JSI Violence 2-col Preprint.doc Pr e- Pu bli ca tio n D ra ft 1 Victim Experiences in Hate Crimes Based on Sexual Orientation* Gregory M. Herek University of California, Davis Jeanine C. Cogan Washington, DC J. Roy Gillis Ontario Institute for Studies in Education, University of Toronto This is a preprint of a paper that will appear in Journal of Social Issues (2002), 58(2), 319-339. Please do not quote or cite without the author’s explicit permission. * Data collection and preparation of this paper were supported by grants to Gregory Herek from the National Institute of Mental Health (R01-MH50185 and K02-MH01455). We thank Robyn Caruso and David Webb for their special assistance with coding and analyzing the interview narratives. We also thank Jennifer Lyle, Camille Barber, Rebecca Blanton, Aaron Carruthers, Mary Ellen Chaney, Yancey Cuyugan, Gricelda Espinoza, Fred Fead, Gregory Gadwood, Adela Gayton, Eric Glunt, Gisi Gonzales, Rebecca Hill, Kim Lasdon, Gerardo Medina, Kyle Pham, Clarmundo Sullivan, Paula Urtecho, and David Welton for their invaluable contributions to the project. Correspondence should be sent to Gregory Herek, Psychology Department, University of California, 1 Shields Avenue, Davis, CA 95616- 8686, USA. Abstract Using interview data from a convenience sample of 450 lesbian, gay, and bisexual adults, the varieties of victim experiences in hate crimes based on sexual orientation are described. Most crimes were perpetrated in public settings by one or more strangers, but victimization also occurred in other locales and perpetrators included neighbors, coworkers, and relatives. In deciding whether a crime was based on their sexual orientation, victims tended to rely primarily on contextual cues and perpetrators’ explicit statements. Victims’ concerns about police bias and public disclosure of their sexual orientation were important factors in deciding whether to report antigay crimes, as were beliefs about the crime’s severity and the likelihood that perpetrators would be punished. Crimes that target individuals because of their sexual orientation – commonly referred to as hate crimes or bias crimes – are now understood to be a serious social problem (Jenness & Broad, 1997). This recognition has been facilitated, in part, by empirical research showing that hate crime victimization is widespread among lesbians, gay men, and bisexuals and that hate crimes are less likely than other crimes to be reported to police authorities (e.g., Berrill, 1992a; Federal Bureau of Investigation, 1999; Herek, 1989, 1993; Herek, Gillis, & Cogan, 1999; Herek, Gillis, Cogan, & Glunt, 1997; Pilkington & D’Augelli, 1995). In addition to documenting the prevalence of antigay hate crimes, social scientists have begun to attempt to understand the consequences of such crimes for the victim. This line of study represents a melding of research traditions, integrating scientific inquiry about the psychological consequences of severe trauma and crime (e.g., Bard & Sangrey, 1979; Janoff- Bulman, 1992), prejudice against sexual minorities (e.g., Herek, 1992, 2000), and the effects of prejudice on its target (e.g., Swim & Pr e- Pu bli ca tio n D ra ft 2 Stangor, 1998). Although still preliminary, theoretical and empirical work suggest that victims of antigay bias crimes are at risk for greater and perhaps longer lasting psychological distress than are gay and lesbian victims of comparable non-bias crimes (e.g., Garnets, Herek, & Levy, 1990; Herek et al., 1999). Because research on the psychosocial impact of antigay crimes is still in its early stages, systematic description of the experiences associated with such crimes is needed. However, such description is difficult to obtain through self-administered questionnaires, the method used in most published studies of antigay hate crime victimization. Questionnaires are an efficient means for collecting readily quantifiable responses from large numbers of gay men, lesbians, and bisexuals (Herek & Berrill, 1992), but yield only a limited understanding of the intricacies of hate crimes. Among their shortcomings are that they require respondents to have at least basic literacy skills, do not allow for follow-up probes, and are practical only when most questions are closed rather than open-ended. For these and other reasons, face-to-face interviews – in which respondents can give spoken, narrative responses and which allow for extended follow- up questions – can be a valuable complement to self-administered questionnaires. This paper describes insights into the nature of hate crimes that we gained from personal interviews with gay men, lesbians, and bisexuals. We focus on three questions: (1) What are the varieties of hate crime victimization experienced by sexual minorities? (2) How do victims decide that a crime is based on their sexual orientation? (3) Why do many victims choose not to report an incident to police authorities? METHOD Interviewees were 450 volunteers from a larger convenience sample. The larger sample (N = 2,259) consisted of sexual minority adults in the greater Sacramento (California) area who completed a lengthy self-administered questionnaire about victimization and mental health. They were recruited through a variety of strategies including outreach at community events, organizational contacts, local advertising, and personal networks. (For details about the larger sample and findings from the questionnaire study, see Herek et al., 1999.) Questionnaire respondents who indicated their willingness to be interviewed were contacted by a member of the research team (matched to the respondent’s sex and race) who explained the interview procedures and, for those deciding to participate, arranged the interview time and location. We originally planned to conduct 150 interviews with victims of antigay hate crimes, 150 with victims of nonbias crimes, and 150 with individuals who had never been criminally victimized. However, most interviewees who reported no crimes in their screening questionnaire subsequently recalled some type of victimization experience during their interview. This discrepancy was probably an effect of the extensive questioning and many memory cues provided during the interview. As a result, about 94% of interviewees described at least one crime experience. Most recalled multiple victimizations (median = 4; maximum = 8). Interviews were conducted by college- educated staff members who had completed extensive training in interview methods and procedures for protecting the welfare of interviewees. They were completed in the participant’s home, our research office in Davis, or another suitable location chosen by the participant. The interview lasted 2-4 hours, depending on the number of victimizations described by the respondent. Interviews were audiotaped with the consent of respondents and answers to open-ended questions were transcribed. Each interviewee was paid $15. The final interview sample (N = 450) consisted of 224 men (204 gay, 20 bisexual) and 226 women (202 lesbian, 24 bisexual), ranging in age from 19 to 73 (mean = 37 years) with a median annual income between $15,000 and $25,000. Approximately 81% were non-Hispanic White, with another 7% Latino, 3% Asian/Pacific Islander, 3% African American, 1% Native American, and 5% of mixed race or ethnicity. One half had earned at least a Pr e- Pu bli ca tio n D ra ft 3 bachelor’s degree. After they briefly described each crime, we asked respondents if they felt “that the incident occurred mainly because the other person(s) believed that you were [lesbian/gay/bisexual], that is, because they wanted to victimize a [lesbian/gay/bisexual] person.” The majority (57% of the sample) described one or