Abstract
Data about the 1972 U.S. Presidential election are used to test theorems from the spatial model of electoral competition, a theory of social choice expressed through the voting mechanism. Four statistical models are derived from the theorems and tested against the data. All four models are supported in the data, predict voting behavior well, and generally appear to satisfy the requisite conditions. The best fitting model is then examined in more detail. This model continues to meet the theoretical conditions under more exacting inquiry. The paper concludes with a discussion of the implications of this analysis for social choice and the spatial models and for understanding electoral behavior.