Adolescent interpersonal violence, including various forms of sexual violence (SV), homophobic name-calling (HNC), and teen dating violence (TDV), are public health problems that harm many adolescents in the United States (U.S.; Basile et al., 2020). These forms of violence often co-occur among adolescents (Espelage et al., 2021a; Humphrey and Vaillancourt, 2020). SV varies from verbal sexual harassment (SH) to unwanted touching to rape and can occur in-person or online (Basile et al., 2014). HNC includes the use of verbal epithets (e.g., fag, gay) to disparage youth perceived to be gender non-conforming or gender/sexual minorities (Meyer, 2008). TDV includes physical, sexual, psychological, or stalking aggression against a dating partner (Centers for Disease Control and Prevention, 2020). Adolescent violence is preventable, but successful prevention requires an understanding of perpetrator characteristics. Although research on adolescent violence perpetration has increased in recent years, little is known about perpetration across gender, racial/ethnic, and sexual identities.

Gender identity and expression, race/ethnicity, and sexual identity are distinct yet overlapping cultural and social identities. A framework of intersectionality (Bowleg, 2012; Warner, 2008) is necessary for examining how multiple social identities, in a reflection of larger systems of privilege and oppression, contribute to differences in interpersonal violence perpetration. In a sample of high school students in Colorado, the present study examined the prevalence of SV, TDV, and HNC perpetration among adolescents based on gender identity, race/ethnicity, sexual identity, and intersecting social identities. Understanding peer violence perpetration within and as a product of systems of privilege and oppression can inform efforts to create safe, affirming schools for all students regardless of personal demographic characteristics.

Racism, sexism, homophobia, and trans-prejudice are societal systems that place youth at high risk for internalized oppression and interpersonal stressors; individuals with multiple oppressed social identities (e.g., sexual/gender minority person of color) are at particularly high risk (Crenshaw, 1991; Nadal et al., 2021; Garofalo et al., 2006; Sokoloff and Dupont, 2005). While not considered direct risk factors for adolescent violence perpetration, race and ethnicity are seen as “markers for a constellation of external and malleable social contexts that are differentially allocated by racial/ethnic status in American Society” (Sampson et al., 2005, p. 224). Many scholars point to socioeconomic inequality embedded in racially segregated neighborhoods where individuals are differentially exposed to key risk and protective factors that contribute to differences in violence perpetration rates across race/ethnicity (Kaufman and Cooper, 2001; Sampson et al., 2005).

Historically excluded groups (e.g., women, people of color, transgender individuals) are often targets of repeated disparaging messages about their identity groups as they interact with peers, educators, and family (Nadal, 2018). These encounters, combined with systemic oppression (e.g., discrimination) and stereotypes about the groups they belong to (David, 2014), can lead to adolescents from historically excluded groups internalizing negative feelings and thoughts about themselves and their own social identity (Nadal et al., 2021). For youth experiencing identity-related internalized oppression and interpersonal stressors, the enforcement of these norms—as modeled by peers, family, social and broader societal forces—can be associated with violence perpetration (Garofalo et al., 2006; Nadal et al., 2021; Sokoloff and Dupont, 2005).

Although the field has documented the overlap between adolescent SV, HNC, and TDV perpetration and identified shared and unique risk and protective factors in general population youth studies (Espelage et al., 2015; Humphrey and Vaillancourt, 2020; Ybarra and Thompson, 2018; Wincentak et al., 2017), the field knows little about variations in perpetration by gender identity, race/ethnicity, and sexual identity given challenges with identifying large enough samples or lack of demographic data. To address this gap, the current study examined perpetration rates between and across gender, race/ethnic, and sexual identities. This study represents an important starting place to better understand perpetrators of specific forms of violence so that prevention efforts can be adapted to improve their effectiveness.

This study examined multiple forms of adolescent interpersonal violence perpetration across cultural and social identities. While varying perpetration rates were noted across race/ethnicity and sexual identity, boys in most intersectional identities reported greater perpetration than girls, except for TDV, where the trend was reversed. The finding that African American students had higher rates of forced sexual contact perpetration than White students may be a valid finding, or it may be a case of Simpson’s paradox - the association between two variables in the population is ameliorated or reversed in all or most subpopulations (Goltz and Smith, 2010). Possible causes are unequal distribution of a confounding variable within subpopulations and low frequency events. Both apply to this study sample. Boys had higher forced sexual contact perpetration rates, albeit non-significantly, than girls (2% vs. 1%); the sample of African American students was disproportionately composed of boys (67%), unlike the gender distribution of White students (50%). Paired with the low frequency of occurrences (n = 5 for African American students), the unequal gender distribution could contribute to Simpson’s paradox when comparing perpetration rates aggregated by identity to rates disaggregated by intersectionality.

