Put at and kept from risk: The sexual risk dilemmas confronting marginalized youth

Abstract

Introduction

The current study examines Bay‐Cheng and Ginn's (2024) claim that marginalized youth confront two intertwined sexual risk dilemmas: being put at risk by unjust social forces and structures that threaten their sexual health and safety; and being kept from risk by those same forces and structures, thus compromising their development and dignity.

Methods

We explored the presence and relevance of these dilemmas using interviews conducted in 2022 with 101 racialized and/or queer (i.e., a sexuality other than exclusively heterosexual) young women and trans or nonbinary youth aged 16–21 and living in Melbourne (Australia), New York City (United States), and Toronto (Canada). We conducted a directed content analysis of the data and examined intersectional group‐based comparisons of code frequencies.

Results

We found that 80% of participants felt endangered (i.e., put at risk) and/or obstructed (i.e., kept from risk) by social and material conditions. Over half felt put at risk through exposure to hostility and harm, and over one‐third of participants described precarious life circumstances as keeping them from taking wanted sexual risks. Group‐based and intersectional comparisons indicated that feeling put at risk is associated with marginalized gender and sexuality identities and corresponding discrimination. We also found tentative signs that feeling kept from risk may be more closely linked to race and racism than marginalization by gender and/or sexuality.

Conclusions

The findings highlight the diverse implications of interlocked social injustices for youths' sexual lives, particularly the infringement on their rights both to safety and to risk.

1. INTRODUCTION

1.1.

For several decades, scholars have critiqued the deficit‐focused preoccupation with youth sexuality for its simplistic, essentializing equation of young people's sexual behavior with detrimental outcomes (e.g., Bay‐Cheng, ²⁰⁰³; Fine, ¹⁹⁸⁸; Fortenberry, ²⁰¹⁴; Lesko, ²⁰¹²). Especially for youth marginalized by interlocked forms of structural oppression (e.g., misogyny, racism, economic injustice, cis‐, and heterosexism), sexual expression and experimentation—particularly partnered sexual behaviors—are presumed to be worrisome, hazardous behaviors that threaten young people's current well‐being and future life prospects. This essentialization of marginalized youths' sexual behavior as inherently problematic deflects attention from the inequitable material conditions and social forces that shape young people's sexual lives (Bay‐Cheng, 2019; Crooks et al., ²⁰²³; Higgins et al., ²⁰²²; Schalet et al., ²⁰¹⁴). It also leaves little room for the possibility that youth sexual behavior, including risk‐taking and experimentation, might not be—or might not only be—a danger, but also a source of pleasure, intimacy, and growth (Fine, 1988; Fortenberry, ²⁰¹⁴; Tolman, ²⁰⁰²).

Scholars have rebutted the simplistic and singular deficit‐focused treatment of youth sexuality by endorsing young people's rights to risk‐taking and pointing out the social injustices at the root of marginalized youths' disproportionate vulnerability to adverse sexual outcomes. Drawing on scholarship identifying sexual expression and self‐determination as human rights (Lottes, 2013; Miller et al., ²⁰¹⁵; World Association of Sexual Health, ²⁰¹⁴), developmental research citing the normativity of sexual experimentation during adolescence (Fortenberry, 2014; Hensel et al., ²⁰¹¹; Rostosky et al., ²⁰⁰⁸; Zimmer‐Gembeck et al., ²⁰¹¹), and/or empirical evidence of the social determinants of sexual health and outcomes (Higgins et al., 2022; Schalet et al., ²⁰¹⁴), they argue for a more discerning and contextualized view of youth sexual risk‐taking. Such an approach would affirm sexual “chance‐taking” (to use Lindroth and Löfgren‐Mårtenson's [2013] positive reframe) as integral to youths' rights and development and trace the dangers associated with it to the circumstances in which sexual experiences occur. Through a critical synthesis of empirical and theoretical literature regarding the interplay among sexual rights, risk, and interlocked social injustices, Bay‐Cheng and Ginn (2024) posited that marginalized women, specifically young women and those labeled with intellectual disability, contend with two distinct yet co‐occurring sexual risk dilemmas: first, of being put at sexual risk (i.e., potentially damaging ones) by unjust social conditions that leave them disproportionately and unnecessarily vulnerable to harm; and second, of being kept from sexual risk (i.e., potentially enriching ones) by unjust social conditions that deny them opportunities for experimentation and daring.

We extend this work by testing the applicability of these two proposed sexual risk dilemmas to youth marginalized by gender (i.e., those identifying as women or trans/nonbinary [TNB]), race (i.e., those with racialized ethnoracial identities), and/or sexuality (i.e., those identifying as something other than heterosexual). Following a summary and further specification of Bay‐Cheng and Ginn's (2024) argument, we share findings from a directed content analysis (Hsieh & Shannon, 2005) of 101 interviews examining if, how, and which participants perceived sexual risk as a danger to which they were subjected (i.e., put at risk) as a function of social injustices (e.g., misogyny, racism, economic injustice, cis‐, and heterosexism, adultism) and/or as a positive, growth‐promoting opportunity those same injustices prevented them from pursuing (i.e., kept from risk).

2. MARGINALIZED YOUTH PUT AT RISK

In their chapter, Bay‐Cheng and Ginn (2024) cited empirical evidence of the role played by misogyny, racism, economic injustice, adultism, and/or ableism both in dramatically increasing the odds of experiencing sexual harm (e.g., STIs, unwanted pregnancy, and diverse forms of violation) and in obfuscating the systemic bases of this disproportionate sexual vulnerability. The sexual risk dilemma confronting young people is not that they are inherently “at risk,” but more precisely that they are put at risk by interlocked social injustices (Bay‐Cheng & Fava, 2014; Foster & Spencer, ²⁰¹¹; Gilbert, ²⁰¹⁴; Kelly, ²⁰⁰⁰). Bay‐Cheng and Ginn organized the ways in which these social injustices endanger sexual rights and well‐being into three categories. First, marginalized people are put at risk when they are disproportionately exposed to avoidable dangers. To cite one example from the United States, the racialized socioeconomic segregation of Black youth in communities with high prevalence of STI means that even if they engage in identical sexual behaviors and decisions as White peers—or if they are safer (i.e., fewer partners, older age of initiating partnered sexual activity, more consistent condom use)—they remain at heightened risk of contracting an STI (Corcoran et al., 2021). Such statistics make clear that sexual outcomes are products not simply of an individual's own sexual behavior and decisions, but of surrounding circumstances and forces beyond their control (Bay‐Cheng, 2019).

Marginalized young people are also put at risk when they are deprived of resources that could buffer them from the risks to which they are unjustly exposed or mitigate the harm of those risks. This includes sexual health care that is inaccessible as a function of cost, transportation, schedules (e.g., incompatibility with school hours), capacity (e.g., limited staff and hours), and/or policy (e.g., parental consent or notification laws; Bay‐Cheng et al., ²⁰²²; Johnston et al., ²⁰¹⁵; Salehi et al., ²⁰¹⁴). Returning to the example of Black youth in the United States effectively segregated in communities with high concentrations of STIs, they are endangered not only through exposure to STI, but also the simultaneous denial of adequate and equal sexual health education, prevention, and treatment.

Last, young people are put at risk when they are compelled by necessity to engage in sexual behaviors and relationships. Consent to unwanted sex, for example, may be provoked by social conditions that collapse young people's volition and options such that engaging in unwanted sexual behaviors becomes an essential strategy for accessing material, social, or psychological resources (Bay‐Cheng, 2019; Joly & Connolly, ²⁰¹⁹; Watson, ²⁰¹¹). Whether exposed, deprived, or compelled, marginalized young people's subjection to sexual risk is not an inevitable consequence of individual behaviors, but instead a by‐product of social injustices.

