| Test | Result |
|---|---|
| Shapiro-Wilk |
This panel is intended to help determine the likelihood that the input data is a sample from a normally distributed population. These tests are hypothesis tests where the null hypothesis is that the data is a sample from a normal distribution. If p is below some threshold (typically p < 0.05), then we reject the null hypothesis and conclude that it is unlikely the data is a sample from a normally distributed population.
| Attribute | Value |
|---|---|
| n | 0 |
| Mean | |
| Standard deviation | |
| Cpk | n/a |
| Ordered data |
On this panel we look at a histogram of the data and overlay any specification limits and tolerance interval. The graphical output gives a quick visual indication of where the data lies relative to the specification limits.
The process capability index, Cpk, is a measure of how closely the data lies to its specified limits. A Cpk of 1 means that the mean of the data is 3 standard deviations away from its closest limit.
The tolerance interval is a range within which a stated proportion P of the population lies, with a given confidence level C. This tolerance interval is unrelated to the engineering tolerances, which we call specification limits here to avoid confusion. The accuracy of the tolerance interval is based on an underlying assumption that the population is normally distributed, so the normality check on this page is a critical starting point.