1 Hate Crimes and Stigma-Related Experiences Among Sexual Minority Adults in the United States: Prevalence Estimates from a National Probability Sample Gregory M. Herek Department of Psychology, University of California, Davis This is a preprint of a paper that has been accepted for publication in the Journal of Interpersonal Violence. Minor changes to the paper may be made in the course of copy editing. Abstract* Using survey responses collected via the Internet from a U.S. national probability sample of gay, lesbian, and bisexual adults (N = 662), this article reports prevalence estimates of criminal victimization and related experiences based on the target’s sexual orientation. Approximately 20% of respondents reported having experienced a person crime or property crime based on their sexual orientation, about half had experienced verbal harassment, and more than one in ten reported having experienced employment or housing discrimination. Gay men were significantly more likely than lesbians or bisexuals to experience violence or property crimes, as well as harassment and verbal abuse. Employment and housing discrimination were significantly more likely among gay men and lesbians than among bisexual men and women. Implications for future research and policy are discussed. In 1989, the National Institute of Mental Health convened an expert panel on antigay violence to review existing knowledge and identify research needs. The panel named collection of prevalence data as a top research priority and urged that such data be obtained from probability samples AUTHOR NOTE. Data collection for this project was supported by a grant from the Gill Foundation. The author thanks Aaron Norton, Thomas Allen, and Charles Sims for their assistance. Address correspondence to Gregory Herek, Psychology Department, University of California, Davis, CA 95616. when possible (Herek & Berrill, 1990). Since then, data collected by the Federal Bureau of Investigation (e.g., Federal Bureau of Investigation, 2005), the National Crime Victimization Survey (Harlow, 2005), and the National Coalition of Anti-Violence Programs (e.g., National Coalition of Anti-Violence Programs, 2005) have shown that criminal enactments of sexual stigma are widespread (Herek & Sims, 2007). However, prevalence data on criminal victimization among lesbians, gay men, and bisexuals are still fragmentary and derived almost entirely from convenience samples. Berrill (1992) compiled data from 24 published and unpublished studies conducted between 1977 and 1991 by academic researchers and community-based organizations, all but one of them using convenience samples of gay men, lesbians, and bisexuals. Across studies, a median of 9% of respondents reported having been the target of an aggravated assault (i.e., assault with a weapon) because of their sexual orientation; 17% reported simple physical assault (i.e., without a weapon); 19% reported vandalism of their personal property; 44% had been threatened with violence; 33% had been chased or followed; 25% reported having objects thrown at them; 13% had been spat upon; and 80% had been verbally harassed (Berrill, 1992). Most of the studies did not report data separately by respondents’ gender or sexual orientation. More recently, in a study of 2,259 lesbian, gay, and bisexual adults in the greater Sacramento (CA) area, 28% of gay men, 19% of lesbians, 27% of bisexual men, and 15% of bisexual women reported having experienced some type 2 of criminal victimization since age 16 because of their sexual orientation (ns = 898, 980, 191, and 190, respectively; Herek, Gillis, & Cogan, 1999). This includes respondents who reported experiencing a simple or aggravated assault (13% of gay men, 7% of lesbians, 11% of bisexual men, and 5% of bisexual women) or a sexual assault based on their sexual orientation (4% of gay men, 3% of lesbians, 7% of bisexual men, and 4% of bisexual women). Other research has focused on particular age groups in sexual minority communities. In a sample of 1,248 young gay and bisexual men (M = 23 years, range = 18-27 years) recruited in three southwestern U.S. cities, 5% reported they had experienced physical violence because of their sexual orientation during the previous 6 months (Huebner, Rebchook, & Kegeles, 2004). In a sample of 194 lesbian, gay, and bisexual youths (age range = 15-21 yrs) recruited from service agencies across the United States, 9% reported at least one aggravated assault based on their sexual orientation, 18% had experienced a simple assault, 22% had been sexually assaulted, and 44% had been threatened with attack (Pilkington & D’Augelli, 1995). In a study of sexual minority youths recruited through community-based organizations in New York City and its suburbs, D’Augelli and his colleagues found that 11% reported physical violence based on their sexual orientation, 9% reported sexual violence, and 78% reported verbal threats or harassment (D’Augelli, Grossman, & Starks, 2006). At the other end of the age continuum, D’Augelli and Grossman (2001) documented the lifetime occurrence of hate crime victimization among older (> 59 years) lesbian, gay, and bisexual adults recruited from across the United States (n = 416). In that sample, 16% had been physically attacked at some time in their life, 7% had been sexually assaulted, 11% reported having had objects thrown at them, and 29% had been threatened with violence. It is difficult to use these studies to derive an estimate of the population prevalence of hate crime victimization against US sexual minorities because of variations in how they categorized crimes, the time frames within which they assessed victimization, and how they reported their data (e.g., some studies reported findings separately for men and women, or homosexuals and bisexuals, whereas others did not). Moreover, because nearly all of the surveys used convenience samples, the extent to which their results describe the entire U.S. gay, lesbian, and bisexual population cannot be determined. Prevalence data collected in three studies with probability samples further confirm that hate crime victimization is widespread. In a 1989 San Francisco Examiner national telephone survey, 5% of gay men (n = 287) and 10% of lesbians (n = 113) reported having been physically abused or assaulted in the previous year because they were gay (Results of Poll, 1989).