Two-Way Analysis of Variance with Interaction

Effects to be entered: A, B, A*B

In the Select Columns list, select two nominal or ordinal effects, A and B.

Select Macros > Full Factorial.

Or:

In the Select Columns list, select two nominal or ordinal effects, A and B.

Click Add.

In the Select Columns list, select A and B again and click Cross.

Example of Two-Way Analysis of Variance with Interaction

Open Popcorn.jmp. You are interested in whether popcorn and batch have an effect on yield.

Select Analyze > Fit Model.

In the Select Columns list, select yield and click Y.

In the Select Columns list, select popcorn and batch.

Select Macros > Full Factorial.

The Fit Model window appears as shown in Figure 2.10.

Figure 2.10 Fit Model Window for Two-Way Analysis of Variance with Interaction

Figure 2.11 shows a profiler plot of the fit for this example. To obtain this plot, click Run in the Fit Model window shown in Figure 2.10. Then, from the red triangle menu for the Fit Least Squares report, select Factor Profiling > Profiler.

In the top plot, popcorn is set to gourmet, and in the bottom plot, it is set to plain. Note how the predicted values for the settings of batch depend on the type of popcorn. This is a consequence of the interaction between popcorn and batch.

Figure 2.11 Model Fit for a Two-Way Analysis of Variance with Interaction