Minitab Cookbook

Finding similarity across columns using cluster variables

Cluster variables work in a manner that is similar to cluster observations. Here, we are interested in the columns and variables in the worksheet rather than grouping the observations and rows.

We will look at the dataset for car fuel efficiency to identify groups of variables, or rather identify the columns that are similar to each other.

This dataset was collected from manufacturer-stated specifications.

How to do it…

1. Open the mpg.MTW worksheet.
2. Go to the Stat menu, click on Multivariate, and select Cluster Variables….
3. Enter the columns for CO2, Cylinders, Weight, Combined mpg, Max hp, and Capacity into the Variables or distance matrix: section.
4. Check the Show dendrogram option.
5. Click on OK