* In a 2000 Kaiser Family Foundation (KFF) survey of 405 lesbian, gay, and bisexual adults residing in major US population centers, 32% of respondents said they had been targeted for violence against their person or property because of their sexual orientation (Kaiser Family Foundation, 2001). In a probability sample of 912 Latino men who have sex with men, recruited from social venues in New York, Miami, and Los Angeles, 10% reported they had experienced violence as an adult because of their sexual orientation or femininity (Díaz, Ayala, Bein, Henne, & Marín, 2001). These pioneering surveys yielded valuable data but are nevertheless limited in the generalizability of their findings. Neither the San Francisco Examiner poll nor the KFF poll were published in a peer-reviewed journal. Few published details are available about the methodology of the Examiner survey, making it particularly difficult to evaluate. The KFF poll sampled respondents only in 15 US cities, which may limit its generalizability, and the survey of Latino men focused on specific venues in only 3 cities. The present paper addresses a gap in current knowledge by reporting data on the prevalence of antigay violence and related experiences in a national probability sample of sexual minority * This is the one study cited by Berrill (1992) that was based on a probability sample. 3 adults. Violence against individuals because of their presumed sexual orientation is conceptualized here as a manifestation of sexual stigma, that is, society’s negative regard for any nonheterosexual behavior, identity, relationship, or community (Herek, 2004, 2008). Sexual stigma is a cultural belief system through which homosexuality is denigrated, discredited, and socially constructed as invalid relative to heterosexuality (Herek, 2008; Herek, Chopp, & Strohl, 2007). As with other forms of stigma, sexual stigma is expressed through society’s institutions (e.g., through discriminatory laws and policies) and by its individual members. Individual enactments of stigma can range from personal ostracism to criminal attacks against people perceived to be homosexual or bisexual. Sexual stigma has important consequences for sexual minority individuals. Whereas being the target of any violent crime can have negative psychological effects, victims of antigay violence are at heightened risk for psychological distress (Herek et al., 1999; Mills et al., 2004; see also McDevitt, Balboni, Garcia, & Gu, 2001). Hate crimes may have especially negative psychological sequelae because they attack a core aspect of the victim’s personal identity and community membership, components of the self that are particularly important to sexual minority individuals because of the stresses created by sexual stigma (Garnets, Herek, & Levy, 1990; Herek et al., 1999, 2007). In addition to stigma’s direct effects, sexual minority individuals’ awareness of its extent and their expectancies about when it will be enacted create a subjective sense of threat. This felt stigma (e.g., Scambler & Hopkins, 1986) can motivate them to engage in a variety of proactive behaviors aimed at shielding themselves from enacted stigma. Such strategies (e.g., concealing their sexual orientation) can protect them from directly experiencing enacted stigma but also restrict their opportunities for having normal social interaction and receiving social support (Herek, 2008). The present study assessed the prevalence of eight forms of enacted stigma and three aspects of felt stigma in a national probability sample of sexual minority individuals. The sample consisted of self-identified lesbian, gay, and bisexual adults randomly selected from an existing panel of more than 40,000 US households. Panel members were recruited by Knowledge Networks (KN) using random-digit dialing (RDD) methods. Upon joining the panel, respondents agreed to participate regularly in on-line surveys, and were provided with free Internet access and equipment if they did not already have it. Thus, in contrast to Internet studies with volunteer samples recruited via the Web, the KN panel includes individuals who would not otherwise have Internet access because of their financial or social situation. Reflecting this fact, KN samples are demographically similar to other RDD samples and more closely match the US population than do other Internet samples (Krosnick & Chang, 2001). Samples drawn from the KN panel have been used extensively in academic and government research (Knowledge Networks, 2007). Method Sample and Procedure Members of the KN panel routinely answer a battery of background questions, including an item about their sexual orientation. Using these data, a sample was drawn of 902 English- speaking adults (> 18 years) who had previously responded affirmatively to the question, “Are you yourself gay, lesbian, or bisexual?” Following standard KN procedures, they each received an e-mail invitation to complete the questionnaire at their convenience. A follow-up e-mail was sent to nonresponders after approximately one week. Neither invitation mentioned sexual orientation. As with all KN surveys, panel members were free to decline to participate. A total of 775 individuals (86%) accessed the questionnaire between September 13 and October 7, 2005. In response to a screening question at the beginning of the survey, 56 indicated they were heterosexual. They were thanked for their assistance and their survey was terminated. This left 719 self-identified lesbian, gay, and bisexual respondents who completed the questionnaire. Within that group, 56 4 households were represented by multiple respondents. In these cases, one respondent was randomly selected from the household for inclusion in the data set, yielding a final sample of 662. Taking into account all attrition that has occurred in the KN panel