ABSTRACT
This chapter introduces the terminology, lists the assumptions, and gives tests for the assumptions used in the analysis of the data from designed experiments. It focuses on experiments containing only one factor. The layout of the experiment is the exact fashion in which the experiment is carried out. The treatment levels of the factor to be applied to the experimental units are pre-specified and the inference of the experiment is only to these fixed levels. The experimental units are randomly selected from some large pool of all possible experimental units with the inference being to all such experimental units. Generally speaking, the F ratio used in the analysis of variance has been shown to be very robust to departures from normality. In addition, lack of normality is often associated with heterogeneity and transformations that improve homogeneity also improve normality.
