Abstract
Lesbian, gay, bisexual, transgender/nonbinary, and queer (LGBTQ+) adolescents experience considerable disparities in dating violence and sexual assault victimization relative to heterosexual and cisgender peers. These disparities may be driven in part by the disruptive effects of heterosexism and cissexism on school-based and family relationships. To quantify the potential roles of these processes and identify priorities for prevention efforts, we estimated the extent to which dating violence and sexual assault victimization in LGBTQ+ adolescents could be reduced by eliminating sexual orientation and gender modality inequities in school adult support, bullying victimization, and family adversity. We analyzed data from a cross-sectional, population-based survey of high school students in Dane County, Wisconsin (N=15,467; 13% sexual minority; 4% transgender/nonbinary; 72% White) using interventional effects analysis, adjusting for grade, race/ethnicity, and family financial status. We found that eliminating inequities in bullying victimization and family adversity could significantly reduce dating violence and sexual assault victimization in LGBTQ+ adolescents, particularly sexual minority cisgender girls and transgender/nonbinary adolescents. For instance, eliminating gender modality inequities in family adversity could reduce sexual assault victimization in transgender/nonbinary adolescents by 2.4 percentage points, representing 27% of the existing sexual assault victimization disparity between transgender/nonbinary and cisgender adolescents (P<.001). Results suggest that dating violence and sexual assault victimization in LGBTQ+ adolescents could be meaningfully reduced by policies and practices addressing anti-LGBTQ+ bullying as well as heterosexism- and cissexism-related stress in LGBTQ+ adolescents’ families.
Keywords: Dating violence, sexual assault, sexual orientation, gender identity, LGBTQ+, bullying, adverse childhood experiences
Heterosexism and cissexism—prejudice and structural disadvantage against sexual minority people (e.g., lesbian, gay, bisexual) and transgender/nonbinary people, respectively—are considerable threats to adolescent health. For sexual minority and transgender/nonbinary adolescents (hereafter, LGBTQ+ adolescents), heterosexism and cissexism appear to increase the risks of experiencing dating violence (Stephenson et al., 2022) and sexual assault (Marx et al., 2021; Murchison et al., 2019). Adolescent dating violence and sexual assault victimization are associated with depressive symptoms, suicidality, and problematic substance use (Exner-Cortens et al., 2013; Kilpatrick et al., 2003), all key health disparity domains for LGBTQ+ adolescents (Johns et al., 2019; Johns et al., 2020). Thus, preventing dating violence and sexual assault may be key to addressing broader health disparities between LGBTQ+ and heterosexual/cisgender adolescents. However, these prevention efforts require a better understanding of the processes by which heterosexism and cissexism impact adolescent dating violence and sexual assault.
Dating violence is psychological, physical, and/or sexual aggression between youth in a dating relationship (Hébert et al., 2019); sexual assault is physical sexual contact against a person’s will (Tillyer et al., 2010). Dating violence and sexual assault can overlap at the event level, with dating partners perpetrating approximately 20% of sexual assaults targeting adolescents (Hamby et al., 2012). In addition, adolescent dating violence and sexual assault victimization share underlying risk factors and are thus strongly correlated at the individual level (Hamby et al., 2012). Cisgender boys are far less likely to be sexually assaulted than cisgender girls or transgender/nonbinary adolescents; gender patterns are less consistent for dating violence (Centers for Disease Control [CDC], 2022; Hamby et al., 2012; Johns et al., 2020).
Among U.S. high school students, LGBTQ+ adolescents experience significantly more dating violence than heterosexual/cisgender peers. For instance, the 12-month prevalence of physical dating violence victimization is greater in sexual minority vs. heterosexual girls (12.1% vs. 8.8%; CDC, 2022), in sexual minority vs. heterosexual boys (15.9% vs. 6.2%; CDC, 2022), and in transgender/nonbinary vs. cisgender students (26.4% vs. 8.0%; Johns et al., 2020). Sexual assault exhibits similar patterns: The lifetime prevalence of physically forced sexual intercourse is considerably higher in sexual minority vs. heterosexual girls (21.0% vs. 9.2%; CDC, 2022), in sexual minority vs. heterosexual boys (15.6% vs. 2.5%; CDC, 2022), and in transgender/nonbinary vs. cisgender adolescents (23.8% vs. 7.4%; Johns et al., 2019). There is growing evidence that heterosexism and cissexism play important roles in these disparities. Quantitative studies have found that transgender/nonbinary adolescents who face bias-based peer victimization (Marx et al., 2021) or stigmatizing school restroom and locker room policies (Murchison et al., 2019) are at elevated risk of sexual assault victimization compared to transgender/nonbinary adolescents not exposed to these forms of cissexism. The effects of heterosexism and cissexism extend beyond bias-motivated violence; for instance, sexual minority adolescent boys have described how heterosexism among peers and families can lead to conflict with dating partners, resulting in dating violence (Stephenson et al., 2022).
Research on adolescent interpersonal violence has often adopted social-ecological perspectives, highlighting how social and cultural contexts impact violence involvement. Espelage’s ecological model of peer victimization (2014) captures key findings from this literature, describing how relationships with school staff, peers, and parents/caregivers shape the risk of victimization. This model is not specific to dating violence or sexual assault, but research on dating violence and sexual assault supports the model’s contentions that close and positive school-based and family relationships are protective against victimization (Hébert et al., 2019; Tillyer et al., 2010), while disrupted (e.g., weak or violent) school-based and family relationships are risk factors for dating violence (Hamby et al., 2012; Hébert et al., 2019; Hipwell et al., 2014; Morris et al., 2015). (The influence of disrupted relationships on sexual assault has received less attention.) The Espelage model highlights various forms of protection that school staff, peers, and parents/caregivers may provide, such as monitoring adolescents’ activities and intervening when violence occurs (Espelage, 2014). Others have suggested additional protective processes, such as modeling healthy relationships (Hébert et al., 2019) and advising adolescents as they navigate risky situations (Hébert et al., 2019). Not only may disrupted relationships provide less protection (Finkelhor, 2008; Morris et al., 2015), they may have psychological impacts that increase vulnerability to victimization (Felix et al., 2019; Hipwell et al., 2014).
The Espelage model holds that school-based and family relationships are shaped by broader contexts, including laws, policies, and broader sociocultural dynamics (Espelage, 2014). The model also holds that interactions among contexts impact victimization risk, e.g., when school staff intervene in peer victimization (Espelage, 2014). As depicted in the conceptual model in Figure 1, these insights suggest processes by which heterosexism and cissexism may contribute to dating violence and sexual assault. In our model, structural heterosexism/cissexism (e.g., laws, policies, stereotypes; Krieger, 2020) can disrupt school-based and family relationships via prejudiced attitudes among school staff, peers, and families (Watson et al., 2021) and, more generally, as a stressor that may strain these relationships. Victimization in one context may also be influenced by heterosexism/cissexism in other contexts, e.g., when school staff ignore victimization of LGBTQ+ students due to their own biases (Snapp et al., 2014).
Figure 1.
Social-ecological conceptual model for the influence of structural heterosexism and cissexism on dating violence and sexual assault victimization in LGBTQ+ adolescents.
