Hate, Threats, and Violence. A Register Study of Persons Suspected of Hate Crime This article was downloaded by: [University of Stockholm] On: 19 December 2011, At: 05:12 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Journal of Scandinavian Studies in Criminology and Crime Prevention Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/scri20 Hate, Threats, and Violence. A Register Study of Persons Suspected of Hate Crime Lena Roxell a a Department of Criminology, Stockholm University, Stockholm, Sweden Available online: 02 Nov 2011 To cite this article: Lena Roxell (2011): Hate, Threats, and Violence. A Register Study of Persons Suspected of Hate Crime, Journal of Scandinavian Studies in Criminology and Crime Prevention, 12:2, 198-215 To link to this article: http://dx.doi.org/10.1080/14043858.2011.623391 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. http://www.tandfonline.com/loi/scri20 http://dx.doi.org/10.1080/14043858.2011.623391 http://www.tandfonline.com/page/terms-and-conditions Hate, Threats, and Violence. A Register Study of Persons Suspected of Hate Crime LENA ROXELL Department of Criminology, Stockholm University, Stockholm, Sweden Abstract Introduction ‘A man who was in the company of his boyfriend was gashed with a knife in the Pride Festival area in Tantolundenon Sunday night. The couple were attacked by three unknown assailants in their twenties’. (Dagens Nyheter 2008) ‘The fires on Parkvägen in Hagforson Friday night were started deliberately. And the motive may be racist. . . . No suspect has yet been identified and investigations are ongoing’ (Värmlands Folkblad 2009) The above media reports describe serious offences that could also represent examples of what are referred to as hate crimes. The hate crime concept is relatively new to both the criminal justice system and the research community. These offences represent what have been referred to as motive crimes, which involve violating people, or groups of people, by means of various criminal acts. People are violated because of their origin, religion, or sexual disposition— because they are perceived to be homosex- uals or Muslims, for example. In Sweden, if a person is subjected to a hate crime, this should lead to the application of the This article focuses on individuals suspected of hate crimes with xenophobic, Islamophobic, and homophobic motives. The objective is to fill a gap in the knowledge left by existing research, which has primarily focused on victims and definitional problems. This article’s genuine contribution to new research is the comparative perspective and the study of co- offending and specialization in offences for persons suspected for hate crimes. To find persons suspected for hate crimes, register data relating to hate- crime-motivated assault and unlawful threats/molestation offences from 2006 have been used. The study is based on a total of 1,910 offence reports together with information from the Registers of Suspected and Convicted Offenders for 558 persons suspected for hate crimes. Xenophobic hate crimes are over-represented in the material by comparison with homophobic and Islamophobic hate crimes. In the reports that have information about the relation between victim and perpetrators, it is more common for the perpetrators to be known than unknown to the victims. In cases where a suspected person has been identified, males are in a clear majority. Those suspected of homophobic hate crimes have the lowest mean age. Only a small number of offence reports include information on suspected co- offenders. Fifty-five per cent of the suspected people have prior registered convictions. It is very uncommon for them to be specialized in violent offences or unlawful threats/molestation, however. It is not possible to generalize the results to perpetrators of hate crimes, because 70% of the offence reports did not have information of suspected persons. KEY WORDS: Islamophobic and homophobic hate crimes, Perpe- trators, Register study, Xenopho- bic 198 Journal of Scandinavian Studies in Criminology and Crime Prevention ISSN 1404-3858 Vol. 12, pp. 198–215, 2011 http://dx.doi.org/10.1080/14043858.2011.623391 q 2011 Taylor & Francis D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 http://dx.doi.org/10.1080/14043858.2011.623391 http://dx.doi.org/10.1080/14043858.2011.623391 sentencing severity clause found at Chapter 29, Section 2, Paragraph 7 of the Swedish Penal Code (Tiby 2006:9). In the media texts presented above, the perpetrators are described as being unknown to both the victims and the police. The question is whether this constitutes a representative depiction of hate crimes and the perpe- trators who commit them. Or does it rather represent a media image of hate crimes always involving serious offences com- mitted by unknown perpetrators (Tiby 2010)? Previous research has shown that hate crimes can vary a great deal with regard to the types of crime that are involved (Brå 2007, 2008, 2009, 2010a). They can take the form of anything from offensive remarks and vandalism, to threats and lethal violence. It has also been noted that the perpetrators may be found in both the immediate environments of the victims and in more peripheral positions (Mason 2005a, 2005b; Tiby 2010). Existing research on hate crimes has primarily focused on victimization (see e.g. Tiby 1999; Brå 2007, 2008, 2009, 2010a; Herek 2009). The issue of victimization has been examined with a focus on different research questions and using a variety of methodologies. In Sweden, hate crime studies have for the most part been directed at charting the offences involved and the experiences of victims, while in the USA research has also focused on different explanatory models for homophobic hate crimes (Dahl 2005:21). Existing research has also emphasized the issue of defini- tional problems and has further included studies of hate crime legislation and of how these laws are applied in practice (Jenness 2001; Sullaway 2004). Some of these studies have adopted a more critical approach to the question of whether it is possible to legislate away feelings such as hatred (Sullaway 2004). Research that has focused on persons suspected of hate crimes is not as common, however, and in order to fill this gap in the knowledge base, the current article focuses on persons suspected of hate crimes. Another issue that has rarely been the subject of research studies involves the different motives underlying hate crimes (Mason 2005a: 843). For this reason the study will also focus on a comparison of xenophobic, Islamophobic, and homophobic hate crimes. The objective of the article is to present a comparative study of persons suspected of threatening