Example: Scatterplot Matrix (SPLOM)

We will create a scatterplot matrix of several different variables so that we can determine whether there are any relationships among the variables in the dataset.

Note: This example uses customer_subset.

  1. Add a Graphboard node and open it for editing.
  2. On the Basic tab, select Age in years, Household income in thousands, and Credit card debt in thousands . (Use Ctrl+Click to select multiple fields/variables.)
  3. Select SPLOM.
  4. Click Run.
  5. Maximize the output window to view the matrix more clearly.
Figure 1. Scatterplot matrix (SPLOM)
Scatterplot matrix (SPLOM)

We can observe the following:

  • The histograms displayed on the diagonal show the distribution of each variable in the SPLOM. The histogram for age appears in the upper left cell, that for income in the center cell, and that for creddebt in the lower right cell. None of the variables appears normally distributed. That is, none of the histograms resembles a bell curve. Also, note that the histograms for income and creddebt are positively skewed.
  • There does not seem to be any relationship between age and the other variables.
  • There is a linear relationship between income and creddebt. That is, creddebt increases as income increases. You may want to create individual scatterplots of the these variables and the other related variables to explore the relationships further.