Paired comparisons and designed experiments

Paired comparisons can be a very effective way of performing measurements of preferences. The statistical analysis can be carried out as an ANOVA although the explanatory variables are not categorical variables. Explanatory variables of subjects and objects can be included in the ANOVA if the experiment is carefully designed. It is often feasible or necessary to reduce the number of comparisons to be performed by subjects, however, such structure can be included in ANOVA if comparisons to be performed are carefully selected.