Introduction

Violence against lesbian, gay, bisexual, transgender, and queer (LGBTQ) individuals is not a new phenomenon, with researchers dating anti-LGBTQ hate violence back to the Middle Ages (Boswell, 1980). Currently, bias-motivated violence against LGBTQ individuals is one of the fastest growing forms of hate violence in the United States (Chibbaro, 2019; Fitzsimons, 2019; Herek, 2017). Potok (2010) found that LGBTQ individuals are 2.4 times more likely to be the victim of a hate crime than Jews, 2.6 times more likely than Blacks, 4.4 times more likely than Muslims, 13.8 times more likely than Latinos, and 41.5 times more likely than whites. Echoing Potok’s conclusions, Park and Mykhyalyshyn (2016) found the LGBT people were the most likely minority group to be targeted for hate crimes in 2014. Despite these alarming statistics, substantial attention has not been applied to research on the phenomenon of sexual orientation and gender identity bias hate crime (Perry, 2009). As hate crimes motivated by racial bias historically represent over half of the annual total of hate crimes, criminological research has generally focused on this form of bias crime (Perry, 2001). While racially based bias crime has amassed a substantial body of criminological literature, it is unwise to assume that findings regarding this form of prejudice carry over to other bias motivations (Gruenewald & Allison, 2018; Perry, 2001).

The long-standing deficiency of empirical research on anti-LGBTQ hate has restricted efforts to develop a sound theoretical foundation (Perry, 2009). However, in recent years a basic theoretical framework of hate crime has emerged (Perry & Scrivens, 2017). Hate crimes against the LGBTQ community in the United States are best understood by examining both the heterosexist culture context in which they occur and the psychological context of those who perpetrate them (Herek, 1990). From a cultural perspective, anti-LGBTQ hate crimes can be viewed as radical extensions of common American heterosexism and its bearing on quests to prove heterosexuality (Herek, 1990). Heterosexist views grounded in American history and continued by modern social institutions promote the stigmatization of nonheterosexual behavior regarding gender and sexuality and encourage the need to establish a socially observed heterosexual identity (Ehrlich, 1990; Herek, 1990). From an individual perspective, attitudes of heterosexism and homophobia serve distinct psychological and social functions. Attacking a homosexual allows the perpetrator to establish his own gender identity by exhibiting his own masculinity and by displaying the subordinate or inappropriate masculinity of his homosexual victim (Messerschmidt, 2012). Gender and sexuality are not simply predetermined features of the self, but something that must be demonstrated. Heterosexuals are under intense pressure to “do” gender, to define their heterosexuality and reestablish their heterosexual identity in various situations and conditions. Gender and sexuality are essentially ongoing accomplishments that must be consciously pursued and demonstrated in the public sphere (Messerschmidt, 2012; West & Zimmerman, 1987). Consequently, many individuals, especially young men, feel compelled to publically display their heterosexuality by adhering to heterosexist gender roles and norms of sexuality and/or expressing disapproval for alternative gender and sexual identities (Herek, 1986, Schrock & Schwalbe, 2009).

Attacking a homosexual allows the perpetrator to establish his own gender identity by exhibiting his own masculinity and by displaying the subordinate or inappropriate masculinity of his homosexual victim. Anti-LGBTQ hate crimes and acts of violence are theorized to be mechanisms for constructing masculinity as they allow for an individual to establish himself as an idealized heterosexual male while subjugating other subordinate forms of masculinity (Messerschmidt, 2012; Perry, 2001). Theoretically, perpetrators of anti-LGBTQ simultaneously seek to visually communicate their prejudice, punish perceived violators of traditional notions of sexual orientation and gender identity, and establish themselves as belonging to acceptable social categories of heterosexual and cisgendered individuals. Perpetrators of anti-LGBTQ hate simultaneously seek to visually communicate their prejudice, punish perceived violators of traditional notions of sexual orientation and gender identity, and establish themselves as belonging to acceptable social categories of heterosexual and cisgendered individuals (Messerschmidt, 1993; Perry, 2001; Schrock & Schwalbe, 2009; West & Zimmerman, 1987).

Offender Substance Use

The role of alcohol in facilitating violent and criminal behavior has been well established. Alcohol is often cited as a factor contributing to cognitive, physiological, and emotional changes that increase the probability of aggression (Bushman & Cooper, 1990). As hate crime is a type of aggression, it too has been associated with alcohol use. Several works of empirical research have asserted that perpetrators of hate crime frequently consumed alcohol prior to attacks (Aronowitz, 1994; Bowling, 1994; Comstock, 1991; Dunbar, 2003; Hamm, 1993; Levin and McDevitt, 1993; Messner, McHugh, & Felson, 2004; Stacey, 2011). The link between alcohol use and aggression suggests that perpetrators of hate violence are likely to consume alcohol prior to the incident.

Independent of the link between alcohol and aggression, alcohol may also incite anti-LGBTQ hate due to an association between drinking alcohol and masculinity. “Alcohol consumption factors into the equation because such behavior is practically universally associated with being a man or achieving manhood…Males equate drinking with masculinity and, perhaps, more importantly, this male bonding exercise is often laced with violence” (Bufkin, 1999, p. 166). From Dyck’s (1980) barroom scrapping to Vigil’s (1998) gang members drinking to get loco to Hamm’s (1993) drinking beer and going berserk, excessive drinking, manliness, and violence are inextricably linked. Drinking is a key element in the construction of masculinity, an element that facilitates not just aggression, but “higher levels of sexual prejudice and masculine gender role stress” (Parrott, Gallagher, Vincent, & Bakerman, 2010, p. 520). The association between alcohol consumption, aggression, and masculinity appears to predict an increased proportion of anti-LGBTQ hate offenders that consume alcohol prior to the commission of the incident.

