The Nature of Co-Occurring Exposure to Violence and of Court Responses to Girls in the Juvenile Justice System

Co-Occurrence of Different Types of Violence Exposure

Studies in different disciplines (criminology, psychology, social work, child welfare) consistently show that girls heavily involved with juvenile courts have histories of childhood mistreatment (e.g., sexual and physical abuse, neglect, abandonment), victimization (e.g., sexual assault, physical attack), and exposure to violence. Many girls in the juvenile justice system have seen a person dead or dying or lost a loved one to death, incarceration, or abandonment (Schaffner, 2006). An especially high proportion has suffered sexual victimization, often by a parent or other family member (Abram et al., 2004; Acoca, 1998; Belknap & Holsinger, 2006; Chesney-Lind & Shelden, 2004; DeHart, 2009; McCormick, Janus, & Burgess, 1986; Rhodes & Fischer, 1993; Wood et al., 2002). One national study revealed that nearly two thirds of girls in correctional settings reported physical abuse (half of them at least 11 incidents), and half reported sexual abuse (a third noting 3-10 instances; Holsinger, 2000). Given the high prevalence of several types of violence exposure among court-involved girls, it is not surprising that many girls experience multiple adversities.

Confirming co-occurrence of girls’ high prevalence of multiple types of childhood violence exposure, a study of nearly 2,000 youth in juvenile detention facilities revealed that girls were at higher risk than boys for multiple types of victimization (i.e., poly-victimization; Ford, Grasso, Hawke, & Chapman, 2013). There also appears to be a tendency toward revictimization, sometimes through new forms of attack; victims of childhood sexual abuse are at heightened risk for sexual assault and intimate partner violence (IPV; Banyard, Arnold, & Smith, 2000; Coid, Petruckevitch, Feder, Chung, & Moorey, 2001; Hebert, Lavoie, Vitaro, McDuff, & Tremblay, 2008). DeHart’s (2009) study of poly-victimization determined the sequencing of different types of violence exposure for a sample of adjudicated delinquent girls. Caregiver violence and witness to violence began at an early age, with a second high point in adolescence. Gang or peer group attacks increased just before pubescence, and dating violence escalated just after. Girls experienced sexual violence at all ages, but especially in adolescence. Violence exposure has the strongest statistical associations with delinquency when it occurs in multiple contexts (family, community, and school), in multiple forms, and repeatedly (Chang, Chen, & Brownson, 2003; Finkelhor, Ormrod, Turner, & Hamby, 2005; Ford, Elhai, Connor, & Frueh, 2010; Hebert et al., 2008; Lin, Cochran, & Mieczkowski, 2011; Mrug & Windle, 2010; Shukla, 2010). As a result, a sizable number of court-involved girls would be expected to have exposure to multiple forms of victimization and violence.

Just a few studies explain why youth experience multiple adversities. Some researchers contend that families with assaultive behavior between some members (e.g., siblings) have high prevalence of aggression of other family member pairings (e.g., parent with parent; Dixon, Howie, & Starling, 2005; Salisbury, Henning, & Holdford, 2009; Smith Slep & O’Leary, 2005). In such families, siblings may assault girls, and girls may see parents assaulting each other. Providing a different explanation from a study of urban girls in a highly disadvantaged area, Kennedy (2005; also see Kulkarni, 2006) concluded that girls abused in their natal homes escape by joining the first available intimate partner, thereby becoming vulnerable to IPV. Alternatively, Kennedy, Bybee, Kulkarni, and Archer (2012) offered two other plausible explanations of the connection between childhood abuse and IPV in poor, urban areas: Witness of abuse and being abused may normalize abuse within relationships, or it may lead to psychological difficulties (e.g., difficulty regulating affect) that contribute to conflict in relationships. Although these explanations are consistent with prior research on the effects of abuse, Kennedy et al.’s (2012) study had a different purpose than generating evidence of these causal mechanisms. A need remains to better understand why seemingly disparate types of childhood adversities (e.g., child sexual abuse, community violence exposure, IPV) co-occur for girls.

