ABSTRACT
For the matched-pairs sign and signed-rank tests of Chapter 5 the data consisted of two samples, but each element in one sample was linked with a particular element of the other sample by some unit of association. This sampling situation can be described as a case of two dependent samples or alternatively as a single sample of pairs from a bivariate population. When the inferences to be drawn are related only to the population of differences of the paired observations, the first step in the analysis usually is to take the differences of the paired observations; this leaves only a single set of observations. Therefore, this type of data may be legitimately classified as a one-sample problem. In this chapter we shall be concerned with data consisting of two mutually independent random samples, i.e., random samples drawn independently from each of two populations. Not only are the elements 232within each sample independent, but also every element in the first sample is independent of every element in the second sample.
