14.5 Customizing the Study
When the original data are not normal, you may use a simple mathematical transformation to obtain data that, for practical purposes, are normally distributed.
Method used to estimate the standard deviation.
Usually, there is no need to change the default options. The within standard deviation is estimated via the average of the sample variances (if the sample sizes differ, each variance is weighted by the degrees of freedom of each sample).
It is well known that the sample variance s2 is an unbiased estimator of the population variance σ2; however, the sample standard deviation s is not an unbiased estimator of the population standard deviation σ. To make it unbiased, it is necessary to apply a bias correction factor which is a constant that depends on the degrees of freedom used to estimate s2. Minitab applies it by default (use unbiasing constants).
When the subgroup size is one, by default the short time variability is computed via the average moving range of order two.
Sometimes, 8σ is used as the width of the bell-shaped curve to compute the capability indexes.