A Comparison of Risk Factors for Various Forms of Trauma in the Lives of Lesbian, Gay, Bisexual and Heterosexual Homeless Youth

Abstract

Numerous homeless youth experience trauma prior to leaving home and while on the street. Bullying and trading sex (i.e. exchanging sexual favors for survival items such as food) are additional forms of trauma experienced by many homeless youth. Although lesbian, gay, and bisexual (LGB) homeless youth experience higher rates of child abuse compared to their heterosexual counterparts, there is a paucity of research comparing these two groups. As such, we compare whether difficulties finding items of necessity (e.g. food) and different forms of trauma including child sexual abuse, being bullied, trading sex, and street sexual victimization significantly differ for LGB and heterosexual homeless youth. Bivariate results reveal that LGB youth have more trouble finding shelter and are more likely to trade sex compared to heterosexual youth. Logistic regression results show that trading sex is correlated with street sexual victimization. In the second model, we find that being female, experiencing more child sexual abuse, and ever having traded sex are all positively linked with street sexual victimization. LGB youth are over-represented among homeless youth populations and are also more likely to trade sex; therefore, these young people need services that are widely accessible and specific to their needs.

Many homeless youth experience numerous forms of trauma, such as child sexual abuse, prior to leaving home (Bender, Brown, Thompson, Ferguson, & Langenderfer, 2014; Tyler & Cauce, 2002). Homeless youth also endure other traumatic experiences while on the street: 21-32% report experiences of street sexual victimization (Bender et al., 2014; Tyler & Beal, 2010). Moreover, between 9 and 20% of homeless youth have traded sex for money, food, drugs, or shelter (Tyler & Beal, 2010; Tyler, Gervais, & Davidson, 2013; Walls & Bell, 2011) since being on the street.

While these prevalence rates are high, research finds that lesbian, gay, and bisexual (LGB) homeless youth experience even higher rates of early childhood trauma (Rew, Whittaker, Taylor-Seehafer, & Smith, 2005; Tyler, 2008; Whitbeck, Chen, Hoyt, Tyler, & Johnson, 2004) and street victimization (Tyler & Beal, 2010) compared to their heterosexual counterparts. Additionally, because LGB youth are more likely to lack family support (Bouris et al., 2010) and to run away or be kicked out of their homes (Cochran, Stewart, Ginzler, & Cauce, 2002), they may be less likely to return home and, consequently, have fewer supports available (Gattis, 2013). Subsequently, LGB homeless youth may be more likely to trade sex compared to their heterosexual peers (Tyler, 2008; Walls & Bell, 2011). Trading sex, or survival sex, involves exchanging sexual favors for specific items youth deem necessary for survival (i.e. shelter) (Heerde & Hemphill, 2016; Whitbeck et al., 2004). This exchange of sexual favors may be consensual or non-consensual (Heerde & Hemphill, 2016). Trading sex is also linked to being sexually victimized on the street (Whitbeck et al., 2004). Although LGB homeless youth may be at heightened risk for various forms of trauma compared to their heterosexual counterparts, there is a paucity of research comparing these two groups. As such, we compare whether difficulties finding items of necessity (e.g. shelter) and various forms of trauma including child sexual abuse, being bullied, trading sex, and street sexual victimization significantly differ for LGB and heterosexual homeless youth.

Different Forms of Trauma

Although all forms of child abuse are prevalent among homeless youth (Tyler & Cauce, 2002), research finds higher rates of sexual abuse among LGB youth compared to heterosexual youth (Rew et al., 2005; Tyler, 2008; Whitbeck et al., 2004). Trading sex is another form of trauma that many homeless youth experience (Tyler et al., 2013) and it is often done as a last resort (Hagan & McCarthy, 1997). Many youths who run away or who are kicked out of their homes find themselves on the streets with little education and limited job skills, and thus are often forced to find a way to support themselves, which may lead to trading sex. Trading sex is typically viewed as an exploitative interaction because homeless youth usually engage in this practice when they have no other viable options to meet basic needs (Hagan & McCarthy, 1997).

