New Supreme Court Database

A new searchable database of Supreme Court opinions and statistics came online last month.  With National Science Foundation funding, longtime Court scholar Harold Spaeth is teaming up with five other law and political science professors  to preserve and expand data he has collected annually since the 1980s.  The website is supremecourtdatabase.org.

The new website features both downloadable datasets about all Court cases since 1953 and interactive search tools that allow users to design and run their own statistical analysis or look up data about individual cases by case name or citation.  Set to the right parameters, the Database churns out answers for questions like “Which decisions in the 1980s involved school desegregation?” or “In how many criminal law cases did Justice Brennan and Chief Justice Burger vote the same way?”

Earlier I talked to Professor Andrew Martin of the Washington University in St. Louis School of Law, one of the architects of the database.  The information for this post comes largely from him.

The source of the data

The project started with Spaeth’s original database.  Spaeth has collected data on Court decisions since the 1980s, and he continues to release statistics for each new Term.  The new Database team wants to preserve his method and his data, which is widely used not only by political scientists and law professors but also – according to the project website – by “virtually all systematic analyses of the contemporary Supreme Court and its members.” The online interface is the culmination of eighteen months of work by Spaeth and five other political science and law professors, Lee Epstein, Ted Ruger, Keith Whittington, Jeffrey Segal, and Martin, as well as graduate students from various universities.

The online Database improves on the original Spaeth database in several ways.  First, the data is accessible to a broader audience.  The original database is hard to use without advanced statistical software training.  By adding a search by case name and removing specialized computer programming language, the Database allows users to analyze data with even basic statistical knowledge.  The downloadable empirical datasets, which remain fairly technical, are still intended primarily for political scientists, but the search tools could be used by other researchers of the Court, such as journalists and undergraduates.  Second, the new website backdates the data to 1953.

Spaeth continues to make all the tough judgment calls for the more qualitative data, like justice alignment and issues decided.  The Documentation section of the project website carefully lays out the criteria Spaeth uses to make decisions for all variables.  According to Professor Martin, the method gets high marks in reliability tests – meaning, in layman terms, that other coders using the same method consistently get the same results.  The familiar search engine Oyez already uses some of the Spaeth data.

The future of the project

Ruger and Whittinger are leading an effort to expand the database’s coverage to all cases the Court has decided since 1792, beginning with the earlier cases.  Users can expect full historical coverage by 2013.

The creators also envision the website as a platform for other researchers to share Court-related data.  For example, Martin mentioned a project by Court librarian Jill Duffy and Elizabeth Lambert on oral dissents as one that he is working to incorporate into the project.  Martin also hopes to integrate with other online databases that collect different kinds of information about decisions, such as Cornell Law School’s database of Court opinion texts.  The “pie in the sky” goal, according to Martin, is to enable Boolean searching, which would allow the type of cross-searching featured on Lexis Nexis and Westlaw.

Feedback and other ideas are welcome.  Each page on the website features a “contact” button, and Martin encourages people to use it!

Martin also appears in a Wall Street Journal story, linked in today’s SCOTUSblog media round-up, about a new computer statistics method that compares justices’ opinions across time.