Gender and Sexual Identity in Adolescence: A Mixed-Methods Study (2021) PDF

1000315 research-article2021 JARXXX10.1177/07435584211000315Journal of Adolescent ResearchHammack et al. Article Journal of Adole...

1000315 research-article2021 JARXXX10.1177/07435584211000315Journal of Adolescent ResearchHammack et al. Article Journal of Adolescent Research Gender and Sexual 2022, Vol. 37(2) 167­–220 © The Author(s) 2021 Article reuse guidelines: Identity in Adolescence: sagepub.com/journals-permissions DOI: 10.1177/07435584211000315 https://doi.org/10.1177/07435584211000315 A Mixed-Methods Study journals.sagepub.com/home/jar of Labeling in Diverse Community Settings Phillip L. Hammack1 , Sam D. Hughes1, Julianne M. Atwood1, Elliot M. Cohen1, and Richard C. Clark2 Abstract Understandings of sexual and gender identity have expanded beyond traditional binaries, yet we know little about adolescents’ appropriation of identity labels across diverse communities. In a mixed-methods study of adolescents recruited from lesbian, gay, bisexual, transgender, and queer (LGBTQ+) spaces in communities differing in support of sexual and gender diversity, seven patterns emerged: (a) frequent use of nonbinary gender identity labels (23.9% of survey sample), especially in high-support communities; (b) greater comfort among adolescents assigned female at birth (AFAB) with diverse gender expression, which informants attributed to pressures to conform to compulsive masculinity for boys; (c) frequent use of plurisexual (60.8%) and asexual (9.9%) labels, especially among those AFAB, and discussion of online settings as a resource; (d) intersectional patterning of “queer” to describe sexual identity (12.4% of survey sample), with White youth in high-support communities signifying an intellectual/ political stance and non-White youth in low-support communities using 1 University of California, Santa Cruz, USA 2 The City University of New York, New York City, USA Corresponding Author: Phillip L. Hammack, Department of Psychology, University of California, Santa Cruz, 1156 High Street, Santa Cruz, CA 95064, USA. Email: [email protected] 168 Journal of Adolescent Research 37(2) queer as an umbrella term; (e) resistance to labeling and ambivalence about labels due to intra-community dynamics; (f) labeling challenges among boys of color; and (g) challenges with stigma, sexualization, and violence for transgender and nonbinary youth. Findings highlight how contemporary adolescents engage with and challenge received conceptions of gender and sexuality and how this process is shaped by intersectional identities. Keywords gender identity, sexual identity, adolescence, masculinity, intersectionality, life course Danny1 looked in the mirror with a puzzled glance, not fully recognizing his reflection. “I noticed when I would look to the side, I would see my breasts, and I would just press them down,” he said. Danny had been assigned female at birth (AFAB), and as puberty commenced, the discon- nect between his body and his sense of gender identity created a chal- lenge—and an opportunity. In high school, Danny began to use the pronoun “he” instead of what had previously been imposed—“she.” Yet, Danny’s story is not as simple as it might first appear. Danny was 17 years old at the time of our interview in late 2016 and iden- tified as nonbinary. Danny reported using male pronouns but feeling neither completely male nor female. Some days he felt like a boy, some days not. He reported attraction to boys and identified as “gay” at the time of our inter- view, but he acknowledged that existing labels did not fully capture his expe- rience. He narrated the experience of turning to online resources like Twitter to “educate” himself and to find support. In sophomore year of high school, Danny cut his hair very short. “That’s when it felt right,” he said. We begin with the story of Danny to foreground the lived experience of adolescents in the midst of labeling their gender and sexuality at a time of change in cultural understandings and opportunities for self-understanding in the United States. Danny’s story centers our life course approach to the study of adolescent identity development by highlighting his development in his- torical time and place and situating the adolescents in our study as part of a particular generation-cohort (e.g., Cohler & Hammack, 2007). Although non- binary experiences of gender and sexuality are not new and have been docu- mented in earlier generations and other cultural contexts (e.g., Bornstein, 1995; Nestle et al., 2002; Vincent & Manzano, 2017), there is increasing visibility for an expanded set of gender and sexual identity labels beyond conventional binaries of “man-woman” and “gay-straight” (e.g., Barker & Hammack et al. 169 Iantaffi, 2019; Mardell, 2016). At the same time, our scientific understanding of the expansion of gender and sexual identity in adolescence is nascent (e.g., Risman, 2018; Russell, 2016; Watson et al., 2019). We know little about the pervasiveness of identity labels among contemporary adolescents, and we have limited knowledge of the ways in which the appropriation of particular labels may diverge across both individual factors (e.g., sex assigned at birth) and contextual factors (e.g., community settings). Knowledge about identity labeling among contemporary adolescents is vital as scientific understandings of gender, sexuality, and adolescent devel- opment shift in light of increased recognition of nonbinary thinking about identity (e.g., Barker & Iantaffi, 2019; Hyde et al., 2019). It is also vital as we reconsider social policies, cultural practices, and institutions which have long assumed binary conceptions of gender and sexuality, thereby obscuring the full range of diversity (e.g., Woolley, 2017). Such knowledge informs, for example, how settings that serve adolescents might best target services and programs for adolescents of particular identities. It also informs sensitivity about curriculum and the material conditions (e.g., bath- rooms), policies (policies highlighting “boys” and “girls”), and practices of institutional spaces (e.g., sex segregation) that have historically considered gender and sexuality in binary terms. Understanding the expansion of iden- tity labels among contemporary adolescents works toward social justice and positive development by honoring the ways in which adolescents engage with and challenge existing taxonomies of self-understanding (see Russell, 2016). Armed with new knowledge about identity labeling, the sci- ence of adolescent development might grow to better capture the diversity of lived experience of gender and sexuality, offering new perspectives to enhance adolescent development in the process. This article presents findings from a mixed-methods study of adolescents conducted from 2015 to 2017 in two distinct regions of California known for their historically2 high (i.e., the San Francisco Bay Area) or low (i.e., the Central Valley) support for sexual and gender diversity. Communities sup- portive of sexual and gender diversity are characterized by such factors as the presence of lesbian, gay, bisexual, transgender, queer (LGBTQ)-supportive businesses and organizations, high proportions of schools with “gay-straight” or “gender-sexuality” alliances (GSAs), more same-sex headed households, and visible rituals such as annual Pride events (e.g., Hatzenbuehler, 2011; Oswald et al., 2010). Our aim was to examine the way in which contempo- rary adolescents in diverse settings engage with shifting conceptions of gen- der and sexual identity. Using a concurrent, equal-status mixed-methods design (Creswell, 2009), we collected both quantitative survey data and qualitative ethnographic and 170 Journal of Adolescent Research 37(2) interview data from adolescent and adult informants affiliated with settings for sexual and gender minority youth (e.g., high school GSAs, community-based organizations, and online settings). Our key questions were as follows: Research Question 1 (RQ1): What labels do contemporary sexual and gender minority adolescents most often employ to describe their identities? Research Question 2 (RQ2): Does the use of particular labels vary by individual factors (e.g., sex assigned at birth) or contextual factors (e.g., community setting)? Research Question 3 (RQ3): How do adolescents across diverse settings experience the process of labeling, and what might account for variability of this experience? Our quantitative survey component was intended to address RQ1 and RQ2, and our qualitative component was intended to address RQ3. Gender and Sexual Identity Diversity in Adolescence: A Life Course Approach A life course approach to the study of adolescent sexual and gender diversity recognizes the significance of social and historical context for identity devel- opment and emphasizes the role of membership in a generation-cohort in the shaping of development (e.g., Cohler & Hammack, 2007; Hammack & Cohler, 2009; Hammack et al., 2018). Our research design centered the social context of development for our adolescent research participants, who were all born at the turn of the century, and the discourse about sexuality and gender to which they were exposed in the course of their development. Consistent with a social constructionist perspective on sexual and gender identity devel- opment (e.g., Yon-Leau & Muñoz-Laboy, 2010), we were interested in the way in which contemporary adolescents appropriate identity labels in dia- logue with existing cultural categories. Figure 1 highlights aspects of the social context of gender and sexual diversity during the lifetime of our participants. Two important larger histori- cal developments occurred during the childhood of this cohort: (a) the mar- riage equality movement and (b) the emergence of social media. Both of these developments likely played a central role in sexual and gender identity development for this cohort, in potentially both positive and negative ways. Yet, because the labeling and identity development process has been so understudied among adolescents of this generation, we know little about how the historical context has impacted development. Hammack et al. 171 Figure 1. Sexual and gender diversity in the 21st century. Note. DSM = Diagnostic and Statistical Manual of Mental Disorders; OED = Oxford English Dictionary; SCOTUS = Supreme Court of the United States; CA = California; MA = Massachusetts; OR = Oregon. The marriage equality movement involved debates about the legitimacy of same-sex relationships, which likely exposed members of this cohort to a