The paper describes the creation of a new scale to measure intimate partner violence (IPV) among gay and bisexual men.
Seven focus group discussions were held with gay and bisexual men, focusing on defining intimate partner violence: 30 forms of IPV were identified. A venue-recruited sample of 912 gay and bisexual men was surveyed, examining definitional understanding and recent experiences of each of the 30 forms of IPV. Participants were also asked questions from the CDC definition of intimate partner violence and the short-form of the Conflicts Tactics Scale (CTS2S). Factor analysis of responses to the definitional questions was used to create the IPV-GBM scale, and the prevalence of intimate partner violence was compared with that identified by the CDC and CTS2S measures of intimate partner violence.
A 23-item scale, with 5 unique domains, was created, with strong internal reliability (Cronbach Alpha >.90). The IPV-GBM scale mirrored both the CDC and CTS2S definitions of intimate partner violence, but contained additional domains such as controlling violence, monitoring behaviors, emotional violence, and HIV-related violence. The new scale identified a significantly higher prevalence of IPV than either of the more commonly used measures.
Citation: Stephenson R, Finneran C (2013) The IPV-GBM Scale: A New Scale to Measure Intimate Partner Violence among Gay and Bisexual Men. PLoS ONE 8(6): e62592. doi:10.1371/journal.pone.0062592
Editor: Nitika Pant Pai, McGill University Health Centre, McGill University, Canada
Received: August 3, 2012; Accepted: March 25, 2013; Published: June 5, 2013
Copyright: © 2013 Stephenson, Finneran. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This original research was supported by funding from the Eunice Kennedy Shriver National Insitute of Child Health & Human Development, grant #5R21HD066306-02. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The emergence of research on intimate partner violence (IPV) among gay, bisexual, and other men who have sex with men (MSM) has demonstrated that IPV occurs in male-male partnerships at rates similar to or higher than opposite-sex partnerships –. Recently, researchers have documented vastly varied, though universally high, rates of IPV among MSM: between 32–78% for any form of IPV , , 12–45% for physical IPV , , and 5–33% for sexual IPV , . Studies of IPV among gay and bisexual men have suffered from a number of methodological limitations , , including a challenge not unique to the field of same-sex IPV research: a lack of an uniform definition of IPV and use of IPV definitions non-specific to MSM –. Existing studies of IPV among MSM have relied upon measures of IPV that were created for use in assumedly heterosexual populations . While IPV is universal across sexual orientations, it is not clear as to the extent to which the typologies of IPV – especially psychological IPV and controlling behaviors – experienced by gay and bisexual men are different to those experienced by heterosexual women, perhaps contributing to the wide range of prevalence estimates found in the literature. This paper describes the development of a new scale to measure IPV among gay and bisexual men, the IPV-GBM scale, contrasting the prevalence of IPV identified among gay and bisexual men using this new scale with that identified with two other commonly used measures of IPV. The new IPV-GBM scale has the potential to significantly improve the accuracy of the measurement of IPV among gay and bisexual men in the US, allowing a more accurate understanding of the relationships between IPV and health outcomes experienced by gay and bisexual men.
In studies of IPV in women, one of the most commonly used measures of IPV is the Revised Conflict Tactics Scale (R-CTS) , developed from the original Conflict Tactics Scale (CTS) , which aimed to measure the extent to which specific tactics, including acts of physical violence, were used in intimate partnerships . The R-CTS updated the original CTS by including measurement of sexual coercion and consequences of physical violence (injury), and is comprised of 39 items contained in five sub-scales: physical assault, psychological aggression, negotiation, sexual coercion, and injury . In 2004, Straus and Douglas updated the R-CTS and created the short-form CTS (CTS2S), a reduced, 10-item scale, including the same five sub-scales included in the R-CTS . Recently, a number of studies have used the CTS2S to identify the prevalence of IPV among gay and bisexual men , , , , .
A number of studies have also relied on single-item questions to capture the experience of IPV among gay and bisexual men –, often based up Center for Disease Control and Prevention (CDC)-developed specific definitions of physical and sexual IPV. Variations of this definition are commonly used to capture recent experiences of physical (In the last “time period”, have any of your partners ever tried to hurt you? This includes pushing you, holding you down, hitting you with a fist, kicking you, attempting to strangle you, and/or attacking you with a knife, gun or other weapon?) and sexual violence (In the last “time period”, have any of your partners ever used physical force or verbal threats to force you to have sex when you did not want to?) .