more nonbias victimizations, but none that were bias motivated. Another 3% had experienced at least one bias victimization but no nonbias incidents. One third (34%) had experienced both types of victimization. Those who experienced a bias crime were asked what it was “that made you realize or believe that the incident was related to your sexual orientation.” We then asked follow- up questions about the crime and whether it was reported to the police. After respondents had thus described each of their victimization experiences (to a maximum of 8), we asked a series of more detailed questions about the one antigay crime in which the interviewee felt he or she was in the greatest danger. For interviewees who did not report any crimes based on their sexual orientation, the follow-up questions referred to the nonbias crime in which they felt they were in the greatest danger. A total of 154 interviewees provided detailed information about a completed or attempted bias crime (52 a property crime, such as robbery or vandalism; 102 an assault, rape, or robbery, hereafter referred to as person crimes) and another 244 described a completed or attempted nonbias crime (111 a property crime; 133 a person crime). Another 25 described no victimization experiences. The remaining 27 interviewees either declined to designate one crime as the most serious or named an incident that we later decided could not unambiguously be classified as a crime, such as witnessing an assault or receiving a harassing telephone call. RESULTS Because of the small number of nonvictims in the sample, we report comparisons between victims of bias and nonbias crimes, and between victims of person and property crimes. Before making such comparisons, we assessed the severity of bias and nonbias crimes. For person crimes, we asked victims how much they feared that they would be injured during the attack, how much they feared they would be killed, and the extent of their actual injuries. Survivors of bias and nonbias person crimes did not differ significantly on any of these indicators. In both groups, about 88% said that they had feared being injured to at least some extent during the incident they described. One half (50%) of nonbias crime survivors and 66% of bias crime survivors said that they “felt a lot of danger” of injury. Approximately one half (51%) of bias crime victims and 42% of nonbias crime victims feared that their life was in danger during the incident, with 27% of nonbias crime victims and 36% of bias crime victims reporting that they felt “a lot of danger” of being killed. In both groups, slightly more than 40% of survivors reported some physical injury as a result of the attack, with 8% of nonbias crime victims and 14% of bias crime victims characterizing their injuries as “fairly serious” or “very serious.” Whereas the severity of bias and nonbias person crimes was comparable, bias and nonbias property crimes differed somewhat in form. Most bias property crimes were acts of vandalism: 81% included vandalism, compared to 39% of nonbias property crimes. By contrast, 64% of the nonbias property crimes involved theft, compared to 27% of bias crimes. This difference should be recalled when property crimes are compared below. Describing Crimes Against Lesbians, Gay Men, and Bisexuals Settings and Perpetrators We categorized each interviewee’s most serious crime according to whether it occurred in a public setting (e.g., a street, a commercial establishment), an interior private setting (a home or office), a school campus, another institutional setting (jail, military base), a rural area, or immediately outside the victim’s home (e.g., a driveway or front yard). Compared to nonbias crimes, bias person crimes occurred disproportionately in public places (60%, versus 32% of nonbias person crimes and 32% of Pr e- Pu bli ca tio n D ra ft 4 nonbias property crimes). By contrast, nonbias person crimes took place significantly more often in an interior private setting than did bias person crimes (53% versus 27%, respectively). Property crimes were more likely than person crimes to occur immediately outside one’s home (respectively, 15% and 12% of bias and nonbias property crimes, compared to 1% and 2% of bias and nonbias person crimes). For the comparison of the four crime types by location, Χ2 (15, n = 398) = 55.67, p < .001. Of the 320 respondents who could specify the number of perpetrators, most (68%) were victimized by a lone assailant. However, bias crimes were more likely than nonbias crimes to have multiple perpetrators. For person crimes, 46% of bias crimes had two or more perpetrators, compared to only 17% of nonbias crimes. And whereas 68% of bias property crimes had two or more perpetrators, this was true of only 33% of nonbias property crimes. For the comparison of crime type by number of perpetrators, Χ2 (3, n = 305) = 33.26, p < .001. In nearly two thirds of crimes (62%), the victim did not know the perpetrator(s). This pattern differed according to crime category, however. In 70% of bias person crimes, 66% of bias property crimes, and 79% of nonbias property crimes, the perpetrators were strangers. By contrast, 58% of nonbias person crimes were committed by someone known to the victim. The difference between nonbias person crimes and the three other types of crime was statistically significant, Χ2 (3, n = 364) = 34.52, p < .001. Most perpetrators were adolescents or young adults. Of the 304 victims who were able to estimate the perpetrator’s age, 61% placed him or her in the range of 13 to 25 years. The majority of perpetrators were White. Of the 302 interviewees who could identify at least one perpetrator’s race, 69% reported one or more White perpetrators, 17% reported at least one Black perpetrator, and 19% reported at least one Hispanic perpetrator. Few respondents reported an Asian-American (n = 4) or Native American/Indian (n = 2) perpetrator. Perpetrator race did not appear to vary according to the victim’s race or ethnicity. However, because most victims in the interview sample were non- Hispanic Whites, we could not adequately assess this relationship. Regardless of the crime type or the victim’s sex, most perpetrators were male. Of the 302 victims who knew the perpetrator’s sex, 99% of men and 90% of women named at least one male perpetrator. The relatively few incidents that included a female perpetrator tended to be female-on-female crimes. Across crime categories, the sex of perpetrators differed significantly between men and women interviewees only for bias person crimes. In that group, 91% of women respondents were victimized by at least one man and 17% were victimized by at least one woman. By contrast, all of the men respondents were victimized by males and only 2% by females. Among the victims of bias person crimes, the difference between men and women respondents was significant for male perpetrators, Χ2 (1, n = 100) = 5.75, p < .05, and for female perpetrators, Χ2 (1, n = 100) = 7.39, p < .01. The Varieties of Hate Crime Victimization Victimization in Gay-Identified Settings Consistent with previous research (e.g., Berrill, 1992a) and the analyses reported above, many interviewees described antigay crimes that were perpetrated by groups of strangers in proximity to a gay-identified venue. The severity of such attacks ranged from harassment and threats to physical assault and, in the case of one interviewee, the murder of a friend. The story told by one lesbian was typical of many incidents described by interviewees. I was parked in the lot of a gay bar and if you were parked there it was obvious you were in the gay bar…. And I came out to my car and my front windshield was smashed in by what appeared to