3. MARGINALIZED YOUTH KEPT FROM RISK

Bay‐Cheng and Ginn (2024) argued that just as being put at sexual risk may be an artifact of social injustice, so may be being kept from sexual risk. Although sexual risk‐taking is normally treated as a danger to be reduced if not eliminated, this ignores other evidence and arguments in support of sexual experimentation. This is particularly salient in developmental literature, including studies demonstrating the benefits of sexual exploration during adolescence, such as greater sexual self‐awareness and subjectivity and better relational and communication skills (Hensel et al., 2011; Rostosky et al., ²⁰⁰⁸; Zimmer‐Gembeck et al., ²⁰¹¹). From this perspective, “sexual learning” (Fortenberry, 2014) is founded on sexual risk‐taking and can be understood as a developmental right (see also Gilbert, ²⁰¹⁴). Disabilities studies literature offers an alternate lens for viewing risk‐taking. Complementing the developmental argument, in which the right to risk is predicated on its potential dividends, Perske (1972) advocated for the “dignity of risk” regardless of its outcomes: that the ability to exercise autonomy in one's life, including the right to take chances, is fundamental to human dignity and should be universally upheld.

Whether framed in terms of development or dignity, marginalized youths' rights to risk are routinely thwarted. Similar to those labeled with disabilities, they may be kept from risk‐taking because they are overprotected. Protectionism supposes that those with limited (or still developing) capacities for thoroughly reasoned decision‐making must be prevented by authorities (e.g., parents, teachers, policymakers) from sexual activity lest they put themselves in harm's way (Jackson & Scott, 1999; Robinson, ²⁰¹²). Other youth may be prohibited from desired and potentially beneficial sexual experiences not to protect them, but to protect others from them, an urge often motivated by racist, classist, and/or cis‐ and heteronormative presumptions of the reckless behavior and corrupting influence of marginalized young people (Dwyer, 2014; Fields, ²⁰⁰⁵; Kelly, ²⁰⁰⁰).

Contrary to protectionism, by which specific authority figures and explicit policies block youth from sexual chance‐taking and learning opportunities, many marginalized youth are hampered in their sexual lives by the precarity of their general and surrounding life circumstances. Countering dominant narratives that tap developmentalist, racist, and/or classist discourses to cast marginalized youth as reckless (Lesko, 2012), there is evidence that youth making their lives on social and material margins may be more conservative than their more privileged peers. For example, socioeconomically disadvantaged young women may eschew sexual and romantic relationships, wary of any distraction from or potential derailment of their academic and professional efforts (e.g., Armstrong & Hamilton, ²⁰¹³). Operating in tandem with or separately from such material concerns may be worries about inviting or confirming derogatory stereotypes. Indeed, marginalized youth may forego sexual exploration and self‐expression in an effort to pre‐empt or rebut sexual stigma born of racism, classism, misogyny, cis‐ and heterosexism, and adultism (Crooks et al., 2023; García, ²⁰¹²; Leath & Mims, ²⁰²³).

4. STUDY OVERVIEW

The opportunity to test Bay‐Cheng and Ginn's (2024) proposed sexual risk dilemmas of being put at and kept from sexual risk‐taking emerged from a trinational, mixed method research project. The project focused on the forms, functions, and meanings of risk during the COVID‐19 pandemic among 16–21 year old women or TNB youth in Melbourne, New York City, and Toronto with marginalized ethnoracial and/or sexual identities. Although the study was not designed to test the proposed sexual risk dilemmas, its general focus on risk and on the sexual lives and perspectives of marginalized youth nevertheless provided a relevant and ample data set for doing so. Specifically, one stage of the project included in‐depth interviews with 101 marginalized young people recounting and reflecting on diverse risk decisions. In the current study, we deductively coded the transcripts for comments referring to feeling put at risk (i.e., exposed, deprived, or compelled) and kept from risk (i.e., being overprotected, precarious, or stereotyped) as a function of social injustice. Table 1 details these themes and constituent codes. We also made use of the sample's size and diversity to compare participants' references to these sexual risk dilemmas at various intersections of race/ethnicity, gender, and sexuality.

Table 1.

Sexual dilemma code categories, definitions, and examples.

Code categories	Definitions	Examples
Put at sexual risk by being
Exposed to risks	Subject to disproportionate and avoidable sexual harm by converging social injustices and societal disregard	I had thoughts about if I should be doing this because, as a woman, as like a Muslim Brown girl, it's not like the safest. Well for women in general to kind of be on the roads late at night when it's dark outside.
Deprived of resources	Denied material and social resources that might lessen either the chances or consequences of sexual harm	There was no support in my first high school, there was not there was like literally zero services for like LGBTQI+ students struggling or whatnot, even externally, I didn't have anything
Compelled to take risks	Driven to potentially harmful sexual behavior by external factors and conditions	I feel like people especially like young woman, like queer women will like take risks that they feel they're forced to, or they feel will benefit them sometimes, even if the consequences aren't always great
Kept from sexual risk by being
Precarious	Prevented from exercising sexual rights and autonomy by social or material insecurity	I'm like, a woman of color, lower income, like, and now, you know, I'm bisexual. I feel like a lot of society is kind of like, against people like me. So it makes me a little more wary of making risks.
Stereotyped	Prevented from exercising sexual rights and autonomy by stereotypes and stigma	I don't want to act in a way that would give someone ammunition to be like, “See, I told you that how these people are”
Overprotected	Prevented from exercising sexual rights and autonomy by intervening authorities on the basis of safety	I understand that my mom is just looking out for the best of me and it's not her like prejudiced views or whatever, even if she might have them, it's her concern that society will treat me differently and I won't be able to be as happy as I could be by just giving up on dating girls

5. METHOD

The current study uses data collected during a large multistage, multisite research project conducted between February 2022 and July 2022. The project focused on the risk and risk‐taking experiences of young people living in Melbourne, New York City, and Toronto who were between the ages of 16 and 21, identified as girls/women or TNB, and who were racialized (i.e., having an ethnoracial identity other than exclusively White) and/or queer (i.e., having a sexuality other than exclusively heterosexual; we use “queer” to encompass the wide range and dynamic nature of participants' sexual identifiers as well as those participants who eschewed the use of any sexual identifiers [Ruberg & Ruelos, ²⁰²⁰]). The research project was first conceived by an interdisciplinary and international team of six senior researchers, all with substantial expertise in critical youth and sexuality studies. To add generational diversity and insight to the project, the senior researchers hired graduate and undergraduate students early in the project's development so that they were active collaborators in research design, data collection, data analysis, and reporting. The research team also solicited feedback regarding data collection approaches and measures on an ad hoc basis from adolescent youth they knew through various community partnerships. The project collected data in three stages: 703 participants completed a brief online survey; 373 of them proceeded to create “timelines” of key life experiences using a website designed for the study; last, 101 participants also agreed to virtual individual interviews. Other than demographic data reported on the survey, the current study draws only on the interview transcripts. Additional details about the project (e.g., data collection, sample, research team) are available from the first author.

5.1. Participants

The interview sample consisted of 101 participants with an average age of 18.76 (SD = 1.69): 29 (29%) living in Melbourne, 38 (38%) in New York City, and 34 (34%) in Toronto. Demographic details of the sample divided by city and including some intersections among gender, sexuality, and ethnoracial identities are presented in Table 2. Reflecting the study's inclusion criteria and focus on the experiences of youth with marginalized gender, sexuality, and/or ethnoracial identities, majorities of participants identified as: women (65; 65%), queer (91; 90%), and/or racialized (75; 74%). On average, participants across the sites rated their socioeconomic status while growing up as “not rich but could afford some luxuries” (see Section 4.2 for further details). The Melbourne sample was demographically distinct from New York City and Toronto in three ways (see Table 2 for details): it had a younger mean age than New York City (F (2, 98) = 4.14, p = .019); a larger proportion was White (χ ² (2, N = 101) = 16.67; p < .001; and no participants identified as heterosexual (χ ² (2, N = 101) = 9.21, p = .01).