Supporting our conceptual model, several studies have identified aspects of LGBTQ+ adolescents’ school-based and family relationships as risk or protective factors for sexual assault or dating violence victimization. Addressing family influences, one study identified adverse childhood experiences (primarily family adversity, i.e., familial abuse and serious parent/caregiver challenges) as a risk factor for dating violence victimization in sexual minority adolescents (Petit et al., 2021), a second identified parental monitoring as a protective factor for dating violence victimization in LGBTQ+ adolescents (Stroem et al., 2021), and a third identified family support as a protective factor for sexual assault victimization in transgender/nonbinary adolescents (Ross-Reed et al., 2019). Regarding school-based relationships, one study linked school support (primarily from adults) to lower dating violence victimization risk in transgender/nonbinary adolescents (Ross-Reed et al., 2019), and a second identified peer bullying and sexual harassment as risk factors for sexual assault victimization in transgender/nonbinary adolescents (Marx et al., 2021).
While characteristics of school-based and family relationships have been identified as risk factors for sexual assault and dating violence in LGBTQ+ adolescents, few studies have addressed the extent to which disruptions to school-based and family relationships could account for the impacts of heterosexism and cissexism on dating violence or sexual assault victimization. Of two studies addressing this question, one found that peer sexual harassment victimization fully mediated the association between stigmatizing school restroom/locker room policies and sexual assault victimization in transgender/nonbinary adolescents (Murchison et al., 2019), but this analysis addressed only one form of structural cissexism and did not explore the roles of school staff or families. A second study found that the association between sexual orientation and dating violence victimization was attenuated by adjusting for an array of risk and protective factors (Petit et al., 2021). This study did not quantify each risk/protective factor’s contribution and thus does not provide a roadmap for prevention efforts. However, its approach—using sexual orientation differences in risk/protective factors as a proxy for the effects of heterosexism—has the advantage of capturing the effects of multiple forms of heterosexism, including those that are subtle or otherwise difficult to measure.
The Present Study
To elucidate the processes by which heterosexism and cissexism contribute to adolescent dating violence and sexual assault, and to identify priority areas for violence prevention efforts with LGBTQ+ adolescents, the goal of this study was to estimate the extent to which addressing specific impacts of heterosexism and cissexism on school-based and family relationships could reduce dating violence and sexual assault victimization in LGBTQ+ adolescents. Because many forms of heterosexism and cissexism are difficult to measure (Krieger, 2020), we operationalized their effects in terms of the less-favorable distributions of risk and protective factors in LGBTQ+ adolescents compared to heterosexual/cisgender peers. Specifically, we estimated how the prevalence of dating violence and sexual assault victimization in LGBTQ+ adolescents might change if LGBTQ+ adolescents experienced the same distributions of school adult support, bullying victimization, and family adversity as heterosexual/cisgender adolescents.
Within the domains of school-based and family relationships, we focused on school adult support, bullying victimization, and family adversity because each is prominent in the literature on LGBTQ+ adolescent health and has previously been linked to adolescent dating violence and/or sexual assault. School adult support reflects students’ perceptions that teachers and other school staff are encouraging, supportive, and fair; responses to peer victimization play key roles in these perceptions (Suldo et al., 2009). School adult support has been identified as a protective factor for dating violence victimization in transgender/nonbinary adolescents (Ross-Reed et al., 2019) and for sexual assault victimization in adolescent boys (Tillyer et al., 2010). Many LGBTQ+ adolescents report that school adults ignore or even blame them for peer victimization they experience (Kosciw et al., 2020; Snapp et al., 2014), suggesting that heterosexism and cissexism may reduce school adult support. Likewise, bullying victimization has been identified as a risk factor for dating violence and sexual assault victimization for adolescents overall (Felix et al., 2019; Hipwell et al., 2014) and for sexual assault victimization in transgender/nonbinary adolescents (Marx et al., 2021). LGBTQ+ students experience considerably more bullying than heterosexual/cisgender peers (Johns et al., 2019; Johns et al., 2020), and many report being targeted based on heterosexism and cissexism (Kosciw et al., 2020). Finally, family adversity—disruptive family experiences, including abuse and serious parent/caregiver difficulties (e.g., mental illness)—is a risk factor for dating violence in heterosexual and sexual minority adolescents (Morris et al., 2015; Petit et al., 2021). LGBTQ+ adolescents report more family adversity than heterosexual/cisgender peers (Andersen & Blosnich, 2013; Craig et al., 2020), likely due to the strain that heterosexism and cissexism can place on family relationships.
We hypothesized that eliminating inequities in school adult support, bullying victimization, and family adversity between LGBTQ+ and heterosexual/cisgender adolescents would be associated with significant reductions in the predicted prevalence of dating violence victimization (Hypothesis 1; H1) and sexual assault victimization (Hypothesis 2; H2) among LGBTQ+ adolescents. We were primarily interested in the effects of changing each risk/protective factor individually (“single-factor estimates”; H1a-c and H2a-c) because these results could indicate which risk/protective factors should be prioritized for intervention. We also estimated the joint effects of changing all three factors (H1d and H2d). Given prior evidence of gender differences in risk/protective factors for dating violence and sexual assault (Tillyer et al., 2010; Yan et al., 2009), we addressed each hypothesis separately in sexual minority cisgender girls, sexual minority cisgender boys, and transgender/nonbinary adolescents. We tested these hypotheses using interventional effects analysis, which is increasingly recommended in health disparities research to estimate the public health impact of achieving equity in risk/protective factors across population groups (Jackson & VanderWeele, 2019). In a health disparities context, one advantage of interventional effects analysis over traditional mediation analysis is that, conceptually, it does not position a minoritized identity as the “cause” of inequities (Jackson & VanderWeele, 2019). We expressed effects as the expected population-level reduction in victimization in the LGBTQ+ group (adjusted prevalence differences) and calculated the proportions of existing disparities that could be eliminated by each change. To contextualize the interventional effects analyses, we first calculated (1) adjusted differences in each risk/protective factor between LGBTQ+ and cisgender/heterosexual adolescents; (2) each risk/protective factor’s adjusted association with dating violence and sexual assault victimization; and (3) adjusted differences in the prevalence of dating violence and sexual assault victimization between LGBTQ+ and cisgender/heterosexual adolescents (existing disparities).
Methods
Data
We analyzed data from the 2018 Dane County Youth Assessment (DCYA), a cross-sectional survey administered in 17 public school districts and one private school in Dane County, Wisconsin in January–March 2018 (Dane County Youth Commission, 2018). Two districts sampled 50% of eligible students while the remainder surveyed all eligible students. Participation exceeded 90% in most districts. Surveys were anonymous and administered electronically. Parents were notified of the survey and could decline consent; students provided voluntary assent/consent. The Harvard T.H. Chan School of Public Health IRB determined that this secondary analysis of DCYA data was not human subjects research.
We excluded participants not in 9th–12th grade (n=143) and those missing data on sex assigned at birth (n=284) or sexual orientation/gender modality (n=414), for an analytic sample of 15,467 participants. Primary dating violence analyses were restricted to students who dated in the prior 12 months (n=7,149); compared to the full sample, these students were in higher grades (P<.001), were less often cisgender sexual minority boys (P<.001) or Asian/Asian American (P<.001), had poorer family financial status (P<.001), and had more family adversity (P<.001). As a sensitivity analysis, we also performed dating violence analyses among all students.
