We've Had Three of Them: Addressing the Invisibility of Lesbian, Gay, Bisexual, and Gender Nonconforming Youths in the Juvenile Justice System "WE'VE HAD THREE OF THEM": ADDRESSING THE INVISIBILITY OF LESBIAN, GAY, BISEXUAL AND GENDER NON- CONFORMING YOUTHS IN THE JUVENILE JUSTICE SYSTEM ANGELA IRVINE, PH.D.* Two researchers were speaking to a high-ranking probation officer from a large city, trying to convince her to participate in a project on lesbian, gay, bisexual and questioning ("LGB"), and gender non- conforming youths.' Her first response was, "I've worked in this system for * Ph.D. in Sociology, Northwestern University; principal researcher of evaluation projects at Ceres Policy Research. Dr. Irvine's research interests include youth development, education, gender specific programming with the juvenile justice system, and parolee reentry. 1 For the purpose of this paper, "lesbian" is defined as a girl or woman who primarily is emotionally, romantically, and sexually attracted to girls or women. "Gay" is defined as a person who is emotionally, romantically, and sexually attracted to individuals of the same sex, typically in reference to boys or men. "Bisexual" is defined as a person who is emotionally, romantically, and sexually attracted to both males and females. "Transgender" is defined as a person whose gender identity (their understanding of themselves as male or female) does not correspond with their birth sex. A transgender girl is a girl whose birth sex was male but who understands herself to be female. A transgender boy is a boy whose birth sex was female but who understands himself to be male. For the purpose of this paper, "gender identity" is defined as a person's internal sense of being a man, boy, woman, girl, or somewhere in between. "Gender expression" is defined as the physical manifestation of one's gender identity, usually through clothing, mannerisms, and chosen names. The term "gender non-conforming" refers to youths who have gender identities or gender expressions that break social norms. Youths with non-conforming gender identities have identities that are different than the genders they were assigned to at birth. For example, a person with male genitalia that identifies as a girl has a non-conforming gender identity. Youths with a non- conforming gender expressions act in ways contrary to gender norms. Such youths may express their gender by, among other things, wearing non-conforming hairstyles, choosing non-conforming names, or behaving in other ways that break social norms. For example, a girl with a non-conforming gender expression may still identify as a girl but choose to wear masculine clothing and hairstyles. Columbia Journal of Gender and Law twenty-five years and in all of that time I think we've had three of [these types of youths]."2 This quote perfectly captures a prevalent myth: the juvenile justice system detains only very few LGB and gender non-conforming youths. Juvenile justice professionals believe this myth because only a handful of LGB and gender non-conforming youths disclose their sexual orientations, act in ways that do not conform to gender norms,3 or have court cases 4 linked to their sexual orientation, gender identity or gender expression. However, many LGB youths, youths who are questioning their sexual orientations or youths who have non-conforming gender identities enter the juvenile justice system unnoticed.5 The disproportionate incarceration of youths of color within the juvenile justice system further reinforces the invisibility of LGB and gender non-conforming youths.6 Many juvenile justice professionals assume that LGB and gender non-conforming youths come from middle class, white families, and therefore, juvenile justice jurisdictions detaining large numbers of youths of color do not serve LGB and gender non-conforming youths] This Article shows that both assumptions are wrong. National survey data presented in this Article shows that fifteen percent of youths in 2 Interview with anonymous source, Juvenile Det. Alternatives Initiatives Conference, in Indianapolis (Sept. 23, 2008). 3 Mary Curtin, Lesbian and Bisexual Girls in the Juvenile Justice System, 19 CHILD & ADOLESCENT SOC. WORK J. 285, 291 (2002); Meda Chesney-Lind & Michele Eliason, From Invisible to Incorrigible: The Demonization of Marginalized Women and Girls, 2 CRIME, MEDIA, CULTURE 29, 36 (2006), available at http://cmc.sagepub. com/cgi!reprint/2/1/29 (follow "Begin Manual Download"). 4 Sarah Valentine, Traditional Advocacy for Nontraditional Youth: Rethinking Best Interest for the Queer Child, 2008 MICH. ST. L. REv. 1053, 1087-1090 (2008). For the purposes of this paper, "sexual orientation" is defined as a person's emotional, romantic and sexual attraction to individuals of the same sex or of a different sex. 5 See Curtin, supra note 3, at 290. 6 For a discussion of the disproportionate incarceration of youths of color, see RICHARD MENDEL, ANNIE E. CASEY FOUND., Two DECADES OF JDAI: FROM DEMONSTRATION PROJECT TO NATIONAL STANDARD: A PROGRESS REPORT 6 (2009); see also James Bell, Juvenile Justice and Race: An Uphill Climb to the Bottom, THE HUFFINGTON POST, Nov. 18, 2009, available at http://wwwv.huffingtonpost.com/james-belU/iuvenile-iustice-andrace b 362283.html. 7 Participant notes of Laura Gamette, Div. Dir. of Adult Prob. in Santa Cruz County, documenting her observation of an institutional staff training in San Francisco County, San Francisco, Cal. (Oct. 10, 2009). 676 [Vol. 19:3 "We've Had Three of Them" the juvenile justice system are LGB, questioning their sexual orientation, transgender or express their gender in non-conforming ways. Moreover, the data shows that equal proportions of white, African American and Latino youths are LGB and gender non-conforming.8 Yet, myths around the nonexistence of LGB and gender non- conforming youths in the juvenile justice system persist, presenting numerous challenges to the equitable treatment of such youths. Juvenile justice professionals need to know that LGB or gender non-conforming youths exist within the system, and that LGB or gender non-conforming youths often enter the juvenile justice system for different reasons than straight youths. For example, the findings presented in this Article show that LGB or gender non-conforming youths are more likely than heterosexual youths to enter the juvenile justice system because they run away from home or placement or because of status offenses such as truancy. 9 Juvenile justice professionals need to know the underlying reasons for failure to remain at home, in placement or truancy in order to identify successful alternatives to detention and out-of-home placements or to assign appropriate terms of probation. Yet, gathering information about clients' sexual orientation, gender identity and gender expression create a difficult conundrum for juvenile justice professionals. While juvenile justice professionals need to know the reasons that youths have entered the juvenile justice system, many youths do not disclose their sexual orientations or gender identities. Research that uncovers high levels of homophobic reprisal from peers, parents, institutional staff or judges provides one possible explanation for such lack of disclosure: youths may fear further victimization if they disclose their sexual orientation or gender identity.' 