Academic journal article
By Hurwitz, Mark S.; Kuersten, Ashlyn
Judicature , Vol. 96, No. 1
Continued examination of the docket of the Courts of Appeals, through use of the Songer Database, reveals trends regarding litigation participation and success and judicial behavior.
Goldman noted nearly a half century ago, "Voting behavior of public decision-makers has been of central concern for political scientists... However, the United States Courts of Appeals, second only to the Supreme Court in judicial importance, have been largely neglected."1 As the decades continued, other scholars ultimately took note of these courts, and there is now a relative wealth of knowledge and understanding of them.2
This path from scholarly neglect of these courts to our current comprehension of them was neither straightforward nor sudden. In fact, for an extended period after Goldman's observation, the Courts of Appeals remained generally unstudied, as the bulk of judicial research focused on the U.S. Supreme Court. Why these courts were not studied more extensively presents a bit of a puzzle. Congress deemed these courts to be sufficiently important in the judicial hierarchy, establishing the original circuit courts in the First Congress and the modern Courts of Appeals in 1891.3 Since then, the number of cases decided by these courts has increased tremendously, giving them "a pivotal position in our political system"4 as their actions "radiate throughout the judicial hierarchy."5 Because of the relative dearth of cases decided each year by the Supreme Court and the appeals courts' mandatory jurisdiction, the Courts of Appeals have become effectively the courts of last resort for the majority of cases.6 They are responsible for both error correction and, increasingly, national decision making since they also supervise the federal district courts.7
Notwithstanding their importance, judicial scholars collectively ignored the Courts of Appeals with the exception of a few notable pioneers.8 In their ground-breaking studies, these scholars had to explain convincingly the importance of these courts within the judicial system, as well as go through the time-consuming process of collecting data on the various circuit courts. By the 1990s, however, these early scholars had persuaded the judicial community of the importance of these courts; consequently, efforts began to coordinate the expensive and laborintensive process of collecting data on all 12 circuits, following Harold Spaeth's protocol for collection of data on the U.S. Supreme Court (hereafter, the "Spaeth Supreme Court Database").9 Certainly, the most comprehensive undertaking of data collection on the appeals courts was the database assembled by Donald Songer.
The Songer Database
Beginning in 1988, with the support from the National Science Foundation and a Board of Overseers, Songer produced a multi-user database of approximately 20,000 published decisions on the Courts of Appeals (hereafter, the "Songer Database").10 Drawing on a probability sample of cases decided from 1925 to 1996 (15 cases per circuit/year from 1925 - 1960 and 30 cases per circuit/year from 19611996), information was coded for 229 variables divided into three major categories. The first grouping of cases in the Database, referred to in the documentation as "basic coding," includes several fundamental variables that provide descriptive general information about each decision in the sample. The Songer Database also provides 36 variables devoted to characterizing the parties or litigants involved in the decision, referred to as "parties coding." In addition to recording the names of the first five listed appellants and respondents for each side, the litigants also are classified into categories. This categorization initially distinguishes between natural persons, private business, nonprofit organizations, federal government/agencies, state governments/agencies, local governments, and fiduciaries/trustees.11
The third set of variables in the Songer Database is designed to capture the issue content of the judicial decision. …