for kim woods only

308

Gender and Runaways Risk Factors, Delinquency, and Juvenile Justice Experiences

Kimberly Kempf-Leonard Southern Illinois University Carbondale

Pernilla Johansson University of Texas at Dallas

Runaway is a status offense for which there is concern that youths, particularly girls, are vulnerable to the “evils of the streets.” Prior research suggests that many personal and family risks of runaway are gendered, that the streets may be meaner, and that the strategies for survival disproportionately serve to criminalize girls. Moreover, patriarchial values may make the juvenile justice response to runaway girls overly harsh “for their own good.” Unfortunately, research has been hampered by small or biased samples. This study examines many risk factors to discern patterns and gender differences among runaway youths for whom there is a status offense charged. Many charged runaway youths are followed to determine the nature of official responses and whether responses appear gendered. The findings have implications for changes in how officials respond to runaway youths, both in terms of what will better serve the youths’ best interests and for improving public safety.

Keywords: runaway; status offense; child abuse; juvenile justice alternatives

The dearth of knowledge about who the girls are who get referred to juvenile court andhow best to respond to girls in juvenile justice is the result of attention focused almost exclusively at the consistently more prevalent male offender population. This situation, described variously as the “tyranny of small numbers” or “persistent invisibility of girls” (Acoca, 1998, p. 562), was finally recognized as problematic in the late 1990s as females became the fastest growing segment of most juvenile and adult criminal justice systems. It is impossible to disentangle whether the rise in female offenders reflects changing social and cultural norms and behaviors among the female population or among parents, police, and court officials, but that females now have the attention of researchers and administra- tors is without dispute.

One recurrent issue that arises in the literature that does exist on females and juvenile justice is the perception of their vulnerability, for which they are often viewed as requiring greater protection than their male counterparts. From the beginning, the designation of a special category of status offenses eligible only to youths because of their age has been a

Youth Violence and Juvenile Justice

Volume 5 Number 3 July 2007 308-327

© 2007 Sage Publications 10.1177/1541204007301293

http://yvj.sagepub.com hosted at

http://online.sagepub.com

Authors’ Note: This article is drawn from a project supported by the Federal Office of Juvenile Justice & Delinquency Prevention Challenge Grant Program through a grant awarded from the Criminal Justice Division, Office of the Governor of Texas. Opinions expressed within the document are those of the authors.

Kempf-Leonard, Johansson / Gender and Runaways 309

way in which girls were disproportionately represented in juvenile justice. Historian Joseph Kett (1977) connects this gendered legal status to a romantic view of childhood with particular fears of modern life, cities, overpressure, and overcivilization. The concept of adolescence was the creation of a distinctive mind set, an expression of a mélange of nos- talgia and anxiety, and in its crudest mold an embodiment of Victorian prejudices about females and sexuality (p. 143).

In the early juvenile courts, girls appeared more often for status offenses related to perceptions of sexual promiscuity, and girls were incarcerated disproportionately more than boys in reform schools (Schlossman & Wallach, 1978). There is evidence that early differential treatment of status offenders was gender biased and class based and linked to anti-immigrant sentiments (Bernard, 1992; Feld, 1999; Odem, 1995; Odem & Schlossman, 1991; Rothman, 1980; Schlossman, 1977; Schlossman & Wallach, 1978; Tanenhaus, 2004).

Although the Victorian and Progressive eras have passed, patterns that reflect similar gendered values continue (Belknap, 2001; Belknap & Holsinger, 1998; Bishop & Frazier, 1995; Chesney-Lind, 1973, 1988, 1995, 1997; Chesney-Lind & Shelden, 1998; Datesman & Scarpetti, 1980; Feld, 1999; Myers & Sangster, 2001; Snyder & Sickmund, 2006). Many critics assert that a double standard of case processing exists for girls. According to Chesney-Lind and Shelden (1998), differential treatment is the way in which juvenile justice ensures “enforcement of girls’ obedience to a special set of expectations about their deportment, their sexuality, and their obedience to familial demands” (p. 6). It is not only police and court officials but also parents who access juvenile justice systems to intervene with their daughters in ways that they do not with their sons (Sussman, 1977, p. 183). Understanding how gender affects juvenile justice processing is further complicated because of prevailing views that victimization and offending converge for girls in ways they do not for boys (Belknap, 2001; Chesney-Lind, 1997; Chesney-Lind & Shelden, 1998).

The offense for which this gendered view is most pronounced is the status offense of runaway. According to the National Academy of Sciences Panel on Juvenile Crime: Prevention, Treatment, and Control, girls have consistently had a higher rate of arrest for running away than have boys. Studies of runaways, however, have found that boys and girls are about equally likely to run away (Finkelhor, Hotaling, Lewis, & Smith, 1990; Kaufman & Widom, 1999). This disproportionate arrest of girls for running away has been explained by “unique and intense preoccupation with girls’ sexuality and their obedience to parental authority” (Chesney-Lind & Shelden, 1998, p. 135), but it could also reflect a greater con- cern for their safety (McCord, Widom, & Crowell, 2001, p. 56).

This article examines how gender affects youths who are referred to juvenile court for runaway offenses. We look at personal and family characteristics, including known risks and other offending. We also analyze the role of gender in shaping how juvenile justice offi- cials intervene with youths who are arrested as runaways. In the following section, we review the relevant prior criminological research on this topic. We then describe empirical findings based on a large and diverse group of arrested youths from which we are able to discern patterns and relationships involving gender and runaway that have not previously been examined.

310 Youth Violence and Juvenile Justice

Prior Research

Runaways, Risks, and Interventions

Runaway is a status offense for which there has been, and continues to be, an acute con- cern about the effect of the evils of the streets for which youths, and girls in particular, may be vulnerable. There is a lot of speculation about runaway youths but very little systematic information about who they are or what happens to them. According to authors of the Juvenile Offenders and Victims: 2006 National Report, “The stereotype of a runaway is a youth roaming the streets of a large, unfamiliar city alone or in the company of drug deal- ers or a pimp” (Snyder & Sickmund, 2006, p. 45). This notion has also been chastised as based on anecdotal and distorted information, to the detriment of understanding the real needs of runaway youths (Moss, 1986). Chesney-Lind and Shelden (1998) also rebuke the current practice of juvenile justice interventions with runaway girls that serve to victimize them a second time, while the real culprits, their abusers, are uninhibited. The only real issue on which there is consensus appears to be that our knowledge of runaways and their experiences is inadequate (Kurtz, Lindsey, Jarvis, & Nackerud, 2000; Rotheram-Borus et al., 1992; Slesnick, 2001).

Although there is a good chance that most children contemplate a runaway experience at sometime during their childhood, the majority never run away, and the actual prevalence of running away is unknown. Available estimates suggest that one out of eight children run away prior to age 18, and 40% do not return to the same living situation they had prior to leaving (Whitbeck & Simons, 1990). Estimates vary by location too, from 2.1% in Houston and 5.7% in New York City to 12% in San Francisco (Rotheram-Borus et al., 1992; Stricof, Kennedy, Nattell, Weifuse, & Novick, 1991). According to the best available information, during any given year in the United States, there are approximately 1.7 million runaway children, with girls and boys equally represented (Hammer, Finkelhor, & Sedlack, 2002). In this section, we review what is known about why youths run away, what happens to them on the streets, and their involvement with juvenile justice systems.

