Table of Contents

Words that Matter: Predicting Persuasive Briefs in the United States Supreme Court

Adam Feldman, University of Southern California, Political Science
Elli Menounou, University of Southern California, Department of Political Science


POLITICAL METHODS: COMPUTATIONAL eJOURNAL

"Words that Matter: Predicting Persuasive Briefs in the United States Supreme Court" Free Download

ADAM FELDMAN, University of Southern California, Political Science
Email:
ELLI MENOUNOU, University of Southern California, Department of Political Science
Email:

In this paper, we generate predictive models of judicial voting using the language from litigants' briefs. Specifically, we use quantitative text classification to generate models of Supreme Court Justices' voting behavior in First Amendment cases. We then propose a multidimensional scaling procedure to assess the justices' preferences through language-based modeling. We utilize a dataset composed of five justices' votes - Justices Scalia, Rehnquist, O'Connor, Stevens, and Kennedy, over a seventeen term period from 1987 through 2004. With this data, we show how focusing on inputs to judicial decision-making such as litigants' briefs necessitates a new means of assessing judicial attitudes. In particular, we show that with the addition of a second dimension of voting in the First Amendment context with a rights/restrictions axis, we can better understand the justices' preferences. The models in this paper achieve near 80% predictive accuracy of the justices' votes which far surpasses the prediction accuracy of models based on ideology scores commonly used in judicial behavior literature.

^top

About this eJournal

This eJournal distributes working and accepted paper abstracts. Papers in this area study the formation, structure and function of networks or apply computational methods or algorithmic game theory to understanding politics.

Submissions

To submit your research to SSRN, sign in to the SSRN User HeadQuarters, click the My Papers link on left menu and then the Start New Submission button at top of page.

Distribution Services

If your organization is interested in increasing readership for its research by starting a Research Paper Series, or sponsoring a Subject Matter eJournal, please email: RPS@SSRN.com

Distributed by

Political Science Network (PSN), a division of Social Science Electronic Publishing (SSEP) and Social Science Research Network (SSRN)

Directors

POLITICAL METHODS EJOURNALS

DAVID A. LAKE
UC San Diego
Email: dlake@ucsd.edu

MATHEW D. MCCUBBINS
Duke University
Email: mathew.mccubbins@duke.edu

Please contact us at the above addresses with your comments, questions or suggestions for PSN-Sub.