Classifying Offenders

• Author: Steven H. Day,Craig S. Schwalbe,Rebecca J. Macy,Mark W. Fraser
• DOI: 10.1177/1541204007313383
• Published date: 01 July 2008
• Date: 01 July 2008
• Subject Matter: Articles

Classifying Offenders

An Application of Latent Class Analysis to

Needs Assessment in Juvenile Justice

Craig S. Schwalbe

Columbia University School of Social Work

Rebecca J. Macy

Steven H. Day

Mark W. Fraser

University of North Carolina at Chapel Hill

Structured needs assessment instruments aid in-service planning in the juvenile justice system.

This article uses latent class analysis to classify juvenile offenders into categories or types

based on need profiles using a sample of 542 youthful offenders. The study identified five

profiles based on scores from a structured needs assessment instrument.

Keywords: juvenile justice; structured decision making; needs assessment; delinquency

At both the federal and state levels, contemporary juvenile justice policies promote the

structured assessments of need for court-involved youths (Howell, 1995). These policies

serve two purposes. First, structured needs assessment informs individual case planning

decisions for youths. Ideally, such assessment helps to identify intervention targets to which

evidence-based practices can be matched. Second, the aggregated findings of structured needs

assessments inform program planning and resource allocation. Although both purposes are

crucial for the functioning of the juvenile justice system, support for the latter has received

less attention. Furthermore, to date, few studies have aggregated data from needs assessments

with the purpose of developing findings to inform program planning and resource allocation.

This article demonstrates the utility of a person-centered analytic approach—specifically

latent class analysis (LCA)—to the classification of juveniles into discrete need profiles. In

applying this approach to needs assessment data from juvenile offenders, we present an

innovative way of using needs data to inform juvenile justice policy and practice.

Needs Assessment of Juvenile Offenders

The widespread use of needs assessment in juvenile court jurisdictions corresponds to

the adoption of the Comprehensive Strategy for Serious, Violent, and Chronic Juvenile

Youth Violence and

Juvenile Justice

Volume 6 Number 3

July 2008 279-294

© 2008 Sage Publications

10.1177/1541204007313383

http://yvj.sagepub.com

hosted at

http://online.sagepub.com

279

Authors’ Note: Correspondence concerning this article should be addressed to Craig S. Schwalbe, 1255

Amsterdam Avenue, New York, NY 10027; e-mail: css2109@columbia.edu.

Offenders by the U.S. Office of Juvenile Justice and Delinquency Prevention (OJJDP;

Howell, 1995, 2003). The Comprehensive Strategy promoted structured decision making as a

way to develop more equitable and efficient allocation of resources. In addition, this strategy

supported the implementation of systems of graduated sanctions designed both to protect

public safety and to divert some youths from institutional-based justice interventions into

community-based rehabilitative programs when appropriate. When fully implemented, four

kinds of assessments guide placement and intervention decisions: offense severity, offense

history, actuarial risk assessment, and structured needs assessment. Whereas the first three

assessment types inform decisions regarding sanction level (i.e., diversion, probation, or

commitment), needs assessment alone informs intervention decisions aimed at rehabilitation.

The Comprehensive Strategy encouraged the development of needs assessment instruments

throughout the United States. Many needs assessment instruments followed an approach

first employed by adult criminal justice agencies and later recommended by OJJDP (Baird,

1984). This approach relies heavily on the experience of local professionals to identify

relevant indicators of need and results in needs assessment instruments tailored to local

conditions. Yet despite their jurisdiction-specific development, many similarities exist

across these instruments. For example, an OJJDP comparison of nine needs assessment

instruments (Howell, 1995) found that all instruments in the review included measures of

substance abuse, family functioning or relationships, emotional stability, school attendance

and behavior, and peer relationships. Most assessment tools also measured health/hygiene,

intellectual ability or achievement, and learning disability.

In contrast, the actuarial approach guided the development of other needs assessment

instruments such as the Youth Level of Service/Case Management Inventory (YLS/CMI)

and the Joint Risk Matrix (JRM; Hoge & Andrews, 2003; Schwalbe, Fraser, & Day, 2007).

These instruments differ from OJJDP-style needs assessment instruments in that item

development and selection was guided by empirical predictive validity rather than by

professional consensus. The YLS/CMI is an assessment of 42 risk factors in eight domains

(offense history, family circumstances/parenting, education/employment, peer relations,

substance abuse, leisure/recreation, personality/behavior, and attitudes/orientation). In contrast,

the JRM is a brief instrument that measures three static risk factors (i.e., factors over which

youths and their families have little control) and 11 dynamic risk factors (i.e., factors that

are potentially malleable through intervention). Though structurally different, the YLS/CMI

and JRM share a similar focus on factors that predict recidivism, and they differ funda-

mentally from consensus-based instruments that tend to incorporate a broader array of

developmental risk factors.

Classification of Offender Needs

As previously discussed, few studies aggregate needs assessment data. In one exception,

Kelly, Macy, and Mears (2005) reported on the need characteristics of juveniles referred to

the Texas juvenile justice system (N=2,100). Needs assessment in Texas is conducted as

part of a joint risk/needs assessment process that uses OJJDP-model instruments. Kelly et al.

(2005) found that the most commonly identified needs were “problems with parental

280 Youth Violence and Juvenile Justice