از بزهکاری تا ارتکاب بدرفتاری با کودک: بررسی عدالت کیفری بزرگسالان و رفاه کودکان درگیری جوانان آزاد شده از امکانات دادرسی نوجوانان

Abstract The present study prospectively tracks 999 juvenile delinquents (499 females) released from New York State correctional facilities in the early 1990s and describes their engagement in two socially problematic behaviors in early adulthood: child maltreatment and crime. By age 28, nearly two-thirds of sample girls were investigated by child protective services for alleged acts of child maltreatment and over half became clients of both the child welfare and adult criminal justice systems. Prevalence of maltreatment perpetration and dual-system contact were lower for boys but still worrisome. Findings add to a growing body of research documenting the overlap between criminal justice and child welfare populations and highlight the need for greater integration between these systems, particularly when dealing with female clients.

. Introduction Acquiring a comprehensive picture of how juvenile delinquents fare once they leave the auspices of the juvenile justice system is crucial to the development and implementation of effective rehabilitation services. Unfortunately, while it is clear that many youth who offend as adolescents will continue to engage in criminal activity as adults (Colman et al., 2009, Eggleston and Laub, 2002 and Ezell and Cohen, 2005), far less is known about how these high-risk youth behave in other areas relevant to healthy adult functioning. Intent on documenting adult criminal involvement, researchers have largely ignored the extent to which youth with histories of delinquency engage in other less publicly visible, but equally problematic types of adult antisocial behavior, such as child maltreatment. Yet, both developmental theory and delinquency research suggest that youth served by the juvenile justice system may be at risk for the perpetration of abuse and neglect. While it has been well documented that crime often begets crime, longitudinal studies of human development indicate that how antisocial tendencies are expressed may also vary over time and across contexts (e.g., Broidy et al., 2003, Huesmann et al., 1984, Pajer, 1998 and Sampson and Laub, 1990). Known as “heterotypic continuity”, this concept refers to the notion that a single, underlying trait may give rise to different types of behaviors as individuals mature, encounter new social contexts, and take on new social roles. Thus, individuals who commit street-based crimes as teens may go on to engage in more family-centered forms of antisocial behavior (e.g., intimate partner violence and child maltreatment) as they enter young adulthood, establish romantic partnerships, and begin to form families of their own. Consistent with this hypothesis, findings from two longitudinal studies indicate that individuals with histories of juvenile delinquency are more likely than their less antisocial peers to engage in family violence in adulthood. In the Dunedin Multidisciplinary Health and Development Study, both aggressive delinquency and juvenile police contact significantly predicted physically abusive behavior toward romantic partners in adulthood (Moffitt & Caspi, 1999). Likewise, Giordano and colleagues found girls' and boys' self-reported level of delinquent activity in adolescence to be significantly related to engagement in relationship violence ten years later (Giordano, Millhollin, Cernkovich, Pugh, & Rudolph, 1999). Although the extent to which these findings extend to violence toward children has not been explored, recent work examining the overlap between different types of family-based violence suggests that intimate partner violence and child maltreatment often go hand in hand. Hazen, Connelly, Kelleher, Lansverk, and Barth (2004) analyzed data from the National Survey of Child and Adolescent Well-Being (NSCAW), a probability study of U.S. children involved in child protective services investigations, and found that nearly half of all female caregivers reported for childhood maltreatment also experienced relationship violence at some point in their lifetime. Moreover, many correlates of juvenile delinquency are also known risk factors for the perpetration of child maltreatment, suggesting that the level of maltreatment risk found within delinquent samples may be particularly high. Indeed, retrospective studies examining the early histories of known juvenile offenders indicate that one-third to two-thirds of youth involved in delinquency have themselves experienced some form of childhood maltreatment (Wiebush, Freitag, & Baird, 2001), placing them at greater risk for later perpetration. Although the association is far from deterministic, numerous studies have shown that individuals who experience maltreatment as children are significantly more likely than their non-maltreated peers to become abusive and neglectful parents later in life (Dixon et al., 2005, Egeland et al., 1988, Kaufman and Zigler, 1987 and Pears and Capaldi, 2001). High rates of early childbearing in delinquent samples may also increase the likelihood that youth with histories of delinquency will maltreat in early adulthood. In the Denver Youth and Rochester Youth Development studies, pregnancy rates by age 17 were high (42% and 29% respectively, particularly among girls involved in some form of delinquent activity (Huizinga, Loeber, & Thornberry, 1993). The proportion of boys heavily involved in delinquent activity who became fathers by age 20 was also considerable, with estimates from two longitudinal studies ranging from 19% to 47% (Thornberry, Wei, Stouthamer-Loeber, & Van Dyke, 2000). As teen parents cope less effectively with the stresses associated with parenting and engage in harsh/punitive parenting practices more often than mature parents (George