You are here :: Home » Research Initiatives » Current Research » Primary Elections for U.S. Office

Current Research at ISERP

Primary Elections for U.S. State and Federal Offices, 1900-2004

by Shigeo Hirano (Political Science)

The emergence of the primary election process during the 20th century is widely thought to have transformed American politics. For example, primary elections allegedly weakened political parties and fueled the rise of personalistic, incumbency oriented politics; and primary elections are thought to have made politicians much more ideologically extreme than the electorate as a whole, because candidates must first win approval of a partisan subset of the electorate before they can even qualify for the general election.

Addressing even the most basic questions about this two-stage process is difficult because no comprehensive database on primary election results in the U.S. currently exists. This project will assemble a database on primary elections for statewide and federal offices. For each state, the investigators will collect election returns for every statewide and federal office for major-party primary elections from the year the first primaries were held to the present. For a subset of states, they will collect data on newspaper endorsements and paid campaign advertisements for primary election candidates.

This project builds on the investigators' existing research into general election results in the American states, which covers a similar time frame. Ultimately, the databases produced from these two projects will be linked allowing scholars to study fully the two-stage election process in the United States. Their interests in collecting these data are three-fold. (1) Measuring Competition in a Two-Stage Electoral System. These data will allow them to gauge the overall competitiveness of elections across states and over time, and the importance of primaries in producing this competition. The data will also help assess whether primary competition leads politicians to respond more to their primary electorates than to their constituency as a whole or party leaders in the legislature. (2) Understanding the Rise of the Personal Vote and the Incumbency Advantage. These data will allow for the estimation of the magnitude and sources of personal voting in primary elections and to determine whether the competitiveness of primaries or the emergence of personal voting in primaries contributed to the increase in incumbency advantages or to the decline of parties. (3) Understanding Internal Party Politics. The county-level and town-level primary election results will enable the study of factionalism within state political parties, to determine whether there are regular divisions within party electorates that reject regionalism, ideology, local party machines, or personal rivalries. What is the political landscape within the Democratic and Republican parties in each of the states, and how has it changed over the last 100 years?