Scaling up effective juvenile delinquency programs by focusing on change levers: Evidence from a large meta‐analysis

• Published date: 01 May 2024
• Author: David B. Wilson,Mark W. Lipsey
• Date: 01 May 2024
• DOI: http://doi.org/10.1111/1745-9133.12663

DOI: ./- .

ORIGINAL ARTICLE

Scaling up effective juvenile delinquency

programs by focusing on change levers:

Evidence from a large meta-analysis

David B. Wilson1Mark W. Lipsey2

Department of Criminology, Lawand

Society, George Mason University,

Fairfax, Virginia, USA

Department of Human and

Organizational Development, Vanderbilt

University, Nashville, Tennessee,USA

Correspondence

David B. Wilson, Department of

Criminology, Lawand Society, George

Mason University  University Drive,

MS F, Fairfax, VA, USA.

Email: dwilsonb@gmu.edu

A prior version of this work was presented

as a keynote address at the Applied

Research in Crime and Justice Conference,

Sydney,Australia, on February , .

Additional supporting information can be

found online in the Supporting

Information section at the end of this

article.

Funding information

Creation and updating of the database

used in this study were supported by

grants from the Office of Juvenile Justice

and Delinquency Prevention

-JN-FX-, the National Institute

of Mental Health MH, MH,

Abstract

Research summary: The primary outcome desired for

juvenile delinquency programs is the cessation of delin-

quent and related problematic behaviors. However, this

outcome is almost always pursued by attempting to

change intermediate outcomes, such as family function-

ing, improved mental health, or peer relations. We can

conceptualize intermediate outcomes that are related

to reduced delinquency as change levers for effective

intervention. A large meta-analysis identified several

school-related change levers, including school engage-

ment (i.e., improved attendance and reduced truancy),

nondelinquent problem behaviors, and attitudes about

school and teachers. In addition, family functioning

and reducing substance use were also effective change

levers. In contrast, effects on youth getting/keeping a

job, peer relationships, and academic achievement were

not associated with reduced delinquency.

Policy implications: Only a small percentage of reha-

bilitative programs provided to youth involved in the

juvenile justice system have been established as evi-

dence based. Moreover, there are constraints on what

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits

use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or

adaptations are made.

©  The Authors. Criminology & Public Policypublished by Wiley Periodicals LLC on behalf of American Society of Criminology.

Criminology & Public Policy. ;:–. wileyonlinelibrary.com/journal/capp 261

262 WILSON  LIPSEY

andMH,theW.T.Grant

Foundation, and the Russell Sage

Founda tion.

local policy makers and practitioners can do regard-

ing the selection, adoption, and implementation of

programs from the available lists of evidence-based

programs. Adopting programs that focus on effective

change levers and avoiding those that concentrate on

ineffective ones has the potential to increase the likeli-

hood that a local agency is engaged in effective program-

ming. Based on our data, programs known to improve

family functioning, attachment to and involvement in

schooling, and reducing substance use are justified by

the change lever evidence, even if these programs’ effec-

tiveness in reducing delinquency has not been directly

proven. In contrast, programs focusing on vocational

skills, academic achievement, and peer relations are

less likely to be beneficial. Furthermore, a change lever

perspective can help frontline staff select appropriate

programs for different juvenile offenders and focus their

quality control efforts on those aspects of a program that

are likely to be essential to maintaining effectiveness.

KEYWORDS

change levers, delinquency, mediators, meta-analysis, program

effectiveness

Evidence-based programming for juvenile delinquency focuses on the effectiveness of specific

programs and program types. For example, the CrimeSolutions.gov (crimesolutions.ojp.gov)reg-

istry lists programs and practices of this sort. Programs are defined as “a specific set of activities

carried out according to guidelines to achieve a defined purpose,” and practices are defined as

“a general category of programs, strategies, or procedures that share similar characteristics with

regard to the issues they address and how they address them.” This paper takes a complementary

approach and focuses on the potential change mechanisms or intermediate outcomes through

which programs attempt to achieve the goal of reduced delinquency.

The only direct control of delinquency is incapacitation, which has significant costs, can make

matters worse, and is not appropriate for most youthful offenders. Most delinquency programs,

therefore, must work indirectly by changing something that, in turn, affects delinquency, for

example, by increasing or improving something desirable that is presumedto be negatively related

to delinquency,such as school attendance, interpersonal skills, prosocial attitudes, or family func-

tioning. One advantage to focusing on these intermediate outcomes is that there are relativelyfew

theoretically plausible change pathways that affect delinquent behavior.

In contrast, the number of distinct intervention programsthat could b e createdto address delin-

quency is limitless, and a considerable number have already been developed. For example, as of

April  CrimeSolutions.gov lists  programs focused on juveniles. That list represents a lot

of programs to evaluate and consider for adoption. And determining the effectiveness of each