Number of Offenders

Similar to other types of group-perpetrated violence, group-perpetrated hate crime has parallel advantages, namely dilution of blame, endorsement of prejudicial attitudes, anonymity, and a lower risk of counterattack (Gruenewald & Allison, 2018). Summed up by Levin and McDevitt (1993),

The hatemongers who instigate an altercation believe that they are less likely to be hurt in a group because they have their friends to protect them. The group also grants a certain degree of anonymity. If everyone participated, then no one person can easily be singled out as bearing primary responsibility for the attack. Because they share it, the blame is diluted. Finally, the group gives its members a dose of psychological support for their blatant bigotry. (p. 17)

These psychological and physical advantages of general crime group offending are likely analogous to hate crime group offending.

In addition to the aforementioned advantages, the theoretical underpinnings of hate violence motivation suggest another possible reason for group offending. Demonstration of hegemonic masculinity in the presence of others is key to establishing an identity that is viewed as in line with sociocultural normative expectations (Gruenewald & Allison, 2018). “To the extent that members of society know their actions are accountable, they will design their actions in relation to how they might be seen and described by others” (West & Fenstermaker, 1995, p. 25). An individual’s commitment to traditional notions of gender is ideally perceived by others and deserving of positive feedback. As such, “bias offending provides an opportunity for a collective of predominantly males to accomplish hegemonic masculinity…[in which] recognition and membership in that group are enhanced when the Other is attacked” (Bufkin, 1999, p. 163). If an act of hate is not perceived by others, it does nothing to further the perpetrator’s outward hegemonic appearance. Furthermore, the decreased likelihood of a counterattack by the victim in a group-perpetrated attack minimizes the potential for an offender to fail, look weak, or lose the fight (Bufkin, 1999) As group-perpetrated hate crime provides the advantages associated with general crime as well as an increased guarantee of establishing hegemonic masculinity, theory predicts more co-offending amongst hate crime and anti-LGBTQ hate crime offenders.

Crime Location

Hate crime, just like general crime, can occur anywhere including home, work, school, and public places. Some hate crime researchers argue that hate crime is most likely occur in the course of routine daily life. “Generally speaking there is no particular place that can be categorized as constituting the typical location for hate crimes…since they tend to occur in many different locations” (Bleich, 2008, p. 48). On the other hand, other researchers suggest that hate crimes may occur more often in public spaces for three distinct reasons. First, hate crimes are more likely than non-bias crimes to be perpetrated by unknown assailants and, therefore, more likely to take place outside the home where strangers interact (Berk, Boyd, & Hamner, 1992; Messner, McHugh, & Felson, 2004; Stacey, 2011). Second, hate crimes, in part, serve to send a message to a minority community that they are of inferior value. Hate crime “is not only to subordinate the victim, but also subdue his or her community, to intimidate a group of people” (Perry, 2001, p. 10). Spreading an ideology of hate is better achieved when there is an audience, in addition to the victim, to hear the message. Third, Comstock (1991) asserted that public locations that were “gay identified” such as lesbian or gay bars, discos, or bathhouses were common sites for an anti-LGBTQ hate crime because victims could more easily be identified as belonging to the LGBTQ community. These theories predict that hate crime is more likely to occur in public locations than general crime.

In recent literature, researchers attempted to disaggregate the motivations behind hate crime. In the examination of anti-LGBTQ hate murder, Kelley and Gruenewald (2015) and Tomsen (2009) argued that while each attack involved the perpetrator’s attempt to establish masculinity, two different scenarios were likely. The first anti-LGBTQ hate murder scenario involved predatory attacks, lacking victim provocation. These incidents, similar to the public bias attacks described above by Berk, Boyd, Hamner (1992), Perry (2001), and Stacey (2011), were either carried out to send a message of hatred toward LGBTQ individuals or to target allegedly weak LGBTQ victims for an instrumental purpose. As spreading an anti-LGBTQ message required an audience to view the offense or to join in participating, it was more likely to occur in a public space. Also, an instrumental crime in which the victim was selected for his or her LGBTQ identity, was likely to occur in a public space where strangers came into contact with one another (Kelley & Gruenewald, 2015; Tomsen, 2009). The second anti-LGBTQ hate murder scenario involved a violent response to a challenge of sexuality. Undesired sexual advances, cases of mistaken gender or sexual identity, or perceptions of wrongdoing on the part of the victim triggered the perpetrator to try and save face and restore his hegemonic masculinity. These attacks, as they involved intimate contact and personal confrontation, were more likely to occur in a private location such as a residence (Kelley & Gruenewald, 2015). The development of two alternate motivations for hate, one public and predatory and one personal and reactive, suggests that the percent of crimes taking place in public may not be as overwhelming as the above earlier developed theories predicted.

