The State of Disunion: Regional Sources of Modern American Partisanship

APPENDIX A. RESEARCH METHOD AND CASE SELECTION

APPENDIX A RESEARCH METHOD AND CASE SELECTION THE RESEARCH in this book involves actors at virtually all levels of the party system : presidents, congressional representatives, interest groups, and voters. The analytical focus, however, is on the House of Representatives. Studies of party conflict typically focus on the House because, of all of the national institutions , the House most closely approximates the general electorate. It is thus safe to assume that regional diÅerences observed in the House represent regional diÅerences in the electorate.1 The analysis in the book relies primarily on information from House roll calls and debates. Most existing studies of congressional party conflict rely heavily on aggregate data analysis. This book, instead, uses the vehicle of historical case studies of policy conflict in order to examine in greater depth the causal linkages between party conflict outcomes and the factors propelling these outcomes. Narrowing the lens to focus on conflict in specific issue areas (disaggregating) introduces a degree of control in that it reduces, though by no means eliminates , the uncertainty associated with aggregate analyses.2 A case study does not an argument make, however. The votes in any one area, in all likelihood, do not constitute a particularly large proportion of congressional votes; and therefore, explaining this pattern of conflict provides only partial leverage on the question of party conflict more generally. Moreover, the politics of trade are diÅerent from the politics of abortion. Because of this, cases were selected to represent the range of foreign, economic, and social policy issues over which the parties have fought in recent decades. Trade, welfare, and abortion policy cases serve as proxies for larger groups of policies.3 Trade policy resembles other economic regulatory policy. Like other economic policies, it has explicit distributional consequences and invites competition between economic groups and associations. For the case of welfare, I focus specifically on the now defunct program Aid to Families with Dependent Children (AFDC). This is a proxy for the general category of social welfare policy . Many of the more general means-tested social welfare services and provisions were, in fact, linked programmatically to AFDC. As a case study, AFDC combines the distributional politics of regions with their racial politics. The dominant scholarship on regions in recent decades has treated regions as proxies for political economy divisions. While both trade and welfare have noneconomic dimensions to their regional politics (namely, foreign policy ideology in the former and racial politics in the latter), abortion is the case that demonstrates the cultural dimensions of regional politics most explicitly. It is the case that most clearly reveals regions to be more than just vehicles for political economy diÅerences but, rather, political entities with distinct material and cultural properties. The case of abortion fulfills another function as well. Since this policy was introduced into national political debate in the 1970s, in the middle of the period under study, it provides a useful illustration of how parties incorporate new issues into existing regional dynamics. Specifically, abortion reveals the potential for new issues to redefine the regional landscape in ways that have profound implications for party coalitions and interparty conflict dynamics. The extent to which an issue succeeds in transforming the party system depends in large part on how the parties construct and define the issue for their regional audiences. Within each of the policy cases, both quantitative and qualitative indicators demonstrate the role of geography in legislative party conflict. Roll call data illustrate party and regional diÅerences on the three issues, and statistical analyses establish geographical voting patterns. Statistical analyses of state-level data on economic, demographic, and party factors help explain variation in the geography of support for issues. These quantitative assessments are meaningful, however, because they are embedded in a historical account of the developmental paths of the regions as well as representatives’ substantive debates on the issues studied. The reAPPENDIX A 182 search here draws broadly on the Congressional Record, newspaper articles, political journals such as the CongressionalQuarterlyWeeklyReport and the NationalJournal, and government produced documents and data, such as that accumulated by the Census Bureau and other executive agencies. Information has also been culled from memoranda, correspondence, surveys, press statements, and internal party documents that exist in the archival holdings of Republican Party leader Robert Michel, Democratic leader Carl Albert, and Nixon aide...