ABSTRACT
This chapter considers factorial experiments replicated in designs other than a completely randomized design (CRD), and looks at some special cases for factorial experiments. With a CRD there are rabc experimental units, and they are randomly assigned to the abc treatments with the only restriction being that r are assigned to each treatment. There are special cases of factorial experiments that occur frequently in applications. Some of these have special structures and formulas that make them especially easy to use. In some cases a factorial experiment can require more treatments or observations than is feasible in terms of time and money. A Latin square design can also be used with a two factor factorial experiment. If factors A and B have a and b levels, respectively, then these "ab" treatments can be used in a Latin square with "ab" rows and columns. A factorial arrangement is used to assign the treatments to the experimental units.