There are many reasons youths run away from home. Generally, running away coincides with decisions that alternative living situations are preferable to the status quo. According to Zide and Cherry (1992), these decisions fall within four categories: (a) running to seek adventure; (b) running from conflicting, alienating, and dangerous family situations; (c) being thrown out by alienated families often following school or legal problems; and (d) being forsaken because their families can no longer support them. Seeking fun and adven- ture seems almost to define the quests of American adolescents who tend to prefer the com- pany of friends to parents and family and for whom romance is a new pursuit. Rather than running to something, however, most researchers attribute running away to flight from family difficulties.

Family difficulties associated with runaways may be a lack of understanding from par- ents (Spillane-Grieco, 1984), family conflict or chaos (Crespi & Sabatelli, 1993; Pipher, 1994; Spillane-Grieco, 2000), dysfunctional families, or families with problem behaviors (Hwang & Bedford, 2003). The most frequently noted family difficulty related to runaway is child abuse, and sexual abuse in particular (Bucy & Obolensky, 1990; Fleming, 1991;

Kempf-Leonard, Johansson / Gender and Runaways 311

Garbarino, Schellenbach, & Seber, 1987; Janus, Burgess, & McCormack, 1987; Jencks, 1994; McCormack, Janus, & Burgess, 1986; Tyler, Hoyt, Whitbeck, & Cause, 2001). Generally, it is assumed that females are the more common sexual abuse victims. In sup- port of this view, more sexual abuse has been reported by female runaway than male run- away and homeless youths (Janus et al., 1987; McCormack et al., 1986). However, Kaufman and Widom (1999) found that childhood sexual abuse was not more often asso- ciated with running away than other forms of abuse or neglect. Besides running away, researchers report that childhood sexual abuse can lead to self-medication via drug use (Finkelhor et al., 1990; Rotheram-Borus et al., 1992; Rutter, 1989; Widom & Ames, 1994). Sexual abuse also nearly doubled the odds of prostitution (McClanahan, McClelland, Abram, & Teplin, 1999); sexual abuse is directly linked to prostitution, independent of run- ning away (Foti, 1995).

Most runaway youths stay nearby and return home within a few days (Nye & Edelbrock, 1980). For other runaway youths, the risks do not end with their flight from home. To the contrary, many problems are attributed to difficulties following running away. These include drug abuse, delinquency, sexually transmitted diseases, teenage pregnancy, dropping out of school, and violence (Hagan & McCarthy, 1997; Kipke, Montgomery, Simon, & Iverson, 1997; Kurtz, Jarvis, & Kurtz, 1991; Whitbeck, Hoyt, Yoder, Cauce, & Paradise, 2001).

It is easy to envision runaway youths without adult supervision as vulnerable, overly exposed to risks, and missing adequate support networks (Bass, 1992; McMorris, Tyler, Whitbeck, & Hoyt, 2002). Many runaway and homeless youths are sexually active, initiate sexual intercourse at young ages (Robertson, 1989; Rotheram-Borus et al., 1992; Tyler, Whitbeck, Hoyt, & Cauce, 2004; Yates, Mackenzie, Pennbridge, & Cohen, 1988), and are consequently at heightened risk of sexually transmitted diseases, including HIV (Miller, Turner, & Moss, 1990). Black and Hispanic runaways may be at higher risk as these minorities already represent higher percentage of adolescent AIDS cases (56%; Centers for Disease Control, 1991). Elsewhere, analyses of these data show that gay, lesbian, and bisex- ual adolescent runaways experienced more victimization, mental health problems, and risky survival strategies than heterosexual counterparts (Whitbeck, Chen, Hoyt, Tyler, & Johnson, 2004). Among other problems, homeless and runaway youths are also at high risk for illicit drug use (Hagan & McCarthy, 1997; McMorris et al., 2002; Yates et al., 1988). In one study, two thirds of the runaways experienced either alcohol abuse or dependence, or drug abuse (Johnson, Whitbeck, & Hoyt, 2005). In another study, substance abuse among homeless and runaways was five times as high as teenagers living at home (Stiffman, Earls, Powell, & Robins, 1987).

Moreover, younger runaways logically are the most vulnerable because potentially they have spent longer time on the streets and thus are more exposed to risks (Chen, Tyler, Whitbeck, & Hoyt, 2004; Hagan & McCarthy, 1997; Obeidallah & Earls, 1999). In the multisite longitudinal study mentioned earlier, researchers have identified a pathway from early abuse to running away while very young and a trajectory for early independence— street life and high-risk behaviors (Tyler et al., 2001). These researchers suggest the risk for young runaways of becoming what Moffit (1993) calls “life-course persistent offenders” (p. 173). Chapple, Johnson, and Whitbeck (2004) observed the highest rate offenders among the runaways who had accumulated arrests prior to their leaving home.

From the research to date, it is unclear whether the risks experienced by runaways are due more to their vulnerable status or to choices they make as part of lifestyle and strategies to survive independently (Rotheram-Borus et al., 1992, p. 16). Survival strategies make some sense too, such as theft for food and clothes, prostitution for economic survival, and drugs to self-medicate the pain (Gilfus, 1992). In their study on homelessness, Hagan and McCarthy (1997) identify both sexual and nonsexual survival strategies for life on the streets. Whether the result of taking risks or merely trying to survive as a runaway, their status is a juvenile offense in every state, and some of them encounter the juvenile justice system.

The juvenile justice response to runaway youths is a contentious policy issue, but also another issue about which there is limited information. As a status offense, runaway falls within the provision for deinstitutionalization of status offenders in the 1974 Juvenile Justice and Delinquency Prevention Act—the DSO provision. Given the lack of viable alternatives, in 1986, the Advisory Board on Missing Children criticized this provision and recommended that police be allowed to detain runaway children who otherwise could leave police stations and shelters “regardless of the risks and dangers of the street” (Moss, 1986, p. 28). Because of the DSO provision, incarceration for runaway is allowable only through reclassification of the status offense as a law violation termed bootstrapping (Chesney-Lind & Shelden, 1998) or by parental consent—and payment—to confine youths in a private facility (Parham v. J. R., 1979). The general practices of reclassification and privatization of status offenders have been frequently criticized (Castellano, 1986, pp. 496-497; Krisberg, Schwartz, Fishman, Eisikovits, & Guttman, 1986; Schneider, 1984; Schwartz, 1989; Weithorn, 1988), and this may hold true more for runaway youths than for those who violate curfews or are otherwise unruly.

A Gendered Pathway

Meda Chesney-Lind (1997) contends that running away plays a unique and salient role in the pathway from delinquency to crime for girls. According to Chesney-Lind (1989; see also Chesney-Lind & Shelden, 1998), the first issue is that adolescent girls are likely to run from violence and abuse in the home. Indeed, the National Academy of Sciences Panel on Juvenile Crime: Prevention, Treatment, and Control agree:

The motivation for running away may differ by gender. For examples, females may be running away to escape physical or sexual abuse or neglect in their homes. For boys, running away may be an indirect consequence of childhood victimization or may be part of a larger constellation of antisocial and problem behaviors. (Luntz & Widom, 1994; McCord et al., 2001, p, 101)

The situation after leaving home also may be more problematic for girls, whether because of risks or survival strategies that are gendered. In their longitudinal study of 428 runaway and homeless youths in eight midwestern cities, Tyler et al. (2001) report sexual predation and victimization of runaway females. Hagan and McCarthy (1997) found that among youths living on the streets, girls were more likely than boys to become involved in prosti- tution. Hwang and Bedford (2003) assert that a common criminal career trajectory for run- away girls is prostitution, but their methodology is somewhat suspect as they base this finding on interviews with only adult prostitutes, many of whom report childhood sexual

312 Youth Violence and Juvenile Justice

abuse. In his developmental theory, Howell (2003) includes runaway as one stepping stone in the female-specific pathway to serious, violent, and chronic offending. Chesney-Lind (1989) refers to this as the “criminalization of girls’ survival strategies” (p. 11).