In a systematic review of the literature around IPV among MSM, Finneran and Stephenson (in press) note that across 28 studies identified, 16 different definitions of IPV were used by researchers in various combinations . The most commonly used scale measures of IPV were the Conflict Tactics Scale  or its derivatives, the Revised Conflicts Tactics Scale  or the CTS2S. Ten studies used definitions of IPV that were unique to the study or did not provide a reference to a validated scale, while several studies used binary measures of the presence of IPV based on the CDC definition of violence. However, none of these measures were developed specifically for gay and bisexual men; hence, it remains unknown whether or not these measures accurately represent IPV in gay and bisexual men. Several authors of studies captured in this systematic review reported having to modify the validated scales post hoc, such as by using gender-neutral language, in order to make the measurement tools appropriate for MSM . In this paper we describe the development of a new scale – the intimate partner violence among gay and bisexual men (IPV-GBM) scale – and compare the prevalence of IPV identified with that identified with the CTS2S and the CDC measures of IPV.
Data collection involved two stages: the first included seven focus group discussions (FGDs) with gay and bisexual men stratified by race (Black/African-American and white) in Atlanta, Ga., and the second stage included a survey of over1000 gay and bisexual men, also in Atlanta, Ga. For both stages, respondents were recruited through venue-based sampling (VBS). VBS is a derivative of time-space sampling, in which sampling occurs within prescribed blocks of time at previously-identified venues at which hard-to-reach populations congregate with greater frequency than elsewhere . In order to reach a diverse population of gay and bisexual men in the Atlanta area, the venue sampling frame used for this study consisted of a wide variety of over 160 gay-themed or gay-friendly venues, including Gay Pride events, gay sports teams events, gay fundraising events, downtown areas, gay bars, bathhouses, an AIDS service organization, an MSM-targeted drop-in center, gay bookstores, restaurants, and urban parks.
Study recruiters stood adjacent to the venue, drew an imaginary line on the ground, and approached every nth man who crossed it; n varied between one and three depending on the volume of traffic at the venue. If he agreed to be screened, he was then asked a series of eight questions to assess his eligibility. Eligible men were then read a short script that described the study process. For men recruited for the first phase, the script described their participation in a FGD. Men were eligible for FGD participation if they reported being aged 18 or older, living in the Atlanta metropolitan area, and identifying as a gay or bisexual man. In total, seven FGDs were held (n = 84); two with only African-American/Black participants, one with only white participants, and four with both participants of all races/ethnicities. Each FGD lasted approximately one hour, and the discussion centered on understanding multiple definitions of IPV. The question guide was based on the CTS2S questions ; respondents were asked if they would consider each item to be IPV if it were to occur in a male-male relationship. Further questions examined participant's definitions of sexual, physical, and psychological IPV and controlling/stalking behaviors. Discussions were recorded and transcribed, with analysis conducted in MAXQDA. The focus of the analysis was on identifying definitions of IPV, and on examining variations in definitions of IPV across white and Black/African-American participants. FGD transcripts were reviewed by three different researchers in order to ensure that all forms of IPV mentioned by the participants were captured, resulting in the identification 46 different potential forms of IPV. Forms of IPV that were inconsistent with the empirical literature regarding the nature of IPV (e.g., slamming a door, giving someone the “silent treatment”) were removed and duplicative forms of IPV (e.g., calling someone names and putting him down) were condensed. As a result of this review, 30 different forms of IPV were used to create the survey questions in order to examine the perceptions of and experience of IPV among the sample of over 1000 gay and bisexual men in Atlanta.
Recruitment for the survey was conducted using the same venue-based approach as was used for FGD recruitment. For survey recruitment, eligible participants were read a script that outlined their potential participation in a web-based survey, approximately 25 minutes in length, that could be completed at home, or, in the case of five venues (the AIDS service organization, the drop-in center, Atlanta Pride, In the Life Pride, and a National Coming Out Day event), at the venue itself on an iPad. Men interested in study participation were then given a card with a web address and a unique identifier, ensuring that the survey could only be completed once per venue-recruited participant.