behaviour, molestation, and assaults involving xenophobic, Islamopho- bic, and homophobic motives. The inten- tion is in part to examine whether there are differences between persons suspected of different types of hate crime, in part to provide a description of the criminal activities of these individuals. The com- parison presents background facts such as the age and sex of the perpetrators, as well as examining previous involvement in crime, co-offending, and specialization in the offence categories assault and threaten- ing behaviour/molestation. This article’s genuine contribution to new research is the comparative perspective and the study of co-offending and specialization in offences for persons suspected for hate crimes. The primary data in the current study are from the national Registers of Suspected and Convicted Offenders. ‘In statistics a person suspected of offences is defined as a person who at the time of committing the offence is 15 years or older and who by police or a prosecutor is considered to be suspected after a completed investigation’ ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 199 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 (Brå 2011a). ‘Statistics relating to persons convicted for criminal offences present those found guilty either by court (county court convictions) or by prosecutors (through prosecutor fines or waivers of prosecution)’ (Brå 2011b). To find persons suspected for hate crimes the offences reported to the police in 2006 that have been classified as hate crime have been used. In this case it means that they have been suspected by the police during the investigation. Problems associated with using this type of data linked to this study will be discussed in the methodology section. Previous research and theoretical perspectives Who commits crimes with hate crime motives? Are there any common denomi- nators among these people? Since this article will be focusing on individuals who have been suspected of committing hate crimes, it is reasonable to begin by looking at existing research in this area. In order to study these persons, they must be known to the victim and/or to the police. In the most recent survey of hate crimes published by the Swedish National Council for Crime Prevention (Brå 2010a), a suspect has been linked to the crime in just under one-third (32%) of reported hate crime incidents. It has also been noted that it is more common for crimes to be reported when the perpetrator is not known to the victim than when perpetrator and victim are acquainted (von Hofer 2006). This means that important information about certain types of perpetrator is missed in research based on officially registered crimes. By utilizing the data provided by victims in questionnaire surveys, for example, it is possible to obtain other types of infor- mation (Tiby 1999). Even in data of the latter kind, however, the problem remains that the victim must have some knowledge of the perpetrator. As was noted in the introduction, there is a perception that the perpetrators of hate crimes are usually unknown to the victims. Over recent years, however, this view has been called into question (Mason 2005a, 2005b; Tiby 2010). There may be vari- ations in how victims define the known and unknown. The way in which known and unknown persons are defined can also vary between different studies, which can itself affect the results. The same is true of differences in the categories employed by researchers and how they choose to present results. Results can also be affected by respondents who have been victimized repeatedly being asked to restrict them- selves to the most serious incident when they describe the perpetrator (Tiby 1999). This can lead to results which produce an image of an unknown perpetrator, rather than of someone closer to the victim who may, for example, be subjecting them to harassment. Instead of focusing on the neighbour or schoolmate who has repeat- edly been harassing the victim, the focus is instead directed at the unknown perpe- trator who has perhaps assaulted the victim on a single occasion. A register study from USA also shows that most of the perpe- trators were unknown to the victims (McDevitt et al. 2010). This finding is problematized by the study’s authors, who argue that it depends on what is meant by being previously unknown to the victim. Several of the perpetrators lived in the same area, went to the same school, or worked in the same organization as the victim. This would indicate that the victims know who the perpetrators are, even though they are ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention200 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 not directly acquainted with them, and that this is the reason they report that the perpetrator is not known to them. In other words, there are a lot of things to take into account in studies of the relations between victims and perpetrators. The annual report on hate crimes published by the Swedish National Council for Crime prevention includes a description of the persons suspected of hate crimes (Brå 2007, 2008, 2009, 2010a). Having only presented information on the suspects’ age and sex distributions in the first report, descriptions of the suspects have included more details in recent reports, including information on the nature of the offences involved and the suspects’ previous regis- tered criminality. The results show that males are suspected of hate crimes more often than females, and that middle-aged suspects are most common in relation to xenophobic/racist and religion-based hate crimes and least common in relation to homophobic and transphobic hate crimes.1 The most common offence types that these suspects have been registered for commit- ting as hate crimes are violent offences and unlawful threats/molestation. The offence distribution differs by comparison with that for registered hate crimes as a whole, which may be due to differences in the risk for detection associated with different types of crime. The results show further that the majority of those suspected of committing hate crimes have no prior convictions for criminal offences (ibid.). Questions have also been posed about perpetrators in questionnaire surveys focused on exposure to homophobic hate crime (e.g. Tiby 1999:153 ff.). For the most part the perpetrators were young males, most often more than one, 20% of the cases involved perpetrators with an over- seas background, and the perpetrators were intoxicated in connection with approximately half of the cases described. Among the female victims in the study, lone perpetrators, perpetrators who were older, and also perpetrators who were acquainted with the victim were