Sample characteristics are presented in Table 1. Of cisgender participants, 10.9% were sexual minority and 89.1% were heterosexual. Transgender/nonbinary adolescents were 3.8% of the sample; most were sexual minority (73.1%). Most participants were White (71.8%), while 5.8% were Asian, 5.3% were Black, 7.6% were Latinx, and 9.5% were multiracial or another race/ethnicity. Most participants (61.2%) reported their family financial status as good.
Table 1.
Characteristics of Dane County, Wisconsin high school students who participated in the 2018 Dane County Youth Assessment, N=15,467, by sexual orientation and gender modality
| Cisgender heterosexual adolescents |
Cisgender sexual minority adolescents |
Transgender & nonbinary adolescents (n=602) |
|||
|---|---|---|---|---|---|
| Girls (n=6,236) | Boys (n=7,014) | Girls (n=1,181) | Boys (n=434) | ||
| n (%) | n (%) | n (%) | n (%) | n (%) | |
| Gender identity | |||||
| Female | 6227 (100.0%) | – | 1179 (100.0%) | – | 57 (9.5%) |
| Male | – | 7002 (100.0%) | – | 432 (100.0%) | 127 (21.1%) |
| Nonbinary | – | – | – | – | 156 (25.9%) |
| Gender fluid | – | – | – | – | 137 (22.8%) |
| Other | – | – | – | – | 125 (20.8%) |
| Sexual orientation identity | |||||
| Straight/heterosexual | 6236 (100.0%) | 7014 (100.0%) | – | – | 125 (26.0%) |
| Asexual | – | – | 34 (2.9%) | 14 (3.2%) | 29 (6.0%) |
| Bisexual | – | – | 689 (58.3%) | 184 (42.4%) | 107 (22.2%) |
| Gay | – | – | 88 (7.5%) | 130 (30.0%) | 69 (14.3%) |
| Pansexual | – | – | 79 (6.7%) | 13 (3.0%) | 78 (16.2%) |
| Questioning | – | – | 291 (24.6%) | 93 (21.4%) | 56 (11.6%) |
| Refused | – | – | – | – | 17 (3.5%) |
| Grade in school | |||||
| 9th | 1749 (28.0%) | 2056 (29.3%) | 330 (27.9%) | 113 (26.0%) | 194 (32.2%) |
| 10th | 1646 (26.4%) | 1868 (26.6%) | 323 (27.3%) | 107 (24.7%) | 161 (26.7%) |
| 11th | 1589 (25.5%) | 1674 (23.9%) | 292 (24.7%) | 128 (29.5%) | 139 (23.1%) |
| 12th | 1252 (20.1%) | 1416 (20.2%) | 236 (20.0%) | 86 (19.8%) | 108 (17.9%) |
| Family financial status | |||||
| Good | 3812 (61.7%) | 4498 (65.3%) | 525 (45.1%) | 221 (51.9%) | 259 (45.3%) |
| Moderate | 2165 (35.0%) | 2213 (32.1%) | 539 (46.3%) | 181 (42.5%) | 254 (44.4%) |
| Poor | 203 (3.3%) | 176 (2.6%) | 100 (8.6%) | 24 (5.6%) | 59 (10.3%) |
| Race/ethnicity | |||||
| Multiracial and additional identities | 588 (9.5%) | 596 (8.5%) | 124 (10.6%) | 49 (11.4%) | 97 (16.4%) |
| Asian | 374 (6.0%) | 387 (5.5%) | 67 (5.7%) | 20 (4.7%) | 46 (7.8%) |
| Black | 288 (4.6%) | 409 (5.9%) | 73 (6.2%) | 11 (2.6%) | 37 (6.3%) |
| Latinx | 485 (7.8%) | 532 (7.6%) | 84 (7.2%) | 33 (7.7%) | 38 (6.4%) |
| White | 4463 (72.0%) | 5054 (72.4%) | 822 (70.3%) | 315 (73.6%) | 373 (63.1%) |
| Family adversity | |||||
| 0 experiences | 4704 (78.8%) | 5294 (81.2%) | 693 (62.6%) | 278 (68.1%) | 309 (59.1%) |
| 1-2 experiences | 1119 (18.8%) | 1099 (16.8%) | 329 (29.7%) | 116 (28.4%) | 158 (30.2%) |
| 3-4 experiences | 115 (1.9%) | 85 (1.3%) | 72 (6.5%) | 10 (2.5%) | 29 (5.5%) |
| 5-6 experiences | 30 (0.5%) | 45 (0.7%) | 13 (1.2%) | 4 (1.0%) | 27 (5.2%) |
| Dating violence victimization | |||||
| No | 4919 (87.7%) | 5318 (89.9%) | 834 (81.2%) | 341 (90.9%) | 364 (77.9%) |
| Yes | 691 (12.3%) | 597 (10.1%) | 193 (18.8%) | 34 (9.1%) | 103 (22.1%) |
| Sexual assault victimization | |||||
| No | 5057 (91.0%) | 5741 (97.5%) | 792 (77.7%) | 332 (89.7%) | 361 (78.1%) |
| Yes | 499 (9.0%) | 148 (2.5%) | 227 (22.3%) | 38 (10.3%) | 101 (21.9%) |
| Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | |
| School adult support | 8.90 (1.70) | 9.14 (1.93) | 8.50 (1.68) | 8.87 (1.89) | 8.32 (2.10) |
| Bullying victimization | 4.97 (1.87) | 5.25 (2.37) | 5.57 (2.49) | 5.93 (2.69) | 6.47 (3.38) |
Note. SD = standard deviation. Counts and proportions in this table are based on complete cases and, consequently, may not sum to the subgroup sample sizes. Adolescents with missing gender identity data who reported not being transgender were assigned to the cisgender subgroups based on their sex assigned at birth, resulting in minor discrepancies between the gender identity tallies and subgroup sample sizes in the cisgender subgroups.
Measures
Sexual Orientation and Gender Modality
Participants reported sexual orientation identity as straight/heterosexual, gay or lesbian, bisexual, pansexual, asexual, questioning, or write-in. Participants reporting a sexual orientation identity other than straight/heterosexual were classified as sexual minority. We determined gender modality (i.e., being transgender/nonbinary or cisgender; Ashley, 2022) based self-reports of gender identity (male, female, nonbinary, gender fluid, other), sex assigned at birth (male, female), and whether the participant identified as transgender. Students were classified as transgender/nonbinary if they identified as transgender, reported female gender identity and male sex assigned at birth or vice versa, and/or reported gender identity as nonbinary, gender fluid, or other. Students were classified as cisgender if they did not identify as transgender and reported the same gender identity and sex assigned at birth, i.e., female/female or male/male.
Outcomes: Dating Violence and Sexual Assault Victimization
Dating violence victimization was measured with four items assessing whether a dating partner had subjected the participant to verbal put-downs, isolation from friends and family, threats or physical violence, and/or forced sexual activity in the prior 12 months. Responses were dichotomized (1=experienced one or more forms of dating violence, 0=did not experience any form of dating violence). Lifetime sexual assault victimization was measured using the item, “Have you ever been forced, either verbally or physically, to take part in a sexual activity?” Responses were dichotomized (1=any sexual assault victimization, 0=no victimization).
