For Whom Does Hate Crime Hurt More? A Comparison of Consequences of Victimization Across Motives and Crime Types

Abstract

Hate crimes have been found to have more severe consequences than other parallel crimes that were not motivated by the offenders’ hostility toward someone because of their real or perceived difference. Many countries today have hate crime laws that make it possible to increase the penalties for such crimes. The main critique against hate crime laws is that they punish thoughts. Instead, proponents of hate crime laws argue that sentence enhancement is justified because hate crimes cause greater harm. This study compares consequences of victimization across groups of victims to test for whom hate crimes hurt more. We analyzed data that were collected through questionnaires distributed to almost 3,000 students at Malmö University, Sweden, during 2013. The survey focused on students’ exposure to, and experiences of, hate crime. A series of separate logistic regression analyses were performed, which analyzed the likelihood for reporting consequences following a crime depending on crime type, perceived motive, repeat victimization, gender, and age. Analyzed as one victim group, victims of hate crime more often reported any of the consequences following a crime compared with victims of parallel non–hate-motivated crimes. And, overall victims of threat more often reported consequences compared with victims of sexual harassment and minor assault. However, all hate crime victim groups did not report more consequences than the non–hate crime victim group. The results provide grounds for questioning that hate crimes hurt the individual victim more. It seems that hate crimes do not hurt all more but hate crimes hurt some victims of some crimes more in some ways.

Keywords

hate crime victimization Sweden consequences

Introduction

Hate crimes have been found to have more severe consequences than other parallel crimes that were not motivated by the offenders’ hostility toward someone because of their real or perceived difference (Iganski & Lagou, 2015). Many countries today have hate crime laws that make it possible to increase the penalties for such crimes. The main critique against hate crime laws is that they punish thoughts. Instead, proponents of hate crime laws argue that sentence enhancement is justified because hate crimes cause greater harm (Iganski & Lagou, 2015). Three aspects in particular are set forth as setting hate crimes apart from other parallel crimes and making hate crimes more severe and hurt more (Levin & McDevitt, 2008). First, hate crimes are proposed to not only affect the initial victim but potentially the entire group with which the victim identifies, via a message effect. When, for example, a homosexual person is attacked because of his or her sexual orientation, it sends a message to the entire LGBTQI1 community that they are not safe, causing the consequences to spread way beyond the initial victim (Iganski, 2001). The second aspect concerns the interchangeability of the victims. Individuals who share an identity characteristic, all risk being the victims of a hate-motivated attack as it is the identity that is the motivation for the crime and not some personal conflict or provocation (Levin & McDevitt, 2008). The third factor is that hate crimes attack the self of the victim. That the motive for an attack is a characteristic that is ascribed, and cannot be changed, is the main explanation to why hate crimes hurt more than other crimes (Craig-Henderson & Sloan, 2003; Herek, Cogan, Gillis, & Glunt, 1997; Iganski, 2001, 2002; Iganski & Lagou, 2015). Because of the randomness of the attack, that is, the victim’s interchangeability, the consequences of being the victim of a hate-motivated attack are also longer lasting compared with other crimes (see, for example, Funnell, 2013). Because hate crimes target people for aspects of the self they cannot change, it creates a distinct reaction from those subjected to crimes without a motive (Avison & Turner, 1988; Craig-Henderson & Sloan, 2003; Thoits, 1991). People who identify with a group that is targeted constantly carry their vulnerability, or risk of attack, with them.

As a means to avoid further victimization, many (potential) victims apply strategies designed to modify some aspects of the self that set them apart, such as changing the style of dress, adjusting the way of speaking, hiding religious symbols, or avoiding certain places (Andersson & Mellgren, 2016a, 2016b). These strategies ultimately limit the activity spaces of potential targets (Funnell, 2015; National Union of Students [NUS], 2011, 2012), and may contribute to further marginalization of already vulnerable groups (Wallengren & Mellgren, 2015).

There is little debate about the fact that hate crimes do hurt individuals and communities and threaten the social fabric of societies. But there are only few empirical studies that investigate the consequences of hate crimes compared with parallel crimes motivated by other factors. Perry (2001) lifted the importance of studying whether, and to what extent, victimization experiences differ between victim identity groups. The decades since Perry’s call for more research comparing consequences between groups has, however, only produced few such investigations (e.g., Garcia, McDevitt, Gu, & Balboni, 2002; Williams & Tregidga, 2014).

To add to the existing literature, this article offers the first comparison of the impact of hate crimes on victims from the following identity groups in Sweden: background/ethnicity, religion/worldview, transgender identity, gender (identifying as man or woman), disability, sexual orientation. Because one of the main arguments for hate crime legislation in the form of enhanced penalties for hate crime offenders is that crimes hurt more compared with parallel crimes (Iganski & Lagou, 2015), this area needs more empirical investigation.

Hate Crime in Sweden

The term hate crime first appeared in Sweden in 1999 in a dissertation by Eva Tiby (1999), although the type of incidents now referred to as hate crimes were recognized before. Hate crime is not a separate offense in Sweden and, thus, has no legal definition, but a hate crime includes all crimes with a motive to offend someone from a particular group. The motive to offend someone because they belong to a particular group, hence referred to as a hate motive, can be seen as an aggravating circumstance under the Swedish Penal Code. The penalty enhancement rule was enacted in 1994, and means that crimes where the perpetrators’ motive was to “violate a person, ethnic group or other such group of persons because of race, color, national or ethnic origin, religion, sexual orientation or other similar circumstance” can result in a penalty enhancement. The writing does not limit penalty enhancement to apply only to certain minority groups but can also apply to majority groups and other groups than the ones particularly mentioned, such as transgender identity.

The penalty enhancement rule was later expanded to apply to crimes motivated by a hostility toward a person’s sexual orientation. The following reasoning was included in Proposition 2001/02: 59, which led to the extension: “Crimes of this nature can be said to have its sting directed against the basic notion of human equality and ultimately against the entire set of values that supports democracy” (Prop. 2001/02: 5911). A study on Swedish university students’ support to penalty enhancement for hate crimes found that only one third agreed that such crimes should be punished harder than parallel crimes without a hate motive (Mellgren, 2016).