discourse of uncertainty about the rights and recognition of sexual diversity at young ages (e.g., Corvino & Gallagher, 2012). This uncertainty was reflected in policy in California (the site of our study) with the initial passage of Proposition 8, which banned same-sex marriage, in 2008, and then its 2010 overruling by a federal judge whose decision explicitly invoked the discourse of diversity in nondiscrimination (Hammack & Windell, 2011). The second important historical development which occurred during the childhood of this cohort was the emergence of social media (Hur & Gupta, 2013; Manago et al., 2015), which facilitates connection with diverse others unbounded by the constraints of geographic proximity (Michikyan & Suárez- Orozco, 2016) and serves an empowering, protective, and community-build- ing role for adolescents (e.g., Craig & McInroy, 2014; Middaugh et al., 2017; Singh, 2013; Wang & Edwards, 2016). This generation is the first to have exposure to social media at very young ages, during a time in the history of social media when there was less regulation and thus more opportunity for diverse content and interaction among many subcultures. For example, Tumblr and Twitter were launched in 2007, when our research participants 172 Journal of Adolescent Research 37(2) were approximately 8 years old. Instagram was launched in 2010, when our participants were entering early adolescence and experiencing puberty. These and other social media platforms created a new social context for interaction and the formation of communities and subcultures (van Dijck, 2013). Although the impact of social media for LGBTQ+ youth can also involve heightened risk of bullying (Abreu & Kenny, 2018), such forms of open com- munication likely played a formidable role in the proliferation of new vocab- ularies and new taxonomies of gender and sexuality for members of this generation, including adolescents like Danny whose narrative of nonbinary gender identity highlighted engagement with social media. Research sug- gests that online settings provide sexual and gender minorities with opportu- nities for exploration with identity presentation (e.g., DeVito et al., 2018) and disclosure (e.g., Haimson et al., 2015) but that the specific affordances of particular settings (e.g., audience management) shapes the experience and the possibilities for development (e.g., Duguay, 2016). In addition, research sug- gests that the experience of online settings varies across subcommunities of gender and sexual minorities, as some online settings can be spaces in which intra-community stigma is enacted or intra-community tensions exacerbated (e.g., biphobia; Walker & DeVito, 2020). By the time this cohort entered adolescence, cultural discourse and policy debates had begun to shift toward transgender visibility and recognition of transgender rights and identities (e.g., Hughto et al., 2015; Levitt & Ippolito, 2014a, 2014b; Reisner et al., 2015). Marriage equality was formally achieved nationwide with the landmark U.S. Supreme Court ruling in Obergefell v. Hodges (2015). However, the election of Donald J. Trump as President of the United States in 2016 represented a setback in this regard, as the Justice Department under the administration of President Barack Obama had explic- itly endorsed transgender rights, but the Trump administration openly opposed transgender rights (e.g., military service). Greater cultural attention has more recently been paid to nonbinary gender identities, with states such as Oregon and California recognizing nonbinary identities in official govern- ment documents such as driver licenses. The adolescence of the current cohort has thus been characterized by heightened visibility of both transgen- der and nonbinary gender identities. Labeling Gender and Sexual Identity in the 21st Century In the context of this significant cultural change in the meaning of gender and sexual diversity, surprisingly little research has examined the appropriation of identity labels or the identity development process among adolescents. Hammack et al. 173 The majority of research on this population has examined social and psycho- logical experiences of having a gender or sexual minority status (e.g., Fish & Pasley, 2015; Reisner et al., 2015; Russell & Fish, 2016), without sufficient attention to the nuances of identity and labeling. Interrogating the process and meaning of identity labeling among contemporary sexual and gender minority adolescents is vital to our evolving understanding of adolescent development. The study of identity labeling draws our attention to distinc- tions within the larger population of sexual and gender minority youth (e.g., between cisgender and transgender youth, between monosexual [e.g., gay] and plurisexual [e.g., bisexual, pansexual] youth, between youth of different racial and ethnic identities; see below) and shifts us away from conflating experience across subcategories of gender and sexual identity. In this section, we review the limited recent research on labeling gender and sexual identity among contemporary adolescents. While labels like “gay,” “lesbian,” and “bisexual” were once the only common options for sexual minorities and “man” or “woman” the only com- mon options for gender identity, today new taxonomies have emerged or become more widespread and have challenged the notion that either sexual or gender identity are inherently binary (see Barker & Iantaffi, 2019; Mardell, 2016). A distinction is now commonly made between monosexual identity labels such as “gay,” “lesbian,” or “straight,” which signify a singu- lar direction in which one’s sexual attractions are oriented, and plurisexual identity labels such as “bisexual,” “pansexual,” and “queer,” which typically signify attraction to multiple genders when referring to sexual identity (e.g., Galupo et al., 2017; Hammack et al., 2019; Mitchell et al., 2015). The label “queer” may signify attraction to multiple genders but also may signify a gender identity (see below) or a political or cultural identity intended to challenge sexual or gender normativity (see Hammack et al., 2019; Miller et al., 2016; Morandini et al., 2017). We now also recognize some sexual identity labels which do not signify the gender(s) to which one is attracted but rather the degree of sexual attrac- tion one experiences. These labels fall under the larger asexual umbrella and include labels such as “graysexual” (limited degree of sexual desire) and “demisexual” (sexual desire occurs only in the context of an emotional con- nection; e.g., Van Houdenhove et al., 2015; Walton et al., 2016). With regard to gender identity, it is increasingly common to distinguish between cisgender (i.e., one’s current gender identity matches the sex assigned at birth) and transgender identities (i.e., one’s current gender iden- tity is different from the sex assigned at birth), as well as to recognize the existence of nonbinary or genderqueer (i.e., one’s gender identity is neither man nor woman) identities and genderfluid (i.e., one’s gender identity is not 174 Journal of Adolescent Research 37(2) experienced as fixed but rather fluctuates over time) identities (e.g., Hegarty et al., 2018; Richards et al., 2016). Simply put, both sexual and gender identity labels increasingly recognize the full spectrum of diversity beyond traditional taxonomic systems that center attraction to a binary gender identity. While we know that these newer identities and labels are now in use (Mardell, 2016; Robertson, 2019), we know little about the nuances of the labeling process or the individual factors associated with appropriation of particular labels. Some research has examined the distinction in use of the labels “bisexual” and “pansexual” because both signify an attraction to mul- tiple genders. These studies find similar reports of attraction and indices of sexual orientation for both pansexual and bisexual individuals (e.g., Flanders et al., 2017; Galupo et al., 2017; Morandini et al., 2017), though there appears to be a cohort effect in the distinction between the meaning and appropriation of these labels. Studies suggest that those who adopt pansexual labels are typically younger (Lapointe, 2017) and more likely to be transgender or gen- der nonconforming (Morandini et al., 2017). Labels such as “queer” (when used as a sexual identity label) and “pansexual” appear to be more common among transgender people and cisgender women (Goldberg et al., 2020; Morandini et al., 2017), which may be related to the way in which male sexu- ality has historically been framed as binary, and thus plurisexuality viewed as insincere or transitional (Yost & Thomas, 2012). Callis (2014) makes the distinction that many who adopt a pansexual identity do so to signal attrac- tion to transgender, nonbinary, or gender nonconforming people. The limited research on labeling among contemporary adolescents sug- gests that youth are creating new identities through new labels and sexual stories (e.g., Eisenberg et al., 2017; Lapointe, 2017; White et al., 2018). At the same time, consistent with research with the previous generation of ado- lescents (e.g., Hammack et al., 2009; Russell et al., 2009), youth continue to identify with binary labels such as “gay” or “lesbian,” but they often seek to challenge the stereotypes associated with these labels and express a desire to be seen as “ordinary” adolescents (e.g., Coleman-Fountain, 2014). In a study of 66 sexual and gender minority adolescents recruited through a commu- nity-based organization in British Columbia, Eisenberg and colleagues (2017) found that two thirds of participants used “traditional” sexual identity labels like gay, lesbian, or bisexual, while one third used “newer” labels such as pansexual or queer. They found that those who used newer sexual identity labels were also more likely to use newer gender identity labels such as gen- derqueer or fluid. In a nonprobability sample of more than 19,000 high school students in the United States, White and colleagues (2018) discovered that most youth continue to identify with conventional labels while also selecting new ways to describe their sexual or gender identity. Hammack et al. 175 Among a convenience sample of 175 LGBTQ youth below the age of 25 years in the United States, Bosse and Chiodo (2016) discovered that one third reported a gender identity distinct from their sex assigned at birth, signaling the rise in gender diversity among members of this generation. Importantly, they found that those AFAB were more likely than those assigned male at birth (AMAB) to report gender identities