1. The F-test in an ANOVA tests the null hypothesis that all of the groups have equal variance.
2. The average of the residuals within a category used in an ANOVA is zero.
3. The within mean square in an ANOVA is another name for s2e, the sample variance of the residuals in the corresponding regression.
4. The F-statistic depends upon which dummy variable defined by a categorical variable is excluded from the regression model.
5. Tukey confidence intervals for the difference between means in an ANOVA are narrower than the corresponding Bonferroni t-intervals.
6. Bonferroni confidence intervals adjust for the effect of multiple comparisons.