ABSTRACT
Chapters 4 and 5 discussed how to analyze the design structure of an experiment. This chapter will discuss methods of analyzing the treatment structure of an experiment. In particular, this chapter will emphasize the analysis of a two-way cross-classi ed treatment structure. Suppose there are two sets of treatments T1, T2, º , Tt and B1, B2, º , Bb. Each one of the T-treatments is to be combined with each one of the B-treatments and assigned to an experimental unit. For convenience, assume that the experimental units are assigned to the treatment combinations completely at random. Alternatively, suppose a survey is taken randomly from a large set of experimental units and that experimental units in the sample can be assigned to categories according to the values of T1, T2, º , Tt and B1, B2, º , Bb. In either of these two situations, a total of bt populations are sampled in a cross-classi ed treatment structure.
