آیا تجارب بی توجهی دوران کودکی بزهکاری در میان رفاه کودکان را تحت تاثیر قرار می دهد؟

Abstract Child neglect, which is the most common form of maltreatment in the United States, has been repeatedly linked to an increased risk of delinquency. However, the existing literature lacks studies that simultaneously investigate how distinct types of neglect differentially influence delinquency among child welfare involved-youth. In addition, few studies of the relationship between neglect and delinquency include measures of ADHD, peer deviance or community violence, even though these variables have been strongly associated with delinquency. This study uses data from 784 11 to 17 year old youth who participated in Wave I of the Second National Survey of Child and Adolescent Wellbeing (NSCAW II) to examine whether supervisory neglect, physical neglect and parental substance abuse affect delinquency after controlling for ADHD, peer deviance, exposure to community violence, and out-of-home placements. We conducted a negative binomial regression to account for the low rates of delinquency among NSCAW II participants. We did not find significant main effects for supervisory neglect, physical neglect or parental substance abuse. Our study found that as youth age the count of delinquency acts increases. Black and Hispanic youth had higher counts of delinquency than youth with White, multi-racial, or “other” racial identities. Youth in out-of-home care had nearly double the rate of delinquency. Youth with more deviant peer affiliations and youth who had been exposed to community violence engaged in more delinquent behaviors. Our findings underscore the importance of the environment surrounding the youth, and the peers with whom the youth affiliates.

1. Introduction While it is well-established that child neglect increases the risk of delinquency (c.f. Chapple et al., 2005, Kazemian et al., 2011, Maxfield and Widom, 1996 and Zingraff et al., 1993), studies have yet to simultaneously investigate how distinct types of neglect, such as supervisory neglect, physical neglect and parental substance abuse differentially influence delinquency. In addition, few studies regarding the relationship between neglect and delinquency include measures that capture the influence of peer relationships (for exceptions see Chapple et al., 2005, Garnier and Stein, 2002 and Kim and Cicchetti, 2010) or the exposure to community violence (for exceptions see Manly, Oshri, Lynch, Herzog, & Wortel, 2013). The absence of these variables is problematic because they have been strongly associated with delinquency. As a result our knowledge about how child neglect impacts delinquency needs further development in these respects. For the past 30 years scholars have referred to the dearth of studies regarding child neglect as the “neglect of neglect” (c.f. Hildyard and Wolfe, 2002, Kendall-Tackett and Eckenrode, 1996 and Wolock and Horowitz, 1984). This study aims to help fill this gap in the literature by investigating how supervisory neglect, physical neglect and parental substance abuse impact delinquency. To ensure that this study elucidates the association between each type of neglect and delinquency, the study also controls for the influences of ADHD, deviant peers, exposure to community violence, and living in out-of-home care, because each has strongly been associated with a risk for delinquency. In 2012, child neglect accounted for 78.3% of substantiated child maltreatment victims in the United States. Child neglect is considered to be an adverse childhood experience with long-term deleterious effects (Duke, Pettingell, McMorris, & Borowsky, 2010). The consequences of neglect may be as detrimental as sexual or physical abuse (Erickson & Egeland, 2011), or more detrimental (DePanfilis, 2006 and Garbarino and Collins, 1999). Approximately 70% of child maltreatment-related deaths were due to child neglect (U.S. Department of Health and Human Services, Administration on Children, Youth and Families, Children's Bureau, 2013). Additionally, child neglect contributes to adverse brain development and compromised neuropsychological and psychosocial outcomes (De Bellis, 2005 and DePanfilis, 2006). These alterations in brain development can affect memory acquisition, cognitive functioning, personality development, social interactions, and how children respond to fear and stress later in life (APSAC [APSAC], 2008). Finally, child neglect constitutes a traumatic experience that causes children to experience anxiety and distress (De Bellis, 2010). 