JMP can handle data with layers of repeated measures. For example, see the Cholesterol.jmp data table. Groups of five subjects belong to one of four treatment groups called A, B, Control, and Placebo. Cholesterol was measured in the morning and again in the afternoon once a month for three months (the data are fictional). In this example, the response columns are arranged chronologically with time of day within month.
1.
Select Help > Sample Data Library and open Cholesterol.jmp.
2.
Select Analyze > Fit Model.
3.
Select April AM, April PM, May AM, May PM, June AM, and June PM and click Y.
4.
Select treatment and click Add.
6.
Click Run.
Treatment Graph
In the treatment graph, you can see that the four treatment groups began the study with very similar mean cholesterol values. The A and B treatment groups appear to have lower cholesterol values at the end of the trial period. The control and placebo groups remain unchanged.
7.
Click on the Choose Response menu and select Compound.
8.
Use the options in Compound Window to complete the window.
Compound Window
9.
With a p-value of 0.6038, the interaction between Time and treatment is not significant. This means that there is no difference in treatment between AM and PM. Since Time has two levels (AM and PM) the exact f-test appears.
With p-values of <.0001, the interaction between Month and treatment is significant. This suggests that the differences between treatment groups change depending on the month. The treatment graph in Treatment Graph indicates no difference among the groups in April, but the difference between treatment types (A, B, Control, and Placebo) becomes large in May and even larger in June.
The interaction effect between Month, Time, and treatment is not statistically significant.
Cholesterol Study Results