be a baseball bat. The side mirrors were beat, it was totally shattered …. Some victims suffered assault but were not seriously injured. One respondent, for example, Pr e- Pu bli ca tio n D ra ft 5 described being followed by a truck as he walked home one night from a gay bar. These 4 guys started yelling these great slurs at me, called me “faggot….” Then I started getting scared. The next thing you know, that truck stops, this guy jumps out. He has an egg in his hand. First I didn’t know it was an egg in his hand. I didn’t see the egg, I saw his hand. Now everybody else didn’t get out of the truck, but you know that was a possibility that could have happened too, wasn’t there? He then called me faggot and he threw the egg and hit me right here, it hit me right here in the chest…. You tell me if I wasn’t scared. None of the other guys got out of the truck, and then he simply calmly got back, jumped back into the truck and sped off. This man’s story highlights the fear associated with such an experience. Most likely, he was aware that such incidents often end in serious injury or even death for the victim. During the crime, he could not know that only one of the truck’s occupants would attack him or that he would be struck with an egg rather than a more lethal weapon. Thus, despite the absence of serious physical harm, the attack evoked considerable fear. Because fears of serious injury and death during a crime are associated with the degree of psychological trauma it inflicts (Kilpatrick et al., 1989), even seemingly minor incidents can have a negative aftermath for the victim. Many attacks caused serious injury and some had deadly consequences. One man described how he was assaulted and his friend was murdered while visiting a Southern city. [We] walked outside the bar and there were about six guys standing on the corner down from the bar. Our car was parked right on the corner of where they were. As we were walking towards them, they saw us coming and started walking towards us. They started calling us “fags,” saying “fags,” “look at the fags” and “nigger fag.” …They said “We’re going to kill us some faggots today” and “we don’t want fags in [this city]” ….They attacked us with bricks and clubs. [My friend] was hit in the head with a brick, and when he went down, they hit him more in the head with bricks and clubs till he stopped moving. I was hit in the legs with a club, and broke my knee cap. The other two friends got away and went back into the bar to call the police, and came out with more people from the bar, and chased the attackers away until the police got there. The police took about 20 minutes to get there and the ambulance almost a half an hour. By that time [my friend] had already expired. He died in my arms. Victimization in Other Public Spaces Going to a gay-identified establishment carries the risk of verbal harassment, vandalism, and physical assault. However, the interviews also made it clear that any public space is potentially dangerous. One woman, for example, described her experience in a public park with a group of women friends when three men walked past and harassed them. When they reached me … they pinched me on the butt. And my girlfriend saw this and said “Get your hands off her.” So they stopped, turned around and said “Want to fight, bitch?” We kind of circled up and, you know, we were all facing them and said “No, leave us alone, get out of here.” We tried to get them to leave and they wouldn’t. When she said she didn’t want to fight, he just stuck his fist out and broke her nose…As he was getting ready to throw the punch … he said “fucking dykes….” One person [got] a cut open on their cheek, face. And another one had her collarbone broken and got knocked unconscious. I got kicked in the knee and upper thigh and was severely bruised. And Pr e- Pu bli ca tio n D ra ft 6 then somebody came by and helped scare them away for us. A man described his experience while jogging near a river, wearing a singlet with a lavender triangle emblem. There were five or six youths between 18 and 22. They were drinking and I had stopped to cool off at the river. They grabbed stones and were cornering me against the river and the dam, with threats and comments…. Their comments were “We hate you gays….” They threw stones, large riverbed stones at me…. To leap into the river would have been to go over the dam, and it’s a very deadly dam.… I knew that was no route to escape. And so the only possibility was to leap and escape running. I tried to distract them in some way, verbally, and move away from the river to sprint away from them. Another woman was harassed while riding her bicycle one afternoon. I heard some guys kinda’ yelling and screaming. I tried to ignore them. Then suddenly beer cans were being thrown at me. They pulled up next to me and they were throwing beer cans at me and making derogatory remarks…. “fucking faggot,” “fucking queer”…. “fucking dyke”…. They stopped in back of me and I continued riding away from them. And they started up again, coming up from behind, continuing with derogatory remarks and then saying “Let’s get her and let’s get that bike.” They made a motion as though some of them were going to get out of their car, and that’s when I took off on my bike and got the hell outta’ there. In these situations, the victims were not in a gay-identified setting but their sexual orientation was assumed by the perpetrators on the basis of contextual cues. The woman in the park was with a group of women, the jogger was wearing a singlet with a gay symbol on it, and the bicyclist was wearing a shirt with the word “lesbian” prominently displayed. Their experiences suggest that, although gay venues are often sites of attacks, being identified as gay or bisexual in any public setting carries a risk. Victimization At Home Nor is one’s home a safe refuge. Many interviewees had their house, car, or other personal property stolen, vandalized, or destroyed in antigay incidents. One man described the events on a day when he and his lover were leaving on a fishing trip. We got up very early that morning and got in the car and backed out of the garage. And it has an automatic closer and it went down and it said – I don’t remember the words but it was something about “fag” or “faggot.” Several words, but I don’t remember what they were, sprayed onto the redwood garage door in black paint. Another woman described how vandals filled her car with grass clippings, put nails in the tires, and “spray painted ‘queer’ all over my car.” In some cases, the incident was part of an ongoing pattern of harassment. One woman described a series of incidents in which teenage neighbors vandalized her mailbox, threw rocks at her daughter, shouted obscenities at them, and left burning dog feces on her front porch. Another woman reported that a new neighbor “took an immediate dislike” to her and her female partner. He “began an endless onslaught of verbal harassment, ‘Dykes, sick love, queers,’” and threatened them and their pets. They began to find their property vandalized in small ways and were convinced that the neighbor was responsible. Several respondents described having their property vandalized after they made a public gesture that identified them as gay. One woman, for example, flew rainbow and American flags outside her house but one day “Somebody burned my rainbow flag and apparently stuck a sticker on my car that was parked in the driveway…. It said ‘homo’ on it.