Institute of Governmental Studies

There are numerous models predicting the partisan distribution of seats in the U.S. House of Representatives using national-level variables. The partisan distribution in Congress is simply an aggregation of 435 separate elections. This paper is an attempt to formulate measures of differential partisan vulnerability in elections from individual election results. The measures constructed are compared to models in the literature. Findings indicate that either the measures are not indicative of the exposure thesis or that, similar to the literature on sociotropic versus “pocketbook” economic voting, the district-level causal mechanism is complex and unclear.