Citation

Labriola M, Rempel M, Cissner A. J. Exp. Criminol. 2010; 6(4): 447-473.

Copyright

(Copyright © 2010, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s11292-010-9102-8

PMID

unavailable

Abstract

Randomized trials represent the most rigorous type of research design to measure the impact of a social policy intervention. However, such designs are difficult to implement and require the consent of multiple parties, including researchers and non-researchers. Unique challenges arise when seeking to implement such a design in a criminal court setting, due to the need to revise legal procedures, uphold due process for defendants, and obtain the direct, ongoing participation of judges and attorneys, among other stakeholders. The current principals recently conducted two randomized trials concerning the court response to intimate partner violence: one testing the impact of court-ordered batterer programs in the Bronx, New York, and a second testing the impact of intensive judicial monitoring in Rochester, New York. Key lessons involved forging collaborative stakeholder relationships, critically assessing the experimental intervention and its contrast with the control condition, ensuring legal due process for defendants, addressing victim safety, setting realistic timetables, adopting a skeptical view towards estimates of study volume, and anticipating substantial variation from original design to final plan, especially in regards to randomization protocols and defendant eligibility criteria. These lessons may prove invaluable in informing future research in court-based and other social settings where random assignment is pursued.