Finally, several authors argue that the response to runaway youths by juvenile justice sys- tems is gendered in a way that is problematic for girls. One specific concern is differential treat- ment resulting from the policy for deinstitutionalization of status offenders may have led to a double standard of justice for girls (Chesney-Lind & Shelden, 1998; Feld, 1999; Wiethorn, 1988). Feld (1999) argues that this policy contributes an element to the “triage” of juvenile jus- tice that “increasingly removes many middle class, White, and female juveniles to private insti- tutions and relegates young Black males to secure residential facilities” (p. 169). He considers this part of a larger difficulty in juvenile justice policy that allows competing visions of youths as vulnerable and dependent or as autonomous and responsible to affect wide discretionary application of the law. More recently, Bloom, Owen, Deschenes, and Rosenbaum (2005) argued that “the high incidence of runaway incidents bears a closer look, as this survival strat- egy may lead to incarceration” (p. 800). The probability of confinement is unknown, but this outcome was observed among a small number of juvenile case files in Washington.

[E]ven as they recognized the poor family environments the girls came from and the abuse they suffered, probation officers repeatedly recommended confinement in juvenile correc- tional facilities as a way for the girls to get out of their homes and to get counseling and treat- ment. (Inderbitzin, 2005, p. 789)

Our review shows that prior research has identified a variety of reasons youths may run away from home and that attention has focused on difficulties with family. Several schol- ars underscore a connection between runaway and child abuse, and especially sexual abuse. Given their vulnerabilities and minimal resources, the lives of many runaways also involve risks to their own well-being and to public safety. There is some indication that the juve- nile justice policy arena in which the government responds to runaway youths lacks both the appropriate means and the relevant knowledge about effective interventions with run- away youths. Moreover, the literature suggests that these difficulties are gendered in ways that exacerbate the problems for girls. The prevailing view on the gendered pathway for runaway girls is perhaps aptly summarized by Chesney-Lind (1997) as follows:

Some girls resort to panhandling and shoplifting; others theft for money, food, and clothes. Some exchange sex for these necessities and become involved in prostitution and drug abuse. Girls’ situations often are worsened by patriarchial law enforcement and justice systems that require girls to obey parents and stay at home. In many cases, girls are sent back to their vic- timizers. Even when taken out of their homes, courts have few placement alternatives for girls and sometimes are left only with returning them home. (p. 5)

The available criminology literature suggests that significant risks of crime and safety exist both in the home and on the streets for runaway youths. Moreover, published works indicate that the difficulties are exacerbated for girls, many of whom also subsequently receive ineffective, biased, and potentially harmful responses from juvenile justice systems. In our review of prior research, the problems of runaway youths seem far reaching and

Kempf-Leonard, Johansson / Gender and Runaways 313

314 Youth Violence and Juvenile Justice

gendered, yet it is also important to recognize that past inquiries fail to identify how far reaching these problems are. One reason this information does not yet exist is that delin- quency and juvenile justice have been secondary concerns in most prior research on run- aways. The majority of studies selected small samples of youths from shelter facilities and used in-depth interviews to obtain detailed information about their current lives and prior experiences. Without a criminological focus, researchers often do not distinguish between runaways and youths who are homeless for other reasons. Although orphans and youths “thrown away” by parents may experience similar vulnerabilities and strategies for survival as runaways, they do not share their legal status or evoke the same responses from juvenile justice systems. In most studies based on participants from shelters, the shelter experience was not court imposed.

Another measurement issue is that although the qualitative interview data provide rich details about the youths’ perceptions and experiences, they are not capable of conveying the generalizability of those perceptions and experiences beyond the immediate samples. This is particularly troubling in the research based on small samples of adjudicated females because we cannot know whether the difficulties observed are female specific. We cannot be certain that the difficulties are disproportionately borne by females when males are not in the sample or are not studied in the same way (Daly, 1994; Visher, 1983).

The runaway information available from official juvenile justice data sources also has suf- fered some methodological weaknesses. First, most court information systems record infor- mation on referrals rather than individual youths, and often it is routine to count only the most serious statute violation per referral. Thus, in these systems, runaway youths are lost when they also are charged as delinquents with misdemeanor or felony offenses. Second, given the referral focus, most systems also do not capture data on individual risk factors suf- ficiently to recreate criteria used by court officials in their decision making. Researchers must overcome these difficulties and learn whether the risks and poor treatment identified in the extant research are typical of runaways and whether the problems are gendered to the detriment of females, if we are to inspire public concern to the plight of these at-risk youths and to justify policy attention for establishing appropriate interventions. In the next section, we describe data that provide a valuable next step in this line of inquiry.

Data and Methods

The data we examine include all arrest referrals between 1997 and 2003 to a juvenile jus- tice system in one of four counties in a metropolitan area in Texas with more than 5 mil- lion people. This area is among the 10 most populous in the nation and among the most diverse. Although among the juveniles arrested, Hispanic and African American youths are overrepresented. Given the size of the population they serve, the juvenile justice system is a formal bureaucracy, with specialized court services and treatment, on-site prosecutorial and public defender staff, and a large secure detention facility. Most importantly for our purposes, of the 42,577 individual youths with charges, there are 6,473—65.3% of them girls—who are charged with the status offense of runaway and processed by the juvenile justice system.

Kempf-Leonard, Johansson / Gender and Runaways 315

Most of the background and initial referral data are recorded at intake screening by court officials who rely on self-reported information from the youths, parents, and police reports. Information about interventions and any changes in status are updated at each meeting with court officials through disposition. All data fields are completed. Although we cannot com- ment on the accuracy or comprehensiveness of these data, we can discern patterns across 7 years, which give us more confidence in the quality than would data for a single year.

Analyses are based on individual youths rather than referrals. We do this for two reasons. First, juvenile justice officials respond to youths rather than only specific charges. Sometimes, cases are merged, and officials base decisions on all information available. Analysis based on youths more closely corresponds to this actual practice of juvenile case processing. Second, all offense codes charged are recorded and not merely the most seri- ous charge, as is often the case in juvenile justice data. Thus, we can examine all types of charges for which the youths are processed. This is particularly important for observing runaway, which may be the lesser charges for runaway youths who come to the attention of police. Prior charges and court experiences are available only for referrals within the 1997 to 2003 time frame.

In addition to offense code information on multiple charges for a single referral, these data are unique in providing information about youths that typically is known to court offi- cials but not recorded. As such, these data enable statistical analysis to more closely repre- sent criteria used in court processing. Several data fields serve as indicators of youth risk or need. We know who the child lives with at each referral, including two parents, single parents, a variety of relatives and friends, foster care, group homes, and institutional set- tings. Information is recorded about child abuse, including physical, emotional, and sexual abuse, and whether the Child Protective Services agency has involvement with the family. Pregnancy and youths who are parents are also recorded. Substance abuse and gang involvement are also reported.

In the initial analyses, we provide descriptive comparisons of the individual characteris- tics, offenses, and juvenile justice experiences across distinct groups of youths: runaways and other youths referred to juvenile justice, female and male runaways, female and male runaways referred to court by age 13 (early onset), and female and male runaways with child abuse. We then examine multivariate models to determine (a) whether the risk factors help to distinguish runaways from other youths and whether gender makes a difference and (b) whether formal juvenile justice interventions with runaway youths differs by gender. Following the advice of Visher (1983) and Daly (1994) that we not merely “add-women- and-stir” (Daly, 1994, p. 121), the multivariate models are tested separately by gender so that the independent effects of all variables can be seen.