1.1. Definitions Although definitions of neglect vary widely in the literature, this study defines neglect as an overarching construct that includes acts of caregiver omission, wherein caregivers fail to provide necessary care for a child (APSAC, 2008). Although experiences of physical neglect, supervisory neglect and parental substance abuse fall within this rubric, each subtype may have distinct etiology and sequelae. Each subtype should accordingly be defined separately (Zuravin, 1999), as this study does. Physical neglect entails situations when a parent fails to protect a child from harm, or does not provide basic necessities, such as food, shelter and clothing (Erickson & Egeland, 2011). Supervisory neglect encompasses situations when a parent does not adequately protect a child from harmful people or situations (Coohey, 2003). Parental substance abuse includes situations where a parent's ability to adequately care for a child is impaired (Child Welfare Information Gateway, 2012). 1.2. Social development model Our study applies the social development model to investigate delinquency among child-welfare involved-youth. The social development model blends key elements of social control, social learning, and differential association theories to explain how risk and protective factors influence problematic behaviors, such as delinquency (Catalano & Hawkins, 1996). According to social control theorists, youth commit delinquent acts when the pleasure (i.e., benefits) associated with the behavior exceeds the punishment (i.e., costs). Thus, social control theorists assert that all youth are equally susceptible to engaging in delinquent behaviors, and the decision whether or not to engage in delinquent behavior depends on social controls (Hirschi, 1986, Hirschi, 2002 and Kornhauser, 1978). Social controls are internal (i.e., guilt or shame) or external (i.e., supervision or surveillance) rewards or punishments that result from conforming to or deviating from societal norms (Hirschi, 2002). Primary sources of social control include a youth's bonds to his or her family and peers, and the community context that surrounds the youth. Naturally, experiences of neglect can impinge upon the development of internal and external social controls. In addition, youth with attention deficit hyperactivity disorder (ADHD) may not be able to focus enough to recognize the consequences of her behavior in advance. Wikström and Loeber (2000) found that youth with ADHD were at greater risk of engaging in delinquent behaviors. Social learning theory assumes that youth engage in delinquent behaviors because youth act out of perceived self-interest that exceeds the risk of punishment. Both long- and short-term consequences of behaviors act as behavioral reinforcers. These reinforcers can provide either punishments or rewards for behavior. Within social learning theory, age, gender, and race determine an individual's location in the social structure. These characteristics can influence the extent to which an individual engages in prosocial or antisocial behaviors through social learning variables, including differential association (e.g., associating with peers who use engage in delinquent behaviors) and modeling (e.g., imitating a parent's disregard for laws, which the parent demonstrates by abusing illegal substances; Akers & Lee, 1999). Crime rate plots by perpetrator age show a steep upward slope during adolescence, which indicates that delinquent behaviors increase with age (Hirschi & Gottfredson, 1983). While several studies have found that males are at greater risk of delinquency than females (Connell et al., 2011, Farrington et al., 2010 and Schwartz et al., 2010), others have found that females who have experienced child maltreatment exhibit more delinquent behaviors than boys (Garbarino, Levene, Walsh, & Coupet, 2009). Both Black (Franke, 2000 and Martin et al., 2011) and Hispanic youth have been found to be a greater risk of engaging in delinquency (Franke, 2000). It is difficult to disentangle the effects of race and class. According to Furstenberg, Cook, Eccles, and Elder (1999), “The effects of race confound differences in resources, opportunities, history, and culture, and combinations of these elements” (p. 57). Differential association theory pinpoints the causal pathways that lead child welfare involved-youth either to engage in delinquent behaviors or prosocial behaviors. Prosocial family and peers can thwart delinquency, while parents and peers who model deviant behaviors, such as substance abuse or delinquency, can function as conduits to delinquency. The social development model also posits that prior experiences, such as child neglect, can have different impacts over the course of development (Catalano & Hawkins, 1996). 