Normality Test Suite
Overview
The Normality Test Suite applies multiple statistical tests to assess whether your data follows a normal distribution. Run Shapiro-Wilk, Anderson-Darling, Kolmogorov-Smirnov, and D’Agostino-Pearson tests simultaneously, visualize your data with Q-Q plots and histograms with normal curve overlays, and get clear traffic light indicators for normality. Essential for choosing appropriate statistical methods and understanding your data’s distributional properties.
Tips
- Shapiro-Wilk is generally the most powerful test for sample sizes under 2000 and is recommended for most applications
- Visual inspection of Q-Q plots is often more informative than p-values alone - look for systematic deviations from the diagonal
- Small deviations from normality may be statistically significant with large samples but practically unimportant
- For sample sizes over 30-40, many parametric tests are robust to moderate non-normality due to the Central Limit Theorem
- Skewness measures asymmetry: values near 0 indicate symmetry, positive values indicate right skew, negative values left skew
- Kurtosis measures tail heaviness: values near 0 (excess kurtosis) indicate normal tails, positive values heavy tails, negative light tails
- If all tests reject normality, consider data transformations (log, square root, Box-Cox) or non-parametric alternatives