In Sweden, almost 7,000 hate crimes with different motives are reported to the police each year (National Crime Preventive Council [BRÅ], 2017). However, the tendency to report hate-based incidents appears to be lower than for parallel crimes without a hate motive (Andersson & Mellgren, 2015). Only a small percentage of reported hate crimes are violent and brutal events. According to statistics provided by the BRÅ, the most commonly reported crime type is assault and molestation (43%), followed by various types of violent crimes (15%), defamation, hate speech, vandalism, graffiti, illegal discrimination, and other offenses. The most common motive is xenophobic or racist (69%). Generally, most crimes are committed in a public place and the offender is unknown to the victim. This pattern contributes to the low proportion of crimes being prosecuted. In only 5% of all reported hate crimes, in 2014, a perpetrator could be tied to the crime.

The Consequences of Hate

As stated in the beginning of this article, one of the aspects of hate crime that set them apart from other types of crime is that hate crimes supposedly do not only affect the individual victim but entire communities and society as a whole (Bell & Perry, 2015; Craig, 1999; Herek, Gillis, & Cogan, 1999; Perry & Alvi, 2012). Iganski discusses the harms of hate as waves: The harms of hate spread from the initial victim to the victim’s community, to other targeted groups, and to society as a whole (Iganski, 2001). This is where hate crimes become message crimes (Iganski, 2002), that is, crimes that send to entire communities the message that they are not secure. The mechanism delivering the effects to communities has, from a collective identity perspective, been proposed to be the threat to self that emanates from sharing identity with other victims. The effects of individual hate crimes on communities is an empirically underresearched area but the few studies that have been conducted to date agree that effects on group members involve increased feelings of vulnerability, anger, and reduced belief in the benevolence of the world (Bell & Perry, 2015; Lannert, 2014).

The “message-effect” may take different forms for individuals who identify with different victim groups. Members of minority groups, being exposed to information about hate crime, have been found to suffer from emotional distress and change their behavior after hearing about others being attacked (Noelle, 2002; Perry & Alvi, 2012). The homophobic murder of Matthew Shepherd, a young American homosexual man, in 1998, received a lot of attention. In a qualitative study of the effects of the Matthew Shepherd murder on homo- and bisexual people, Noelle (2002) found that the event had a traumatization effect on people who identified as belonging to the same minority group as the victim. Related to the “message-effect” is the hate crime victims’ interchangeability (McDevitt, Balboni, Garcia, & Gu, 2001), and this contributes to a sense of loss of control and elevated worry in entire groups.

Existing research about the consequences of hate crime for the individual comes to one main conclusion: Hate crimes do seem to hurt more. This conclusion is, however, supported mainly by studies that treat hate crime victims as one collective group and compare this group of victims with victims of other crimes, and by studies focusing on samples consisting solely of hate crime victims. Studies that are able to distinguish between identity groups and, at the same time, include victims of non–hate-motivated parallel crimes are scarce. Also, studies on the consequences of hate crime vary in respects that make it difficult to compare results between studies and between victim groups: what consequences respondents are asked about, and the samples used. Studies have asked about diverse consequences such as being frightened, or scared (Garofalo, 1997); anxiety; sleep difficulties (Ehrlich, 1992); or anger (Herek, Gillis, Cogan, & Glunt, 1997). Barnes and Ephross (1994) performed focus group interviews with 59 individuals with varying ethnicities who had been victims of various types of physical and nonphysical bias-motivated crimes. Reported consequences included sadness, anger, and behavioral changes such as moving from the neighborhood and increased personal and property security. McDevitt et al. (2001) used the Horowitz Impact of Events Scale to measure, for example, difficulties overcoming the incident, nervousness, and depression. This scale was also used by Ehrlich et al. (1994) in their study on the effects of ethnoviolence. Although research on consequences for hate crime victims has been centered on the United States, Canada, and the United Kingdom, there are studies from other European countries such as Latvia (Dzelme, 2008) and Sweden (Tiby, 2001).

Garcia et al. (2002) employed a random sample of assault victims of both bias and nonbias crimes in Boston. They were unable to demonstrate differences in behavioral reactions between victim groups but bias victims reported higher levels of psychological distress and were less likely to feel safe following an attack. Unfortunately, the response rate was very low, and the final sample only included a total of 134 victims of which about two thirds were victims of hate crime. In an ethnographic study using observational and interview data, Funnell (2015) showed that victims of hate crimes, because they worried about people close to them, began isolating themselves and their significant others. As such, hate crimes alter victims’ everyday lives and limit their access to public places. Most studies have focused on consequences of homophobic and racist crimes but Garland and Hodkinson (2014) found similar experienced consequences among victims of crimes that were motivated by belonging to a subculture and those belonging to the groups generally covered under hate crime laws. More specifically, they analyzed data from 21 interviews with persons identifying as Goths.

Many of the aforementioned studies use small, nonrandom samples but, more recently, hate crime studies have started to use larger random samples that allow for comparisons between groups (e.g., Iganski, 2008; Iganski, & Lagou, 2015; Smith, Lader, Hoare, & Lau, 2012; Williams & Tregidga, 2014). Herek et al. (1999) analyzed survey data from approximately 2,300 gay men, lesbians, and bisexuals in Sacramento, and compared the consequences between those who reported a hate-motivated victimization experience with those who had experiences of crimes without such motives. Hate crime victims reported significantly more anger, symptoms of depression, and anxiety compared with nonbias victims. Smith et al. (2012) analyzed two waves of the British Crime Survey (2009/2010 and 2010/2011), and found that hate crime victims were more likely than victims of parallel crimes without a hate motive to report that they had been emotionally affected by the crime. Williams and Tregidga (2013, 2014) compared consequences of hate crime across seven victim groups (disability, race/ethnicity, religion/belief, sexual orientation, transgender status/gender identity, age, gender) based on survey data from the All Wales Hate Crime Project, one of the largest studies on hate crime in the United Kingdom. Findings revealed that impacts of victimization were not homogeneous across victim groups. They did, however, not compare the consequences of hate crime victims with victims of parallel crimes without a hate motive. Overall, those who suffered several victimizations by the same offender and those who were victims of disability- and transgender-related hate crimes were found to be the most psychologically and physically affected. These results were in line with a growing number of studies showing that transgendered people are one of society’s most vulnerable groups (Jauk, 2013; Moran & Sharpe, 2004; Namaste, 2000). Examples of other factors that were related to reported consequences of a hate-motivated victimization were where the crime occurred (proximity to home), whether the victim was alone, whether the crime was violent or not, and the victim’s sense of belonging in the neighborhood. Victims of hate crimes are often repeatedly victimized, which further strengthens the effects (Andersson & Mellgren, 2016a, Lannert, 2014; NUS, 2011, 2012; Peri, Ben-Shakhar, Orr, & Shalev, 2000).