all more common than they were among the male victims. Studies of lethal violence towards homo- sexual men show the same thing with regard to the perpetrators (Lander 1996; Bartlett 2007). They were young men with different kinds of problems like drug problems, criminal records, and unemployment. Perpetrators of hate crimes have also been studied in the USA (McDevitt et al. 2010). A study based on 2,976 hate crimes registered between 1997 and 1999 found that it was for the most part males that were registered as perpetrators in connec- tion with hate crimes. The majority of them were young men and were previously unknown to the victim. In three-quarters of the reported cases, the perpetrator was alone at the time of the offence. The study’s authors argue, however, that it should be noted that more than one person may be involved in the crime, even though they did not participate directly at the time the crime was actually committed. Others may have provided help and support to the perpetrators, for example. When studying the perpetrators of hate crime it is important to understand the underlying factors that lead to the commis- sion of hate crimes. Research has noted that youths with intolerant attitudes towards one minority often also have intolerant attitudes 1 In 2009 the perpetrators’ mean age was 31 for xenophobic/racist and religion-based hate crimes, 28 years in relation to homophobic hate crimes, and 26 for transphobic hate crimes (Brå 2010a). ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 201 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 towards other minority groups (Ring and Morgentau 2004; Ring 2006). High levels of intolerance appear to be associated with low levels of parental educational attainment and social class, problematic family con- ditions, individual-level factors such as restlessness, aggression, and a lack of empathy, and low levels of school achieve- ment. Stereotyped gender norms, feelings of being an outsider in relation to society, and frequently associating with friends in the evening are also associated with intolerance. These same factors are also viewed as involving a risk for criminal behaviour (Loeber and Farrington 1998). It has been noted in the literature that theoretical explanations of the origins of hate crime are a rarity (Craig 2003). Thus the core question in this research field— that of why hate crimes occur—remains under-theorized. One of the reasons for this may be that much of the focus of the research community has been directed at the issues of unclear definitions and the difficulties associated with attempts to measure the extent of the problem (Perry 2003:93). Another reason may be that the perspective of the victim has been central to theoretical efforts, which means that the perpetrators, along with the related issue of the causes of hate crime, have been neglected. Since the subject of hate crimes is a complex one, covering a range of offence types that can take place anywhere and can be committed by anyone, and where the strength of the causal links between hate and the criminal acts can vary substantially, it would probably be overly optimistic to expect to produce a single theory that can explain the entire field (Tiby 2010:191). One perspective that has been dis- cussed, however, involves the view that there is a hegemonic conception of ‘how the world should be’ which is based on dominant structures such as work, power, sexuality, and culture (Perry 2003:100). Within these structures, norms are con- structed as to how individuals should be. This means that we differentiate between people who follow these norms and those who deviate from them. A hate crime then involves the perpetrator demonstrating his/her identity and at the same time punishing the victim because he or she is perceived to deviate from the established norms (ibid.:105). It may also be the case that the perpetrator of a hate crime wants to make it clear that he/she is not a member of the group to which the victim belongs (Craig 2003). Viewed in this way, hate crimes may be regarded as a special form of aggression. The act has a symbolic function. For the perpetrators it does not matter whether or not the victim identifies him/herself as a member of a specific group. Instead, what is important to the perpetrators is making the statement that they are not themselves part of the group of which they perceive the victim to be a member. Craig also means that hate crimes differ from other aggres- sive offences because they typically involve multiple perpetrators. She also notes that hate crimes appear to be crimes carried out by young men against young men. Hate crimes can also be discussed on the basis of the concept of ‘message crimes’ (Iganski 2003). These crimes not only injure the initial victim but also the group to which the victim belongs or is presumed to belong. Since people are subjected to crimes on the basis of who they are perceived to be, or the group to which they belong, this causes greater suffering, which should result in a more severe sanction (ibid.). ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention202 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 On the basis of register data, Levin and McDevitt have produced a typology of hate crime perpetrators (Levin and McDevitt 2002; McDevitt et al. 2003). The data were drawn from the Boston police, where investigators have been trained to identify hate crimes. Levin and McDevitt’s point of departure was to look at the motives behind the hate crimes. In the majority (66%) of cases, the motive for the hate crime was sensation seeking. In 25% of the cases the motive was defensive, in 8% revenge, and in 1% of cases the perpe- trators were on a mission to ‘save the world’ from undesirable groups. What makes the study interesting is that it makes an attempt to look at the underlying motives for hate crimes. That is to say, there is an assumption that the hate motive itself is not the critical element in whether or not the crime is committed. However, the study does not problematize the question of what the underlying factors behind the hatred might be, or how important these factors might be, for whether or not a hate crime will be committed. McDevitt et al. (2010) have subsequently argued that the typology needs to be developed further. Amongst other things, they note that one motive need not exclude another, and that a hate crime may rather be a product of several motives. They also feel that attention should be focused on individuals who do not directly participate in hate crimes but who are present to provide support. McDevitt et al. also discuss whether incidents such as the 9/11 attacks may affect the