since the earliest stage of RDD recruitment, the response rate for the present study was 30% (American Association for Public Opinion Research, 2006 [Formula 3]). This is a relatively high rate for contemporary commercial surveys (Holbrook, Krosnick, & Pfent, 2008). Measures The present data are based on a subset of questions from a larger questionnaire. Only the relevant variables are discussed here. Enacted Stigma Eight questions were posed about how often respondents had experienced different forms of enacted stigma since age 18 because someone perceived them to be lesbian or bisexual (female respondents) or gay or bisexual (male respondents). The response options were never, once, twice, and three or more times. The questions assessed experiences of enacted stigma within three general categories: criminal victimization, harassment and threats, and discrimination. To assess criminal victimization, respondents were asked how often they had experienced a crime against their person (“You were hit, beaten, physically attacked, or sexually assaulted”) or property (“You were robbed, or your property was stolen, vandalized, or purposely damaged”) or an attempted crime (“Someone tried to attack you, rob you, or damage your property, but they didn’t succeed”) based on their sexual orientation. To assess harassment and threats, respondents were asked about their experiences with antigay threats (“Someone threatened you with violence”) and harassment (“Someone verbally insulted or abused you” and “Someone threw an object at you”). To assess discrimination, respondents were asked about their experiences with sexual orientation discrimination in employment (“You were fired from your job or denied a job or promotion”) and housing (“You were prevented from moving into a house or apartment by a landlord or realtor”). Felt Stigma Felt stigma was assessed with three statements, each accompanied by a 5-point Likert-type response scale ranging from strongly agree to strongly disagree. Utilizing Web software capabilities, the item wording was customized by inserting respondents’ preferred term for characterizing their own sexual orientation (gay, lesbian, bisexual, queer, etc.); each respondent had selected this term earlier in the questionnaire. The three statements (as worded for respondents who indicated gay was their preferred self-identifying term) were (1) “Most people where I live think less of a person who is gay.” (2) “Most employers where I live will hire openly gay people if they are qualified for the job.” (3) “Most people where I live would not want someone who is openly gay to take care of their children.” Demographic Data The survey included a question about the respondent’s specific sexual orientation (bisexual or homosexual). Other demographic data – including respondents’ gender, age, race and ethnicity, and highest educational level completed – had been collected previously by Knowledge Networks. Weighting Because the KN panel was recruited using RDD methods, the initial sample design yielded a simple random sample with equal probability of selection for all US households with a telephone. However, the actual probability of selection for individual respondents was affected by multiple factors (e.g., differences in household size and the number of telephone lines in a single household). Following standard procedures for RDD samples (e.g., Kish, 1965), design weights were calculated and assigned to each case to adjust for unequal probability of selection.* * Design weights were computed to account for the following: (a) variations among respondents in the number of adults in the household; (b) variations in 5 Because the use of weighted data necessitates special analytic procedures to correct standard errors (Lee & Forthofer, 2006), analyses were conducted using statistical software packages that permit such correction (STATA and SPSS Complex Samples). Results Sample Characteristics The final sample consisted of 311 women (152 lesbians, 159 bisexuals) and 351 men (241 gay men, 110 bisexuals). When design weights were applied, the weighted sample was 14.6% lesbian (95% Confidence Interval [CI] = 11.6% - 18.2%), 34.8% gay male (CI = 28.9% - 41.2%), 23.7% bisexual female (CI = 18.8% - 29.3%), and 26.9% bisexual male (CI = 19.1% - 36.4%). Unless otherwise indicated, the weighted data are used hereafter. The respondents’ mean age was 39 years (CI = 37.2 - 40.8). Gay men were significantly older (M = 45.3) than lesbians (M = 40.1), bisexual men (M = 36.6), and bisexual women (M = 31.8 years). In addition, lesbians were significantly older than bisexual women (for all statistically significant differences reported here, p < .05). The sample was 65.4% non-Hispanic White, 15.6% non-Hispanic Black, and 12.5% Hispanic, with the remaining 6.5% from other races or of mixed race or ethnicity. Compared to lesbians and bisexual women, significantly fewer bisexual men were non-Hispanic White (43.0%, compared to 74.4% of lesbians and 77.5% of bisexual women). Bisexual men were also substantially less likely than gay men (70.5%) to be non-Hispanic White but the the number of telephone lines in the household; (c) undersampling of telephone numbers for which matching addresses could not be obtained; (d) oversampling of Black and Hispanic populations; (e) oversampling of households with personal computers and Internet access; (f) undersampling of households in areas not covered by MSN Web TV; (g) slight overrepresentation of Chicago and Los Angeles, cities where early pilot testing for the panel was conducted; (h) oversampling of the 4 most populous states (CA, NY, FL, TX) and Central regional states early in the life of the panel. difference was not statistically significant. Bisexual men were more likely than other respondents to be Hispanic (20.6%) or non- Hispanic Black (28.6%), but the differences were not statistically significant. Most respondents had earned a bachelor’s degree (32.9%) or attended some college (31.4%). Only 7.3% did not have a high school diploma or equivalent. Compared to gay men and lesbians, bisexual men were