Legislative Organization and Ideal-Point Bias

43 Pages Posted: 21 Nov 2012 Last revised: 29 Jul 2014

See all articles by Keith Krehbiel

Keith Krehbiel

Stanford Graduate School of Business

Zachary Peskowitz

Independent

Date Written: November 8, 2013

Abstract

Four pure types of legislative organization are characterized as data generating processes for commonly used measures of preferences or, in the spatial vernacular, ideal points. The types of legislative organization are differentiated by their partisan versus nonpartisan nature of agenda formation, and by whether the amendment process is open or closed. For each organization, roll call voting data are Monte-Carlo generated and used as input for four different ideal point measures: standard percent-correct interest group ratings (IGRs), linear factor analysis scores (LFAs), W-NOMINATE ratings (NOMs), and Markov chain Monte Carlo measures (MCMCs). Three questions motivate and are addressed the analysis. Do estimated ideal points differ significantly across forms of legislative organization? Are some ideal point estimates consistently more accurate than others? Are there patterns of substantively relevant, persistent bias in ideal point estimates? The answers are all affirmative.

Suggested Citation

Krehbiel, Keith and Peskowitz, Zachary, Legislative Organization and Ideal-Point Bias (November 8, 2013). Stanford University Graduate School of Business Research Paper No. 14-28. Available at SSRN: https://ssrn.com/abstract=2178654 or http://dx.doi.org/10.2139/ssrn.2178654

Keith Krehbiel (Contact Author)

Stanford Graduate School of Business ( email )

655 Knight Way
Stanford, CA 94305-5015
United States

Zachary Peskowitz

Independent ( email )

No Address Available
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
60
Abstract Views
501
rank
384,509
PlumX Metrics