Victim/Offender Relationship

Most definitions of hate crime include that the victim was chosen purely based on his or her membership in a particular minority group (Mason, 2005; Medoff, 1999; Perry, 2001). Two explanations of why stranger-perpetrated hate crimes are likely to occur dominate the literature. First, the individual identity of the victim was not significant to the offender. “The victim simply represents the Other in generic terms. That he or she is a member of the hated or demonized group is enough to leave them vulnerable to attack. Further knowledge of their identity, personality, or intent was unnecessary” (Perry, 2001, p. 29). The symbolic status of the victim, not his or her personal identity or relationship to the offender, was important. Franklin (1998) argued that the victim’s identity, other than his or her membership in a particular minority group, could be of so little importance that the victim was fundamentally a “dramatic prop” or a “vehicle for ritualized conquest” (p. 12). Second, Berk, Boyd, and Hamner (1992) suggested that preexisting stereotypes dominated the perpetrator’s view of the victim. Without particular information about the victim that contradicted negative stereotypes, it was easier to express hostility and aggression. The victim’s humanity and other positive attributes are not acknowledged due to the interpersonal distance between the victim and offender.

Despite the frequent categorization of hate crime as a form of “stranger danger,” Mason (2005) asserted that several empirical studies have effectively challenged this conception. Perpetrators may, in fact, be more likely than not to be someone the victim knows (Bowling, 1993; Mason, 1997; Stanko, 2001; Tomsen & Mason, 2001; von Schulthess, 1992). Stanko (2001) suggested that conceptualizing hate crime as only motivated by prejudice was misleading. Hate may intersect with social context and interpersonal relationships. “The use of race or homophobic hatred is somehow ‘purely’ political and discriminatory, uncontaminated by social contexts that may characterize [a] dispute as really about argumentative neighbors than about intolerance toward a person who is racially, ethnically different or different because of sexuality” (Stanko, 2001, p. 322). Hate crime may happen in entirely ordinary situations, distinctly different from the stranger danger scenario, and involving perpetrators the victim knows. Labeling a hate crime offender as a stranger serves to “manage that uncontrollable fear by displacing the figure of the stranger and thereby revaluing it, controlling it, marginalizing it, willing it away” (Moran, 2007, p. 434). Simultaneously, this creates a safe space occupied by known individuals in which the potential victim exists with minimal risk (Chakraborti & Garland, 2009; Moran, 2007).

Regarding anti-LGBTQ hate crimes, research is just as mixed. Several researchers cited strangers as being the primary perpetrators of anti-LGBTQ hate (Berk, Boyd, & Hamner, 1992; Comstock, 1991; Perry, 2001). However, just as with hate crimes in general, other researchers asserted the role that known offenders can play in perpetrating anti-LGBTQ hate crime. Schoolmates, workmates, relatives, intimates, friends, and acquaintances have all been included in the pool of potential anti-LGBTQ hate offenders.

Opposing views regarding the relationship between a hate crime perpetrator and his or her victim make predictions difficult. While the stranger danger paradigm predicts more unknown perpetrators the theory that hate is likely to occur during routine activities between people who at least vaguely know each other would not predict such a large proportion of unknown perpetrators.

Methods

Examining data from four national level datasets, this study aimed to investigate the situational dynamics of anti-LGBTQ hate crime in an effort to provide a comprehensive picture of how anti-LGBTQ hate transpires. As three of these datasets also include data for general crime and hate crimes of all motivations (race, religion, ethnicity/nationality, disability, gender, sexual orientation, and gender identity), comparisons were made across general crime, hate crime, and anti-LGBTQ hate crime. The previously discussed theory served as a framework for reviewing the offending and situational characteristics of anti-LGBTQ hate crime and how they compared to general crime and hate crime.

Research Question 1:

What are the predominant situational dynamics (offender substance use, number of offenders, crime location, victim/offender relationship) associated with anti-LGBTQ hate according to four national crime datasets?

Research Question 2:

How do the situational dynamics of anti-LGBTQ hate differ from the situational dynamics of general crime and from hate violence including all bias motivations?

Datasets

This study utilized four national crime datasets.

  1. The Uniform Crime Report (UCR) is compiled by the Federal Bureau of Investigation and includes crimes known to city, university/college, county, state, tribal, and federal law enforcement agencies. The UCR collects summary data on eight Part I offenses reported to law enforcement including murder and nonnegligent manslaughter, forcible rape, robbery, aggravated assault, burglary, larceny-theft, motor vehicle theft, and arson. Since 1990, the UCR has also collected data about crimes motivated by biases regarding race, gender, gender identity, religion, disability, sexual orientation, and ethnicity. As the UCR’s hate crime data collection was structured akin to the National Incident-Based Reporting System (NIBRS), the FBI’s other data collection program, it deviates from original UCR’s methodology and makes comparisons across general crime, hate crime, and anti-LGBTQ hate crime limited.
  2. The National Incident-Based Reporting System (NIBRS) is also compiled by the Federal Bureau of Investigation, but was designed to collect more comprehensive crime data than the summary data included in the UCR. It provides data on 23 offense categories comprised of 49 specific Group A offenses for both bias and non-bias motivated offenses as well as a multitude of crime elements, of which several are related to the scope of this study.
  3. The National Crime Victim Survey (NCVS) is a data collection of crime victimization survey results gathered by the Bureau of Justice Statistics. Data is obtained from “a nationally representative sample of about 90,000 households, comprising nearly 160,000 persons, on the frequency, characteristics, and consequences of criminal victimization in the United States” (Bureau of Justice Statistics, 2016a, para. 3). The NCVS includes victim reported hate crimes categorized by bias motivation allowing for the analysis of general crime, hate crime, and anti-LGBTQ hate crime across situational variables.
  4. Since 1998, the NCAVP has issued an annual report on LGBTQ hate violence “to document and raise awareness of the prevalence of this violence, advocate for public policy and funding changes that will increase resources to address LGBTQ violence, and recommend strategies to prevent, respond to, and end this violence” (NCAVP, 2012, para. 4). Each annual NCAVP Hate Violence Report included data collected by NCAVP member and ally organizations during an individual calendar year. The number of contributing organizations varied by year ranging from thirteen to sixteen organizations per year for each year of data included in the current study. Each contributing organization collected information from surviving victims of anti-LGBTQ hate violence. Such information was obtained after the victim contacted the organization in person, via a telephone hotline, or by filling out a report online. As the NCAVP dataset only included anti- LGBTQ hate incidents, comparisons could not be made within this source regarding general crime and hate crimes of other motivations.