…” In other Pr e- Pu bli ca tio n D ra ft 7 cases, respondents were threatened and harassed after they had been publicly identified with a local gay or lesbian event or organization. One man, for example, received threatening telephone calls after his name was published as a contact for the gay student organization on his campus. Some incidents were potentially deadly. One man described having his house fire bombed and the windows smashed on two of the family cars. I was asleep on the front porch and a Molotov cocktail was lobbed up onto the second story front porch where I was at…. And it immediately ignited the porch. I was asleep in that porch. As the building was burning I could hear the windows being broken out of the cars. And the people doing it laughing and screaming “faggot” at the top of their lungs…. There was a note attached to the windshield of my car: “The faggot that lives here will be dead within a week.” Victimization in Schools Middle and secondary schools routinely are sites of harassment for students who are gay or deviate from gender norms (e.g., Smith, 1998). One interviewee related how he had been the target of ongoing antigay harassment in his high school. Late one night, he went to a convenience store to buy milk for the family’s breakfast. He saw two classmates in the store’s parking lot. And as I got out of the car, they came up to the car, and approached me and said the usual “Hey, faggot” and started asking me what I was doing out…how would I like to be beat up. And they cornered me against the car and started beating on me. He escaped into the store. When he asked the clerk to phone the police, the clerk refused. I had a black eye, bloody nose, split lip, and some other bruises on my body. My parents called the police. They didn’t even want to take a report. Their comment was “They’re just a couple of good all-American boys out for some fun….” Both the guys who assaulted me often called me “faggot” and things like that in [my high school]. Such harassment occurs in college as well (Berrill, 1992a, 1992b; D’Augelli, 1989; Herek, 1993). One respondent and his lover experienced ongoing harassment on campus. Just on a daily basis we were taunted, called “fags” to the point where … we didn’t even want to go to the cafeteria, we’d just go out to eat. Then nightly we pretty much, we never knew what we were going to wake up to or be awakened by in the night, like someone urinating on the door or sticking stuff, something on the door. …When people write “fag” on your door and that was one of the main things…. Put a gay sticker on the door with one guy bent over and another behind him with a big line through it…. It was so awful. The narrative also illustrates the stresses that such harassment places on the victims. In this case, that stress had a negative impact on the victims’ relationship. According to the interviewee: [The harassment] made… the two of us ... fight all of the time because our stress level was so high. I actually quit school twice but ended up going back ’cause of all this…. It was to the point where you’d be in your dorm room and you felt like a prisoner because you didn’t even want to go out, because it just seemed everyone was your enemy and no one would help. Victimization In The Workplace Most descriptions of crime on the job came from women respondents working in historically male settings. For example, one woman described how she was harassed and her car was vandalized with anti-lesbian graffiti while she was training to become a police officer. Another woman related her experiences Pr e- Pu bli ca tio n D ra ft 8 in the Army after her sergeant read some love letters written to her by another woman. The sergeant… called me into his office, and told me to explain the letters. When I refused to explain the letters, he grabbed me by the collar and threw me up against the wall and told me he wasn’t going to have any dykes in his company. He took me outside and said get in the jeep. Took me to the top of [a mountain]. Got out, pushed me against the side of the jeep and said he was going to show me what a real man could do. Said he was sick of all these lesbians. He said what I needed was a real man to bring me out of this gay shit. I refused to cooperate, to make a long story short, and I threatened to kill him or us both. I don’t know, I guess that scared him, cause I was looking real mean about that time and a little scared. He said just forget it, he got in his jeep and drove off. And I walked back down the mountain, which was about 40 miles. Victimization By Friends and Family Members Family members, friends, and acquaintances also were perpetrators of harassment and violence. Several women described incidents in which a male acquaintance or friend sexually assaulted them, sometimes after a failed attempt to seduce them. Many reported that their attacker seemed to be trying to prove that the woman was actually bisexual or simply needed the right man in order to become heterosexual. One woman, for example, described what happened when, during dinner with a male former school friend, she refused his sexual invitation. I tried to talk to him and explain that I didn’t want any man at all. He figured I was bi. I kept trying to leave because I had things to do and errands to run, and he wouldn’t let me out. I became very frightened and I broke a window to get out. I got out the window, he chased me down and grabbed me, threw me up on a vehicle and, as I was sliding down, he grabbed me again and threw me down so that my head hit the speed bump. Then proceeded to beat me. In other cases, a former husband or male partner attacked a woman, seeking revenge for his perceived rejection by the victim. Parents and siblings also were perpetrators. One man recounted being accosted by his father and brother after they overheard him talking to his mother about his weekend: They got up, both very angry – you could see it in their face with the blood vessels sticking out of their necks and on their foreheads. And it was what they said and their body language, it was pretty scary: “Queer, faggot, you’re going to get what you deserve, you’re going to get that AIDS….” They came in here, there was some grabbing of my clothing….I was able to defuse it, and let them know I was going to stand my ground. I was going to defend myself. A woman described what happened after her mother read her diary, which included details about her sexual involvement with another woman. She went out and got drunk and then came home and started yelling and screaming at me about sleeping with women. Actually, it was girls at the time…. She was pissed off at me because I was gay or I chose to sleep with women…. She said stuff like “bumping pussies” and just stuff like that…. And how sick it was, unnatural. … And then she just started getting really crazy and hitting me and knocking me around. And actually I was pretty scared…. Summary Mass media depictions of antigay crimes often focus on street attacks that are perpetrated by groups of young men who have no prior acquaintance with the victim. Among our interviewees, bias crimes were indeed most Pr e- Pu bli ca tio n D ra ft 9 likely to occur in those circumstances. But many other scenarios were also described. In addition to strangers, the perpetrators included parents, siblings, former spouses, peers, supervisors, coworkers, and neighbors. Many incidents occurred in gay-identified venues and other public settings, but others took place in and around the victim’s home or in campus or workplace settings. Clearly, antigay crimes occur in many settings and are perpetrated by individuals with a variety of relationships to the victim. Labeling the Crime Researchers who study antigay crimes inevitably must rely on victims’ self reports to determine that a crime was antigay. But how do people determine that a crime is based on their sexual orientation? Does such labeling result from unambiguous