SPSS: K-means analysis. What criteria can I use to state my choice of the number of final clusters I choose. Using a hierarchical cluster analysis, I started with 2 clusters in my K-mean analysis. However, after running many other k-means with different number of clusters, I dont knwo how to choose which one is better. Is there a general method of choosing the number of clusters that is scientifically right.
1 Answers
0
votes
Are you using SPSS Modeler or SPSS Statistics? Because I created an extension to determine the optimal number of clusters.
It is based on R: Cluster analysis in R: determine the optimal number of clusters