1.3. Physical neglect Of all of the forms of neglect, physical neglect has been most clearly linked with poverty (Sedlak & Broadhurst, 1996). Eamon and Kopels (2004) review of court cases found that youth who lived in poverty were more likely to be placed in out-of-home care. Unfortunately, being placed in out-of-home care has been associated with a greater likelihood of delinquency (Ryan & Testa, 2005). Jonson-Reid, Drake, and Kohl (2009) found that poor youth with child maltreatment reports are at substantially greater risk of negative outcomes, including delinquency. Although poverty has been linked with physical neglect it is important to emphasize that poverty is not synonymous with physical neglect. It is also important to acknowledge that impoverished communities may expose youth to more violence, which may contribute to delinquency. Thus, it is necessary to differentiate whether physical neglect or exposure to community violence are correlated with delinquency. Although the extant literature has furthered our understanding of the relationship between physical neglect and delinquency, the few studies that have focused explicitly on physical neglect as a risk factor for delinquency have some noteworthy limitations. While most youth offenders initiate delinquent behaviors between 12 and 13 years old (Thornberry, 1996), much of the literature regarding physical neglect focuses on young children. In addition, existing studies have used fairly small samples drawn from limited geographical areas. Manly et al. (2013) followed 101 urban low-income children in upstate New York from four to nine years old, and found that the severity of physical neglect was positively associated with externalizing behavior. They also found that rates of neighborhood crime mediated the relationship between neglect and externalizing behavior. Using data from cohorts of 310 first graders and 361 fifth graders from a mid-sized metropolitan area, Knutson, DeGarmo, and Reid (2004) found that physical neglect predicted antisocial behavior both at baseline and five years later. Erickson and Egeland (2011) had teachers complete the Child Behavior Checklist for 267 children in grades 1, 2, 3 and 6 in Minneapolis. They found that teachers rated youth who experienced physical neglect with higher delinquency ratings on the child behavior checklist compared to children who had not been maltreated. Furthermore, youth who had been physically neglected were more likely to be expelled from school or dropout, and use alcohol. Chapple et al. (2005) found that physical neglect predicted adolescent violence. They also found that youth who experienced physical neglect were rejected by their peers, and that peer rejection led to violent behaviors. Chappell and colleagues concluded that peer rejection functioned as the impetus for forming relationships with deviant peers. 1.4. Supervisory neglect Although supervisory neglect is the most common form of neglect (Coohey, 2003 and Mennen et al., 2010), it is one of the least studied forms of maltreatment (Coohey, 2003). APSAC (2008) suggests considering supervisory neglect as a separate subtype because it differs from other types of neglect in important ways. While other forms of neglect can be characterized by their chronicity and insidious harm, supervisory neglect involves discrete events that can have immediate and sometimes disastrous consequences. Knutson et al. (2004) study of the first and fifth grade cohorts found evidence of a relationship between supervisory neglect and delinquency among youth in the fifth grader cohort, but not the first grade cohort. These results suggest that the consequences of supervisory neglect differ over the course of development. Maughan and Moore (2010) used data from 389 participants of the Cambridge Study in Delinquent Development (CSDD) to form factors of neglectful behavior. They found that when the male participants were 24 years old the parental supervision factor predicted a greater risk of being convicted for a crime. 