Trying to expand on previous research that had mainly focused on consequences for whole groups and between groups, Iganski and Lagou (2015) studied within-group differences in a group of victims of racist-motivated hate crimes using data from the Crime Survey for England and Wales. They analyzed the impacts of 441 incidents reported by 398 respondents as being racially motivated. While the results confirmed previous research showing that hate crime victims, in general, do suffer more than parallel victims of crime, they also found, by analogy of those by Funnell (2015), consequences to vary within the group of victims of hate crime. Funnell concludes that “ . . . there is no shared experience of victimization, but rather . . . it impacts upon each individual differently” (Funnell, 2015, p. 81). Within the group of racially motivated hate crime victims, Iganski and Lagou (2015) could show that men, Whites, and young persons were more likely to report externalized reactions such as anger. Anger was also more commonly expressed among property crime victims than victims of personal crimes. Almost the opposite pattern was found for internalized reactions such as anxiety and fear. Women, non-White, and middle aged were more likely to experience such consequences following an attack. The likelihood of experiencing both types of impacts was greatest for women, younger persons, non-White, persons with a lower socioeconomic status, and repeat victims. One possible problem associated with analyzing one particular motive in a larger survey and comparing that group of victims with all others is that all other possible motives, such as misogyny, homophobia, and so forth are not being considered. The “nonhate victims” group in such studies presumably includes victims of hate crimes with other motives, which may affect the results.

Funnell (2015) raises an important question that emanates from research on the differential impact of victimization across victim identity groups: whether or not all hate crime offenders shall be punished more severely or whether all incidents must be individually tried based on the consequences for the victim. Iganski and Lagou (2015) provide a similar discussion. Based on their findings that impacts vary within the group of victims of racially motivated crimes, they question the basis for sentencing enhancement applied equally for all hate crimes. If sentencing enhancement is motivated by the greater harm caused by hate crimes but studies find great variability in victims of the same offenses and everyone does not experience greater harm, then where does that leave us? Iganski and Lagou (2015) propose the solution that “. . . the victim impact to be demonstrated on a case-by-case basis to the satisfaction of the court that the impact of the hate crime on the victim is greater than would be the case in a parallel crime” (p. 14). They do, however, recognize that this solution may cause further harm to the victim, by having to relive in detail the event and its consequences to motivate a penalty enhancement. They also point out that because hate crimes are still assumed to spread to others than the direct victim, the argument that hate crimes hurt more is still valid.

In a theoretical paper, Chakraborti and Garland (2012) discuss that vulnerability and “difference” of a victim should be the focus of the hate crime debate and not groups and identity strands. By focusing on specific groups, they argue, some victims who are especially vulnerable to attack are forgotten, such as those belonging to a subculture, or homeless people, people targeted for being different from the norm and who, perhaps, because of being left out of policy discussions on victimization, are the most vulnerable in society.

The Present Study

Against this background, the present study seeks to expand previous knowledge in mainly two ways. This is the first study on the consequences of hate crime in a Swedish context that compares the reported victim impacts following hate crimes with different motives with parallel crimes without a hate motive. Because samples used in hate crime studies often focus on one identity group and one motive, few international studies, to date, have been able to analyze consequences for several groups using the same methodology and asking each group about the same consequences to compare for whom hate crimes hurt more. As such, this study contributes with important knowledge about crime victims in a Swedish as well as an international context. Second, we include a gender-based category, which is currently not included in Swedish hate crime legislation. Gender-based hate crimes, also called misogynistic crimes, are currently included under hate crime statutes in some U.S. states and in Canada, and some police departments in the United Kingdom register gender-based crimes as hate crimes. Although Sweden has well-developed legislation and policy against violence against women, there is a current lack of knowledge about the potential hate motives behind these crimes and the consequences for the victims. If victims of crimes motivated by misogyny are more likely to report consequences than parallel nonmisogynistic crimes, the exclusion of gender as a protected ground in Swedish hate crime legislation may be questioned.

Method

Data Collection

This study analyzed data that were collected through questionnaires distributed to students at Malmö University, Sweden, during 2013. The survey focused on students’ self-reported exposure to, and experiences of, hate crime, and included questions about fear of crime, perceived risk of victimization, and consequences of victimization. Questionnaires were available in Swedish and in English and distributed in classrooms. In total, 4,649 students were reached; of these, 2,853 answered the questionnaire, giving a response rate of 61% (for a further description of the study, see Andersson & Mellgren, 2016b).

The sample closely reflected the student population at Malmö University and the general student population in Sweden for percentage men and women, age, and national background (Swedish National Agency for Higher Education, 2008; Statistics Sweden, 2014). Compared with the general population in Malmö and in Sweden, there was a lack of age variation in the sample with half of the participants between the ages of 21 and 26 years, with a mean age of 25 years (median = 23 years). The distribution of those who described their gender as male or female was skewed with a large proportion of female students.² Forty percent were born in another country than Sweden, had at least one parent born in another country than Sweden, or belonged to a national minority³ (of these, 15 belonged to a national minority but were born in Sweden with Swedish-born parents, and 140 students both had an immigrant background and belonged to a national minority). The respondents who stated that they belonged to another minority were primarily African Swedish. The corresponding figure (definitions vary slightly) for Malmö was 43% (born in another country than Sweden or born in Sweden with both parents born in another country than Sweden) in 2014 according to statistics provided by the city of Malmö and for Sweden about 30% (Statistics Sweden, 2013).

A majority of the sample reported heterosexual as sexual orientation (89%), and a majority of students identified as Christians or atheists (74%). There were no significant gender differences in immigrant background. But there seems to be an overlap between gender identity, sexual orientation, and religion, with significantly more students who report having a transgender identity also reporting another religion/worldview than Christianity or Atheism and another sexual orientation than heterosexual. The number of students who reported having a transgender identity was small (n = 71). In total, 54.2% of the sample belonged to one or more minority groups. About 20% of the sample belonged to two or more minority groups.

