Agglomerative cluster analysis
|
|
Top Previous Next |
|
Hierarchical agglomerative cluster analysis is selected by choosing Clustering: Agglomerative from the drop-down menu. The following methods for linking groups are available:
Whichever method you select, in the Setup for Cluster box, CAP then offers a choice of 15 distance measures: Euclidean, Geodesic, Whittakers, Ave. Distance, Manhattan, Canberra, Chord, Mean Character Difference (Czekanowski), Bray-Curtis, Squared Chord, Mahalanobis, 1-Jaccard, 1-Sørensen and Square root 1-Sørensen and Renkonen distance. Each of these distance measures is different, and will influence the outcome of cluster analysis. If you have presence-absence data, it is worthwhile choosing a distance measure specifically designed for it, such as Jaccard's or Sorenson's.
Once a measure has been selected the cluster analysis will immediately be run. Output is presented on a series of tabbed pages. These are described in turn below.
|