1.5. Parental substance abuse Several studies have found that children of substance abusing parents experience difficulty adapting socially, and engage in more externalizing or delinquent behaviors than their peers (Barnard and McKeganey, 2004, Hill et al., 2011, Hussong et al., 2007 and Solis et al., 2012). As the result of being incapacitated by the substances they abuse, parents may be less involved and responsive to their children (Suchman & Luthar, 2000). In turn, the parent–child relationship may be undermined and the attachment may be insecure (Barnard & McKeganey, 2004). Not surprisingly, youth with parents who abuse substances experience a multitude of related problems including being removed from their homes on a recurring basis, and remaining in out-of-home placements longer than youth without substance abusing parents (Brook and McDonald, 2009, Fuller and Wells, 2003, Miller et al., 2006 and Wolock and Magura, 1996). Marmorstein, Iacono, and McGue (2009) used data from 1252 17 year olds and their parents who participated in the Minnesota Twin Family Study to assess the relationship between parental substance abuse and youths' externalizing disorders including oppositional defiance disorder (ODD), conduct disorder (CD), and adolescent antisocial behavior. Specifically, they investigated whether parental alcohol dependence (adjusted for parental drug dependence), parental drug dependence (adjusted for parental alcohol dependence), parental cannabis dependence, or parental non-cannabis drug-dependence increased the risk of ODD, CD or antisocial behavior. They found that, with the exception of parents who abused cannabis, significant relationships existed between parental drug dependence and each of the youth externalizing disorders. Although Marmostein et al.'s study provides key insights, one limitation is that the results may not be generalizable to a larger population; the sample was predominantly Caucasian, and was limited to individuals living in Minnesota at the time of birth. Not all studies have linked parental substance abuse to delinquency. A study by Clark, Cornelius, Wood, and Vanyukov (2004) used data from 1167 6 to 14 year old youth and their mothers and fathers to determine whether parental substance use disorders or the presence of parental psychopathology predicted child psychopathology, including CD. They found that parental substance abuse was not associated with conduct disorders after controlling for other forms of psychopathology. One limitation of Clark et al.'s study is that the sample was 63% male. Another limitation is that the study was limited to one geographic area, which may limit the generalizability of the results. It is important to point out that while the studies by Marmostein et al. and Clark et al. further our understanding of the relationship between parental substance abuse and externalizing behavior they both focus on CD, rather than delinquency. Thus, their focus was on the most extreme cases. We believe that it would be informative to examine a range of delinquent behaviors because this would allow for an earlier intervention point. 1.6. Current study The purpose of our study is to investigate whether experiencing supervisory neglect, experiencing physical neglect, or having a parent who abuses substances are independently associated with delinquency beyond the effects of deviant peer affiliation, substance abuse, and demographic characteristics among youth involved in the child welfare system. Our study asks: Do supervisory neglect, physical neglect or parental substance use increase delinquent behaviors among child welfare involved-youth? Our study has the following hypotheses: (1) Hypothesis: We expect that the results of this study will show that experiencing physical neglect will increase adolescent delinquency even after controlling for ADHD, peer deviance, exposure to community violence, out-of-home placements and demographic characteristics (i.e., gender, race, and age). (2) Hypothesis: We also anticipate study results will indicate that experiencing supervisory neglect abuse will increase adolescent delinquency even after controlling for ADHD, peer deviance, exposure to community violence, out-of-home placements, and demographic characteristics (i.e., gender, race, and age). (3) Hypothesis: We predict that the results of this study will show that experiencing parental substance abuse will increase adolescent delinquency even after controlling for ADHD, peer deviance, exposure to community violence, out-of-home placements, and demographic characteristics (i.e., gender, race, and age).