Variables

Victimization experiences

This study departs from a victim perspective and all experiences are self-reported. Victimization experiences include both such acts that would be considered criminal according to the law and low-level incidents that would not qualify as a crime if reported to the police. Respondents were asked whether they had experienced any of the following acts in the 12 months prior to the survey: threat (bullying, threats, or harassment), minor assault (followed, chased, pushed, had your hair pulled, or hit with an open hand), sexual harassment (been exposed to unwanted physical contact that was carried out in an offending and or/unpleasant way, for example, touching, kissing, grabbing, or fondling), online threats (bullying, threats, or harassment on the Internet), assault (kicked, punched with a closed fist, or something else, like a weapon), sexual crimes (been forced to have sex against your will, raped, sexually abused, or subjected to similar acts), and robbery or burglary. Respondents were asked a series of follow-up questions.

First, for each type of act, all respondents were asked to indicate whether they perceived the incident to be founded on prejudice or hostility against them based on (any of) their (yes/no) gender, gender identity, origin/national background, religion, sexual orientation, disability. Next, those respondents who had suffered a hate-motivated crime were asked to indicate which motives they perceived to be involved in the act. Respondents could report several motives for each crime type. All reports rely on the victim’s perception of the act and the motive behind it. This means that there may be both overreporting and underreporting of incidents and motives because of different perceptions of what constitutes a crime or an act serious enough to report and what the motive was. Because victims may be reluctant to frame their experiences as hate motivated (Williams & Tregidga, 2013), neither the words hate nor crime were used anywhere. It should also be noted that these categories are generic and may mask the individual experiences by all different groups who are covered by, for example, national background/ethnicity or sexual orientation (Garland, Spalek, & Chakraborti, 2006; Sherry, 2010). The survey information allows for a breakdown of categories into specific national backgrounds, sexual orientations, and worldviews but such categorization does not hold for statistical analyses. The categories gender and gender identity were collapsed into one category because the categories can be difficult to separate. Ethnicity/national background and religion were collapsed into one category for the same reason. It can be difficult for the victim to determine whether they were attacked because they are simply perceived to be foreigners or because they are perceived to be Muslims.

Respondents were also asked to state how many times they had experienced the act. This variable was coded 0 if the respondent had only experienced the act once and 1 if two or more experiences.

Consequences of victimization

For each incident type, the respondents were asked to indicate whether they had experienced one or several of the following consequences (yes/no): anger; worry about future victimization; whether they had made adaptions to their everyday life by changing the way they dressed, their body language, or general behaviors; avoided certain places to avoid being victimized.

Ethical Considerations

This study has been approved by the regional ethical review board in Lund, Sweden (reference no. H15 2013/221) and was conducted in accordance with recommendations for safe and ethical research. Given the sensitive nature of the questions, several measures were taken to assure the integrity of the respondents. First, students were informed that the study would take place prior to administering the survey, so that students who did not wish to participate could choose not to attend the class at the time of the survey and did not feel pressured to take part in the study. To guarantee the safety and integrity of the respondents, we distributed five different versions of the survey. The questions asked were the same in every version of the survey, but they came in a different order, so that no one would know what question a particular student answered. Another advantage with this method is that it does not take longer to fill in the survey for those who has no experiences of victimization because the survey also contained questions directed only toward those students who had not been exposed to any crime.

Analysis

The main analyses in this study included only the following crime types: threats or harassment (n = 287), minor assault (n = 233), and sexual harassment (n = 525). The other crime types were excluded from further analysis because of the low rate of hate-motivated incidents. The analysis was performed in two steps. First, we investigated the prevalence of the different consequences of victimization by crime type and for hate-motivated crimes and crimes without a hate crime motive separately. Next, a series of separate logistic regression analyses were performed, which analyzed the likelihood for reporting each of the consequences for each incident depending on crime type, perceived motive, repeat victimization, gender, and age. Statistical significance was set at p < .05.

Results

Of the 1,032 reported incidents of threat, minor assault, and sexual harassment, 434 were motivated by at least one hate motive. Thirty-one percent were motivated by the victim’s gender or gender identity, 8.4% were motivated by the victim’s national background/ethnicity or religion, and 2.5% were motivated by the victim’s sexual orientation (see Table 1).

Table 1.

Frequencies of Type of Crime and Motive.

Offense/Motive	No Motive	Gender and Gender Identity	National Background or Religion	Sexual Orientation
Threat (n = 278)	52.5	23.7	19.4	4.3
Minor assault (n = 230)	67.0	22.6	9.1	1.3^a
Sexual harassment (n = 524)	56.9	38.7	2.3	2.1
Total % (N = 1,032)	57.9 (598)	31.1 (321)	8.4 (87)	2.5 (26)

Missing information about type of motive in eight cases.

Less than five cases.

The motive gender or gender identity was the most common among all included crime types. Four out of 10 of the reported cases of sexual harassment were experienced by the victim to be motivated by the offender’s hostility toward his or her gender, and in a majority of cases, the victim was a woman. About one in five incidents of threat was motivated by the victim’s national background or religion and one in 10 assaults. A person’s sexual orientation was the motive only in a small number of incidents. The low frequency of this type of bias crimes was surprising given that homophobic hate crimes have been the focus of much Swedish research and public debate on hate crime. The number of hate crimes motivated by a person’s sexual orientation does, however, seem to decrease when the source is crimes reported to the police (BRÅ, 2017). One explanation may be that the general tolerance for different sexual orientations in society has increased.

The impact of victimization for victims of hate crime, considered as one group regardless of motive, and victims of parallel crimes without a perceived hate motive is presented per crime type in Table 2. Victims of hate-motivated offenses as one group significantly more often reported having experienced any of the consequences compared with victims of crimes without a perceived hate motive.

Table 2.

Cross Tabulation of Type of Crime and Consequences of Crime.

Consequence	Threat (n = 287)		Minor Assault (n = 233)		Sexual Harassment (n = 525)
Consequence	Not Hate Crime	Hate Crime	Not Hate Crime	Hate Crime	Not Hate Crime	Hate Crime
No consequence	23.9	13.8*	45.3	16.4*	45.3	25.3*
Worry	31.7	37.7 ns	23.6	42.5*	16.4	27.6*
Anger	57.0	65.4 ns	34.5	58.9*	40.4	61.1*
Avoid places	12.0	24.6*	15.5	35.6*	8.4	13.6*
Behavior adaptations	9.2	14.6 ns	8.8	6.8 ns	10.1	14.0 ns

Significant differences between those who perceived that the crime was motivated by any one of the “hate” motives compared with crimes with otherwise motives.

