Statistical Analysis of Presidential Pardons

BJS' FY 2012 Statistical Examination of Presidential Pardons was issued to seek applications for an examination of the processing of petitions for Presidential Pardons. In the federal justice system, commutation of sentence and pardon are different forms of executive clemency, which is a broad term that applies to the President's constitutional power to exercise leniency toward persons who have committed federal crimes. The purpose of this program is to examine how petitions for pardon are adjudicated by the Justice Department's Office of the Pardon Attorney. The Pardon Attorney reviews and investigates pardon applications, and prepares the Department's recommendation to the President.

Through this award, the Rand Corporation will carry out the envisioned statistical examination involving the following aspects:

(1) The development of a conceptual model of the Office of the Pardon Attorney (OPA) deliberative process. The assumption is that this process is composed of one or more stages of decision making that begin or end with a decision to finalize a recommendation or continue the investigative process. By necessity, it will be important for the model to identify what information is available at each decision making point in the deliberative processes.

(2) The identification of the pardon applications included in the statistical examination and the access to the relevant OPA files for review. The population of petitions that this study speaks to includes all petitions that reached final disposition, (i.e. were granted, denied or closed) between October 1, 2001 and April 30, 2012. During this period approximately 211 pardon petitions were granted, 2,748 were denied and 834 were closed without action. This statistical examination will include all pardon petitions that were granted and a stratified random sample of approximately 1,000 petitions that were either denied or closed without Presidential action.

(3) The development and testing of a coding scheme for the information to be obtained from the Pardon Attorneys' files that address the five principal factors taken into account by the Pardon Attorney. Available demographic and case information will also be coded which, at a minimum, will include: petitioner age, race, ethnicity, gender, the crime involved, and sentence imposed. The coded information will be entered into a data file created by the successful applicant and statistical models will be used to understand the pardon decision process.

(4) The models are to be used by the successful applicant to test the primary hypothesis that all other things being equal African Americans and other minorities are less likely to progress in the pardon adjudication process than applicants of other races. Applicants may propose one or more secondary hypotheses for testing with clear and convincing justification.

(5) The development of a publishable quality final report describing the methodology, data collection, coding, modeling, statistical analyses, and hypothesis testing.

(6) The project will be guided by BJS and a group of knowledgeable advisors designated by the Office of the Pardon Attorney through regular (at least bi-weekly) meetings and discussions. The successful applicant will be expected to convene this steering committee, and prepare relevant meeting materials including general minutes that document discussions and decisions.
(ca/ncf)