Of 4,309 men approached, 2,936 (59.9%) agreed to be screened for the survey. Of these, 2,093 (71.3%) were eligible for study participation. Men were eligible for study participation if they reported being 18 years of age or older, being male, identifying as gay/homosexual or bisexual, living in the Atlanta Metro Area, and having had sex with a man in the previous six months. Of eligible participants, 1,965 (93.9%) were interested in study participation. A total of 1,074 men completed the survey; thus 21.9% of men approached and 51.4% of eligible men completed the survey. Approximately one-third (33.7%) completed the survey at a venue, while the remaining two-thirds (66.3%) of respondents completed the survey at home. A total of 912 men had complete data for all covariates of interest and were included in the final analysis.
The self-administered, web-based survey, hosted on SurveyGizmo, contained several domains of questions regarding demographics (e.g., age, education, and race) and recent sexual behavior with male partners. To measure IPV, the survey included 30 items taken from the FGDs: participants were asked if they considered each one of the items to be IPV (“Would you consider it violent if a male partner of yours were to…”), and if they had experienced it from or perpetrated it against a male partner in the past 12 months. The survey also included the CTS2S and the binary questions based on the CDC definitions measuring the experience and perpetration of physical and sexual IPV.
Creating a new scale to measure IPV among gay and bisexual men.
Rotational factor analysis was conducted to identify which of the 30 items were to be included in the IPV-GBM scale based upon respondent's answers to whether or not they would consider an individual item presented to be violent if it happened to them from a male partner. The factor structure of the IPV-GBM scale was determined using principal components analysis with oblique rotation using a promax solution. The Kaiser-Meyer-Olkin test and Bartlett's test of sphericity were calculated prior to the exploratory factor analysis (EFA) in order to assure the appropriateness of EFA. The factor analysis was conducted for the total sample, and then separately for white and Black/African-American respondents to identify racial variations in scale content. There were insufficient numbers of Latino/Other respondents to allow factor analysis to be performed for this group. Reliability of each definitional scale (overall, white respondents, Black/African-American respondents only) was assessed by calculating Cronbach's alpha to assess the internal consistency of the items. Adequate reliability was indicated if Cronbach's alpha was >0.70.
The analysis sample.
Table 1 shows the characteristics of participants in the final analysis sample. The mean age was approximately 34 years, with the majority reporting post-secondary education (83.9%), current employment (79%), negative HIV status (69%), and homosexual sexual orientation (90%). Approximately 48% of the sample was white non-Hispanic, 40% Black/African-American non-Hispanic, and 12% Latino/Hispanic or other, including Asian/Pacific Islander and Native American/Alaska Native.
Definitions of IPV.
There was significant variation reported among survey participants as to what constituted IPV (Table 2). While more than 90% of respondents agreed that hitting, punching, kicking, rape, slapping and intentional damage to property were forms of IPV, fewer than 40% of participants reported that preventing the victim from seeing his friends or family, putting the perpetrator's sexual needs before the victim's, asking/telling the victim to act straight around others, criticizing the victim's clothes, or calling the victim fat were considered IPV. Thus, definitions of IPV tended to focus more on physical and extreme forms of sexual IPV (e.g., rape), whereas controlling behaviors were less likely to be viewed as IPV. Latino/Other men endorsed an average of 20 of 30 items as IPV and Black/African-American an average of 19 of 30, both significantly higher than the mean 17 endorsed by white men. There were racial variations in the definitions of IPV: Black/African-American participants were less likely to report that hitting, punching, kicking, rape, slapping, intentionally transmitting HIV and intentional damage to property were forms of IPV, although the vast majority of all respondents affirmed that these were forms of IPV. Conversely, Black/African-American and Latino/Other participants were more likely to report that doing something sexual for which you hadn't given consent, preventing someone from seeing their family or friends, refusing to wear a condom during sex, calling someone names, and cheating were forms of IPV. Non-white participants were also more likely to report that controlling behaviors, such as demanding access to a cell phone or email, reading text messages or email, and preventing someone from seeing his friends were forms of IPV. Whereas white men were more likely to report physical violence or extreme sexual violence as IPV, non-white men were more likely to report psychological violence or controlling behaviors as forms of IPV.