0 The policy recommendations provided at the end of this Article can help create safe conditions that lead to disclosure about sexual orientation and gender identity, but there will always be youths who prefer to maintain their privacy.' 8 See infra Part 1.D & I.E. 'See infra Part I.B. 10 Anthony R. D'Augelli, Mental Health Problems Among Lesbian, Gay, and Bisexual Youth Ages 14 to 21, 7 CLINICAL CHILD PSYCHOL. AND PSYCHIATRY 433, 434-435 (2002) [hereinafter D'Augelli]; Anthony R. D'Augelli & Arnold H. Grossman, Disclosure of Sexual Orientation, Victimization, and Mental Health Among Lesbian, Gay, and Bisexual Adults, 16 J. INTERPERSONAL VIOLENCE, 1008, 1009-21 (2001). " See infra Part 111. 2010] Columbia Journal of Gender and Law Faced with the need to serve LGB and gender non-conforming youths while respecting their privacy, national survey data can help inform practice even when the sexual orientation or gender identity of individual youths are not known in individual instances. By analyzing 2,100 surveys that have been collected from six jurisdictions across the country, this Article: " provides the only existing estimates of the number of LGB and gender non-conforming youths who enter the juvenile justice system; " describes patterns of incarceration for LGB and gender non- conforming youths compared with their heterosexual and gender conforming peers to help juvenile justice professionals understand the social context around the detention of many LGB and gender non-conforming youths; and " offers suggestions to juvenile justice professionals on how to address the needs of LGB and gender non-conforming youths, whether visible or invisible within the juvenile justice system. I. ESTIMATING THE NUMBER OF LESBIAN, GAY, BISEXUAL AND GENDER NON-CONFORMING YOUTHS IN THE JUVENILE JUSTICE SYSTEM Collecting accurate data on the number of LGB and gender non- conforming youths in the juvenile justice system has been difficult to date. There are no federal or state agencies that require the collection of data on sexual orientation. As a result, we, at Ceres Policy Research, failed to identify any jurisdictions that collect formal data linking the sexual orientation and gender identity of an individual youth to his or her probation record.12 Our own previous research provides some clues regarding the number of LGB and gender non-conforming youths in the juvenile justice system. Questions about sexual orientation and gender identity in two small 12 National and state data collection requirements shape the collection of data on the local level. The Department of Justice does not require the reporting of sexual orientation or gender identity in their annual census. Nor is this data collected in any of the states with Juvenile Detention Alternative Initiative sites: CA, OR, WA, NV, ID, MT, AZ, NM, TX, LA, MS, AL, GA, FL, VA, DC, MD, DE, NJ, MA, NH, IN, IL, MO, IA, or MN. Colleagues in NY and UT report that jurisdictions in these states also do not collect data on sexual orientation or gender identity. [Vol. 19:3678 2ulu] "We've Had Three oJ Them" 679 program evaluations for the Santa Cruz and Sonoma County Probation Departments in California provide two limited snapshots. We added questions about sexual orientation and gender identity to one needs assessment of 230 youths in Santa Cruz County, CA. 13 We found that fourteen percent of youths disclosed being LGB or questioning their sexual orientation.' 4 No youths reported having gender identities other than male or female.' 5 In Sonoma County, CA we asked questions about sexual orientation in surveys and interviews for an evaluation of a gender specific program for girls.' 6 When we asked 176 participants about their sexual orientation on an anonymous survey, thirteen percent of girls disclosed having a lesbian or bisexual sexual orientation or being unsure.' 7 When we asked a smaller sample of 26 girls about their sexual orientation during qualitative interviews, seven percent of participants disclosed bisexual sexual orientations.' 8 Yet, the number of youths in each of these studies was small, making broad generalities about LGB and gender non-conforming youths in the juvenile justice system difficult. Even if jurisdictions decided to collect system data on sexual orientation or gender identity, understanding exactly how many LGB and gender non-conforming youths enter the juvenile justice system as well as their detention patterns presents a unique challenge; youths can hide their gender identity and sexual orientation, making the accurate collection of data that links gender identity and sexual orientation to juvenile justice outcomes difficult.' 9 At the same time, peers, family members, and juvenile 13 Angela Irvine, Laura Garnette, & Cynthia Chase, Ceres Policy Research, Healthy Returns Initiative: Santa Cruz County Update, PowerPoint slide #27 (2008) (on file with author). 14 Id. 15 Id. 16 JESSICA ROA & ANGELA IRVINE, CERES POLICY RESEARCH, TITLE I SONOMA: GIRLS CIRCLE ANNUAL 2007-08 REPORT 4 (May, 2008). 17 Id. 18 JESSICA ROA, ANGELA IRVINE & CHRISTINE KALINOWSKI, CERES POLICY RESEARCH, TITLE 11 SONOMA: CIRCLES ACROSS SONOMA QUALITATIVE RESULTS 3 (Apr. 2009). 19 For a discussion of how LGB and gender non-conforming youths may hide their sexual orientations or gender identities, see KATAYOON MAJD, JODY MARKSAMER & CAROLYN REYES, HIDDEN INJUSTICE: LESBIAN, GAY, BISEXUAL AND TRANSGENDER YOUTH IN JUVENILE COURTS 131 (2009), available at http://www.equityproject.org/pdfs/ hiddeninjustice.pdf; see also Stephen Tropiano, Playing It Straight: Reality Dating Shows OJ .............. 680 Columbia Journal of Gender and Law [Vol. 19:3 justice professionals often inaccurately make assumptions about youths' gender identities and sexual orientations based on social norms.20 Researchers must rely on youths themselves to disclose their sexual orientation and gender identity. Unfortunately, youths often resist disclosing their sexual orientations or gender identities. One explanation for this resistance is fear of reprisal from parents, teachers, or juvenile hall staff. Researchers have documented that LGBT youths experience rejection, bullying and harassment from parents, teachers or juvenile hall staff when they disclose their sexual orientations or gender identities.21 Given the difficulty of tracking LGB and gender non-conforming youths, all three of the previous studies of LGB and gender non-conforming youths in the juvenile justice system have relied on convenience samples. 