ABSTRACT
The purpose of this paper is to examine one approach to a fundamental problem confronting the geographer interested in developing models of voting behavior -- that of more satisfactorily integrating spatial and behavioral approaches given the types of data availability constraints typically encountered in electoral geography. In an attempt to overcome this problem, a simple model is developed and tested empirically. Results suggest that derivatives of the model have considerable potential in analyses of political processes over space, particularly those directed toward discerning the impact of localisms on a voting response surface.
