Sexual Orientation Differences in Gun Ownership and Beliefs About Gun Safety Policy,General Social Survey 2010–2016

Introduction

Access to a gun in the home is a crucial predictor of suicide and homicide (Anglemyer et al. 2014). Whether there are sexual orientation differences in access to a gun in the home and gun ownership is largely unknown, but may represent unrecognized differences in distal risk for gun-related violence. This is a significant knowledge gap given sexual minorities' (i.e., lesbians, gay men, and bisexuals) elevated risk for self-directed violence and intimate partner violence (Edwards et al. 2015; Haas et al. 2010). Relatedly, no studies of which we are aware have investigated gun-specific policy beliefs among sexual minorities. Thus, we drew upon information from the General Social Survey (GSS) to investigate sexual orientation-related patterns of gun ownership and beliefs about gun safety policy.

Materials and Methods

We used cross-sectional data from multiple years of the GSS (2010–2016), a biennial survey that uses a multistage sampling design to obtain a representative sample of noninstitutionalized adults (≥18 years old) in the United States. In 2008, the GSS added a question to gather respondents' self-identified sexual orientation. However, military veteran status, a strong modifier of gun ownership (Blosnich et al. 2013), was not collected in that cycle; thus, we analyzed data from the GSS collected in 2010, 2012, 2014, and 2016. Response rates for the GSS ranged from 71.4% (in 2012) to 61.3% (in 2016). We focused on three dependent variables: (1) beliefs about gun safety policy [“Would you favor or oppose a law which would require a person to obtain a police permit before he or she could buy a gun?”], (2) presence of household-based guns [“Do you happen to have in your home or garage any guns or revolvers?”], and (3) among those reporting household guns, whether the respondent personally owned the gun [“Do any of these guns personally belong to you?”]. Analyses included respondents who were administered the following questionnaire items: gun-related outcomes, sexual orientation identity (lesbian, gay, bisexual, and heterosexual), and an item assessing military veteran status (N = 4554). Additional demographic characteristics considered included gender, race/ethnicity, age, educational attainment, household size, urbanicity, and political views. We dichotomized sexual orientation as sexual minority versus heterosexual.

Statistical analyses were conducted taking into account the complex sampling design and weighting of the GSS. Multivariable logistic regression models were used to evaluate the adjusted association of sexual minority status with each of the three outcomes in the overall sample and in gender-stratified samples. Analyses controlled for several possible confounders of gun safety beliefs and gun ownership including gender, age, race/ethnicity, education level, size of household, urbanicity, military veteran status, and political views as well as survey year. Results were evaluated at p ≤ 0.05. We report adjusted odds ratios (aORs) and 95% confidence intervals (CIs). This study was exempt from institutional review board review.

Results

Demographic characteristics are given in Table 1. In brief, 3.5% (standard error [SE] = 0.005) of men and 3.7% (SE = 0.004) of women identified as a sexual minority. For both men and women, sexual minority status was significantly associated with younger age and more liberal political views.

Table 2 characterizes gun safety beliefs and gun ownership by gender and sexual orientation and reports partial results from the regression analyses. As compared with heterosexual men, sexual minority men evidenced more than three times the odds (aOR = 3.24, 95% CI = 1.79–5.88) of favoring a gun safety law and about one-third the odds (aOR = 0.34, 95% CI = 0.18–0.65) of reporting a gun in the household. Personal gun ownership among men did not differ significantly by sexual orientation. Among women, gun safety beliefs and reporting a gun in the household did not differ significantly by sexual orientation, although reported presence of a household gun was somewhat higher among heterosexual (31.5%) than sexual minority (23.6%) women. But, among women who reported a gun in the household, sexual minority women demonstrated nearly four times the odds (aOR = 3.97, 95% CI = 1.43–11.03) of personal gun ownership compared with heterosexual women.

Discussion

Our results are consistent with a previous study conducted in a sample of U.S. veterans, which found that same-sex partnered individuals were significantly less likely to report having a gun in the household compared with opposite-sex partnered individuals (Blosnich et al. 2013). In this study, we expand on these findings to show that sexual minority women were significantly more likely to report personal gun ownership than did heterosexual women. Individuals who personally own guns report gun ownership with greater accuracy than nongun owners who live in a household with guns; thus, injury prevention experts have called for research focused on personal gun ownership (Hepburn et al. 2007).

Elevated gun ownership among sexual minority women compared with heterosexual women might be partially explained by the populations' disproportionate risk of stigma-related victimization (Herek 2009). Perceived risk of victimization is associated with obtaining a gun for self-protection (Kleck et al. 2011). However, sexual minority men in our sample demonstrated reduced risk of gun ownership compared with heterosexual men despite the fact that sexual minority men also experience elevated risk of victimization (Herek 2009). These results highlight the importance of recognizing gender differences in the sexual minority population.

Gender norms may be one alternate explanation for personal gun ownership among sexual minority women: compared with heterosexual women, sexual minority women are more likely to report self-ascribed masculinity and traditionally male-associated hobbies and occupational preferences (Lippa 2002). Indeed, concordant with other studies (Blosnich et al. 2013), sexual minority women in our sample were more likely to report military veteran status, a masculine-associated occupation, than heterosexual women. Across studies, masculine-linked gender norms have been associated with specific patterns of health risk behaviors (Fleming and Agnew-Brune 2015), including higher rates of gun ownership (Hepburn et al. 2007). Our findings call for additional research on the ways in which sex differences among sexual minorities shape health risks and inform tailored interventions.

Physicians tend to underestimate gun ownership among their patients (Becher and Christakis 1999), but gun storage counseling by physicians is effective in improving patient gun storage practices (Albright and Burge 2003). Personal gun owners are more likely to be responsible for gun storage and safety practices (Coyne-Beasley et al. 2005). Given that sexual minorities report favorable gun safety policy beliefs and generally liberal political views, an elevated propensity for personal gun ownership among sexual minority women could be unrecognized by physicians. Thus, clinicians working with sexual minority women should be cognizant of the populations' elevated propensity for personal gun ownership given that sexual minority women experience disproportionate risk for violence in the home, including suicide attempt (Haas et al. 2010) and intimate partner violence (Edwards et al. 2015).

Our finding that sexual minority status was associated with gun safety policy beliefs invites several other questions regarding sexual minority gun owners. For instance, because sexual minorities favor gun safety policy, do they have safer gun ownership practices than heterosexuals (e.g., keeping a gun locked, storing ammunition separately)? This may have relevance to gun-related violence risk reduction efforts.

Three limitations should be considered in contextualizing our results. First, we aggregated lesbian, gay, and bisexual individuals into a single sexual minority group to increase statistical power. Relatedly, because of our relatively small sample size, the frequencies of our subgroup outcomes are small, which reduced power to detect differences. Second, 2.3% of respondents refused to answer the initial question pertaining to guns in the household; elsewhere research finds that ‘refusals’ are more likely to be gun owners (Smith et al. 2014; Urbatsch 2018). In analyses not shown we evaluated the impact of this effect on study outcomes and found no evidence that the refusal rate could have impacted our conclusions. Third, the measures of gun safety attitudes in the GSS were limited to a single question. In addition, gun safety practices within gun-owning households were not assessed. Additional studies with both larger sample sizes and more thorough assessment of gun-related safety concerns can resolve these concerns.

Conclusions

We recommend that clinicians inquire about gun ownership and gun storage practices in regular screening and counseling sessions with both heterosexual and sexual minority patients, especially women. In addition, our results underscore the public health imperative of gender-stratifying behavioral risk research among sexual minority populations.