ABSTRACT
Let X and Y be the random variables denoting the responses under trea tm ents A and B, respectively. Let F(.) and G(.) be the cd f’s of the X and Y variables. We want to test the null hypothesis
In Chapter 2 we considered th is testing problem in relation to independent samples on X and Y . However, in some situations the d a ta are obtained in blocks. Often, it is somewhat difficult to get many homogeneous experimental un its for obtain ing independent samples. In those cases, it may be convenient to obtain a small number of homogeneous un its and use a set of such groups of homogeneous units, which are usually called blocks. In th is chapter we consider studies where such blocks are used to investigate the effects of two treatm ents. We only consider the case of blocks consisting of two units. The two trea tm en ts are random ly assigned to the un its in each block.