While UCR, NIBRS, and NCVS datasets are publically available and frequently utilized in academic research, the data behind the National Coalition of Anti-Violence Programs’ (NCAVP) Report on Lesbian, Gay, Bisexual, Transgender, Queer, and HIV-Affected Hate Violence has only been evaluated in self-published annual reports and has never been released for outside analysis prior to this study. Extended communication with the NCAVP resulted in the contribution of seven years of anti-LGBTQ hate incident data ranging from 2007 through 2013, with conversations continuing to date to achieve the release additional, more recent data. UCR, NIBRS, and NCVS data was compiled from public databases from 2007 through 2013 as well to allow for comparisons across datasets in the same time frame. The inclusion of seven years of data helped to increase the sample size of select variables. Compiling data for seven years helped to create a large enough sample to be confident in the results of the current analysis. Furthermore, the inclusion of two law enforcement report datasets and two self-report datasets allowed for a more robust collection of data including both reported and unreported incidents.

Variables

Situational dynamic variables were selected based on two criteria: (1) theoretical relevance; and (2) inclusion in the NCAVP dataset. While UCR, NIBRS, and NCVS data have been analyzed before in various contexts, this is the first time the NCAVP data has been made available for outside analysis. Special consideration was given to ensure optimal analysis of the NCAVP dataset since this is the first time its data was analyzed for scholarly research. Consequently, variables were selected from the UCR, NIBRS, and NCVS datasets for inclusion in this study if they were included in the NCAVP dataset. The formation of variable categories in the UCR, NIBRS, and NCVS datasets was dictated by the variable categories utilized in the NCAVP dataset (see Appendix A for variable categories).

Analytical Framework

The descriptive portion of this study utilized three crime type categories: general crime, hate crime, and anti-LGBTQ hate crime. As hate crime is a form of general crime, incidents of hate crime were included in the general crime category. As anti-LGBTQ hate crime is a form of hate crime, incidents of anti-LGBTQ hate crime were included in the hate crime category. Proportions of situational dynamic characteristics were calculated based on these crime types. Since all variables in the current study were categorical, chi-square tests were appropriate to elucidate the statistical significance of the relationship between crime type and each situational dynamic variable. In order to conduct these tests of statistical significance, the crime type categories had to be modified. Each crime type needed to be coded based on its membership to one crime type category (general crime = 1, hate crime = 2, and anti-LGBTQ hate crime = 3). However, despite the membership of hate crimes to both groups 1 (general crime) and 2 (hate crime) according to the definitions utilized in the descriptive portion of this study, they could only be included in one crime type category for this analysis. Hate crimes, therefore, were coded as belonging to group 2 (hate crime) and were excluded from group 1 (general crime). Similarly, anti-LGBTQ hate crimes belonged to both group 2 (hate crime) and group 3 (anti-LGBTQ hate crime) according to the definitions utilized in the descriptive portion of this study. Anti-LGBTQ hate crimes were coded as belonging to group 3 (anti-LGBTQ hate crimes) and were excluded from group 2 (hate crime). Since anti-LGBTQ hate crimes comprised a minor portion of the hate crime group (15.35%, 16.78%, and 19.43%) and hate crimes comprised an even smaller portion of the general crime group (0.05%, 0.06%, and 1.39%), it was unlikely that these nested subgroups drove the higher-order group they were removed from. Even though removing hate crimes from the general crime type category and anti-LGBTQ hate crime category created different groups than the descriptive analysis, forming distinct crime type categories was necessary to conduct the chi-square analyses and likely resulted in minimal divergence.

Results

From 2007–2013, the UCR recorded 73,774,079 Part I Offenses, NIBRS recorded 43,004,865 Group A Offenses, and NCVS recorded 60,109 criminal victimizations. During this same time frame, the UCR recorded 47,399 hate crimes, NIBRS recorded 21,555 hate crimes, and NCVS recorded 834 criminal victimizations identified as hate crimes. When these hate incidents were classified by bias motivation, the UCR recorded 9,209 anti-LGBTQ hate crimes, NIBRS recorded 3,618 anti-LGBTQ hate crimes, NCVS recorded 128 anti-LGBTQ hate victimizations, and NCAVP recorded 17,999 anti-LGBTQ hate victimizations. General crime and hate crime totals are absent for the NCAVP dataset as it only included anti-LGBTQ hate incidents (see Table 1).

Table 1

Crime category totals.

UCR NIBRS NCVS NCAVP

General Crime 73,774,079 43,004,865 60,109 NA
Hate Crime 47,399 21,555 834 NA
Anti-LGBTQ Hate Crime 9,209 3,618 128 17,999

Offender Substance Use

The percentages calculated regarding offender substance use did not present a uniform picture on the extent of offender substance use before the commission of an act of anti-LGBTQ hate (see Table 2). While the NIBRS and NCAVP datasets revealed a somewhat similar breakdown of 13.36% and 19.05% of offenders identified as being under the influence of a substance during an anti-LGBTQ attack, the NCVS shows an alarming 83.33% of offenders were identified as being under the influence. The number of cases in the NCVS dataset in which an anti-LGBTQ hate crime victim reported whether the offender used substances prior to the incident was relatively small at 54. This may be responsible for the skewed results, but further investigation is needed to make a more substantial conclusion.