significantly less likely to have a bachelor’s degree: 15.9% of bisexual men had a degree, compared to 46.4% of gay men and 40.9% of lesbians. A more detailed demographic description of the sample will be reported elsewhere. Experiences With Enacted Stigma Table 1 reports response frequencies and 95% confidence intervals (CIs) for each question about enacted stigma for the four sexual orientation groups. The response options are collapsed into three categories: once, twice or more, and never. Table 2 translates the Table 1 data into more easily interpretable categories. It details the proportions of respondents in each sexual orientation group who reported ever experiencing enacted stigma (combining the response options of once, twice, and three or more times), with specific forms of enacted stigma combined into larger categories when appropriate. ______________________________ Insert Tables 1 and 2 about here ______________________________ As shown in the last column of Table 2, 13.1% of the sample reported having experienced violence against their person based on their sexual orientation at least once during their adult life, and 14.9% had experienced a property crime. Approximately 1 in 5 reported experiencing one or both types of crime, and this proportion increased to about 25% when attempted crimes were included. Overall, 12.5% of respondents reported having objects thrown at them because of their sexual orientation, 23.4% had been threatened with violence, and 49.2% had experienced verbal abuse. More than one respondent in ten (11.2%) reported having experienced housing or employment 6 discrimination because of her or his sexual orientation. Across the sexual orientation groups, gay men reported the highest levels of enacted stigma. They were significantly more likely than others to report experiences of antigay violence (24.9%) and antigay property crimes (28.1%). More than one third of gay men (37.6%) reported experiencing one or both types of crimes, compared to 12.5% of lesbians, 10.7% of bisexual men, and 12.7% of bisexual women. As indicated by the overlapping CIs in Table 1, differences among the latter groups were too small to be considered reliable. Gay men were significantly more likely than bisexual men to report having had objects thrown at them (21.1% vs. 5.6%). They were significantly more likely than lesbians and bisexual women to have been threatened with violence (35.4% of gay men vs. 17.3% of lesbians and 14.1% of bisexual women), and significantly more likely than bisexual women to report verbal abuse because of their sexual orientation (63.0% vs. 34.3%). Employment and housing discrimination were significantly more likely among gay men and lesbians (reported by 17.7% and 16.3%, respectively) than among bisexual men and women (3.7% and 6.8%, respectively). As noted above, the groups differed significantly in age, race and ethnicity, and educational level, which might account for the observed differences on enacted stigma. To test this hypothesis, separate logistic regression equations were computed for each type of enacted stigma (violence, objects thrown, etc.) as well as for the combined categories of criminal victimization (comprising violence, property crime, and attempted crime) and harassment (comprising threats, objects thrown, and verbal abuse). In each equation, the dependent variable was dichotomized (ever experiencing that form of enacted stigma vs. never experiencing it). The independent variables were age, education, race and ethnicity (coded as non-Hispanic White vs. Hispanic or non-White), and sexual orientation (coded as gay male vs. other groups). In the equations for the combined categories of criminal victimization and harassment, the odds ratios (ORs) for sexual orientation were statistically significant, indicating gay men were significantly more likely to report experiencing both categories of enacted stigma, even when demographic differences are controlled. (For criminal victimization, OR = 2.45, 95% CI = 1.36 – 4.41; for harassment, OR = 2.10, CI = 1.23 – 3.60.) The same pattern was observed for all of the individual forms of enacted stigma except attempted crimes (for which none of the variables yielded a statistically significant OR). A similar procedure was followed to assess whether lesbians and gay men’s greater likelihood of experiencing employment or housing discrimination was explained by group differences in age, education, and race. In this equation, sexual orientation was coded as homosexual versus bisexual. The OR for sexual orientation (3.29, CI = 1.58 – 6.85) was significant, indicating that the differences in discrimination were not due to demographic differences. Felt Stigma As shown in Table 3, substantial minorities of respondents expressed some degree of felt stigma. More than one third (34.6%) agreed (strongly or somewhat) that most people where they live think less of a sexual minority individual, 25.5% disagreed that most employers will hire qualified sexual minority individuals, and 40.6% agreed that most people would not want a sexual minority individual to care for their children. Overall, a majority of respondents (54.7%, CI = 47.1 – 62.1%) gave at least one response indicative of felt stigma. As indicated by the overlapping CIs in Table 3, there were few significant differences among the sexual orientation groups in their responses to the individual felt stigma items. The only exception is that gay men were significantly less likely than lesbians to perceive sexual stigma in hiring: 54.8% agreed that most employers in their area will hire a qualified sexual minority person, compared to 32.3% of lesbians who agreed with this statement. Responses to the three statements were summed (with hiring question responses reversed) and divided by the number of items to create a felt 7 stigma scale (α = .71, using unweighted data). Scores could range from 1 (low felt stigma) to 5 (high felt stigma). As shown in the first row of Table 3, gay men scored the lowest on the scale and lesbians the highest (Ms = 2.79 and 3.11, respectively), but differences