Table 2.

Participant demographics by site (N = 101).

	Melbourne (n = 29)	New York City (n = 38)		Toronto (n = 34)	Total (N = 101)
	M (SD)	M (SD)		M (SD)	M (SD)
Age (range: 16–21)	18.07 (1.53)	19.24 (1.79)		18.59 (1.62)	18.76 (1.69)
Subjective SES (range: 1–6a)	3.64 (0.73)	3.38 (1.11)		3.79 (0.93)	3.59 (0.96)
	N (%)	N (%)		N (%)	N (%)
Education
High school student	12 (41.4)	12 (31.6)		10 (29.4)	34 (33.7)
Postsecondary student	16 (55.2)	23 (60.5)		23 (67.6)	62 (61.4)
Nonstudent	1 (3.4)	3 (7.9)		1 (2.9)	5 (5.0)
Race/ethnicity
Black	0	14 (36.8)		6 (17.7)	20 (19.8)
East/Southeast Asian	9 (31.0)	11 (28.9)		7 (20.6)	27 (26.7)
Latinx	0	5 (13.2)		0	5 (5.0)
Multiracial/“Brown”/“Person of Color”	1 (3.5)	3 (7.9)		3 (8.8)	7 (6.9)
South/Southwest Asian	4 (13.8)	2 (5.3)		10 (29.4)	16 (15.8)
White	15 (51.7)	3 (7.9)		8 (23.5)	26 (25.7)
Gender
Cisgender	16 (55.2)	26 (68.4)		23 (67.7)	65 (64.4)
Transgender	1 (3.4)	0		1 (2.9)	2 (2.0)
Nonbinary	12 (41.4)	12 (31.6)		10 (29.4)	34 (33.7)
Sexuality
Heterosexual	0 (0)	5 (13.2)		8 (23.5)	13 (12.9)
Lesbian	6 (20.7)	9 (23.7)		10 (29.4)	25 (24.8)
Plurisexual	15 (51.7)	20 (52.6)		12 (35.3)	47 (46.5)
“Queer”	5 (17.2)	4 (10.5)		3 (8.8)	12 (11.9)
Aromantic/asexual	7 (24.1)	3 (7.9)		3 (8.8)	13 (12.9)
Questioning/unspecified	2 (6.9)	0 (0)		2 (5.9)	4 (4.0)
Race*gender*sexuality
Racialized, cisgender, heterosexual	0	5 (13.2)		9 (26.5)	14 (13.9)
Racialized, cisgender, queer	10 (34.5)	20 (52.6)		10 (29.4)	40 (39.6)
Racialized, TNB, queer	4 (13.8)	10 (26.3)		7 (20.6)	21 (20.8)
White, cisgender, queer	6 (20.7)	1 (2.6)		4 (11.8)	11 (10.9)
White, TNB, queer	9 (31.0)	2 (5.3)		4 (11.8)	15 (14.9)

Note: Frequencies presented are percentages within each column. Frequencies for sexuality exceed sample size because some participants used multiple labels (see Table 3 for examples). Asterisks are used to signify intersectionality.

Abbreviations: queer, any sexual identity other than exclusively heterosexual; SES, socioeconomic status; Unless, used as an explicit label (indicated by quotation marks); TNB, transgender or nonbinary.

^a 1 = didn't have enough to get by; 2 = barely enough to get by; 3 = enough to get by but couldn't afford extras; 4 = not rich, but could afford some luxuries; 5 = financially well‐off; 6 = wealthy.

5.2. Procedures and measures

Study procedures and measures were approved by ethics review boards at five universities (i.e., principal investigators' home institutions) in Australia, Canada, and the United States. Melbourne, New York City, and Toronto were selected as recruitment sites since they were the three most populous and diverse cities in each nation. Recruitment was initiated in February 2022 and concluded in July 2022 when target samples sizes were reached. The research team used convenience sampling through social media, emails through their professional networks, and physical flyers posted in youth‐ and student‐centered locations in each city. We also used snowball sampling, asking participants to circulate study information and social media posts. Recruitment notices directed prospective participants to an online screening survey.

Eligible and consenting participants completed an online survey that took an average of 12 min to complete and collected the demographic data used in the current study, including open‐ended responses to the questions, “How do you identify in terms of [race, gender, sexuality]?” Participants were also asked their age in years, whether they were students and if so, in what type of school (e.g., secondary, vocational, university). Following Bay‐Cheng (2017), we assessed subjective socioeconomic status using the question, “How would you describe your financial situation during most of your life before COVID?” and six ordinal response options: 1 = Didn't have enough to get by; 2 = Barely enough to get by; 3 = Enough to get by but couldn't afford extras; 4 = Not rich, but could afford some luxuries; 5 = Financially well‐off; 6 = Wealthy.

After completing the survey, participants were invited to create timelines of various life experiences over the past 3 years using a website designed for the study and modeled on Bay‐Cheng's (2017) sexual life history calendar. The website used a series of close‐ and open‐ended questions to prompt participants to share any memorable landmark life events, associated emotions, and long‐term impact or meaningfulness. When participants indicated they did not wish to add any more events, they received a gift card worth $50 AUD, $40 USD, or $50 CAD.

Immediately upon completing a timeline, participants were invited to participate in a third stage of data collection consisting of individual virtual interviews. Those who consented were contacted within 2 weeks of completing their timelines to arrange an interview. These were scheduled at the convenience of the participant and interviewers were assigned based on availability. The vast majority of the interviews (93; 92%) were conducted by doctoral and undergraduate student members of the research team; the remaining eight were conducted by the project's principal investigators. All interviewers were trained in affirming and trauma‐informed interview methods. Resembling the participant sample, all interviewers identified as women or TNB and most identified as queer and/or racialized. Before an interview, the assigned interviewer reviewed the participant's timeline to identify experiences they wished to learn more about and to prepare for a participant's possible need for further support (e.g., if a participant disclosed prior trauma and/or mental health concerns).

The semistructured interview protocol was designed by one of the project's senior researchers who had particular expertise in qualitative methodologies. It was then extensively vetted and revised by the research team, including undergraduate members, to ensure its relevance and intelligibility. The protocol consisted of 12 core questions, including asking participants: to elaborate on various events on their timelines in which they took risks; to describe their attitude or stance toward risk‐taking in general; what factors support or stop their risk‐taking; and whether their identities and circumstances affect their risk‐taking. During the interview, participants were invited to open their timelines or to view images of their timeline screenshared by the interviewer. Participants were also invited to respond verbally or to use the zoom chat function if preferred (e.g., if they worried about being overheard). Interviews averaged 62 min (range: 27–109) in length. Participants received a second gift card worth $45 AUD, $40 USD, or $50 CAD following the interview. Interviews were audio recorded and first transcribed using Otter.ai. Research team members then corrected and deidentified the final transcripts used in analysis.

5.3. Data analysis

We conducted a directed content analysis (Hsieh & Shannon, 2005), deductively coding the transcripts using Bay‐Cheng and Ginn's (2024) proposed sexual risk dilemmas of marginalized youth being subjected to sexual danger (i.e., put at risk) and/or being prevented from sexual experimentation and expression (i.e., kept from risk). Codes were not mutually exclusive, allowing for excerpts to have multiple codes assigned to them. We coded all comments related to a participant's experiences taking or refraining from risks related to sexual behavior, relationships, and/or identities. This included both explicit reflections focused specifically on a sex‐related risk experience as well as more general comments that included sex and sexuality. During coding, we made one exception to the condition that we would code only comments related to a participant's personal experiences. Although no participants disclosed having taken a sexual risk out of necessity (e.g., trading sex for resources), five referred to such circumstances in generalized terms and/or knowing people with such experiences. Given that participants might have been referring to their own experience but framing it as others' to avoid stigma or possible repercussions (e.g., that the interviewer might report an incident to authorities) and the meaningfulness of their comments, we opted to code such content while noting that it pertained to others' behavior.

