One-Way Analysis of Variance

When comparing more than two groups it is necessary to use a statistical technique called analysis of variance (ANOVA). This statistic does not evaluate differences between two groups but rather assesses how much one might expect the different group means to spread out if they all came from the same population. This chapter simplifies what is meant by an F score and takes the mystery out of the term homogeneity of variance. One of the more complicated tasks following an ANOVA is deciding on which statistic to use to actually compare possible differences between group means. Because there are more than two groups when using an ANOVA, there are many more different comparisons that can be made. The different multiple comparison procedures (post hoc tests) are described in detail and compared with each other. There is also a discussion of repeated measure analysis and the concept of sphericity.