among the sexual orientation groups were not statistically significant. Association of Felt Stigma with Enacted Stigma Felt stigma scores were higher among respondents who reported having experienced person or property crimes or attempted crimes based on their sexual orientation compared to those who had not (Ms = 3.2 vs. 2.8, respectively), those who had experienced verbal threats or harassment compared to those who had not (Ms = 3.0 vs. 2.7, respectively), and those who had experienced employment or housing discrimination compared to those who had not (Ms = 3.2 and 2.8, respectively). None of these differences were statistically significant. However, significant differences were observed on these variables within some sexual orientation groups. Lesbians who said they had experienced crimes or attempted crimes scored significantly higher on felt stigma compared to those who had not (Ms = 3.6 vs. 3.0, respectively). The difference was also significant among bisexual men (Ms = 3.6 vs. 2.6, respectively), but not gay men (Ms = 2.9 vs. 2.8) or bisexual women (Ms = 3.7 vs. 2.9). Felt stigma was significantly higher among lesbians who reported having experienced employment or housing discrimination (M = 3.7 vs. 3.0 among those who had not). The difference was also significant for bisexual men (3.9 vs. 2.8), but not for gay men (2.9 vs. 2.8) or bisexual women (3.2 vs. 2.9). Discussion The present study yields the most reliable estimates to date of the prevalence of antigay victimization in the United States. The data indicate that approximately 20% of the US sexual minority population has experienced a crime against their person or property since age 18 that was based on their sexual orientation. With attempted crimes included, the proportion increases to roughly 25%. Harassment is considerably more widespread, with about half of sexual minority adults reporting verbal abuse at some time in their adult life as a consequence of their sexual orientation. More than one sexual minority adult in ten has experienced housing or employment discrimination because of her or his sexual orientation. The likelihood of experiencing victimization clearly is not uniform among sexual minorities. Gay men are at the greatest risk for person and property crimes. Approximately 38% of gay men in the present sample reported they had experienced one or both types of criminal victimization. Gay men also were more likely than lesbians and bisexuals to be harassed because of their sexual orientation. This pattern is consistent with previous findings that sexual minority men are at greater risk for antigay victimization than are sexual minority women (D’Augelli & Grossman, 2001; Herek et al., 1999). Several factors may account for it. Men are more likely than women to be victims of violent crime in general, especially crimes committed by strangers (Catalano & Bureau of Justice Statistics, 2005). Most such crimes are perpetrated by heterosexually-identified men, who tend to hold more hostile attitudes toward sexual minority males than toward sexual minority females (Herek, 2002a, 2002b). In addition, gay men may be more visible targets than sexual minority women and bisexual men because, for example, they may be more likely to frequent gay-oriented venues and the public spaces around them. Lesbians and gay men were significantly more likely than bisexuals to report discrimination based on their sexual orientation. This pattern probably cannot be attributed to attitudinal differences among the agents of discrimination because heterosexuals’ attitudes toward bisexuals tend to be somewhat more negative than their attitudes toward gay men and lesbians (Herek, 2002b). Instead, homosexual adults’ greater visibility probably makes them more vulnerable to discrimination in workplace and housing settings, compared to bisexuals. Additional data collected in the present study 8 suggest that bisexual men and women are less likely than gay men and lesbians to disclose their sexual orientation to others in a variety of social contexts, including the workplace. In addition, to the extent that homosexual adults are more likely than bisexuals to cohabit with a same-sex partner (because many coupled bisexuals have a different-sex partner), the former are probably more readily labeled as gay by landlords and realtors and thus are more subject to discrimination. As noted above, comparisons of prevalence estimates across previous studies are problematic because of differences in sampling strategies, question wording, time frames, and data reporting conventions. Estimates of lifetime victimization from the present data are higher than those reported in some published studies (e.g., D’Augelli & Grossman, 2001; Herek et al., 1999) but lower than in others (Kaiser Family Foundation, 2001). The difference between the current study and the KFF survey may be due to the fact that the latter was conducted only with urban residents, who are more likely than non- urban residents to experience crime of all sorts (Herek & Sims, 2007). Differences between the present study and those conducted by D’Augelli and Grossman (2001) and Herek et al. (1999) may result from the latter’s use of convenience samples whose representativeness cannot be determined; the current data probably provide a more accurate estimate of the extent of victimization experiences within the sexual minority population. The current study’s estimates of the extent of felt stigma in the sexual minority population are another unique contribution. About 55% of respondents manifested some degree of felt stigma. It tended to be higher among respondents who had experienced enactments of stigma. For criminal victimization and employment discrimination, this pattern was due mainly to significant differences among lesbians and bisexual men who had experienced enacted stigma versus those who had not. Because the data reported here are retrospective and cross- sectional, the causal and temporal direction of these relationships, if any, cannot be determined. Experiencing enacted stigma is likely to increase an individual’s subjective