ABSTRACT
How persons in interaction combine their individual preference patterns into a social choice is a fundamental question about which there has been much speculation, and an amount of mathematical work; but there have been few, if any, empirical studies dealing directly with the question. 1 The problem is ubiquitous; examples abound. The eleven members of the Security Council of the United Nations must decide to take or not take some action; a president of the United States must be selected from the millions of eligible citizens, but particularly from the dozen or so likely possibilities; a university research institute must plan a research program that will interest its members who have diverse preferences and satisfy its sources of support; consumer goods must be produced and distributed to satisfy the preferences of millions of consumers; Congress must apportion the $ 50 billion defense budget among the Army, Navy, Air Force, and Marines; and a family must decide whether it is to live in the city or a suburb. All of these diverse decisions have something in common: In some way the preferences of different individuals or groups must be summed up or amalgamated into a social or group choice.