Results

Of the 42,577 youths processed by the juvenile justice system between 1997 and 2003, 6,473 youths had at least one referral for running away from home. Moreover, gender is associated with referrals for runaway. As Table 1 shows, females constitute the majority (65.3%) of youths with a runaway referral but constitute the minority (27.7%) of youths

316 Youth Violence and Juvenile Justice

referred to the juvenile court with no charges for runaway. There are other differences between the two groups as well. Somewhat more of the youths referred for a runaway vio- lation are White. Among both groups, most of the youths lived with a single parent, but a higher percentage of runaways lived with other relatives or family, or in foster care, group homes, or other institutional settings. Runaways also had more recorded child abuse than other referred youths for each of the five measures: emotional abuse; physical abuse; sex- ual abuse; formal involvement with the family by the state agency for abuse; and one or more of any of the four (29.7% runaways, 6.6% nonrunaways). A slightly higher percent- age of runaway youths than other youths in juvenile justice were either currently parents themselves or pregnant (girls only, obviously). Substance abuse was higher among run- aways, but gang involvement was similar in the two groups.

Table 1 also shows that there is a statistically significant difference (t tests) in the risk factors between females and males among only runaway youths. The highest proportions

Table 1 Percentages for Demographic and Risk Characteristics of Youths by Trait

All Youths Referred to Juvenile Justice Runaways

No Runaway Runaway Females Males Trait (36,101) (6,473) Significance (4,228) (2,244) Significance

Sex p < .001 Female 27.7 65.3 Male 72.3 34.7

Race or ethnicity White 26.6 38.3 p < .001 39.2 36.6 p < .001 Black 36.0 30.8 p < .001 27.8 36.5 p < .001 Hispanic 35.0 29.0 p < .001 30.9 25.4 p < .001 Other 2.3 1.9 NS 2.1 1.6 NS Age, 1st charge ≤ 13 15.4 15.9 NS 14.1 19.3 p < .001

Lives with Two parents 21.7 18.2 p < .001 19.5 15.7 p < .001 Single parent 51.2 46.3 p < .001 48.1 43.0 p < .001 Friend or relative 17.8 23.1 p < .001 22.5 24.3 NS Foster care, group home,

or institution 4.7 7.4 p < .001 5.2 11.6 p < .001 Other or unknown 4.6 5.0 NS 4.8 5.3 NS

Risk factors Emotional abuse or neglect 2.1 13.9 p < .001 16.8 8.4 p < .001 Physical abuse 3.4 18.2 p < .001 20.4 14.2 p < .001 Sexual abuse 3.4 14.0 p < .001 19.6 3.5 p < .001 Child welfare agency involved 3.4 17.5 p < .001 20.2 12.6 p < .001 Any abuse 6.6 29.7 p < .001 34.8 20.0 p < .001 Pregnant or already a parent 2.9 5.2 p < .001 6.9 1.9 p < .001 Drug abuse 36.2 45.5 p < .001 42.4 52.4 p < .001 Gang involvement 7.6 7.9 NS 5.3 12.8 p < .001

Note: NS = not significant.

Kempf-Leonard, Johansson / Gender and Runaways 317

of both girls and boys were White, but then differences emerge with more runaway Black boys and Hispanic girls. Foster care, group homes, and institutional living were higher among runaway boys. Child abuse, according to all available indicators, was higher among runaway girls, as was pending or current parenthood. Both groups, but particularly boys, had a high level of substance abuse indicated. Gang involvement among boys was more than double that recorded for girls.

The distributions of offense characteristics and juvenile justice interventions are shown in Table 2. Most youths were referred only once for a single offense, however, higher pro- portions of runaways than other youths involved with the court had multiple charges. Multiple charges were significantly higher among runaway boys than girls. Three fourths of the runaways had only one runaway offense, but one fourth of them were charged mul- tiple times with runaway violations. There was no difference by gender. One youth accu- mulated 15 runaway charges (not shown).

Table 2 also shows several types of offenses for which the youths were charged. There were 111 homicide charges, including 7 among youths who were runaways. Serious charges, including homicide, robbery, burglary, aggravated assault, and weapons violations, were not as common among runaways as other youths involved with the juvenile justice system. More than 10% of runaways also were charged with theft or simple assaults sometime between 1997 and 2003. Half of the youths charged with prostitution were girls who also had a run- away charge. When only youth with at least one runaway charge are compared, the findings show a statistically significant difference with boys more involved than girls with other types of offending, particularly theft, assault, burglary, drug violations, and weapons.

Finally, Table 2 shows the most restrictive outcome the youths ever received from the juvenile justice system during the 1997 to 2003 time frame. All charges were dismissed for 7.4% of runaways—slightly higher at 8.5% among girls compared to 5.5% among boys— and 13.5% of youths with no runaway charges. More than one third of other youths and 11.7% of runaways had their referrals deferred, which meant they were set aside and would have a petition filed only if a new referral occurred. The most common juvenile justice intervention with runaway youths, particularly girls, was a stern warning cautioning them against further offending. The remaining interventions involve formal outcomes following petitions and adjudication. Approximately 3% of the youths had no recorded disposition for referrals during this period, but some of these had cases merged with other referrals in years outside those studied. Probation was the most restrictive intervention for 27.1% of other youths and 17.2% of all runaways, but 28.0% of runaway boys and only 11.5% of runaway girls. Out of home commitment was the outcome for 4.4% of other youths and 3.6% of run- aways, but 8.2% of runaway boys and only 1.1% of girls. Just more than 1% of youths were certified as adults, and the highest level was among runaway boys (1.7%).

Table 3 shows the results of logit regressions of runaway on risk factors for females and males separately. The table compares the effects of the risk factors for youths charged with at least one runaway status offense to all other youths charged with offenses but no run- away. The regression analysis for females shows that with respect to family structure, females who live with a single parent are less likely to have a runaway charge than those who live with two parents. In contrast, living in foster care, group home, or institution improves the chances of being charged as a runaway. Analysis of the effect of race indicates

318 Youth Violence and Juvenile Justice

that African American and Hispanic females are significantly less likely to have a runaway charge than White females.

The effect of child abuse experiences on likelihood of a runaway referral is positive and significant after controlling for other relevant risk factors. Among females, odds of being a runaway increases 3.4 times with abuse victimization. Similarly, the coefficient for substance abuse is statistically significant and increases the likelihood of being a runaway.