Both the Kaiser-Meyer-Olkin test (0.903) and Bartlett's test of sphericity (χ2 = 17817.0, p<0.000) indicated that the variation present in the data was well-suited to EFA, both for the overall sample and for factor analysis by race (White men, KMO: 0.866, Bartlett's test p<0.000) and Black men, KMO: 0.901, Bartlett's test p <0.000). The factor analysis yielded five unique factors with eigenvalues >1.0: physical and sexual IPV, monitoring behaviors, controlling behaviors, HIV-related IPV, and emotional IPV (Table 3). The same five factors were identified for each racial group, although the content of the factor varied by race. Five items did not load into any factor (i.e., alpha<0.50 in any one factor): name-calling, refusing to wear a condom during sex, revealing the victim's sexual orientation to others (“outing” him), doing something sexually for which the victim had not given his prior consent, and unintentionally transmitting HIV to the victim.
Factor One: Physical and Sexual IPV.
For the total sample, and for white respondents, this factor was comprised of slapping, punching, hitting, kicking, pushing, coerced sex, rape, and damage to property: however, for white respondents, pushing and coerced sex did not load into this factor. The factor explained 32% of total variance for the total sample: 31% for white men and 34% for Black/African-American men.
Factor Two: Monitoring Behaviors.
The same items loaded for all groups: demanding access to a cell phone, demanding access to email, reading text messages or email(s) without knowledge, and repeatedly posting on victim's social networking pages (e.g., Facebook, Twitter), explaining approximately 14% of total variance.
Factor Three: Controlling Behaviors.
Again, the same items loaded for all groups: preventing a victim from seeing his family or friends, and preventing victim from seeing his partner's family or friends, explaining approximately 5% of the variance in each group.
Factor Four: HIV-related IPV.
For the total sample and for white men, the items loading in this factor were lying about HIV status to a partner, not revealing HIV positive status to a partner before sex, and intentionally transmitting HIV, which collectively explained 5% of the total variance. For Black/African-American men, cheating and the perpetrator putting his sexual needs first also loaded on this factor, although the percentage of variation explained remained approximately 5%.
Factor Five: Emotional IPV.
For the total sample and for Black/African-American men, the following items loaded: calling the victim fat, asking/telling the victim to “act straight,” and criticizing the victim's clothes, explaining approximately 4% of the variation. For white men, the perpetrator putting his sexual needs before the victim's also loaded on this factor. This factor explained only 4% of the total variance.
The experience of IPV.
Among the total sample, the most commonly experienced forms of IPV in the past 12 months were criticizing of clothes (18.5%, emotional IPV), reading text messages without permission (16.2%, monitoring behavior), and pushing/shoving (14.7%, physical and sexual IPV). The least commonly experienced forms of IPV were rape (2.8%, physical and sexual IPV), preventing victim from seeing his family (3.3%, monitoring behaviors), and intentionally transmitting HIV (3.2%, HIV-related IPV) (Table 4).
There were clear racial variations in reporting experience of IPV. Black/African-American respondents were more likely to report experiencing any form of physical and sexual IPV and all forms of HIV-related IPV. Black/African-American respondents were also more likely to report being prevented from seeing their family by a male partner, but were not more likely than white men to report any other controlling behaviors. There were no significant racial differences in the reporting of emotional violence between Black/African-American and white men. Similar variations were present in the variation of reporting recent perpetration of IPV, with Black/African-American men significantly more likely to report any recent perpetration of IPV.
The CDC definition-based measure of IPV consistently generated the lowest prevalence of IPV (experience 13.5%, perpetration 7.9%). The CTS2S measure generated higher prevalences (experience 28.2%, perpetration 18.6%), while the newly developed IPV-GBM scale generated significantly higher prevalences than the other two measures (experience 45.8%, perpetration 32.1%). The same patterns in the reporting of IPV experience and perpetration across the three measures were observed in each racial group; however, while the CDC measurement and the CTS2S measurement suggested significantly higher prevalences of both experience of IPV and perpetration of IPV among Black/African-American and Latino/Other men compared to white men, these differences were not statistically significant when using the IPV-GBM scale (Figure 1).
