ABSTRACT
The extension to more than two factors in a factorial experiment is fairly simple in terms of the calculation of the analysis of variance; the difficulty is in the interpretation. Suppose that we have an experiment in r blocks with factors A, B, and C, having a, b, and c levels, respectively. The treatment SS for the total set of abc experimental treatments can be subdivided into
(i) SS due to the main effect of each factor, e.g., for factor A:
where T
is the total of all units at level 1 of factor A, summing over all levels of B and C.