Juvenile Court Practices

Federal agencies, departments, and offices have adopted trauma-informed approaches in response to the high proportion of girls (and women) who are both charged with offenses and are also victims of crimes (Women and Trauma Federal Partners Committee & United States of America, 2013). A trauma-informed juvenile justice system “understands that youth who are chronically exposed to trauma are often hyper vigilant and can be easily triggered into a defensive or aggressive response toward adults and peers” (Buffington, Dierkhising, & Marsh, 2010, p. 20). This understanding requires that the juvenile justice system takes steps to improve youths’ feelings of safety and decrease reminders of past violence exposure, and it provides evidence-based treatment to youth with histories of childhood adversity. The Office of Juvenile Justice and Delinquency Prevention promotes state and local agency adoption of these types of trauma-informed approaches.

Research on specific juvenile court jurisdictions has uncovered several negative effects of some responses to girls with histories of childhood adversities. Using 1999 data, separate studies in Maricopa County, Arizona, and Philadelphia, Pennsylvania (Gaarder, Rodriguez, & Zatz, 2004; Simkins & Katz, 2002) showed that girls’ case files often lacked information about prior history of abuse, and judges made decisions without such background information. Studies also find that commonly used detention and residential facility practices, including isolation of girls, staff insensitivity, use of restraints, and lack of privacy, bring back painful memories of abuse and thereby revictimize girls (Griffin, 2002; Hennessey, Ford, Mahoney, Ko, & Siegfried, 2004; Quinn, Poirier, & Garfinkel, 2005). Treatment of trauma from childhood adversities rather than confinement to facilities focused on punishment and incapacitation reduces revictimization, delinquency, and status offending (Adams, 2010; Holman & Ziedenberg, 2006). The variation between juvenile court jurisdictions and the possible changes since data for these articles and reports were collected provide the rationale for the present study’s focus on a contemporary juvenile court in one locale.

Setting and Sample

The research was conducted in one Michigan county court jurisdiction with a rural, suburban, and small city mix. The state’s Juvenile Code (712A.1) mandates that juvenile courts provide guidance and control conducive to youth’s welfare and the state’s best interests and sets 16 as the upper age for juveniles. Except for detention centers, both private and public facilities are intended to meet treatment and educational needs of youth. The Code also specifies that youth under the court’s supervision should remain with their families if at all possible. To place youth in treatment settings or refer them to treatment programs, the court studied in the present research routinely assessed youth with the widely used Youth Level of Service/Case Management Inventory (YLS/CMI; Schmidt, Hoge, & Gomes, 2005). Although state laws are relatively punitive in regard to the end of juvenile court jurisdiction once a youth is 16, the court maintained a strong treatment orientation through its use of the YLS/CMI and its extensive use of referrals to residential and non-residential treatment programs for juvenile offenders.

To recruit a varied group of girls heavily involved in the court system, two sampling approaches were used: (a) a census of girls currently above 16 and previously placed in the county-based group home most often used for residential placement, and (b) a maximum variation sample of girls not placed in the group home, but with extensive court involvement. Maximum variation sampling, a form of purposeful sampling, involves selecting a wide range of cases so common patterns can be documented across diverse circumstances (Creswell, 2007; Patton, 2002). The sample was restricted to girls older than 16, since in Michigan, youth above 16 years are considered adults. For the census, interviews were completed with 12 girls placed just in the group home and five girls placed in both the group home and another residential setting. One other age-eligible group-home resident declined participation. For the maximum variation sample, court personnel identified five girls placed in other residential programs, three who received intensive supervision and counseling, one in a day school/treatment program, and one in foster care. Seven other court-referred girls declined participation or could not be located, so the sample size was 27 (77% participation rate).