noncongruent with sex assigned at birth, as well as fluid and nonbinary gender identities. They also found that youth who reported gender identities noncongruent with sex assigned at birth were more likely to identify with a newer sexual identity label such as pan- sexual or queer, compared with more conventional labels such as gay, les- bian, or bisexual. The finding that youth who identify as nonbinary are more likely to be AFAB also emerged in the Canadian Trans Youth Health Survey, in which 82% of nonbinary youth were AFAB (Clark et al., 2018). In a large U.S. national sample of gender and sexual minority adolescents (ages 13–17 years), Watson and colleagues (2019) found that 24% identified with nontraditional labels such as pansexual and nonbinary. Appropriation of labels varied by both gender and racial identity groups, revealing consider- able heterogeneity within the larger category of sexual and gender minority adolescents. Youth with trans and nonbinary gender identities were less likely than their cisgender peers to identify with a traditional sexual identity label such as gay, lesbian, or bisexual. Cisgender girls with multiple ethnic/racial identities were more likely to identify with new labels such as pansexual, queer, and asexual. Confirming the emerging pattern that those AFAB are more likely to be challenging binary conceptions of gender and sexuality, 8% of their sample consisted of transgender boys versus only 1% transgender girls, and 20.9% identified as “nonbinary transmasculine” versus 2.8% “non- binary transfeminine.” An intersectional approach to the study of gender and sexual identity label- ing interrogates the way in which the constellation of identities one inhabits influences lived experience, particularly with regard to privilege and power (e.g., Cole, 2009). Such an approach dismantles uniform notions of LGBTQ identity or experience and considers variability on the basis of multiple cate- gory membership, such as distinctions based on race, class, gender, and ability (e.g., Coston & Kimmel, 2012; Hulko & Hovanes, 2018; Singh, 2013; Toft et al., 2020). An intersectional approach has rarely been centered in research on identity labeling among contemporary sexual and gender minority adoles- cents. A notable exception is the What’s Your Issue? project, which used criti- cal participatory action research methods to obtain a large, highly diverse national sample of LGBTQ+ adolescents (Fine, Torre, Frost, & Cabana, 2018; Fine, Torre, Frost, Cabana, & Avory, 2018). Survey respondents chal- lenged traditional categories of gender and sexual identity, revealing more 176 Journal of Adolescent Research 37(2) than 34 gender identity labels and even more sexual identity labels (Fine, Torre, Frost, Cabana, & Avory, 2018). They also found that queer and trans youth of color who have less support face more discrimination, which can lead to challenges with mental health and identity development. In sum, there is compelling evidence that contemporary adolescents in the United States, Canada, and other Anglophone societies are increasingly using an expanded vocabulary to describe their gender and sexual identities. The expansion of labeling options better captures diversity in gender and sexual- ity and shifts understandings away from traditional binaries which histori- cally obscured the full range of experience. Research has begun to document this cultural phenomenon, though the majority of empirical work with this population continues to focus on factors other than identity development or the labeling process itself. It is vital to understand the way in which the changing cultural landscape for gender and sexual diversity is impacting the discourse adolescents themselves use to describe their identities. Russell (2016) has argued that researchers need to revisit their assumptions about human development as adolescents construct new identities and use new labels and categories. A social justice approach calls researchers to challenge normative concepts such as the ontological stability of social categories related to sexuality and gender (Russell, 2016), and an intersectional approach challenges researchers of gender and sexuality in adolescence to recognize how experience and development varies depending upon unique configura- tions of identities which confer relative power and privilege (e.g., Santos & Toomey, 2018). In other words, researchers should be open to the way in which social categories are in constant states of creative redefinition from adolescents themselves (Hammack & Toolis, 2015), and they should reject singular approaches that eschew complexity and flatten power dynamics in identity development. The utility and vitality of the knowledge we produce about gender and sexual diversity in adolescence is predicated on an under- standing of the shifting language adolescents appropriate to make meaning of their experience of gender and sexual identity. The Current Study The current study sought to produce new knowledge about gender and sexual diversity among adolescents in diverse community settings by examining identity labeling in two distinct communities using a concurrent, equal-status mixed-methods design (Creswell, 2009). Using quantitative survey methods, we asked (a) what identity labels sexual and gender minority adolescents most often employed to describe their identities? and (b) whether use of par- ticular labels varied by individual factors (e.g., sex assigned at birth) or Hammack et al. 177 contextual factors (e.g., community setting)? Using qualitative methods, we asked (c) how adolescents across diverse settings experienced the process of labeling and what might account for the variability in experience? Through the triangulation of both quantitative and qualitative data, we sought to con- tribute new knowledge about the identity labeling process for sexual and gen- der minority adolescents in diverse community settings. Method Overview and Researcher Descriptions The study used a concurrent, equal-status mixed-methods design (Creswell, 2009) involving 20 months of ethnographic fieldwork and a self-report sur- vey administered in communities identified as either historically high (i.e., San Francisco Bay Area) or historically low (i.e., Central Valley) in support of gender and sexual diversity. In this type of design, qualitative and quantita- tive data are collected simultaneously over the same approximate period, and their value is weighted equally (Creswell, 2009). In the case of this study, the methods addressed distinct questions about gender and sexual identity label- ing, as noted above. All research procedures were approved by the Institutional Review Board at the University of California, Santa Cruz. For reflexivity, we note that the five authors hold diverse configurations of identities of relevance to the project, including cisgender (AFAB and AMAB), genderfluid trans femme (AMAB), gay (AMAB), bisexual (AFAB), straight (AMAB), and queer (AMAB). The authorship team consisted of individuals who identify as White and Black and from diverse social class backgrounds. The researchers were committed to a process of constant reflexivity in which they reflected on the ways in which these identities and the absence of other perspectives on the research team might impact data collection or analysis. Consistent with guidelines for qualitative and mixed-methods research (Levitt et al., 2018), several procedures were undertaken to integrate consid- erations of reflexivity throughout the data collection and analysis processes. During data collection, regular meetings among the authors included a larger research team with members who held other diverse gender, sexual, and racial identities. The perspectives of the full research team during data collec- tion thus included a broader range of voices, and considerable effort was made to ensure that the research included a diverse representation of partici- pants. During data collection and data analysis, the researchers created reflexivity memos in which they consistently reflected on their own identities and relative positions of power in relation to the research participants. The research team met regularly to discuss these memos. Our constructionist 178 Journal of Adolescent Research 37(2) epistemology recognizes that the knowledge we produce is the result of co- constructed meaning making between us and our participants (Madill et al., 2000). This reflexive process was intended to ensure methodological integ- rity and fidelity in our interpretive process (Levitt et al., 2017, 2018). Qualitative Data Collection We identified our field sites using a method of community climate assess- ment for support of sexual and gender diversity (see Oswald et al., 2010). Community climate refers to the level of support of a particular context when it comes to sentiments and policies related to gender and sexual diversity. It has been formally operationalized as “the level of community support for [sexual diversity], and indicated by objectively measurable phenomena such as religious and political affiliations, legal rights, workplace opportunities and policies, and the presence of GLBT community members and services” (Oswald et al., 2010, p. 215). In two regions of California known for their historic high (i.e., the San Francisco Bay Area) or low (i.e., the Central Valley) support for sexual and gender diversity, we randomly selected3 10 counties for in-depth analysis using a modified method of community climate assessment (Oswald et al., 2010). This method uses publicly available data such as the number of same- sex headed households in a county, the number of registered democrats, and the proportion of high schools with a GSA to calculate a quantitative index of support for sexual and gender diversity. To maximize our ability to examine the relationship between community climate and social psychological pro- cesses, we targeted counties at the two poles of the spectrum of support for sexual and gender diversity. Across the 10 counties, Alameda County (adja- cent to San Francisco; population 1.66 million) was determined to be the most supportive. Madera, Kings, and Tulare counties (clustered around Fresno) were determined to be the least supportive and were combined for an analytic focus to include a larger population (combined population: 771,484). Ethnographic fieldwork. The third and fourth authors served as field research- ers based in the Central Valley and the San Francisco Bay Area, respectively. Ethnographic fieldwork occurred over a 20-month period beginning in November, 2015, and