events during the crime, or does it reflect the victim’s interpretations, inferences, or projections? This question has implications not only for social psychological research but also for policy makers and criminal justice personnel, who often determine whether an incident will be classified as a hate crime. ________________________________ Insert Table 1 about here ________________________________ To shed light on the attributional processes underlying self-reports of victimization, we asked each respondent who described a bias- motivated crime “What was it that made you realize or believe that the incident was related to your sexual orientation?” Interviewees with multiple bias victimizations were asked this question about each crime. The responses described 272 crimes, and displayed many of the same general patterns as noted in an earlier pilot study (Herek et al., 1997). We identified 6 categories: (1) explicit statements by the perpetrator or other unambiguous information from a third party such as the police; (2) the location of the crime in a gay-identified venue; (3) other contextual cues; (4) the victim’s inference or hunch; (5) revenge from a male former partner (women only); and (6) victimization in conjunction with a sexual pickup (men only). In addition, some respondents characterized a crime as based on their sexual orientation because it was perpetrated by another gay person (and thus did not fit our definition of a hate crime). Transcripts were coded by two independent raters and responses were assigned to one of these categories; inter-rater agreement was 92% (disagreements were discussed and jointly coded). Table 1 reports the frequency with which each explanation was cited. Explicit Statements or Other Unambiguous Information In slightly more than half of the crimes, respondents had an unambiguous indication that the incident was antigay, usually from the perpetrator’s explicit statements. One woman, for example, described an attack while she and a companion were driving in an urban area. I was with another woman who was very dykey looking and very androgynous. We were in a car and these young teenage boys started screaming “Dykes! Look at the dykes,” blah blah. Then one got out of his car and physically attacked our car, shattering its windows, kicking them in….There was glass everywhere. It looked as if we had crashed. A man who was the victim of a physical assault reported that he was “leaving a gay establishment and the men were yelling obscenities and derogatory anti-gay names, name-calling.... ‘You fucking faggot.’” In several cases, an act of vandalism or burglary included antigay epithets or threats. One male burglary victim reported that he “came home from work, my front door was kicked in and everything was gone. Just a note tacked on the door facing: ‘This is what happens to faggots.’” In a few cases, the victim received information from the police that the crime was one in a series of anti-gay offenses by the perpetrator. After being assaulted and robbed, for example, one man was informed by the prosecuting official that his attackers had similarly victimized other gay people. A robbery victim reported, “When I went to the police and Pr e- Pu bli ca tio n D ra ft 10 identified him in their notebooks, they told me that he was known as someone who preyed on gay people.” Crime Location In 9% of the crimes, victims decided that the incident was based on their sexual orientation because it occurred outside a gay-identified bar, church, or similar venue. A male assault victim, for example, reported “I was leaving a gay bar. They attacked me. They were going to beat the tar out of me and I got away from them.” Theft and vandalism were also common in such settings, especially involving respondent’s vehicles. One woman, for example, reported: I parked my truck in a parking lot right next to a lesbian bar in downtown [Sacramento]. Went in, came out a couple hours – no, two hours – later. My truck was gone. There was broken glass in the parking lot where my truck had been. She added that vandals broke the windows of several cars parked outside the same bar on the following evening. Other Contextual Cues Other respondents based their attribution on the fact that the attack occurred after they (or others with them) had somehow made their gay or lesbian identity visible. For example, the victim was holding hands with a same-sex partner, had affixed a gay bumper sticker to her or his car, or displayed a rainbow flag at her or his home. Contextual cues of this sort were used by the victim in categorizing 10% of the crimes. One woman reported: I was driving in my car, and me and my girlfriend kissed at a stop light. Then this car of guys pulls up next to us and is screaming and yelling and flailing about and throwing eggs at us…. They followed us for awhile. [Why attributed to sexual orientation?] Because it happened right after I kissed my girlfriend. There was a rainbow flag sticker on the car. A gay man reported: Someone during the night egged my car…. I had a gay flag on my car.… And my car was the only one with any type of gay decal or anything… and it was the only one that was egged. We were also known in the neighborhood as being a couple, a gay couple… and the car was definitely identified with me. A lesbian reported that her car was one of several that were all vandalized in a single incident. “My car was vandalized, a bunch of cars were vandalized because of the pink triangle…. We had our windows smashed in, where the pink triangle was.” The common theme was that all of them displayed a pink triangle sticker, which is widely recognized as a symbol of the lesbian and gay community. Hunch In 16% of the crimes, respondents based their judgment simply on a hunch or inference. Such hunches often reflected the timing of the incident or a belief that the perpetrators recognized the respondent’s sexual orientation. One woman, for example, related how several young men attempted to assault her in a parking lot. When asked why she attributed the incident to her sexual orientation, she replied, “Because of where I was, I was in a place in the city where there was a lot of gay bashing going on and I looked like a dyke.” Some victims noted that they were singled out for attack. One lesbian described a recurring pattern of vandalism: Mailboxes here are in clusters, and there are three mailboxes together. Two of them belong to lesbian households. Routinely someone would come along with a baseball bat and bash two of them. We’d replace them, they’d bash them, and on the fourth of July, they blew them up. And they only blew up the ones of the lesbians’ houses. Revenge From a Male Former Partner (Women Only) In 4% of the crimes against women, the victim reported that a former husband or male lover attacked her out of revenge. One woman, Pr e- Pu bli ca tio n D ra ft 11 for example, was assaulted by her ex-husband: We were getting a divorce over my lesbian relationship and he was really angry….