If we know the exact consequences of each action, then we can select an action with the largest value of the objective function. In practice, we often only know these values with interval uncertainty. If two intervals intersect, then some people may prefer the alternative corresponding to the first interval, and some prefer the alternative corresponding to the second interval. How can we describe the portion of people who select the first alternative? In this paper, we provide a new theoretical justification for Hurwicz optimism-pessimism approach, and we show how this approach can be used in group decision making.