Clustering representations of group overlap

A distinction is made between implicit and explicit group overlap in sociological data, and literature is briefly reviewed in terms of this distinction. The conclusion is drawn that for implicit overlap, the method of data analysis should use continuous input, while yielding output in a discrete form of (possibly overlapping) subsets. Such a method of clustering (ADCLUS) is presented briefly and is applied to the communication structure of a biomedical area of specialization. Notes I am indebted to Scott A. Boorman for his detailed and helpful suggestions and criticism. I have also benefited from discussions of this work with Paul Levitt, Ronald L. Breiger, and Harrison C. White, and from technical assistance supplied by Dan C. Knutson. This research was supported by NSP Grant GS‐2689 (Co‐principal Investigators: Harrison C. White and Scott A. Boorman) and funds from the Graduate School and the Computing Facility of the University of Minnesota. Current address: Department of Psychology, Elliott Hall, University of Minnesota, Minneapolis, MN 55455.