[He] accused me and [a woman friend] of being lesbians. Beat the shit out of me. I was holding the baby and he shoved me into the bathtub. I was amazed at how I saw stars when I was hit in the face. Entrapment (Men Only) Two percent of the crimes against men were perpetrated by another man in a sexual context, such as a sexual cruising area. The perpetrator(s) made a sexual advance and, in some cases, had sex with the victim before robbing or assaulting him. I had gone to a rest stop… and a male asked me if I wanted to have sex. I said yes…. There were two males, I followed both of them. We went up in the woods and got out of the car. We – I started having sex with one male. Without my knowledge, the other male came up behind me with a bowie knife type of weapon, and cut both of my hands, stole my car keys, and stole my car. I got away from them and the [police] found my car the next day, burnt – completely…. Misclassified Crimes Finally, it was clear from some interviewees’ responses that the incident they described was not a bias crime. Most such incidents were perpetrated by another gay or lesbian individual, and many involved unwanted sexual advances or contacts. Based on our pilot study results (Herek et al., 1997), we tried to minimize this type of response by framing the question about the basis for the crime to refer specifically to incidents in which the perpetrator “wanted to victimize a [lesbian/gay/bisexual] person.” Some respondents nevertheless interpreted the item differently than we intended. This finding points to the value of collecting data in face-to- face interviews. Without the respondents’ narratives about the crime and their reasons for categorizing it as based on their sexual orientation, we would not have identified this small subset of responses. Conclusions The patterns observed here are similar to those noted in our pilot study (Herek et al., 1997) and suggest that most victims who categorize their crime as antigay have good reasons for doing so. Typically, the perpetrators made explicit statements, the attack occurred in a gay-identified location, or the crime was closely associated with behaviors by the victim that identified her or him as gay. Only 16% of interviewees based their attribution merely on a hunch, and in many cases their reasons for believing that they were targeted because of their sexual orientation seemed highly plausible. Reporting the Crime Service providers, researchers, and activists have all noted that hate crimes are less likely than other crimes to be reported to law enforcement authorities (e.g., Berrill, 1992a; Herek, 1989). Estimates of nonreporting among gay and lesbian hate crime victims have ranged as high as 90% (Berrill & Herek, 1992; Herek, 1989). The explanation most commonly offered for the high rate of nonreporting is victims’ fear of what Berrill and Herek (1992) labeled secondary victimization: discrimination and mistreatment by police authorities, or negative consequences as a result of having one’s sexual orientation publicly revealed. Systematic data documenting the actual extent of nonreporting and the reasons for it have been lacking. Several questionnaire studies have estimated the prevalence of nonreporting (Berrill & Herek, 1992), but many of them combined data about crimes with data about minor incidents (e.g., name-calling). Although often distressing, the latter are usually not criminal offenses and in many cases are probably perceived by the target as too trivial to report. Studies that count such incidents probably overestimate nonreporting rates. In addition, previous studies have not obtained data about reporting rates for crimes not based on sexual orientation. Such comparison data are important for determining whether nonreporting is a problem specifically for hate crimes, or if all gay and bisexual crime victims are reluctant to Pr e- Pu bli ca tio n D ra ft 12 interact with police. To the extent that members of sexual minorities fear secondary victimization, they may avoid reporting any crime, regardless of whether or not it is hate- motivated. ________________________________ Insert Table 2 about here ________________________________ In our questionnaire study, therefore, we assessed nonreporting for both bias and nonbias crimes and found that it was disproportionately associated with bias crime victimization. As shown in Table 2, hate crimes were less likely than nonbias crimes to be reported to police authorities by lesbians (who reported 36% of person and property hate crimes to police versus 68% of other victimizations), gay men (who reported 46% versus 72%), bisexual women (35% versus 62%), and bisexual men (24% versus 61%). In contrast to the questionnaire sample, the interviewees’ reporting rates did not differ between bias and nonbias person crimes. They reported fewer than one third of person crimes (32% of nonbias person crimes, 29% of bias person crimes) to the police. Nonbias property crimes were significantly more likely to be reported (68%) than bias property crimes (54%) and person crimes, Χ2 (3, n = 397) = 45.01, p < .001. Although the interviewees did not display the same patterns as the questionnaire sample, their comments provide insights into the reasons for nonreporting. For those who did not report their most serious victimization, we asked about the extent to which concerns about police prejudice influenced their decision. Such concerns played an important role for bias crime victims but not for nonbias crime victims. Among victims of bias person crimes who did not report, 68% said that concerns about the police played at least some role in their decision, compared to 18% of victims of nonbias person crime. Among victims of bias property crimes, 44% said that such concerns affected their decision, compared to 6% of nonbias property crime victims. The difference between bias and nonbias crimes was statistically significant, Χ2 (3, n = 216) = 57.25, p < .001. The interview narratives demonstrate respondents’ concerns. One woman said she “didn’t want to deal with police in West Sacramento; they were known to not be friendly to gays.” A man recalled a case of a local lesbian police officer who was fired because of her sexual orientation, and said that he consequently did not trust the police to act properly. Another man commented “Being a gay male, I figured the cops wouldn’t really give a shit about it anyway.” Another man said that it hadn’t occurred to him to report his victimization to the police because he “felt as though that kind of went along with the territory of being gay.” A woman said that she didn’t report her victimization because of “the fact we were lesbians. They wouldn’t take it seriously at that time, and the fact that we were in a straight bar and therefore were ‘asking for it.’” Another man stated: “I was afraid that the police was going to gay bash me, and that everybody would know that I’m gay because I was in the closet.” One consequence of reporting a bias crime is that the victim’s sexual orientation is likely to become known to the police. We asked interviewees who had reported a crime whether they believed that the police had recognized their sexual orientation. Significantly more bias crime victims believed that the police had perceived them to be gay, lesbian, or bisexual: 79% of bias person crime victims versus 29% of nonbias person crime victims; 56% of bias property crime victims versus 21% of nonbias property crime victims (totals include 9 participants who volunteered that they were unsure whether the police had perceived them to be gay, lesbian, or bisexual). For the comparison across crime categories, Χ2 (6, n = 171) = 37.81, p < .001. Of those who believed the police had perceived (or might have perceived) that they were not heterosexual, victims of bias person crimes were the most likely to believe that this perception possibly affected their treatment by the police (52% believed it affected the way they were treated, compared to 42% of nonbias person crime victims, 23% of bias property Pr e- Pu bli ca tio n D ra ft 13 crime victims, and 14% of nonbias property crime victims). The differences among crime groups were marginally significant, Χ2 (3, n = 60) = 6.48, p = .09. Another reason for not reporting was that victims did not wish to be outed. Most interviewees who cited concerns about public exposure of their sexual orientation focused on the consequences for their employment. One woman, for example, noted that “I was in my [work] uniform and I wasn’t out to the people at [work].” Another man responded “In those days you didn’t want anybody to know you were gay.