The trustworthiness of such specific coding decisions and the analytic process overall was strengthened by the first two authors' respective standpoints and complementary insights. Bay‐Cheng offered professional experience as a senior researcher in the field of youth sexuality and as a former practitioner who worked with young women with histories of sexual assault. She had also collaborated on the chapter proposing the sexual risk dilemmas used to frame the current analyses. However, she had not conducted any of the interviews and as a professional in midlife, stood at a distance from the participants. As an undergraduate student at the time of the study, Sutton did not have the same professional experience as the first author but had been an integral member of the research team, including conducting nine interviews, and was a generational peer to the participants. Both authors also shared common ground with the participants in their identification as queer and/or racialized and as women.

The coding process was guided by best practices in qualitative content analysis, (e.g., Elo et al., ²⁰¹⁴; Hruschka et al., ²⁰⁰⁴; Hsieh & Shannon, ²⁰⁰⁵; Neuendorf, ²⁰¹⁸). The first two authors had become immersed in the data over the course of transcribing and “tagging” the interviews for topics of interest to the research team (e.g., stories of coming out, COVID‐related worries). It was during this open‐ended exploration of the interview transcripts that the first author recognized the potential for a data‐driven test of the claims they made in an earlier work. The multistage analytic process for the current study began with the first author drafting a codebook with two themes based on the proposal that marginalized youth confront two primary sexual risk dilemmas: (1) being put at risk, with the specific codes of being exposed to harm, deprived of resources, and compelled to take risks; and (2) being kept from risk, with the specific codes of being precarious, likely to be stereotyped, and being overprotected. The second author acted as the primary coder. She piloted the first draft of the codebook by coding 17 randomly selected transcripts, tracking questions and feedback along the way. Following a discussion, the first author used these comments to clarify code definitions and coding conventions (i.e., when and what to code, as described in the previous paragraph). The second author then used Dedoose to code all the transcripts according to the revised codebook, meeting with the first author twice during that period to ask clarifying questions. The first author also coded the units identified by the second author and compared their respective codes, noting instances of disagreement. Their intercoder reliability at that point was acceptable, with Krippendorff's alphas (calculated using Marzi et al.'s [2024] K‐Alpha Calculator) of .94 for the put at risk codes and .81 for the kept from risk codes. The authors collaboratively reviewed all remaining coding discrepancies to reach consensus on all assigned codes. Table 1 provides final code definitions, conventions, and examples. Throughout the Section 5, we include verbatim quotes from participants to illustrate and contextualize the codes. Table 3 provides the self‐reported demographic characteristics of quoted participants.

Table 3.

Quoted participants' demographics and identities.

City	Pseudonym	Age	Subjective SES	Race/ethnicity (verbatim)	Gender (verbatim)	Sexuality (verbatim)
Melbourne	Amal	18	(Skipped)	Middle Eastern/SWANA	Agender/femme	Bisexual/queer
	Ash	16	Enough but no extras	Indian	Bigender (nonbinary and a woman)	Aroace and queer
	Avery	17	Enough but no extras	White	Undecided	Bisexual
	Cindy	18	Not rich, some luxuries	Southeast Asian	Cis woman	Lesbian
	Isha	16	Not rich, some luxuries	Indian	Female	Bisexual ace
	Sally	17	Not rich, some luxuries	Asian	Female	Lesbian/bisexual, fem leaning
	Serena	20	Enough but no extras	Asian–Australian	Female	Bisexual
	Tai	17	Enough but no extras	Chinese	NonBinary	Bisexual/pansexual
New York City	Brooklyn	21	Barely enough	Black	Nonbinary	Queer asexual
	Camila	21	Enough but no extras	Hispanic/Latina	Cis female	Bisexual
	Fatima	19	Barely enough	Black	Woman	Lesbian
	Kayla	21	Not rich, some luxuries	Black, African	Female	Heterosexual
	Jiaxin	21	Not rich, some luxuries	Asian	Female	Straight
	Jess	20	Not rich, some luxuries	White	Cis woman	Lesbian
	June	17	Not rich, some luxuries	Asian American	Female	Bisexual
	Justice	17	Not rich, some luxuries	Black	Female	Asexual
	Melanie	17	Enough but no extras	Asian	female	Lesbian/asexual
	Renee	20	Not rich, some luxuries	Black	Female	Queer
	Sammy	16	Financially well‐off	White	Nonbinary	Bisexual
	Ziggy	16	Not rich, some luxuries	Asian/Hispanic	Gender fluid/gender queer/nonbinary	Bisexual/pansexual
Toronto	Aubrey	17	Not rich, some luxuries	White	Woman	Lesbian
	CJ	19	Not rich, some luxuries	Mixed Black and White	Nonbinary	Bisexual and greysexual
	Cora	20	Enough but no extras	Chinese	Cisgender	Bi questioning
	Dev	18	Financially well‐off	South Asian	Cis, questioning gender queer	Bisexual
	Jules	18	Enough but no extras	East Asian	Gender fluid	Lesbian
	Nat	20	Not rich, some luxuries	Mixed Black and South Asian	Nonbinary but many of my experiences have been influenced by being raised as a girl	Bisexual
	Shaima	18	Enough but no extras	Pakistani	Woman	Lesbian, asexual

Although quantification is not a priority or objective of qualitative analysis, quantitative analysis can be a means to assess trustworthiness (Elo et al., 2014) and identify patterns among findings. Therefore, once coding was finalized, the first author used SPSS 29 to compare code frequencies between participant groups (see Table 4), applying the Fisher‐Freeman‐Halton Exact test (FFH) given the small number of expected observations per cell in some cases. We used pairwise z tests to identify which cell proportions differed significantly at p ≤ .05.

Table 4.

Code frequencies in the overall sample (N = 101) and by subgroups.

	Put at risk				Kept from risk
	Any	Exposed	Deprived	Compelled	Any	Precarious	Stereotyped	Overprotected
	N (%)	N (%)	N (%)	N (%)	N (%)	N (%)	N (%)	N (%)
Overall	62 (61.4)	56 (55.4)	26 (25.7)	5 (5.0)	61 (60.4)	37 (36.6)	18 (17.8)	14 (13.9)
By site
Melbourne	21a (72.4)	20 (69.0)	7 (24.1)	0	18 (62.1)	11 (37.9)	4 (13.8)	3 (10.3)
New York City	26a (68.4)	21 (55.3)	13 (34.2)	4 (10.5)	25 (65.8)	16 (42.1)	9 (23.7)	8 (21.1)
Toronto	15b (44.1)	15 (44.1)	6 (17.6)	1 (2.9)	18 (52.9)	10 (29.4)	5 (14.7)	3 (8.8)
By race/ethnicity
Black	11 (55.0)	8 (40.0)	6 (30.0)	2 (10.0)	15 (75.0)	11 (55.0)	7 (35.0)	3a (15.0)
East/Southeast Asian	20 (74.1)	19 (70.4)	5 (18.5)	2 (7.4)	14 (51.9)	7 (25.9)	4 (14.8)	5a (18.5)
Latinx	5 (100)	4 (80.0)	3 (60.0)	1 (20.0)	2 (40.0)	2 (40.0)	1 (20.0)	0b
Multiracial/unspecified	3 (42.9)	3 (42.9)	2 (28.6)	0	5 (71.4)	3 (42.9)	1 (14.3)	3a (42.9)
South/Southwest Asian	6 (37.5)	6 (37.5)	2 (12.5)	0	12 (75.0)	7 (43.8)	2 (12.5)	3a (18.8)
White	17 (65.4)	16 (61.5)	8 (30.8)	0	13 (50.0)	7 (26.9)	3 (11.5)	0b
By gender*sexuality
Cisgender, heterosexual	4a (28.6)	3a (21.4)	1a (7.1)	1 (7.1)	9 (64.3)	4 (28.6)	3 (21.4)	5a (35.7)
Cisgender, queer	34b (66.7)	31b (60.8)	11b (21.6)	3 (5.9)	31 (60.8)	21 (41.2)	8 (15.7)	6b (11.8)
TNB, queer	24b (66.7)	22b (61.1)	14b (38.9)	1 (2.8)	21 (58.3)	12 (33.3)	7 (19.4)	3b (8.3)
By race*gender*sexuality
Racialized, cishet	4a (28.6)	3a (21.4)	1 (7.1)	1 (7.1)	9 (64.3)	4 (28.6)	3 (21.4)	5a (35.7)
Racialized, TNB/queer	41b (67.2)	37b (60.7)	17 (27.9)	4 (6.6)	39 (63.9)	26 (42.6)	12 (19.7)	9a (14.8)
White, TNB/queer	17b (65.4)	16b (61.5)	8 (30.8)	0	13 (50.0)	7 (26.9)	3 (11.5)	0b

