The design and analysis of factorial experiments

Factorial experiments are experiments which include all combinations of several different sets of treatments or "factors." Information is thus simultaneously obtained on the responses to the different factors, and also on the effects of changes in the level of each factor on the responses to the others.

This Technical Communication has not been written with the object of convincing experimenters of the need for employing factorial designs, but rather for those who, while fully conscious of the advantages of such designs, find difficulty in laying them out and in analysing the results. It is,_ in fact, an attempt to give a comprehensive survey oi the simpler types of design at present available, and a description of the appropriate methods of analysis. The reader who has not done so is advised to first read Prof. R. A. Fisher's Design of Experiments, where he will find a full account of the logical basis of the whole technique of modern experimental design.