A Randomized Control Trial of a Targeted High-Risk Offender Program Across Three Jurisdictions

• Author: Mariel Shutinya,Julie Schnobrich-Davis,Craig D. Uchida,Christine Connor,Marc Swatt,Daniel Wagner
• Published date: 01 June 2019
• Date: 01 June 2019
• DOI: http://doi.org/10.1177/1098611118807771
• Subject Matter: Articles

Article

A Randomized Control

Trial of a Targeted

High-Risk Offender

Program Across

Three Jurisdictions

Craig D. Uchida

1

, Marc Swatt

1

,

Julie Schnobrich-Davis

2

,

Christine Connor

1

, Mariel Shutinya

1

,

and Daniel Wagner

3

Abstract

This study reviews findings from Project Regional Analytics for the Safety of Our

Residents, a modified focused deterrence program operated across three jurisdic-

tions in Massachusetts. Unlike most other evaluations of targeted high-risk offender

programs, the impact on individual-level offending is examined. Data from records

management systems in three police agencies were extracted and used to create

social harm risk scores. The top 150 offenders were selected for inclusion in this

study. Subjects were blocked into three groups according to their composite score

and then randomly assigned into treatment and control groups. All treatment group

offenders were invited to a notification meeting where they decided whether to

participate in the program with a police and social worker case management

team. The outcome examined in this study is time to a new arr aignment.

Nonparametric and semiparametric methods detected no significant difference

between groups after approximately one year. Implications for future research and

practice are then discussed.

1

Justice & Security Strategies, Inc., Silver Spring, MD, USA

2

Central Connecticut State University, New Britain, CT, USA

3

Cambridge Police Department, Cambridge, MA, USA

Corresponding Author:

Marc Swatt, 6138 S 102nd Street Omaha, NE 68127, USA.

Email: marc.swatt@gmail.com

Police Quarterly

2019, Vol. 22(2) 192–216

!The Author(s) 2018

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1098611118807771

journals.sagepub.com/home/pqx

Keywords

focused deterrence, randomized experiment, targeted high-risk offenders

Introduction

Focused deterrence continues to be a popular intervention for jurisdictions

struggling with violent crime problems. Following the successes of Boston’s

Operation Ceasefire (Kennedy, 1997), a number of cities have employed the

“pulling levers” approach to address a number of crime problems, including

gang violence, drug markets, and substance abuse (see Braga & Weisburd,

2012a; Braga, Weisburd, & Turchan, 2018). A meta-analytic review of these

projects suggests that such focused deterrence approaches are associated with

statistically significant reductions in crime (Braga & Weisburd, 2012a; Braga

et al., 2018). Targeting high-risk offenders for crime reduction purposes contin-

ues to be a viable strategy for police organizations throughout the country.

While many of these programs have been successful, important gaps in the

research on these interventions remain. First, most of these studies have relied

on place-based assessments of the impact of these programs (see Lattimore

et al., 2016 for an exception). While many of these programs are oriented

around particular locations, the key theoretical mechanism behind focused

deterrence remains at the individual level (see discussion in Kennedy, 1997).

Because the pulling levers approaches involve increasing the probability of

detection and the severity of punishment associated with criminal activity (see

Kennedy, 1997), pulling levers should be associated with deterrent effects for

individual offenders. To date, individual-level research on focused deterrence

approaches remains limited.

This project takes a slightly modified version of focused deterrence and

implements the concepts of rational choice theory without the leverage that

traditional focused deterrence programs have used. The current program

called, Operation RASOR, which stands for Regional Analytics for the Safety

of Our Residents, seeks to increase the risks of getting caught and convicted

while also providing case management and services to targeted high-risk

offenders. While other studies used focused deterrence strategies in one jurisdic-

tion, this study implemented the program among three contiguous cities in

Massachusetts to focus on chronic cross-jurisdictional high-risk offenders.

Second, researchers found few of the evaluations of focused deterrence

approaches have involved randomized experimental designs (Braga &

Weisburd, 2012a; Braga et al., 2018). Quasi-experimental designs are often nec-

essary when randomization is either inappropriate or impossible. While these

designs offer some control over confounding variables (see Shadish, Cook, &

Campbell, 2002), only true experimental designs offer sufficient control over

Uchida et al. 193