22 These researchers contacted LGB-and-gender-non-conforming-youth- serving agencies and conducted focus groups or interviews with youths who have disclosed being LGB or having non-conforming gender identities.23 However, these findings fail to document the experiences of youths who choose to keep their sexual orientations or non-conforming gender identities hidden. Large anonymous surveys that target the general juvenile justice population can come closer to accurately measuring the number of LGB and gender non-conforming youths in the juvenile justice system, since youths may be more likely to disclose their sexual orientation or gender and the Construction of Heterosexuality, 37 J. OF POPULAR FILM AND TELEVISION 60, 69 (2009); Jay Clarkson, The Limitations of the Discourse of Norms: Gay Visibility and Degrees of Transgression, 32 J. OF COMM. INQUIRY 368, 369 (Oct. 2008). 20 For a discussion how families, the child welfare system, the law, the media and the workplace inaccurately assume people are heterosexual unless they disclose their sexual orientation, see Gill Valentine, Negotiating and Managing Multiple Sexual Identities: Lesbian Time-Space Strategies, 18 TRANSACTIONS OF THE INST. OF BRITISH GEOGRAPHERS 237, 238 (1993). 21 See D'Augelli, supra note 10. 22 "Convenience samples" are a method of selecting samples for research. In contrast to random samples, convenience samples select people that are readily available. Convenience samples are used when populations are difficult to identify or contact. There have only been three previous studies of LGBT youths in the juvenile justice system: see RANDI FEINSTEIN ET AL., THE LESBIAN & GAY YOUTH PROJECT OF THE URBAN JUSTICE CENTER, JUSTICE FOR ALL? A REPORT ON LESBIAN, GAY, BISEXUAL AND TRANSGENDERED YOUTH IN THE NEW YORK JUVENILE JUSTICE SYSTEM 57 (2001); MAJD ET. AL., supra note 19; Curtin, supra note 3, at 288. 23 See MAD ET. AL., supra note 19; see also Curtin, supra note 3, at 288. "We've Had Three of Them" identity if this information will be kept from parents, peers and juvenile justice staff members.24 By surveying youths within the juvenile justice system, researchers can also link sexual orientation and gender identity to detention patterns. Previously, juvenile justice systems did not collect or analyze data about the sexual orientations or gender identities of individual youths, making statistical disaggregation along these two variables impossible. By sampling the general population in the juvenile justice system, researchers ensure the inclusion of LGB and gender non- conforming youths who have not disclosed their sexual orientations or gender identities or who have not accessed LGB-and-gender-non- conforming-youth-serving agencies. This Article presents the findings of such a large national survey. Contrary to popular belief, survey results show that fifteen percent of youths being detained prior to adjudication are LGB, questioning their sexual orientation, transgender, have non-conforming gender identities or express their gender in non-conforming ways.25 This is a significant number of LGB and gender non-conforming youths within the juvenile justice system each year. A. Survey Methods This Article presents findings from a survey of 2,100 youths in six juvenile justice jurisdictions across the country. The research sites were selected among jurisdictions receiving funds from the Annie E. Casey Foundation as part of their Juvenile Detention Alternatives Initiative.26 Survey sites were selected to represent a range of jurisdictional size, youth demographics and geographic location. The final sites were Minneapolis, 24 See D'Augelli, supra note 10. 25 See infra tbl. 1. 26 This initiative began in 1997 to create a systemic method for reducing the number of youths detained in the juvenile justice system. Participating sites first adopt structured decision-making programs for detention decisions. Subsequent reforms generally include the adoption of programs to reduce disproportionate minority contact and the unnecessary detention of girls. See DAVID STEINHART, ANNIE E. CASEY FOUND., PATHWAYS TO JUVENILE DETENTION REFORM: PLANNING FOR JUVENILE DETENTION REFORMS-A STRUCTURED APPROACH 41 (1999); FRANCINE T. SHERMAN, ANNIE E. CASEY FOUND., PATHWAYS TO JUVENILE DETENTION REFORM-DETENTION REFORM AND GIRLS: CHALLENGES AND SOLUTIONS 41 (2005). 2010] Columbia Journal of Gender and Law MN, Albuquerque, NM, Las Vegas, NV, Santa Cruz, CA, Birmingham, AL and Portland, OR. Institutional or health staff members surveyed the entire pre-trial detention population at each site. The survey included questions about demographics, histories of bullying and harassment, histories with the child welfare system, histories of suspension and expulsion from school and detention histories.27 We piloted the survey with youths who attend an alternative education high school in Santa Cruz County, CA, as well as youths who have been involved in the Santa Cruz County juvenile justice system. This process ensured that we had created an instrument with an appropriate reading level and culturally relevant vocabulary. We also developed a data collection protocol that ensured confidentiality. Institutional or health staff members offered every youth a survey, a cover sheet explaining the purpose of the survey and an envelope. Youths could fill out the survey or leave the survey blank, place the survey sheet into the envelope, seal the envelope, and then place their envelope in a locked box. When the locked box was full, a research liaison at each site then placed the sealed surveys into a manila envelope and sent the completed surveys to the researchers. Finally, we entered the data sent from each of the research sites into a spreadsheet and then analyzed it using Statistical Package for the Social Sciences statistics software (SPSS). We calculated the percentage of respondents falling into divisions based on the following categories: race/ethnicity, gender, age, sexual orientation and county of origin. We completed additional statistics tests in order to determine whether particular subsets of respondents were more likely to have histories of home removal, group or foster home placement or homelessness. Finally, we analyzed detention histories to determine whether LGB and gender non-conforming youths are more likely to be detained for violent offenses, weapon charges, property offenses, drug and alcohol offenses, running away, having sex with someone of the same gender, prostitution, warrants or probation violations. B. Respondents 'Ages, Races and Ethnic Identities Survey respondents varied across age. The age of respondents ranged from eleven to twenty-one-years-old, with an average age of 15.7 27 See infra app. A (containing the full instrument). [Vol. 19:3682 "We've Had Three of Them" years old. Respondents also varied across race and ethnic identity. Within our sample, thirty-four percent (n=665) of respondents identified as