregated journal-journal citation reports of the Institute of Scientific Information (ISI), 1578 social science journals are clustered by an emerging hierarchical clustering method with single-link, complete-link and average-link as cluster-cluster similarities. The statistical properties of the clustering results, detailed comparisons of the clustering results and graph-theoretic interpretations show that social science has little interdisciplinary property, which demonstrates the intra-domain compactness of social science.