ABSTRACT
This chapter discusses the comparison of means from multiple groups, with a primary emphasis on one-way analysis of variance. A motivating example of the effect of THC on mice runs throughout much of the chapter. Notation and the one-way ANOVA test statistic are given, with simulations used to justify claims made about the distribution of the test statistic under the null hypothesis. The robustness of one-way ANOVA to departure from assumptions is assessed via simulations, with special attention given to the interaction between unequal variance and unequally sized groups. The R function oneway.test is introduced and applied to a data set which does not meet the assumptions of one-way ANOVA. Family-wide error rates and p-value adjustments for multiple comparisons are discussed in the context of pairwise t-tests. The chapter concludes with a vignette on reproducibility.
