Clustering of scientific journals

A cluster analysis procedure is described in which 288 journals in the disciplines of physics, chemistry and molecular biology are grouped into clusters. Most of the clusters are easily identified as subdisciplinary subject areas. The data source was the cross citing amongst the journals derived from the Journal Citation Index (JCI), a file derived in turn from the Science Citation Index (SCI)®. The JCI consists of journal by journal tabulation of citings to and from each journal processed in the SCI. Two‐step citation maps linking the clusters are presented for each discipline. Within the disciplines the clusters of journals form fully transitive hierarchies with very few relational conflicts.