Table 2 Percentages for Offense Characteristics/Juvenile Justice Interventions by Trait

All Youths Referred to Juvenile Justice Runaways

No Runaway Runaway Females Males Trait (36,101) (6,473) Significance (4,228) (2,245) Significance

Total # charges p < .001 p < .001 1 66.6 47.1 54.2 33.7 2 to 4 27.7 38.9 37.3 41.9 5 or more 5.7 14.0 8.5 24.4

# runaway charges NS 1 76.5 75.5 78.4 2 or more 23.5 24.5 21.6

Offense types Homicide (104) (7) (2) (5) Robbery 3.0 1.6 p < .001 .7 3.4 p < .001 Burglary 14.6 7.2 p < .001 2.0 17.0 p < .001 Aggravated assault 8.4 3.6 p < .001 2.2 6.1 p < .001 Other assault 15.5 12.2 p < .001 9.2 18.0 p < .001 Theft 30.1 17.5 p < .001 13.3 25.3 p < .001 Drug violation 15.6 6.7 p < .001 3.1 13.5 p < .001 Property damage 6.3 3.3 p < .001 1.3 7.1 p < .001 Disturbance 2.0 1.3 p < .001 .9 2.0 p < .001 Forgery or fraud 1.3 1.5 NS 1.7 1.3 NS Truancy 5.7 4.4 p < .001 2.2 8.6 p < .001 Curfew .8 1.6 p < .001 1.1 2.4 p < .001 Weapons violation 3.4 1.3 p < .001 .5 2.9 p < .001 Prostitution (37) (36) (36) (0)

Most restrictive outcome Dismissed 13.5 7.4 8.5 5.5 Deferred 36.1 11.7 12.0 11.1 Cautioned 14.4 56.8 64.1 43.0 Consolidated

or no disposition 3.3 2.3 2.3 2.4 Probation 27.1 17.2 11.5 28.0 Commitment 4.4 3.6 1.1 8.2 Certified 1.2 1.0 .5 1.7

Note: Because of the small number of cases, numbers are reported for homicide and prostitution rather than percentages. NS = not significant.

Kempf-Leonard, Johansson / Gender and Runaways 319

Early onset of offending (a court referral by age 13), being pregnant or already a parent, or being involved with a gang are not significantly related to being a runaway among females.

The regression results for males show that early onset is positively related to having a referral as a runaway, unlike the model for females in which the early onset coefficient was insignificant. Regarding family structure, several alternative family constellations are pos- itively related to being a runaway. Living with a relative or friend and living in a foster fam- ily, group home or institution, and unknown or other residential situations make the odds of having a runaway charge more likely than living with two parents for boys. However, living with a single parent does not have a significantly different effect for the risk of being a runaway.

As was shown for girls, boys are more likely to have a runaway charge if they are White than if they are African American, Hispanic, or members of another racial group. Boys with known or suspected child abuse are also more apt to runaway. Substance abuse also is more common among runaway boys. In contrast to the girls, the coefficient for gang involvement is positive and statistically significant for boys, indicating that for them, gang involvement is a risk factor associated with running away.

In sum, for both girls and boys, the likelihood of runaway referrals to juvenile court increases if they are White, are known or suspected victims of child abuse, have substance

Table 3 Runaway Regressed on Risk Variables by Gender.

Females Males

B SE Exp(B) B SE Exp(B)

Constant –.943** .054 –2.76** .068 Early onset .123 .057 1.13 .201* .058 1.22 Family structure

Single parenta –.134* .052 .875 .091 .066 1.09 Relative or friend .135 .062 1.14 .457** .074 1.58 Foster, group, institution .345* .113 1.41 .604** .093 1.83 Other or unknown .183 .099 1.20 .445** .114 1.56

Race or ethnicity African American –.641** .049 .53 –.292** .054 .747 Hispanic –.374** .048 .69 –.608** .060 .544 Other raceb –.335 .136 .72 –.566* .180 .568

Parent or pregnant .006 .080 1.01 –.117 .168 .890 Any abuse 1.230** .049 3.42 1.359** .063 3.89 Substance abuse .481** .042 1.62 .411** .047 1.51 Gang involvement –.006 .096 .99 .280** .073 1.32 Log likelihood –7951.5385 –7375.2552 Pseudo R-squared .0798 .0565 N 14,204 28,192

aCategory is living with two parents. bReference category is White. *p < .01. **p < .001.

320 Youth Violence and Juvenile Justice

abuse problems, and live in a group setting such as foster care. Runaway girls are less likely to live with a single parent than both parents. Although not significant factors for girls, early involvement with juvenile justice prior to age 13, gang involvement, and living with friends or relatives instead of parents help to distinguish runaway boys from others involved with juvenile justice.

Next, we try to determine what factors affect juvenile justice interventions with run- away youths. This is somewhat difficult because, as was shown in Table 2, many runaway youths also are involved with juvenile justice for other types of offenses. Moreover, it is not uncommon for youths to have juvenile justice involvement throughout their adoles- cence. Table 4 and Table 5 shows that although most youths had offense referrals during only a single year, male runaways had more multiple year offending than other youths. The results also show that for the majority of youths with a runaway charge, this occurred during their initial year of juvenile justice involvement. During the initial year, the most restrictive response from the juvenile justice system to runaway youths was a cautionary warning (60.4%), followed by an equally informal warning via deferring the case (17.2%), and probation (10.1%). Cautionary warnings are higher among girls, and probation is higher among boys. However, the more restrictive outcomes for more boys may be because of their other charges. This appears even more likely when looking at Year 1 run- aways with no other types of charges. Among the 2,871 girls and 612 boys (72.8% of 3,944 and 56.2% of 1,862, respectively), there was almost no variation from cautionary warnings (not shown).

To identify the factors that affect juvenile justice interventions with runaways simulta- neously controlling for all others, we restrict the analysis to youths with a runaway refer- ral during their 1st year of involvement. For their most restrictive juvenile justice response

Table 4 Percentages for Initial Referral Year Experiences

All Youths Referred to Juvenile Justice Runaways

No Runaway Runaway Females Males Trait (36,101) (6,473) Significance (4,228) (2,245) Significance

Single year 76.9 62.9 p < .001 69.3 50.9 p < .001 Multiple year 23.1 37.1 30.7 49.1 Runaway 1st year 89.7 93.3 83.0 p < .001 1st year most restrictive outcome p < .001 p < .001

Dismissed 14.7 9.4 10.0 8.3 Deferred 36.5 17.2 15.9 19.5 Cautioned 21.9 60.4 66.1 49.6 Consolidated or no disposition 3.2 2.1 1.7 2.8 Probation 21.3 10.1 5.8 18.2 Commitment 1.6 .4 .2 1.0 Certified .7 .4 .3 .6

Kempf-Leonard, Johansson / Gender and Runaways 321

during that initial year, we distinguish formal outcomes (probation, commitment, certifi- cation = 1) from informal outcomes (cautionary warnings, deferred cases = 0) and omit youths whose cases were dismissed. The logit regression results are shown separately by gender in Table 6.

For both girls and boys, formal juvenile justice interventions are more likely when run- aways are living in group settings without two parents, are African Americans, have sub- stance abuse problems, gang involvement, and other charges. Age is not a significant factor for interventions with girls; the likelihood of formal dispositions for runaway boys increase when they are older than age 13. Boys living with friends and relatives rather than two par- ents also have more chance of a formal outcome for runaway. Although the negative coef- ficient suggests informal warnings, child abuse victimization is not significant in juvenile court responses to runaway youths regardless of gender.

To summarize, the findings show that the gender relationship among most juvenile court referrals reversed itself for runaway charges, which were more common for girls. A race effect is also evident with more runaways who were White. Child abuse was indicated more often among runaways than youths with other charges. Approximately one third of the female runways had known or suspected child abuse, but the same was true for only one fifth of the male runaways. Substance abuse was also a problem for many runaway youths. Running away is not the only juvenile offense for many youths, and additional charges and referrals are more common for boys. When all variables are controlled simultaneously in the multivariate analyses, findings show that runaways, both girls and boys, are signifi- cantly different in terms of risk factors than youths with other reasons for juvenile court involvement.