Bullying is another form of trauma homeless youth experience either while attending school or while on the street. Kidd and Shahar (2008), for example, established a positive link between homeless youth’s sexual minority status and elevated experiences of bullying from other students. Among a sample of lesbian, gay, bisexual, and transgender homeless youth, Bidell (2014) found that 16% reported frequent verbal harassment while in school, and 10% reported recurring physical harassment. Relatedly, Coates and McKenzie-Mohr (2010) found that homeless youth experienced high rates of bullying both prior to and after becoming homeless. Regardless of social environment, bullying can have adverse consequences for homeless youth (Coates & McKenzie-Mohr, 2010).

Theoretical Framework

We use a life stress framework (Lin & Ensel, 1989; Pearlin, 1989), which emphasizes multiple levels of influence (individual, family, and environment) and both distal (primary stressors) and proximal (secondary stressors) risk factors. Both elements are fundamental to understanding the relation between early and later forms of trauma among homeless youth. Additionally, the life stress framework assumes that individuals exposed to one serious stressor (e.g. child sexual abuse) will be exposed to additional stressors, which can then cluster together (Pearlin, 1989). Applied to the current study, primary (i.e. distal) stressors (i.e. the number of times youth have run away from home, child sexual abuse, and bullying) are expected to be positively correlated with secondary (i.e. proximal) stressors (i.e. trading sex and street sexual victimization). At the individual level, occupying the social status of LGB is another likely source of stress for these young people due to their socially stigmatized position within society (Kelleher, 2009). Homeless sexual minority youths’ exposure to stigma may be exacerbated by both their marginalized social status and structural vulnerability (Bruce, Stall, Fata, & Campbell, 2014). Subsequently, these youths are more likely to experience multiple trauma (Meyer, 2015). We also expect that, because of the cumulative nature of risk (Tyler & Whitbeck, 2004), homeless youth who are exposed to one extreme life stressor (e.g. child sexual abuse) in their formative years will undergo subsequent trauma, such as experiencing street sexual victimization.

Hypotheses

In this study, we compare whether difficulties finding items of necessity and various forms of trauma significantly differ for LGB and heterosexual homeless youth. As such, we hypothesize the following: Hypothesis #1: LGB youth will have more trouble finding food, money, clothing, and shelter compared to heterosexual youth. Hypothesis #2: LGB youth will be more likely to trade sex for food, money, drugs, and shelter compared to heterosexual youth. Hypothesis #3: LGB youth will experience more trauma including more child sexual abuse, more bullying, and more street sexual victimization compared to heterosexual youth. Hypothesis #4: LGB youth will be more likely to trade sex for (a) food, (b) money (c) drugs, and (d) shelter when controlling for other variables compared to heterosexual youth. Hypothesis #5: LGB youth will experience more street sexual victimization when controlling for other variables compared to heterosexual youth.

Method

We interviewed 150 youth in shelters and on the streets from July 2014 to October 2015 in two Midwestern cities. All youth ages16 to 22 and who were homeless or a runaway on the night prior to screening were eligible to participate. Homeless includes those who lack permanent housing such as spending the previous night in a shelter, public place, on the street, staying with friends or in a transitional facility, or other places not intended as a domicile (National Center for Homeless Education, 2010). Runaway refers to youth under age 18 who have spent the previous night away from home without parental permission (Ennett, Bailey, & Federman, 1999).

Four trained and experienced interviewers conducted the interviews. Participants were recruited through three local agencies, which offer various services (e.g. emergency shelter). Interviewers approached youth at these different venues, varying the times of the day, on both weekdays and weekends. This sampling protocol was conducted repeatedly over the course of the study. Interviewers obtained written informed consent from respondents and told them their responses were confidential and participation was voluntary. Most interviews were conducted in shelter conference rooms with fewer completed at a public library and outside in a park if weather permitted. The structured interview lasted approximately 45 minutes and participants received a $20 gift card for their time. Referrals for shelter, counseling services, and food services were offered to all youth at the time of the interview. The Institutional Review Board at the first author’s institution approved this study.

