Committee Voting Under Alternative Procedures and Preferences: An Experimental Analysis

Abstract

This paper reports on four series of experiments in a five-person committee voting under majority rule. Each of two voting procedures was paired with each of two types of preference sets. The types were characterized as high or low intensity. Every set of preferences had a Condorcet point and that point was the best alternative for one (and only one) voter. When the high intensity preferences were used, committees operating under either voting procedure selected the Condorcet point more than 90% of the time; when low intensity payoffs were used, the success rate was less than 51%. A theory is suggested which predicts which preference sets should successfully induce selection of the Condorcet point and which should not; in the latter case, the same theory predicts that the choice will be confined to a certain collection f the other points. Our observations are consistent with this theory.