Computing Fitted Factorial Effects in JMP®
Chapter 4, Printout 6
- Launch JMP and create a new data table.
- Enter the data.
- Rename Column 1 to "Joint"
- Double-click the column header, "Joint"
- In the dialog box that follows, pull the "Data Type" menu down to "Character." The "Modeling Type" menu should read, "Nominal."
- Enter the joint type data from the Joint column in Table 4.11 (p. 164).
- Add a second column, Wood, and enter the wood type data from the Wood column in the table.
Note: when JMP adds a column, it assumes that column's data type to be the same as that of the previous column. In this case, the Wood column also contains nominal data, so it is unnecessary to set the data type. The data type of the next column will have to be changed manually.
- Add a third column, "y"
- Double-click the column header, "y"
- In the dialog box that follows, pull the "Data Type" menu down to "Numeric." Pull the "Modeling Type" menu down to "Continuous."
- Enter the y data from the Stress at Failure column in the table.
- Compute fitted effects.
- Select "Analyze --> Fit Model."
- In the dialog box that follows, click "y" in the "Select Columns" list. Then click the "Y" button to designate y as the response variable.
- Click "Joint" in the "Select Columns" list; then click "Add" to add the Joint column to the list under "Construct Model Effects."
- Repeat the above step, this time adding the Wood column.
- In the list under "Construct Model Effects," click "Joint." In the "Select Columns" list, click "Wood." Then click "Cross" to add the cross of the Joint and Wood columns to the list."
- Click "Run Model."
JMP displays the fit summary, an actual by predicted plot, and leverage plots of for Joint, Wood, and Joint*Wood.
Screenshot (summary and actual by predicted):
Screenshot (leverage plots):