sense of vulnerability related to her or his sexual orientation (Herek et al., 1999), which could result in a positive correlation between enacted and felt stigma. However, other explanations for the pattern are also plausible. For example, persons with high levels of felt stigma may have a heightened sensitivity to the occurrence of stigma enactments and consequently may be more likely than others to attribute ambiguous incidents to stigma. Previous research on the cognitive strategies used by sexual minority crime victims for assessing their attackers’ motives, however, suggests that relatively unambiguous cues (e.g., antigay verbal abuse) often accompany hate crimes based on sexual orientation (Herek, Cogan, & Gillis, 2002). Despite the methodological advance represented by its use of a national probability sample, the present study has limitations. The data reported here may not accurately describe the experiences of sexual minority members of the KN panel who did not disclose their sexual orientation in response to the initial screening question. Moreover, the questions about victimization were brief and presented in a quasi-checklist format. This approach, which was necessitated by the limited resources available for the present study, did not permit detailed assessment of specific incidents nor did it provide respondents with extensive memory cues (e.g., questions about the date and types of victimization in each incident) to assist them in accurately recalling their experiences with stigma. Whereas experiences with dramatic enactments of stigma, such as serious assaults, are not likely to be forgotten, relatively mundane episodes may be more difficult to recall (Herek, Gillis, Cogan, & Glunt, 1997). In future research, therefore, it will be important to replicate the present findings using more detailed queries about the events surrounding each experience of enacted stigma, and to provide respondents with a variety of memory aids to maximize the accuracy of their self-reports. In addition, as with any probability sample, some error due to sampling can be assumed to be associated with the present data. The great advantage of the use of a probability sample is 9 that, in contrast to convenience samples, sampling error can be quantified and expressed in the form of confidence intervals. Nevertheless, other sources of error may also affect the data. As with most modern surveys, for example, the initial sampling frame excluded US households without a telephone as well as group living quarters (e.g., barracks and dormitories). These limitations to the sample coverage should be kept in mind when interpreting the data. In addition, measurement error may have resulted from the specific wording of the survey questions or from other factors such as the order in which questions were presented. As with all surveys, the data represent a snapshot of the population at the time the study was fielded (autumn of 2005). For these reasons, more studies with national probability samples are needed. Even in the absence of additional surveys with probability samples, however, the present data have important policy implications. They demonstrate that the experience of violence and property crime is disturbingly widespread among sexual minority adults, especially gay men. Thus, they highlight the ongoing need for criminal justice programs to prevent and deter such crimes, as well as the need for victim services that will help to alleviate the physical, economic, social, and psychological consequences of such crimes (e.g., Herek et al., 1999; Herek & Sims, 2007). In addition, the psychological toll of antigay hate crimes and harassment should be considered by mental health professionals and by researchers conducting studies of psychological distress and well being in this population. Some research, for example, suggests that individuals who have engaged in homosexual behavior may be at greater risk than exclusively heterosexual adults for some forms of psychological distress (Herek & Garnets, 2007). Most studies, however, have not assessed how experiences of victimization and harassment might explain this pattern (for an exception, see Mays & Cochran, 2001). Given the association between antigay victimization and heightened psychological distress (Herek et al., 1999) and the present study’s finding that such victimization has been experienced by roughly 1 in 8 lesbians and bisexuals, and nearly 4 gay men in 10, it seems likely that the associations observed in past research between sexual orientation and psychological problems are attributable, at least in part, to such victimization. Moreover, the fact that more than half of the respondents in the present study experienced some degree of felt stigma related to their sexual orientation further highlights the extent to which sexual minorities are subjected to stressors that heterosexuals do not experience (e.g., Meyer, 2003). Ever since the Hate Crimes Statistics Act became law in 1990, marking the federal government’s first official recognition of the problem of violence and crime against people because of their sexual orientation, researchers have attempted to document the extent and prevalence of antigay victimization. The present study makes an important contribution to this effort. Such data can assist law enforcement agencies, service providers, and sexual minority communities in alleviating and preventing the problems created by sexual stigma. References American Association for Public Opinion Research. (2006). Standard definitions: Final dispositions of case codes and outcome rates for surveys. Lenexa, KS: Author. Berrill, K. (1992). Antigay violence and victimization in the United States: An overview. In G. Herek & K. Berrill (Eds.), Hate crimes: Confronting violence against lesbians and gay men (pp. 19-45). Thousand Oaks, CA: Sage Publications. Catalano, S., & Bureau of Justice Statistics. (2005). Criminal victimization, 2004. Retrieved February 19, 2006, from http://www.ojp.usdoj.gov/bjs/abstract/ cv04.htm D’Augelli, A., & Grossman, A. (2001). Disclosure of sexual orientation, victimization and mental health among lesbian, gay, and bisexual older adults. Journal of Interpersonal Violence, 16, 1008-1027. D’Augelli, A., Grossman, A., & Starks, M. (2006). Childhood gender atypicality, victimization, and PTSD among lesbian, gay, and bisexual youth. Journal of Interpersonal Violence, 21, 1462-1482. 