Table 1.

Statistical tests of normality: known distributions

Distribution
ProcedureNormalLog-NormalUniform
Anderson-Darling0.640.0010.003
Cramér-von Mises0.670.0010.01
Lilliefors0.500.0010.08
Shapiro-Francia0.650.0010.005
Shapiro-Wilk0.620.0010.001
  • Values represent P values associated with procedures designed to assess the null hypothesis that sample observations are consistent with having come from a normal distribution. The sample observations are the 100 observations drawn from each distribution in Fig. 3. A value of 0.001 denotes P < 0.001. The commands in lines 202–235 of Advances_Statistics_Code_Normal.R analyze the samples drawn from each distribution in Fig. 3. Your P values will differ slightly. To obtain the analyses summarized in this table, highlight and submit the lines of code from Table 1: first line to Table 1: last line.