Specify a transformation for a Normal Capability Analysis for Multiple Variables

Stat > Quality Tools > Capability Analysis > Multiple Variables (Normal) > Transform
You can transform your data to fit a normal distribution in order to satisfy the assumptions for the analysis.
  • No transformation: Do not use a transformation if your data already follow a normal distribution. To determine the distribution of your data, or whether a transformation will be effective if your data are nonnormal, use Individual Distribution Identification.
  • Box-Cox power transformation (W = Y^λ): Use the Box-Cox transformation if your nonnormal data are all positive (> 0) and you want to obtain estimates of within-subgroup (potential) capability as well as overall capability. The Box-Cox transformation is a simple, easy-to-understand transformation.
    Select the lambda (λ) value that Minitab uses to transform the data.
    • Use optimal λ: Use the optimal lambda, which should produce the best fitting transformation. Minitab rounds the optimal lambda to 0.5 or the nearest integer.

      To use an exact value instead of a rounded value for optimal λ, choose File > Options > Control Charts and Quality Tools > Other and deselect Use rounded values for Box-Cox transformations when possible.

    • λ = 0 (ln): Use the natural log of your data.
    • λ = 0.5 (square root): Use the square root of your data.
    • Other (enter a value between -5 and 5): Use a specified value for lambda. Other common transformations are square (λ = 2), inverse square root (λ = −0.5), and inverse (λ = −1). In most cases, you should not use a value outside the range of −2 and 2.
  • Johnson transformation (for overall analysis only): Use the Johnson transformation if your nonnormal data contain negative values (or 0) or if the Box-Cox transformation is not effective. The Johnson transformation function is more complicated than Box-Cox, but is very powerful for finding an appropriate transformation.
    P-Value to select best fit

    Enter a value between 0 and 1. The value that you enter defines the significance level for a normality test of the data before and after the transformation. A higher value makes the criteria for normality more rigorous. A lower value makes the criteria for normality less stringent.