Victims of hate-motivated threat reported, to a significantly lesser degree, that they suffered no consequence following an attack and significantly more often reported that they avoided places to avoid future victimization. Victims of hate-motivated minor assault were significantly more likely to report worry, anger, and avoiding places as a consequence of the victimization compared with victims of minor assault without a hate motive. Victims of hate-motivated sexual harassment more often reported worry, anger, and avoiding places following an incident. Relatively few of the victims, regardless of the motive, reported that they made behavioral adaptations, such as changing body language or clothes, following an incident. This far, the argument that hate crimes hurt more than parallel crimes without a hate motive when treating all victims of hate crimes as one group, and which has been established by a body of previous research (e.g., Iganski & Lagou, 2015), was confirmed in this study.

Next, we wanted to investigate whether the argument would still hold if we compared the likelihood of experiencing the different consequences for each of the hate victim groups: gender or gender identity, national background or religion, sexual orientation, with the group of victims of non–hate-motivated incidents. We fitted four separate logistic regression models, one for each consequence to estimate the “effect” of the different motives. “No motive” was used as reference category. Each reported crime was included, which means that one individual can appear up to three times if the person has been the victim of all three crimes: threat, minor assault, and sexual harassment. Forty-four cases were excluded from the analysis. Of these, 33 cases had missing information on the outcome variables, 13 were excluded because the motive was disability and the group was too small for statistical analysis. Information on gender was missing in three cases. Eighty-six respondents did not state the number of incidents. These were replaced with one victimization experience (coded 0), which was the mode of the sample. The final number of cases left for final analysis was 998.

Results from the logistic regressions are presented in Tables 3 to 7. Overall, comparisons between groups revealed differences between subgroups of hate crime victims.

Table 3.

Logistic Regression—Dependent Variable: No Consequences.

Independent Variables	B Coefficient	SE	p	Odds Ratio
Constant	0.770	0.471	.102	2.160
Gender
Male
Female	−1.025	0.204	.001	0.359
Other	2.087	0.523	.001	0.124
Age	−0.039	0.017	.018	0.962
Crime type
Threat
Minor assault	0.784	0.223	.001	2.191
Sexual harassment	1.115	0.203	.001	3.048
Repeat victimization	−0.350	0.149	.018	0.705
Motives
No motive
Gender identity and expression	−0.735	0.170	.001	0.479
National background/religion	−0.802	0.318	.012	0.448
Sexual orientation	−1.544	0.756	.041	0.214

Table 4.

Logistic Regression—Dependent Variable: Worry.

Independent Variables	B Coefficient	SE	p	Odds Ratio
Constant	−2.506	0.445	.001	0.082
Gender
Male
Female	0.618	0.233	.008	1.856
Other	0.909	0.393	.021	2.481
Age	0.037	0.014	.009	1.037
Crime type
Threat
Minor assault	−0.088	0.205	.667	0.916
Sexual harassment	−0.739	0.182	.001	0.478
Repeat victimization	0.386	0.158	.014	1.471
Motives
No motive
Gender identity and expression	0.423	0.167	.011	1.526
National background/religion	0.225	0.265	.397	1.252
Sexual orientation	1.076	0.430	.012	2.933

Table 5.

Logistic Regression—Dependent Variable: Anger.

Independent Variables	B Coefficient	SE	p	Odds Ratio
Constant	−0.844	0.389	.030	0.430
Gender
Male
Female	0.686	0.196	.001	1.985
Other	1.233	0.382	.001	3.430
Age	0.016	0.013	.224	1.016
Crime type
Threat
Minor assault	−0.691	0.194	.001	0.501
Sexual harassment	−0.584	0.168	.001	0.558
Repeat victimization	0.146	0.137	.286	1.157
Motives
No motive
Gender identity and expression	0.621	0.150	.001	1.861
National background/religion	0.540	0.252	.032	1.716
Sexual orientation	1.089	0.487	.025	2.971

Table 6.

Logistic Regression—Dependent Variable: Avoid Places.

Independent Variables	B Coefficient	SE	p	Odds Ratio
Constant	−0.2275	0.506	.001	0.103
Gender
Male
Female	0.281	0.277	.310	1.325
Other	0.370	0.470	.310	1.325
Age	−0.002	0.016	.891	0.998
Crime type
Threat
Minor assault	0.427	0.238	.073	1.524
Sexual harassment	−0.629	0.231	.006	0.533
Repeat victimization	0.272	0.194	.161	1.313
Motives
No motive
Gender identity and expression	0.705	0.208	.001	2.025
National background/religion	0.623	0.307	.008	3.495
Sexual orientation	1.251	0.471	.008	3.495

Table 7.

Logistic Regression—Dependent Variable: Changing Behavior.

Independent Variables	B Coefficient	SE	p	Odds Ratio
Constant	−4.128	0.592	.247	1.831
Gender
Male
Female	0.529	0.363	.145	1.697
Other	1.764	0.487	.001	5.834
Age	0.047	0.016	.004	1.048
Crime type
Threat
Minor assault	−0.328	0.324	.311	0.720
Sexual harassment	0.020	0.252	.937	1.020
Repeat victimization	0.529	0.229	.021	1.698
Motives
No motive
Gender identity and expression	0.080	0.233	.731	1.083
National background and religion	−0.015	0.389	.968	0.985
Sexual orientation	0.605	0.523	.247	1.831

No Consequences

All three motive categories were significantly less likely than the category “no motive” to report no consequences of threat and harassment. As expected, incidents where the victim was a woman or a person reporting another gender identity than male or female were significantly less likely to result in no consequences than incidents where the victim was male (Table 3). Younger persons were also less likely than older persons to report no consequences. Looking at the crime variables revealed that being exposed to sexual harassment or minor assault significantly increased the likelihood of reporting no consequences compared with being exposed to threats. Those who were victimized repeatedly were less likely than those who were only victimized once to report no consequences, which seems to support the idea that hate crime victimization has a cumulative effect (Bowling, 1998).

Worry About Future Victimization

Of the three possible motives, those having experienced an incident that was motivated by their gender or gender identity, or their sexual orientation, were significantly more likely to report feeling worried compared with those who did not perceive the incident to be hate motivated. Women and those not identifying as male or female were significantly more likely to worry about future victimization following an incident than men, as were those who were older (Table 4). Reporting worry about future victimization was significantly less likely among those who suffered sexual harassment compared with those who had been exposed to threats. Repeat victimization emerged as a significant predictor of reporting worry.

