Family contact and recidivism: a longitudinal study of adjudicated delinquents in residential care.

Author:Ryan, Joseph P.

Families are critical to understanding and interrupting patterns of delinquent and criminal behavior. The objective of this study was to determine whether specific types of contact with family members are associated with a reduced risk of recidivism after a long-term residential program for juvenile delinquents. Family contact reports were analyzed for a random sample of 90 adjudicated delinquents released from a long-term residential facility in Michigan. Two years after their release from residential care, official arrest reports were provided by the Michigan State Police Department. Thirty-six percent of the youths were arrested at least once. Cox regression coefficients indicate that age, race, and family contact were related to subsequent involvement with the adult correctional system. In-home counseling, campus visits initiated by family members, and home visits initiated by the family service worker reduced the likelihood of recidivism. The implications for social work policy and practice are discussed.

KEY WORDS: delinquency; family; recidivism; residential care; youths


Approximately 109,000 juvenile offenders were in residential placement in the United States in 1999 (Sickmund, 2002).The primary objectives of residential placement are to understand and interrupt patterns of delinquent behavior. Treatment providers offer a range of services, including but not limited to individualized assessments, group work, family work, and aftercare. Research indicates that many of these services are related to the stated objectives of placement. For example, there is a broad range of literature that explores the risk assessment process (Funk, 1999; Kawahara, 2002) and the effectiveness of aftercare services (Mtschuler & Armstrong, 1995; Altschuler, Armstrong, & Roberts, 1998; Ryan, Davis, & Yang, 2001; Sarri, Rollin, Stephens, & Wolfson, 1996). In part, the findings indicate that risk factors differ for male and female offenders and that the provision of aftercare services significantly reduces the risk of recidivism. These studies are important contributions in the profession's commitment to advance intervention knowledge and understand the treatment mechanisms associated with long-term success. But social work researchers have yet to investigate the effectiveness of all components of residential placement. In particular, research on the benefits of family work is noticeably absent from the literature.

The family system is important in the development and continuation of delinquent behavior. Yet, there is relatively little research on the effectiveness of direct practice with families in the context of residential care for juvenile offenders. The absence of such research leaves many questions unanswered. For example, are regular phone contacts related to the successful completion of program goals? Are campus visits related to family reunification? Are overnight visits associated with a reduction in criminal behavior? The purpose of our study was to address the gap in knowledge regarding family work and the likelihood of recidivism. We explored the treatment activities that involved family members and investigated the relationship between the frequency of such activities and recidivism.


A relationship should exist between the individual activities of a social work intervention and the problem the intervention is to address. This relationship is based on assumptions about the causes of a problem (Martin & Kettner, 1996). With regard to the present study, a relationship should exist between treatment activities with family members and a reduction in the likelihood of delinquent behavior. The association between family contact and delinquency is based on the assumption that certain family factors or characteristics (for example, levels of functioning and parenting practices) are associated with delinquent and criminal conduct. Ideally, these assumptions are grounded in theory. Theory guides the development of social work intervention and helps practitioners identify which problem areas to address and which treatments are most likely to be efficacious.


Social learning and social control are dominant theories used to understand the etiology of delinquency. Social learning theory identifies family members and significant others as important contextual factors in understanding the etiology of human behavior (Akers, 1985). Social learning theory starts with the premise that individuals learn behaviors through their associations with others. Such behaviors can either conform to or deviate from social norms. Three mechanisms are believed to support the learning of delinquent behavior: (1) availability of a delinquent model to imitate, (2) teaching beliefs and attitudes favorable to delinquency, and (3) reinforcement of such attitudes, beliefs, and behaviors (Akers, 1998). Family members are important elements in this process because they often model and reinforce inappropriate behaviors. According to social learning theorists, children tend to repeat behaviors that are reinforced and avoid behaviors that are punished (Agnew, 2001).

The empirical evidence supports interventions that address delinquency from a social learning perspective. These interventions often focus on parental management strategies. Social learning theorists argue that parental management encourages self-control, which in turn reduces the likelihood of delinquent behavior. The mechanisms of parental management believed to produce self-control include consistent monitoring, identification of delinquent behavior, and appropriate punishment (Gottfredson & Hirschi, 1990). Gibbs, Giever, and Martin (1998) tested these relationships and concluded that parental management has an indirect effect on deviance through self-control. Similarly, researchers at the Oregon Social Learning Center reported that interventions to improve monitoring reduce the likelihood of antisocial behavior (Patterson, Reid, & Dishion, 1992). Such interventions generally consist of training modules to establish clear expectations, reinforce compliance, and provide consistent discipline (Smith & Stern, 1997).

Social learning theorists attempt to understand why people engage in delinquent behavior. In contrast, social control theorists attempt to explain why people do not engage in delinquent behavior (Agnew, 2001).The premise behind control theory is that social bonds prevent children from engaging in delinquent behavior. These theorists argue that we all share a similar propensity to engage in delinquency. However, we do not share similar levels of control when confronted with opportunities to engage in nonconforming or undesirable behavior. In part, family relationships and parenting practices explain the variable levels of control. When an individual's bond with society in general and family specifically is weak, the likelihood of delinquency is increased (Hirschi, 1969). Attachment, commitment, involvement, and beliefs are the primary components of the social bond. Control theorists argue that greater levels of parental attachment are associated with greater levels of control and thus a reduced likelihood of delinquency.

Interventions grounded in social control theory often address parent--child relationships. The premise behind these interventions is that if practitioners can increase closeness, warmth, and affection between family members and the adolescent, the likelihood of continued delinquent behavior decreases. Delinquent behaviors decrease because adolescents are hesitant to risk a valued relationship for activities that would be met with disapproval (Hirschi, 1969).

Certainly no single theory can explain all the variance associated with delinquent behavior. Thus, it is not surprising that delinquency interventions often attempt to combine theoretical models and approach problematic behavior from an integrated perspective. These perspectives often assume that delinquent behavior is determined by child, parent, and contextual factors. Child characteristics related to delinquency include but are not limited to gender, race (Archwamety & Katsiyannis, 1998), and age at first offense (Towberman, 1994). Family factors related to delinquency include antisocial parents, rejecting parents...

To continue reading