and Lee, 1997 and Stier et al., 1993), early transitions into parenting roles may increase both opportunity and risk for engaging in child maltreatment. Finally, other problems frequently found in delinquency samples—mental health disorders (Cauffman et al., 1998, Teplin et al., 2002 and Ulzen and Hamilton, 1998), substance abuse (McClelland, Teplin, & Abram, 2004), and post-release employment difficulties (Bullis et al., 2002 and Sampson and Laub, 1990)—may also interfere with youth's ability to successfully transition into healthy caregiving roles. Both population and clinical-based studies of child maltreatment consistently report higher rates of mental health disorders and substance use among parents who maltreat (Debellis et al., 2001, Kelleher et al., 1994 and Walsh et al., 2003). Similarly, unemployment rates and financial hardship have been linked to harsh parental behavior and higher child maltreatment rates in numerous child welfare studies (Gillham et al., 1998 and McLoyd et al., 1994). In short, findings from the various lines of research reviewed above suggest that the family sphere represents a likely, albeit largely unexamined, context for the expression of antisocial behavior in early adulthood. Research examining how youth with histories of delinquency fare within this domain is therefore needed in order to gain a more comprehensive picture of these youth's early adult functioning. 1.1. Gender and the expression of adult antisocial behavior In particular, exploring the extent to which youth with histories of delinquency become abusive and neglectful caregivers may help to broaden our understanding of the long-term consequences and costs of female delinquency. Studies examining criminal recidivism rates among known offenders typically find that women are less likely than men to reoffend (Benda, 2005, Cottle et al., 2001, Langan and Levin, 2002, Mazerolle et al., 2000, Minor et al., 2008 and Soothill et al., 2003). This finding is often taken to indicate that female offenders are less persistent in their antisocial behavior than male offenders, and hence of lesser societal concern. However, it is also possible that differences in adult gender roles simply alter the landscape in which men and women have the opportunity to misbehave. Women are more likely than men to live with children and spend considerably more time engaged in caregiving tasks than their male counterparts (Fields, 2003, Kreider, 2008 and Zick and Bryant, 1996). Consequently, while this greater pull toward family responsibilities may play a potentially influential role in fostering women's desistence from crime as some research suggests (Broidy and Cauffman, 2006, Graham and Bowling, 1995 and Rumgay, 2004), it may also serve to create new family-based opportunities for deviance that are simply less likely to trigger the attention of the criminal justice system. Indeed, research from both criminology and child welfare indicates that women's misbehavior is more likely than men's to center around the home. Women are more likely than men to target their violent behavior toward intimates, parents, and family members (Franke et al., 2002, Greenfeld and Snell, 1999, Snyder and Sickmund, 2006 and Zahn et al., 2008) and are more likely to be identified as perpetrators of abuse and neglect within the general population (Sedlak & Broadhurst, 1996). Thus while it appears that girls with histories of delinquency are less persistent in their antisocial behavior than their male counterparts, the veracity of this assertion may depend upon the context in which youth's behavior is examined. Apparent differences between boys' and girls' behavior may be reduced, or even disappear, if we expand our exploration of adult deviance to include abusive and neglectful acts within the family domain. Cross-domain research aimed at simultaneously documenting boys' and girls' involvement in multiple forms of adult antisocial behavior, such as crime and family violence, is therefore needed to help determine how sex and context interact to shape our understanding of adult outcomes. 1.2. The current study In hopes of providing a more comprehensive picture of the long-term functioning of youth with histories of delinquency, the present study examines the extent to which male and female youth served by the juvenile justice system engage in both child maltreatment and crime during their early adult years. Three core aims guide our research. First, as little is currently known about how youth with histories of delinquency behave within the family domain, we begin by providing gender-specific information on the prevalence, frequency, and type of youth's adult involvement with child protective services (CPS) up to age 28. We then briefly describe youth's criminal involvement during this same period and explore whether involvement in adult antisocial behavior varies across domain and sex. Specifically, we hypothesize that girls with histories of delinquency will be more likely than boys with histories of delinquency to become perpetrators of child abuse and neglect, while boys with histories of delinquency will be more likely than girls with histories of delinquency to enter the adult criminal justice system. Third, we explore how knowledge of maltreatment outcomes adds to our understanding of the prevalence and scope of adult antisocial behavior within delinquent samples by describing patterns of cross-system involvement in early adulthood in a sample of youth released from juvenile justice facilities. How many of these youth avoid contact with the adult criminal justice system, but go on to maltreat the children under their care? How many offend in both domains? Do patterns of cross-systems involvement vary by gender and, if so, what are the theoretical and practical implications of these differences?