Anger

The results from the present study indicate that, when looking at specific motives, hate crime victims are significantly more likely to feel wronged by an attack. Women and those not identifying as either male or female were significantly more likely to report feeling angry after an incident (Table 5). Compared with threats, minor assault and sexual harassment crimes were less likely to result in anger. Being a repeat victim or not did not influence the likelihood of reporting anger. All hate crime victims were significantly more likely to report anger following an incident than victims of parallel crimes without a hate motive. Iganski and Lagou (2015) describe anger as “ . . . a natural response, a basic human reaction when a person feels wronged, or when they have been threatened or attacked” (p. 10).

Avoiding Places

Those who experienced an incident to be motivated by a hostility toward them because of their gender or gender identity, national background or religion, or sexual orientation were more likely to worry about future victimization than those who experienced the incident to be motivated by another factor. The analysis found no differences in the tendency to report avoiding places based on gender or gender identity, or age (Table 6). It was less likely for respondents to report worry following sexual harassment compared with threats but there were no differences between threats and minor assault.

Changing Behavior

There were no differences between any of the motive categories compared with the nonmotive group, in the likelihood of reporting changing behavior following an incident. Overall, few reported that they had made behavioral changes such as changing clothes or body language (Table 7). Those who did not identify as male of female were more likely to make behavioral changes after being victimized.

Discussion

The results from this study have provided partial support for the argument that hate crimes hurt more. Analyzed as one victim group, victims of hate crime more often reported feeling any of the consequences following a crime compared with victims of parallel non–hate-motivated crimes, except for changing behavior. And, overall victims of threat more often reported consequences compared with victims of sexual harassment and minor assault. But this study also partly questions the assumption that this argument is valid for all groups as the results indicate that not all victims experience the same consequences. Similar to Iganski and Lagou (2015), who found within-group variability in a group of victims of racially motivated attacks, the between-group variability suggests that all types of hate crimes do not hurt all types of hate crime victims more than parallel crimes without a hate motive. This is also in line with Williams and Tregidga (2014). Taken together, empirical evidence suggests that there is not only within- and between-group variability but also crime-type variability. Some hate crimes do hurt some more, but not all.

The assumption that hate crimes hurt more, on an individual, group, and societal level, is used as an argument for harsher sentences for hate crime offenders (Iganski & Lagou, 2015). Analyzed as one group of victims, victims of hate crimes really do seem to hurt more than the non–hate victim group but this study does not support that this is true for all victims and for all crime types. Maybe, it does not hurt equally to be attacked for all parts of an identity. Maybe then, it is time for hate crime scholars to move away from analyzing categories of victims to analyzing victims in terms of vulnerability (Chakraborti & Garland, 2012). Compared with victims of nonhate incidents, the hate crime victims, regardless of motive category, were less likely to report that they had not experienced any consequences following the event and more likely to experience anger, and to avoid places. Incidents where the motive was sexual orientation, gender or gender identity was significantly more likely to lead to worry about future victimization, whereas there were no differences between the groups regarding changing behavior. Overall, those who reported the motive to be their sexual orientation were more likely than any of the other categories to report any of the consequences. This suggests that although the number of reported hate crimes against persons with another sexual orientation than heterosexual seems to decrease in Sweden, and the number of reported crimes was low in this study; this group seems to be the most affected by victimization. A decrease in number should, therefore, not be mistaken for a decrease in impact or magnitude of problem.

Garland (2012) and Garland and Hodkinson (2014) suggest that mediating and moderating factors other than group membership, such as risk and vulnerability, may also be important to understand victimization experiences. For example, a study on lesbians and posttraumatic stress disorder (PTSD) symptoms (Szymanski & Balsam, 2011) found that PTSD symptoms following a hate crime were partially mediated by self-esteem, suggesting that not all victims of homophobic hate crimes react the same because of buffering personal characteristics. Characteristics other than group membership are often left unmeasured in studies about hate crime; and, based on the findings from this study, we encourage larger surveys of victims of both hate crimes and parallel crimes without a hate motive to also measure mediating factors related to vulnerability.

This study also highlights one motive that is currently not covered in the Swedish penalty enhancement rule: gender. Gender-based crimes are, however, covered under hate crime statutes in other countries. For example, Canada and some states in the United States include gender as a protected category. And, scholars who argue for the general inclusion of gender under hate crime statutes make the claim that these crimes, as well as other bias crimes, hurt more (Carney, 2001) and that many hate crimes are gendered (Mason-Bish, 2014). However, the gendered aspect of hate crimes is often overlooked because gender is not covered by hate crime laws (Mason-Bish, 2014). Victims in this study who experienced the incident to be motivated by the offender’s hostility or prejudice against their gender or gender identity, an overwhelming majority being women, were more likely to experience anger and worry and to make changes to their everyday life by avoiding places than those who did not make the same interpretation of the same type of event. Sexual harassment against women has gained a lot of media attention during 2016, both in Sweden and internationally, following the incidents in Cologne, Germany, on New Year’s Eve (2015), when groups of men harassed a large number of women, and also in 2017 after thousands of women worldwide shared their experiences of sexual assault under the hashtag metoo in different social media. In the wake of these events, sexual harassment has been put on the agenda both in Sweden and internationally. This study finds that experiencing that you are attacked based on your gender can result in similar consequences as being attacked based on other hate motives and, perhaps, now it is time to consider including gender as a motive in Swedish hate crime legislation? The main arguments to treat gender-based crimes as hate crimes are that they too are a result of wider social structures that allow for the subordination of women (Weisburd & Levin, 1994), and gender-based acts are used as a means to control and keep women subordinate (Isaacs, 2001; McPhail, 2002). To include gender under Swedish hate crime statutes and viewing misogyny as an aggravating circumstance were suggested already in 1995 by a commission that investigated violence against women (SOU, 1995). The acts that were suggested to be treated as misogynistic hate crimes were those acts that “ . . . offend another person based on gender and has its foundation in the notion of women’s lower value in relation to men” (SOU, 1995, p. 426).

