Evaluating science outside the trial box: applying Daubert to the Federal Sentencing Guidelines' criminal history score.

A limited amount of research exists examining the ability of the Criminal History Score of the Federal Sentencing Guidelines (Guidelines) to achieve one of its most essential objectives: prediction of recidivism. Building on the work of Schopp [Schopp, R. (2001). Competency, condemnation, and commitment: An integrated theory of mental health law. Washington D.C.: American Psychological Association], it is suggested that the scientific admissibility framework and the underlying principles announced by the United States Supreme Court in Daubert v. Merrell Dow Pharmaceutical [Daubert v. Merrell Dow Pharmaceuticals Inc., 507 U.S. 579, 113 S.Ct. 2786 (1993)] should be expanded beyond the constraints of the evidentiary admissibility phase of trial and should apply to legislative and administrative rules that have: a) an empirically testable purpose and b) a substantial impact on the rights of individuals. Such an analysis offers a useful mechanism for understanding the strengths and weaknesses of social science being used by legal institutions. Based upon a hypothetical Daubert analysis, the scientific validity of the Guidelines' Criminal History Score is assessed and demonstrated to be insufficient. The law and policy implications of this finding are discussed.