k-means clustering

k-means clustering

Main options input clustering

Load file

File must be a SPSS Statistics (*.sav) file, first column must be observation ID, remaining columns are variables to cluster on

Choose SPSS File

Browse...

Cluster solution input

Select the variables to cluster on (default: all variables except first column are selected)

Number of clusters in cluster solution (default: 3)

Run cross validation to help decide on number of clusters (note: may take a while to run)

Developed by Peter Ebbes, HEC Paris, for classroom purpose (version: 2.0).