Risk and Protective Factors in School-Based and Family Relationships
School adult support was measured using the mean of three items from the DCYA School Connection scale (Koenig, 2006; sample: “Teachers and other adults treat students fairly;” 1=Strongly disagree, 4=Strongly agree), with alpha=.72. Bullying victimization was measured using the mean of four items assessing peer victimization in the prior 30 days (Espelage & Holt, 2001; sample: “Other students picked on me;” 0=Never, 3=5 or more times), with alpha=.87. Family adversity was measured with the six-item DCYA Adverse Childhood Experiences scale (Dane County Youth Commission, 2018), asking students whether they had experienced parental mental health difficulties, parental incarceration, physical abuse by a parent, regular parental drunkenness, parental drug abuse, and parental intimate partner violence. Affirmative responses were summed and recoded as ordinal (0=0 experiences, 3=5-6 experiences).
Potential Confounders
We considered race/ethnicity and family financial status as potential confounders given that racism and socioeconomic marginalization were expected to impact the risk/protective factors and victimization outcomes. We also considered grade in school as a potential confounder given developmental differences in awareness of sexual orientation/gender identity, risk/protective factors, and victimization outcomes. For race/ethnicity, students selected one of nine options, combined into five categories for analysis: Black, Hispanic/Latinx, Non-Hispanic White, Asian, and Multiracial and additional identities). Family financial status was measured with the question “How would you describe your family’s current financial situation?” (1=We’re struggling with not having enough money, 2=Things are tight but we’re doing fine, 3=Money is not a problem for my family right now). Students self-reported their grade in school.
Results
Analyses were conducted in R (R Core Team, 2017). After examining missing data patterns (see Online Resource 1), missing data were multiply imputed (m=50) using the mice package (van Buuren & Groothuis-Oudshoorn, 2011). For analyses comparing LGBTQ+ participants to heterosexual/cisgender participants, we compared cisgender sexual minority girls to cisgender heterosexual girls, cisgender sexual minority boys to cisgender heterosexual boys, and transgender/nonbinary adolescents to cisgender adolescents. For transgender/nonbinary adolescents, we conducted initial analyses stratified by sex assigned at birth; results were comparable to those presented below. Because few transgender/nonbinary adolescents were heterosexual (n=125), we did not disaggregate transgender/nonbinary participants by sexual orientation. All models were adjusted for race/ethnicity, family financial status, and grade.
Preliminary Analyses
The prevalence of sexual assault victimization was 9.0% in heterosexual cisgender girls and much lower, 2.5%, in heterosexual cisgender boys. For LGBTQ+ participants, the prevalence of sexual assault victimization was 22.3% in sexual minority cisgender girls, 10.3% in sexual minority cisgender boys, and 21.9% in transgender/nonbinary adolescents. In students who had dated in the prior 12 months (see Online Resource 2 for subsample characteristics), the prevalence of dating violence victimization was 22.4% in heterosexual cisgender girls, 18.7% in heterosexual cisgender boys, 34.4% in sexual minority cisgender girls, 20.5% in sexual minority cisgender boys, and 39.0% in transgender/nonbinary adolescents. There was notable individual-level overlap between participants with sexual assault victimization and those with dating violence victimization: Nearly half (48.0%) of those with sexual assault victimization reported dating violence victimization, compared to 8.9% of those without sexual assault victimization. Similarly, 32.9% of participants with dating violence victimization reported sexual assault victimization, compared to 4.2% of those without dating violence victimization. There was no appreciable school-level clustering for these outcomes (see Online Resource 1).
To contextualize the interventional effects analyses, we calculated adjusted mean differences in the risk/protective factors between each LGBTQ+ group and its heterosexual/cisgender reference group using linear regression (Table 2). As expected, all LGBTQ+ groups had lower school adult support, greater bullying victimization, and greater family adversity than their heterosexual/cisgender reference groups, except that there was no significant difference in school adult support between sexual minority and heterosexual cisgender boys. We then calculated adjusted associations between risk/protective factors and outcomes via logistic regression (Table 3). All associations were significant in the expected directions, with small associations for school adult support (OR=1.10-1.11; reverse scored) and moderate associations for bullying victimization and family adversity (OR=1.32-1.52). Subsequently, we calculated adjusted differences in the prevalence of sexual assault and dating violence victimization between the LGBTQ+ and heterosexual/cisgender groups (observed disparities; see Table 4). For dating violence, there were disparities of 9.49 percentage points (pp; P<.001) for sexual minority vs. heterosexual cisgender girls and 13.72 pp (P<.001) for transgender/nonbinary vs. cisgender adolescents, but no significant disparity among cisgender boys. Sexual assault disparities ranged from 6.34 pp to 11.55 pp (all P<.001).
Table 2.
Adjusted mean differences in risk and protective factors, comparing subgroups of LGBTQ+ adolescents to heterosexual/cisgender adolescents, among high school students who participated in the 2018 Dane County Youth Assessment, N=15,467
| Cisgender girls: Sexual minority (n=1,181) vs. heterosexual (ref; n=6,236) |
Cisgender boys: Sexual minority (n=434) vs. heterosexual (ref; n=7,014) |
Transgender & nonbinary (n=602) vs. cisgender (n=14,865) |
|
|---|---|---|---|
| Adjusted mean difference (95% CI), P |
Adjusted mean difference (95% CI), P |
Adjusted mean difference (95% CI), P |
|
| School adult support (reverse scored) | 0.19 (0.07, 0.32), P=.002 | 0.09 (−0.12, 0.30), P=.405 | 0.23 (0.05, 0.42), P=.013 |
| Bullying victimization | 0.32 (0.18, 0.46), P<.001 | 0.45 (0.20, 0.69), P<.001 | 0.75 (0.54, 0.96), P<.001 |
| Family adversity | 0.14 (0.11, 0.17), P<.001 | 0.10 (0.05, 0.15), P<.001 | 0.22 (0.17, 0.27), P<.001 |
Note. LGBTQ+ = lesbian, gay, bisexual, transgender/nonbinary, queer, and related identities. Estimates are based on linear regression models. All estimates are adjusted for the remaining risk/protective factors and potential confounders (grade in school, family financial status, and race/ethnicity). Estimates comparing transgender and nonbinary adolescents to cisgender adolescents are also adjusted for sex assigned at birth.
Table 3.
Adjusted odds ratios for the associations of risk and protective factors with dating violence victimization and sexual assault victimization among high school students who participated in the 2018 Dane County Youth Assessment, N=15,467
| I. Dating violence victimization (students who dated in the past 12 months; n=7,149) | ||
|---|---|---|
| OR (95% CI) | P | |
| School adult support (z score; reverse scored) | 1.11 (1.04, 1.18) | .001 |
| Bullying victimization (z score) | 1.52 (1.44, 1.61) | <.001 |
| Family adversity (z score) | 1.32 (1.25, 1.40) | <.001 |
| II. Sexual assault victimization (all students; n=15,467) | ||
| OR (95% CI) | P | |
| School adult support (z score; reverse scored) | 1.10 (1.02, 1.19) | .010 |
| Bullying victimization (z score) | 1.49 (1.40, 1.58) | <.001 |
| Family adversity (z score) | 1.45 (1.37, 1.53) | <.001 |
Note. OR = odds ratio. CI = confidence interval. Estimates are adjusted for the remaining risk/protective factors, gender/sexual orientation subgroup, race/ethnicity, grade, and family financial status.
