Blog Post: How Liberal was Barack Obama as a State Senator?
Stronger Bridges: Putting Congress and State Legislatures in Common Ideological Space
Working Paper
Abstract: In an earlier paper, I addressed the problem of putting state legislatures on a cross-institutional common space with Congress. I relied on new roll call data from these legislatures, and a bridging technique involving state legislators who served in Congress later in their careers. This bridging approach is inherently limited because the supply of those who “graduate” is inherently limited by state size, the incumbency advantage, and the limited extent of the state roll call data. I adopt a new approach, employing the Votesmart questionnaire administered to campaigning politicians every election year. The number of respondents and the broad range of questions asked allow me to treat respondents as bridge actors. I expand the number of states in common space to all 50. This new data allows me to answer important questions about polarization, policy output, and party strength.
All Together Now: Putting Congress, State Legislatures, and Individuals in a Common Ideological Space
Working Paper
Abstract: The main tool measuring ideology in public opinion has historically been self-reported. But this is now well known to be severely plagued by measurement error. While Ansolabehere et al (2008) correctly address this problem with multiple measures of issue preferences, a major problem remains. How can we put representatives and the public on a common scale? While a common space exists for politicians’ ideal points, no such space exists that incorporates voters. The solution lies in bridging; the key is finding some survey which asks the exact same questions of politicians and voters, and then to use these questions as “glue” in estimating comparable ideal points for both types. I locate this in the Project Votesmart questionnaire which has been administered to members of Congress and state legislatures since the early 1990s. With its broad range of policy preference questions, it is extremely amenable to ideal point estimation. I administer a survey consisting of the exact same questions to 3,400 respondents, representative at the state level. Thus, I am able to put state voters, state legislators, and members of Congress on the same ideological scale. Furthermore, I employ the new common space scores in vote choice models for 2008, and find that my measures of ideology compare favorably to alternatives, and similarly to party identification.
A Bridge to Somewhere: Mapping State and Congressional Ideology on a Cross-Institutional Common Space
(with Nolan McCarty and Christopher Berry)
Revise and Resubmit at Legislative Studies Quarterly
Abstract: Two major problems exist in applying ideal point estimation techniques to state legislatures. First, there has been a scarcity of available longitudinal roll call data. Second, even where such data exists, scaling ideal points within a single state suffers from a basic defect. No comparisons can be made across institutions, whether to other state legislatures or to the US Congress. Our project is a solution to both of these problems. We use a new comparative data set of state legislative roll calls beginning in the mid-1990s to generate ideal points for legislators. We take advantage of the fact that state legislators sometimes go on to serve in Congress to create a common ideological scale between Congress and the various legislatures. These “bridge actors” are similar in concept to members of the House who go on to serve in the Senate, thereby providing the “glue” necessary to scale the House and Senate together. We use this approach for California, Colorado, Florida, Illinois, Michigan, Missouri, Ohio, Pennsylvania, and Texas. Using these bridge actors, we create a new state-federal congressional common space ideological scores. We conclude by using these common space scores to address important topics in legislative politics.
Methodological Issues in Bridging Ideal Points in Disparate Institutions in a Data Sparse Environment
(with Nolan McCarty and Christopher Berry)
Working Paper
Abstract: In earlier work, we created Congressional common space scores for multiple state legislatures using bridge actors who served in both institutions. Here, we employ simulations to explore the general issues involved in bridging institutions in data-sparse environments, where only a few bridge actors exist to allow inter-institutional comparisons. We find that only a few such bridges are necessary to improve ideal point estimates of rescaled legislative chambers.
European National Parliaments in European Parliament Common Space
Abstract: Numerous attempts have been made to put European political parties on a common ideological scale, including content analysis of political statements, hand coding of party manifestos, and surveys of expert opinion. At the same time, there have been numerous attempts at using roll call-based ideal point estimation techniques on national and transnational parliaments. These have the satisfying property of using newly available roll call data, which due to their presumed consequentiality may reveal latent preferences better than other methods. However, we cannot compare within-state parliament estimates to those of other countries or the European Parliament. The solution to the problem lies in the existence of bridge actors–in this case, substantial numbers of national parliamentarians who also serve in the EP before and after their national service. Their revealed preferences in the form of roll call votes on the European Parliament’s agenda allow us to rescale solely national members of parliament with a common space score. I test this approach with roll call data on several national parliaments and all sessions of the EP. The common space estimates allow me to explore important questions in comparative politics.