10 Díaz, R., Ayala, G., Bein, E., Henne, J., & Marín, B. (2001). The impact of homophobia, poverty, and racism on the mental health of gay and bisexual Latino men: Findings from 3 US cities. American Journal of Public Health, 91, 927-932. Federal Bureau of Investigation. (2005). Hate crime statistics 2004. Washington, DC: U.S. Department of Justice. Garnets, L., Herek, G., & Levy, B. (1990). Violence and victimization of lesbians and gay men: Mental health consequences. Journal of Interpersonal Violence, 5, 366-383. Harlow, C. (2005). Hate crime reported by victims and police. Washington, DC: U.S. Department of Justice Retrieved July 4, 2006, from http://www .ojp.usdoj.gov/bjs/pub/pdf/hcrvp.pdf Herek, G. (2002a). Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly, 66, 40-66. Herek, G. (2002b). Heterosexuals’ attitudes toward bisexual men and women in the United States. Journal of Sex Research, 39, 264-274. Herek, G. (2004). Beyond “homophobia”: Thinking about sexual stigma and prejudice in the twenty-first century. Sexuality Research and Social Policy, 1(2), 6-24. Herek, G. M. (2008). Understanding sexual stigma and sexual prejudice in the United States: A conceptual framework. In D. Hope (Ed.), Contemporary perspectives on lesbian, gay and bisexual identities: The 54th Nebraska Symposium on Motivation. New York: Springer. Herek, G., & Berrill, K. (1990). Anti-gay violence and mental health: Setting an agenda for research. Journal of Interpersonal Violence, 5, 414-423. Herek, G., Chopp, R., & Strohl, D. (2007). Sexual stigma: Putting sexual minority health issues in context. In I. Meyer & M. Northridge (Eds.), The health of sexual minorities: Public health perspectives on lesbian, gay, bisexual, and transgender populations (pp. 171-208). New York: Springer. Herek, G., Cogan, J., & Gillis, J. (2002). Victim experiences in hate crimes based on sexual orientation. Journal of Social Issues, 58, 319-339. Herek, G., & Garnets, L. (2007). Sexual orientation and mental health. Annual Review of Clinical Psychology, 3, in press. Herek, G., Gillis, J., & Cogan, J. (1999). Psychological sequelae of hate crime victimization among lesbian, gay, and bisexual adults. Journal of Consulting and Clinical Psychology, 67, 945-951. Herek, G., Gillis, J., Cogan, J., & Glunt, E. (1997). Hate crime victimization among lesbian, gay, and bisexual adults: Prevalence, psychological correlates, and methodological issues. Journal of Interpersonal Violence, 12, 195-215. Herek, G., & Sims, C. (2007). Sexual orientation and violent victimization: Hate crimes and intimate partner violence among gay and bisexual males in the United States. In R.J. Wolitski, R. Stall, & R.O. Valdiserri (Eds.), Unequal opportunity: Health disparities among gay and bisexual men in the United States. New York: Oxford University Press. Holbrook, A. L., Krosnick, J. A., & Pfent, A. (2008). The causes and consequences of response rates in surveys by the news media and government contractor survey research firms. In J. Lepkowski, C. Tucker, J.M. Brick, E. D. de Leeuw, L. Japec, P.J. Lavrakas, M.W. Link, & R.L. Sangster (Eds.), Advances in telephone survey methodology (pp. 499- 528). New York: Wiley. Huebner, D., Rebchook, G., & Kegeles, S. (2004). Experiences of harassment, discrimination, and physical violence among young gay and bisexual men. American Journal of Public Health, 94, 1200- 1203. Kaiser Family Foundation. (2001). Inside-out: A report on the experiences of lesbians, gays, and bisexuals in America and the public’s view on issues and politics related to sexual orientation. Menlo Park, CA: Author. Retrieved November 14, 2001, from http://www.kff.org . Kish, L. (1965). Survey sampling. New York: Wiley. Knowledge Networks. (2007). Knowledge Networks bibliography: Articles and presentations based on KN collected data, analysis, methodology. Retrieved April 30, 2007, from http://www.knowledgenetworks .com/ganp/docs/ Krosnick, J.A. & Chang, L. (2001). A comparison of the random digit dialing telephone survey methodology with Internet survey methodology as implemented by Knowledge Networks and Harris Interactive. Columbus, OH: Department of Psychology, Ohio State University Retrieved November 1, 2002, from http://www .knowledgenetworks.com/ganp/docs/OSU Mode Effects 2001.pdf 11 Lee, E., & Forthofer, R. (2006). Analyzing complex survey data (2nd ed.). Thousand Oaks: Sage. Mays, V. M., & Cochran, S. D. (2001). Mental health correlates of perceived discrimination among lesbian, gay, and bisexual adults in the United States. American Journal of Public Health, 91, 1869-1876. McDevitt, J., Balboni, J., Garcia, L., & Gu, J. (2001). Consequences for victims: A comparison of bias- and non-bias-motivated assaults. American Behavioral Scientist, 45, 697-713. Meyer, I. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin, 129, 674-697. Mills, T., Paul, J., Stall, R., Pollack, L., Canchola, J., Chang, Y., Moskowitz, J., & Catania, J. (2004). Distress and depression in men who have sex with men: The Urban Men’s Health Study. American Journal of Psychiatry, 161, 278-285. National Coalition of Anti-Violence Programs. (2005). Anti-lesbian, gay, bisexual, and transgender violence in 2004. New York: Author. Retrieved May 1, 2006, from http://www.ncavp.org/ Pilkington, N. W., & D’Augelli, A. R. (1995). Victimization of lesbian, gay, and bisexual youth in community settings. Journal of Community Psychology, 23, 34-56. Results of poll. (1989, June 6). San Francisco Examiner, p. A-19. Scambler, G., & Hopkins, A. (1986). Being epileptic: Coming to terms with stigma. Sociology of Health and Illness, 8, 26-43. 