ending in July, 2017, and consisted of active participant observation in the counties, including at high schools, community-based organizations, and community rituals for LGBTQ+ youth (e.g., Pride). Field researchers completed ethnographic field notes following procedures out- lined by Emerson et al. (2011). The field researchers met weekly via video conferencing with the principal investigator (the first author) and the project coordinators (the second and fifth authors). Hammack et al. 179 Adult informant interviews. In addition to participant observation, ethno- graphic fieldwork involved semi-structured interviews with adult and youth leaders in the communities who served as informants. Because the larger study sought data about community climate for youth, we wanted to capture the perspectives of not just youth themselves but also adults who work closely with them and could provide a broader historical perspective on the community. This approach to sampling informants was based on our a priori focus on the role of community climate in the identity labeling process and is consistent with established tradition in ethnographic research (e.g., Gold, 1997; Johnson, 1990). We sought informants who were most likely to have cultural expertise on community climate and thus best able to address our research questions. Adult informants (N = 24) were recruited by the field researchers based on two criteria: (a) affiliation with an organization or institution that serves LGBTQ+ youth, and (b) at least 3 years of direct experience with LGBTQ+ youth in some capacity. We sought to ensure a diversity of representation among adult informants interviewed in terms of gender, sexual, and racial/ ethnic identities. Following standards in qualitative data collection (e.g., Guest et al., 2006; Levitt et al., 2018), we did not have a target sample size for interviews but rather collected data until the research team agreed we had reached saturation in terms of key themes emerging in the interviews. Interviews were conducted in a location selected by the interviewee, most commonly in a private office at their place of employment. Field researchers described the project as “looking to better understand what local community contexts are like across California for LGBTQ youth today,... what resources exist in local communities as well as the general climate in different com- munities with regard to sexual and gender identity diversity.” All interview- ees provided written informed consent. Interviewees completed a brief demographic form providing information about their age, race/ethnicity, gender identity, pronouns, assigned sex at birth, sexual identity, and length of residence in the community in which the interview was conducted. The interview protocol covered such topics as their perceptions of community climate toward gender and sexual diversity, dis- crimination/violence incidents, and health and mental health concerns and resources for LGBTQ+ youth. Interviews ranged from approximately 1 to 3 hours (M = 101 minutes, SD = 37 minutes), were recorded using a digital audio recorder, were transcribed by a professional transcription company, and were uploaded to a secure server with identifying information redacted. Interviewees were provided with a US$20 cash incentive for participation. Field researchers completed case reports following each interview which served as another source of qualitative data. These reports included 180 Journal of Adolescent Research 37(2) observational data surrounding the interview and provided a space for the field researchers to engage in a reflexivity process in which they consid- ered the role of their configuration of identities and personal reactions in the interview encounter. The reports also provided space for initial reflec- tion on recurring themes, which was used to determine saturation for data collection. Youth informant interviews. The same informant sampling approach used for adults (Gold, 1997; Johnson, 1990) informed our recruitment of youth inter- viewees. Because we were guided by the a priori concern with community climate, we sought youth who were most likely to have extensive cultural knowledge of the community by assuming a leadership role. (We note that this sampling approach presents a limitation in that the data are not intended to generalize to all LGBTQ+ youth in the community. See the section “Dis- cussion” for more details.) Youth informants (N = 28) were recruited both through nominations provided by adult informants, as well as through recommendations of field researchers based on participant observation. Youth were eligible to be an informant if they were (a) between the ages of 14 and 21 years, (b) consis- tently active in LGBTQ+ spaces (e.g., GSAs, community-based organiza- tions), and (c) currently or previously held a leadership role in the community (though they did not need to have an official organizational title or position). Although our focus was on high-school aged adolescents, we expanded eligibility criteria for our sample of youth informants up to age 21 years because most spaces and organizations in the field considered youth programs to serve individuals up to age 21 years. All but two infor- mants were 18 years old or younger, with a median age of 16 years. As we recruited youth informants, we sought representation of a diversity of gen- der, sexual, and racial/ethnic identities within communities, and we ceased data collection upon saturation. Interviews were conducted in a location of the interviewee’s choice and included public library conference rooms, empty school classrooms, private rooms in community centers, and rented office space. Field researchers described the project using the same language as for the adult interviews, as noted above. A waiver of parental consent was granted by the institutional review board, based on the rationale that the requirement of parental consent for sexual and gender minority youth can represent a high level of risk depending on their disclosure status with parents (Mustanski, 2011). Although we recognized that youth leaders were more likely to be public about their identities and thus at lower risk, we did not want to be limited in our ability to recruit informants and thus sought and received the waiver of parental Hammack et al. 181 consent. Youth interviewees below 18 years provided written assent to par- ticipate in the research, and youth 18 years and older provided written con- sent to participate. Interviewees completed a demographic form, providing information on age, race/ethnicity, gender identity, pronouns, assigned sex at birth, sexual identity, and length of residence in the community. The interview protocol consisted of the same content as for adult interviews noted above. Interviews ranged from approximately 1 to 3 hours (M = 72 minutes, SD = 25 minutes), were recorded using a digital audio recorder, were transcribed by a profes- sional transcription company, and were uploaded to a secure server with iden- tifying information redacted. Interviewees received a US$20 cash incentive for participation. As for the adult interviews, field researchers completed case notes following each youth interview, documenting their overall impressions of the interview, emerging themes, and their own experience of reflexivity and personal reactions. Qualitative Data Analysis Qualitative data analysis proceeded in three stages. First, the first, third, and fourth authors completed “pre-analysis reflexivity memos” which provided space to reflect upon their own gender and sexual identity development, per- sonal connection to sexual and gender diversity, feelings about the research project, expectations of data analysis, and intersectionality (i.e., how the identities we hold or the intersection of our own social identities might impact data analysis). The three coders met several times to discuss these memos, in an effort to create a unified interpretive community, aware of its positionality vis-à-vis the focus of study. This approach was consistent with our interpre- tive epistemology and was intended to ensure fidelity to the data analytic process (see Levitt et al., 2017, 2018; Madill et al., 2000). In the second phase of analysis, the three coders established a preliminary codebook based on the main foci of the larger study. Codes were constructed as descriptive and for use in content analysis (e.g., Hsieh & Shannon, 2005). For example, we constructed codes about “community climate” to flag any content discussing the community climate toward sexual and gender diver- sity. In this article, we focus on qualitative data content related to identity and labeling, defined as “text related to gender or sexual identities, processes of using labels, or meaning of particular labels.” We created a child code to identify content specifically related to nonbinary gender or sexual identity labels, defined as “text related to nonbinary gender or sexual identity labels, such as pansexual, queer, nonbinary, genderqueer, etc.” We also created a content code for any data related to gender, defined as “text related to gender 182 Journal of Adolescent Research 37(2) diversity,” with three child codes related to gender dysphoria, gender presen- tation/conformity, and gender roles/ideology. Finally, we created a content code for any data related to the transgender experience, with child codes related to distinction from the LGB community, trans-specific resources, and transgender identity development. In the third phase of analysis, the three coders collaboratively coded a selection of data using Dedoose qualitative analytic software to establish interpretive and technical consistency in the application of content codes. Once we reached a point of consistency, two coders (the third and fourth authors) independently applied codes to the data. The first author then served as an external auditor of the remaining coding, and the fifth author served as an auditor for all coding presented in the article. This general analytic approach is commonly employed in qualitative research on sexual and gen- der diversity (e.g., Galupo et al., 2019). Survey Data Collection Participants and procedure. We used a venue-based purposive sampling strat- egy to obtain a nonprobability, community-based sample of adolescents residing in the four counties targeted for study. This strategy is considered ideal to capture a diversity of perspectives among those who identify as gen- der or sexual minorities (see Krueger et al., 2020; Meyer & Wilson, 2009). The survey was conducted from October, 2016, to June, 2017. Respondents