Another possible explanation for the findings for African American students is found in minority stress theory. Extant literature suggests that minority stress — chronic stress resulting from actual, perceived, or anticipated interpersonal prejudice and discrimination –increases the likelihood of violence perpetration (Edwards and Sylaska, 2013). Minority stress experiences are associated with psychological symptoms such as depression, anxiety, and low self-esteem (Brooks, 1981; Meyer, 2003), all previously identified risk factors of violence perpetration. For instance, Edwards and Sylaska (2013) found that, among LGBTQ college students, TDV perpetration was associated with identity concealment, internalized homophobia, and sexual orientation-focused victimization. Also, the prevalence findings for this study are generally consistent with previous literature (Clear et al., 2014; Fedina et al., 2016; Valido et al., 2021a).

This study builds on previous studies by assessing perpetration rates across intersections of gender identity, race/ethnicity, and sexual identity (Birkett and Espelage, 2015; Clear et al., 2014; Edwards and Sylaska, 2013; Fedina et al., 2016; Wincentak et al., 2017). Given findings that boys reported higher perpetration rates than girls across violence types, a continued emphasis on prevention efforts that focus on boys and address hypermasculinity, patriarchy, misogyny, and homophobia is needed. Prevention strategies that seek to change social norms around violence and gender, such as bystander approaches or approaches focused on engaging men and boys as allies in violence prevention (Basile et al., 2016; Niolon et al., 2017), coupled with antiracist practices to create intersectional approaches (Garnett et al., 2014), are necessary to comprehensively address violence against and by adolescents with historically excluded intersectional identities.

However, there is a lack of evidence-based prevention approaches specifically tailored for racial/ethnic, sexual, and gender minority adolescents. Future studies are needed to examine the effectiveness of prevention efforts for youth with historically excluded intersectional identities and to evaluate adaptations of existing evidence-based approaches for their effectiveness among students with diverse cultural and social identities. Prevention strategies must center intersectional experiences and the structural systems of both racism and sexism (Bowleg, 2012) to address the larger structural conditions and systems of power that contribute to and perpetuate violence against and by students with historically excluded intersectional identities, such as school policies or social norms that discriminate against some students. Future research with larger, representative samples that allow an examination of violence perpetration by youth with historically excluded intersectional identities, including socioeconomic status, in greater detail would be beneficial for informing the content of prevention efforts. Additional perpetration-specific data on gender, sexual, and racial/ethnic minority youth would be useful for developing prevention strategies, shifting away from the existing practice of relying on victimization data to understand these groups and inform prevention strategies.

Although the current study fills essential gaps in the literature, several limitations exist. First, results are descriptive and employ cross-sectional data from a non-representative convenience sample of Colorado high schools, limiting the generalizability of findings. Compared to state demographics, the study sampled a higher percentage of female, Hispanic, multiracial, and AI/ANstudents and a lower percentage of male, African American, White, and Asian American/NHPI (National Center for Education Statistics, 2022). Additionally, the sample size for some intersections was not large enough to be included in the contrasts analysis. Second, victim characteristics were not assessed. It is possible that the high rates observed among some of the groups were due to other participant characteristics not examined in this study. Future longitudinal studies that measure gender, racial/ethnic, and sexual identity and assess victim characteristics would improve understanding of risk factors and potential intervention points for perpetration. Third, violence perpetration was self-reported, which might introduce bias given social desirability; however, other informants (e.g., teacher, official records) would also introduce bias. Our intentional assessment of violence perpetration through numerous behaviorally-specific questions was employed to increase chances of disclosure. Sexual orientation was also captured through self-reported sexual identity alone; adolescents who engage in same-sex sexual behavior or attraction but do not identify as LGBTQ were not captured in this sample. Further, the gender variable was intended to capture gender identity rather than sex assigned at birth, but it cannot be confirmed whether all respondents interpreted the question and answered as such. Fourth, the HNC variable did not ask students about their intent behind the use of homophobic epithets; some students who identify as LGBTQ may use homophobic slurs as a form of reclamation (e.g., change in the meaning of queer; Hess, 2020; Rand, 2014) rather than bullying. Future studies could include more detailed measures that capture students’ intent for using homophobic language. Finally, the TDV measure was limited to three items representing verbal, physical, and sexual TDV; items were combined into one overall measure, so findings may mask differences across different TDV types. Future studies could examine different types of TDV separately utilizing multiple items.

Adolescent interpersonal violence perpetration is a critical public health problem. The present study adds to the dearth of literature on perpetration by students of diverse social/cultural identities. Boys reported perpetration most commonly for most types of violence, which held true across sexual identity and race/ethnicity. These findings contribute to the field’s understanding of how students from historically excluded social/cultural communities, whom literature has often identified as frequent victims of interpersonal violence, are also sometimes perpetrators. It is important to understand these findings in the context of the structural and social power hierarchies that exist in society. More data and research are needed in this area to advance prevention efforts, such as adapting proven prevention approaches that have not yet been tested in diverse samples, to get closer to societal goals of making school a safe haven for all students regardless of their social/cultural identities.