Results 3.1. Perpetration of abuse and neglect Consistent with notions of heterotypic continuity and study hypotheses, sample youth were highly likely to come into contact with CPS during their early adult years, generating over 1300 maltreatment investigations during the follow-up period. As shown in Table 1, 62% of girls released from juvenile justice facilities were named as an alleged perpetrator of abuse and neglect in at least one CPS investigation prior to age 28. Prevalence of alleged perpetration was lower among sample boys, but still worrisome, with 17% alleged to have maltreated the children under their care. In keeping with general trends within the child welfare system, allegations of child neglect were most common, followed by physical and then sexual abuse. When investigation outcome was taken into account, 68% of reported girls and 54% of reported boys had credible evidence confirming the alleged maltreatment within at least one CPS report, bringing the prevalence of confirmed perpetration to 42% for sample girls and 9% for sample boys. Table 1. Early adult CPS and criminal justice involvement of youth formerly served in juvenile justice facilities. Type of system contact Males (n = 500) Females (n = 499) Significance test Prevalence (%) Rate Prevalence (%) Rate Chi-square F-test CPS investigation Alleged perpetrator 17 2.06 62 3.95 213.10⁎⁎⁎ 22.43⁎⁎⁎ Neglect 16 2.04 59 3.79 197.26⁎⁎⁎ 19.62⁎⁎⁎ Physical abuse 5 1.16 24 1.84 71.92⁎⁎⁎ 6.14⁎ Sexual abuse 2 1.75 6 1.11 11.57⁎⁎⁎ 5.40⁎ Confirmed perpetrator 9 1.62 42 2.54 145.18⁎⁎⁎ 9.24⁎⁎ Criminal justice Arrest 89 8.97 81 5.95 10.77⁎⁎⁎ 54.54⁎⁎⁎ Felony 83 5.16 63 2.72 47.91⁎⁎⁎ 121.04⁎⁎⁎ Violent 77 3.53 57 2.21 43.75⁎⁎⁎ 62.85⁎⁎⁎ Property 66 3.47 55 3.28 14.29⁎⁎⁎ .51 Drug 61 3.36 36 2.72 59.18⁎⁎⁎ 5.32⁎ Conviction 85 5.72 69 4.09 37.61⁎⁎⁎ 28.37⁎⁎⁎ Incarceration 71 3.87 32 1.53 153.18⁎⁎⁎ 85.99⁎⁎⁎ ⁎ p ≤ .05. ⁎⁎ p ≤ .01. ⁎⁎⁎ p ≤ .001. Table options Rate of CPS contact was also high, as youth who came to the attention of the child welfare system typically did so on more than one occasion. Over two-thirds (68%) of allegedly perpetrating girls were named in two or more CPS calls during the follow-up window, racking up an average of 3.95 investigations per girl. Similarly, 47% of boys identified as an alleged perpetrator were named in multiple CPS reports, generating an average of 2.06 investigations per alleged perpetrator. As suggested by the disparate numbers reported above, chi-square and one-way anova comparisons confirmed that both the prevalence and rate of CPS contact varied significantly by sex. Consistent with study hypotheses, girls were approximately 3.5 times more likely than boys to be identified as an alleged perpetrator of child abuse and neglect during their early adult years, and when reported were investigated more often. 3.2. Adult crime In keeping with the high recidivism rates observed in other samples of youth with histories of serious delinquency, sample youth were also highly likely to enter the adult criminal justice system (see Table 1). By age 28, 89% of sample boys had been arrested on adult charges at least once, as had 81% of sample girls. In addition, recidivating youth tended to be serious, repeat offenders. Eighty-three percent of the boys and 69% of the girls that recidivated were arrested on more than one occasion, accumulating an average of 8.97 and 5.95 adult arrests respectively. Felony-related charges were common, with 83% of sample boys and 63% of sample girls charged with at least one felony-level offense. In addition, a substantial proportion of both our male and female sample experienced a period of incarceration during our follow-up period, potentially diminishing the rate of criminal arrest and maltreatment perpetration in both groups. However, despite the exceedingly high recidivism rates observed for both boys and girls, prevalence and rate of youth's criminal involvement still varied significantly by sex. As hypothesized, boys were significantly more likely than girls to experience all types of adult arrest. They also tended to accumulate more arrest charges when they did recidivate. In addition, substantial sex differences in the prevalence of adult conviction and incarceration also emerged. Eighty-five percent of sample boys were convicted on at least one occasion, compared to only two-thirds of sample girls. Similarly, boys with delinquency histories were far more likely than girls with delinquency histories to experience incarceration. 