Note: Column frequencies reflect within‐group proportions. Within columns and comparison groups, proportions with the same subscript do not differ significantly at the .05 level using the Fisher‐Freeman‐Halton Exact test with the Bonferroni correction. Asterisks are used to signify intersectionality.

Abbreviations: Cishet, cisgender and heterosexual; Queer, any sexual identity other than exclusively heterosexual; TNB, transgender or nonbinary.

6. RESULTS

We identified 292 excerpts among 83 participants (82.2% of the sample) in which sexual risk dilemma themes appeared: 40 participants referred both to being put at risk and kept from risk; 22 participants made comments related only to being put at risk; and 21 referred only to being kept from risk. As a preliminary test of the codes' relevance and applicability to marginalized youth across cities, we compared the 18 participants who made no references being put at risk or kept from risk and the 83 participants who referred to one or both themes. We found no significant mean (i.e., age, subjective SES) or proportional (i.e., student status, being racialized, being queer) differences between them. In the next sections, we detail findings related to each theme and its constituent codes, including through illustrative quotes from participants (see Table 3 for participants' characteristics). We then compare code frequencies using the full sample of 101 participants (presented in Table 4) to understand the codes' relative salience between groups and their intersections. Sample sizes could not support meaningful comparisons among identity positions within ethnoracial, gender, and sexuality groups. Therefore, we examined the three possible permutations of dichotomized ethnoracial (i.e., racialized or not) and gender × sexuality (i.e., TNB/queer or not) variables.

6.1. Theme 1: Put at risk

6.1.1. Exposed to harm

Feeling exposed to various dangers appeared to be the dominant mode by which participants felt put at risk, with 55% of interviewees expressing fairly substantial and pervasive worries about sexual and gender‐related violence. Participants' apprehension was based on their own experiences, of those close to them, and/or media reports of sexual harassment, violence, and hate‐motivated attacks:

[I]t's just like, the collective fear that women have going out by themselves at night, knowing that we could be harassed or assaulted just because we're women. Just, it's like that fear has festered, like you don't even need to know stories, you just know, by the way that men look at you or how they talk about you on the internet. So, it's just really gross and scary. [Isha]

I don't have the worry someone's gonna hate‐crime me for being White, but I do have the worry someone's gonna hate‐crime me for being with my girlfriend. [Aubrey]

In addition to the heightened anti‐Asian racism during COVID mentioned by 10 Asian participants (four in Melbourne, three in New York City, three in Toronto), five referred to the fetishization of Asian women, including by White men with “yellow fever” (Cora), stereotypes about being “submissive” (Cindy), and hearing comments about “Chinese women and like, their vaginas” (Sally), referring to tropes about the shape of Asian women's genitalia.

6.1.2. Deprived of resources

Just over one‐quarter of participants (26; 25.7%) referred to lacking resources needed to avoid or mitigate risks. Seven participants, all from New York City, pointed to formal policies and legislation as the cause. Most understood reproductive rights restrictions and antitrans legislation as happening in other states or at the federal level rather than locally, but they were nevertheless concerned about what might one day happen in New York City. One commented at length on how age‐based policies had prevented her from accessing therapy earlier in her life because she did not have parental consent and were the reason why she had sold sexual photos of herself online since she could not find a good‐paying job as a minor. Beyond policy‐based concerns and obstructions, six participants described encounters with cis‐ and heterocentric services and providers that were unhelpful if not hostile, while three felt unprotected by police and service providers when they sought support. Fourteen participants described lacking practical access to high‐quality social services, including mental and sexual health care, even if formally entitled to these. Sammy's attempt to get confidential STI testing illustrates many of the barriers and breakdowns in services for young people:

I know that a lot of places usually have like, free STD testing and stuff, things like Planned Parenthood and other like, sexual health places and doctor places, I guess. So I was like, “Okay, this shouldn't be that difficult. You know, I live in New York City, it's like an urban place, there's plenty of those things.” But actually, like, almost all of them were closed at the time that I was looking for them because of COVID. […] But after like some more research, I found some place that's like, it was maybe like, half an hour on the train from where I lived. But it was in kind of a really sketchy neighborhood. Honestly, that was kind of dangerous.

To allay their concerns, Sammy asked their brother to accompany them to the clinic. However, they never received the results, despite calling to request them. Sammy was happy to be tested some time later at a check‐up with their family doctor, but was dismayed when the results (which were negative) were sent to their mother, violating their privacy rights under New York State law.

6.1.3. Compelled to risk

As noted in the explanation of our coding process and decisions, no participants directly disclosed that they had made sexual risk decisions to obtain needed resources. Three made somewhat oblique comments which, given other things they shared about their lives (e.g., material insecurity, familial rejection), may have been self‐referential, such as Camila's comment: “I know that, I know of other like, sex workers, just like other students who are doing it [sex work], you know, to pay their tuition and stuff. […] I feel like, so many people have to do that work to survive, and I feel like, it can really mess with your head.” Two others referred to compelled sexual risk‐taking while distancing themselves from such situations:

It's hard to speak about something I personally haven't experienced, but I would just say generally that, a woman who decides to take risks or has to take risks because of her situation, that there should be policies in place that protect women. […] Nobody is deserving of harm because of the decisions that they have made or have been forced to make. [Kayla]

But like, if somebody is like Black and poor, then they have to take more risks, because it's necessary for them to feed their family and stuff. But since I don't have to take that risk, because like, I'm well‐off enough to just be comfortable in my house, you know. [Justice]

Given the paucity of coded references, we did not include these in the statistical group comparisons we report in following section. However, we note that all five participant who made comments relevant to being compelled to take risks were racialized, and four also identified as queer. Four lived in New York City and one in Toronto.

6.1.4. Put at risk: Between‐group comparisons

We examined frequencies to identify group differences based on participants' characteristics (see Table 4). Comparisons indicated no differences by age, subjective SES, or ethnoracial group. We observed one significant difference by site, with less than half of Toronto participants referring to feeling put at risk (any code) compared to those from New York City (over two‐thirds) or Melbourne (almost three‐quarters; FFH p = .042).

Looking at the intersection of gender and sexuality (i.e., gender × sexuality, wherein the asterisk represents intersectionality), we found a significant difference between the proportion of cishet (i.e., cisgender and heterosexual) participants who referred to being put at risk (approximately 29%) and the cisgender queer and TNB queer participants who did (approximately 66% of each group; FFH p = .030). Similarly, a smaller proportion of cishet participants than cisgender and TNB queer participants felt exposed to harm (FFH p = .024) and deprived of resources (FFH p = .051).