were eligible if they were between the ages of 14 and 18 years, had lived in one of the four counties for at least 1 year, and identified as transgender or nonbinary (or any other label associated with nonbinary or fluid gender) and/ or lesbian, gay, bisexual, pansexual, queer, or any other nonheterosexual identity, or any asexual identity. A target sample of approximately 150 per region was determined following consultation with statisticians about planned quantitative analyses. Respondents received a US$20 gift card to an online retailer as an incentive for their participation. Sampling venues were selected to ensure a wide diversity of cultural, political, racial/ethnic, gender, and sexual representation within demograph- ics of interest. To avoid sampling bias related to particular venues, we recruited participants from each of the seven following mechanisms: (a) commercial establishments (e.g., coffee shops, book stores), (b) outdoors (e.g., parks, streets), (c) groups (e.g., community-based organizations and groups organized around a variety of activities or interests such as sports, politics, culture, racial, ethnic, or national interests), (d) events (e.g., Gay Pride), (e) schools, (f) online settings (e.g., Instagram, Facebook), and (g) snowball sampling (i.e., referral from other respondents). Hammack et al. 183 Survey respondents consisted of 314 adolescents (aged 14–18 years), who resided in one of the four counties of focused study. Respondents completed the survey online using SurveyGizmo. They first completed an eligibility screener, which determined that respondents met the eligibility criteria related to age (14–18 years old), residence within one of the four target counties for at least 1 year, and identification criteria outlined above. Measures Assigned sex at birth. Participants were asked to indicate their assigned sex at birth (female or male) on their original birth certificate. Gender identity. Following the recommended approach to assess gender identity using two items (Gender Identity in U.S. Surveillance Group, 2014), participants were asked to select among five options any gender identity label(s) that they thought best described them: girl/woman, boy/ man, transgender girl or woman/male-to-female (MTF), transgender boy or man/female-to-male (FTM), and nonbinary/genderqueer. They were offered an open-response “other” option if their current gender identity label did not match one of these five. Sexual identity. Following best practices for the assessment of sexual orien- tation in survey research (Sexual Minority Assessment Research Team, 2009), participants were asked to select among a list of eight sexual identity labels any terms with which they identified (i.e., they could select more than one): straight/heterosexual, lesbian, gay, bisexual, queer, pansexual, trans-amorous/trans-attracted, and asexual. They were offered an open- response “other” option where they could include other sexual identity labels not included in our list. Sexual attraction. Participants were asked to indicate their current sexual attraction on a 3-point scale (1 = not at all attracted, 3 = very attracted) to the following groups: woman (non-transgender), man (non-transgender), transgender woman, transgender man, and nonbinary/genderqueer. Racial/ethnic identity. Participants were asked to indicate which race/ethnicity terms applied to them from a list of eight options: American Indian or Alas- kan Native, Asian/Asian American, Black/African American, Biracial/multi- racial, Hispanic/Latino or Spanish origin, Middle Eastern/North African, Native Hawaiian/Pacific Islander, and White/Caucasian/European American. They could select all that applied. They were offered an open-response 184 Journal of Adolescent Research 37(2) “other” option where they could include a race/ethnicity label not offered on our list. Parental education. Participants were asked to indicate the highest level of education completed by the parent who went the farthest in school. They were provided with five options: did not finish high school, graduated from high school, attended college but did not complete 4-year degree, graduated from college, don’t know. Analytic Strategy Our analytic plan to address RQ1 (What identity labels do sexual and gender minority adolescents employ to describe their identities?) was to examine frequencies of response for gender and sexual identity labels. To address RQ2 (Does use of particular labels vary by individual factors or contextual fac- tors?), we examined proportions of respondents within label categories who possessed the particular individual (i.e., assigned sex, race/ethnicity, parental education) and contextual (i.e., community) characteristics of interest in the study. We computed confidence intervals (CIs) for phi coefficients between each sexual and gender identity label with each of the dichotomous variables (i.e., assigned sex at birth, White vs. non-White race, high-support vs. low- support community). We also computed CIs for point-biserial correlations between each sexual and gender identity label and the ordinal variable (i.e., parental education). Both effect sizes carry similar interpretations to a Pearson correlation. Results Our mixed-methods approach integrated qualitative and quantitative data to examine the appropriation of sexual and gender identity labels in diverse community contexts. Addressing RQ1 and RQ2, we used the quantitative survey data to examine the frequency of particular labels across community settings, examining similarities and differences in appropriation of labels by community, sex assigned at birth, race/ethnicity, and parental education. Addressing RQ3, we used the qualitative data to interrogate the meaning and experience of labeling and to provide nuance and understanding beyond the survey data. Data triangulation occurred among the authors during meetings through- out the data analysis and manuscript preparation phases. This integrative pro- cess involved addressing the three research questions by identifying themes that emerged across analyses of the data sets, consistent with triangulation Hammack et al. 185 Figure 2. Patterns of findings based on triangulated data sources: Pattern 1. Note. AFAB = assigned female at birth. Figure 3. Patterns of findings based on triangulated data sources: Pattern 2. Note. LGBTQ = lesbian, gay, bisexual, transgender, queer; AFAB = assigned female at birth; AMAB = assigned male at birth. procedures for a concurrent, equal-status mixed-methods design (Creswell, 2009). This process resulted in seven patterns of findings which are high- lighted in Figures 2 to 6. As indicated in the figures, four of these patterns emerged across the data sources, whereas three emerged solely from analysis of the qualitative data (see Figure 6). We organize the presentation of findings by pattern, integrating quantitative and qualitative data to reflect the princi- ples of our concurrent, equal-status mixed-methods design. For each pattern, we highlight points of elaboration. 186 Journal of Adolescent Research 37(2) Figure 4. Patterns of findings based on triangulated data sources: Pattern 3. Note. AFAB = assigned female at birth. Figure 5. Patterns of findings based on triangulated data sources: Pattern 4. Note. AFAB = assigned female at birth. Pattern 1: Nonbinary Gender Identity Labels Used Frequently The finding that adolescents used nonbinary (i.e., neither man nor woman) gender identity labels frequently emerged from three data sources: (a) eth- nographic field notes, (b) survey data, and (c) informant interviews (see Figure 2). Field notes revealed numerous instances of adolescents in Hammack et al. 187 Figure 6. Patterns of findings exclusively from qualitative data sources. Note. LGBTQ = lesbian, gay, bisexual, transgender, queer. LGBTQ+ spaces who identified as nonbinary, genderqueer, or some other nonbinary gender identity label, as well as frequent use of they/them pro- nouns. Table 1 reports demographics for the survey sample, including fre- quencies of all gender identity labels. Nearly one quarter of the sample (23.9%, n = 75) indicated a nonbinary gender identity by selecting “nonbi- nary/genderqueer” and/or writing in a label such as “genderfluid” (n = 5), “bigender” (n = 2), or “demiboy” (n = 2). We discovered two points of elaboration in this pattern. First, those who identified with a nonbinary gender identity label were more likely to be AFAB (n = 59, 78.7%) than AMAB, ϕ =.151, 95% CI = [.047,.255], p <.01. Second, the use of nonbinary gender identity labels was more frequent among youth residing in high-support communities (29.6% in high-support vs. 18.1% in low-support communities, ϕ =.135, 95% CI = [.024,.245], p <.05) and among youth whose parents had a higher education level, rpb =.113, 95% CI = [.002,.219], p <.05. Qualitative data revealed the value and meaning of a nonbinary identity for those adolescents who identified as such. Danny (he/him/his), a 17-year- old Latinx gay nonbinary (AFAB) youth from a low-support community, nar- rated his experience of coming to identify as nonbinary:  Table 1. Sample Demographics and Survey Findings. 