distribution over time of the motives that are most commonly associ- ated with hate crimes. Following this incident it is conceivable that hate crimes motivated by defence or revenge may have become more common. As has been noted above, survey studies have shown that people who have a hostile attitude towards one minority group are usually also hostile towards other minorities (Ring and Mor- gentau 2004; Ring 2006). Thus people with xenophobic attitudes are commonly also hostile towards LGBT people. In order to make progress and advance our knowl- edge of this phenomenon it is possible that other methods than those based on register data and questionnaire surveys may be required, such as interviews, for example. Previous research shows that there are some findings about perpetrators or sus- pected people of hate crimes. But there is no research about registered co-offending and specialization in crime for people suspected for hate crimes. Therefore it is relevant to do this study to get more knowledge about people who are suspected for hate crime. To answer this question, data from the national Registers of Suspected and Convicted Offenders will be used. Although this type of data has certain limitations, it is preferable instead of using interviews, for example. The aim is to find some patterns in the suspected people’s criminality, which motivates the choice of data. What type of results can be expected in the current study on the basis of the research and theoretical assumptions pre- sented above? The finding that individuals who are intolerant towards one minority group usually also hold intolerant attitudes towards several minorities would suggest that we may find perpetrators in the data set who have been suspected of committing offences against a number of different minority groups. There might also be an expectation that the suspected persons will be relatively young. It may be more ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 203 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 important for younger people to act in a way that shows them to be distancing themselves from groups like homosexuals or people of a different religion. By means of such actions the perpetrator establishes that he does not belong to the group in question. There might also be an expec- tation that it will be common with co- offenders. Method In order to find people suspected2 for hate crimes the offences reported to the police in 2006 that have been classified as hate crime have been used. The classification is made by the Swedish National Council for Crime Prevention on the basis of the offence descriptions contained in the offence reports (Brå 2007:31). On the basis of offence description they also categorize the relationship between the victim and the offender. This study is limited to hate crimes with xenophobic, homophobic, and Islamophobic motives.3A further restric- tion has also been made with regard to the offence types included in the study. These are unlawful threats, molestation, and assault. Unlawful threats and molestation are the most common offences types found among the classified hate crimes, followed by assault (ibid.:43). Offence types such as defamation and agitation against a national or ethnic group have been excluded. In other words, the study includes the most common hate crime motives and the most common and more serious offence types found among the hate crimes. In 2006 a total of 1,910 offence reports relating to assault or unlawful threats/ molestation were coded as hate crimes with xenophobic, Islamophobic, or homopho- bic motives (Brå 2007:42). Table 1 presents the distribution across the different hate crime motives. As can be seen from the table, xenopho- bic hate crimes are over-represented by comparison with those based on other motives. Only 7% of the hate crimes have been classified as Islamophobic. For the purposes of this study, incidents involving assault, unlawful threats, and molestation have been selected. The distri- bution of these offence across the different motive categories (xenophobic, homophobic, and Islamophobic) is presented in Table 2. Table 2 shows that unlawful threats or molestation are more common than assault among registered hate crimes. The pro- portion of incidents of unlawful threats and molestation is higher among the Islamophobic hate crimes than it is among the two other categories of hate crimes. The primary data have been collected from the national Registers of Suspected Table 1. Number of reported hate crimes in the form of assault and unlaw- ful threats/molestation, by xeno- phobic, Islamophobic, and homophobic motives, 2006 (%) n % Xenophobic 1320 69 Islamophobic 140 7 Homophobic 450 24 Total 1910 100 2 In this case it means that they have been suspected by the police during the investigation. 3 Anti-Semitic hate crimes were excluded since the number of such offences was very small. These offences comprised only 4% of the total number of hate crimes (Brå 2007:42). ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention204 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 and Convicted Offenders. At the time the data were collected from these registers, the Register of Suspected Offenders con- tained information for the period 1995 – 2006. The Register of Convicted Offenders contained information for the period 1973 – 2006. Data from the first of the two registers have been employed to study the question of possible suspected co- offenders. Data from the convictions register have been employed to examine the issues of prior involvement in crime and specialization in offending among those suspected of hate crimes. Previous studies that have examined specialization in offending have done so on the basis of different points of departure (see e.g. Wikström 1987; Kyvsgaard 1998). In part the crime types examined have varied, and there are differences in the definitions of specialization employed. The current study focuses on the offence types assault, unlawful threats, and molestation. Given that these are the offence types chosen for the study, it is of interest to see whether the suspected offen- ders are specialized in just these types of crime. According to previous research I have decided the following: To be included in the analysis of specialization, the offender has to have been convicted on at least three occasions. This restriction was introduced in part in order to avoid classifying indivi- duals with only one or two convictions as specialists, in part because it has been used in other studies focused on the issue of specialization (Wikström 1987). Further, the individuals are only included in the analysis if they have been convicted