Conduct a Permutation Test
First, you construct a formula that randomly shuffles the pain measurements among the three drugs. Under the null hypothesis of no effect, any of these allocations is as likely as any other. It follows that the F ratios obtained in this manner approximate the distribution of F ratios under the null hypothesis. Finally, you compare the observed value of the F ratio to the null distribution obtained by simulation.
Define the Simulation Formula
1.
Select Help > Sample Data Library and open Analgesics.jmp.
2.
Select Cols > New Columns.
3.
Type Pain Shuffled for Column Name.
5.
In the function list, select Row > Col Stored Value.
8.
From the list of functions, select Random > Col Shuffle.
Completed Formula
9.
Click OK in the Formula Editor window.
10.
Click OK in the Column Info window.
1.
Select Analyze > Fit Y by X.
2.
Select pain and click Y, Response.
3.
Select drug and click X, Factor.
4.
Analysis of Variance Report
Notice that the F ratio is 6.2780.
9.
Next to Number of Samples, enter 1000.
10.
(Optional) Next to Random Seed, enter 456.
Completed Simulate Window
11.
In the table of simulated results, the columns for C. Total and Error are empty, since the F Ratio value in the Analysis of Variance table only applies to drug.
Simulated Distribution of F Ratios under the Null Distribution
The observed F ratio value of 6.2780 is represented with a red line in the histogram. This value falls in the upper 0.5% of the simulated null distribution of F ratios. This presents strong evidence that the three drugs differ in their effects on pain.

Help created on 9/19/2017