Working with Correlation Matrices :: SAS(R) Visual Analytics 7.2: User's Guide

About Correlation Matrices

A correlation matrix displays the degree of correlation between multiple intersections of measures as a matrix of rectangular cells. Each cell in the matrix represents the intersection of two measures, and the color of the cell indicates the degree of correlation between those two measures.

A correlation matrix can either compare within a single set of measures or it can compare between two sets of measures.

The correlation values are calculated by using Pearson’s product-moment correlation coefficient. Correlation values are identified as weak, moderate, or strong as follows:

Weak

the absolute value is 0.3 or lower

Moderate

the absolute value is greater than 0.3 and less than or equal to 0.6

Strong

the absolute value is greater than 0.6

Data Roles for a Correlation Matrix

The basic data role for a correlation matrix is a measure. You must assign at least two measures.

Note: The maximum number of measures is 60.

The Show Correlations option specifies whether the correlation matrix uses a single set of measures (Within one set of measures) or two sets of measures (Between two sets of measures.)

If you select Between two sets of measures, then assign measures to the X axis and Y axis roles.

Specify Properties for a Correlation Matrix

On the Properties tab, you can specify the following options:

Name

specifies the name of the visualization.

Title

specifies the title that appears above the graph.

Note: The Title option is disabled if you select Generate graph title.

Generate graph title

specifies that the graph title is generated automatically based on the data items in the visualization.

Show borders

specifies that the borders between cells are visible.

Rotate axis labels

displays the axis labels at an angle.

Color gradient

selects the gradient colors for the visualization.

Sort Correlation Values

For a correlation matrix between two sets of measures, you can sort by the correlation values for a measure.

To apply sorting, right-click the Measures button on the axis that you want to sort, and then select Sort. You can select any of the measures on the selected axis, and you can select whether to sort the correlation values in ascending or descending order.

Explore the Data for a Cell

For any cell (crossing) in the correlation matrix, you can explore the measures for that cell as a heat map.

To explore the measures, right-click the cell that you want to explore, and then select Explore measure by measure.

A new heat map visualization displays the two measures from your selected cell.