Approximately 29% of racialized cishet participants referred to feeling put at risk, a significantly smaller proportion than White TNB/queer participants (almost two‐thirds) and racialized TNB/queer participants (over two‐thirds; FFH p = .034). A significantly smaller proportion of racialized cishet participants commented that they felt exposed to harm (FFH p = .023). There was no significant difference among participant subgroups in resource deprivation.

Participants' comments aligned with quantitative evidence of the salience of feeling put at risk, specifically by being left exposed, among TNB/queer participants across cities and ethnoracial groups. Ten participants described being subject to homophobic and transphobic harassment at school (e.g., bullying, being outed) without effective protection or intervention, five felt endangered by policies and/or authorities (e.g., police, social services), and five felt unsafe in their familial homes as a result of their gender or sexuality. In addition to potential conflict or violence at home, parental disapproval could push TNB/queer youth into public spaces, inviting other threats. As Ziggy explained:

The thing is, like most queer kids', like most gay kids' houses are not available [for being with sexual partners]. So like, I've had many of my friends being like, “Oh, you can just like, if their house isn't available, there's a back alley somewhere,” you know? Yeah. Like, which I think is pretty unsafe. Maybe don't have sex in a back alley or like, a park. Maybe don't do that. But at the same time, you might have to do what you have to do.

The dangers of TNB/queer youths' use of outdoor spaces to see each other was amplified by 13 participants who feared being targeted while in public, especially if also racialized:

A lot of my choices tend to carry more risk than with other people. Like being out, walking around campus, it's very different for a White student, a White man walking around campus. That means that it's more easy to target me, if people want to target me. It's easy to target me even within the queer community. […] Just going out for a walk, that's risky. [Fatima]

6.2. Theme 2: Kept from risk

Sixty percent of participants referred in some way to having been prevented from taking wanted sexual risks, with 37% citing their precarious circumstances and roughly equivalent proportions citing stereotypes and being overprotected (18% and 14%, respectively; see Table 4).

6.2.1. Too precarious

Participants articulated the experience and effect of social and material precarity in no uncertain terms: having “less of a buffer,” as someone from a working class background (Serena); being unable to live a “carefree life” in which one's “parents' money will like, always bail me out” (Isha); not having a “safety net” or “the luxury of choice” and therefore having “to stick to the safe thing” (June); and how fear “definitely dampens any abilities I have to just be more free” (Cindy). All 37 participants who referred to precarity linked it directly to being marginalized on the basis of race/ethnicity, immigration, class, gender, and/or sexuality:

[Being in an Asian immigrant family] definitely ties into me being very measured and being very, like, analytical of what risks I take, because if I do something wrong, if I do something and it turns out, like, catastrophically wrong, and it comes back to my family, I can't recover as quickly as a White person […]. There's just so much more riding on it, you know? [Cindy]

Knowing I belong to multiple different categories of identity that literally just have lower lifespans, there's more danger factors that can affect many groups that I belong to, it kind of factors into the way that I navigate the world […] in a different way than some of my White cis friends. Definitely I think that's part of the reason why I am a bit more reserved with stuff. Because I feel like in certain scenarios, I'm not really given the same kind of benefit of the doubt or the same safety net. [CJ]

You know being like queer, being Black, being nonbinary, that really like halts me in taking risks. Because I just feel like there's more, like sometimes it's more dangerous for me to take risks in certain scenarios, because I know that the consequences if this goes wrong will be ten times worse for me than it would be for like, my White cisgendered counterpart. [Renee]

Like Cindy who attributed her “measured” and “analytical” approach to tenuous circumstances, others also saw the impact of precarity on their self‐concepts and life orientation. As Serena put it, “I'm not socially produced to take risks,” while Tai thought that fears of racist, gender‐based, and/or homophobic violence had “shaped me into being, like, a very tamed person.” Nat mused in their interview, “Sometimes, I wonder if I was, I don't know, a big, strong, heterosexual, rich, White man—that was a mouthful!—if I was that, if I would be taking more risks?”

6.2.2. Potentially stereotyped

Eighteen participants, half of whom lived in New York City, described refraining from certain modes of sexual behavior or self‐expression as they tried to anticipate and pre‐empt prejudice. Three White participants referred to needing to defend against sexist and heterosexist stereotypes (e.g., not pursuing sexual interests to avoid slut‐shaming, not talking openly with friends about their sexual relationships with women to avoid fetishizing comments). Fourteen of the remaining 15 participants, all of whom were racialized, referred to being constrained by intermingled racism and sexism, whether because of their hypervisibility (“I would stick out more like a sore thumb, if I had taken that risk” [CJ]), because they were trying to avoid confirming negative stereotypes (e.g., “I don't want to act in a way that would give someone ammunition to be like, ‘See, I told you that's how these people are.’” [Kayla]), or because they worried they would face disproportionately negative consequences (“Whenever I take the risk, I often have to think about, will the fact that my race is a certain color already make like any negative impact from that risk two times worse?” [Ash]; “White people or straight people don't really get questioned for their actions. But like, if a person of color were to do something risky, it would be like seen as a crime or something.” [Shaima]). As in the earlier quote from Renee, Brooklyn pointed out the interplay between stereotypes and precarity:

You feel like put down in society because of, because your gender at birth and being Black. Like, you gotta work extra hard to be taken seriously. And so it's like, sort of hard taking risks because, because like you're worried about, like, how is this going to affect you, affect you in like later in life. […] I would say, like definitely say, there's a financial aspect of it. I feel like a lot of people don't want to take risks, because like, there's like a lot to lose, at stake, especially in the financial aspect, like, housing, employment, food.

6.2.3. Too overprotected

All 14 participants who felt hampered by protectionism identified as racialized. One felt policymakers and educators underestimated young women as naïve, while the remaining 13 focused on their parents. Ten attributed their parents' restrictions to being immigrants with norms from their cultures of origin and fears for their children's safety. For example, Jiaxin described trying to reconcile her parents' prohibitions with her frustration:

So, my parents are like, really strict. I mean, that's typical for Asian parents. […] My parents are stricter on me just because I'm the only daughter. And it's understandable. It's not because they don't trust me, it's because they don't trust society, the people out there. But it's really upsetting because it feels like, I'm missing out on a lot.

Three queer participants said their parents tried to insist they remain closeted for their own safety. Like Jiaxin, they worked to rationalize their parents' perspectives:

My dad was more like, he didn't want me to be outcasted. […] It was like, borderline homophobic, but it was also, he just wanted me to be safe. I'm like, you know, that's great. It's just that I'd rather be authentic, than have to hide for another couple of years. [Amal]

[My mother thinks] society will treat you differently if you are not heterosexual, like if you are not straight, she says people will treat you differently. You know, you could get hate crimed, you could get restricted from job opportunities. […] But I do understand that my mom is just looking out for the best for me and it's not her like prejudiced views or whatever, even if she might have them. It's her concern that society will treat me differently and I won't be able to be as happy as I could be by just giving up on dating girls. [June]

Participants' comments shed light on the tension they experience between self‐expression and others' authority—even if seemingly well‐intentioned—over them.

6.2.4. Kept from risk: Between‐group comparisons

As indicated in Table 4, we found no differences by age, SES, or site among participants who referred to the overall kept from risk theme or its constituent codes. The only significant differences between groups pertained to protectionism (i.e., being prevented from taking desired sexual risks by intervening authorities). Although only 14 participants referred to feeling overprotected, comparisons consistently indicated a relation to race/ethnicity: in ethnoracial group comparisons, no White or Latinx participants referred to being overprotected (FFH p = .031); a marginally larger proportion of cishet participants, all racialized (as per the logic of the eligibility criteria), referred to feeling overprotected compared to cisgender and TNB queer participants (FFH p = .053); and race/ethnicity × gender × sexuality intersectional comparisons indicated a significant difference between White TNB/queer participants and racialized participants, whether cishet or TNB/queer (FFH p = .005).

