Interest Groups and State Legislative Polarization

The Democratic and Republican parties in the US states have grown apart ideologically over the past several decades as the national parties have similarly diverged. Despite having implications for both governance and the overall health of US democracy, polarization in the states remains much less well-understood than polarization in Washington. This dissertation employs two sources of data relating to interest group activity on the state level—machine-learning classifications of groups based on their contributions to candidates for state legislative office and group ratings of individual legislators on different issues—to provide new information on the extent and nature of state legislative polarization and to assess the influence that organized groups exert on this process. In the first paper, I develop and implement a machine-learning model that utilizes campaign finance data to determine the degree to which groups behave in an ideological manner, showing that this approach results in stable classifications that align with theoretical expectations and finding that conventional methods substantially understate the amount of ideological group involvement in state legislative elections. Using the results of this model, in the second paper I demonstrate that ideological group activity in state elections is positively and robustly associated with polarization, providing evidence that interest groups have played a role in state legislative polarization over the past several decades. Finally, in the third paper I analyze group scorecards grading state legislators on their roll-call voting behavior to determine that although polarization has occurred in most issue areas, the parties have moved furthest apart most quickly on a cluster of issues relating to family roles and the place of religion in public life. Taken together, these papers advance our understanding of the nature and causes of partisan polarization in contemporary state legislatures.