African American, thirty percent (n=575) of respondents identified as Hispanic, Latino, Chicano, Mexican or Mexican American, eighteen percent (n=345) of respondents identified as white or Caucasian, four percent (n=76) of respondents identified as Native American, one percent (n=27) of respondents identified as Japanese, Chinese, Samoan or Pacific Islander, and thirteen percent (n=245) of respondents identified with multiple racial or ethnic categories or a racial or ethnic category other than those reported above. C Gender Identity and Expression We asked respondents a series of questions about gender identity and expression. As discussed above, researchers face a number of challenges around the accurate collection of data on gender identity.28 Some youths resist disclosing non-conforming gender identities. Other youths may not identify with distinct gender categories. Our survey question on gender identity provided the option of circling "boy," "girl," or filling in a blank field labeled "other." Eighty percent (n=1557) of respondents identified as boys, twenty percent (n=397) of respondents identified as girls, and three-tenths of one percent (n=5) of respondents identified as something other than boy or girl.29 We also wanted to identify youths who express their gender identities in a way that is non-conforming. In order to measure gender expression that falls outside societal norms, we asked whether the youths had ever been bullied or harassed for not being feminine or masculine enough. Ninety-four percent (n=17 9 6) of respondents answered no, while six percent (n=1 17) of respondents said yes. Four of the five respondents who did not identify as girl or boy answered yes. Girls were more likely to answer yes than boys: eight percent (n=32) of girls reported that they had been bullied or harassed for not being feminine enough compared with five percent (n=79) of boys who reported that they had been bullied or harassed 28 See supra Part 1. 29 These five youths did not specifically identify as transgender. Instead, two youths wrote "nothing," one youth circled both "boy" and "girl," one youth wrote "sex change" and one youth wrote "sometimes I wonder." 2010] Columbia Journal of Gender and Law for not being masculine enough. There were too few respondents to conduct an analysis for differences among race and ethnic identities. D. Sexual Orientation We asked a series of questions to measure respondents' sexual orientations. As with gender identity, researchers face a number of challenges around the accurate collection of data on sexual orientation.30 Some youths resist disclosing their sexual orientation if they are not heterosexuals. Other youths may not identify with distinct sexual orientation categories. Our first question provided the option of reporting traditional categories of sexual orientation: straight, lesbian/gay, bisexual, questioning and other. A second question asked respondents whether they are sexually attracted to boys/men, girls/women or other. A third question asked respondents, "Have you ever been bullied or harassed at school because of your sexual orientation (being lesbian, gay, etc.)?" We also asked, "Have you ever been kicked out of your home or ran away because of your gender identity (being transgender) or sexual orientation (being lesbian, gay, etc.)?" We added these questions so that youths who were uncomfortable disclosing their sexual orientations or who do not identify with distinct sexual orientation categories could identify themselves as existing outside societal norms around sexual orientation. When you compile the results from respondents who answered "'yes" or "no" to questions regarding sexual orientation, eighty-nine percent of respondents can be categorized as heterosexual while eleven percent of respondents are LGB, sexually attracted to the same gender, have been bullied or harassed by peers for their perceived sexual orientations or have been kicked out of their home or ran away due to their sexual orientations. These eleven percent of participants will from this point forward be referred to as LGB. Responses on sexual orientation vary by gender. While eight percent of boys are gay, bisexual or questioning, twenty-four percent of girls and eighty percent of youths who do not identify as either boy or girl are LGB. These responses also vary by race and ethnic identity. Ten percent of white, Latino and African American respondents are LGB. This means that an equal proportion of white, Latino, and African American youths 30 See supra Part I. [Vol. 19:3 "We've Had Three of Them" disclosed LGB sexual orientations. Youths outside of these racial identities have higher disclosure rates: twelve percent of Asian, twenty-four percent of Native American and eighteen percent of respondents with multiple race and ethnic identities are LGB. E. Combining Gender Identity, Gender Expression and Sexual Orientation Data Combining the data on gender identity, gender expression and sexual orientation, we can estimate the number of LGB and gender non- conforming youths who enter the juvenile justice system. We arrive at this data by completing two steps. First, we combine gender identity and expression into one category: gender conformity. For this Article, youths who have non-conforming identities or express themselves in non- conforming ways are considered "gender non-conforming." We then separate gender conformity from sexual orientation. We do this because of the complex ways that gender identity and expression interact with sexual orientation. For example, a girl could identify as a lesbian and conform to gender norms while a boy could identify as heterosexual but express himself in a feminine way. In order to present the data in an easily digestible way, we create four distinct categories: heterosexual and gender conforming; heterosexual and gender non-conforming; LGB and gender conforming; and LGB and gender non-conforming. Table 1, infra, reports the number and percentage of respondents who fall into each of these four categories: TABLE 1: Number and Percentage of Respondents by Sexual Orientation and Gender Conformity 2010] 685 686 Columbia Journal of Gender and Law [Vol. 19:3 Heterosexual LGB31 Gender 85% 9% Conforming n=1638 n=167 Gender Non- 3% 3% conforming32 n=66 n=53 These results provide a superficial snapshot of the four combinations that sexual orientation and gender identity can take. Nevertheless, this data provides an estimate of the number of LGB and gender non-conforming youths in the system. In this case, the number equates to those youths who are not heterosexual and gender conforming. Adding the number of LGB youths who are gender conforming, the number of LGB youths who are gender non-conforming, and the number of heterosexual youths who are gender non-conforming, we find that fifteen percent of youths can be categorized as LGB or gender non-conforming. This data also helps us understand