Analysis of Variance in Different Experimental Designs

This chapter introduces analysis of variance (ANOVA) for various experimental designs, where an outcome variable is measured in several categories (levels) of one or more factors. Hypothesis testing asks whether the population mean of the variable differs among the levels of each factor. ANOVA for randomized block design can be used to address such a deficiency. The ANOVA for randomized block design should be simplified to a one-way ANOVA. The ANOVA for factorial design analyzes “interaction effect” through the crossed combination of multiple factors at each level. The chapter focuses on the ANOVA for two-factor factorial design experiments. It introduces the ANOVA method for one-sample repeated measures design data, which is the most commonly used in practice. The chapter considers data obtained using a two-treatment and two-period crossover design, taking a pharmaceutical trial as an example to introduce the ANOVA technique.