188 Region N % Bay Area 159 50.6 Central Valley 155 49.4 Tulare County 91 29.0 Madera County 35 11.1 Kings County 29 9.2 Gender identity N % Cisgender boy 81 25.8 Chose “Boy/Man” and “AMAB” only 77 24.5 Chose “Boy/Man,” “AMAB,” and “Nonbinary/Genderqueer” 2 0.6 Chose “Boy/Man,” “AMAB,” and other labels 2 0.6 Cisgender girl 136 43.3 Chose “Girl/Woman” and “AFAB” only 127 40.4 Chose “Girl/Woman,” “AFAB,” and “Nonbinary/Genderqueer” 9 2.9 Transgender boy 26 8.3 Chose “Transgender Boy or Man/FTM” and “AFAB,” but not “Boy/Man” 10 3.2 Chose “Boy/Man,” “AFAB,” and “Transgender Boy or Man/Female-to-Male (FTM)” 6 1.9 Chose “Boy/Man” and “AFAB,” but not “Transgender Boy or Man/ Female-to- Male (FTM)” 5 1.6 Chose “Boy/Man,” “AFAB,” and some nonbinary gender identity (e.g., nonbinary/genderqueer, bigender, genderfluid) 3 1.0 Chose “Boy/Man,” “AFAB,” “Transgender Boy or Man/Female-to-Male (FTM),” and “Nonbinary/Genderqueer” 1 0.3 Chose “Transgender Boy or Man/Female-to-Male (FTM),” “AMAB,” “Nonbinary/Genderqueer,” and wrote in “Demiboy” 1 0.3 Transgender girl 13 4.1 Chose “Transgender Girl or Woman/MTF” and “AMAB,” but not “Girl/Woman” 8 2.5 Chose “Girl/Woman,” “AMAB,” and “Transgender Girl or Woman/MTF” 1 0.3 Chose “Girl/Woman” and “AMAB,” but not “Transgender Girl or Woman/MTF” 4 1.3 Nonbinary gender identitya 75 23.9 Chose “Nonbinary/Genderqueer” and/or wrote in a nonbinary gender (“genderfluid,” “bigender,” “demiwoman”), and chose “AFAB” 59 18.8 Chose “Nonbinary/Genderqueer” and/or wrote in a nonbinary gender (“demiboy”), and chose “AMAB” 16 5.1 (continued)  Table 1. (continued) Sexual identity N % Asexual spectrum identity 31 9.9 Chose “Asexual” only 14 4.5 Chose “Asexual” with other sexual identity labels 14 4.5 Wrote in one or more asexual spectrum labels but did not choose “Asexual” (e.g., demisexual, graysexual) 3 1.0 Bisexual 121 38.5 Chose “Bisexual” only 88 28.0 Chose “Bisexual” with other sexual identity labels 32 10.2 Wrote in “Bicurious” 1 0.3 Gay 68 21.7 Cis or trans boy who chose “Gay” only 45 14.3 Cis or trans boy who chose “Gay” with other labels to signify plurisexual attraction (e.g., bisexual, pansexual) 15 4.8 Cis or trans boy who chose “Gay” and wrote in “Transgender” for sexual identity and reported plurisexual attraction pattern 1 0.3 Cis or trans girl who chose “Gay” only 2 0.6 Cis or trans girl who chose “Gay” and “Lesbian” 3 1.0 Nonbinary who chose “Gay” only and reported plurisexual attraction pattern 2 0.6 Lesbian 41 13.1 Chose “Lesbian” only 28 8.9 Chose “Lesbian” with other sexual identity labels to signify plurisexual attraction (e.g., bisexual, pansexual) 9 2.9 Chose “Lesbian” and “Gay” 3 1.0 Chose “Lesbian” with an asexual spectrum identity and indicated plurisexual attraction pattern 1 0.3 Pansexual 69 22.0 Chose “Pansexual” only 46 14.6 Chose “Pansexual” with other sexual identity labels to signify plurisexual attraction (e.g., bisexual) 15 4.8 Chose “Pansexual” with an asexual spectrum identity 6 1.9 Wrote “Polysexual” 2 0.1 (continued) 189 190 Table 1. (continued) Sexual identity N % Queer 39 12.4 Chose “Queer” only 14 4.5    Chose only “Queer” and had plurisexual attraction patternb 10 3.2    Chose only “Queer” but reported monosexual attractionb 1 0.3    Chose only “Queer” but attraction patterns did not indicate plurisexuality or monosexualityb 3 1.0 Chose “Queer” with other sexual identity labels to signify plurisexual attraction (e.g., bisexual, pansexual) 16 5.1 Chose “Queer” with “Gay” or “Lesbian”c 11 3.5 Chose “Queer” with an asexual spectrum identityc 4 1.3 Straight 15 4.8 Chose “Straight” only 12 3.8 Chose “Straight” with other sexual identity labels to signify plurisexual attraction (e.g., bisexual, pansexual) 1 0.3 Chose “Straight” with an asexual spectrum identity 2 0.6 Transamorous 1 0.3 Chose “Transamorous” only 0 0 Chose “Transamorous” with “Bisexual,” “Gay,” “Pansexual,” and “Queer” 1 0.3 Educational attainment of most educated parent N % Did not finish high school 59 18.8 Graduated from high school 46 14.6 Attended college but did not complete 4 years 54 17.2 Graduated from college 71 22.6 Attended graduate school or advanced degree 9 2.9 Graduated from advanced degree program 74 23.6 No response 1 0.3 (continued)  Table 1. (continued) Race/ethnicity N % American Indian or Alaskan Native 23 7.3 Asian/Asian American 55 17.5 Biracial or Multiraciald 36 11.5 Black/African American 10 3.2 Hispanic, Latino, or Spanish Origin 121 38.5 Middle Eastern/North African 6 1.9 Native Hawaiian/Pacific Islander 11 3.5 White/Caucasian/European American 146 46.5 Note. AMAB = assigned male at birth; AFAB = assigned female at birth; FTM = female-to-male; MTF = male-to-female. a This category includes participants who selected “Nonbinary/Genderqueer” and/or wrote in nonbinary labels such as bigender, genderfluid, demiboy, or demiwoman. b Among participants identifying only as “Queer,” these categories were determined by looking at the extent to which participants reported they were sexually attracted to three groups: (a) men (cis and/or trans), (b) women (cis and/or trans), and (c) nonbinary people. Participants indicating they were somewhat or very attracted to two or more of these three groups were considered to have a plurisexual attraction pattern. Indicating attraction to just one of the three groups suggested a monosexual attraction pattern, and indicating attraction to none of the three groups was considered to have neither a plurisexual nor monosexual attraction pattern. cA small number of participants in these categories also indicated a plurisexual identity label, so this grouping totals more than 100%. dThis category only includes people who explicitly identified as Biracial or Multiracial. Participants who identified with multiple racial groups, but did not identify as Biracial or Multiracial are not included here. 191 192 Journal of Adolescent Research 37(2)... When I got [to] high school, when I started learning more about these terms that I didn’t really know of, like nonbinary, or being trans, in general. Because I knew that I didn’t feel all the way like a boy. I did feel like a boy, but I didn’t at the same time. I was always, like, I don’t know what it is. When I finally cut my hair when I was around sophomore year, that’s when it felt right. That’s when it was like, okay, this feels a lot better. Adolescents like Danny revealed the value of expanded language and opportu- nities to identify beyond traditional binaries of “man” and “woman,” often accompanied by shifts in gender presentation to embody a nonbinary identity. While both our quantitative and qualitative data revealed a prominent level of visibility for adolescents who identified as nonbinary, our qualitative data also revealed challenges related to intelligibility. Marina (she/her/hers, they/them/their), a 17-year-old Latinx gay genderqueer (AFAB) youth from a low-support community, narrated this challenge: I just don’t wanna have to explain myself, I guess. When you say you’re gay, people know what gay means. People don’t know what genderqueer means. There’s so many different genders. People think sexuality is complicated. Gender is 10 times more complicated than sexuality is, in my opinion. I feel like it’s such—it’s different for anyone. My definition of genderqueer can be somebody else’s definition of genderqueer. I don’t want to be like, “I’m this way,” and have to explain myself and my interpretation of gender. I don’t really think of it as gender. I think of it as more of a feeling. If I look at [a form] where I have to fill out a gender, I’m not like, “I should’ve picked male. I should’ve picked female.” I’m automatically gonna pick female. I feel genderqueer, but I don’t wanna change my name and stuff like that. It’s just different for anyone. I don’t talk about it a lot, because I just don’t wanna have to explain myself. Marina’s narrative revealed the way in which youth experience gender in myriad ways that may not be easily subsumed within a single category of “nonbinary” or “genderqueer,” as they acknowledged their own experience of gender is unique. Furthermore, they illustrated the potential exhaustion and stress those who identify with nonbinary gender labels may experience at having to educate or explain their identities in every interaction. Pattern 2: AFAB Adolescents More Comfortable With Diverse Gender Expression; AMAB Adolescents Experience Pressures to Conform to Compulsive Masculinity The second pattern to emerge across multiple data sources concerned the distinction of experience between AFAB and AMAB adolescents in gender Hammack et al. 193 identity labeling and gender expression more broadly (see Figure 3). Ethnographic field notes consistently revealed the predominance of AFAB adolescents in LGBTQ+ spaces such as GSAs and community-based organi- zations. Across communities, field notes described challenges to recruit AMAB adolescents for both the survey and informant interviews. Considerable effort was undertaken to obtain adequate representation of AMAB adolescents in those data sources, and our concurrent design allowed us to integrate questions about the reduced visibility of AMAB adolescents, especially in our informant interviews. As indicated, the vast majority of survey respondents who identified with a nonbinary gender identity label were AFAB (78.7%). Notably more survey respondents could be classified as a transgender boy (n = 26, 8.3%) than a transgender girl (n = 13, 4.1%) based on their responses (see Table 1). These quantitative findings suggested greater comfort with gender diversity and flu- idity among AFAB adolescents in our sample. Qualitative data obtained in informant interviews affirmed this pattern and provided elaboration on experiences related to gender socialization which may explain the distinction between AFAB and AMAB adolescents. Dalton (he/him/his), a 27-year-old cisgender gay White male (AMAB) adult infor- mant from a high-support community, shared his impressions of this phe- nomenon based on his extensive work with LGBTQ+ youth: I do think it’s a trend that female-identified folks or even female-presenting folks oftentimes feel less pressure to choose a label. In my experience, female- identified or female-presenting folks say, “Well, I don’t know what my gender is,” or, “I don’t really know what my sexuality is. I may be just queer. I don’t know. My gender is just whatever it is right now.” There is a fluidity to it. There is an uncertainty to it. I do not see that—I rarely see that—in male-identified and male-presenting folks... Folks who are [assigned] female at birth and who are cultured to be female, have more—it’s safer for them. It’s perceived as safer for them to be flexible, to be flexible in their sexuality, in their gender identity, in their gender expression. Whatever it is, it’s safer for them to be flexible. Whereas boys don’t feel safe being flexible. They don’t feel safe. I think they feel less safe changing, being flexible, or choosing something outside of the norm and sticking to it. They feel less safe either way, but they feel definitely less safe just being flexible, being undecided, being questioning. I think that that is, yeah, internalized belief about how they’re supposed to behave. That’s ingrained into them from a very young age. Dalton attributed the higher proportion of nonbinary and transgender youth assigned female at least in part to gender socialization—the idea that those assigned female feel safer and more comfortable with flexibility and fluidity 194 Journal of Adolescent Research 37(2) in the identity development process than those assigned male. The language of “safety” seems especially important in Dalton’s narrative, as it indicates the fear of violence that those AMAB experience for gender nonconformity or diverse gender expression. In our survey data, we noted the relatively lower proportion of boys who could be classified as cisgender (n = 81, 25.8%) compared with cisgender girls (n = 136, 43.3%) affiliating with LGBTQ+ spaces, further suggesting greater comfort with gender and sexual