The results of sensitivity analysis are summarized in Table 5. For all but two forms of IPV (rape and forced sex), the CTS2S measurement was more sensitive at identifying victims of IPV compared to the CDC measurement. For example, only 27% of participants who had recently experienced emotional violence from a male partner were classified as experiencing violence by the CDC measurement, whereas the CTS2S correctly classified 66% of these participants. Overall, the 64.9% and 41.7% of participants with recent experience of IPV per the IPV-GBM scale were not classified as such by the CDC measurement and the CTS2S measurement, respectively.
The results demonstrate the increased sensitivity of scale to specifically measure the experience of IPV among MSM in the US. The IPV-GBM scale, consisting of 23 items in five unique domains of IPV, showed strong internal reliability, and more than 60% of variance in definitions of IPV was explained by the scale items. Although there were some variations in the content of the scale by race, these were minimal, and the scale seems appropriate for use in both white and Black/African-American gay and bisexual male populations. A small sample size limited the ability to create the scale for Latino or other racial/ethnic groups, and further work must examine the extent to which the IPV-GBM scale is applicable to other racial and ethnic groups of gay and bisexual men in the U.S Further work is required to test associations between IPV as measured in the IPV-GBM and other health outcomes, especially health outcomes thought to have strong correlations with IPV such as substance abuse , –, depression , , , and HIV risk. Much has been written recently about the possible links between IPV and HIV among gay and bisexual men in the US , , , , , , , although findings of a statistical association have been mixed , , , , , .
Participants largely conceptualized IPV as including physical violence and extreme sexual coercion, items that are included in both the CDC definition of IPV and the variations of the CTS. However, the IPV-GBM also identified areas of IPV not included in other measures that gay and bisexual men reported as constituting IPV. These included HIV-related IPV, monitoring behaviors (such as observing emails/texts) and controlling behaviors (including limiting access to friends or family). The IPV-GBM identified a significantly higher prevalence of IPV than either of the other two measures tested, suggesting that the inclusion of items in an IPV scale that more closely reflects the lived experiences of gay and bisexual men may lead to a more accurate, although higher, estimation of the prevalence of IPV. Of course, it is also possible that these additional forms of IPV are also prevalent in heterosexual populations, and further work is required to establish whether the newly created scale is unique to gay and bisexual men, or represents a generally more accurate measure of IPV that can be used in wider populations.
Interestingly, Black/African-American men were less likely to report that physical acts constituted IPV while they were more likely to report controlling behaviors as IPV. However, Black/African-American men were more likely to report recent experience of all forms of physical and sexual IPV, leading to a generally higher prevalence of IPV among Black/African-American gay and bisexual men. Disconnect exists between what Black/African-American men in this sample think of as IPV and what they report actually experiencing, suggesting that physical and sexual IPV were perhaps more tolerated (and hence not conceived of as IPV as often) among Black/African-American men in this sample. Despite this, the majority of men in the sample endorsed physical and sexual violence as forms of IPV. Thus, further work is needed to understand the factors driving the racial variations observed in both the conceptualization of and the experience of IPV.
There are a number of limitations to the current study. The sampling procedures relied on venue-based sampling rather than random sampling: however, there is increasing evidence that this form of sampling produces a sample of similar diversity as is found with random sampling methods , , . The data are specific to the metro-Atlanta area, and there may be regional differences in how gay and bisexual men experience and conceptualize IPV; thus, further work is needed to replicate this work with men in other parts of the U.S. Additionally, the stem question used to determine whether or not something was IPV (“Would you consider it violent if a male partner of yours were to…”) was intentionally oblique. Participants may have considered that certain acts would not always necessarily constitute violence (particularly in cases of consensual violence, such as occurs in bondage, domination, and sadomasochism [BDSM] relationships); this may partly explain why no single IPV item received 100% endorsement.
The results presented here provide encouraging evidence for a new, more accurate, measure of IPV among gay and bisexual men in the U.S. The large number of items in the scale enhances content sensitivity and reliability, and provides the ability to differentiate between five domains of IPV. The IPV-GBM utilizes interspersed item order to limit response set bias , and the referent time period can be adjusted from 12 months to meet the needs of the research question (e.g., last 3 months). The IPV-GBM requires a 6th grade reading level, and takes approximately 10–15 minutes to complete. These characteristics are similar to those of the variations of the CTS , , which have gained in popularity and frequency of use in the IPV research community. Given the increased attention to IPV among gay and bisexual men, a more accurate measure of IPV that is grounded in the lived realities of gay and bisexual men is vital. Further work is now required to test this scale on larger samples of gay and bisexual men, and to explore the extent to which the IPV-GBM scale is applicable to other racial/ethnic groups and is associated with other health outcomes.