Table 4.
Estimated contributions of risk and protective factor inequities to LGBTQ+ dating violence and sexual assault victimization among high school students who participated in the 2018 Dane County Youth Assessment, N=15,467
| I. Dating violence victimization (students who dated in the past 12 months) | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Sexual minority cisgender girls, n=566 (reference: heterosexual cisgender girls, n=3,040) |
Sexual minority cisgender boys, n=151 (reference: heterosexual cisgender boys, n=3,123) |
Transgender & nonbinary adolescents, n=269 (reference: cisgender adolescents, n=6,880) |
|||||||
| Adjusted prevalence difference in percentage points (95% CI) |
Percent of disparity |
P | Adjusted prevalence difference in percentage points (95% CI) |
Percent of disparity |
P | Adjusted prevalence difference in percentage points (95% CI) |
Percent of disparity |
P | |
| Dating violence victimization disparity | 9.49 (4.88, 14.09) | – | <.001 | 1.40 (−5.56, 8.36) | – | .693 | 13.72 (7.04, 20.40) | – | <.001 |
| Estimated reductions in dating violence victimization prevalence (in LGBTQ+ group) and disparity if risk/protective factor inequities were eliminateda | |||||||||
| School adult support | −0.49 (−1.16, 0.18) | 5.2% | .150 | −0.36 (−1.42, 0.70) | b | .505 | −0.17 (−1.68, 1.33) | 1.2% | .819 |
| Bullying victimization | −1.86 (−3.00, −0.72) | 19.6% | .001 | −0.25 (−1.94, 1.43) | b | .767 | −4.22 (−6.78, −1.66) | 30.8% | .001 |
| Family adversity | −1.88 (−3.06, −0.71) | 19.8% | .002 | −0.75 (−2.06, 0.55) | b | .259 | −4.11 (−6.65, −1.58) | 30.0% | .001 |
| All three factors (joint) | −4.21 (−2.55, −5.86) | 44.4% | <.001 | −1.40 (−3.48, 0.69) | b | .189 | −8.53 (−12.02, −5.04) | 62.2% | <.001 |
| Reference group prevalence of dating violence vict.: 22.4% in cisgender heterosexual girls, 18.7% in cisgender heterosexual boys, 21.7% in cisgender adolescents. | |||||||||
| II. Sexual assault victimization (all students) | |||||||||
| Sexual minority cisgender girls, n=1,181 (reference: heterosexual cisgender girls, n=6,236) |
Sexual minority cisgender boys, n=434 (reference: heterosexual cisgender boys, n=7,014) |
Transgender & nonbinary adolescents, n=602 (reference: cisgender adolescents, n=14,865) |
|||||||
| Adjusted prevalence difference in percentage points (95% CI) |
Percent of disparity |
P | Adjusted prevalence difference in percentage points (95% CI) |
Percent of disparity |
P | Adjusted prevalence difference in percentage points (95% CI) |
Percent of disparity |
P | |
| Sexual assault victimization disparity | 11.55 (8.89, 14.22) | – | <.001 | 6.34 (3.40, 9.29) | – | <.001 | 8.86 (5.27, 12.44) | – | <.001 |
| Estimated reductions in sexual assault victimization prevalence (in LGBTQ+ group) and disparity if risk/protective factor inequities were eliminateda | |||||||||
| School adult support | −0.14 (−0.58, 0.30) | 1.2% | .523 | −0.01 (−0.33, 0.30) | 0.2% | .937 | −0.33 (−0.99, 0.33) | 3.7% | .321 |
| Bullying victimization | −1.06 (−1.61, −0.52) | 9.2% | <.001 | −0.37 (−0.92, 0.18) | 5.8% | .183 | −1.89 (−2.92, −0.87) | 21.3% | <.001 |
| Family adversity | −1.33 (−1.98, −0.67) | 11.5% | <.001 | −0.71 (−1.28, 0.14) | 11.2% | .014 | −2.40 (−3.54, −1.27) | 27.1% | <.001 |
| All three factors (joint) | −2.50 (−3.38, −1.61) | 21.6% | <.001 | −1.10 (−1.84, −0.36) | 17.4% | .003 | −4.41 (−5.93, −2.89) | 49.8% | <.001 |
| Reference group prevalence of sexual assault vict.: 9.0% in cisgender heterosexual girls, 2.5% in cisgender heterosexual boys, 7.1% in cisgender adolescents. | |||||||||
Note. CI = confidence interval. All estimates are adjusted for potential confounders (grade in school, family financial status, and race/ethnicity). Estimates comparing transgender and nonbinary adolescents to cisgender adolescents are also adjusted for sex assigned at birth. Effect estimates for individual risk/protective factors assume that the remaining risk/protective factors are held constant. Effect estimates for individual risk/protective factors do not sum to the joint effect for all three factors because the joint effect also incorporates interaction among the risk/protective factors; however, interaction effect estimates were small in magnitude and are not shown here.
In these scenarios, the distributions of risk/protective factors in LGBTQ+ groups are made equal to their observed distributions in the reference groups.
Percent of disparity estimates are suppressed when total effect is non-significant and when the mediation effect and total effect are not in the same direction.
Estimated Reductions in LGBTQ+ Dating Violence and Sexual Assault Victimization Prevalence and Disparities if Risk/Protective Factor Inequities Were Eliminated
To test our primary hypotheses (H1a-d and H2a-d), we used the interventional effects estimators and g-estimation methods proposed by Moreno-Betancur and colleagues (2021). These estimators accommodate exposure-mediator interaction and interrelated mediators (Moreno-Betancur et al., 2021), which was necessary because our conceptual model suggested that risk/protective factors could influence one another. Our primary estimates of interest were expected changes in victimization prevalence in the LGBTQ+ groups if the distribution of a risk/protective factor were changed to its observed distribution in the heterosexual/cisgender group, holding other risk/protective factors constant. These single-factor estimates (H1a-c and H2a-c) suggest the potential impact of eliminating the influence of heterosexism/cissexism on each risk/protective factor, which is informative in prioritizing specific intervention targets. However, single-factor estimates conservatively exclude downstream effects on other risk/protective factors (Moreno-Betancur et al., 2021), e.g., effects of school adult support on bullying victimization. Thus, we also estimated the joint effects of changing all risk/protective factors (H1d and H2d). Notably, these estimates (and the term “interventional effects”) refer to hypothetical changes in risk/protective factors, which may not be feasible in real-world interventions (Moreno-Betancur et al., 2021). We fit six separate models, one for each gender-stratified group and each outcome. Online Resource 1 presents details of these analyses.
Interventional effects results are presented in Table 4. There were no significant effects associated with school adult support (H1a; H2a), but eliminating inequities in bullying victimization (H1b; H2b) and family adversity (H1c; H2c) were associated with significant reductions in LGBTQ+ dating violence and sexual assault victimization, each accounting for 9–31% of observed disparities. Effects varied by LGBTQ+ subgroup, as discussed below.
