EITM: Collaborative Research on Legislative Party Switching: Integrating Formal & Comparative Empirical Analyses

Carol A Mershon
Department of Politics

Party switching in legislatures is important. Recent US history drives home the point. When Senator Jeffords left the Republican Party in 2002 to serve in the Senate as an independent, he fundamentally altered the policy-making landscape. In leaving the Republican Party and handing control of the Senate to the Democrats, Jeffords transformed a unified government into a divided government.

The change in partisan control of the Senate changed policy making and outcomes. His action thus demonstrated the importance of political parties, even in the United States where the two major parties often are seen as weak and essentially indistinguishable from one another.

Political scientists of all stripes agree that parties are essential to representative democracy. Parties organize legislatures, articulate choices, aggregate preferences, and, above all, provide labels under which candidates vie for public office. Scholars commonly examine such issues as the value of party labels or party discipline for voters' decision making, but rarely ask how these affect legislators' choices. Students of US politics debate the degree to which parties condition legislators' voting, for example, but few have trained either empirical or theoretical attention on such other aspects of legislative behavior as the choice of party affiliation. Only a few formal models focus on politicians' party switching. A more extensive but still limited empirical literature, most of it divorced from formal research, examines legislators' moves among parties.

This project investigates the causes and consequences of party switching. It seeks to answer such questions as Why would a legislator decide to change party affiliation? Under what conditions would a party choose to accept a defector from another party? Most broadly, what are the effects of party switching?

Answers to these questions will both illuminate party switching and advance our understanding of parties, party discipline, and party systems. They also will advance understanding of how interdependent, individual decisions interact to create a social dynamic. To pursue this agenda, the project integrates formal and empirical approaches to the study of politicians' choices of, and changes in, party affiliation.

The co-Principal Investigators will build a formal model of party switching, specify the model's testable hypotheses, develop a research design for empirical evaluation of those hypotheses, and conduct empirical tests of the model in selected national settings. Their plans are animated by the conviction that the key to fostering scientific progress in this area (and others) is to bridge the chasm between formal and empirical analyses.

The co-PIs have successfully recruited ten prominent political scientists in order to establish a Research Work Group on Party Switching. Group members will propose and discuss refinements or alternatives to the co-PIs' model, hypotheses, and research design. They also will assess the empirical plausibility of the model, examining evidence from a wide range of settings, including Brazil, Eastern Europe, Russia, Italy, Spain, the European Parliament, the United States, and Japan. The Work Group will meet twice, in summer 2004 and summer 2005. The co-PIs will make publicly available the Group's datasets on the party affiliations of legislators and candidates as well as seek to arrange publication of the Group's papers in a special issue of a refereed journal.

The co-PIs and the Group will contribute to: (1) the dissemination of scholarly results (via publication of journal articles and sharing of databases); (2) the enhancement of research infrastructure (via scholarly collaboration across national boundaries); (3) the teaching and training of graduate students (discussion with graduate students will be an integral part of the Group's meetings); and (4) the broadening of participation of underrepresented groups (women constitute one-third of the Research Work Group). We have no plans directly to undertake undergraduate curriculum development as part of this project, but we believe it can and should have (5) substantial indirect impact on undergraduate education. For example, the flurry of media attention to the causes and consequences of Jeffords's move made it clear that party switching is a puzzle. This project should help scholars better understand party switching as a general phenomenon. They then can better teach their students why, when, and how choices of party affiliation and, by implication, parties matter.

More information at www.virginia.edu

Project Sponsored By: U.S. National Science Foundation
Start Date: 5/1/2004 - End Date: 8/31/2007

I am Carol A Mershon and I would like to this information.