On the Perils of Drawing Inferences About Supreme Court Justices From Their First Few Years of Service

Published in 2008. In Judicature 91: 168-179.
Winner of an “Exemplary Legal Writing” Honor, Long Article category. Presented by Green Bag, 2008.

Lee Epstein
Andrew D. Martin
Kevin M. Quinn
Jeffrey A. Segal

Is the behavior of a Supreme Court justice in the first few terms of a service a reliable indicator of future voting behavior? Despite decades of study, judicial specialists cannot offer a conclusive answer.

In what follows, we hope to bring a fresh eye to this seemingly age-old but nonetheless on-going debate. Deploying Martin and Quinn's (2002) measurement model, we first contemplate whether we can reach high-quality inferences about the justices' long-term ideological preferences based on one-year's---the first-year's---worth of observations. Finding that the vast majority of justices exhibit statistically significant ideological movement from their initial preferences, we turn to questions of substantive importance. We demonstrate that drift away from first-year ideal points occasionally manifests itself in consequential doctrinal change---so consequential that a vote in favor of, say, restricting privacy rights or permitting prayer in school in the first term might translate into a vote in opposition before the justice concludes a decade of service.

Click here for the page proofs.
Click here for the data (Stata .dta file) (data also available on Andrew D. Martin's website). Note: Martin and Quinn recommend using the posterior mean (post_mn) as the estimate the ideal point of each justice in each term.
Click here for the website of related article, Ideological Drift Among Supreme Court Justice.