Interviews

Data were collected through in-depth, retrospective interviews with court-involved girls to examine the co-occurrence of childhood adversities and girls’ experience of juvenile court interventions. One of three trained graduate students, women in their 20s, completed each interview in private settings, usually in two or three sessions to build rapport and avoid interview fatigue. During interviews, a life calendar method was used (Hanks & Carr, 2008; Roberts & Horney, 2010). To implement this method, interviewers asked girls to create a timeline of easy-to-recall events, such as ages for each place of residence, schools attended, jobs worked, arrests and court contacts, and other key life events that could increase recall of event timing (Freedman, Thornton, Camburn, Alwin, & Young-DeMarco, 1988; Krienert, 2003). During the interviews, the interviewer and participant constructed, periodically reviewed, and improved the life calendar (Hanks & Carr, 2008; Roberts & Horney, 2010) to clarify the timing and order of events. Supporting the validity of this approach, Habermas and Bluck (2000; also see Ruspini, 2002) found that most adolescents are adept at assessing the causal ordering of important life events.

Interview questions relevant to the present analysis covered, from age 12: each neighborhood and school context; periods living apart from parents, reasons, and whereabouts; obstacles and difficulties; and periods when life was and was not going well. Questions about obstacles and difficulties were followed by probes about specific problems and solutions related to a safe place to live, adequate food and clothing, romantic relationships, school, exposure to violence or death, pressured or forced sex, and peers. Specifically, if a girl indicated a specific problem (e.g., lack of a safe place to live), the interviewer asked for a detailed description of why that occurred and whether, and if yes how, the problem was solved. In the example given, girls therefore were encouraged to identify any instances of being “kicked out” of their homes. For reported adversities, interviewers specifically asked about programs and professionals that intervened, and the nature, result, and adequacy of intervention. Final questions centered on girls’ assessment of help they needed at the time of the interview. ¹

Credibility was increased by asking a parallel set of questions of a professional who girls indicated knew them well. For 23 girls consenting to contact with a professional, 21 interviews were completed. Professionals worked as probation officers (5), group-home staff (13), and staff in other programs (3; for example, shelters). In this article, information from professionals is denoted by referring to “the professional.” Quotes from girls are presented with numbers assigned to each girl to show variation and repetition in girls quoted. Audio-recorded interviews were transcribed and analyzed with NVivo software. In many cases, a girl and a professional filled in gaps left by the other or validated each other’s comments. To assess girls’ patterns of delinquency, reports of the professional and the girl were considered to identify the fullest range of illegal behaviors; girls reported more delinquent activity than did professionals.

Analysis

The author carried out a theoretical thematic analysis (Braun & Clarke, 2006) to identify the types of adversities girls and professionals described, co-occurrence of adversities, and experiences of court placements and referrals. Belknap and Holsinger (2006) concluded from a study of risk factors for delinquency that definitions of childhood adversities should be expanded beyond common categories (e.g., child sexual abuse, neglect). To allow this expansion, the author coded whatever girls described as adverse childhood experiences regardless of whether they were identified in prior literature. Also coded were sections of transcripts where a study participant described reasons why adversities co-occurred, and connections of adversities to how girls experienced interventions. These coded passages and the life calendars were the basis for generating study findings.

Analysis included using the matrix query function (adversity BY adversity) of NVivo software to identify the co-occurrence of different adversities. Also, the matrix query function (case BY themes) was used to output a quantitative data set, with each theme coded as present (1) or absent (0) in the girls’ responses; this output allowed for the use of SPSS to generate descriptive statistics.

Description of the Sample

Girls described themselves as White/Caucasian (13), Black/African American (6), multiracial (4), Hispanic (3), or Asian (1). They ranged from age 17-19. Most families were poor or working class, but a few had at least one parent who was a professional. Based on both girls’ and professionals’ responses, most girls had used drugs or alcohol (21), been truant (18), or had run away (16). Some also engaged in physical fighting (12), committed larceny shoplifting (9), or assaulted a parent or sibling (7). Just four girls had more serious, repeated offenses including selling drugs, prostitution, felony theft, and robbery.

