The aims of many experiments require assessment of the effects of two or more different factors on the response.
When more than one factor will be varied in an experiment, it is particularly important that the experiment is well designed. A good design is needed both to allow a full assessment of how the factors affect the response and for efficiency (getting the same accuracy with fewer experimental units).
The design, modelling and analysis of experiments in this chapter introduces concepts that will be extended in later chapters.
This chapter still assumes that nothing is known about the characteristics of the experimental units before the start of the experiment — they either appear identical or we choose to ignore any differences. (Later chapters introduces experimental designs for situations where there is structure in the experimental units.)