Using the Group By option, you can select a classification (grouping) variable. When a grouping variable is in effect, the Bivariate platform computes a separate analysis for each level of the grouping variable. The scatterplot shows the fitted models (lines, curves, contours, or ellipses) for each grouping variable. The fit for each level of the grouping variable is identified beneath the scatterplot, with individual pop-up menus.
When a grouping variable is in effect, the Group By option is checked in the Bivariate Fit red triangle menu. You can change the grouping variable by first selecting the Group By option to remove (uncheck) the existing variable. Then, select the Group By option again and respond to its window as before.
The Group By option enables you to generate separate analyses by a grouping variable while plotting the fits on the same scatterplot. This enables you to visually compare the fits across groups.
Note: For examples of this option, see Example of Group By Using Density Ellipses and Example of Group By Using Regression Lines.