Table 2

Descriptive statistics: Offender substance use.

UCR NIBRS NCVS NCAVP

General Crime N = 39,442,390 N = 8,020
Yes 9.51% 48.68%
No 90.49% 51.32%
Hate Crime N = 19,911 N = 316
Yes 11.41% 60.13%
No 88.59% 39.87%
Anti-LGBTQ Hate Crime N = 3,346 N = 54 N = 168
Yes 13.36% 83.33% 19.05%
No 86.64% 16.67% 80.95%

Chi-square analyses revealed that the relationship between offender substance use and crime type within the NIBRS and NCVS datasets were statistically significant (NIBRS: χ2 (2, N = 39,544,686) = 84.35, p < .001, NCVS: χ2 (2, N = 8,773) = 34.06, p < .001) (see Table 3). Both the NIBRS and NCVS datasets, while showing disparate representations of anti-LGBTQ hate crime offender substance use, showed the same offender substance use trends across general crime, hate crime, and anti-LGBTQ hate crime. The NIBRS dataset showed that 9.51% of general crime offenders, 11.41% of hate crime offenders, and 13.36% of anti-LGBTQ hate crime offenders used substances prior to the incident. The NCVS dataset showed that 48.68% of general crime offenders, 60.13% of hate crime offenders, and 83.33% of anti-LGBTQ hate crime offenders used substances prior to the incident. While these findings show a considerably dissimilar picture of substance use among offenders of general crime (9.51% vs. 48.68%), hate crime (11.41% vs. 60.13%), and anti-LGBTQ hate crime (13.36% vs. 83.33%), they both show increased offender use of substances among hate crime offenders and even more so among anti-LGBTQ hate crime offenders.

Table 3

Chi-square results: Offender substance use.

N χ2 df p Statistically Significant Cramér’s V

UCR
NIBRS 39,544,686 84.35 2 .00 Yes .001
NCVS 8,773 34.06 2 .00 Yes .062

Number of Offenders

All four datasets included information regarding the number of offenders responsible for perpetrating each anti-LGBTQ hate crime. The UCR, NIBRS, NCVS, and NCAVP found that 70.46%, 81.05%, 57.95%, and 73.38% of anti-LGBTQ hate incidents were perpetrated by only one offender. The remaining 29.54%, 18.95%, 42.05%, and 26.62% of anti-LGBTQ hate incidents were perpetrated by two or more offenders (see Table 4). The four datasets showed significant variation between the number of single offender and multiple offender anti-LGBTQ hate crimes. NIBRS reported the highest number of single offender anti-LGBTQ hate crimes at 81.05% and the lowest number of multiple offender anti-LGBTQ hate crimes at 18.95%. Conversely, the NCVS reported the lowest number of single offender anti-LGBTQ hate crimes at 57.95% and the highest number of multiple offender anti-LGBTQ hate crimes at 42.05%. These two datasets diverged by a significant 23.1%. Of note, while the NCVS dataset did reveal a sizable portion of multiple perpetrator anti-LGBTQ hate crimes, the sample size for this crime category was considerably smaller than the other datasets at only 88.

Table 4

Descriptive statistics: Number of offenders.

UCR NIBRS NCVS NCAVP

General Crime N = 28,474,213 N = 15,566
1 79.76% 81.48%
≥2 20.24% 18.52%
Hate Crime N = 31,356 N = 13,522 N = 599
1 75.11% 79.69% 63.44%
≥2 24.89% 20.31% 36.56%
Anti-LGBTQ Hate Crime N = 7,001 N = 2,290 N = 88 N = 4,253
1 70.46% 81.05% 57.95% 73.38%
≥2 29.54% 18.95% 42.05% 26.62%

The NIBRS dataset indicated that 79.76% of general crimes, 79.69% of hate crimes, and 81.05% of anti-LGBTQ hate crimes involved one offender and 20.24%, 20.31%, and 18.95% involved two or more offenders. A chi-square analysis revealed that the relationship between number of offenders and crime type within the NIBRS dataset was not statistically significant (NIBRS: χ2 (2, N = 28,364,905) = 3.95, p = .14) (see Table 5). However, chi-square analyses revealed that the relationship between number of offenders and crime type within the UCR and NCVS datasets were statistically significant (UCR: χ2 (1, N = 31,356) = 104.27, p < .001, NCVS: χ2 (2, N = 15,727) = 112.15, p < .001). Regarding hate crimes, the UCR found 75.11% were perpetrated by one offender and 24.89% were perpetrated by two or more offenders, while 70.46% of anti-LGBTQ hate crimes were perpetrated by one offender and 29.54% were perpetrated by two or more offenders. The UCR only included a co-offending variable in its collection of hate crime information, making the number of offenders of general crime unavailable for analysis. The NCVS found 81.48% of general crimes, 63.44% of hate crimes, and 57.95% of anti-LGBTQ hate crimes were perpetrated by one offender, while 18.52% of general crimes, 36.56% of hate crimes, and 42.05% of anti-LGBTQ hate crimes were perpetrated by two or more offenders. As compared to hate crime, the UCR shows an increase in the number of multiple perpetrator anti-LGBTQ hate crimes. Similarly, the NCVS shows that multiple perpetrators were involved in a higher percentage of hate crimes than general crimes, and even more anti-LGBTQ hate crimes. While the number of hate crimes and anti-LGBTQ hate crimes committed by multiple perpetrators did not constitute the predicted majority, these two datasets showed that anti-LGBTQ hate crimes were more likely to involve multiple perpetrators than hate crimes and general crimes.

