ABSTRACT
In this chapter, we cover some more advanced mathematical topics. We begin with the general techniques and rules for deriving sums of squares, degrees of freedom, and expected mean square terms for a given ANOVA design. We consider nonorthogonal sums of squares, the situation when predictors are themselves correlated. We examine ANOV A in terms of a general linear model; this is the way computers actually perform ANOV As these days, and conceptually the most advanced and useful way to think about ANOV A. Then we discuss effect size.
