PRINT Subcommand (QUICK CLUSTER command)

QUICK CLUSTER always displays in a Final Cluster Centers table listing the centers used to classify cases and the mean values of the cases in each cluster and a Number of Cases in Each Cluster table listing the number of weighted (if weighting is on) and unweighted cases in each cluster. Use PRINT to request other types of output.

  • If PRINT is not specified or is specified without keywords, the default is INITIAL.

INITIAL. Initial cluster centers. When SPLIT FILES is in effect, the initial cluster center for each split file is displayed. This is the default.

CLUSTER. Cluster membership. Each case displays an identifying number or value, the number of the cluster to which it was assigned, and its distance from the center of that cluster. This output is extensive when the number of cases is large.

ID(varname). Case identification. The value of the specified variable is used in addition to the case numbers to identify cases in output. Case numbers may not be sequential if cases have been selected.

DISTANCE. Pairwise distances between all final cluster centers. This output can consume a great deal of processing time when the number of clusters requested is large.

ANOVA. Descriptive univariate F tests for the clustering variables. Since cases are systematically assigned to clusters to maximize differences on the clustering variables, these tests are descriptive only and should not be used to test the null hypothesis that there are no differences between clusters. Statistics after clustering are also available through procedure DISCRIMINANT or GLM (GLM is available in the Advanced Statistics option).

NONE. No additional output. Only the default output is displayed. NONE overrides any other specifications on PRINT.

Example

QUICK CLUSTER A B C D E
  /CRITERIA=CLUSTERS(6)
  /PRINT=CLUSTER ID(CASEID) DISTANCE.
  • Six clusters are formed on the basis of the five variables A, B, C, D, and E.
  • For each case in the file, cluster membership and distance from cluster center are displayed. Cases are identified by the values of the variable CASEID.
  • Distances between all cluster centers are printed.