the ways that LGB youths might enter and exit the juvenile justice system without being noticed. Gender non-conforming youths are the most likely to be noticed. Only three percent of youths who are LGB are also gender non-conforming. The same percentage of youths (three percent) has heterosexual sexual orientations and do not conform to gender norms. However, a much larger percentage of youths (nine percent) are LGB and gender conforming. This is the group that is most likely to enter the juvenile justice system and remain invisible. Lesbian, bisexual, questioning and gender non-conforming girls remain more invisible than gay, bisexual, questioning and gender non- conforming boys. Tables 2 and 3, infra, separate the data for respondents 31 Respondents were categorized as LGB if they disclosed having a lesbian, gay, bisexual or other sexual orientation, questioning their sexual orientation, having same-gender sexual attraction, having a history of running away or being kicked out of their home because of their gender identity or sexual orientation or having been bullied or harassed at school because of their sexual orientation. 32 Gender non-conforming respondents responded "yes" to the question, "Have you ever been bullied or harassed at school because people don't think you are masculine enough or feminine enough?" or reported that they have neither a boy nor girl gender identity. "We've Had Three of Them" with girl and boy gender identities. Comparing these two groups, we see that a high number of girls fall into the three categories used to estimate the numbers of LGB and gender non-conforming youths 33: twenty-seven percent of girls compared with eleven percent of boys can be categorized as lesbian, bisexual and questioning and gender non-conforming. In addition, a higher proportion of these girls fall into the category of youths that are most likely to remain invisible; nineteen percent of girls are lesbian, bisexual and questioning girls and gender conforming compared to six percent of boys. Notably, even though the percentages of gay, bisexual, questioning and gender non-conforming boys are lower than girls, the numbers are higher. This happens because the juvenile justice system detains many more boys than girls.34 Combining the data on girls and boys in all of the categories we use to measure the number of LGB and gender non- conforming youths, our six research sites detained 286 LGB and gender non-conforming respondents over the course of two months. " See infra Part I.E. 34 In 2001, girls represented 19% of detained youths. See SHERMAN, supra note 26, at 10. 2010] 687 Columbia Journal of Gender and Law TABLE 2: Number and Percentage of Girl Respondents by Sexual Orientation and Gender Conformity Heterosexual LGB35 Gender 73% 19% Conforming n=285 n=76 Gender Non- 4% 4% conforming 36 n= 15 n=17 TABLE 3: Number and Percentage of Boy Respondents by Sexual Orientation and Gender Conformity Heterosexual LGB3 7 Gender 89% 6% Conforming n=1353 n=91 Gender Non- 3% 2% conforming 38 n=50 n=29 35 See supra note 31. 36 See supra note 32. 37 See supra note 31. 38 See supra note 32. [Vol. 19:3688 "We've Had Three of Them" II. DETENTION PATTERNS OF LGB AND GENDER NON- CONFORMING YOUTHS In addition to estimating the numbers of LGB and gender non- conforming youths in the juvenile justice system-an estimate of fifteen percent-findings from this study allow us to determine whether the juvenile justice system detains LGB and gender non-conforming youths for different reasons when compared with heterosexual and gender conforming youths. Findings from the survey data document that LGB and gender non- conforming youths in the juvenile justice system are twice as likely to have experienced child abuse, group and foster home placement and homelessness when compared with their heterosexual and gender conforming peers. Findings also show that LGB and gender non- conforming youths are more likely to be held in pre-trial detention for truancy, warrants, probation violations, running away and prostitution. This suggests that LGB and gender non-conforming youths are caught in a pernicious cycle of abuse, family rejection and detention. The findings from our study reinforce existing research on LGB and gender non-conforming youths. Previous research on LGB and gender non-conforming youths uncovered high rates of child abuse, homelessness and conflict with parents. 39 For example, one study found that over thirty percent of gay men and lesbians reported suffering physical violence at the hands of a family member as a result of their sexual orientation.4 ° When this occurs, these youths may be removed by child protective services. Another study found that twenty-six percent of gay adolescents were forced to leave home after disclosing their sexual orientation.41 In both cases, conflict with family members leaves LGB and gender non-conforming youths more dependent on other social institutions such as group homes, foster homes or 39See COLLEEN SULLIVAN, SUSAN SOMMER & JASON MOFF, LAMBDA LEGAL DEF. & EDUC. FUND, YOUTH IN THE MARGINS: A REPORT ON THE UNMET NEEDS OF LESBIAN, GAY, BISEXUAL, AND TRANSGENDER ADOLESCENTS IN FOSTER CARE 11 (2001); NICHOLAS RAY, NAT'L GAY & LESBIAN TASK FORCE POL'Y INST. & THE NAT'L COALITION FOR THE HOMELESS, LESBIAN, GAY, BISEXUAL AND TRANSGENDER YOUTH: AN EPIDEMIC OF HOMELESSNESS 17-21 (2006); Caitlin Ryan, David Buebner, Rafael M. Diaz & Jorge Sanchez, Family Rejection as a Predictor of Negative Health Outcomes in White and Latino Lesbian, Gay, and Bisexual Youth Adults, 123 PEDIATRICS 346, 350 (2009). 40 SULLIVAN, SOMMER & MOFF, supra note 39, at 11. 41 See id. 2010] 689 Columbia Journal of Gender and Law homeless shelters.42 However, only three previous studies have directly linked these variables to the involvement of LGB and gender non- conforming youths in the juvenile justice system.43 The methodology used in this study also strengthens the existing literature. Given the difficulty in identifying LGB and gender non- conforming youth in the juvenile justice system, the three existing studies on this topic relied on convenience samples.44 In other words, researchers from these previous studies contacted LGB-and-gender-non-conforming- youth-serving agencies and interviewed youths who had previously disclosed their sexual orientations or gender identities.45 However, this represents only a small minority of LGB and gender non-conforming youths. Moreover, these studies do not document the experience of heterosexual and gender conforming youths, and therefore, fail to provide a comparison between LGB and gender non-conforming youths and their peers.46 Surveys that include heterosexual and gender conforming youths provide helpful baseline data to judge the severity of LGB and gender non- conforming youth outcomes.47 We base the findings of this Article on a larger sample of youths that includes heterosexual, gender conforming, LGB and gender non- conforming youths. This sample includes LGB and gender non-conforming youths who have not disclosed their sexual orientations or gender identities. By surveying the general juvenile justice population, we can develop more accurate links between home removal, group and foster home placement, homelessness and juvenile justice involvement for LGB and gender non- conforming youths. We can compare the experiences of heterosexual and 42 See RAY, supra note 39, at 17-21. 