The implication of race/ethnicity in protectionism in these group comparisons aligns with the substance of participants' comments reviewed earlier. Importantly, parents' anxieties reflect not only different cultural norms, but also their awareness of racism (e.g., parents warning their children about race‐related hate crimes) and in the case of queer/TNB racialized young people, homo‐ and transphobia, as described in the earlier comments from Amal and June. Jess, who was White, realized the impact of racial discrimination and privilege on parents' protectionist inclinations when comparing her experience to her ex‐girlfriend's, who was South Asian:

[W]hen I think of a victim of a hate crime towards LGBTQ people, I do not think of a White lesbian. So, that was never really an issue for me. [Description of ex‐girlfriend's family and parents.] Like, they themselves were not homophobic, but a big concern for them was like, they were so worried that she would be a victim of a hate crime, which I thought was kind of interesting. Because that was never a concern of my parents.

Dev also zeroed in on the inhibitory effect of marginalization on sexual risk‐taking, especially but not only by race/ethnicity:

And I think maybe, like risk kind of looks different in White people than Brown queer people. Because I guess Brown queer people, like their risks are different because they have more barriers. And then maybe like a White person, it's different for them. Maybe they're more willing to, like, engage in sexual activity at a young age and explore their sexual identity and like, do things that… you know, that's not really in the cards for me. Like, even a straight Brown girl probably wouldn't explore her sexual identity and have sex and do things like that at a young age, compared to like, White people, I guess.

As many of the preceding excerpts show, being put at risk and kept from risk were not mutually exclusive circumstances. To the contrary and as explained by Avery, their intertwining puts marginalized young people in a seemingly unnavigable predicament:

Being diverse in that way limits you from taking some risks, because of like, the other risks it brings along. Because you're already at like, I don't know, heightened chance to, you know, have violence against you. […] So for some people, it's harder to take risks and they don't really have the opportunity to if they want to, like, you know, survive.

7. DISCUSSION

Our relatively large and rich qualitative data set permitted novel insight into the risk dilemmas confronting marginalized youth at various intersections of race/ethnicity, gender, and sexuality. Overall, findings supported Bay‐Cheng and Ginn's (2024) proposed sexual risk dilemmas, with over 80% of participants pointing to social conditions that endangered them (particularly exposing them to risk or depriving them of resources) and/or hampered them (specifically by making them too precarious, prone to stereotypes, or overprotected) in their sexual lives. Whether put at sexual risk, kept from it, or both, the development and dignity of racialized and/or queer young women and TNB youth are compromised by social and material conditions that endanger their well‐being and infringe on their rights, including to taking risks.

In their interviews, participants described considerable anxiety about their vulnerability to violence. While exposure was one of the key components of being put at risk in Bay‐Cheng and Ginn's (2024) original formulation of the sexual risk dilemmas, they were primarily informed by literature pertaining to structural environmental factors (e.g., high community rates of STIs; Corcoran et al., ²⁰²¹). The predominant concern of participants in this study, however, was their susceptibility to interpersonal hostility and harm. For some, this stemmed from personal experiences of threat and violence on the basis of gender (whether as femme‐presenting or for gender nonconformity), sexuality, and/or race/ethnicity, while others referred to a more generalized fear of such dangers without having direct experiences. The pervasive and heightened apprehension among participants may be artifacts of the sociopolitical period during which we collected data and our adolescent and young adult participants' identities had been forged, including what Khanlou et al. (2021) identify as the “syndemic” of COVID‐19, gender‐based violence, and manifest racism (see also Velez & Hostinar, ²⁰²³).

Popular and political efforts to curtail the rights and resources of marginalized individuals and communities (e.g., Fuentes, ²⁰²³; Hastings & Hodge, ²⁰²³) were not lost on participants, either. As detailed in the Results, the 26 participants who felt put at risk by deprivation framed this in terms of rights restrictions, untrustworthy services and service providers, and a scarcity of high‐quality, affirming, and practically accessible services. Many participants linked this to sexism, cis‐ and heterosexism, racism, and socioeconomic disparities, though age also blocked their access to transportation, housing, money, credible information, and confidential services. As illustrated by Sammy's efforts to exercise their right to confidential STI testing, formal policy provisions may be nullified by adultist systems, practices, and norms that privilege parents' authority over youths' rights (Bay‐Cheng et al., 2022).

Participants were critically aware of and articulate about the ways in which they were kept from sexual risk‐taking by intermingled gendered, ethnoracial, sexual, and class‐based precarity and stereotypes. The extent of their conscious efforts to anticipate, ward off, and counteract situations that could derail their lives or diminish them in the eyes of others was not only a significant energy drain, but also an imposed constraint that, as Cindy put it, “dampens any abilities I have to just be more free.” Participants' risk aversion is a potent rebuttal to racist, classist, cis‐ and heterosexist, and/or adultist stereotypes of marginalized youth as reckless and irresponsible pleasure‐seekers (Lesko, 2012). However, this should not be celebrated as or mistaken for a freely‐chosen path of cautious self‐protection. Instead, as Cindy articulated, “being very measured” is demanded by social and material conditions and ultimately restricts participants' self‐expression, experimentation, and autonomy. This is precisely the cost that disability studies scholars warn that people pay when denied the dignity of risk (Perske, 1972), and represents a clear developmental toll given how crucial exploration is to young people's growth during adolescence (Fortenberry, 2014; Harden, ²⁰¹⁴; Tolman & McClelland, ²⁰¹¹).

We were surprised by two findings that diverged somewhat from Bay‐Cheng and Ginn's (2024)'s conceptualization of the sexual risk dilemmas. First, no participants disclosed being compelled to take risks, such as sex trading or strategic consent to unwanted sex. Only five referred to such circumstances at all, and none pertained to their own experiences. There are several possible explanations for this, ranging from sample self‐selection bias (i.e., only those with some amount of privilege could or chose to participate) to the effects of social desirability, particularly the contemporary neoliberal imperative to present oneself as in control and invulnerable (Bay‐Cheng, 2015). It is also possible that needing to take sexual risks to meet basic needs is relatively uncommon. We do not see this relative infrequency as grounds for discounting or removing it from consideration (i.e., dropping it as an outlier), since the prevalence—or rarity—of an experience is not the sole determinant of meaningfulness. Instead, we think an important research implication of our study is that sampling and data collection must be intentionally designed to explore what could be a significant albeit minority experience of being compelled by circumstance to take sexual risks (as models, see Watson's [2011] study of survival sex among youth in Melbourne and Joly and Connolly's [2019] of street‐involved youth in Toronto).

Second, we expected protectionism to figure more prominently in participants' perceptions of feeling prevented from taking desired risks. However, only 14% of participants described feeling hemmed in by protectionist norms and intervening authorities. The relative unimportance of adult authority in participants' reflections on sexual risk‐taking may be a function of: participants' statuses as older adolescents and/or young adults; that many were above the age of legal majority in each of the study's sites; and/or that two‐thirds were already out of high school and therefore felt less accountable to adults. On a related note, some participants' deprivation of resources appeared linked to, if not caused by, age‐based restriction and adultism, yet this also was not a dominant theme among participants. However, a competing explanation is that adultism is so thoroughly naturalized in our social worlds and relationships as to be imperceptible and unremarkable (Lesko, 2012). We did reflect on the comments of participants who felt kept from taking risks by protectionism, in almost all cases by parents' prohibitions. As indicated in the illustrative excerpts—and almost all other coded segments, too—participants who felt their social and sexual lives were constrained by parents were not simply frustrated or resentful; instead, they interpreted their parents' prohibitions as evidence of caring and good intentions, misdirected as they may be. The participants' sympathetic positions vis‐à‐vis parental protectionism had not been anticipated by Bay‐Cheng and Ginn (2024) and adds needed depth and complexity to the sexual risk dilemmas. Indeed, it illustrates Cense's (2019) conceptualization of “bonded agency” to accommodate relationality as a priority for many young people, and provokes critical reflection on the privileging of individualism in conventional models of youth development.