Table 5

Chi-square results: Number of offenders.

N χ2 df p Statistically Significant Cramér’s V

UCR* 31,356 104.27 1 .00 Yes .058**
NIBRS 28,364,905 3.95 2 .14 No .000    
NCVS 15,727 112.15 2 .00 Yes .084    

* Only includes two crime types: hate crimes and anti-LGBTQ hate crimes.

** As chi-square is 2 × 2, Cramér’s phi is the appropriate measure of association.

Crime Location

All four datasets presented relatively analogous information regarding anti-LGBTQ crime location (see Table 6). Each dataset indicated that the most amount of anti-LGBTQ hate crime occurred in residential settings and in open spaces such as streets, sidewalks, and parking lots. The UCR, NIBRS, NCVS, and NCAVP reported 36.45%, 40.26%, 50.00%, and 33.39% of anti-LGBTQ hate crimes occurred in the victim’s home or someone else’s home. An additional 36.31%, 31.61%, 28.13%, and 27.45% occurred in open spaces. The third most likely location for anti-LGBTQ crime was commercial locations. Smaller percentages occurred at schools and in other location such as government buildings, religious sites, and in law enforcement custody.

Table 6

Descriptive statistics: Crime location.

UCR NIBRS NCVS NCAVP

General Crime N = 36,741,116 N = 60,109
Residence 47.52% 66.31%
Commercial 21.06% 6.08%
Open Space 26.44% 15.44%
School 3.23% 6.99%
Other 1.75% 5.18%
Hate Crime N = 13,575 N = 18,699 N = 821
Residence 36.36% 38.21% 47.87%
Commercial 12.01% 16.16% 9.38%
Open Space 31.34% 29.34% 24.00%
School 13.22% 10.03% 11.21%
Other 7.07% 6.26% 7.55%
Anti-LGBTQ Hate Crime N = 8,161 N = 3,157 N = 128 N = 8,295
Residence 36.45% 40.26% 50.00% 33.39%
Commercial 14.23% 16.85% 9.38% 21.53%
Open Space 36.31% 31.61% 28.13% 27.45%
School 10.53% 8.93% 8.59% 8.04%
Other 2.49% 2.34% 3.91% 9.58%

Both the NIBRS and NCVS datasets included crime location information for all three crime categories, while the UCR included crime location for the hate crime and anti-LGBTQ hate crime categories. All three datasets yielded statistically significant relationships between crime type and crime location (UCR: χ2 (4, N = 13,575) = 965.94, p < .001, NIBRS: χ2 (8, N = 44,629,351) = 3,450,82, p < .001, NCVS: χ2 (8, N = 60,109) = 130.02, p < .001) (see Table 7). The most substantial trend was the increased number of anti-LGBTQ hate crimes occurring in open spaces such as streets, sidewalks, parks, and parking facilities. UCR documented 31.34% of hate crimes and 36.31% occurring in open spaces. NIBRS documented crime in open spaces occurring in 26.44% of general crime, 29.34% of hate crime, and 31.61% of anti-LGBTQ hate crime. NCVS documented crime in open spaces occurring in 15.44% of general crime, 24.00% of hate crime, and 28.13% of anti-LGBTQ hate crime.

Table 7

Chi-square results: Crime location.

N χ2 df p Statistically Significant Cramér’s V

UCR* 13,575 965.94 4 .00 Yes .267**
NIBRS 44,629,351 3,450.82 8 .00 Yes .006    
NCVS 60,109 130.02 8 .00 Yes .033    

* Only includes two crime types: hate crimes and anti-LGBTQ hate crimes.

** As chi-square is 2 × 2, Cramér’s phi is the appropriate measure of association.

Victim/Offender Relationship

The three datasets that offered information on anti-LGBTQ hate crime victim/offender relationship created a disparate picture (see Table 8). NIBRS found that 81.69% of victim knew their offender, NCVS found that 64.47% knew their offender, and NCAVP found that 57.86% knew their offender. While these varied considerably, they did present the important conclusion that many anti-LGBTQ hate crime victims were victimized by people they know.

Table 8

Descriptive statistics: Victim/Offender relationship – Known/Unknown.

UCR NIBRS NCVS NCAVP

General Crime N = 9,794,574 N = 14,311
Known 83.96% 65.73%
Unknown 16.04% 34.27%
Hate Crime N = 4,748 N = 538
Known 84.34% 61.45%
Unknown 15.66% 38.55%
Anti-LGBTQ Hate Crime N = 830 N = 76 N = 10,444
Known 81.69% 64.47% 57.86%
Unknown 18.31% 35.53% 42.14%

When known offenders were classified based on their previous relationship with the victim, all three datasets indicated that offenders were most likely to be a friend or acquaintance (ex. neighbor, workmate, etc.) (see Table 9). Family members comprised 20.65% of anti-LGBTQ hate offenders in the NIBRS dataset, 12.24% in the NCVS dataset, and 9.41% of the NCAVP dataset. The small sample size of NCVS offenders did not include any intimates or former intimates of the victims, however, both NIBRS and NCAVP did. NIBRS and NCAVP found 27.43% and 10.18% of offenders were the current or former lover of the victim.

