On the Nature of Political Issues: Insights from a Spatial Analysis

From citizen evaluations of candidates we construct a spatial representation of the 1968 and 1972 elections. This representation differs from that of Weisberg and Rusk (1970) in that it includes both voters and candidates, enabling us to examine the mass-elite relationship. The spaces we recover provide an excellent overview of the two elections. However, contrary to expectations based on spatial theory, we find no political figure located in the center of the distribution of voters. This result can be explained if issues operate in a dispositional manner. The thrust of the dispositional model is that it is the direction of a candidate's policy that is critical to developing his support base, not his absolute position. The model implies no candidate will receive his strongest support from those who are centrist on an issue. We test this model on a variety of issues with respect to several candidates and find it well supported.