7.1. Sexual risk dilemmas at intersections

Analyses generally confirmed that many participants felt put at unwanted sexual risk and kept from taking wanted ones (low frequencies of compelled risk‐taking notwithstanding), though not necessarily in the same ways or to the same degrees across geographic and social locations. Significantly smaller proportions of Toronto participants felt either put at risk or kept from it, suggesting that the city may offer youth a greater sense of safety and support than Melbourne or New York City. This is an interesting finding and invites further investigation, particularly since larger cross‐national analyses do not indicate a significantly better quality of life for young people in Canada, especially compared to Australia (Pullen Sansfaçon et al., 2023; Shah, et al., ²⁰¹⁹) and Toronto‐based studies reveal significant gaps in youth sexual well‐being and supports (e.g., Narushima et al., ²⁰²⁰; Toronto Public Health, ²⁰¹⁹).

In looking at codes across social identities and intersections thereof, we saw some signs that feeling put at risk may be particularly related to marginalized gender and sexual identities and corresponding discrimination. In group comparisons by gender × sexuality and by intersections of race/ethnicity × gender × sexuality, equivalent proportions (i.e., roughly two‐thirds) of racialized and White TNB/queer participants felt put at risk, both significantly larger than the 29% of cishet participants of color who felt this way. Given the salience of worries about hostility and being “hate‐crimed,” fears about homophobic and transphobic harassment may have been at the root of participants' feelings of endangerment (specifically being left exposed). By no means does this dismiss or minimize concerns about racially motivated harms. As evident in several participants' comments, many were acutely sensitive to the interplay between racism and other forms of bias, particularly anti‐Asian and anti‐Black racism, heightening both their vulnerability and consequent anxiety. Yet as revealed by some of the participants' comments, racism was often interlaced with misogyny (e.g., the fetishization of Asian women) and many queer participants of color saw their race as heightening their visibility and concomitant vulnerability to homo‐ and transphobic harassment. These findings are useful indicators of how under threat and underresourced young queer/TNB people feel as they make their lives, and of the need for dedicated, affirming, practical support from others.

Group and intersectional analyses of participants who referred to being kept from risk yielded fewer statistically significant or conceptually clear findings than for being put at risk, though some code frequencies suggested a link to race and racism. Proportions of White TNB/queer participants who felt they were prevented from risk‐taking were consistently smaller than racialized participants, whether TNB/queer or cishet. Participants' comments, such as Jess's comparison of her parents' minimal concern about her safety with the fears of her South Asian ex‐girlfriend's parents, Dev's observation that “even like a straight Brown girl probably wouldn't explore her sexual identity and have sex and do things like that at a young age, compared to like, White people…,” and CJ's point that if they were to take a risk they would “stick out like a sore thumb,” frame risk‐taking as a luxury afforded by racial privilege. This may be a manifestation of the “White man effect” (Finucane et al., 2000): that those with more cultural and socioeconomic capital perceive fewer risks, while those with less capital perceive risks to be greater in magnitude and number. As we explored earlier, being prevented from taking risks is also related to race/ethnicity through cultural norms and fears of racism. While the current data set offers only tentative signs that racism substantially hinders youths' sexual self‐expression and experimentation, we see this as a crucial direction for future research and, as needed, advocacy for equal sexual rights and dignity.

7.2. Limitations and future directions

We have already identified some paths for future research, such as more intentional studies of compelled risk‐taking, greater attention to young people striving both for independence and the maintenance of familial ties, and attention to differences in the social and policy contexts of Toronto compared to Melbourne and New York City. The limitations of our study point to several other future directions as well. For one, although some participants referred to class and material insecurity, there was not enough socioeconomic variance in the data to explore this more thoroughly or reliably. We expect that more dedicated and detailed studies of socioeconomic status would have much to contribute to understanding how young people—and which young people—are put at or kept from risk. Furthermore, although our sample included significant ethnoracial, gender, and sexual diversity, we also made use of larger umbrella groups (e.g., racialized or not, TNB/queer or not) in our comparative intersectional analyses. We acknowledge the conceptual, empirical, and political problems associated with such aggregation (e.g., obscured distinctiveness, false and dehumanizing overgeneralization) and see the need for more intentional sampling to enable finer‐grained analyses. Nevertheless, insofar as being treated as a generalized “other” is a common mechanism and experience of marginalization, we see some meaning and utility in considering these aggregated groups (i.e., not White, not cishet).

Finally, we are mindful of more common limitations stemming from the study's research design, substantive focus, and analytic approach. Retrospective self‐reports collected from a geographically varied convenience sample through recorded individual interviews conducted by different interviewers (i.e., research design), pertaining to complex, intimate, and often stigmatized issues (i.e., substantive focus), and examined through the lenses of both scholarly expertise and personal standpoint (i.e., analytic approach) can be replete with biases (e.g., self‐selection, recall, social desirability, confirmation, etc.). As detailed in the Section 4.3, we followed best practices for qualitative analyses and directed content analysis more specifically (e.g., Elo et al., ²⁰¹⁴; Hruschka et al., ²⁰⁰⁴; Hsieh & Shannon, ²⁰⁰⁵; Neuendorf, ²⁰¹⁸) to address some of these and increase the trustworthiness of our treatment and interpretations of the data. One specific factor potentially compromising the credibility of our analyses is that Bay‐Cheng may have been too invested in confirming the sexual risk dilemmas she and a collaborator had previously proposed (Bay‐Cheng & Ginn, 2024). This concern is offset somewhat by the authors' attention to findings that surprised them (e.g., the dearth of compelled sexual risks, that protectionism was not more prominent). Furthermore, Sutton played a key role in countering Bay‐Cheng's potential bias. Although “junior” in scholarly expertise and experience, she did not hesitate to assert differing views and interpretations of the data based on her generational and identity affinity with the participants and experience conducting some interviews. Indeed, Sutton's confidence and comfort as a member of our intergenerational research team throughout the larger project made her an important collaborator in the current study.

7.3. Conclusion

Adults are right to worry about young people's sexual risk, though perhaps not only in the ways or for the reasons we most commonly do. Rather than oversimplify risk as inevitably detrimental and inherent to youth and their sexualities (and therefore beyond our control and responsibility), we must also keep in mind the dignity and dividends of risk‐taking. In this study, queer and/or racialized young women and TNB youth from cities in three countries described dilemmas in their sexual lives that stemmed from their social and material conditions: how their circumstances left them vulnerable to harm, surrounding them with threats and denying them resources; and how those same circumstances left many of them with no margin for error, foreclosing their opportunities for sexual experimentation and expression. The perspectives shared by participants offer not only insight into their lives, but also the imperative to rectify the inequities that jeopardize their sexual rights both to safety and to risk.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS STATEMENT

Study procedures and measures were approved by ethics review boards at five universities (i.e., principal investigators' home institutions) in Australia, Canada, and the United States.

Bay‐Cheng, L. Y. , Sutton, E. R. , & 4tRecord Research Team. (2025). Put at and kept from risk: The sexual risk dilemmas confronting marginalized youth. Journal of Adolescence, 97, 434–450. 10.1002/jad.12430

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.