Table 9

Descriptive statistics: Victim/Offender relationship – Known type.

UCR NIBRS NCVS NCAVP

General Crime N = 8,223,471 N = 8,309
Acquaintance 49.11% 63.57%
Intimate 31.34% 20.98%
Family 19.55% 15.45%
Hate Crime N = 4,013 N = 108
Acquaintance 49.61% 84.01%
Intimate 31.50% 7.06%
Family 18.89% 8.92%
Anti-LGBTQ Hate Crime N = 678 N = 35 N = 5,961
Acquaintance 51.92% 87.76% 80.41%
Intimate 27.43% 0.00% 10.18%
Family 20.65% 12.24% 9.41%

NIBRS and NCVS also presented varying descriptions of victim/offender relationships as they changed across general crime, hate crime, and anti-LGBTQ hate crime (see Table 10). NIBRS showed that 83.96% of general crime offenders were known to their victims, 84.34% of hate crime offenders were known to their victims, and 81.69% of anti-LGBTQ hate crime offenders were known to their victims. On the other hand, NCVS showed that 65.73% of general crime offenders were known to their victims, 61.45%% of hate crime offenders were known to their victims, and 64.47% of anti-LGBTQ hate crime offenders were known to their victims. The relationship between victim/offender relationship: known/unknown offenders and crime type was not statistically significant in either the NIBRS or NCVS datasets (NIBRS: χ2 (2, N = 8,136,241) = .82, p = .66, NCVS: χ2 (2, N = 14,308) = 4.04, p = .13).

Table 10

Chi-square results: Victim/Offender relationship – Known/Unknown.

N χ2 df p Statistically Significant Cramér’s V

UCR
NIBRS 8,136,241 .82 2 .66 No .000
NCVS 14,308 4.04 2 .13 No .017

Within the NCVS dataset, anti-LGBTQ hate crimes involved more acquaintances as known offenders at 87.76% than hate crime at 84.01% and general crime at 63.57% (see Table 11). Anti-LGBTQ hate crimes were also the least likely to involve current or former intimates at 0.00% as compared to 7.06% and 20.98% of hate and general crime, respectively. The relationship between known offender type and crime type was statistically significant in the NCVS dataset (NCVS: χ2 (4, N = 8,165) = 57.81, p < .001). In the NIBRS dataset, acquaintances committed a slightly higher percentage of anti-LGBTQ hate crimes at 51.92% as opposed to 49.11% of general crime and 49.61% of hate crime. Intimates committed fewer anti-LGBTQ hate crimes at 27.43% than hate crime and general crime at 31.50% and 31.34%, respectively. However, despite reflecting a similar directional trend to the NCVS data in this regard, the relationship between known offender type and crime type was not statistically significant in the NIBRS dataset (NIBRS: χ2 (4, = 7,040,976) = 4.98, p < .29).

Table 11

Chi-square results: Victim/Offender relationship – Known type.

N χ2 df p Statistically Significant Cramér’s V

UCR
NIBRS 7,040,976 4.98 4 .29 No .000
NCVS 8,165 57.81 4 .00 Yes .059

Discussion

The theoretical notion that anti-LGBTQ hate is a unique form of interpersonal aggression forecasted differences between its situational dynamics and those evident in other forms of violence. Analysis of situational dynamics variables revealed several significant distinctions between anti-LGBTQ hate crime, hate crime, and general crime. First, the relationship between offender substance use and crime type was statistically significant in both datasets that contained data on offender substance use. While percentages of substance use by anti-LGBTQ hate offenders varied greatly across the NIBRS and NCVS datasets, both datasets found that anti-LGBTQ hate crime offenders were more likely to be under the influence during the commission of their crimes than hate crime offenders and even more so than general crime offenders. These findings support the theory that substance use may fuel anti-LGBTQ hate more than general forms of violence and other forms of bias motivated hate. Beyond facilitating aggressive behavior, alcohol consumption is uniquely associated with male bonding, establishing masculinity, and sexual prejudice.

Second, the relationship between number of offenders and crime type was statistically significant in two of the three datasets with number of offender data. In the NCVS dataset, anti-LGBTQ was more likely to involve multiple offenders than hate crime and general crime. In the UCR dataset, anti-LGBTQ hate crime was more likely to involve multiple offenders than hate crime. The UCR’s exclusion of number of offender data for general crime made comparisons to general crime impossible. The NIBRS dataset yielded a nonsignificant relationship between number of offenders and crime type. Viewed collectively, the data supports the notion that multiple perpetrator anti-LGBTQ hate crimes are more likely than other forms of crime, potentially to allow the offenders to visibly demonstrate masculinity to others with lowered risk of failure, injury, feelings of culpability, and identification as a suspect.

Third, the relationship between known victim type and crime type was statistically significant in the NCVS dataset. Acquaintances comprised higher proportions of anti-LGBTQ hate crime known offenders than hate crime known offenders and general crime known offenders. Intimates also comprised lower proportions of anti-LGBTQ hate crime known offenders than hate crime known offenders and general crime known offenders. NIBRS data mimicked these trends, to a lesser extent, but did not yield statistically significant results.