41 See FEINSTEIN ET. AL., supra note 22, at 11; see also MAJD ET. AL., supra note 19, at 93-144; Curtin, supra note 3, at 287-288. 44 See FEINSTEIN ET AL., supra note 22, at 57; see also MAJD ET. AL., supra note 19, at 17; Curtin, supra note 3, at 288. 45 See FEINSTEIN ET AL., supra note 22, at 57; see also MAJD ET. AL., supra note 19, at 17; Curtin, supra note 3, at 288. 46 See FEINSTEIN ET AL., supra note 22, at 57; see also MAID ET. AL., supra note 19, at 17; Curtin, supra note 3, at 288. 47 For a longer discussion of the importance of control groups and the history of experimental and quasi-experimental research designs, see William R. Shadish & Thomas D. Cook, The Renaissance of Field Experimentation in Evaluating Interventions, 60 ANN. REV. PSYCHOL. 607, 608 (2009). [Vol. 19:3690 "We've Had Three of Them" gender conforming youths to LGB and gender non-conforming youths in these areas. This particular survey also collects information about the specific reasons youths are detained, allowing us to compare LGB, gender non-conforming, heterosexual and gender conforming youths. A. Measuring Home Removal, Group and Foster Home Placement and Homelessness Our survey asked respondents three questions related to child abuse, home removal and homelessness. In order to determine whether youths had ever been removed from their homes for child abuse, we asked, "Have you ever been removed from your home by a social worker?" In order to measure whether youths had ever been placed in a group or foster home, we asked, "Have you ever lived in a group or foster home?" In order to measure whether youths had ever been homeless, we asked, "Have you ever been homeless after being kicked out of home or running away?''48 In order to understand whether there are differences across sexual orientation, we used statistical analysis to determine whether LGB and gender non-conforming youths have different outcomes when compared with their heterosexual and gender conforming peers. In order to understand whether there are differences across gender, we used statistical analysis to compare gender non-conforming boys and girls to their gender conforming peers. .49 Results show that LGB and gender non-conforming youths are more than twice as likely as their heterosexual and gender normative peers to answer "yes" to these questions. L Home Removal and Sexual Orientation Twenty percent of gay, bisexual and questioning boys have been removed from their home because someone was hurting them compared with nine percent of heterosexual boys. Thirty-three percent of lesbian, bisexual and questioning girls have been removed from their home because 48 See infra app. A. 49 Comparisons across race and ethnic identity were not possible due to a large number of race and ethnic identity categories and a small number of LGB and gender non- conforming respondents. 2010] Columbia Journal of Gender and Law someone was hurting them compared with nineteen percent of heterosexual girls. 2. Home Removal and Gender Conformity Twenty-three percent of gender non-conforming boys have been removed from their home because someone was hurting them compared with nine percent of gender conforming boys. Forty-two percent of gender non-conforming girls have been removed from their home because someone was hurting them compared with twenty percent of gender conforming girls. 3. Group Home and Foster Home Placements by Sexual Orientation Twenty-five percent of gay, bisexual and questioning boys have lived in a group or foster home compared with seventeen percent of heterosexual boys. Forty-five percent of lesbian, bisexual and questioning girls have lived in a group or foster home compared with twenty-seven percent of heterosexual girls. 4. Group Home and Foster Home Placements by Gender Conformity Thirty-four percent of gender non-conforming boys have lived in a group or foster home compared with sixteen percent of gender conforming boys. Forty-seven percent of gender non-conforming girls have lived in a group or foster home compared with thirty percent of gender conforming girls. 5. Homelessness by Sexual Orientation Thirty-two percent of gay, bisexual and questioning boys have been homeless after being kicked out of or running away from home compared with sixteen percent of heterosexual boys. Forty percent of lesbian, bisexual and questioning girls have been homeless after being kicked out of or [Vol. 19:3 "We've Had Three of Them" running away from home compared with twenty-three percent of heterosexual girls. 6. Homelessness by Gender Conformity Forty-three percent of gender non-conforming boys have been homeless after being kicked out of or running away from home compared with sixteen percent of gender conforming boys. Forty-one percent of gender non-conforming girls have been homeless after being kicked out of or running away from home compared with twenty-six percent of gender conforming girls. B LGB and Gender Non-Conforming Youths and Juvenile Detention The survey asked respondents whether they had been detained within the past twelve months for crimes against persons, property crimes, drug and alcohol offenses, running away, missing school, prostitution or any other reason. In order to understand whether there are differences across sexual orientation, we used statistical analysis to determine whether LGB youths have different outcomes when compared with their heterosexual and peers. In order to understand whether there are differences across gender, we used statistical analysis to compare gender non- conforming boys and girls to their gender conforming peers.. Findings from the survey show that LGB and gender non- conforming youths are twice as likely to be held in secure detention for truancy, warrants, probation violations, running away and prostitution. Notably, there were no differences in the prevalence of detention for violent offenses, weapon charges, property offenses or alcohol or drug offenses. Instead, LGB and gender non-conforming youths are more likely to be detained for non-violent offenses with direct links to out-of-home placement and homelessness. 50 Comparisons across race and ethnic identity were not possible due to a large number of race and ethnic identity categories and a small number of LGBT and gender non- conforming respondents. 