… I had a secret clearance for where I was working and I would have lost that and my job and everything else.” Other interviewees were on active military duty and noted the likelihood of secondary victimization if they had reported the incident: I was in the military, in the Army. And I would’ve gotten more trouble than him [the perpetrator] for being a lesbian. I would have gotten 7 years imprisonment for homosexual tendencies and he knew that. We were in a foreign country and we were in the military. When in the military you don’t go to the police and say “Oh, they harassed us because we are gay.” Concerns about police harassment and public disclosure of one’s homosexuality clearly are important reasons why gay men, lesbians, and bisexuals do not report victimizations to the police. Yet, the interview narratives also highlighted many other considerations that play a role in deciding whether or not to report. We transcribed responses to the open-ended question about why a crime wasn’t reported and coded them for the presence or absence of 23 different reasons. Two independent raters coded the transcripts (rate of agreement = 83%), with disagreements resolved by a third coder. Other than fear of secondary victimization, the reasons most often cited for not reporting were: (1) The incident did not seem sufficiently important, or reporting seemed futile because the police were unlikely to catch the perpetrators. (2) The victim considered the incident a personal matter (e.g., between friends or family members) rather than a reportable crime. (3) The victim judged the incident to be her or his own fault or was embarrassed about being victimized (for reasons not having to do with her or his sexual orientation). (4) When the incident happened, the victim did not believe it was a reportable offense. (5) The victim took some action on her or his own to avoid future incidents or to alleviate the problems caused by the incident, and consequently did not believe it was necessary to involve the police. In general, victims appear to have engaged in a cost-benefit analysis. They tended not to report the crime if reporting seemed unlikely to produce a satisfactory outcome but could be unpleasant, time consuming, or even risky. Fear of secondary victimization was an important factor, but was not the only consideration. GENERAL DISCUSSION The data reported here corroborate and extend the findings of past research. They also paint a rich portrait of hate crime victimization. As in earlier studies, the hate crimes described by interviewees most commonly occurred in public locations and were perpetrated by one or more males who were strangers to the victim. Yet, as the narratives make clear, it would be inaccurate to conceptualize bias victimization only in terms of street crimes. Members of sexual minorities face harassment and violence in schools, in the workplace, and in and around their homes. Whereas they are often targeted by strangers, they are also victimized by neighbors, schoolmates, coworkers, and relatives. Indeed, the respondents’ stories dramatically show that people risk victimization whenever they are labeled gay, lesbian, or bisexual. Although hate crime victims in the interview sample reported slightly higher levels of fear during their attack, bias-motivated and nonbias person crimes did not differ significantly in their general severity. Nevertheless, we were struck by the physical and psychological brutality of the hate crimes described in the interviews. This brutality has important consequences. For victims, it results in Pr e- Pu bli ca tio n D ra ft 14 heightened and prolonged psychological distress after the crime. In our questionnaire study we found that victims of recent (i.e., occurring in the previous 5 years) bias person crimes manifested significantly higher levels of psychological distress than did victims of nonbias person crimes during the same period. We speculated that this difference might be due in part to slower recovery among bias crime victims (Herek et al., 1999). The brutality of hate crimes also has consequences for the entire gay community. It is not an exaggeration to conclude that bias- motivated attacks function as a form of terrorism, sending a message to all lesbians, gay men, and bisexuals that they are not safe if they are visible. Thus, even when one does not personally know the victim, hate crimes can threaten the illusion of invulnerability that is so important in one’s daily life (Janoff-Bulman, 1992). In addition, the ubiquity of hate crimes might make even minor instances of harassment more frightening for the victim. As noted above, a gay or bisexual person who encounters an expression of hostility because of her or his sexual orientation does not know in advance how the incident will end. She or he may be attacked with words, a raw egg (like one of the interviewees), or a deadly weapon. Consequently, an incident that appears minor in retrospect might nevertheless have considerable psychological impact on the victim. Lesbian, gay, and bisexual people appear not to have difficulty recognizing when a crime is based on their sexual orientation. In most incidents described by interviewees, the perpetrators made explicit antigay statements in the course of the crime. The remainder often occurred in a context or setting in which the victim’s sexual orientation was made known to the perpetrator through the victim’s actions, the presence of gay-identified symbols on the victim’s clothing or property, or the setting of the crime itself (e.g., outside a gay bar). Thus, the incidents described by most respondents appear to have been bias crimes. Nevertheless, follow-up questioning revealed that a small proportion of the incidents that respondents characterized as based on their sexual orientation were not bias crimes. This finding highlights one of the difficulties inherent in hate crime research. Directly asking respondents if they were the victim of a hate crime or bias crime is problematic because those terms may have different meanings for different respondents. In addition, some victims may avoid explicitly labeling their experience a hate crime out of a need to preserve a sense of personal safety or a feeling of control over events in their life. Research with women and ethnic minorities suggests that labeling an incident as discrimination may reduce an individual’s sense of personal control (Ruggiero & Taylor, 1997). Consistent with this finding, we have observed that gay men and lesbians who generally attribute negative events in their lives to sexual prejudice have a lower sense of personal mastery and more psychological distress than those who do not make such attributions (Herek et al., 1999). Thus, labeling an incident a hate crime may have a disempowering effect on the victim. Researchers face the dilemma that using terms such as hate crime in interview or questionnaire items may lead to underreporting, but framing the question more broadly and in neutral terms is likely to result in some participants reporting incidents that are not hate crimes. We do not have a ready solution for this problem, but recommend that other researchers be aware of it when interpreting their findings. We note that it can be detected more readily in a face-to-face interview than in a self- administered questionnaire. In our questionnaire study, we found that bias crime victims were considerably less likely to report the incident to police than were victims of nonbias