Conceived and designed the experiments: RS CF. Performed the experiments: RS CF. Analyzed the data: CF. Wrote the paper: RS CF.
- 1. Tjaden P, Thoennes N, Allison CJ (1999) Comparing violence over the life span in samples of same-sex and opposite-sex cohabitants. Violence and Victims 14: 413–425.
- 2. Blosnich JR, Bossarte RM (2009) Comparisons of Intimate Partner Violence Among Partners in Same-Sex and Opposite-Sex Relationships in the United States. American journal of public health 99: 2182.
- 3. Messinger AM (2011) Invisible Victims: Same-Sex IPV in the National Violence Against Women Survey. Journal of Interpersonal Violence 26: 2228.
- 4. Houston E, McKirnan DJ (2007) Intimate partner abuse among gay and bisexual men: Risk correlates and health outcomes. Journal of Urban Health 84: 681–690.
- 5. Pantalone DW, Schneider KL, Valentine SE, Simoni JM (2011) Investigating Partner Abuse Among HIV-Positive Men Who have Sex with Men. AIDS and Behavior: 1–13.
- 6. Stephenson R, Khosropour C, Sullivan P (2010) Reporting of intimate partner violence among men who have sex with men in an online survey. Western Journal of Emergency Medicine 11: 242.
- 7. Craft SM, Serovich JM (2005) Family-of-origin factors and partner violence in the intimate relationships of gay men who are HIV positive. Journal of interpersonal violence 20: 777–791.
- 8. Greenwood GL, Relf MV, Huang B, Pollack LM, Canchola JA, et al. (2002) Battering victimization among a probability-based sample of men who have sex with men. American journal of public health 92: 1964.
- 9. Relf MV (2001) Battering and HIV in men who have sex with men: A critique and synthesis of the literature. Journal of the Association of Nurses in AIDS Care 12: 41–48.
- 10. Murray CE, Mobley AK (2009) Empirical research about same-sex intimate partner violence: A methodological review. Journal of Homosexuality 56: 361–386.
- 11. Saltzman L, Fanslow J, McMahnon P, Shelley G (2002) Intimate partner violence surveilliance: uniform defintions and recommended data elemts, version 1.0. Atlanta, GA: Centers for Disease Control and Prevention National Center for Injury Prevention and Control.
- 12. Bonomi A, Thompson R, Anderson M, Rivara F, Holt V, et al. (2006) Ascertainment of intimate partner violence using two abuse measurement frameworks. Injury Prevention 12: 121–124.
- 13. Finneran C, Stephenson R (in press) Intimate Partner Violence among Men Who Have Sex with Men: A Systematic Review. Trauma, Violence, & Abuse.
- 14. Skogan WG, Statistics USBoJ, Affairs NUCfU (1981) Issues in the Measurement of Victimization: US Department of Justice, Bureau of Justice Statistics.
- 15. Plichta SB (2004) Intimate partner violence and physical health consequences. Journal of Interpersonal Violence 19: 1296–1323.
- 16. Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB (1996) The revised conflict tactics scales (CTS2). Journal of family issues 17: 283–316.
- 17. Straus MA (1979) Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. Journal of Marriage and the Family: 75–88.
- 18. Straus MA, Douglas EM (2004) A short form of the Revised Conflict Tactics Scales, and typologies for severity and mutuality. Violence and victims 19: 507–520.
- 19. Stephenson R, Rentsch C, Salazar LF, Sullivan PS (2011) Dyadic Characteristics and Intimate Partner Violence among Men Who Have Sex with Men. Western Journal of Emergency Medicine 12: 324.
- 20. Balsam KF, Rothblum ED, Beauchaine TP (2005) Victimization over the life span: a comparison of lesbian, gay, bisexual, and heterosexual siblings. Journal of Consulting and Clinical Psychology 73: 477.
- 21. Ghanem A, Little SJ, Drumright L, Liu L, Morris S, et al.. (2011) High-Risk Behaviors Associated with Injection Drug Use Among Recently HIV-Infected Men Who Have Sex with Men in San Diego, CA. AIDS and Behavior: 1–9.
