Cont Scaled Colored Bubble Plot
 |
|
Description: This demonstration shows how a third variable can be introduced to a two-variable scatter plot by means of color and area. You cannot do this using the Rows > Color and Mark By command. This script replaces the points with circles but otherwise leaves the full Bivariate functionality as it is.
Instructions: Open the data table below (or another that contains at least three continuous variables). Open and run the script. Choose the response and predictor variables, then choose a scaling variable. For example, choose Oxy as the response, Weight as the predictor, and Runtime as the scale. Notice that the scatter of the circles does not indicate a strong relation between Oxy and Weight but the circle color and area generally decrease when you scan the plot from left to right. (Actually the color does not decrease in the same sense as the area changes but there is more or less a regular, continuous change in the color across the horizontal axis.) This pattern indicates that a more useful plot might be Oxy by Runtime.
Requirements: Fitness.JMP (included)
SAS Campus Drive, Building S, Cary, NC, 27513 • Phone: 1.919.677.8000© 2008 SAS Institute, Cary, NC, USA. All Rights Reserved. Terms of Use • Privacy Policy