Estimated Reductions in Dating Violence Victimization
Eliminating inequities in bullying victimization (H1b) was estimated to reduce dating violence victimization by 1.86 pp in sexual minority cisgender girls (19.6% of disparity; P=.001) and by 4.22 pp in transgender/nonbinary adolescents (30.8% of disparity; P=.001). Similarly, eliminating inequities in family adversity (H1c) was estimated to reduce dating violence victimization by 1.88 pp in sexual minority cisgender girls (19.8% of disparity; P=.002) and 4.11 pp in transgender/nonbinary adolescents (30.0% of disparity; P=.001). Jointly eliminating inequities in all three risk/protective factors (H1d) was estimated to reduce dating violence victimization by 4.21 pp in sexual minority cisgender girls (44.4% of disparity; P<.001) and 8.53 pp in transgender/nonbinary adolescents (62.2% of disparity; P<.001). As a sensitivity analysis, dating violence analyses were also performed in the full sample, i.e., not restricted to students who had dated in the past 12 months (see Online Resource 3). All effects that were significant in the primary dating violence analyses were also significant in the full sample sensitivity analysis.
Estimated Reductions in Sexual Assault Victimization
Eliminating inequities in bullying victimization (H2b) was estimated to reduce sexual assault victimization by 1.06 pp in sexual minority cisgender girls (9.2% of disparity; P<.001) and 1.89 pp in transgender/nonbinary adolescents (21.3% of disparity; P<.001). Eliminating inequities in family adversity (H2c) was estimated to reduce sexual assault victimization by 1.33 pp in sexual minority cisgender girls (11.5% of disparity; P<.001), 0.71 pp in sexual minority cisgender boys (11.2% of disparity; P=.014), and 2.40 pp in transgender/nonbinary adolescents (27.1% of disparity; P<.001). Jointly eliminating inequities in all three factors (H2d) was estimated to reduce sexual assault victimization by 2.50 pp in sexual minority cisgender girls (21.6% of disparity; P<.001), 1.10 pp in sexual minority cisgender boys (17.4% of disparity; P=.003), and 4.41 pp in transgender/nonbinary adolescents (49.8% of disparity; P<.001).
Discussion
This study estimated the impact of sexual orientation and gender modality inequities in school adult support, bullying victimization, and family adversity on the prevalence of dating violence and sexual assault victimization in LGBTQ+ adolescents, providing crucial information to guide violence prevention efforts with this population. Prior studies had linked these factors, and other characteristics of school-based and family relationships, to dating violence and sexual assault victimization in LGBTQ+ adolescents—but few had assessed the impact of heterosexism- and cissexism-related inequities in these domains. Moreover, prior studies focused on individual-level risk, without clear implications for population-level interventions such as laws, school policies, or other institutional practices. By quantifying the potential population-level impacts of achieving equity in each risk/protective factor, we identified two key inequities—bullying victimization and family adversity—on which violence prevention efforts with LGBTQ+ adolescents should focus. Further, our gender-stratified analyses suggest directions for violence prevention efforts with specific subgroups of LGBTQ+ adolescents.
Our results suggest that dating violence and sexual assault victimization in LGBTQ+ adolescents are partly driven by disproportionate bullying victimization relative to heterosexual/cisgender peers. Accordingly, schools should implement best practices for promoting LGBTQ+ students’ safety, including Gender & Sexuality Alliance groups ("GSAs"; Marx & Kettrey, 2016) and policies prohibiting anti-LGBTQ+ harassment (Hall, 2017). Further, private all-gender restrooms may discourage anti-LGBTQ+ bullying by signaling support for LGBTQ+ students and allowing transgender/nonbinary students to avoid harassment in gender-segregated restrooms (Porta et al., 2017). Though we did not find that eliminating inequities in school adult support would reduce LGBTQ+ dating violence or sexual assault victimization if bullying victimization and family adversity were held constant, school staff can be key allies in reducing bullying victimization of LGBTQ+ students (Hall, 2017).
Our results further suggest that the impacts of heterosexism and cissexism on family adversity play a key role in sexual assault and dating violence victimization for LGBTQ+ adolescents. For LGBTQ+ adolescents, family adversity may result from, or be exacerbated by, family members’ heterosexist/cissexist attitudes (Shelton & Bond, 2017). Given that structural heterosexism and cissexism can be considerable stressors for LGBTQ+ adolescents and their families (Kuper et al., 2022), heterosexism- and cissexism-related stress could also contribute to adversity even when family members do not hold heterosexist/cissexist beliefs, though this process remains understudied. Thus, violence prevention efforts should include advocacy against structural heterosexism and cissexism, particularly laws, policies, and practices targeting families of LGBTQ+ young people (e.g., classifying gender-affirming medical care as child abuse; Kuper et al., 2022). Also indicated are psychosocial interventions that help LGBTQ+ adolescents’ families cope with heterosexism- and cissexism-related stress (e.g., Lozano et al., 2022).
We found notable variations in the roles of risk/protective factor inequities across subgroups of LGBTQ+ adolescents. In particular, we found that eliminating inequities in bullying victimization could significantly reduce dating violence and sexual assault victimization in sexual minority cisgender girls and transgender/nonbinary adolescents—but not cisgender sexual minority boys. We found that eliminating inequities in family adversity could significantly reduce sexual assault victimization in sexual minority boys, but this effect accounted for just 11% of the sexual assault victimization disparity between sexual minority and heterosexual cisgender boys. These null and limited effects were surprising, particularly given that sexual minority cisgender boys reported significantly more bullying victimization and family adversity than heterosexual peers. Accordingly, while addressing bullying victimization and family adversity inequities may have other benefits for cisgender sexual minority boys, these strategies may not be adequate for preventing dating violence and sexual assault in this group. Qualitative and quantitative studies describing sexual minority boys’ experiences with dating violence and sexual assault (e.g., perpetrators, settings, tactics) will be an important step towards identifying intervention strategies for this population. We recommend including both cisgender and transgender boys in this work, attending to similarities and differences between these groups.
The present study highlights the value of interventional effects analysis in LGBTQ+ health disparities research. Interventional effects analysis can identify inequities in risk and protective factors that, if addressed, would meaningfully reduce population-level disparities in health outcomes (Jackson & VanderWeele, 2019). By using group differences in risk/protective factors as a proxy for the effects of heterosexism and cissexism, this approach captures the consequences of multiple forms of heterosexism and cissexism, including forms that are difficult to measure (Krieger, 2020). This work should occur in concert with studies addressing specific forms of heterosexism and cissexism (e.g., stigmatizing school policies; Murchison et al., 2019), which may also be key targets for prevention efforts.