Measures

Gender was measured by asking youth, “What is your gender?” Two youth reported being “transgender” but did not choose a corresponding gender/sex and we could not impose this upon them; thus, the two transgender respondents were not included in this analysis. This variable was coded 0 = male and 1 = female (Walls & Bell, 2011).

Number of times run was a single item that asked youth for the total number of times that they had ever run away or left home (adapted from Whitbeck & Simons, 1990).

Sexual orientation was measured by asking youth to describe their sexual orientation (0 = lesbian, gay, or bisexual; 1 = straight or heterosexual) (Whitbeck et al., 2004).

Child sexual abuse included seven items (adapted from Whitbeck & Simons, 1990) which asked youth, “Before you were on your own, how often did any adult or someone at least five years older than you...” for example, “ask you to do something sexual” (0 = never to 6 = more than 20 times). All items loaded on a single factor (α = .92). Due to skewness, the seven items were dichotomized (0 = never; 1 = at least once) and then a count variable was created; a higher score indicates more types of sexual abuse experienced.

Bullying included three items created by the first author. Youth were asked, for example, how often they had been bullied at school (0 = never to 3 = many times). All items loaded on a single factor (α = .73). Due to skewness, the three items were dichotomized (0 = never; 1 = at least once) and then a count variable was created; a higher score indicates more types of bullying experienced.

Trouble finding necessities included four items created by the first author. For example, how often youth had trouble finding food or a place to stay (0 = never to 4 = every day). All items loaded on a single factor (α = .86). Due to skewness, each item was dichotomized (0 = never; 1 = at least 1-2 days per week) and then summed; a higher score indicates more necessities youth had trouble finding.

Trading sex included four items, such as how often youth had ever traded sex in return for food (0 = never to 3 = many times). Due to skewness, the four items were dichotomized (0 = never; 1 = at least once). Additionally, we created a single variable, “ever traded sex for any item” (0 = never; 1 = at least once) (Tyler et al., 2013; Walls & Bell, 2011).

Street sexual victimization included four items such as, how often youth have been forced to do something sexual since leaving home (0 = never to 3 = many times). All items loaded on a single factor (α = .90). Due to skewness, the four items were dichotomized (0 = never; 1 = at least once) and then a count variable was created; a higher score indicates more sexual victimization experienced (Tyler & Beal, 2010; Whitbeck et al., 2004).

Results

Participant characteristics

Ages of participants ranged from 16 to 22 years (M = 19.4 years). One-half of our sample (N = 77; 51%) were female and 33 youth (22%) identified as LGB. In terms of race/ethnicity, 41.3% were White, 26% Black or African American, 10% Hispanic or Latino, 4% American Indian or Alaskan Native, 13.3% were bi-racial and 5.3% multi-racial. Youth reported running away from home between one and 35 times (M = 4.9 times; SD = 6.32).

Table 1 results revealed that LGB youth were more likely to report trouble finding shelter (χ² = 4.38; p < .05) compared to heterosexual youth, which is consistent with hypothesis #1. LGB youth were also more likely to trade sex for food (χ² = 8.69; p < .01), money (χ² = 10.27; p < .01), and drugs (χ² = 5.08; p < .05) and for any item (χ² = 6.09; p < .05) compared to heterosexual youth, which is consistent with hypothesis #2. As some of the cell sizes for trading sex are small, we recommend caution in interpreting the findings in Table 1. Table 2 results revealed that experiences of child sexual abuse (t = 3.06; p < .01), bullying (t = 2.61; p < .05), and street sexual victimization (t = 3.80; p < .01) were significantly higher among LGB youth than heterosexual youth, which is consistent with hypothesis #3 (see Supplemental Online Material for Tables 1 and 2).

