Goodness-of-Fit Measures for Two-mode Cluster Analyses

Unterreitmeier, Andreas und Schwaiger, Manfred (2002): Goodness-of-Fit Measures for Two-mode Cluster Analyses. 24th Annual Conference of the Gesellschaft für Klassifikation e.V., University of Passau, March 15—17, 2000. Gaul, Wolfgang und Ritter, Gunter (Hrsg.): In: Classification, Automation, and New Media. Proceedings of the 24th Annual Conference of the Gesellschaft für Klassifikation e.V., Berlin; Heidelberg: Springer. S. 401-408

Abstract

Two-mode cluster analyses take pleasure in increasing distribution not only in psychological but also in management applications, for example in the controlling of advertising effects (Schwaiger 1997a). Until today, only a few index numbers exist that are able to measure the goodness-of-fit of a two-mode classification, like VAF, TIC and CCC.

This paper is meant to consider how heterogeneity within a two-mode cluster can be quantified by referring to intramode similarity relations. An approach by (1997a, p. 122 f.) calculates the square deviations between an original matrix and matrices of cluster-centroids that originate by relating elements of one mode (combined in a two-mode cluster) in each case to the vector of the mode-specific class-centroid. In the strict sense, this index number can only be used to compare two classifications with an identical number of clusters. The subject of the discussion will be a suggestion that extends this procedure. Moreover, it is able to compare partitions with different numbers of clusters.

• Dokumententyp: Konferenzbeitrag (Paper)
• Fakultät: Betriebswirtschaft > Institut für Marktorientierte Unternehmensführung
• Themengebiete: 300 Sozialwissenschaften > 330 Wirtschaft
• Ort: Berlin; Heidelberg
• Sprache: Englisch