Types and Combinations of Childhood Adversities

Consistent with prior research, most girls experienced multiple adversities. Based on coding into 13 categories, the mean number of types is 4.15 (SD = 2.46, range = 0-9). Most common, 18 girls described at least one type of non-consensual sex: sexual assault incidents when they were teenagers (9 girls), one or more unconnected (i.e., a different perpetrator) childhood sexual abuse incidents (8), repeated childhood sexual abuse by the same perpetrator (6), and/or sex with an adult (5). Sexual assault incidents typically occurred when girls were at parties, on dates, or alone with a man they knew. Childhood sexual abuse incidents ranged from inappropriate touching to forced sex, and typically involved a relative or someone else visiting or living in the household. Sex with an adult involved girls agreeing to sex when they were below 18 with a man who was older than 18. Also, an adult man kidnapped and forced one girl to prostitute for several weeks to provide him with money and drugs. Other adversities were parental neglect and IPV, each described by 13 girls. Witnessing violence in the community or school (11), seeing or knowing a person dead or dying of unnatural causes (e.g., drug overdose, shooting; 10), being “kicked out” of one’s home (10), and experiencing behavior of a mentally ill parent as disturbing (e.g., screaming, passing out, rapid mood changes; 9) were also common. Seven girls described ongoing mutual combat between themselves and a parent (i.e., parent and child hit each other), five described physical abuse by a parent (i.e., parent abuses the child), and six saw family members physically hit each other.

Girls experienced not just a wide range, but a pervasiveness of adversities over time and in many settings. As an example, one girl answered a question about seeing violence:

Oh, I see people getting hurt all of the time. I done seen people get bottles cracked over their head, get hit with guns, get shot at. Um, seen people get cut, thrown down stairs. I almost saw someone get drowned in a pool. I seen a girl get her face smashed on the boardwalk. Um, I done seen people get jumped. People fighting one-on-one. Hit with locks, anything, any kind of weapon you can think of is there. I watched my sister get her face stomped in before. Um, I watched my mom and my dad get jumped by a whole bunch of Mexicans before. And, that’s basically it throughout my family. I mean, I done watched my sister get hit by her boyfriend before, her ex-boyfriend. But, family, I really didn’t see too much violence with them except for when my mom and dad used to fight. (#6)

Although the present study was conducted in a county that was not highly disadvantaged and that included a mix of small city, town, and rural areas, the extent of violence exposure that girls like Participant #6 described was similar to reports by girls from high-crime and highly disadvantaged urban areas (e.g., Jones, 2010; Schaffner, 2007).

Study participants identified two forms of traumatic events often omitted from research on youth in juvenile court and from juvenile needs-assessment inventories—being “kicked out” of the home and IPV. Experiences of being “kicked out” were extreme; for example, having all belongings packed, and the bedroom reallocated for a new use (#16). Several girls also described severe IPV; for instance, Participant #21’s account of very controlling behavior (being locked in the house, only allowed to go on the front porch, and isolated from family and friends). In other accounts of severe IPV, Participant #9 described serious injuries involving a broken nose, black eye, bump on the nose when thrown down stairs, and being thrown into walls, refrigerators, and beds; Participant #10 said that highly injurious behavior was sparked by “every little thing.” These patterns of violence mirror the controlling and physically abusive behavior described in studies of serious IPV against adult women (Basile, Chen, Black, & Saltzman, 2007).

Adversities Creating Vulnerability for Sexual Violations and IPV

As shown by the detailed accounts below, girls exposed to considerable violence, with no adult ensuring and monitoring their safety (and thus who were molested or attacked in their homes), who left home due to conflict with parents or neglect, or who were affected by more than one of these dynamics became vulnerable to sexual assault and IPV outside of the natal home. Thus, the adversities of in-home violence exposure, neglect, and parent–child fighting contributed to girls’ sexual victimization and IPV outside the home.

Girls’ Experiences With Interventions

Data analysis provided four insights into the interventions that girls with histories of adversities experienced in court-ordered placements and programs. First, some girls and professionals felt girls’ histories of violence exposure were unrecognized or not addressed. Second, detention promoted depression and low self-worth in girls who, due to childhood adversities, were already emotionally distraught. Third, girls described interventions that directly addressed their trauma and negative emotions as helpful, and interventions that focused solely on changing their behavior as unhelpful. Finally, termination of relationships with helpful professionals left several girls newly traumatized and still negatively affected by prior or current adversities.