2010] Columbia Journal of Gender and Law 1. Running A way by Sexual Orientation Twenty-eight percent of gay, bisexual and questioning boys were detained for running away compared with twelve percent of their heterosexual peers. Thirty-seven percent of lesbian, bisexual and questioning girls were detained for running away compared with eighteen percent of their heterosexual peers. 2. Running A way by Gender Conformity Twenty-two percent of gender non-conforming boys were detained for running away compared with twelve percent of their gender conforming peers. Thirty-three percent of gender non-conforming girls were detained for running away compared with twenty-one percent of their gender conforming peers. 3. Prostitution by Sexual Orientation Ten percent of gay, bisexual and questioning boys were detained for prostitution compared with one percent of their heterosexual peers. Eleven percent of lesbian, bisexual and questioning girls were detained for prostitution compared with five percent of their peers. 4. Prostitution by Gender Conformity Seven percent of gender non-conforming boys were detained for prostitution compared with one percent of their gender conforming peers. Six percent of gender non-conforming girls were detained for prostitution compared with seven percent of their gender conforming peers. 5. Truancy, Warrants and Violations of Probation by Sexual Orientation [Vol. 19:3694 "We've Had Three of Them" Twenty-two percent of gay, bisexual and questioning boys were detained for truancy, warrants or violations of probation compared with twelve percent of their heterosexual peers. Fifteen percent of lesbian, bisexual and questioning girls were detained for truancy, warrants or violations of probation compared with nine percent of their peers. 6. Truancy, Warrants and Violations of Probation by Gender Conformity There were no significant differences in detention for truancy, warrants or violations of probation by gender conformity. Notably, the numbers of gender non-conforming youths who had been detained for these reasons were very small. A larger sample size may yield significant results in this area in the future. III. ADDRESSING THE NEEDS OF INCARCERATED LGB AND GENDER NON-CONFORMING YOUTHS Juvenile justice professionals such as defense attorneys, prosecutors, judges and probation officers need accurate information about youths in the juvenile justice system in order to provide high quality defense, recommend appropriate alternatives to detention and treatment services, and assign appropriate terms of probation. Accurate data, however, is difficult to gather from LGB and gender non-conforming youths. Youths who do not want to disclose their sexual orientations or gender identities will not provide accurate histories related to family conflict, truancy and running away out of fear of reprisal. Faced with this central dilemma, juvenile justice professionals can follow several best practices to reduce the use of incarceration for LGB and gender non-conforming youths caught in a cycle of family conflict and juvenile detention. NEVER ASSUME. The Office of Juvenile Justice and Delinquency Prevention reports that there were 92,854 youth juvenile offenders detained in the United States in 2006. 51 5' MELISSA SICKMUND, T.J. SLADKY & WEI KANG, OFFICE OF Juv. JUST. & DELINQ. PREVENTION, CENSUS OF JUVENILES IN RESIDENTIAL PLACEMENT DATEBOOK (2008), 2010] 695 Columbia Journal of Gender and Law According to this Article, fifteen percent of this population, or 13, 928 youths, is LGB or gender non-conforming. Again applying the findings of this Article, nine percent of this population, or 8,356 youths, is invisible to the juvenile justice system. Given these large numbers, juvenile justice professionals should never assume that they know the sexual orientation or gender identity of an individual youth. Instead, they should ask neutral questions such as, "Do you have a boyfriend or girlfriend?" They should find out if the youth is afraid for their safety while detained and why. They should ask about school attendance, and if the youth is not attending school, ask why. If a youth does not get along with family members, they should ask why. If a youth raises issues related to sexual orientation or gender identity, juvenile justice professionals should remain open and supportive, and explain that they will not reveal this information to anyoneS•52 else unless the youth gives them permission. DEVELOP POLICIES TO ENSURE THE EQUITABLE TREATMENT AND SAFETY OF LGB AND GENDER NON-CONFORMING YOUTHS: Juvenile justice professionals should insist on respect for LGB and gender non-conforming youths in the courtroom, law offices, probation departments and juvenile detention facilities. Juvenile justice professionals should also consistently respond to instances of discrimination based on sexual orientation, gender identity or gender expression.53 In order to do this, they should adopt policies that enumerate ways to ensure the equitable treatment of LGB and gender non- conforming youths in the care of group homes, foster homes and secure detention. 54 Model anti-discrimination policies have been developed and implemented in New York and Hawaii. 55 Similar policies should be developed elsewhere. http://www.ojjdp.ncjrs.gov/ojstatbb/cjrp/ (follow "US & State Profiles," then select "Age on census date by sex") (last visited Feb. 5, 2010). 52 See MAJD ET AL., supra note 19, at 124. 53 Id. at 139. 54 For a model policy, see id. at 149-55, 159-65. 55 See N.Y. STATE OFF. OF CHILD. AND FAM. SERVICES, POLICY ON LESBIAN, GAY, BISEXUAL, TRANSGENDER, AND QUESTIONING YOUTH (PPM 3442.00) (Mar. 17, 2008), available at http://srlp.org/files/LGBTQYouthPolicy_ PPM_3442_00.pdf. 696 [Vol. 19:3 2010] "We've Had Three of Them" 697 PROVIDE TRAINING: Juvenile justice professionals should be educated about the reasons that LGB and gender non- conforming youths are detained, how to provide services in a culturally competent way, the role that alternatives to detention can play in reducing youths' returns to detention and the benefits of resolving cases outside the formal court process.5 6 Juvenile justice professionals should also require that all subcontracting direct service providers be trained in the same content areas. DEVELOP FAMILY CRISIS PROTOCOLS: Juvenile justice professionals should develop family crisis protocols for pre-trial detention. Family rejection is at the heart of many offenses committed by LGB and gender non- conforming youths. Yet, counseling and support for families can help mediate the negative effects of rejection and increase the well-being of LGB and gender non-conforming youths.57 Family crisis protocols can be developed with community-based organizations that specialize in parent-teen mediation. These family crisis protocols should be developed for all youths. Such organizations help parents and teens communicate more effectively and help create safety plans for youths. 