Table 1.

Logistic Regression Models for Correlates of Trading Sex for Food, Shelter, Money, and Drugs

	Model 1		Model 2		Model 3		Model 4
n = 144
	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
	For food		For money		For drugs		For shelter
Female	.05	.00 – .57*	.38	.09 – 1.55	.16	.02 – 1.38+	.08	.01 – .57*
Number of times run	1.17	1.04 – 1.31*	1.05	.95 – 1.15	.99	.87 – 1.13	1.06	.96 – 1.17
Heterosexual	.11	.01 – .82*	.26	.07 – .94*	.16	.02 – 1.31+	.69	.13 – 3.57
Child sexual abuse	1.43	1.00 – 2.03*	1.25	.98 – 1.59+	.88	.57 – 1.36	1.10	.80 – 1.50
Bullying	.57	.24 – 1.39	1.37	.76 – 2.46	2.35	.76 – 7.29	1.83	.91 – 3.69+
Street sexual victimization	2.06	1.16 – 3.65*	1.66	1.12 – 2.47*	2.25	1.17 – 4.31*	2.47	1.44 – 4.23**
LRχ2	40.97		74.53		36.49		58.63
d.f.	6		6		6		6
Cox & Snell R2	.20		.19		.13		.23
Nagelkerke R2	.50		.37		.39		.47

Note: OR = odds ratio; CI = confidence interval.

^**p ≤ .01;

^*p ≤ .05;

⁺p < .10.

Table 2.

OLS Regression Models for Correlates of Street Sexual Victimization

	Model 1		Model 2		Model 3		Model 4
n = 144
	β	95% CI	β	95% CI	β	95% CI	β	95% CI
Female	.19	.10 – 1.02*	.17	.05 – .94*	.20	.14 – 1.03**	.23	.23 – 1.08**
Number of times run	.13	−.01 – .07	.11	−.01 – .06	.04	−.03 – .04	−.01	−.04 – .03
Heterosexual	−.25	−1.44 – −.32**	−.20	−1.24 – −.13*	−.15	−1.09 – .02+	−.11	−.91 – .17
Child sexual abuse			.25	.06 – .25**	.22	.04 – .24**	.17	.01 – .20*
Bullying					.21	.07 – .53*	.15	−.02 – .44+
Ever traded sex							.29	.52 – 1.74**
Adjusted R2	.11		.17		.20		.26

Note: CI = confidence interval.

^**p ≤ .01;

^*p ≤ .05;

⁺p < .10.

Logistic regression results for Model 1 (trading sex for food), revealed that heterosexual females had a 95% lower odds of trading sex for food compared to heterosexual males (OR = .05; p ≤ .05). For each additional time a youth ran away, their odds of trading sex for food increased by 17% (OR = 1.17; p ≤ .05). For each additional type of child sexual abuse experienced, the odds of trading sex for food increased by 43% (OR = 1.43; p ≤ .05), whereas for each additional type of street sexual victimization experienced, youths’ odds of trading sex for food increased by 106% (OR = 2.06; p ≤ .05). Heterosexual youth had an 89% lower odds of trading sex for food compared to LGB youth (OR = .11; p ≤ .05), which is consistent with hypothesis #4 (part a).

In Model 2 (trading sex for money), for each additional type of street sexual victimization experienced, youths’ odds of trading sex for money increased by 66% (OR = 1.66; p ≤ .05). Heterosexual youth had a 74% lower odds of trading sex for money compared to LGB youth (OR = .26; p ≤ .05), which is consistent with hypothesis #4 (part b). In Model 3 (trading sex for drugs), for each additional type of street sexual victimization experienced, youths’ odds of trading sex for drugs increased by 125% (OR = 2.25; p ≤ .05). There was no difference between LGB and heterosexual youth for trading sex for drugs; thus, hypothesis #4 (part c) was not supported. In Model 4 (trading sex for shelter), heterosexual females had a 92% lower odds of trading sex for shelter compared to heterosexual males (OR = .08; p ≤ .05). For each additional type of street sexual victimization experienced, youths’ odds of trading sex for shelter increased by 147% (OR = 2.47; p ≤ .01). There was no difference between LGB and heterosexual youth for trading sex for shelter; thus, hypothesis #4 (part d) was not supported.

