This article shows that the bicriteria preference indices of the MAPPAC method may be classified in two categories: those of a compensatory type in cases of discordant evaluations, methodologically in line with the MAUT approach, and those which underline partial dominance relations in cases of concordant evaluations, in analogy with noncompensatory outranking methods and with the ELECTRE methods in particular. MAPPAC may therefore be considered a 'mixed' or 'intermediate' method, a sort of compromise between the approaches mentioned above. It will also be shown that the aggregated index of MAPPAC is a function not only of the weight of the criteria and of the entity of the difference in evaluation, but also, all other conditions remaining unvaried, of the number of criteria according to which one action is preferred or indifferent with respect to another. It is possible to note a number of similarities and differences between the indices of this method and the concordance indices of the ELECTRE methods.