of one of the three offence types noted above. In this study, all the individuals included have been suspected of assault, unlawful threats, or molestation. This does not necessarily mean that they have been convicted of these offences, however. The suspected hate crime offenders’ convictions have been divided into four categories: violent offences, threats and molestation, violence and threats/molesta- tion against public servants, and other offences.4 For a person to be classified as a specialist, some of his/her convictions must involve the same types of crime. In this study, the cut-off has been set at two-thirds (following Wikström 1987). This means that in order for an individual to be classified as a specialist, over 66% of the individual’s convictions must be violent offences, unlaw- ful threats or molestation, or violence/ threats/molestation against public servants, respectively. The use of the two-thirds cut-off means that for a person convicted of three offences, two of these convictions must relate to the same category for the individual to be counted as a specialist. As has already been noted, the Swedish National Council for Crime Prevention has Table 2. Number of hate crimes involving assault or unlawful threats/mo- lestation by hate crime motive— xenophobic, Islamophobic, and homophobic, 2006 (%) Assault Unlawful threats or molestation n % n % Xenophobic 470 36 850 64 Islamophobic 34 24 106 76 Homophobic 163 36 287 64 Total 667 35 1243 65 4 The categorization employed means that it is not possible to differentiate between violence and threats/molestation against public servants on the basis of offences that have resulted in convictions. ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 205 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 studied hate crimes every year since 2006. This might indicate that the current study is superfluous. The motivation for conduct- ing the current study, however, is that the National Council’s 2006 study presents very few results about suspected people (Brå 2007:86 ff.). This article will also examine the issues of co-offending relations and specialization which have not been studied by the National Council. Problems associated with studies of this kind Using registered crime in the context of research is associated with certain limi- tations. It is widely known that not all crime is brought to the attention of the police or other authorities and thus remains unregis- tered (Brå 2006; von Hofer 2006). It is also known that some types of crime are registered more often than others. More serious crime is generally reported more often than less serious offences. Reporting propensities are also affected by the relationship between perpetrator and victim. Crimes are more often reported where the victim is not acquainted with the perpetrator. Confidence in the justice system may also influence the extent to which victims report offences, and the resources and prioritiza- tions of the police and other agencies also affect which offences are registered. When it comes to hate crimes, there are additional factors that affect the validity of the data, i.e. that we are really measuring what we intend to measure. For the National Council for Crime Prevention to register an offence as a hate crime, the offence must either have been registered under one of the offence codes which are included in the National Council’s hate crime sampling frame, or have been identified as a suspected hate crime in the police’s offence registration system RAR (Brå 2010a:20). The police must also have recorded relevant information in the offence description included in the police report, and have included words from the list employed in the National Council’s electronic search procedure. Finally, the coder must interpret the incident and assess it to constitute a hate crime. At the present time we have no information as to the significance of the various stages in the hate crime registration process for the question of whether or not an offence will finally be registered as a hate crime (ibid.:22). The National Council for Crime Prevention states that users of the hate crime statistics need to bear in mind that the statistics present a minimum estimate of the number of hate crimes reported to the police. It could also be a problem to separate Islamophobic hate crimes from xenopho- bic hate crimes, if it is not clear that the victim is a Muslim. The consequence will be that some Islamophobic hate crimes will be registered like xenophobic hate crimes (Brå 2007:26). Previous studies have shown that the dark figure for hate crimes is large (Tiby 2006). It is also conceivable that the size of this figure varies between different types of hate crime. This knowledge, together with the factors described above as affecting the hate crime statistics in various ways, means that it is important to bear in mind that this article focuses on registered crime and nothing else. As you will see in the findings, of the 1,910 hate crime reports included in the analysis, a suspected person has been identified in connection with 579, or 30%. This means that it is difficult to generalize the results to perpetrators of hate crime. Instead it will be better to talk about trends in this study. ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention206 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 Findings This section presents the results relating to suspected persons who were registered for hate crimes in 2006. The presentation begins by describing the relationship between the victims and the perpetrators. This is followed by results relating to the age and sex of the suspected persons, suspected co-offenders, previous convictions, and possible specialization in crime. Relationship between the victims and the perpetrators The offence reports sometimes contain information on the relationship between the victims of hate crimes and the perpetrators. Swedish National Council for Crime Prevention categorizes the relationship between the victim and the offender. Table 3 presents the distribution for both the different hate crime motives and the number of hate crimes in total. The different kinds of relations have also been categorized into four groups according to the Swedish National Council for Crime Prevention (Brå 2007, 2008, 2009, 2010a). The groups are Related, Superficially acquainted, Unknown, and Missing data. Table 3 shows that the perpetrator was unknown to the victim in one-third of the reported hate crimes. In almost one-quarter of the offence reports there is no infor- mation on the relationship between the victim and the perpetrator. In nearly 43%, there is information showing that the identity