crimes. As Table 2 shows, the exact magnitude of this discrepancy varied by victim sex and sexual orientation but was substantial in all groups. The interview data clearly show that concern about secondary victimization is an important reason for nonreporting but not the sole basis for it. The reasons cited by interviewees suggest a complex calculus in which victims considered the other costs and benefits of reporting (e.g., whether or not the perpetrators could be apprehended and Pr e- Pu bli ca tio n D ra ft 15 punished) and whether the crime could appropriately be considered a police matter. Because the interview and questionnaire data included crimes that occurred in the distant past, we do not know if they accurately describe current patterns of crime reporting. During the past decade, many police departments in the United States have taken measures to respond to the problem of hate crimes, often with assistance from state and federal governmental agencies. Police officials increasingly are working with minority communities to improve their response to bias crimes. Undoubtedly, police personnel in many jurisdictions still need clearer policies and better training for dealing effectively with hate crimes based on sexual orientation. But to the extent that nonreporting persists as a problem, effective remedies will have to come from the gay community as well as the criminal justice system. Outreach to gay men, lesbians, and bisexuals is necessary to overcome their longstanding suspicions of the police. Such efforts will have to originate not only in criminal justice agencies but also in community organizations. Because the present study was conducted with a convenience sample, generalizations from our findings must be made with caution. We hope that other researchers will attempt to replicate our results with samples from other geographic areas that include lesbians, gay men, and bisexuals from diverse backgrounds. We also hope that hate crime researchers will conduct more studies using face-to-face interviews as an alternative to or in conjunction with self-administered questionnaires. As we have tried to demonstrate here, bringing victims’ voices directly into our research yields a more differentiated and nuanced understanding of the nature of hate crimes. References Bard, M., & Sangrey, D. (1979). The crime victim's book. New York: Basic Books. Berrill, K.T. (1992a). Antigay violence and victimization in the United States: An overview. In G. Herek & K. Berrill (Eds.), Hate crimes (pp. 19-45). Thousand Oaks, CA: Sage. Berrill, K.T. (1992b). Organizing against hate on campus: Strategies for activists. In G. Herek & K. Berrill (Eds.), Hate crimes (pp. 259-269). Thousand Oaks, CA: Sage. Berrill, K., & Herek, G. M. (1992). Primary and secondary victimization in anti-gay hate crimes. In G. Herek & K. Berrill (Eds.), Hate crimes (pp. 289-305). Thousand Oaks, CA: Sage. D’Augelli, A. (1989). Lesbians’ and gay men’s experiences of discrimination and harassment in a university community. American Journal of Community Psychology, 17, 317-321. Federal Bureau of Investigation. (1999). Hate crime statistics 1997. Washington, DC: U.S. Department of Justice. Garnets, L., Herek, G. M., & Levy, B. (1990). Violence and victimization of lesbians and gay men: Mental health consequences. Journal of Interpersonal Violence, 5, 366-383. Herek, G. M. (1989). Hate crimes against lesbians and gay men: Issues for research and policy. American Psychologist, 44, 948-955. Herek, G. M. (1992). Psychological heterosexism and anti-gay violence. In G. Herek & K. Berrill (Eds.), Hate crimes (pp. 149-169). Thousand Oaks, CA: Sage. Herek, G. M. (1993). Documenting prejudice on campus: The Yale Sexual Orientation Survey. Journal of Homosexuality, 25(4), 15-30. Herek, G. M. (2000). The psychology of sexual prejudice. Current Directions in Psychological Science, 9, 19-22. Herek, G. M., & Berrill, K. (1992). Documenting the victimization of lesbians and gay men: Methodological issues. In G. Herek & K. Berrill (Eds.), Hate crimes (pp. 270-286). Thousand Oaks, CA: Sage. Herek, G. M., Gillis, J. R., & Cogan, J. C. (1999). Psychological sequelae of hate crime victimization among lesbian, gay, and bisexual adults. Journal of Consulting and Clinical Psychology, 67, 945-951. Herek, G. M., Gillis, J. R., Cogan, J. C., & Glunt, E. K. (1997). Hate crime victimization among lesbian, gay, and bisexual adults: Prevalence, psychological correlates, and methodological issues. Journal of Interpersonal Violence, 12, 195-215. Janoff-Bulman, R. (1992). Shattered assumptions. New York: Free Press. Jenness, V., & Broad, K. (1997). Hate crimes: New social movements and the politics of violence. New York: Aldine de Gruyter. Kilpatrick, D. G., Saunders, B. E., Amick- McMullan, A., Best, C. L., Veronen, L. J., & Resnick, Pr e- Pu bli ca tio n D ra ft 16 H. S. (1989). Victim and crime factors associated with the development of crime-related post-traumatic stress disorder. Behavior Therapy, 20, 199-214. 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. Ruggiero, K. M., & Taylor, D. M. (1997). Why minority group members perceive or do not perceive the discrimination that confronts them: The role of self-esteem and perceived control. Journal of Personality and Social Psychology, 72, 373-389. Smith, G. W. (1998). The ideology of “fag”: The school experience of gay students. Sociology Quarterly, 39, 309-335. Swim, J.K., & Stangor, C. (1998). Prejudice: The target’s perspective. San Diego, CA: Academic Press. Pr e- Pu bli ca tio n D ra ft 17 Table 1 Why Respondents Perceived Crime Was Based on Sexual Orientation (Interview Sample) Women Men Total Reason N of % N of % N of % crimes crimes crimes ___________________________________________________________________________ Explicit statements or other unambiguous information 56 51% 89 55% 145 53% Crime location 11 10% 14 9% 25 9% Other contextual cues 19 17% 8 5% 27 10% Hunch 19 17% 25 15% 44 16% Revenge from former male partner (Females) 4 4% 0 0% 4 2% Entrapment (Males) 0 0% 3 2% 3 1% Misclassified crime 1 1% 23 14% 24 9% TOTAL 110 100% 162 100% 272 100% ___________________________________________________________________________ Note. Table entries reflect the number and percentage of victimizations that respondents attributed to their sexual orientation for each reason. Based on interview responses from 76 women and 90 men who reported one or more criminal victimizations because of their sexual orientation. Pr e- Pu bli ca tio n D ra ft 18 Table 2 Proportion of Bias and Non-Bias Adulthood Victimizations Reported to Police Authorities (Questionnaire Sample) Type of Crime Lesbians Gay Men Bisexual Women Bisexual Men ____________________________________________________________________________________________________________ Crimes Against Person Bias 30.4% 38.4% 28.6% 17.2% (n = 79) (n = 138) (n = 14) (n = 29) Non-Bias 47.5% 58.0% 45.1% 55.3% (n = 118) (n = 88) (n = 51) (n = 38) Property Crimes Bias 41.9% 58.2% 44.4% 35.3% (n = 74) (n = 79) (n = 9) (n = 17) Non-Bias 76.3% 77.2% 78.9% 68.8% (n = 316) (n = 250) (n = 52) (n = 32) ____________________________________________________________________________________________________________ Any Crime Bias 36.0% 45.6% 34.8% 23.9% (n = 153) (n = 217) (n = 23) (n = 46) Non-Bias 68.4% 72.2% 62.1% 61.4% (n = 434) (n = 338) (n = 103) (n = 70) ____________________________________________________________________________________________________________ Each cell reports the percentage of respondents in that category who reported the crime to law enforcement authorities and, in parentheses, the total number who experienced that type of crime and were asked about police reporting.