ABSTRACT
In the last chapter we concentrated on the treatments of the experiment, and looked at ways of introducing structure into the treatment effects to answer specific questions. There are two other aspects of experimental design, and these will be introduced in this chapter. Often an experimenter can group experimental units in such a way that the units within groups are more likely to be uniform than units in different groups. Treatments can be allotted to take this grouping into account so that treatment comparisons are made between experimental units which are as uniform as possible. A typical experimental plan would be as in Figure 7.1.1. Here six treatments (perhaps six different crop varieties) are being compared on 24 experimental units which have been grouped into four blocks of six contiguous plots. This should seem sensible, because six contiguous plots are more likely to be uniform than the whole 24 plot area. Then treatments have been allocated to plots by making a block contain one complete set of treatments, or one replicate, randomly arranged within the block.