of the person who has subjected the victim to the hate crime is known. In this case it means that the perpetrator has been classified to be related or superficially acquainted to the victim. In the reports where the Swedish National Council for Crime Prevention have been able to categorize the relationship between victims and perpetrators, it is slightly more com- mon for the victim to be related or super- ficially acquainted with the perpetrator than it is for the perpetrator to be unknown to the victim. It should be remembered in this context that it is more common to report offences when the perpetrator is unknown to the victim than when he or she is an acquaintance of the victim (Brå 2006; von Hofer 2006).The results are in line with previous research that has shown that the perpetrators may be found in both the immediate environments of the victims and in more peripheral positions (Mason 2005a, 2005b; Tiby 2010). Persons suspected for hate crimes Of the 1,910 hate crime reports included in the analysis, a suspected person has been identified in connection with 579, or 30%. This means that the suspected individual’s personal identification number is registered in the offence report. The material includes a total of 558 such individuals. Of these, 17 have been registered in connection with two or more of the offence reports. One of the 17 has been registered as a suspect in connec- tion with one crime with a xenophobic motive and one with a homophobic motive. Another is suspected for one xenophobic hate crime, one Islamophobic, and one with a homophobic motive.5 The remainder of the individuals in the material have been suspected of offences with a single motive. In other words this is an unexpected result, given that previous research has shown that those with intolerant attitudes towards one minority group are usually also intolerant towards other minorities (Ring and 5 This means that in some calculations the number of suspects will be 561, since two individuals appear in two and three hate crime categories respectively. ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 207 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 Morgentau 2004; Ring 2006). Given this we might have expected to find more individuals who had been suspected of hate crimes with different underlying motives. It may also be that intolerant attitudes need not necessarily lead to criminal activities. It should be remembered that in the majority of the reports, 70%, the police have not identified a suspected person. Therefore it will be difficult to generalize the results to perpetrators of hate crimes. But it is also an interesting result that the police have found a suspected person in only 30% of the cases. The reports with an Islamophobic motive include a smaller proportion of registered suspects, at only 21%, than the reports associated with the other two hate crime motives. The corresponding figure for both the xenophobic and the homo- phobic hate crimes is 30%. The sex distribution of the suspected persons of hate crime has also been examined. Table 4 presents the results by the different hate crime motives. Since there are individuals who have been suspected of crimes with different motives, the same individual may be present in two or three hate crime categories. As expected, men are over-represented among the persons suspected of hate crimes. This is in line with the findings presented in previous research (see e.g. McDevitt et al. 2010). The mean age of the individuals suspected of hate crimes is 31 years. Looking to the different hate crime motives, those suspected Table 3. Relationship between victims and perpetrators in reported hate crimes in the form of assault and unlawful threats/molestation, by crimes with xenophobic, Islamophobic, and homophobic motives, 2006 (%) Xenophobic Islamophobic Homophobic Total n % n % n % n % Related 78 5.9 8 5.7 39 8.7 125 6.5 Cohabitant/partner/ spouse 7 0.5 1 0.7 0 0 8 0.4 Former partner 35 2.7 1 0.7 8 1.8 44 2.3 Family member/relative 12 0.9 4 2.9 10 2.2 26 1.4 Acquaintance/friend 24 1.8 2 1.4 21 4.7 47 2.5 Superficially acquainted 507 38.4 33 23.6 150 33.3 690 36.1 Colleague 12 0.9 3 2.1 1 0.2 16 0.8 Schoolmate 101 7.7 7 5.0 46 10.2 154 8.1 Neighbour 165 12.5 7 5.0 38 8.4 210 11.0 Known person 229 17.3 16 11.4 65 14.4 310 16.2 Unknown 423 32.0 60 42.8 172 38,2 655 34.3 Public agency staff/ Service provider 62 4.7 2 1.4 9 2.0 73 3.8 Unknown person 361 27.3 58 41.4 163 36.2 582 30.5 Missing data 312 23.6 39 27.9 89 19.8 440 23.0 Total 1320 100 140 100 450 100 1910 100 ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention208 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 of homophobic hate crimes have the lowest mean age, at 25 years. The corresponding mean ages for persons suspected of xeno- phobic and Islamophobic hate crimes are 34 and 32, respectively. These figures also correspond with those presented in the National Council’s reports of hate crimes over recent years (Brå 2008, 2009, 2010a). It is interesting that those suspected of homo- phobic hate crimes are on average younger than those suspected of crimes based on the other hate crime motives. One possible explanation could be that young people are keen to make it clear that they are not homosexuals (Craig 2003). This would then lead to a younger mean age among those suspected of homophobic hate crimes by comparison with the other types of hate crime. However, this explanation needs to be explored further. Twenty of the reports include infor- mation on suspected co-offenders. This represents just 3% of the 579 offence reports which include a personal identifi- cation number for a suspect. The material includes a total of 25 suspected co- offenders. Together these individuals are suspected of a total of 24 offences. The number of co-offenders associated with a single offence varies between one and three. Thus at most four individuals have been suspected of committing a hate crime together in this study. In other words, it is uncommon for individuals to be registered as suspected co-offenders regarding hate crimes. This result is not in line with previous research (Levin and McDevitt 2002; Craig 2003; McDevitt et al. 2003). Tiby found in a survey about homosexuals that several perpetrators were more com- mon than lone perpetrators (Tiby 1999: 154). However, one should consider that the information about perpetrators/suspected people is from different sources, from victims in Tiby’s survey and from the police in this study. It is difficult to draw conclusions about reported hate crimes involving co-offen- ders on the basis of the distribution presented in Table 5, nor is there very much to be gained from a more