Analysis of Variance (ANOVA). Comparison of multiple sample means T-test: 1 = 2 ; or 1 - 2 = 0 ANOVA: 1 = 2 = 3 = 4 Multiple T-tests result in probability of a Type I error being too great . Assumptions .
When one-way (simple) ANOVA is used to compare the means of several samples, the variability between samples is compared to the variability within samples. All differences in sample means are judged statistically significant (or not) by comparing them to the variation within samples.
Suppose we have three groups of kids that are evaluated by judges on a particular skills test after being taught by three different methods (tape alone, teacher, teacher + tape). We want to know if the scores from the three groups (methods) differ.
K = number of groups
N = total number of subjects across all groups
If the calculated F ratio is greater than the critical F value from the F table, then we can state that there is a difference between at least two of the means.