Second-degree plans
Design of experiments

So far, we've considered factors that are continuous or discrete but take only two levels. In this paragraph, we will study continuous factors taking more than two levels. The mathematical model adopted in the previous paragraphs was of the first degree with respect to each of the variables. If this first-degree model is valid, further experimentation is rarely necessary. On the other hand, if it is not valid, it is necessary to move on to a higher-degree model. We limit ourselves here to the second-degree model, which proves sufficient in most cases.