ABSTRACT
Suppose an experimenter wants to compare the effects of several different treatments, such as the effects of different drugs on people’s heart rates or the yields of several different varieties of wheat. Often the rst step in analyzing the data from such experiments is to use a statistical method, known as a one-way analysis of variance model, to describe the data. The model on which the one-way analysis of variance is based is one of the most useful models in the eld of statistics. Many experimental situations are simply special cases of this model. Other models that appear to be much more complicated can often be considered as one-way models. This chapter is divided into several sections. In the rst two sections, the one-way model is de ned and the estimation of its parameters is discussed. In Sections 1.3 and 1.5, inference procedures for speci ed linear combinations of the treatment effects are provided. In Sections 1.7 and 1.9, we introduce two basic methods for developing test statistics. These two methods are used extensively throughout the remainder of the book. Finally, in Section 1.11, we discuss readily available computer analyses that use the above techniques. An example is used to demonstrate the concepts and computations described in each section.
