ABSTRACT
This chapter presents one-way between-groups analysis of the variance (ANOVA). In the example outlined, people's first reaction might be to use multiple t-tests to compare pairs of means. In the t-test, the null hypothesis states that there is no difference between the two means. If the calculated t-ratio leads to the conclusion of retaining the null hypothesis, this signifies that the difference between the means is attributable to chance. Just as with the t-test, several assumptions underlie the use of Anova. As with the t-test, ANOVA is reasonably robust, even when these assumptions are not totally met. Having equal or nearly equal numbers of subjects in each group also increases the robustness of ANOVA. The decision as to whether to use a contrast-based or family-based multiple comparison procedure rests on the researcher's perception of which type of error, First Type or Second Type, would be the more serious to make.
