STATISTICS / INDEPENDENT GROUPS
ANOVA & p-values
Compare group means, inspect assumptions and export the complete statistical record.
Data & settings
Synthetic example data
Results
Preparing the labelled synthetic example…
Method
p = Pr(F distribution ≥ observed F)
One-way ANOVA assumes independent observations, a common residual variance and an appropriate residual distribution. Welch ANOVA permits unequal group variances. Two-way analysis includes interaction and uses Type III partial sums of squares with sum-to-zero contrasts; all factor combinations require replication. A p-value is not the probability that the null hypothesis is true and does not measure practical importance.
NIST: one-way ANOVA · NIST: Levene/Brown–Forsythe · SciPy: classical and Welch definitions · statsmodels: Type III ANOVA
Additional methods and scope
This browser module computes classical/Welch/two-way ANOVA, variance tests, confidence intervals, effect sizes where defined, Holm-adjusted pairwise Welch tests and residual diagnostics. It does not compute Shapiro–Wilk, Tukey, Games–Howell, Dunnett or Dunn. Those methods remain in the original Python workbench, which requires a configured Python service. Repeated-measures and mixed-effects models are outside this module’s scope.
Export
Download full-precision results and inputs, or save a record in My work. Figure exports are beside each plot.