Data and code to replicate the results presented in this article are available at http://dvn.iq.harvard.edu/dvn/dv/mkellermann. Earlier versions of this article were presented at the 2009 Annual Meetings of the Midwest Political Science Association and the Harvard Applied Statistics Workshop. Thanks to Kevin Quinn, Ken Shepsle, Simon Jackman, Olivia Lau, Dan Hopkins, Ian Yohai, Ben Goodrich, Justin Grimmer, David Morgan, seminar participants, and three anonymous referees for helpful comments. This article was revised with support from a Naval Academy Research Council grant. This article does not express the views of or endorsement by the United States Naval Academy or the United States government.
Estimating Ideal Points in the British House of Commons Using Early Day Motions
Version of Record online: 22 MAR 2012
© 2012, Midwest Political Science Association
American Journal of Political Science
Volume 56, Issue 3, pages 757–771, July 2012
How to Cite
Kellermann, M. (2012), Estimating Ideal Points in the British House of Commons Using Early Day Motions. American Journal of Political Science, 56: 757–771. doi: 10.1111/j.1540-5907.2012.00587.x
- Issue online: 16 JUL 2012
- Version of Record online: 22 MAR 2012
This article develops a new method for estimating the ideological preferences of members of the British House of Commons. Existing methods produce implausible results due to high levels of party cohesion and strategic voting on the part of opposition parties. To circumvent these problems, this article estimates MP preferences using Early Day Motions (EDMs) as an alternative to roll-call votes. The Bayesian ideal point model for the decision to sign an EDM takes into account both policy preferences and signing costs. The estimates obtained have greater face validity than previous attempts to measure preferences in the House of Commons, recovering the expected order of parties and of members within parties. The estimates successfully predict voting behavior in the House of Commons. As with other Bayesian ideal point methods, this approach produces natural uncertainty estimates and allows for easy calculation of quantities of interest such as member ranks.