Clustering Seven Data Sets by Means of Some or All of Seven Clustering Methods

The semantics of space: experiential and linguistic aspects of selected English spatial terms

To investigate relationships between spatial terms and the human experience of space, participants in three separate experiments were asked to sort 35 English spatial terms into groups on the basis of which items seemed most similar to one another. Participants also were asked to provide a descriptive rationale for each of their groupings. Once all groupings were recorded, they were subject to hierarchical clustering analysis and results of all three studies produced five different clusters centering around issues of (1) barrier and passage, (2) distance, (3) embodiment, (4) boundary and containment, and (5) verticality. Meanings for each cluster were named on the basis of participant descriptions and from independent evaluations made by the research team. Using these thematic meanings as a guide, a first-person description of spatial experience was developed and related to current analyses of the Whorf Hypothesis concerning issues of spatial experience and memory.

Properties of the Hubert-Arabie adjusted Rand index

This article provides an investigation of cluster validation indices that relates 4 of the indices to the L. Hubert and P. Arabie (1985) adjusted Rand index--the cluster validation measure of choice (G. W. Milligan & M. C. Cooper, 1986). It is shown how these other indices can be "roughly" transformed into the same scale as the adjusted Rand index. Furthermore, in-depth explanations are given of why classification rates should not be used in cluster validation research. The article concludes by summarizing several properties of the adjusted Rand index across many conditions and provides a method for testing the significance of observed adjusted Rand indices.

Differentielle Profile der Belastungsbewältigung und Wohlbefinden

Zusammenfassung: In einer Stichprobe erwerbstätiger Personen (n = 137) wurden differentielle Profile der Belastungsverarbeitung und deren Zusammenhänge mit Indikatoren des psychischen und physischen (Wohl-)Befindens untersucht. Als Erhebungsinstrumente wurden eine Kurzform des Streßverarbeitungsfragebogen (SVF, Janke, Erdmann & Kallus, 1984 ), die deutschsprachige Version des Profile of Mood States ( Biehl, Dangel & Reiser, 1986 ) sowie Ratingskalen zur Einschätzung von Belastungen und Bewältigungskompetenzen eingesetzt. Die Coping-Skalen wurden einer hierarchischen Clusteranalyse nach Ward mit anschließender iterativer Partionierung nach dem “k-means”-Algorithmus unterzogen. Es resultierte eine Aufteilung in drei Cluster: (1) “aktiv-flexibles Coping” (n = 53), (2) “rigid-instrumentelles Coping” (n = 40) und (3) “resigniert-distanziertes Coping” (n = 44). Die Reliabilität dieser Clusterlösung konnte durch mehrere Tests auf innere Konsistenz sowie auch bei Wiederholungsmessung nach drei Monaten bestätigt werden. Ferner erwies sich die Clusterzugehörigkeit als unabhängig sowohl von soziodemographischen Basisvariablen als auch von der Häufigkeit perzipierter beruflicher und privater Belastungen. Multivariate Varianzanalysen zeigten Unterschiede im Befinden zwischen den drei Clustern unter der Bedingung hoher subjektiver Belastung. Forschungsarbeiten zur Effektivität von Bewältigung sollten zukünftig anstelle von einzelnen Bewältigungsstrategien verstärkt differentielle Profile der Bewältigung evaluieren.

The Effects of Within-group Covariance Structure on Recovery in Cluster Analysis: I. The Bivariate Case

In a preliminary investigation, two Monte Carlo studies investigated the effects of within-group covariance structure on subgroup recovery by I0 hierarchical clustering methods. Data sets were 100 bivariate observations from two subgroups. Study 1 manipulated subgroup size, within-group correlation, within-group variance, and distance between centroids. Negative within-group correlation yielded much poorer recovery for all clustering methods. In addition, clustering method interacted with within-group variance. Study 2 manipulated subgroup size, within-group correlation, direction of the vector separating subgroup centroids, and distance between subgroup centroids. Superior recovery was associated with within-group correlation that matched the direction of subgroup separation. Results are interpreted according to the weakness of Euclidean distance as a measure of (dis)similarity.