Juvenile justice interventions tended to be informal warnings more often for runaways than other youths. For both girls and boys, warnings were the most common response, but runaway boys received probation and commitment more than runaway girls. These gender differences in the initial findings likely reflect the other offending, including serious charges of homicide, robbery, burglary, aggravated assault, and weapon violations that was more common among runaway boys than girls. When serious and other offending is con- trolled for statistically, juvenile justice dispositions were related to race—Black youths with more restrictive outcomes, in addition to substance abuse, gang involvement, and liv- ing situation for both runaway girls and boys. Child abuse victimization did not signifi- cantly affect responses to runaways from the juvenile justice system.

Table 5 Runaway During 1st Year and Seriousness of Charges That Year

Females (%) Males (%) (N = 3,944) (N = 1,862)

Only runaway 72.8 56.2 Runaway and nonserious charges 25.2 32.9 Runaway and serious chargesa 2.1 11.0

aSerious charges include homicide, aggravated assault, robbery, burglary, and weapons violations.

322 Youth Violence and Juvenile Justice

Discussion and Conclusions

This study has examined characteristics, offending, and juvenile justice interventions among a large group of youths referred for a runaway offense. Runaways have been com- pared to other youths involved with juvenile justice. The large number of youths has also facilitated gender comparisons among runaways. Some of the findings support relation- ships that have been identified in previous studies, but others have never before been exam- ined to provide information that is new.

Unlike most of the research on runaways, which identify participants from residential shelters, this study examines only runaways identified by legal charges of runaway. The data used in this study provide information on several risk factors of interest, offense charges, and juvenile justice interventions. They are valuable both in scope and quality compared to information routinely available in many juvenile justice systems. The data are not without limitations, however, including two which have implications for our findings.

Table 6 Offense Seriousness Controlled by Gender

Females Males

B SE Exp(B) B SE Exp(B)

Constant –5.958** .374 –4.720** .333 Early onset .037 .228 1.03 –.936** .238 .39 Family structure

Single parenta .058 .254 1.06 .416 .256 1.51 Relative or friend .110 .284 1.12 .810* .267 2.25 Foster, group, institution 1.257** .331 3.51 1.667** .303 5.30 Other or unknown .482 .472 1.62 –.043 .533 .96

Race or ethnicity African American .548* .202 1.73 .649* .195 1.91 Hispanic .092 .216 1.10 .226 .220 1.25 Race otherb –.693 .870 .50 .340 .653 1.41

Parent or pregnant .602 .263 1.83 –.079 .484 .92 Any abuse –.093 .168 .911 –.312 .198 .73 Substance abuse .705** .177 2.02 .576* .170 1.78 Gang involvement .950** .240 2.59 .957** .208 2.60 Delinquency Year 1

Not serious charges 3.373** .285 29.16 2.689** .225 14.71 Serious charges 5.467** .372 236.82 4.793** .282 120.67 Log likelihood –520.97133 –516.29168 Pseudo R-squared .3637 .4098 N 3,496 1,672

Note: Youths with a runaway referral during first referral year only. Total sample size n = 5,168 because of omission of cases dismissed at disposition (n = 626) and missing cases in age variable (n = 13). aReference category is living with two parents. bReference category is White. cReference category is only runaway charges during 1st year. *p < 0.01. **p < 0.001.

Kempf-Leonard, Johansson / Gender and Runaways 323

First, most of the indicators used to measure risk factors and other characteristics of the youths were recorded by intake court staff members noting information they received directly from youths, their parents, or police officers. The extent to which this information was verified for accuracy through cross-checks is apt to have been minimal. Child abuse is one example. Some of the youths had no recorded child abuse, yet the child welfare agency was known to be involved with the family. This discrepancy might be explained by siblings. Most importantly, there is no way to discern whether gender differences in recorded abuse are valid or because of gender differences in self-reports or perceptions of gender differences on the part of intake staff.

Second, when youths have numerous referrals, the information recorded at the most recent intake screening supplants the original records for most of the risk measures. This may pose a causal order dilemma in that substance abuse, gang involvement, and residential setting may follow rather than precede the runaway experience. This situation is less apt to be problematic in the analyses of juvenile justice outcomes because they were restricted to the year of the initial runaway year, and thus the information was more current.

Despite limitations, these data provide a valuable foundation for criminological research on runaways, including information on gender that has not previously been available. The large number of runaways allow for observation of patterns while controlling simultaneously for risks and offending. The abilities to test models separately by gender and to control for diver- sity by race and ethnicity, which also may differ by gender, are particularly valuable. The busy bureaucracy of this juvenile justice system includes many court officials, thus patterns of responses are not likely to reflect biases of individual intake officers, prosecutors, or judges.

The results support those found elsewhere that girls are more often brought to court for runaway status offenses. More runaway girls have been victims of child abuse, including sexual abuse, than other girls who are involved in juvenile justice. This is one offense for which the White majority is represented, even after risk factors are controlled. Single par- ents make no difference for sons but may serve as a protective factor over two parents for daughters. It may be that problematic parent-child relationships, including sexually abusive fathers, involve the parent not in residence or that single parents are equally and sometimes better at providing the parental warmth previously found important (Chapple et al., 2004). Runaway girls also have higher rates of substance abuse problems and gang involvement, although this may occur after running away. The findings also support concern that at least some runaway girls, whether because of vulnerabilities or as strategies for survival on the streets, engage in prostitution, theft, forgery, and fraud.

In addition, some of our findings differ somewhat from relationships that previous authors led us to expect. Rather than incarceration (Bloom et al., 2005; Inderbitzin, 2005), the juvenile justice response to runaway girls is an informal caution against further offend- ing, after which girls are sent home and court involvement with the youths is discontinued. Runaway girls who do receive formal dispositions more often are African American than White. The risk factors of child abuse, substance abuse, gang involvement, foster care, and other group living situations are factors for runaway boys just as they are for runaway girls. Juvenile justice interventions with runaway boys, even controlling for other offending, is somewhat more restrictive than it is for girls. This is likely associated with their greater number of arrests, which is a gendered finding similar to Chapple et al. (2004). Juvenile justice interventions with runaway youths are affected by substance abuse, gang involve- ment, and living situation, but surprisingly not by child abuse.

324 Youth Violence and Juvenile Justice

The results of this study suggest implications both for future research to explain what happens to runaway youths and for procedural changes within juvenile justice systems that will make interventions with runaways more beneficial for the individual youths and reduce their potential for subsequent offending. One aim for future research is to determine whether risky behaviors among runaways more often reflect survival strategies or predatory victimization because of their vulnerable status. This knowledge would enable police and child welfare officials to develop prevention and intervention strategies that would be more effective. It also would be beneficial to extend the study of runaway youths into adulthood with longitudinal investigations of long-term effects of their childhood experiences. Both of these types of inquiries need to include girls and boys.

Our findings also provide fairly compelling evidence that child abuse victimization con- stitutes a risk for runaway differently than for other juvenile offending. The nature of that abuse may differ for girls and boys, and the impact may also be gendered in other ways, but the effect promoting runaway is similar. Substance abuse is also a factor; although whether drug use is another effect of victimization is not identified herein. What is partic- ularly noteworthy, especially given the literature that led us to expect more gender differ- ence, is the finding that juvenile justice interventions with runaways are very minimal and do not reflect attention to child abuse. Thus, it seems clear that juvenile justice officials should be encouraged to pay attention to the abuse experiences of youths, as well as their substance abuse, gang involvement, and living situations.

