Child Maltreatment and Offending Behavior

• Author: Joshua P. Mersky,James Topitzes,Arthur J. Reynolds
• DOI: 10.1177/0093854811398578
• Published date: 01 May 2011
• Date: 01 May 2011
• Subject Matter: Articles

492

CRIMINAL JUSTICE AND BEHAVIOR, Vol. XX, No. X, Month 2007 492-XXX

DOI:

© 2007 American Association for Correctional and Forensic Psychology

CRIMINAL JUSTICE AND BEHAVIOR, Vol. 38 No. 5, May 2011 492-510

DOI: 10.1177/0093854811398578

© 2011 International Association for Correctional and Forensic Psychology

AUTHORS’ NOTE: Chicago Longitudinal Study grants from the National Institute of Child Health and

Human Development (No. R01HD34294-06) and the Doris Duke Charitable Foundation (No. 20030035)

supported the research reported herein. The Institute of Education Sciences, U.S. Department of Education

provided additional support with a dissertation grant (No. R305C050055) administered through the University

of Wisconsin–Madison. Correspondence concerning this article should be addressed to James Topitzes,

University of Wisconsin–Milwaukee, Enderis Hall Room 1057, P.O. Box 786, Milwaukee, WI 53201; email:

topitzes@uwm.edu.

CHILD MALTREATMENT

AND OFFENDING BEHAVIOR

Gender-Specific Effects and Pathways

JAMES TOPITZES

JOSHUA P. MERSKY

University of Wisconsin–Milwaukee

ARTHUR J. REYNOLDS

University of Minnesota–Twin Cities

This study assessed the association between child maltreatment (ages 0–11) and offending behavior within gender-specific

models. Prospectively collected data, including official measures of maltreatment and offending, were derived from the

Chicago Longitudinal Study, a panel study of 1,539 low-income minority participants. Multivariate probit analyses revea led

that maltreatment significantly predicted delinquency for males but not females yet forged a significant relation to adult

crime for both genders. Exploratory, confirmatory, and comparative analyses suggested that mechanisms linking maltreat-

ment to adult crime primarily differed across gender. For males, childhood-era externalizing behavior and school com-

mitment along with adolescent-era socioemotional skills, delinquency, and educational attainment explained the

maltreatment-crime nexus. For females, childhood-era parent factors along with adolescent indicators of externalizing

behavior, cognitive performance, mobility, and educational attainment partially mediated the maltreatment–crime relation.

Implications of results were explored.

Keywords: child maltreatment; juvenile delinquency; adult crime; gender; mediation analysis

Evidence from studies conducted over the past several decades consistently links child

maltreatment (CM) to juvenile delinquency (e.g., Smith & Thornberry, 1995). A small

number of studies published relatively recently also suggests that the criminogenic effects

of CM extend into adulthood (e.g., English, Widom, & Brandford, 2001). It appears that

the association between CM and adolescent or adult offending manifests for both males and

females (Maxfield & Widom, 1996), although gender-specific analyses are rare (Herrera &

McCloskey, 2001).

Select criminological theory, such as James Howell’s (2003), suggests that pathways to

offending may differ to some extent between males and females. Howell’s theory states that CM

(or child abuse and neglect) is more relevant for female versus male offending and implies that

different routes link CM to offending across gender (Johansson & Kempf-Leonard, 2009).

Topitzes et al. / CHILD MALTREATMENT 493

Thus, male CM victims may engage in crime partly as a result of negative peer affiliations,

whereas maltreatment-related mental health problems may promote crime involvement

particularly for female CM victims (e.g., Bender, 2010). The field, however, lacks an

adequate body of knowledge to support or challenge these hypotheses.

Empirical insight into the similar or differential nature of the association between CM

and offending behavior across gender helps to address unresolved questions and to test

unsupported hypotheses. It also yields important implications for intervention design and

public policy. Accordingly, the current study investigated gender-specific models of the

association between CM and offending behavior with the following research questions:

1. Does CM predict juvenile delinquency and adult crime in males and females?

2. What mechanisms mediate the CM–crime association within gender subgroups?

3. Do these gender-specific mediation models differ significantly?

Data derived from the Chicago Longitudinal Study (CLS), a prospective investigation of

more than 1,500 low-income, primarily African American participants representing a cohort

of students who graduated from Chicago Public School kindergarten in 1986. Prospectively

administered participant, parent, and teacher assessments along with ongoing admin-

istr ative record searches produced multiple measures spanning child, adolescent, and early

adult years.

LITERATURE REVIEW

Whether measured retrospectively among adolescent or adult offenders or prospectively

among child cohorts, child abuse and neglect appear to increase the risk for offending

behavior. A number of landmark studies have helped to establish a significant link between

CM and juvenile delinquency (e.g., Smith & Thornberry, 1995; Widom, 1989). Extending

investigations into adulthood, researchers have found that incarcerated adult offenders tend

to report significantly higher rates of CM relative to nonoffending adult samples (e.g., Lake,

1995; Weeks & Widom, 1998). Buttressing these results, several prospectively designed

studies have shown that—although the majority of CM victims avoid adult offending—CM

significantly increases the risk for criminality (see English et al., 2001; Fagan, 2005; Maxfield

& Widom, 1996). Furthermore, observed associations between CM and adolescent and/or adult

offending appear consistent across racial/ethnic subgroups (English et al., 2001; Maxfield &

Widom, 1996).

Theorists, empiricists, and practitioners alike have been motivated to understand the ways

in which the CM–offending association varies across gender. Accounting for this interest at

least in part are increases in female delinquency rates over the past few decades at a time when

male offending is declining (Wolf, Graziano, & Hartney, 2009). These divergent trends have

prompted scholars to consider the possibility that female and male offending may have

unique etiologies.

Feminist theorists in particular assert that child victimization is an especially salient

precursor of delinquency and crime for females (Daly & Chesney-Lind, 1988). It may be,

in other words, a strain or life course event that increases the likelihood of offending pri-

marily among females. To a limited extent, empirical evidence supports this claim.

A number of studies, for example, revealed that significantly more female delinquents