diversity among those AFAB. Qualitative data revealed the way in which cisgender boys experienced pres- sures related to compulsive masculinity (Pascoe, 2007) that discouraged them from being visible. The narrative of John (he/him/his), a 16-year-old multira- cial gay boy (AMAB) from a high-support community, illustrated as follows: Masculinity, especially at my school, is very important. I feel like a lot of guys, when it comes to if they want to name call or something like that, usually attacks masculinity, attacks their manhood. In his interview, John narrated the experience of feeling constrained by pres- sures to conform to a masculine standard, and he and others in our ethno- graphic work credited the lower visibility of cisgender boys in LGBTQ+ spaces to this issue. Jade (she/her/hers), a 16-year-old White queer female (AFAB) from a high-support community, echoed the sentiments of boys like John: Many more female-identified people are out than male-identified people at this school. I think, as a guy in high school, it’s a lot harder to come out cause masculinity’s just a crap shoot. Jade noted particularly her impression of a rise in transgender visibility at her school, while suggesting greater challenges for trans girls: More trans people have been coming out the past year. [One of our teachers], he’s trans. I have trans friends here. If you look at—so, we have an independent study section. The number of trans students in independent study, the percentage of the total, and the number of trans students percentage of the total, and the regular school, is a huge disparity there. Because trans students don’t feel safe here, especially [trans girls]. Jade’s awareness of the high proportion of trans peers in independent study (i.e., out of the classroom), and particularly her claim that these are mainly trans girls, speaks to the heightened sensitivity of gender expression and the denigration of feminine expression or embodiment for those AMAB. Hammack et al. 195 Marina reflected on the challenges of visibility and disclosure for those AMAB and diverse in their gender or sexual identities: I believe it all has to do with gender norms and stuff, and stereotypes. Males have to be masculine. When you think of gay, you think of feminine. You think gay equates femini[nity]... I feel like it all has to do with masculinity and stuff like that. For a girl, it’s easy. You always have guy friends. Guys always have their guy friends who will be afraid of the whole like if you’re gonna do something... Even for transgender people, I feel like it’s easier to go from female to male than from male to female, because it’s so much easier to portray a guy, I guess. Because when you have to portray female, it’s really hard to do that. You have to have the hair, the body, the makeup. It’s just a female structure is so much harder. For me, I just cut my hair off, and everybody thinks I’m a boy. It’s just easier. [Chuckles] In all aspects. It’s hard to just portray femini[nity]. Guys, I just think guys have it harder. Marina noted that those AMAB struggle more with compulsory notions of masculinity and suggested a kind of openness within femininity that facili- tates visibility for those AFAB in the LGBTQ+ community. Dalton, the adult informant from a high-support community, linked these compulsory notions of masculinity to fear, offering a reflection based on his work with youth in the community: I think [young] men are much more... afraid of the process of figuring out what’s going on in them. They’re very nervous about the process itself, what’s gonna happen on the other side of that process, what’s it gonna tell them about themselves, all this stuff. I don’t see that fear in the women that come to group or the young women that come to group or the young female-identified folks that come to group. Dalton’s narrative revealed the way in which those assigned male undergo a process characterized by more fear and anxiety about the implications of their identity development process for how they are treated by peers and in society. Taken together, these narratives reveal the psychological injustice of compul- sive masculinity—the fear of direct violence that emerges from boys’ con- cerns about gender nonconformity and diverse gender expression. JB (they/them/their), a 26-year-old Black queer transgender male (AFAB) adult informant from a high-support community, spoke directly about the lower visibility of cisgender boys in LGBTQ+ spaces: I think that it’s just really due to—at least for our young men, it’s not something that’s accepted in their schools. I mean you can just get your ass beat for being feminine or gay or... I think it makes it harder to wanna come out, and if you 196 Journal of Adolescent Research 37(2) fear that, that’s gonna be the reaction. I think that’s the general part of just a lack of safety and fear of either violence or retaliation. Every once in a while, I have a boy or two that’s just... it will be so obvious to me that it’s a queer kid, and—but he won’t identify that way. JB provided greater specificity about the source of fear for many boys: the fear of direct violence on account of gender nonconformity or sexual minority status. These sentiments about the way in which compulsive notions of masculin- ity impact cisgender nonheterosexual boys were echoed by Thomas (he/him/ his), a 52-year-old Chicano gay male (AMAB) adult informant in a high- support community: I guess in a larger way, the way that society polices masculinity along heteronormative lines and cisnormative lines, if you see that as being a particular struggle that might inhibit or make it harder for LGBTQ boys and men... I think when they feel—from listening to these young gay men, it’s when their male friends, when their straight male friend’s masculinity feels threatened is when they feel uncomfortable. Thomas shared his experience of young gay boys referencing themselves to their straight male peers, perhaps more than to peers in the larger LGBTQ community. This observation is consistent with our ethnographic fieldwork, in which we found a lower proportion of cisgender gay boys engaging in LGBTQ spaces. Thomas added a further layer of reflection on the experience of gay boys rooted in his age and membership in a cohort of gay men decimated by the AIDS epidemic:... My generation, those of us who are in our late 40s and 50s, we’re not as present because of this gap that was created by the [AIDS] pandemic because they were busy fighting for their lives for those of us who had come out either right before or during the cocktails. It changed everything in terms of what that meant in terms of getting positive... There’s very few of us, that’s why I said there’s very few of us that are in the schools doing the work in the community doing this kind of work because we were somewhat involved in that as well back in the 90s when we were in our 20s. Thomas attributed the struggle of some contemporary cisgender gay boys to see themselves in the LGBTQ community to the lower visibility of cis- gender gay men in leadership positions. He especially noted the invisibility of adult men of color in these spaces, saying, “[Young people will say,] Hammack et al. 197 ‘I have to have someone that looks like me, someone who I see as a peer, as a role model, as a mentor, and they’re just not there.’ They’re just not there.” It is noteworthy that this pattern did not emerge in our ethnographic work in the Central Valley, where adult male leaders were present in community spaces, though less commonly in GSAs and in high schools in general. Thomas and other adult informants suggested that many cisgender gay boys congregated in spaces focused on sexual health and HIV prevention rather than general community spaces, but he and others expressed concerns about reliance on these types of spaces:... You have to have a specifically identified group where cisgender gay men can feel comfortable getting support. I don’t think they’re getting it. I don’t see that space for them, and I haven’t seen that space in a very long time outside of [HIV] prevention. That’s the only space I see it for them is prevention, which then, now you pathologize them, say you’re at risk, this perpetuation that you’re at risk for HIV. Thomas expressed concern that the primary spaces where young cisgender gay boys tend to gather are those focused on HIV prevention because these spaces often bring with them a discourse of “risk” and behavioral surveil- lance associated with sexual health. Thomas also noted the use of social media dating and hookup apps (intended for adult use) among gay boys with whom he works, expressing concern that these online spaces do not necessar- ily serve a community support function. There was evidence in our qualitative data that not just cisgender boys but also transgender boys struggled as they engaged with compulsory notions of masculinity. Rob (he/him/his), a 14-year-old Black transgender boy (AFAB) who identified as gay and was from a high-support community, narrated his initial reluctance to identify as a man due to perceived lack of conformity with masculinity: Rob: I wanna say three years ago I kind of started exploring what that would mean [to identify as a trans guy], but I thought it was nonbinary. Because I didn’t really know if I can be—call myself a trans guy and be not necessarily super masculine. I thought nonbinary made sense, but I know it definitely isn’t what I am, so. Interviewer: That’s really interesting, so what—you thought that if you were to identify as male you would have to embody certain ideas of masculinity? Like what? Rob: Mm-hmm. I don’t know, just like I guess also it would make sense if—I was born female of course, people don’t see me and read me as 198 Journal of Adolescent Research 37(2) male. I guess it’s like I should, I try to identify how people read me, but that doesn’t like necessarily mean anything, how people read you. In this excerpt, Rob revealed the way in which he initially was reluctant to identify as a boy because of how others might perceive his “credibility” as male, given some of the ways in which he wanted to present (e.g., his desire to wear makeup) his own sense that he is not “masculine enough.” Pattern 3: Adolescents Used Plurisexual and Asexual Identity Labels Frequently The third pattern of findings which emerged across multiple data sources concerned the relative frequency of both plurisexual and asexual identity labels among adolescents across communities (see Figure 4). Ethnographic field notes revealed observations of frequent use of identity labels such as bisexual, pansexual, and