27.03.2013 Views

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

1594 TWOSTEP CLUSTER<br />

Example<br />

TWOSTEP CLUSTER<br />

/CONTINUOUS VARIABLES = INCOME<br />

/CATEGORICAL VARIABLES = GENDER RACE<br />

/NUMCLUSTERS AUTO 10 AIC<br />

/PLOT VARCHART NONPARAMETRIC CONFIDENCE 80.<br />

• Clusters will be based on the values of INCOME, GENDER, and RACE, where INCOME is<br />

specified as a continuous variable, and GENDER and RACE are specified as categorical<br />

variables.<br />

• The optimal number of clusters, up to a maximum of 10, will be determined automatically<br />

by the procedure using the Akaike Information Criterion to choose between models.<br />

• For each cluster, a chart will be produced that compares the importance of the variables<br />

using a nonparametric measure and a confidence level of 80%.<br />

Variable List<br />

The variable lists specify those variables to be clustered. The first variable list specifies only<br />

continuous variables, and the second list specifies only categorical variables (that is, the two<br />

lists are disjoint).<br />

CATEGORICAL Subcommand<br />

The CATEGORICAL subcommand specifies a list of categorical variables.<br />

Example<br />

TWOSTEP CLUSTER<br />

/CATEGORICAL VARIABLES = RACE GENDER CITIZEN<br />

/PRINT SUMMARY COUNT.<br />

This tells TWOSTEP CLUSTER to cluster the categorical variables RACE, GENDER and<br />

CITIZEN. Summary statistics by cluster and cluster frequencies are output in tables.<br />

CONTINUOUS Subcommand<br />

The CONTINUOUS subcommand specifies a list of scale variables.<br />

Example<br />

TWOSTEP CLUSTER<br />

/CATEGORICAL VARIABLES = RACE GENDER CITIZEN<br />

/CONTINUOUS VARIABLES = INCOME<br />

/PRINT SUMMARY COUNT.<br />

This tells TWOSTEP CLUSTER to cluster the categorical variables RACE, GENDER, and<br />

CITIZEN, and the numeric variable INCOME. Summary statistics by cluster and cluster<br />

frequencies are output in tables.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!