Our findings should be interpreted in light of several limitations. First, because we used cross-sectional data, we were unable to determine whether the hypothesized risk/protective factors temporally preceded dating violence and sexual assault victimization. Consequently, results should be interpreted with caution and should not be assumed to reflect causal effects. Our research suggests that peer victimization is prospectively associated with elevated risk of dating violence victimization (Murchison et al., 2022), supporting the temporal ordering proposed here; future studies should use longitudinal data to make similar assessments for the remaining risk/protective factors and outcomes. Second, we focused on inequities in risk/protective factors as a proxy for the impact of heterosexism/cissexism. While this approach may be more comprehensive than measuring heterosexism/cissexism directly, it cannot identify the specific forms of heterosexism and cissexism most relevant to each risk/protective factor, nor the processes by which heterosexism/cissexism impact school-based and family relationships. Third, our interventional effects analysis modeled hypothetical scenarios in which risk/protective factor inequities were eliminated; such changes are likely not achievable through any single policy or practice—or even a set of multiple policy and practice changes. Accordingly, estimates are intended to guide prioritization of interventions addressing specific risk/protective factors, not to predict the impact of real-world interventions. Fourth, our sample was drawn from a single U.S. county, and results may not be generalizable to other geographic areas. Fifth, most participants were White and non-Hispanic; because racism may intersect with anti-LGBTQ+ prejudice to produce unique disruptions for LGBTQ+ adolescents of color, particularly in school-based relationships (Snapp et al., 2014), it is not clear whether our conclusions apply across race and ethnicity. Sixth, to achieve adequate power, we grouped asexual, bisexual, gay, lesbian, pansexual, and questioning students together for analysis. Prior research suggests that, compared with adolescents reporting only same-gender partners, adolescents with partners of multiple genders report more exposure to a range of dating violence risk factors (Petit et al., 2021). Accordingly, there is a need for further research on risk processes specific to plurisexual (e.g., bisexual, pansexual) adolescents, including those stemming from monosexism (prejudice and structural disadvantage against people attracted to multiple genders). Finally, we used brief measures of sexual assault and dating violence, which may be less reliable or valid than detailed measures. These limitations are counterbalanced by a large sample, ensuring adequate sample sizes in the three LGBTQ+ groups, and a population-based sampling strategy with low non-response, limiting the risks of sampling bias and selection bias.
Conclusion
This study is among the first to quantify the potential contributions of inequities in specific risk/protective factors to dating violence and sexual assault victimization among LGBTQ+ adolescents. We found that eliminating inequities in bullying victimization and family adversity between LGBTQ+ and cisgender/heterosexual adolescents could considerably reduce dating violence and sexual assault victimization in LGBTQ+ adolescents. Accordingly, efforts to prevent dating violence and sexual assault in LGBTQ+ adolescents should include advocacy for policies and practices that mitigate the effects of heterosexism- and cissexism on school-based and family relationships—including school policies that promote safety for LGBTQ+ students as well as supports to help LGBTQ+ adolescents’ families cope with heterosexism and cissexism.
Supplementary Material
Acknowledgments
The authors are grateful to the Dane County Youth Commission, Dane County Department of Human Services, and participating students for providing the data used in this study. They also wish to thank the peer reviewers who provided valuable feedback on this manuscript.
Funding
G.R. Murchison was supported by the National Institute of Mental Health under award number T32-MH020031 (PI: Kershaw). S.B. Austin was supported by US Maternal and Child Health Bureau, Health Resources and Services Administration grant T76-MC00001.
Footnotes
Ethics Approval
As a secondary analysis of anonymous data, this study was determined not to be human subjects research by the Harvard T.H. Chan School of Public Health IRB.
Conflicts of Interest
None.
Consent to Participate
All Dane County Youth Assessment participants gave voluntary informed consent/assent. Parents and guardians were informed of the survey and had the option to decline consent.
References
- Andersen JP, & Blosnich J (2013). Disparities in adverse childhood experiences among sexual minority and heterosexual adults: Results from a multi-state probability-based sample. PloS One, 8(1), Article e54691. 10.1371/journal.pone.0054691 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ashley F. (2022). ‘Trans’ is my gender modality: A modest terminological proposal. In Erickson-Schroth L (Ed.), Trans bodies, trans selves: A resource by and for transgender communities (pp. 22). Oxford University Press. [Google Scholar]
- Centers for Disease Control and Prevention. (2022). 1991-2019 High School Youth Risk Behavior Survey data. Retrieved February 7, 2022 from http://nccd.cdc.gov/youthonline/ [Google Scholar]
- Craig SL, Austin A, Levenson J, Leung VWY, Eaton AD, & D’Souza SA (2020). Frequencies and patterns of adverse childhood events in LGBTQ+ youth. Child Abuse and Neglect, 107, Article 104623. 10.1016/j.chiabu.2020.104623 [DOI] [PubMed] [Google Scholar]
- Dane County Youth Commission. (2018). 2018 Dane County Youth Assessment. https://www.dcdhs.com/documents/pdf/Youth/YouthCommission/2018-hs.pdf [Google Scholar]
- Espelage DL (2014). Ecological theory: Preventing youth bullying, aggression, and victimization. Theory Into Practice, 53(4), 257–264. 10.1080/00405841.2014.947216 [DOI] [Google Scholar]
- Espelage DL, & Holt MK (2001). Bullying and victimization during early adolescence. Journal of Emotional Abuse, 2(2-3), 123–142. 10.1300/J135v02n02_08 [DOI] [Google Scholar]
- Exner-Cortens D, Eckenrode J, & Rothman E (2013). Longitudinal associations between teen dating violence victimization and adverse health outcomes. Pediatrics, 131(1), 71–78. 10.1542/peds.2012-1029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Felix ED, Holt MK, Nylund-Gibson K, Grimm RP, Espelage DL, & Green JG (2019). Associations between childhood peer victimization and aggression and subsequent victimization and aggression at college. Psychology of Violence, 9(4), 451–460. 10.1037/vio0000193 [DOI] [Google Scholar]
- Finkelhor D. (2008). Childhood victimization: Violence, crime, and abuse in the lives of young people. Oxford University Press. 10.1093/acprof:oso/9780195342857.003.0003 [DOI] [Google Scholar]
- Hall W. (2017). The effectiveness of policy interventions for school bullying: A systematic review. Journal of the Society for Social Work and Research, 8(1), 45–69. 10.1086/690565 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hamby S, Finkelhor D, & Turner H (2012). Teen dating violence: Co-occurrence with other victimizations in the National Survey of Children's Exposure to Violence (NatSCEV). Psychology of Violence, 2(2), 111–124. 10.1037/a0027191 [DOI] [Google Scholar]
- Hébert M, Daspe M-È, Lapierre A, Godbout N, Blais M, Fernet M, & Lavoie F (2019). A meta-analysis of risk and protective factors for dating violence victimization: The role of family and peer interpersonal context. Trauma, Violence, & Abuse, 20(4), 574–590. 10.1177/1524838017725336 [DOI] [PubMed] [Google Scholar]