58 These organizations, however, should be trained on the needs of LGB and gender non-conforming youths. Such training should cover how to respectfully ask all youths about the source of conflict in their families and how to communicate specifically about conflict around sexual orientation or gender identity if these issues arise. " COLLABORATE: In the search for solutions, juvenile justice professionals should create, lead or join community- based collaborations and task forces to reduce the detention 56 See FRANCINE SHERMAN & ANGELA IRVINE, ANNIE E. CASEY FOUND., USING JDAI STRATEGIES TO REDUCE THE DETENTION OF GIRLS: A PRACTICE GUIDE TO JUVENILE DETENTION REFORM 46 (2010); MAJD ET. AL., supra note 19, at 137. 57 Ryan et. al., supra note 39, at 351. 58 Examples of family crisis protocols have been developed specifically for girls in the juvenile justice system with similar patterns of family conflict and runaway behavior. See SHERMAN & IRVINE, supra note 56, at 45. The Conflict Resolution Center in Santa Cruz County, C.A. provides an example of a community-based organization that specializes in teen-parent mediation. See Conflict Resolution Center, Restorative Justice: Parent Teen Mediation, http://www.crcsantacruz.org/restorative-justice.html#PTM (last visited Feb. 5, 2010). 698 Columbia Journal of Gender and Law [Vol. 19:3 of LGB and gender non-conforming youths. Collaborations that include juvenile justice stakeholders, mental and physical health providers, alcohol and drug abuse services, and community-based organizations can share a broader range of expertise about the needs of LGB and gender non- conforming youths. 59 Communities can also strengthen continua of services for LGB and gender non-conforming youths when organizations are regularly communicating with one another. 60 ALTERNATIVES TO DETENTION: Alternatives to detention should be developed for all youths who are not a risk to the community or at risk of running away. Alternatives to detention, such as evening reporting centers, can keep young people tied to their communities, build positive relationships with adults and peers, and keep youths monitored at the lowest necessary level of supervision . Some of these alternatives can be designed to serve LGB and gender non-conforming youths specifically. Coordinators of these alternative initiatives should train staff on the needs of LGB and gender non-conforming youths, since most LGB and gender non-conforming youths will not disclose their sexual orientation or gender identity. 61 " REVIEW DATA IN AN ONGOING WAY: Decisions on system reforms should be based on a systemic review of data.6 2 Data on LGB and gender non-conforming youths, however, can be difficult to collect. Juvenile justice professionals should use caution when collecting information about sexual orientation or gender identity, avoiding data that can be linked to individual youths. Anonymous surveys and needs assessments can provide jurisdictions with concrete information about their own cities that can guide the creation of alternatives to detention as well as prevention programs. 59Examples of multi-agency collaborations include Girls Task Forces and the Santa Cruz County Queer Youth Task Force. See SHERMAN & IRVINE, supra note 56, at 41; The Diversity Center, Queer Youth Taskforce, http://diversitycenter.org/programs/ affiliates/Igbtiq-youth-task-force/ (last visited Feb. 5, 2010). 60 See MAJD ET AL., supra note 19, at 138. 61 See SHERMAN & IRVINE, supra note 56, at 42 62 See SHERMAN & IRVINE, supra note 56, at 8; STEINHART, supra note 26, at 13. "We've Had Three of Them" APPENDIX A: DATA COLLECTION INSTRUMENT Juvenile Detention Alternatives Survey: Information Sheet Your county is participating in a research project and would like your help. Why is this study being done? We want to understand what happens to lesbian, gay, bisexual, questioning and/or transgender youth in the juvenile justice system. Why did you choose me? We are surveying every youth who enters juvenile detention facilities. What will my involvement be? We would like you to complete this survey. After you are finished, fold the survey and put it in the locked survey box. What will happen to the survey and interview data? A researcher named Angela Irvine, who doesn't work for your Probation Department, will complete a report that will be shared with juvenile detention facilities across the country as well as the Annie E. Casey Foundation, who is funding the study. Your answers will help improve services for all youths. Who will know what answers I give? We are not asking for your name. Only the researcher will see the answers to this survey. She will not know who filled out which survey. Please do not put your name on the survey. What if I choose not to take part? You don't have to finish this survey if you don't want to. If you refuse, just put the blank survey in the secured box. If you decide to complete this survey, you may stop at any time without giving a reason. Your participation will not affect your treatment during detention. 2010] 699 Columbia Journal of Gender and Law Al. How old are you? years A2. What county do you live County in? A3. What is your gender (circle male female other all that apply)? (explain) A4. What is your sexual straight lesbian/gay bisexual orientation (circle all that questioning other (explain) apply)? A5. Who are you attracted to boys/men girls/women other (circle all that apply)? (explain) A6. What is your race/ethnic Latino/ African Native White all that apply)? Asian other (explain)__ identity (circle Hispanic American American B1. How many times have you been 0 1 2 3 4 5 6 7 8+ suspended from school? B2. How many times have you been 0 1 2 3 4 5 6 7 8+ expelled from school? B3. Have you ever been removed from yes no not sure your home by a social worker? B4. Have you ever been kicked out of your yes no not sure home or run away because of your gender identity (being transgender) or sexual orientation (being lesbian, gay, etc.)? B5. Have you ever lived in a group home yes no not sure or foster home? B6. Have you ever been bullied or yes no not sure harassed at school because of your sexual orientation (being lesbian, gay, etc.)? [Vol. 19:3700 "We've Had Three of Them" B7. Have you ever been bullied or yes no not sure harassed at school because people don't think you are masculine enough or feminine enough? B8. Have you ever been homeless after yes no not sure being kicked out of home or running away? C1. In the past 12 months, have you been held in juvenile detention for: a violent offense (hurting yes no not sure someone)? a weapon charge (possession of a yes no not sure weapon)? a property offense (stealing yes no not sure something)? an alcohol or drug offense yes no not sure (possession, dirty test)? running away from where you yes no not sure live? having sex with someone of the yes no not sure same gender? prostitution? yes no not sure skipping school? yes no not sure other (please explain): C2. If you have been held in juvenile detention, how many times have you been held in the last year? 2010]