Data considerations for Box-Cox Transformation

To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.

For additional data considerations, consult the appropriate topic for your specific control chart.

All values must be greater than 0
If you have data values less than 0, use a Johnson Transformation.
Data must be for continuous variables
You can use the Box-Cox transformation with continuous process data, including multivariate process data. Data can be collected in subgroups or as individual observations. The Box-Cox transformation is not appropriate for attributes data or rare event data. For more information, go to Types of data for control charts.
The process must be stable
Create an Individuals Chart to determine if the process is stable before you assess normality.