. Results The descriptive statistics for variables used in the analyses are provided in Table 1. Delinquency rates were quite low among the sample. The most common form of neglect was supervisory neglect. Few youth had delinquent peers, while most youth had witnessed community violence. The rates of ADHD exceeded national incidence rates, which were estimated to be 9% in 2009 (Akinbami, Liu, Pastor, & Reuben, 2011). Table 1. Weighted sample characteristics. Actual sample size 784 Weighted sample size 544,084 Percent 100.00 % or Mean (SE) Minimum Maximum Delinquencya 1.75 (0.22) 0 36 Age 13.61 (0.13) 11 17 Gender Male 38.57 Race/ethnicity Black 19.26 Hispanic 23.42 White 49.38 Biracial/multiracial 5.74 Other 2.20 Living situation Out-of-home care 15.40 0 1 Lack of supervision 20.26 0 1 Physical neglect 6.17 0 1 Parental substance abuse 9.04 0 1 ADHD 22.95 0 1 Peer deviance 8.97 (0.19) 6 30 Community violence exposure 55.14 0 1 Note: Above are frequencies for categorical variables and means and standard errors for continuous variables. a Youth were asked 36 questions regarding a range of delinquent or criminal activities (i.e., status, property, and violent offenses) using a modified self-report of delinquency drawn from the Denver Youth Survey. Table options A correlation matrix of the variables used in the study is presented in Table 2. Because none of the variables we used in our analyses were highly correlated (r > 0.60; see Kweon & Oh, 2011) the results indicate that collinearity is not a problem. Table 2. Correlation matrix of delinquency and predictor variables (n = 784). 1 2 3 4 5 6 7 8 9 10 11 12 1. Delinquency 1 2. Age 0.16 1 3. Male 0.01 − 0.09 1 4. Black 0.04 − 0.04 0.04 1 5. Hispanic 0.00 0.00 − 0.02 − 0.32 1 6. Live in out-of-home care 0.12 0.14 0.05 0.12 0.01 1 7. Physical neglect − 0.02 0.01 0.03 0.04 − 0.06 0.03 1 8. Supervisory neglect − 0.07 − 0.04 0.06 0.02 0.09 0.01 − 0.12 1 9. Substance abusing parent 0.06 − 0.02 0.01 − 0.08 − 0.01 0.02 − 0.08 − 0.13 1 10. Peer deviance 0.37 0.14 − 0.07 0.03 − 0.02 0.05 − 0.03 0.01 0.03 1 11. Community violence exposure 0.17 0.10 − 0.13 − 0.07 0.01 0.10 0.00 − 0.06 0.06 0.27 1 12. ADHD 0.00 − 0.08 0.18 − 0.03 − 0.12 0.00 − 0.05 0.01 − 0.03 0.02 0.05 1 Note. All values were slightly above zero, but the rounding masks these values. Table options The bivariate results of the relationship between the dependent variable, delinquency, and each independent variable are presented in Table 3. Significant bivariate relationships existed between delinquency and the following variables: mean-centered age (F(6,777) = 4.87, p < 0.01), living in out-of-home care (F(1,782) = 4.87, p < 0.01), peer deviance (F(21,762) = 8.05, p < 0.001) and community violence exposure (F(1,782) = 23.10, p < 0.001). It is important to note that relationships between the independent and dependent variables may be hidden until other variables are added to the model. In a regression model each covariate's unique effect is determined after adjusting for all of the variables in the model ( Henry & Muthén, 2010). Table 3. Bivariate results of the relationship between the dependent variable and each independent variable at a bivariate level. F Mean-centered age 4.87 ** Male 0.05 Black 1.32 Hispanic 0.00 Out-of-home care 10.63 ** Physical neglect 0.32 Supervisory neglect 3.57 Substance abusing parent 2.79 ADHD 0.01 Peer deviance 8.05 *** Exposure to community violence 23.10 *** Significant results are bolded. Table options Table 4 provides our results for our negative binomial regression. We found that the count of delinquency acts increases as youth age (IRR = 1.05, SE = 0.05, p < .01). Black and Hispanic youth were more likely to engage in delinquency than youth with White, multi-racial, or “other” racial identities (IRR = 1.82, SE = 0.48, p < .05 and IRR = 1.52, SE = 0.31, p < .05 respectively). Youth who lived in out-of-home care were nearly twice as likely to engage in delinquency (IRR = 1.94, SE = 0.46, p < .01). As youth increased the number of deviant peer affiliations they engaged in higher counts of delinquency (IRR = 1.12, SE = 0.02, p < .0001). Exposure to community violence also contributed to delinquency counts (IRR = 1.71, SE = 0.26, p < .001). Table 4. Negative binomial model regressing delinquency on predictors (n = 784). IRRa S.E. t p-value 95% CI Mean-centered age 1.15 0.05 3.00 0.004 [1.05, 1.26] Male 1.07 0.16 0.45 0.651 [0.79, 1.45] Black 1.82 0.48 2.25 0.027 [1.07, 3.09] Hispanic 1.52 0.31 2.04 0.045 [1.01, 2.29] Out-of-home care 1.94 0.46 2.80 0.007 [1.21, 3.11] Physical neglect 0.86 0.40 − 0.32 0.749 [0.34, 2.18] Supervisory neglect 0.67 0.14 − 1.87 0.066 [0.44, 1.03] Substance abusing parent 1.34 0.36 1.08 0.284 [0.78, 2.30] ADHD 0.89 0.12 − 0.89 0.375 [0.67, 1.17] Peer deviance 1.12 0.02 7.50 0.000 [1.09, 1.16] Exposure to community violence 1.71 0.26 3.49 0.001 [1.26, 2.33] Constant 0.04 0.02 − 5.04 0.000 [0.01, 0.14] For ease of interpretation we have chosen to present the results as the incident rate ratio (IRR), which signifies the change in the dependent variable in terms of a percentage increase or decrease, rather than using the coefficient, which is the log of incident counts. Significant results are bolded.