ABSTRACT
In Chapter 4 we introduced the concept of hypothesis testing. To illustrate the hypothesis testing problem, and to study concepts related to Type I and Type II errors, we examined the test of hypotheses for a single sample mean. In research, it is rare that an experiment will involve a single sample. In this chapter we will concern ourselves with comparisons of means and variances from two samples. This situation is of considerably greater interest in research than the single sample tests. Data for the two-sample case may arise from either a survey involving two groups or from a designed experiment involving the comparison of two treatments. The use of the word “treatment” is in a general sense, in that one of the treatments could be a nontreated group or other appropriate control group. In Chapters 7 through 10 we shall concern ourselves with a technique with more general application known as the analysis of variance.