Juvenile justice officials must have ways to respond to runaway youths that are effective. Our findings suggest that at least in this system, but also quite likely in most juvenile justice systems, the main problem is lack of alternatives and understanding about how best to respond to runaways. Although there are gender differences in how adolescents develop their self-identities in relation to their experiences in the world (Gilligan, 1982) that need to be addressed in gender-specific ways, both runaway girls and boys also need caring, professional intervention that provides concrete assistance in setting boundaries and teaching responsibil- ity and with communication, family relationships, and living arrangements (Kurtz et al., 2000). Beyond a cautionary warning, runaways merit attention regardless of other behavior. This does not mean that the intervention must involve residential placement, but it does indi- cate that formal supervision of the youths’ residential situation, whether in the home, else- where, or independently, and services for healthy adolescent development should be required. Without it, subsequent referrals including more serious offenses, are quite likely.

References

Acoca, L. (1998). Outside/inside: The violation of American girls at home, on the streets, and in the juvenile justice system. Crime & Delinquency, 44(4), 561-589.

Bass, D. (1992). Helping vulnerable youths: Runaway and homeless adolescents in the United States. Washington, DC: National Association of Social Workers Press.

Belknap, J. (2001). The invisible woman: Gender, crime, and justice (2nd ed.). Belmont, CA: Wadsworth. Belknap, J., & Holsinger, K. (1998). An overview of delinquent girls: How theory and practice have failed and

the need for innovative change. In R. T. Zaplin (Ed.), Female offenders: Critical Perspectives and effective interventions (pp. 31-64). Gaithersburg, MD: Aspen.

Bernard, T. J. (1992). The cycle of juvenile justice. New York: Oxford University Press.

Kempf-Leonard, Johansson / Gender and Runaways 325

Bishop, D., & Frazer, C. (1995). Gender bias in juvenile justice processing: Implications of the JJDP Act. Journal of Criminal Law and Criminology, 82, 1162-1186.

Bloom, B., Owen, B., Deschenes, E. P., & Rosenbaum, J. (2005). Developing gender-specific services for delin- quency prevention: Understanding risk and resiliency. In R. Muraskin (Ed.), It’s a Crime: Women and Justice (4th ed., pp. 792-819). Upper Saddle River, NJ: Pearson Education.

Bucy, J., & Obolensky, N. (1990). Runaway and homeless youths. In M. Rotheram-Borus, V. Bradley, & N. Obolensky (Eds.), Planning to live: Evaluation and treating suicidal teens in community settings (pp. 333-353). Tulsa: University of Oklahoma Press.

Castellano, T. C. (1986). The justice model in the juvenile justice system: Washington State’s experiences. Law & Policy, 8, 397-418.

Centers for Disease Control. (1991, June). HIV/AIDS surveillance report. Atlanta, GA: Centers for Disease Control. Chapple, C. L., Johnson, K, D., & Whitbeck, L. B. (2004). Gender and arrest among homeless and runaway youth:

An analysis of background, family, and situational factors. Youth Violence and Juvenile Justice, 2, 129-147. Chen, X., Tyler, K. A., Whitbeck, L. B., & Hoyt, D. R. (2004). Early sexual abuse, street adversity, and drug

use among female homeless and runaway adolescents in the Midwest. Journal of Drug Issues, 1, 1-22. Chesney-Lind, M. (1973). Judicial enforcement of the female sex role. Issues in Criminology, 8, 51-70. Chesney-Lind, M. (1988). Girls and status offenses: Is juvenile justice still sexist? Criminal Justice Abstracts,

20, 144-165. Chesney-Lind, M. (1995). Girls, delinquency, and juvenile justice: Toward a feminist theory of young women’s

crime. In B. R. Price & N. J. Sokoloff (Eds.), The criminal justice system and women: Offenders, victims, and workers (pp. 71-88). New York, McGraw-Hill.

Chesney-Lind, M. (1997). Female offenders: Girls, women, and crime. Thousand Oaks, CA: Sage. Chesney-Lind, M., & Shelden, R. (1998). Girls, delinquency, and juvenile justice (2nd ed.). Pacific Grove, CA:

Brooks/Cole. Crespi, T. D., & Sabatelli, R. M. (1993). Adolescent runaways and family strife: A conflict induced resolution

framework. Adolescence, 28, 867-878. Daly, K. (1994). Gender and punishment disparity. In M. Myers & G. Bridges (Eds.), Inequality, crime, and

social control (pp. 117-133). Boulder, CO: Westview. Datesman, S. K., & Scarpetti, F. R. (1980). Unequal protection for males and females in the juvenile court. In

S. K. Datesman & F. R. Scarpitti (Eds.), Women, crime and justice (pp. 300-319). New York: Oxford University Press.

Feld, B. C. (1999). Bad kids: Race and the transformation of the juvenile court. New York: Oxford University Press. Finkelhor, D., Hotaling, G. T., Lewis, I. A., & Smith, C. (1990). Sexual abuse in a national survey of adult men

and women: Prevalence, characteristics, and risk factors. Child Abuse & Neglect, 14, 19-28. Foti, S. (1995). Child sexual abuse as a precursor to prostitution. Dissertation Abstracts International, 55, 8b.

(UMI No. AAM9433836) Garbarino, H., Schellenbach, C. J., & Seber, J. (1987). Troubled youth, troubled families. New York: Alpine De

Gruyter. Gilfus, M. E. (1992). From victims to survivors to offenders: Women’s routes of entry and immersion into street

crime. Women and Criminal Justice, 4, 63-90. Gilligan, C. (1982). In a different voice: Psychological theory and women’s development. Cambridge, MA:

Harvard University Press. Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homelessness. New York: Cambridge

University Press. Hammer, H., Finkelhor, D., & Sedlack, A. (2002). Runaway/thrownaway children: National estimates and

characteristics. (NISMART: National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children). Washington, DC: U.S. Dept. of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Howell, J. C. (2003). Preventing and reducing juvenile delinquency: A comprehensive framework. Thousand Oaks, CA: Sage.

Hwang, S., & Bedford, O. (2003). Precursors and pathways to adolescent prostitution in Taiwan. Journal of Sex Research, 40(2), 201-210.

326 Youth Violence and Juvenile Justice

Inderbitzin, M. (2005). The impact of gender on juvenile justice decisions. In R. Muraskin (Ed.), It’s a crime: Women and justice (4th ed., pp. 782-790). Upper Saddle River, NJ: Pearson Education.

Janus, M., Burgess, A., & McCormack, A. (1987). Histories of sexual abuse in adolescent male runaways. Adolescence, 22, 405-417.

Jencks, C. (1994). The homeless. Cambridge, MA: Harvard University Press. Johnson, K. D., Whitbeck, L. B., & Hoyt, D. R. (2005). Substance abuse disorders among homeless and run-

away adolescents. Journal of Drug Issues, 2, 799-816. Kaufman, J. G., & Widom, C. S. (1999). Childhood victimization, running away, and delinquency. Journal of

Research in Crime and Delinquency, 36(4), 347-371. Kett, J. F. (1977). Rites of passage: Adolescence in America 1790 to present. New York: Basic Books. Kipke, M. S., Montgomery, S. B., Simon, T. R., & Iverson, E. F. (1997). Substance abuse disorders among run-

away and homeless youth. Substance Use and Misuse, 32, 969-986. Krisberg, B., Schwartz, I. M., Fishman, G., Eisikovits, Z., & Guttman, E. (1986). The incarceration of minor-

ity youth. Minneapolis, MN: Hubert Humphrey Institute of Public Affairs. Kurtz, P. D., Jarvis, S. V., & Kurtz, G. (1991). The problems of homeless youth: Empirical findings and human

service issues. Social Work, 36, 309-314. Kurtz, P. D., Lindsey, E. W., Jarvis, S., & Nackerud, L. (2000). How runaway and homeless youth navigate

troubled waters: The role of formal and informal helpers. Child and Adolescent Social Work Journal, 17 (5), 381-402.

