About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline



14 J. Experimental Criminology 1 (2018)

handle is hein.journals/jexpcrm14 and id is 1 raw text is: J Exp Criminol (2018) 14:1-17                                         CrossMark
https://doi.org/10.1007/s11292-017-9309-z
Opting out of treatment: Self-selection bias
in a randomized controlled study of a focused deterrence
notification meeting
Benjamin Hamilton' - Richard Rosenfeld'-
Aaron Levin'
Published online: 19 October 2017
O Springer Science+Business Media B.V 2017
Abstract
Objectives This study investigated the role of self-selection in an evaluation of the
impact of a focused deterrence notification meeting on subsequent arrests.
Methods We conducted a randomized controlled study that randomly assigned proba-
tioners and parolees to a treatment group asked to attend a focused deterrence notification
meeting and a control group that was not asked to attend the meeting. A sizable proportion
of the treatment group did not attend the meeting. We estimated intent-to-treat, average
treatment, and local average treatment models to evaluate the effect of attending the
notification meeting on future arrests and the effect of self-selection on the results.
Results Subjects who attended the notification meeting were less likely than those who
did not receive treatment to be arrested over the following 17 months. The results were
not significantly affected by selection effects.
Conclusions Future evaluations of focused deterrence and related criminal justice
interventions should be based on randomized controlled research designs that address
selection effects on the outcome.
Keywords Focused deterrence - Notification meeting - Randomized controlled
experiment - Selection effects
The deterrence doctrine in criminology has undergone an important change in recent years
with the addition of the concept and practice of focused deterrence. Focused deterrence
condenses the deterrence message in two ways, first by targeting high-risk criminal
offenders and then by delivering the message in face-to-face meetings with them
W Richard Rosenfeld
richard rosenfeld@umsl.edu
University of Missouri-St. Louis, St. Louis, MO, USA

4_ Springer