Sigmapedia - The Free Online Lean Six Sigma Encyclopedia

English |  Español |  Français |  Português |  Deutsch |  中文

Normality Plot

Go Back


This is the probability plot produced by statistical software when a normality test is performed. Perfectly normal data would form a straight line on the probability plot. Typically, a normality test such as the Anderson-Darling Normality Test statistic is shown on the graph to aid in its interpretation.


The image shows normal probability plots for three different datasets. The one on the left comes from an Uniform distribution, the one in the center comes from a Normal distribution, and the one on the right comes from an Exponential distribution.

Although there are more sophisticated ways of telling whether or not a sample comes from a normal population, a common informal test is the ‘Fat Pencil Test’. Take the normal probability plot and lay an imaginary fat pencil across the plotted data. If the pencil covers all the data points, the data are most likely from a normal distribution.