Fourth, the relationship between crime location and crime type was significant in all three datasets that included crime location data. The UCR, NIBRS, and NCVS datasets found that anti-LGBTQ hate crimes contained higher proportions of crimes that occurred in open spaces. The increased likelihood of anti-LGBTQ hate crime occurring in open spaces such as streets, sidewalks, and parks is in agreement with the theory that anti-LGBTQ hate is a mechanism for establishing an individual’s masculinity, not just to himself, but to those witnessing the incident. Furthermore, anti-LGBTQ hate incidents that occur in open spaces allow for the spread the message that nonnormative sexual orientations and gender identities will not be tolerated in the public domain.

Based on the above analysis, some overarching statements about anti-LGBTQ hate crime can be put forward. Incidents of anti-LGBTQ hate were most likely committed by single offenders known to their victims in open space locations. Anti-LGBTQ hate crimes were more likely than hate crimes and general crimes to involve offenders who were under the influence of alcohol or drugs during the time of the incident. While lone offenders committed the majority of anti-LGBTQ hate incidents, anti-LGBTQ hate crimes were more likely to involve multiple perpetrators than hate crimes and general crimes. Anti-LGBTQ hate crimes were also more likely to involve known offenders described as acquaintances and less likely to involve current or former intimate partners.

A key set of findings worthy of its own discussion focuses on the heterogenous nature of anti-LGBTQ hate violence. The existing literature largely characterizes anti-LGBTQ hate as motivated by pure disgust and prejudice, in which an unknown offender selects his victim solely based on membership to a minority sexual orientation or gender identity group. The offender commits this act in a public location in order to spread a message of hate and intolerance. Instead, the current data analysis shows that many anti-LGBTQ hate incidents occurred in private residences, areas where there is less interaction amongst strangers and where there is a smaller audience for displays of hate and masculinity. In addition, substantial percentages of anti-LGBTQ hate crimes were perpetrated by offenders known to the victim. Most of these known offenders were acquaintances such as friends, neighbors, schoolmates, and workmates, but others were family members and former and current intimate partners. Kelley and Gruenewald (2015) and Tomsen (2009) offer a more plausible explanation for these findings, theorizing that expressions of anti-LGBTQ hate are not homogenous and may be reactionary as well as being predatory. Anti-LGBTQ hate crimes can be unplanned, involving a response to conscious and unconscious victim provocation. Instead of selecting a nonconforming LGBTQ individual as prey in order to establish masculine dominance, anti-LGBTQ hate may evolve out of a personal confrontation. Consequently, these forms of anti-LGBTQ hate are more likely to involve known offenders including friends, family members, and intimates in more private settings. Arguing that some anti-LGBTQ hate crimes are reactionary instead of predatory in nature does not automatically exclude the significant role that masculinity plays in the incident. Interpersonal conflicts that spur reactionary hate may involve challenges to the offender’s gender or sexuality such as undesired sexual advances, mistaken gender identity, or any other insult that is exacerbated by the fact that it came from an individual belonging to a subordinate social group. Committing an act of anti-LGBTQ hate becomes a way to save face after a situation diminished the offender’s hegemonic masculinity.

Conclusion

The purpose of this study was to conduct an in-depth analysis of anti-LGBTQ hate crime dynamics in order to further understand anti-LGBTQ hate as a unique form of aggression. Analysis of situational dynamic variables provided support for the notion that anti-LGBTQ hate is a distinct type of criminal incident. Increased proportions of offender substance use, number of offenders, unknown offenders, acquaintances as known offenders, and crimes taking place in open spaces substantiates the theory that anti-LGBTQ hate crime is qualitatively unique, typified by different characteristics than other forms of crime and aggression. In addition to uncovering distinct situational characteristics of anti-LGBTQ hate crime, the data in this study proposes that anti-LGBTQ hate crime is not a homogenous phenomenon. The theoretical literature on hate crime emphasizes that construction of hegemonic masculinity prompts visible, predatory acts of anti-LGBTQ hate committed by strangers, yet this study yielded data that suggests that anti-LGBTQ hate crime is more heterogeneous. Significant numbers of anti-LGBTQ hate crimes committed by known offenders including friends, family, and intimates and crimes committed in private locations such as residences suggest that another dynamic process may underlie this type of crime.

Chakraborti (2015) argues that responses to hate crime are characterized by a disconnect between scholarship and policy. Hate crime scholars and practitioners need to engage in dialogue in order for academic research to inform interventions that result in effective outcomes. With the 2016 Pulse nightclub massacre in the not so distant past, the need for successful interventions and prevention strategies aimed to prevent anti-LGBTQ hate is palpable. While most anti-LGBTQ hate crimes are not carried out on as large of a scale as that in Orlando, anti-LGBTQ hate crimes substantially impact the lives of victims, the LGBTQ community, and the larger society around them. Front line responders including law enforcement, medical personnel, and social service workers, should be aware of the complex nature of anti-LGBTQ hate in order to appropriately classify, intervene, and provide assistance. A lack of understanding by service providers may result in reduced hate crime reporting, improper crime classification, misidentification of victim and/or offender, and failure to provide essential victim services. Conceptualizing anti-LGBTQ hate crime as simply another form of bias-motivated crime and as an unvarying incident with carbon copy crime characteristics is a mistake. Real world responses must be empirically-driven and should take into account its distinct and heterogeneous nature. A complete picture of anti-LGBTQ hate must include that anti-LGBTQ hate is a unique form of victimization typified by its own set of situational dynamics and that anti-LGBTQ hate is not homogenous, but includes distinct narratives potentially indicative of complex underlying motivations and interpersonal dynamics.

Additional File

The additional file for this article can be found as follows:

APPENDIX A

Dataset Variable Categories. DOI: https://doi.org/10.33972/jhs.154.s1