In recent years, the legal academy has experienced a surge of interest in quantitative empirical analysis. Unfortunately, this enthusiasm has not always been accompanied by careful analysis of what the tools and resources of quantitative analysis can tell us about law and legal doctrine. As this Article demonstrates, the findings of some studies therefore unwittingly reflect the limitations of those tools and resources rather than providing insight into the workings of courts.
Specifically, this Article provides a long-overdue critical analysis of the most influential source of data about the Supreme Court, the Original Supreme Court Database, created by political scientist Harold J. Spaeth. The Database, which codes every opinion issued by the Supreme Court since 1953, contains coding for legal provisions considered by the court and for what Spaeth calls issue and issue area. Although numerous scholars - within both political science and law - rely on them, these codes do not report reliable information about the role that law and legal doctrine plays in the Supreme Court's cases. The Database does not reliably report the legal provisions or doctrines relied upon or at issue; it does not attempt to report legal issues at all, instead describing the "public policy context" of the case; and by design, it generally reports only one issue per case. These limitations have important, but poorly understood, implications for the many, many scholars who rely on the Database, and the Article describes a number of specific studies whose results are unreliable because of the way they use the Database.
This critique of the Database and the ways scholars use it can help scholars to be smarter and more accurate in their use of the Database. At the same time, the Article explores ways to incorporate law and legal doctrine into empirical legal scholarship. To further both goals, the Article presents the results of my Recoding Project of a random sample of recent Supreme Court cases. The findings of the Recoding Project confirm that significant information about law and doctrine is omitted from the databases. Furthermore, the findings suggest that the databases systematically underreport law and doctrine related to courts in particular and to the structure and operations of government in general - issues that may be very salient to the justices in at least some cases. By demonstrating what information is missing or misstated in the Database and by exploring ways to develop more comprehensive and law-focused coding protocols, this Article helps positive scholars - whether political scientists or legal academics - to consider how to take account of law. The Article concludes by discussing implications for future research.
Coding Complexity: Bringing Law to the Empirical Analysis of the Supreme Court,
Available at: https://scholarship.kentlaw.iit.edu/fac_schol/763