Exposure, the degree to which a party has more or fewer than its normal complement of seats going into the election, is a particular factor that must be taken into account in interpreting congressional elections. The exposure concept has already been incorporated into some analyses of U.S. Senate contests but not in U.S. House elections, where the effect is more subtle. Accordingly, interpretations of partisan shifts in House composition are often incomplete or inaccurate. We demonstrate 1) that exposure is highly predictive of congressional seat turnovers, even ignoring short-term causal factors, 2) that it accounts equally for presidential and midterm elections, without requiring any special midterm assumptions or theories such as those embodied in "Surge and Decline," and 3) that multivariate models that include short-term economic and political variables are enhanced by the addition of exposure. We then use exposure to forecast the outcome of the 1986 House election and produce a prediction that differs from the usual expectation for midterm elections.
The Legislative Studies Quarterly is an international journal devoted to the publication of research on representative assemblies. Its purpose is to disseminate scholarly work on parliaments and legislatures, their relations to other political institutions, their functions in the political system, and the activities of their members both within the institution and outside. Contributions are invited from scholars in all countries. The pages of the Quarterly are open to all research approaches consistent with the normal canons of scholarship, and to work on representative assemblies in all settings and all time periods. The aim of the journal is to contribute to the formulation and verification of general theories about legislative systems, processes, and behavior. The editors encourage contributors to emphasize the cross-national implications of their findings, even if these findings are based on research within a single country. The Legislative Studies Quarterly is the official journal of the Legislative Studies Section of the American Political Science Association.
Washington University creates an environment to encourage and support an ethos of wide-ranging exploration. Washington University’s faculty and staff strive to enhance the lives and livelihoods of students, the people of the greater St. Louis community, the country and the world.
This item is part of JSTOR collection
For terms and use, please refer to our Terms and Conditions
Legislative Studies Quarterly
© 1986 Washington University
Request Permissions
