ABSTRACT
Suppose we have a set of data presented in the form of a complete two-way layout of I rows and J columns, with one entry in each of the IJ cells. In the sampling situation of Chapter 10, if the independent samples drawn from each of I univariate populations were all of the same size J, we would have a complete layout of IJ cells. However, this would be called a one-way layout since only one factor is involved, the populations. Under the null hypothesis of identical populations, the data can be considered a single random sample of size IJ from the common population. The parallel to this problem in classical statistics is the one-way analysis of variance. In this chapter we shall study some nonparametric analogs of the two-way analysis-of-variance problem, all parallel in the sense that the data are presented in the form of a two-way layout which cannot be considered a single random sample because of certain relationships among elements.
