Name: John Harnisher
Home office: SPSS New York
About John: John, a senior instructor, has a master’s degree in organizational psychology from New York University, from which he will soon receive his doctorate. John has been an SPSS instructor since 1997.
Using TwoStep Cluster Analysis in SPSS Advanced Models
The TwoStep Cluster procedure in SPSS Advanced Models offers a major advantage over other clustering methods. Using TwoStep clustering, you can easily calculate the appropriate number of clusters for a given segmentation study. TwoStep’s default setting calculates the appropriate number of clusters by comparing the values of a model-choice criterion across different clustering solutions. In contrast, the k-means and hierarchical methods require you to choose the number of clusters, which involves additional analysis.
Another TwoStep advantage is its ability to manage both categorical and continuous data. In traditional cluster analysis, you must dummy code categorical variables. TwoStep doesn’t require dummy coding, so you have more time to interpret the clusters. TwoStep also offers a dialog box with separate points for categorical and continuous variables, and the TwoStep output is also separated.
These advantages make TwoStep a convenient exploratory technique.
To run the TwoStep procedure, simply choose the following menu options:
For the initial attempt, all of the default settings in the main dialog box are used (see Figure 1). Place continuous variables (represented by a ruler) in the bottom box, and place categorical variables in the top box. Then hit OK and begin to interpret the resulting graphs and charts.
Figure 1
We offer SPSS courses at locations around the world.
Find a course in the location nearest to you.
Predictive Analytics
can make your organization
more
successful
Resources