- 22. Welles SL, Corbin TJ, Rich JA, Reed E, Raj A (2010) Intimate Partner Violence Among Men Having Sex with Men, Women, or Both: Early-Life Sexual and Physical Abuse as Antecedents. Journal of Community Health: 1–9.
- 23. Conron KJ, Mimiaga MJ, Landers SJ (2010) A population-based study of sexual orientation identity and gender differences in adult health. American journal of public health: AJPH. 2009.174169 v174162.
- 24. Rhodes S, McCoy TP, Wolfson M (2009) Behavioral risk disparities in a random sample of self-identifying gay and non-gay male university students. Journal of Homosexuality 56: 1083–1100.
- 25. Friedman MS, Marshal MP, Stall R, Cheong JW, Wright ER (2008) Gay-related development, early abuse and adult health outcomes among gay males. AIDS and Behavior 12: 891–902.
- 26. Bimbi DS, Palmadessa NA, Parsons JT (2008) Substance use and domestic violence among urban gays, lesbians and bisexuals. Journal of LGBT Health Research 3: 1–7.
- 27. Feldman MB, Díaz RM, Ream GL, El-Bassel N (2007) Intimate partner violence and HIV sexual risk behavior among Latino gay and bisexual men. Journal of LGBT health research 3: 9.
- 28. Koblin B, Torian L, Xu G, Guilin V, Makki H, et al. (2006) Violence and HIV-related risk among young men who have sex with men. AIDS care 18: 961–967.
- 29. Muhib FB, Lin LS, Stueve A, Miller RL, Ford WL, et al. (2001) A venue-based method for sampling hard-to-reach populations. Public Health Reports 116: 216.
- 30. Koblin BA, Torian L, Xu G, Guilin V, Makki H, et al. (2006) Violence and HIV-related risk among young men who have sex with men. AIDS care 18: 961–967.
- 31. Relf MV, Huang B, Campbell J, Catania J (2004) Gay identity, interpersonal violence, and HIV risk behaviors: An empirical test of theoretical relationships among a probability-based sample of urban men who have sex with men. Journal of the Association of Nurses in AIDS Care 15: 14–26.
- 32. Stall R, Mills TC, Williamson J, Hart T, Greenwood G, et al. (2003) Association of co-occurring psychosocial health problems and increased vulnerability to HIV/AIDS among urban men who have sex with men. American journal of public health 93: 939.
- 33. Klitzman RL, Pope Jr HG, Hudson JI (2000) MDMA (“Ecstasy”) abuse and high-risk sexual behaviors among 169 gay and bisexual men. The American journal of psychiatry 157: 1162.
- 34. Nieves-Rosa LE, Carballo-Diéguez A, Dolezal C (2000) Domestic abuse and HIV-risk behavior in Latin American men who have sex with men in New York city. Journal of Gay & Lesbian Social Services 11: 77–90.
- 35. Stephenson R, de Voux A, Sullivan PS (2011) Intimate Partner Violence and Sexual Risk-taking among Men Who Have Sex with Men in South Africa. Western Journal of Emergency Medicine 12: 343.
- 36. Kalichman S, Rompa D (1995) Sexually coerced and noncoerced gay and bisexual men: Factors relevant to risk for human immunodeficiency virus (HIV) infection. Journal of Sex Research 32: 45–50.
- 37. Kalichman S, Benotsch E, Rompa D, Gore-Felton C, Austin J, et al. (2001) Unwanted sexual experiences and sexual risks in gay and bisexual men: Associations among revictimization, substance use, and psychiatric symptoms. Journal of Sex Research 38: 1–9.
- 38. MacKellar D, Valleroy L, Karon J, Lemp G, Janssen R (1996) The Young Men's Survey: methods for estimating HIV seroprevalence and risk factors among young men who have sex with men. Public Health Reports 111: 138.
- 39. Xia Q, Tholandi M, Osmond DH, Pollack LM, Zhou W, et al. (2006) The effect of venue sampling on estimates of HIV prevalence and sexual risk behaviors in men who have sex with men. Sexually transmitted diseases 33: 545.
- 40. Dahlstrom WG, Brooks JD, Peterson CD (1990) The Beck Depression Inventory: Item order and the impact of response sets. Journal of Personality Assessment 55: 224–233.