Results from stepwise OLS regression (Table 2) revealed that though LGB youth experienced more street sexual victimization compared to their counterparts (see Models 1 and 2), sexual orientation was no longer significant when bullying and trading sex were added. Thus, these results do not support hypothesis #5.

Discussion

Our study compared whether difficulties finding items of necessity and various forms of trauma significantly differ for LGB and heterosexual homeless youth. Consistent with our hypotheses, LGB youth experience more child sexual abuse, bullying, and street sexual victimization compared to heterosexual youth. Additionally, LGB youth have more trouble finding shelter and are more likely to trade sex for food, money, and drugs compared to heterosexual youth. Though both LGB and heterosexual homeless youth have trouble finding basic necessities, LGB youth have a more difficult time accessing shelter. One possible explanation is that some LGB homeless youth avoid using shelter services because they fear that they might experience prejudice and discrimination from service providers (Burwick, Oddo, Durso, Friend, & Gates, 2014; Hunter, 2008). LGB youth also are more likely to trade sex compared to heterosexual youth. It may be that some LGB youth believe certain service providers may be prejudiced against them because of their sexual orientation and this may result in LGB youth resorting to trading sex for survival. Previous research underscores homeless youths’ heightened perceptions of repeated exposure to sexual exploitation on the street, particularly from people they met through service agencies (Fogel, Martin, Nelson, Thomas, & Portia, 2017). These experiences could lead some homeless youth who are doubly marginalized (i.e. LGB and homeless) to avoid services altogether. Moreover, lack of basic necessities could also make them more vulnerable to trading sex, which is supported by our findings.

Though LGB youth are more likely to be bullied compared to heterosexual youth, being bullied was not associated with any form of trading sex, contrary to hypothesis #3. One possible explanation is that because many homeless youth have experienced severe forms of child physical and sexual abuse prior to leaving home (Tyler & Cauce, 2002), as well as several forms of street victimization (Tyler & Beal, 2010), they may be less likely to view minor forms of harassment as bullying victimization. Tyler and Melander (2009) found that homeless youth did not consider “minor” acts of child physical abuse (e.g. being pinched) or “non-contact” sexual abuse (e.g. being asked to do something sexual) as abuse, but instead, reserved this label for moderate to severe violence (e.g. hit with a fist) and “contact” sexual abuse (i.e. being touched sexually). Given the lack of research on bullying among homeless youth, more work is needed to identify unique forms of bullying that street youth experience and their perceptions of this abuse.

Experiencing more types of street sexual victimization was linked to all four forms of trading sex. One possible explanation is that homeless youth who have previously been exploited are more vulnerable to re-victimization (Tyler et al., 2013). Exposure to repeated forms of violence can amplify homeless youth’s risk of sexual revictimization, particularly for youth who have deviant peers and are street-involved for longer periods of time (Tyler, Hoyt, & Whitbeck, 2000). Spending time in the street environment also increases homeless youths’ chances of encountering individuals who may take advantage of them, through street sexual victimization (Tyler & Beal, 2010) or through trading sex, especially if potential perpetrators know youth are vulnerable and lack basic necessities (Fogel et al., 2017; Hagan & McCarthy, 1997).

Though prior work has found higher rates of sexual victimization among sexual minority youth (Whitbeck et al., 2004), we found no difference in street sexual victimization by sexual orientation when controlling for gender, number of times youth have run away from home, and other forms of trauma. Specifically, the addition of bullying and ever trading sex to our model reduced the sexual orientation variable to non-significance. It is possible that because the majority of homeless youth experience multiple traumas and generally have equitable exposure to street life, all homeless youth are at elevated risk for becoming a victim. It may also be that the risk-laden experience of homelessness works as an equalizer across social differences, such as sexual orientation, and overrides other risk factors that could distinguish young people’s experiences in the general population.

