1. All analysis of variance procedures requires that each of the populations being compared follows the normal probability distribution.
2. All analysis of variance procedures requires that the populations being compared have equal variances.
3. One-way ANOVA is used when we are looking at the influence that one factor has on the data values.
4. If the null hypothesis is true, the mean square total is a good estimate for the population variance.
5. The degrees of freedom for the sum of squares between for one-way ANOVA equal the number of populations being compared minus one.
6. Unlike the critical z-score and critical t-score, the critical F-score can only take on positive values.
7. Analysis of variance compares the variance between samples to the variance within those samples to determine if means of populations are different.
8. ANOVA provides a lower probability of a Type I error when compared to multiple t-tests when comparing three or more population means.
9. Analysis of variance only tests to see if any pair of population means are different. To find out which population mean pairs are different, we need to perform a multiple comparisons test.
10. If the size for each sample is the same for the one-way ANOVA, the Tukey-Kramer critical range for all pairs would be the same.