ABSTRACT
In the previous chapters on factorial experiments, we came across factors that were either quantitative or qualitative. We know that when a factor in a system is quantitative, it is often possible to represent the responses in terms of a function of the factor plus random error. We illustrated this method when we used orthogonal polynomials and regression analysis to represent the responses in terms of the levels of the explanatory variables (see Example 2.3). The same was true of the regression approaches to ANOVA we adopted elsewhere in which we employed orthogonal contrasts. The aim in using the orthogonal polynomials or contrasts in the above examples was to enable us find a suitable function, which could approximate the responses in terms of the levels of the factors in the designed experiment.