This study has some limitations. The student sample employed includes both victims and nonvictims, with different sociodemographic backgrounds, representing many different identities. It may be that where the study succeeds in diversity in the sample, the lack of specificity leads to an underrepresentation of the most visible victims, and probably most at-risk persons (Herek et al., 1999). For example, many studies on LGBTQI recruit participants from interest organizations or forums especially devoted to LGBTQI issues, probably sampling a higher degree of visible persons engaged more openly in LGBTQI issues. Other minorities such as the Roma are underrepresented in higher education and, thus, their experiences are not represented in this study. The diversity of the sample also leads to a small number of victims in each group, which does not make a within-group analysis reliable. We do, however, believe that the diversity of the sample is primarily a strength. People have multiple identities, regardless of study design, and a study that only asks for one motive, for example, racism or homophobia, is likely to be influenced by the many other identities masked by what is asked for (or not) in the survey.

Another limitation is with regard to the consequences respondents were asked to report. A qualitative interview study would be able to explore the width of experienced consequences better, but would not allow for comparing the many different groups of victims. There may be significant differences in reactions to crimes between different victim groups that we have failed to identify. Also, we did not consider the extent of the impact, that is, how much anger someone felt, but only that the victim felt angry. Iganski and Lagou (2015) found that although there were differences between victims of racially motivated crime and other victims in frequency of reported consequences, this gap widened further when considering the extent of the impact. Thus, we may underestimate differences between groups. There may be another explanation to the small differences between victim groups. Miers (1989), Fattah (2010), and, most recently, Williams and Tregidga (2014) argue that only using a survey to study impact on victims may not be enough. Not only does it risk not capturing all impacts because of the often times fixed alternatives of consequences, as in this study, but also it may be that victims of hate crime do not report consequences because they have gotten too used to them and, thus, have neutralized their impact.

Taken together though, in light of the most recent research in the area cited in this article, together with the empirical results provided here, there seems to be grounds for questioning that hate crimes hurt the individual victim more. A formulation more in line with the results would be that hate crimes hurt some victims of some crimes more in some ways.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Caroline Mellgren is associate professor in criminology with the Department of Criminology, Malmö University. Her research interests include victimization, hate crime, and fear of crime. Her work has been published in journals such as International Review of Victimology, Journal of Scandinavian Studies in Criminology and Crime Prevention, and she has an article in press for Women and Criminal Justice.

Mika Andersson is currently a PhD student with the Department of Criminology, Malmö University. The subject of her dissertation is experiences and exposure to hate crime, which is based on the same material as used in the present study submitted for review.

Anna-Karin Ivert is a postdoc researcher with the Department of Criminology, Malmö University. Her research interests include fear of crime, immigrants and mental health, and youth and crime. She has published in journals such as Ethnicity and Health, Child Indicators Research, and Social Psychiatry and Psychiatric Epidemiology.

References

Andersson

Mellgren

(2015). Anmälningsbenägenhet vid utsatthet för hatbrott [Tendency to report hate crime victimization]. Socialvetenskaplig Tidskrift, 3-4, 283-301.

Andersson

Mellgren

(2016a). Consequences of bias-motivated victimization among Swedish university students with an immigrant or minority background. Irish Journal of Sociology, 24, 226-250.

Andersson

Mellgren

(2016b). Studenters utsatthet och upplevelser av hatbrott [Students experiences and exposure to hate crime] (FoU-rapport 2016: 1). Malmö, Sweden: Malmö University.

Avison

W. R.

Turner

J. R.

(1988). Stressful life events and depressive symptoms: Disaggregating the effects of acute stressors and chronic strains. Journal of Health and Social Behavior, 29, 253-264.

Barnes

Ephross

P. H.

(1994). The impact of hate violence on victims: Emotional and behavioral responses to attacks. Social Work, 39, 247-251.

Bell

J. G.

Perry

(2015). Outside looking in: The community impacts of anti-lesbian, gay, and bisexual hate crime. Journal of Homosexuality, 62, 98-120.

Bowling

(1998). Violent racism: Victimization, policing and social context. Oxford, UK: Oxford University Press.

Carney

K.M.

(2001). Rape: The paradigmatic hate crime. St. Johns Law Review, 75(2), 315-355.

Chakraborti

Garland

(2012). Reconceptualizing hate crime victimization through the lens of vulnerability and “difference.” Theoretical Criminology, 16, 499-514.

10.

Craig-Henderson

Sloan

L. R.

(2003). After the hate: Helping psychologists help victims of racist hate crime. Clinical Psychology: Science and Practice, 10, 481-490.

11.

Craig

K. M.

(1999). Retaliation, fear, or rage: An investigation of African American and white reactions to racist hate crimes. Journal of Interpersonal Violence, 14, 138-151.

12.

Dzelme

(2008). Psychological effects of hate crime—Individual experience and impact on community. Latvian Centre for Human Rights. Retrieved from http://cilvektiesibas.org.lv/site/attachments/30/01/2012/Naida_noziegums_ENG_cietusajiem_Internetam.pdf

13.

Ehrlich

H. J.

(1992). The ecology of antigay violence. In Herek

G. M.

Berrill

(Eds.), Hate crimes: Confronting violence against lesbians and gay men (pp. 105-122). Newbury Park, CA: Sage.

14.

Ehrlich

H.J.

Larcom

B.E.K.

Purvis

R.D.

(1994). The traumatic effects of ethnoviolence. Towson, MD: Prejudice Institute, Center for the Applied Study of Ethnoviolence.

15.

Fattah

(2010). The evolution of a young, promising discipline: sixty years of victimology, a retrospective and prospective look. In Shoham

S.G

Knepper

Rett

(Eds.), International handbook of victimology (pp. 43-94). Boca Raton, FL: CRC Press.

16.

Funnell

(2013). Racism by stealth: The construction of racist hate crimes. Academic dissertation, Cardiff University. Retrieved from http://eprints.uwe.ac.uk/24021

17.

Funnell

(2015). Racist hate crime and the mortified self: An ethnographic study of the impact of victimization. International Review of Victimology, 21, 71-83.

18.

Garcia

McDevitt

Balboni

(2002). Psychological and behavioral effects of bias and non-bias motivated assault. Washington, DC: National Institute of Justice, U.S. Department of Justice.

19.

Garland

Spalek

Chakraborti

(2006). Hearing lost voices: Issues in researching: “Hidden” minority ethnic communities. British Journal of Criminology, 46, 423-437.

20.

Garland

(2012). Difficulties in defining hate crime victimisation. International Review of Victimology, 18(1), 25-37.

21.

Garland

J. M.

Hodkinson

(2014). “F**king Freak! What the hell do you think you look like?” Experiences of Targeted Victimisation Among Goths and Developing Notions of Hate Crime. The British Journal of Criminology, 54(4), 613-631.