- Hipwell AE, Stepp SD, Xiong S, Keenan K, Blokland A, & Loeber R (2014). Parental punishment and peer victimization as developmental precursors to physical dating violence involvement among girls. Journal of Research on Adolescence, 24(1), 65–79. 10.1111/jora.12016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jackson JW, & VanderWeele TJ (2019). Intersectional decomposition analysis with differential exposure, effects, and construct. Social Science and Medicine, 226, 254–259. 10.1016/j.socscimed.2019.01.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Johns MM, Lowry R, Andrzejewski J, Barrios LC, Demissie Z, McManus T, Rasberry CN, Robin L, & Underwood JM (2019). Transgender identity and experiences of violence victimization, substance use, suicide risk, and sexual risk behaviors among high school students - 19 states and large urban school districts, 2017. MMWR: Morbidity and Mortality Weekly Report, 68(3), 67–71. 10.15585/mmwr.mm6803a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Johns MM, Lowry R, Haderxhanaj LT, Rasberry CN, Robin L, Scales L, Stone D, & Suarez NA (2020). Trends in violence victimization and suicide risk by sexual identity among high school students - Youth Risk Behavior Survey, United States, 2015-2019. MMWR Supplements, 69(1), 19–27. 10.15585/mmwr.su6901a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kilpatrick DG, Ruggiero KJ, Acierno R, Saunders BE, Resnick HS, & Best CL (2003). Violence and risk of PTSD, major depression, substance abuse/dependence, and comorbidity: Results from the National Survey of Adolescents. Journal of Consulting and Clinical Psychology, 71(4), 692–700. 10.1037/0022-006X.71.4.692 [DOI] [PubMed] [Google Scholar]
- Koenig BR (2006). Focus on school engagement - Dane County Youth Assessment. Dane County Youth Commission. https://www.dcdhs.com/documents/pdf/Youth/YouthCommission/youth-2005-engage.pdf [Google Scholar]
- Kosciw J, Clark C, Truong N, & Zongrone A (2020). The 2019 National School Climate Survey: The experiences of lesbian, gay, bisexual, transgender, and queer youth in our nation’s schools. GLSEN. https://www.glsen.org/sites/default/files/2021-04/NSCS19-FullReport-032421-Web_0.pdf [Google Scholar]
- Krieger N. (2020). Measures of racism, sexism, heterosexism, and gender binarism for health equity research: From structural injustice to embodied harm–an ecosocial analysis. Annual Review of Public Health, 41, 37–62. 10.1146/annurev-publhealth-040119-094017 [DOI] [PubMed] [Google Scholar]
- Kuper LE, Cooper MB, & Mooney MA (2022). Supporting and advocating for transgender and gender diverse youth and their families within the sociopolitical context of widespread discriminatory legislation and policies. Clinical Practice in Pediatric Psychology, 10(3), 336–345. 10.1037/cpp0000456 [DOI] [Google Scholar]
- Lozano A, Estrada Y, Tapia MI, Dave DJ, Marquez N, Baudin S, & Prado G (2022). Development of a family-based preventive intervention for Latinx sexual minority youth and their parents. Cultural Diversity and Ethnic Minority Psychology, 28(2), 227–239. 10.1037/cdp0000506 [DOI] [PubMed] [Google Scholar]
- Marx RA, Hatchel T, Mehring CB, & Espelage DL (2021). Predictors of sexual victimisation and suicidal ideation among transgender and gender-nonconforming adolescents. Psychology & Sexuality, 12(1-2), 79–95. 10.1080/19419899.2019.1690034 [DOI] [Google Scholar]
- Marx RA, & Kettrey HH (2016). Gay-straight alliances are associated with lower levels of school-based victimization of LGBTQ+ youth: A systematic review and meta-analysis. Journal of Youth and Adolescence, 45(7), 1269–1282. 10.1007/s10964-016-0501-7 [DOI] [PubMed] [Google Scholar]
- Moreno-Betancur M, Moran P, Becker D, Patton GC, & Carlin JB (2021). Mediation effects that emulate a target randomised trial: Simulation-based evaluation of ill-defined interventions on multiple mediators. Statistical Methods in Medical Research, 30(6), 1395–1412. 10.1177/0962280221998409 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Morris AM, Mrug S, & Windle M (2015). From family violence to dating violence: Testing a dual pathway model. Journal of Youth and Adolescence, 44(9), 1819–1835. 10.1007/s10964-015-0328-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Murchison GR, Agénor M, Reisner SL, & Watson RJ (2019). School restroom and locker room restrictions and sexual assault risk among transgender youth. Pediatrics, 143(6), Article e20182902. 10.1542/peds.2018-2902 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Murchison GR, Austin SB, Reisner SL, & Chen JT (2022). Middle school psychological distress and sexual harassment victimization as predictors of dating violence involvement. Journal of Interpersonal Violence. Advance online publication. 10.1177/08862605221135166 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Petit M-P, Blais M, & Hébert M (2021). Dating violence victimization disparities across sexual orientation of a population-based sample of adolescents: An adverse childhood experiences perspective. Psychology of Sexual Orientation and Gender Diversity. Advance online publication. 10.1037/sgd0000518 [DOI] [Google Scholar]
- Porta CM, Gower AL, Mehus CJ, Yu X, Saewyc EM, & Eisenberg ME (2017). “Kicked out”: LGBTQ youths' bathroom experiences and preferences. Journal of Adolescence, 56, 107–112. 10.1016/j.adolescence.2017.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- R Core Team. (2017). R: A language and environment for statistical computing. (Version 3.4.4) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/ [Google Scholar]
- Ross-Reed DE, Reno J, Peñaloza L, Green D, & FitzGerald C (2019). Family, school, and peer support are associated with rates of violence victimization and self-harm among gender minority and cisgender youth. Journal of Adolescent Health, 65(6), 776–783. 10.1016/j.jadohealth.2019.07.013 [DOI] [PubMed] [Google Scholar]
- Shelton J, & Bond L (2017). “It just never worked out”: How transgender and gender expansive youth understand their pathways into homelessness. Families in Society, 98(4), 284–291. 10.1606/1044-3894.2017.98.33 [DOI] [Google Scholar]
- Snapp SD, Hoenig JM, Fields A, & Russell ST (2014). Messy, butch, and queer: LGBTQ youth and the school-to-prison pipeline. Journal of Adolescent Research, 30(1), 57–82. 10.1177/0743558414557625 [DOI] [Google Scholar]
- Stephenson R, Darbes LA, Rosso MT, Washington C, Hightow-Weidman L, Sullivan P, & Gamarel KE (2022). Perceptions of contexts of intimate partner violence among young, partnered gay, bisexual and other men who have sex with men in the United States. Journal of Interpersonal Violence, 37(15-16), NP12881–NP12900. 10.1177/08862605211001472 [DOI] [PubMed] [Google Scholar]
- Stroem IF, Goodman K, Mitchell KJ, & Ybarra ML (2021). Risk and protective factors for adolescent relationship abuse across different sexual and gender identities. Journal of Youth and Adolescence, 50(8), 1521–1536. 10.1007/s10964-021-01461-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Suldo SM, Friedrich AA, White T, Farmer J, Minch D, & Michalowski J (2009). Teacher support and adolescents' subjective well-being: A mixed-methods investigation. School Psychology Review, 38(1), 67–85. 10.1080/02796015.2009.12087850 [DOI] [Google Scholar]
- Tillyer MS, Wilcox P, & Gialopsos BM (2010). Adolescent school-based sexual victimization: Exploring the role of opportunity in a gender-specific multilevel analysis. Journal of Criminal Justice, 38(5), 1071–1081. 10.1016/j.jcrimjus.2010.07.010 [DOI] [Google Scholar]
- van Buuren S, & Groothuis-Oudshoorn K (2011). Mice: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45(3), 1–67. 10.18637/jss.v045.i03 [DOI] [Google Scholar]
- Watson RJ, Fish JN, Denary W, Caba A, Cunningham C, & Eaton LA (2021). LGBTQ state policies: A lever for reducing SGM youth substance use and bullying. Drug and Alcohol Dependence, 221, Article 108659. https://doi.org/ 10.1016/j.drugalcdep.2021.108659 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yan FA, Howard DE, Beck KH, Shattuck T, & Hallmark-Kerr M (2009). Psychosocial correlates of physical dating violence victimization among Latino early adolescents. Journal of Interpersonal Violence, 25(5), 808–831. 10.1177/0886260509336958 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.