Though bullying was initially significant, it did not remain a significant correlate of street sexual victimization when the trading sex variable was added to the model. It may be that though bullying matters for street sexual victimization, perhaps its magnitude does not matter as much as having to trade sex, which is often an exploitative interaction homeless youth engage in when they have no other available options (Hagan & McCarthy, 1997). Moreover, trading sex likely puts young people in dangerous and vulnerable situations with little protection from violent individuals and others who may exploit them and thus, trading sex increases their risk for street sexual victimization (Tyler, 2008).

Our results are generally consistent with a life stress framework. Numerous study youth experienced a primary or distal stressor (e.g. child sexual abuse) in their formative years. Moreover, many of these young people made numerous transitions away from home at a time when they were least prepared to do so. These primary or distal stressors are exacerbated over time, as many youth experience more proximal stressors, including trading sex and being sexually victimized. We also find that stressors tend to cluster (Pearlin, 1989) among homeless youth, such that those exposed to one serious stressor, such as child sexual abuse, are exposed to additional stressors, including street sexual victimization. Moreover, LGB youth experience unique stressors to their well-being based on their socially marginalized status (Bruce et al., 2014), which is congruent with our finding that homeless sexual minority youth experienced elevated rates of sexual victimization and trading sex.

Some limitations should be noted. First, all data are self-reported and some measures are retrospective, which may have resulted in recall bias. Relatedly, given the sensitive nature of the topics, some youth may have been unwilling to disclose such experiences (Tyler & Melander, 2009). Also, our sample is one of convenience; therefore, our findings cannot be generalized to other homeless youth populations. Because our findings are cross-sectional, we do not know the timing of events. Thus, it is possible that street sexual victimization occurred during a trading sex encounter. Finally, the small number of youth that reported trading sex likely limited our ability to explain trading sex and find significant differences by sexual orientation.

Despite these limitations, our paper has numerous strengths and contributes to our understanding of the relationships between different forms of trauma for both heterosexual and non-heterosexual social groups. Specifically, our work shows that there are differences between LGB and heterosexual homeless youth in difficulties finding shelter and likelihood of trading sex. Moreover, experiencing street sexual victimization is positively correlated with trading sex for both groups of homeless youth. Additionally, experiencing more child sexual abuse, ever trading sex for any item, and being a heterosexual female are all risk factors for street sexual victimization. Though services for this population exist, we find that many homeless youth of diverse sexual orientations have trouble finding most basic necessities. As such, it is possible that at least some youth are not accessing services for various reasons, which may include unawareness of their existence or fear of sexual orientation-related prejudice and discrimination. Another possible explanation is that some homeless youth may resist using services to avoid being judged by conventional cultural expectations and alternatively, work hard to make it on their own (Schmitz & Tyler, 2017).

Policy Implications

At the policy level, our findings show that early exposure to sexual abuse is associated with trading sex for food and street sexual victimization. Bullying is another form of trauma that homeless youth experience, but we know little about how it is similar and/or different from other types of victimization. It may benefit service providers to know about the distinct forms of trauma experienced by homeless youth, that the origins of trauma may be unique, and that level of exposure varies. Additionally, we find that LGB youth are at higher risk for trading sex for food and money and have more trouble finding shelter compared to heterosexual youth, suggesting that some youth may be unaware of services or afraid to use them for fear of prejudice or discrimination (Hunter, 2008). Thus, it may benefit providers to advertise services specifically tailored to the needs of LGB youth. Informing LGB youth that they have a safe place to go with available services may reduce the risk of further trauma. Additionally, LGB staff members who are open about their non-normative sexual orientation may also increase the likelihood that this group will seek services (Choi, Wilson, Shelton, & Gates, 2015).