3.3. Cross-system involvement Fig. 1 displays the patterns of cross-system involvement found for both boys and girls. Consistent with the findings reported above, 72% of sample boys were single-system clients, offending only in the criminal justice domain. Sixteen percent came into contact with both CPS and criminal justice services, 11% avoided contact with either system, and only three boys, or approximately 1% of our male sample, were investigated by CPS but never touched the criminal justice system. Conversely, girls were most likely to be dual-system clients, with 53% coming to the attention of both the CPS and criminal justice systems between the ages of 16 and 28. Just over a quarter experienced only adult arrest, 10% had no system contact, and 9% were investigated by CPS exclusively. Thus, while domain-specific findings indicate that girls were more likely than boys to maltreat and boys were more likely than girls to criminally offend, the overall proportion of youth with delinquency histories engaging in some form of adult antisocial behavior—either the perpetration of child maltreatment or crime—was highly comparable across the sexes (89% for sample boys and 90% for sample girls). Moreover, girls were more likely than boys to experience the co-occurrence of problems in early adulthood (53% and 16%, respectively) and to have evidence of their adult antisocial behavior “missed” by criminal justice-based indicators. Forty-three of the 94 girls who avoided contact with the adult criminal justice system were investigated by CPS, cutting the number of girls who appeared “successful” from a criminal justice perspective nearly in half. Cross-systems involvement. Fig. 1. Cross-systems involvement. Figure options Table 2 displays the CPS and adult criminal profiles of the boys and girls assigned to various cross-systems groups. Descriptive statistics for CPS Only boys are not presented, as only three boys fell into this category. Dual Contact boys entered the adult criminal justice system several years before their first report to CPS. They were arrested significantly more often (F (1,441) = 7.93, p < .01) than their CJ Only peers, but were no more likely to be charged with a violent, property or drug-related charge. Prevalence of conviction was similar across groups; however, boys in the Dual Contact group spent substantially less time incarcerated than boys in the CJ Only group (F (1,441) = 6.23, p < .05). Table 2. Cross-system profiles of youth involved in CPS and adult criminal justice. Type of system contact Males Females CJ Only (n = 362) Dual (n = 81) CPS Only (n = 43) CJ Only (n = 139) Dual (n = 266) CPS Age at first CPS report n/a 23.38 22.3 n/a 21.68 Total CPS reports n/a 2.06 3.19 n/a 4.08 Ever alleged perpetrator of Neglect n/a 93% 95% n/a 94% Physical abuse n/a 30% 33% n/a 39% Sexual abuse n/a 10% 5% n/a 10% Ever confirmed perpetrator n/a 54% 67% n/a 68% Criminal justice Age at first adult arrest 17.46 17.25 n/a 18.43 19.19 Total arrests 8.6 10.62 n/a 5.01 6.43 Ever arrested for Felony offense 94% 91% n/a 78% 78% Violent offense 86% 90% n/a 69% 71% Property offense 74% 79% n/a 60% 71% Drug offense 69% 64% n/a 41% 46% Ever convicted 96% 98% n/a 80% 88% Ever incarcerated 80% 84% n/a 32% 43% Years incarcerated 4.13 2.79 n/a 1.85 1.40 Table options Dual Contact and CPS Only girls were similar in the timing, nature, and degree of their CPS involvement, with no significant differences found between groups on age at first contact, number of CPS reports, types of allegations raised, or prevalence of confirmed perpetration. Dual Contact girls did differ, however, from their CJ Only peers on several measures of criminal activity. Dual Contact girls were older at first criminal arrest (F(1, 403) = 7.99, p < .01) and they experienced adult arrest (F (1, 403) = 5.31, p < .05), conviction, (F(1,403) = 4.80, p < .05) and incarceration (F (1, 403) = 4.53, p < .05) significantly more often than girls who migrated solely into the adult criminal justice system. Girls in the Dual Contact group were also more likely than girls in the CJ Only group to be arrested for a property-related crime (F (1,403) = 5.76, p < .05), but were no more likely than their peers to be brought up on violent or drug-related charges.