12 Table 1 Self-Reported Experiences With Enacted Stigma: Population Estimates for Individual Items Group (Unweighted N) Type of Enacted Stigma Frequency Experienced Gay Men (n = 241) Lesbians (n = 152) Bisexual Men (n = 110) Bisexual Women (n = 159) Total (n = 662) Violence Once 13.3% 2.1% 5.2% 5.7% 7.7% 7.7 – 22.1 0.8 - 5.4 2.0 – 12.5 2.6 – 12.1 5.1 -11.4 11.6% 5.0% 1.7% 1.0% 5.5% Twice or More 6.6 - 19.6 2.2 - 10.9 0.5 - 5.4 0.3 - 3.4 3.4 - 8.6 Never 75.1% 92.9% 93.1% 93.3% 86.9% 65.5 - 82.7 86.9 - 96.3 85.5 - 96.9 87.0 – 96.7 82.4 - 90.3 Once 13.1% 3.8% 4.1% 5.3% 7.5% 8.1 – 20.5 1.7 – 8.4 1.6 – 10.2 1.5 – 17.7 5.0 - 11.1 15.0% 6.4% 3.6% 1.0% 7.4% Twice or More 10.0 - 21.8 3.2 - 12.4 1.5 - 8.6 0.2 – 4.8 5.3 – 10.3 Never 71.9% 89.8% 92.3% 93.7% 85.1% Property Crime 63.4 - 79.1 83.2 - 94.0 84.9 – 96.2 82.2 - 98.0 80.7 - 88.6 Once 12.3% 3.1% 6.2% 3.8% 7.3% 7.5 - 19.5 1.2 - 7.7 2.7 – 13.7 1.4 - 9.9 5.0 - 10.5 9.2% 5.2% 10.0% 1.8% 7.1% Twice or More 4.7 - 17.2 1.7 - 14.5 2.4 – 34.1 0.5 - 6.6 3.8 - 13.1 Never 78.5% 91.7% 83.7% 94.4% 85.6% Attempted Crime 69.7 - 85.3 83.2 - 96.1 64.0 – 93.7 88.0 – 97.5 79.8 - 89.9 Once 11.4% 6.6% 3.5% 3.4% 6.7% 7.3 - 17.6 2.8 - 14.7 1.1 – 10.1 1.4 - 8.0 4.6 - 9.5 9.6% 8.0% 2.1% 3.4% 5.9% Twice or More 4.8 - 18.4 4.4 -14.2 0.7 - 6.0 1.3 - 8.2 3.7 - 9.2 Never 78.9% 85.4% 94.4% 93.2% 87.5% Objects Thrown 70.2 -85.6 77.0 – 91.1 87.6 – 97.6 87.5 – 96.4 83.4 – 90.6 Table continues 13 Table 1 (continued) Group (Unweighted N) Type of Enacted Stigma Frequency Experienced Gay Men (n = 241) Lesbians (n = 152) Bisexual Men (n = 110) Bisexual Women (n = 159) Total (n = 662) Once 14.0% 3.5% 5.7% 9.0% 9.1% 9.0 - 21.2 1.8 - 6.7 2.4 - 12.8 3.8 - 19.8 6.4 - 12.8 21.4% 13.8% 13.3% 5.2% 14.3% Twice or More 14.7 - 30.1 8.0 - 22.8 4.3 - 34.5 2.6 - 10.0 10.0 - 20.0 Never 64.6% 82.7% 81.0% 85.9% 76.6% Threatened With Violence 55.4 - 72.8 73.8 - 89.0 61.9 - 91.8 75.8 - 92.2 70.5 - 81.8 Once 16.7% 10.8% 7.8% 10.3% 11.9% 10.7 - 25.0 6.4 - 17.7 3.8 - 15.6 5.9 - 17.5 8.9 – 15.8 46.4% 43.7% 33.6% 23.9% 37.3% Twice or More 37.5 - 55.5 33.9 - 54.1 16.4 - 56.6 16.0 - 34.1 30.5 - 44.6 Never 37.0% 45.5% 58.6% 65.7% 50.8% Verbal Abuse 28.6 - 46.2 35.7 - 55.6 37.3 - 77.1 54.9 - 75.2 43.6 - 58.0 Once 9.4% 13.3% 3.1% 4.1% 7.0% 6.0 – 14.3 7.8 - 21.6 1.0 - 9.2 2.0 - 8.1 5.1 - 9.5 6.3% 2.7% 0.3% 2.3% 3.2% Twice or More 2.7 - 13.7 1.1 - 6.7 0 - 2.0 0.9 - 5.7 1.7 - 5.8 Never 84.4% 84.0% 96.6% 93.6% 89.8% Job Discrimination 77.1 - 89.6 75.6 - 89.9 90.7 - 98.8 89.0 - 96.4 86.5 - 92.4 Once 4.6% 3.8% 1.5% 0.9% 2.8% 1.8 - 11.4 1.4 - 9.8 0.2 - 10.0 0.2 - 3.6 1.4 - 5.4 1.9% 1.3% 0.3% 0.4% 1.0% Twice or More 0.8 - 4.7 0.5 - 3.6 0 - 2.3 0.1 - 2.6 0.5 - 2.0 Never 93.5% 94.9% 98.2% 98.8% 96.2% Housing Discrimination 87.1 - 96.8 89.2 - 97.7 90.8 - 99.7 96.1 - 99.6 93.6 - 97.7 Table reports parameter estimates and 95% confidence intervals for each response category for individual questions about enacted stigma, using weighted data. 14 Table 2 Proportion of Respondents Who Reported Ever Experiencing Each Category of Enacted Stigma Group (Unweighted N) Type of Enacted Stigma Gay Men (n = 241) Lesbians (n = 152) Bisexual Men (n = 110) Bisexual Women (n = 159) Total (n = 662) Violence 24.9% 7.1% 6.9% 6.7% 13.1% 17.3 - 34.5 3.7 - 13.1 3.1 - 14.5 3.3 - 13.0 9.7 - 17.6 Property Crime 28.1% 10.2% 7.7% 6.3% 14.9% 20.9 - 36.6 6.0 - 16.8 3.8 - 15.1 2.0 - 17.8 11.4 - 19.3 Violence or Property Crime 37.6% 12.5% 10.7% 12.7% 20.9% 29.1 - 46.9 7.9 - 19.3 5.6 - 19.5 6.6 - 23.1 16.5 - 26.1 Attempted Crime 21.5% 8.3% 16.3% 5.6% 14.4% 14.7 - 30.3 3.9 - 16.9 6.3 - 36.0 2.5 - 12.0 10.1 - 20.2 39.0% 15.4% 20.1% 14.6% 24.8% Violence, Property Crime, or Attempted Crime 30.4 - 48.3 9.5 - 24.0 9.0 - 38.9 8.0 - 25.1 19.4 - 31.0 Objects Thrown 21.1% 14.6% 5.6% 6.8% 12.5% 14.4 - 29.8 8.9 - 23.0 2.4 - 12.5 3.6 - 12.5 9.4 – 16.6 Threatened With Violence 35.4% 17.3% 19.0% 14.1% 23.4% 27.2 - 44.6 11.0 - 26.2 8.2 - 38.1 7.8 - 24.2 18.2 - 29.5 Verbal Abuse 63.0% 54.5% 41.4% 34.3% 49.2% 53.8 - 71.4 44.4 - 64.3 22.9 - 62.7 24.8 - 45.1 42.0 - 56.4 17.7% 16.3% 3.7% 6.8% 11.2% Job or Housing Discrimination 12.1 - 25.0 10.3 - 24.7 1.4 - 9.6 3.9 - 11.7 8.5 - 14.6 Table reports parameter estimates and 95% confidence intervals for proportion of respondents experiencing each form of enacted stigma at least once, using weighted data. 15 Table 3 Felt Stigma Scale Scores and Responses to Individual Items Group (Unweighted N) Item Gay Men (n = 241) Lesbians (n = 152) Bisexual Men (n = 110) Bisexual Women (n = 159) Total (n = 662) Scale Score (Mean) 2.79 3.11 2.83 2.96 2.89 2.65 - 2.94 2.92 – 3.29 2.31 - 3.34 2.64 - 3.27 2.72 - 3.06 Agree 33.2% 44.0% 28.6% 37.7% 34.6% 25.3 - 42.1 34.3 - 54.1 16.5 - 44.9 27.7 - 48.9 28.9 - 40.9 In Middle 23.0% 24.6% 24.8% 19.1% 22.8% 15.6 - 32.6 16.4 - 35.2 10.2 - 48.8 12.1 - 28.8 16.9 - 30.0 Disagree 43.8% 31.4% 46.6% 43.2% 42.6% “Most people where I live think less of a person who is [L/G/B]” 35.1 - 52.9 23.0 - 41.1 26.6 - 67.7 31.8 - 55.3 35.3 - 50.2 Agree 54.8% 32.3% 49.1% 35.4% 45.4% 45.5 - 63.7 24.3 - 41.5 29.0 - 69.6 24.6 - 48.0 38.1 - 52.9 In Middle 27.5% 40.6% 19.6% 35.2% 29.1% 19.3 - 37.6 30.8 - 51.2 10.4 - 33.8 25.7 - 46.1 23.7 - 35.2 Disagree 17.7% 27.1% 31.3% 29.4% 25.5% 12.7 - 24.1 19.0 - 37.1 16.9 - 50.6 20.1 - 40.9 20.2 - 31.6 “Most employers where I live will hire openly [L/G/B] people if they are qualified for the job” Agree 41.2% 44.1% 39.1% 39.1% 40.6% 32.5 - 50.6 34.5 - 54.2 22.5 - 58.6 28.9 - 50.4 34.0 - 47.5 In Middle 32.1% 30.3% 20.8% 30.6% 28.4% 24.0 - 41.5 21.3 - 40.9 10.9 - 35.9 21.5 - 41.5 23.1 - 34.4 Disagree 26.7% 25.6% 40.1% 30.3% 31.0% “Most people where I live would not want someone who is openly [L/G/B] to take care of their children” 20.0 - 34.6 17.9 - 35.2 20.1 - 64.2 19.6 - 43.6 23.5 - 39.7 Table reports weighted percentage of respondents (with 95% confidence intervals) in each response category, and weighted mean scores (with 95% confidence intervals) for combined felt stigma scale. “Strongly agree” and “somewhat agree” responses are combined into “Agree” category; “strongly disagree and disagree somewhat” responses are combined into “Disagree” category. In each item, the respondent’s preferred label for her or his own sexual orientation (e.g., “gay”) was substituted for “[L/G/B].”