A Prefatory Note from Harold J. Spaeth
The database to which this introduction pertains spans all four centuries of the Court's decisions, from its first decision in 1792 to the Court's most recent decision.
The initial version of this database, which began with the Warren Court in 1953 and was back dated to te beginning of the Vinson Court in 1946, dates from the mid-1980's at the dawn of the desktop computing revolution and relies on pre-microcomputing and pre-internet conditions. As such, users need knowledge of statistical software packages and the codified variables that the database contains. This new version, however, recognizes the existence of the 21st century by eliminating acquaintance with statistical software packages and coded variables. Plain English rules! But do note that the database can be uploaded into statistical packages to perform advanced calculations if a user so desires.
Aside from the foregoing, the major feature of this version of the database is an interface that is in line with modern technology and which will allow users to directly calculate and view relationships among the variables in the database.
As such, the database may now be treated as justice centered. The original database only allowed for the analyses of judicial decisions and the votes of the individual justices. It is now possible for the individual justice's vote to be the unit of analysis rather than the case. That is, a user may, for example, easily compare the behavior of one or more of the justices with that of others. The original version of the database was not programmed to do so because it was exclusively case centered.
I have specified decision rules governing the entry of data into the various variables, most particularly the legal provisions governing the Court's decisions and the issues to which cases pertain. These, however, are not set in concrete. You, of course, are free to redefine any and all variables on your copy of the database. If convention applies, I adhere to it. But for many variables and their specific entries, none exists.
Because the database now extends over four centuries, it is necessary to add, alter, and adjust a number of variables. I do so to keep the legacy cases (those decided between 1792 and the Court's acquisition of discretionary jurisdiction as a result of the Judges' Bill of 1925) as congruent as possible with the Court's modern decisions. These changes primarily apply to the issues the Court decides. Most notable is the addition of a set of common law issues. These account for the bulk of the Court's heritage decisions especially those of the 19th century and have little applicability to any but the parties to these cases.
In specifying the issue in the legacy cases I have chosen the one that best accords with what today's Court would consider the issue to be. For example, "prize cases," those in which vessels were captured on the high seas and brought into U.S. ports, are categorized either as Fifth Amendment takings clause cases or as cases pertaining to the jurisdiction of the federal district or appeals courts, depending on which issue the Court based its decision.This was done to provide a basis for continuity in the Court's decision making and to avoid undue segmentation of the Court's decisions. The same rule applies to various provisions pertaining to the Bill of Rights even though the Fourteenth Amendment had not been ratified and no guarantees of the Bill of Rights had been made binding on the state and local governments.
Do recognize, however, that the foregoing paragraph applies only to the issue(s) the Court addressed and not to the legal provisions decided by the Court. The latter were nonexistent at the time of decision. The early legacy decisions generally rested either on the common law or judicial fiat.
Because of current concerns I have given primacy to issues involving women and native Americans in cases in which they are involved.
I wish to thank Professor Jeffrey Segal of the State University of New York at Stony Brook for his extremely valuable comments and suggestions on all phases and aspects of the database since its creation. I also thank Harriet Dhanak, the former programming and software specialist in the Department of Political Science at Michigan State University, for her expert guidance and assistance. Her successor, Lawrence Kestenbaum, continued and extended the stellar services on which I had become dependent. Most recently I have relied on the superb technical knowledge and skills of John Schwarz of the Michigan State University Institute for Political and Social Science Research and his successor, Jess Sprague. Professor Tim Hagle of the University of Iowa continues to systematically inform me of errors and missing data that I have overlooked. My former graduate students, now well tenured scholars--Sara C. Benesh and Wendy L. Martinek--have shepherded me through the more arcane byways of current versions of statistical software packages. And though this feature of the database is now passe, their previous assistance has been key.
I also deeply appreciate the support provided me by the Michigan State University College of Law and that of Millie Shiraev of the University's Institute for Political and Social Science Research.
Three outstanding individuals are most responsible for this version of the database. Lee Epstein, whose wide-ranging scholarly productivity is unmatched in the world of judicial scholarship; Andrew D. Martin, chair of the Department of Political Science, professor of law, and Director of the Center for Empirical Research in Law (CERL) at Washington University in St. Louis, whose methodological competence knows no bounds; and Troy DeArmitt, CERL's masterful research technician par excellence.
Compilation of this database has been supported by grants from the National Science Foundation. Without its assistance, the database would not exist.
Notes to All Users
1. The Supreme Court Database's research team continuously updates the database. Accordingly, we urge you to pay attention to the date your version appeared on the website and to check whether it is the current one.
2. The codebook now provides five pieces of information for each variable: the name of the variable as it appears in the current version of the Database, the name Spaeth used in previous versions (if applicable), any normalization (changes we made when converting from Spaeth's format to the new web version), and, of course, a description of the variable and a list of its values.
Voting & Opinion Variables