22.

Garofalo

(1997). Hate crime victimization in the United States. In Davis

R. C.

Lurigio

A. J.

Skogan

W. G.

(Eds.), Victims of crime (pp. 134-145). Thousand Oaks, CA: Sage.

23.

Herek

G. M.

Cogan

J. C.

Gillis

J. R.

Glunt

E. K.

(1997). Correlates of internalized homophobia in a community sample of lesbian and gay men. Journal of the Gay and Lesbian Medical Association, 2, 17–25.

24.

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, 19-215.

25.

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(6), 945-951.

26.

Iganski

(2001). Hate crimes hurt more. American Behavioral Scientist, 45, 626-638.

27.

Iganski

(2002). The hate debate: Should hate be punished as a crime? London, England: Profile Books.

28.

Iganski

(2008). Hate crime and the city. Bristol, UK: Policy Press.

29.

Iganski

Lagou

(2015). Hate crimes hurt some more than others: Implications for the just sentencing of offenders. Journal of Interpersonal Violence, 30, 1696-1718.

30.

Isaacs

(2001). Domestic violence and hate crimes: Acknowledging two levels of responsibility. Criminal Justice Ethics, 20, 31-43.

31.

Jauk

(2013). Gender violence revisited: Lessons from violent victimization of transgender identified individuals. Sexualities, 16(7), 80-825.

32.

Lannert

B. K.

(2014). Traumatogenic processes and pathways to mental health outcomes for sexual minorities exposed to bias crime information. Trauma, Violence & Abuse, 16(3), 1-8.

33.

Levin

McDevitt

(2008). Hate crimes. In Kurtz (Ed.), Encyclopedia of violence, peace, and conflict (pp. 89-102). San Diego, CA: Academic Press.

34.

Mason-Bish

(2014). We need to talk about women: Examining the place of gender in hate crime policy. In Chakraborti

Garland

(Eds.), Responding to hate crime: The case for connecting policy and research (pp. 169-182). Bristol, UK: Policy Press.

35.

McDevitt

Balboni

Garcia

(2001). Consequences for victims: A comparison of bias and non-bias motivated assaults. American Behavioral Scientist, 45, 697-713.

36.

Mellgren

(2016). University students’ support to penalty enhancement. Journal of Scandinavian Studies in Criminology and Crime Prevention, 17, 185-202.

37.

Miers

(1989). Positivust victimology: A critique. International Review of Victimology, 1(1), 3-22.

38.

Moran

L. J.

Sharpe

A. N.

(2004). Violence, identity and policing: The case of violence against transgender people. Criminology & Criminal Justice, 4, 395-417.

39.

Namaste

V. K.

(2000). Invisible lives: The erasure of transsexual and transgendered people. Chicago, IL: The University of Chicago Press.

40.

National Crime Preventive Council. (2017). Hatbrott [Hate crime]. Stockholm, Sweden: Brottsförebyggande rådet.

41.

National Union of Students. (2011). No place for hate crimes and incidents in further and higher education: Disability. London, England: Author.

42.

National Union of Students. (2012). No place for hate crimes and incidents in further and higher education: Race and ethnicity. London, England: Author.

43.

Noelle

(2002). The ripple effect of the Matthew Shepard murder: Impact on the assumptive worlds of members of the targeted group. American Behavioral Scientist, 46, 27-50.

44.

Peri

Ben-Shakhar

Orr

S. P.

Shalev

A. Y.

(2000). Psychophysiologic assessment of aversive conditioning in posttraumatic stress disorder. Biological Psychiatry, 47, 512-519.

45.

Perry

(2001). In the name of hate: Understanding hate crimes. New York, NY: Routledge.

46.

Perry

Alvi

(2012). “We are all vulnerable”: The in terrorem effects of hate crimes. International Review of Victimology, 18, 57-71.

47.

Sherry

(2010). Disability hate crimes: Does anyone really hate disabled people? Farnham, UK: Ashgate.

48.

Smith

Lader

Hoare

Lau

(2012). Hate crime, cyber security and the experience of crime among children: Findings from the 2010/11 British crime survey. London, UK: Home Office.

49.

SOU. (1995) 60 (Official reports of the Swedish government). Kvinnofrid (Safety of women). Socialdepartementet.

50.

Statistics Sweden (SCB). (2013). One in five children has a foreign background. Retrieved from http://www.scb.se/sv_/Hitta-statistik/Artiklar/Vart-femte-barn-har-utlandsk-bakgrund/

51.

Statistics Sweden (SCB). (2014). Swedish and foreign background among students and doctoral students in higher education 2014/15. Retreived from http://www.scb.se/Statistik/UF/UF0211/2014L15/UF0211_2014L15_SM_UF19SM1601.pdf

52.

Swedish National Agency for Higher Education (Högskoleverket). (2008). Kvinnor och män i högskolan [Women and men in higher education] (Report 2008:20 R). Stockholm: Högskoleverket.

53.

Szymanski

D. M.

Balsam

K. F.

(2011). Insidious trauma: Examining the relationship between heterosexism and lesbians’ PTSD symptoms. Traumatology, 17(2), 4-13.

54.

Thoits

P. A.

(1991). On merging identity theory and stress research. Social Psychology Quarterly, 54, 101-112.

55.

Tiby

(1999). Hatbrott? Homosexuella kvinnors och mäns berättelser om utsatthet för brott [Hate crime? Homosexual women and mens’experiences of victimization] Academic dissertation. Kriminologiska institutionen, Stockholm, Sweden.

56.

Tiby

(2001). Victimization and fear among lesbians and gay men in Stockholm. International Review of Victimology, 8, 217-243.

57.

Wallengren

Mellgren

(2015). The role of visibility for a minority’s exposure to (hate)crime and worry about crime—A study of the traveller community. International Review of Victimology, 21, 303-319.

58.

Weisburd

Levin

(1994). On the basis of sex. Stanford Law & Policy Review, 5, 21-47.

59.

Williams

M. L.

Tregidga

(2013). All Wales hate crime research project. Cardiff University. Retrieved from http://orca.cf.ac.uk/60690/13/Time%20for%20Justice-All%20Wales%20Hate%20Crime%20Project.pdf

60.

Williams

M. L.

Tregidga

(2014). Hate crime victimization in Wales: Psychological and physical impacts across seven hate crime victim-types. British Journal of Criminology, 54, 946-967.