cor(spe.ch, spe.ch.ward.coph, method = "spearman")
To illustrate the relationship between a dissimilarity matrix and a set of
cophenetic matrices obtained from various methods, one can draw Shepard-like
diagrams (Legendre and Legendre 2012 p. 414) by plotting the original dissimilarities against the cophenetic distances (Fig. 4.7):
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
0.0 0.2 0.4
0.6 0.8 1.0 1.2 1.4
Single linkage
Cophenetic correlation = 0.599
Chord distance
Cophenetic distance
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
0.0 0.2
0.4
0.6 0.8
1.0 1.2
1.4
Complete linkage
Cophenetic correlation = 0.766
Chord distance
Cophenetic distance
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
UPGMA
Cophenetic correlation = 0.861
Chord distance
Cophenetic distance
-1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
0.0 0.5 1.0 1.5
2.0
2.5 3.0 3.5
Ward
Cophenetic correlation = 0.8
Chord distance
Cophenetic distance
Fig. 4.7 Shepard-like diagrams comparing chord distances (species data) to four cophenetic
distances. A LOWESS smoother shows the trend in each plot
72
4 Cluster Analysis