It is my great pleasure to be a guest blogger on the ELS website. Kudos and thanks to Jason and his colleagues for establishing this terrific site, chock full as it is with interesting information for all of us interested in ELS.
In this first entry, I thought I'd address a question that may be of interest to many readers about the Administrative Office Database, now archived in computer readable format by the Federal Judicial Center and available through the ICPSR. Ted Eisenberg has noted that some have called the AO database "'by far the most prominent' database used by legal researchers for statistical analysis of case outcomes" (78 Notre Dame L.R. at 1458). One frustrating thing about the AO database, as we all know, is that it has the judge ID's expurged prior to its distribution to the public. My understanding of this situation, learned during my stint at the FJC, is that the decision to release the data without the judge identification information was initially rendered by Chief Justice Rehnquist in consultation with the AO. The concern was that statistical analysis of individual judges' decisions would threaten judicial independence and possibly even the safety of judges as well. Nevertheless, this decision led to the paradoxical situation in which the NSF ultimately spent hundreds of thousands of dollars to fund the Songer Database (the U.S. Courts of Appeals Database)--although no similarly publicly funded database for the District Courts has yet to be developed. Although the Songer Database includes information not available in the AO data, it would have been far simpler for Don to have had access to the judge codes as an important starting point in the development of his database.
At this point, it seems the time might be right for the scholarly community to petition the federal judiciary to release the judge codes based on several prevailing factors. First, we have a new Chief Justice who may be more willing to consider such a request. Second, sophisticated computer programs now exist to gather vast amounts of information directly from Westlaw, including the identity of judges rendering the decisions. These computer programs have been used, for example, to produce complex network analyses of precedent based on thousands of cases in the Westlaw database (Jim Spriggs and his colleagues have initiated such analyses, for example). Third, the Songer data exists in the public domain with considerable information about the decisional patterns of individual judges readily available. Fourth, the ludicrous situation in which the federal government gives with one hand while taking away with the other--at a high cost to taxpayers--might be a compelling argument in support of such a request. Finally, from a scholarly viewpoint, analyses of case level decision making without controls for ideology make little sense given the state of our knowledge about attitudinal factors in judicial decision making. Moreover, the AO data includes information about unpublished opinions--presently not available in the Songer database.
One problem, of course, is that the federal judiciary is exempt from FOIA. Thus, the only way to obtain the information is for us to petition the AO to release it. The Executive Committee of the APSA Law and Courts Section might be a place to start in the development of such a request, but I welcome suggestions from readers, who might have other insights into this issue.