K-Means Clustering

Performs k-means clustering to determine a user-defined number of clusters; both variables (columns of the data matrix) and cases (rows of the data matrix) can be clustered. Final assignment of objects (cases, variables) to clusters are reported, and can be saved for further analyses.

Element Name Description
General
Detail of computed results reported Detail of computed results; if Minimal detail is requested, only summary statistics are computed; if All results is requested, spreadsheets showing the members of each cluster are also reported.
Cluster cases or variables Specifies whether to cluster the cases or rows of the input data based on the variables or columns, or the variables or columns of the input data based on the cases or rows.
Number of clusters Specifies the number of clusters for the final k-means clustering solution.
Initial cluster centers Specifies how to determine the initial cluster centers.
Number of iterations Maximum number of iterations for k-means clustering algorithm.
Missing data deletion Specifies how to treat missing values (observations); missing data can be deleted casewise or substituted by means.