queer, though these observations were more com- mon in high-support communities. Reference to asexual identity labels was infrequent in ethnographic field notes across communities but notable in the survey data. Table 1 reports frequencies of sexual identity labels among the entire sam- ple. One quarter (n = 78, 24.8%) of the sample identified exclusively with a monosexual gay or lesbian identity label (i.e., they selected “gay” or “les- bian” with no other labels that might signify plurisexual attraction). Noteworthy was the preponderance of plurisexual identity labels among ado- lescents: 38.5% (n = 121) identified as bisexual, 22% (n = 69) as pansexual, and 12.4% (n = 39) as queer. When removing those who identified as queer but did not report plurisexual attraction, the total proportion of the sample reporting a plurisexual identity label to signify attraction to more than one gender was 60.8% (n = 191). Plurisexual identity labels were especially common among participants who were AFAB (78%), when compared with participants identifying with a monosexual label (42.9% AFAB), ϕ =.352, 95% CI = [.229,.474], p <.001. There were no differences in the use of plurisexual identity labels across community types in the survey data, although discussion of plurisexual iden- tities emerged more often in qualitative data in high-support communities. This pattern of findings suggested that there is perhaps greater open dialogue about plurisexual labels in high-support communities but that adolescents are still exposed to plurisexuality in low-support communities, typically through social media. Survey data revealed a notable proportion of adolescents using labels associated with asexuality (n = 31, 9.9%; see Table 1). Discussion of asexual Hammack et al. 199 identities emerged infrequently in the qualitative data. Loretta (she/her/hers), a 39-year-old Hispanic straight cisgender female (AFAB) adult informant from a low-support community, discussed the challenges she observed with one adolescent she worked with who identified as asexual: I had one girl who was asexual. She was just like, “I don’t like being touched. I don’t like being hugged. People think that there’s something wrong with me. I’m assuming that there is something wrong with me because that’s not normal.” I’m just like, “Your normal is normal. Everybody has a different normal.” She went through a long sad phase. Loretta’s encounter with this student revealed the way in which, although more youth may be identifying as asexual, they continue to engage with dominant narratives of sexuality and intimacy that denigrate asexuality. We found that adolescents AFAB constituted a high proportion of those who identified as asexual (67.7%), bisexual (70.2%), pansexual (87%), and queer (87.2%). The narrative of Gene (they/them/their), a 16-year-old White nonbinary (AFAB) youth informant from a high-support community, illustrated the vocabulary related to asexuality accessible to this cohort of adolescents: Interviewer: You indicated that you identify as demiromantic, panroman- tic and ace flux. Can you tell me a little bit more about coming to iden- tify with those labels? Gene: Yeah,... having two moms made the whole coming out as pan thing not a huge deal. I don’t know. With my friends and stuff, it’s more of an issue [of] defining it, cause a lotta people don’t—some people know of bisexuality, but pan is, “What does that even mean?” Just being like, “Gender doesn’t matter, basically.” Just defining that. Yeah. I’m pretty comfortable when it comes to sexuality and stuff, because I’ve grown up in such a queer community, that I’m very com- fortable being, like, “I’m gay.” Or just saying whatever. Because I also can use gay interchangeably for pan. I don’t know. It works for me. Yeah, that’s always been fairly simple. Coming out hasn’t been hard for that, cause I’m so casual about it. With being ace or demi, again, it’s more of an issue of defining it. Because a lot of people don’t know what it is. Then yeah, that’s mostly. Then again, just explaining. It’d be like, “I just don’t have crushes all that often. Or I have an emotional connection before I feel attracted to someone.” Yeah... Yeah, espe- cially with labels and stuff. I find them fascinating. The reason why I’m like, “Here’s 20 labels that all fit me,” is because I’ve spent so 200 Journal of Adolescent Research 37(2) much time researching the different terms and the different subcatego- ries... of all the different sexualities, cause there are so many now. I just find them super interesting... Raised by a lesbian couple in a community supportive of sexual and gender diversity, Gene felt emboldened to explore labels related to the direction of their desire (i.e., to multiple genders, “panromantic”), the conditions in which they experience desire (i.e., only in the context of a strong emotional connec- tion, “demiromantic”), and the extent to which they experience desire (i.e., less frequently but fluctuating, “ace flux”). Furthermore, Gene is explicit about their identity labels as signifying romantic, rather than sexual, iden- tity—signaling a distinction more frequently being made in people’s intimate lives (e.g., Fine, Torre, Frost, Cabana, & Avory, 2018; Mardell, 2016) but rarely acknowledged in research on sexual diversity (cf. Diamond, 2003, 2004). As indicated in Figure 4, qualitative data from informant interviews revealed the experience of challenging binary and normative thinking in sex- ual identity for those adolescents who identified with plurisexual identity labels. Tiana (she/her/hers), a 17-year-old African American cisgender pan- sexual female (AFAB) informant from a high-support community, narrated her experience of coming to identify as pansexual: I guess it’s a long story but when I was 12, I know I didn’t really—you’re not really ready for a relationship of any kind at a young age, so I just identified myself as asexual... I guess as I grew older I noticed that dating a man, or dating a woman, or dating a transgender—dating a transgender my first year of high school is the same for me, it’s just based on people that I meet and how I interact with you, [the] spiritual level I feel with you. Tiana described a process of sexual identity development that began with an identification as asexual in her early adolescence, revealing both the avail- ability of this vocabulary for adolescents but also that Tiana felt sufficiently self-aware of her sexuality to need to claim an identity at a young age. She ultimately came to select “pansexual” as her dating experiences revealed that she was attracted to individuals who may assume any variety of cisgender or transgender identities. Some youth narrated challenges related to the intelligibility of newer labels such as pansexual among their non-LGBTQ peers and community members. Sue (she/her/hers), a 17-year-old cisgender pansexual Asian girl (AFAB) from a high-support community, explained this experience: Hammack et al. 201 When it does come up, I usually am faced with just confusion about, “What does that word mean?” I’m not usually fond of explaining it, but I do. Usually, to avoid that, I end up having to say that I’m queer or say that I’m gay, even though that’s not the word that I usually most closely identify with. In terms of people accepting it, yeah, I don’t get a lot of... hate for it or anything. I get a lot of tolerance and confusion. Sue’s narrative revealed the exhaustion that many who hold diverse gender or sexual identities experience at having to regularly educate others about the meaning of their particular label. It is noteworthy that youth narrated the discovery of language to more authentically describe their sexual identities in online spaces. For example, Sue learned about the meaning of “pansexual” through her exploration of online spaces: I was about 12 years old, I think. It wasn’t like a person in particular. It was more a realization of gender doesn’t really define anything other than gender. That doesn’t define any of the things I care about, like personality... The way I found the label was the Internet, which is really helpful, actually. [Laughter] I spent a lot of time online, messed around with a couple of labels until I found one that I felt fit me cause gender wasn’t the thing that I cared about. Through her engagement in online communities, Sue came to fit her experi- ence of gender and sexual attraction with the label pansexual, revealing the way in which the labeling process occurs in a larger context of immersion in online communities for this generation of adolescents. Pattern 4: Use of “Queer” Revealed Complexity, Variability, and Intersectionality Across survey and qualitative interview data sources, the use of the label “queer” revealed complexity, variability, and intersectionality among adoles- cents (see Figure 5). As indicated in Table 1, among the 39 survey respon- dents who described their sexual identity as “queer,” 26 (66.7%) reported plurisexual attraction or also selected a plurisexual identity label such as “bisexual” or “pansexual.” The remaining third reported monosexual attrac- tion or also identified with a gay, lesbian, or asexual spectrum identity (and not a plurisexual identity label). This pattern suggests that some identify as queer to signify plurisexual attraction, while others may use queer as an umbrella community term or possibly an intellectual or political term. This interpretation was supported by our qualitative data and is discussed below. 202 Journal of Adolescent Research 37(2) Respondents who identified as queer in our survey (12.4% of total sample) were more likely to be AFAB (16.4% AFAB vs. 4.7% AMAB, ϕ =.169, 95% CI = [.068,.270], p <.01), more likely to reside in a high-support commu- nity (18.2% in a high-support community vs. 6.5% in a low-support commu- nity, ϕ =.179, 95% CI = [.064,.294], p <.01), and more likely to report higher parental education, rpb =.172, 95% CI = [.056,.281], p <.05. White respondents in high-support communities were more likely to identify as queer than non-White respondents (24.7% of White respondents in a high- support community vs. 9.1% of non-White respondents), whereas non-White respondents in low-support communities were more likely to identify as queer than White respondents (8.8% of non-White respondents vs. 1.9% of White respondents in a low-support community, ϕ =.217, 95% CI = [.084,.351], p <.01). The narrative of Ethan (she/her/hers), a 19-year-old White nonbinary (AMAB) queer youth informant in a high-support community, illustrated the way in which some adolescents come to appropriat