TRAC assembled data from a number of sources for this study, publicly available information as well internal data obtained by TRAC and others from the agency under the Freedom of Information Act.
The Social Security Administration and its Office of Disability Adjudication and Review (ODAR) should be commended for their recent moves to provide greater public transparency through posting data each month on their website beginning in 2010 on ODAR's operations. The tabular information includes details on the cases disposed of by each administrative law judge. Additional data at the hearing office level summarizes other useful indicators including receipts, dispositions, cases pending, and processing times in each office (see http://www.ssa.gov/appeals/publicusefiles.html). These data provided the principal source for the individual judge records used in assessing decision disparity during the FY 2010-2011 period.
We are also indebted to Bryan Denson and Brent Walth, reporters for The Oregonian newspaper, who agreed to share the data for the FY 2005-2008 period they had earlier obtained after lengthy Freedom of Information efforts to convince the Social Security Administration to release these judge-by-judge data. The Oregonian published a series of investigative articles based on these data in 2008. We restricted our focus to the records covering just the first two years (FY 2005-2006) from this FY 2005-2008 data so that we could assess disparity levels before the launch of the agency's 2007 initiatives. We also compared the Oregonian data with that from delawareonline.com which also independently obtained judge-by-judge data under FOIA from the agency for the FY 2005-2008 period.
Additional information helped us ensure that year-to-year variations in how the names of individual judges were recorded in these records were made consistent so that the same judge's record for each year could be properly matched, and that the hearing office the ALJ was assigned to was correctly identified and standardized.
We further checked to see how the number of decisions recorded for each judge in these separate data series compared for similar periods, and also then compared the counts derived against internal agency records which we independently obtained under the Freedom of Information Act. These multiple sources allowed us to perform a number of cross-validation checks to assure ourselves as to the quality and coverage of the data we were using for both the FY 2010-2011 and the FY 2005-2006 periods.
In addition to these data, we gathered information through conversations with individuals inside and outside of government about the court's operations, and also reviewed scholarly literature and the extensive array of congressional and other government reports, including those of the Social Security Administration itself and an informative series of General Accountability Office reports going back more than thirty years noting numerous problems in the operations of this body.
The early stages of TRAC's project began several years ago. Support for work during this past year which culminated in the publication of this two-part series was provided by Syracuse University, the Park Foundation, and the Rockefeller Family Fund.