Examining Sexual Identity in the Context of General Strain Theory, School-Based Strain, Depression, and Delinquency

ABSTRACTThe current study expands on existing scholarship by exploring General Strain Theory to understand the relationships between school-based strain (feeling isolated at school and having trouble with people at school) and delinquency while considering the mediating role of depression among gay/lesbian, bisexual, mostly heterosexual, and heterosexual youth. Using the restricted Add Health data (N = 10,897) findings indicated that gay/lesbian youth report more delinquency than heterosexual youth and bisexual youth, yet mostly heterosexual youth report the highest levels of delinquency. Inconsistent with General Strain Theory’s predictions, depression did not mediate the relationship between school-based strain and delinquency when controlling for sexual identity. Further, our results suggested that certain school-based strains are predictive of delinquency for certain sexual identity groups (heterosexual and bisexual) but not others (mostly heterosexual and gay/lesbian). This work highlights the need for intersectional theoretical extensions, policy applications, and more research on strains, emotions, and delinquency to unravel the nuances of these relationships among the LGBTQ+ community.KEYWORDS: Delinquencygendergeneral strain theoryintersectionalityqueer criminologynegative emotionssexual identityLGBTQ Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1. This work focuses on sexual identity, which is one dimension of a person’s sexual orientation. The other dimensions of sexual orientation include sexual attraction and sexual behavior (Fricke & Sironi, Citation2017; Priebe & Svedin, Citation2013). It is important to not conflate attraction or behavior with identity, and thus the findings from this study apply solely to sexual identity.2. Button (Citation2015, p. 489, Citation2016, p. 546) and Button and Worthen’s (Citation2014, p. 280) research used 5 items to construct a victimization scale: (1) missed school during the past month because they felt unsafe, (2) been threatened with a weapon at school during the past year, (3) had property stolen or damaged at school during the past year, (4) been in a physical fight during the past year, and (5) been injured in a physical fight which had to be treated by a nurse or doctor during the past year. Button and Worthen (Citation2017, p. 815) deviated from this by considering items two, three, and four listed above individually. All five of these items measure extreme types of school-based strains.3. However, Cesare Lombroso, one of the earliest criminologists, had a problematic obsession with LGBTQ+ people. Thus, criminology also has roots in the criminalization and demonization of queerness (Groombridge, Citation1999; Woods, Citation2015).4. The genetic sample includes biological siblings (Chen and Harris, Citation2020).5. As per Chen and Harris (Citation2020), we use the cross-sectional weight for Wave 2, which is the wave of our outcome variables. We also remove all missing cases on the weight variable (WSWGT2) as per their instructions.6. All items used in the sales along with range, mean, and Cronbach’s alpha are contained in the Appendix.7. Scholars note that “coming out” is a process that can start in childhood but take many years to reach full identity integration (Coleman, Citation1982). For example, in a 2013 study, the median age LGBT adults reported realizing they might be “something other than heterosexual or straight” was age 12, the median age at which they “knew for sure” was age 17, and the median age at which they first shared that information was 20 (Pew Research Center, Citation2013). Thus, using a retrospective measure from Wave IV to identify sexual identity has certain advantages.8. Respondents were able to choose from six options, including (1) 100% heterosexual, (2) mostly heterosexual but somewhat attracted to people of your own sex, (3) bisexual, (4) mostly homosexual but somewhat attracted to people of the opposite sex, (5) 100% homosexual, and (6) not sexually attracted to either males or females. Given that the term “homosexual” is outdated, we substitute the word “gay/lesbian” throughout this manuscript. As discussed, those who reported they were not sexually attracted to either males or females were dropped from the analysis.9. We ran supplementary analyses (available upon request) to consider the impact of bad temper on delinquency in our models and subgroup models and did not find bad temper to be significantly related to delinquency in any of our models.10. For more discussion on the appropriate measurement of sexual identity and gender identity see Tate et al. (Citation2013) and Truman et al. (Citation2019).11. To date, much of the literature on GST and LGBQ people has centered on the experiences of youth (Button, Citation2015, Citation2016, Citation2019; Button & Worthen, Citation2014, Citation2017). Future studies should explore the experiences that LGBQ adults have with strain, negative emotions, and criminal behavior, as the theory’s applicability may differ.12. There were 2,673 cases missing sexual identity, race, ethnicity, gender, and/or age. These cases were removed prior to imputation to avoid imputing identity characteristics.