Luntz, B. K., & Widom, C. S. (1994). Antisocial personality disorder in abused and neglected children grown up. American Journal of Psychiatry, 151, 670-674.

McClanahan, S., McClelland, G., Abram, K., & Teplin, L. (1999). Pathways into prostitution among female jail detainees and their implications for mental health services. Psychiatric Services, 50, 1606-1613.

McCord, J., Widom, C. S., & Crowell, N. A. (2001). Juvenile crime, juvenile justice. Washington, DC: National Academy.

McCormack, A., Janus, M., & Burgess, A.W. (1986). Runaway youths and sexual victimization: Gender dif- ferences in adolescent runaway populations. Child Abuse & Neglect, 10, 387-395.

McMorris, B. J., Tyler, K. A., Whitbeck, L. B., & Hoyt, D. R. (2002). Familial and “on-the-street” risk factors associated with alcohol use among homeless and runaway adolescents. Journal of Studies on Alcohol, 63, 34-43.

Miller, H. G., Turner, G. F., & Moss, L. E. (1990). AIDS: The second decade. Report of the National Research Council. Washington DC: National Academy Press.

Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental tax- onomy. Psychological Review, 100(4), 674-701.

Moss, D. C. (1986, December 1). Getting tough with juveniles: Justice department urges changes. ABA Journal, p. 28.

Myers, T., & Sangster, J. (2001). Retorts, runaways and riots: Patterns of resistance in Canadian reform schools for girls, 1930-60. Journal of Social History, 34, 669-697.

Obeidallah, D. A., & Earls, F. J. (1999). Adolescent girls: The role of depression in the development of delin- quency. Research preview, July. Washington, DC: U.S. Department of Justice, National Institute of Justice.

Odem, M. (1995). Delinquent daughters: Protecting and policing adolescent female sexuality in the United States, 1885-1920. Chapel Hill: University of North Carolina Press.

Odem, M. E., & Schlossman, S. (1991). Guardians of virtue: The juvenile court and females delinquency in early 20th century Los Angeles. Crime and Delinquency, 37, 186-203.

Parham v. J. R., 442 U.S. 609 (1979). Pipher, M. (1994). Reviving Ophelia: Saving the selves of adolescent girls. New York: Ballantine. Robertson, M. J. (1989). Homeless youth in Hollywood: Patterns of alcohol use. Berkeley, CA: School of

Public Health, University of California–Berkeley. Rotheram-Borus, M. J., Meyer-Bahlburg, H. F. L., Koopman, C., Rosario, M., Exner, T. M., Henderson, R., et al.

(1992). Lifetime sexual behaviors among runaway males and females. Journal of Sex Research, 29(1), 15-29. Rothman, D. (1980). Conscience and convenience: The asylum and its alternative in progressive America.

Boston: Little, Brown.

Kempf-Leonard, Johansson / Gender and Runaways 327

Rutter, M. (1989). Pathways from childhood to adult life. Journal of Child Psychology & Psychiatry, 30, 23-51. Schlossman, S. (1977). Love and the American delinquent: The theory and practice of “progressive” juvenile

justice. Chicago: University of Chicago Press. Schlossman, S., & Wallach, S. (1978). The crime of precocious sexuality: Female juvenile delinquency in the

Progressive era. Harvard Educational Review, 48, 65-94. Schneider, A. (1984). Divesting status offenses for juvenile court jurisdiction. Crime & Delinquency, 30, 347-370. Schwartz, I. M. (1989). (In)justice for juveniles: Rethinking the best interests of the child. Lexington, MA:

Lexington Books. Slesnick, N. (2001). Variables associated with therapy attendance in runaway substance abusing youth:

Preliminary findings. American Journal of Family Therapy, 29, 411-420. Snyder, H., & Sickmund, M. (2006). Juvenile offenders and victims: 2006 national report. Pittsburgh, PA:

National Center for Juvenile Justice. Spillane-Grieco, E. (1984). Characteristics of a helpful relationship: A study of empathic understanding and

positive regard between runaways and parents. Adolescence, 73, 63-75. Spillane-Grieco, E. (2000). From parent verbal abuse to teenage physical aggression. Child and Adolescent

Social Work Journal, 17(6), 411-430. Stiffman, A. R., Earls, F., Powell, J., & Robins, L. N. (1987). Correlates of alcohol and illicit drug use in

adolescent medical patients. Contemporary Drug Problems, 14, 295-314. Stricof, R., Kennedy, J., Nattell, T., Weifuse, I., & Novick, L. (1991). HIV seroprevalence in a facility for

runaway and homeless adolescents. American Journal of Public Health, 81, 50-53. Sussman, A. (1977). Sex-based discrimination and PINS jurisdiction. In L. E. Tietelbaum & A. R. Gough

(Eds.), Beyond control: Status offenders in the juvenile court (pp. 179-199). Cambridge, MA: Ballinger. Tanenhaus, D. S. (2004). Juvenile justice in the making. New York: Oxford University Press. Tyler, K. A., Hoyt, D. R., Whitbeck, L. B., & Cauce, A. M. (2001). The impact of childhood sexual abuse on

later sexual victimization among runaway youth. Journal of Research on Adolescence, 11(2), 151-176. Tyler, K., Whitbeck, L., Hoyt, D., & Cause, A. (2004). Risk factors for sexual victimization among male and

female homeless and runaway youth. Journal of Interpersonal Violence, 19, 503-520. Visher, C. (1983). Gender, police arrest decisions, and notions of chivalry. Criminology, 21, 5-28. Weithorn, L. A. (1988). Mental hospitalization of troublesome youth: An analysis of skyrocketing admission

rates. Stanford Law Review, 40, 773-838. Whitbeck, L. B., Chen, X., Hoyt, D., Tyler, K. A., & Johnson, K. D. (2004). Mental disorder, subsistence strate-

gies, and victimization among gay, lesbian, and bisexual homeless and runaway adolescents. Journal of Sex Research, 41(4), 329-342.

Whitbeck, L. B., Hoyt, D., Yoder, K., Cauce, A., & Paradise, M. (2001). Deviant behavior and victimization among homeless and runaway adolescents. Journal of Interpersonal Violence, 16, 1175-1204.

Whitbeck, L. B., & Simons, R. I. (1990). Life on the streets: The victimization of runaway and homeless ado- lescents. Youth & Society, 22 (1), 108-125.

Widom, C. S., & Ames, M. A. (1994). Criminal consequences of childhood sexual victimization. Child Abuse & Neglect, 18, 303-318.

Yates, G. L., Mackenzie, R., Pennbridge, J., & Cohen, E. (1988). A risk profile comparison of runaway and non- runaway youth. American Journal of Public Health, 78, 820-821.

Zide, M., & Cherry, A. (1992). A typology of runaway youths: An empirically based definition. Child and Adolescent Social Work Journal, 9, 155-168.

Kimberly Kempf-Leonard is professor and chair of the Center for Study of Crime, Delinquency, and Corrections at Southern Illinois University Carbondale. Her research examines inequality, criminal and juvenile justice policy, and criminal career development.

Pernilla Johansson is assistant director of the Eugene McDermott Scholars Program at the University of Texas at Dallas. Her research examines the life course perspective, gender and crime, and social control.