This procedure performs a Box-Cox transformation of the data. It can be used to transform non-normal data into normal data.
1. Choose ProcessMA > Control Charts > Box-Cox Transformation.
2. In Variable, select the column containing the data you want to transform.
· For constant subgroup size:
- Check Constant size.
- In Subgroup, enter a value for the subgroup size.
· For non-constant subgroup size:
- Uncheck Constant size.
- In Subgroup, select the column containing the subgroup codes.
3. Click OK.
Optional
4. Check Display transformed data, if you wish to generate and print out the transformed data.
5. In Specify lambda, enter a value if you want to use a specific lambda value to transform the data.
Note To select a column of data into a textbox, double-click on any of the column names shown in the list on the left of the dialog box while in the textbox.
Variable: Numeric.
Subgroup (For non-constant subgroup size): Text or Numeric; Must contain equal number of data points as the Variable.
Subgroup (For constant subgroup size): Integer.
Specify lambda: Numeric; between -5 and 5.
You have some data in subgroup sizes of 10. The data is not normal and with outliers. You want transform the data for further analysis.
1. Open worksheet ProcessMA > Tools > Data Files > Control.xls.
2. Choose ProcessMA > Control Charts > Box-Cox Transformation.
3. In Variable, select A – Boxcox.
4. Check Constant size.
5. In Subgroup, enter 10.
6. Click OK.
Interpretation
To correct for non-normality, you can transform the data to the power of about 1.5.