detailed examination of these offence reports. What can be noted, however, is that it is very uncommon for two or more individuals to be registered for having committed hate crimes together. Prior convictions The next stage of the analysis involves studying whether the suspected people of hate crime had been convicted of offences during the period 1973 – 2006. A total of 345 of the hate crime suspects had records in the convictions register.6 These had together been registered in connection with Table 4. Sex distribution among individ- uals suspected of assault and unlawful threats/molestation offences with hate crime motives, 2006. Presented by xenophobic, Islamophobic, and homophobic hate crimes (%) Males Females n % n % Xenophobic 321 81 73 19 Islamophobic 26 87 4 13 Homophobic 111 81 26 19 Total 458 82 103 18 6 Of these 345 individuals, 1 had been suspected both of a hate crime with a xenophobic motive and of one with a homophobic motive. This individual is therefore counted twice, and the number of individuals with prior convictions sums to a total of 346 in certain contexts. ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 209 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 a total of 2,620 convictions. Table 6 presents the prior convictions, i.e. for the period prior to the hate crime, of those registered as hate crime suspects.7 The results have been broken down on the basis of the number of convictions for which the individuals in question had been registered prior to the commission of the suspected hate crime included in the study. It can be seen from the table that 55% of the persons registered in 2006 for hate crimes in the form of assaults, threats, and molestation have prior convictions. This finding does not correspond with the results of the National Council’s more recent studies (Brå 2008, 2009, 2010a). The National Council’s studies include all types of crime. This may be due to the fact that the current study is focused only on more serious offences. It is conceivable that those suspected of more serious crimes have prior convictions to a greater extent than others. Among those suspected of homophobic hate crimes, however, it is more common for the suspects to have no prior criminal record. This may be linked Table 5. Number of offence reports involving one suspect offender and suspected co- offenders in connection with assaults and unlawful threats/molestation with hate crime motives, 2006. By crimes with xenophobic, Islamophobic, and homophobic motives (%) One suspect Suspected co-offenders Total n % n % n % Xenophobic 394 97 14 3 408 100 Islamophobic 31 94 2 6 33 100 Homophobic 134 97 4 3 138 100 Total 559 97 20 3 579 100 Table 6. Number of prior convictions among individuals suspected of hate crimes involving assault or unlawful threats/molestation in 2006, by whether the hate crime was based on a xenophobic, Islamophobic, or homophobic motive (%, 1973 – 2006) Xenophobic Islamophobic Homophobic Total No. of prior Convictions n % n % n % n % 0 162 41 12 40 76 55 250 45 1 – 2 88 22 7 23 35 26 130 23 3 – 4 42 11 4 13 9 7 55 10 5 – 9 43 11 2 7 6 4 51 9 10 – 59 15 5 17 11 8 75 13 Total 394 100 30 100 137 100 561 100 7 This analysis includes a total of 2,472 convictions. ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention210 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 to the fact that this group is on average somewhat younger than those suspected of hate crimes with xenophobic or Islamo- phobic motives. It is thus possible that other measures are taken in connection with their offending rather than prosecut- ing and convicting them. One other explanation could be that they maybe are less criminal. The question of whether there are differences between the different categories of hate crime suspects with regard to the types of crime for which they are com- monly convicted has also been examined.8 Among those suspected of xenophobic hate crimes, the most common type of prior conviction is for motoring offences, fol- lowed by theft offences and violent offences. For those suspected of Islamo- phobic hate crimes, it is theft offences, violent offences, and drug offences. The corresponding ranking for those suspected of homophobic hate crimes is theft offences, followed by motoring offences and violent offences. In other words, the different groups of hate crime suspects have prior convictions for the same categories of crime. The ranking of the different categories of prior convictions varies somewhat between the different groups of hate crime suspects, however. If we compare these results with the official crime statistics relating to all persons convicted of offences over recent years, we find that the same offence types are also common there (Brå 2010b). In other words, the population examined in this article, i.e. persons suspected of hate crimes, is not distinctive as regards the types of offences for which the suspects have previously been convicted. The final analysis focuses on whether the individuals suspected of hate crimes were specialists in the crime categories violence or threats and molestation. The analysis includes those hate crime suspects con- victed of at least three offences and who had also been convicted of violent crimes and/or threats or molestation—a total of 141 individuals. In order to be classified as a specialist, two-thirds of the individual’s offences had to have comprised incidents of violence or threats/molestation, respect- ively. The results show that four individ- uals, or 3%, qualify as specialists according to this definition. All four are males. Three of them are specialized in violent offences, and one in threats and molestation. All four have been convicted of three offences, of which two were violent offences or threats/molestation, respectively. In the original hate crime material, all four are suspected of hate crimes with xenophobic motives. The analysis thus shows that it is uncommon for persons suspected of hate crimes to be specialized in assault or threats/molestation. This finding corre- sponds with the results of existing research on specialization in general, which also show that it is uncommon for individuals to be convicted repeatedly in connection with only one specific type of crime (Wikström 1987; Kyvsgaard 1998). Summary and discussion In summary, this study has shown that there was an over-representation of xeno- phobic hate crimes in relation to homo- phobic and Islamophobic hate crimes among the molestation/threat and assault offences reported to the police in 2006. 8 The analysis focuses only on the principal offence in a given conviction, i.e. the offence associated with the most severe sanction in its sentencing scale. This means that less serious offences will be under-represented in relation to more serious types of crime. ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 211 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 In those cases where the reports contain information on the relationship between the victim and the perpetrator, it is more common for the perpetrator to be known than to be unknown to the victim. Thirty per cent of the offences have been linked to an identified suspect. As mentioned before, it leads to difficulties to generalize the results to perpetrators of hate crime. Not unexpectedly, there is an over-represen- tation of males among the perpetrators. The average age of the suspects is lowest among those suspected of homophobic hate crimes. Only a few reports, 3% of those which included an identified suspect, contain information about suspected co- offenders. The study has also shown that 55% of the suspects have prior convictions. Those with records in the convictions register have primarily been convicted of motoring offences, thefts, violent offences, and drug offences. Finally, it could be noted that is rare for those suspected of hate crimes to be specialized in assault or molestation/threat offences. What tendencies can we see in these results? Firstly, we see no conspicuous differences between those suspected of xenophobic, Islamophobic, or homophobic hate crimes with regard to age, sex, or prior involvement in crime. Secondly, the reported hate crime offences are almost exclusively linked to a single suspect. This is to say that it is uncommon for two or more individuals to be suspected of committing hate crimes together. Thirdly, the suspected persons of these types of hate crimes are not ‘specialists’. The study has not shown that the perpetrators are specialized and focus on the commission of a certain type of crime against certain types of individual. In order to examine this question in more detail, however, it would be of interest to study hate crimes over a longer period of time. In order to understand better the findings presented in this study, it would also be appropriate to conduct interviews with persons suspected of hate crimes, for example (see Tiby 2006). This is particu- larly the case in relation to the question of the underlying causes of this type of criminality, since interview studies could examine the questions of what gives rise to feelings of hatred towards certain groups, and what it means to feel this kind of hatred. Are these crimes always governed by feelings such as hatred? Or are they governed by other feelings, other factors, or other combinations of feelings and factors? In order to improve the available knowledge on hate crime it is also important to discuss these questions. One of the factors that make this subject a complicated one to study is that the crimes, in addition to the presence of criminal intent, are also characterized by a specific underlying motive. The investi- gation of different types of crime can be more or less complex. In the context of hate crimes, it is not only the offence itself that has to be examined, but also the motive behind the crime. Research into hate crimes thus requires us to ask different questions and to use other methods than those commonly employed in criminological research. As has already been noted, it is important to remember that this article focuses exclusively on information relating to registered crime. This involves limi- tations when it comes to interpreting and understanding these crimes. Thirty per cent of the offence reports included in this study include registered information on sus- pected offenders. This naturally gives rise to the question of who the offenders ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention212 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 are in the 70% of cases where we have no information. There are also very few reports that include information on sus- pected co-offenders. It would therefore be interesting to look in more detail at individuals who are perhaps not directly involved in the hate crime incident itself, but who are in the company of the suspected persons (see McDevitt 2010). What is the significance of the support provided by these people prior to, during, and after the hate crime itself? A major problem is that we cannot follow hate crimes from police report through to court judgement by means of register data. Thus we cannot see whether the individuals registered as suspects are also convicted of these crimes. Nor does the convictions register contain information on whether or not the convictions relate to hate crimes. In other words, it would be very useful if the data registered by the justice system were to contain more information that facili- tated research of this kind. References Bartlett P (2007). Killing Gay Men, 1976 – 2001. British Journal of Criminology 47(4):573 – 595. Brå (2006). Konsten att läsa statistik om brottslighet [How to Read Statistics on Crime]. Available at: http://www.bra.se/ extra/faq/?module_instance¼2& action¼question_show&id¼349& category_id¼0 (Accessed 27.06.2011). Brå (2007). Hatbrott 2006. En sammanställn- ing av polisanmälningar med främlingsfien- tliga, islamofobiska, antisemitiska och homofobiska motiv [Hate Crimes 2006.A Summary of Reports to the Police of Crimes Motivated by Xenophobia, Islamo- phobia, Anti-Semitism, and Homophobia]. Report 2007:17. Stockholm: Brottsförebyg- gande rådet. Brå (2008). Hatbrott 2007. En sammanställning av polisanmälningar med främlingsfientliga, islamofobiska, antisemitiska och homofo- biska motiv [Hate Crimes 2007.A Summary of Reports to the Police of Crimes Motivated by Xenophobia, Islamophobia, Anti-Semit- ism, and Homophobia]. Report 2008:15. Stockholm: Brottsförebyggande rådet. Brå (2009). Hatbrott 2008. Polisanmälningar där det i motivbilden ingår etnisk bakgrund, religiös tro, sexuell läggning eller könsöver- skridande identitet eller uttryck [Hate Crimes 2008. Reports to the Police where the Motivation for Crime Includes Ethnic Back- ground, Religious Faith, Sexual Orientation, or Transgender Identity or Expression]. Report 2009:10. Stockholm: Brottsförebyg- gande rådet. Brå (2010a). Hatbrott 2009. Statistik över polisanmälningar där det i motivbilden ingår etnisk bakgrund, religiös tro, sexuell läggn- ing eller könsöverskridande identitet eller uttryck [Hate Crimes 2009. Statistics of Reports to the Police where the Motivation for Crime Includes Ethnic Background, Religious Faith, Sexual Orientation, or Transgender Identity or Expression]. Report 2010:12. Stockholm: Brottsförebyggande rådet. Brå (2010b). Personer lagförda för brott. Slutlig statistik för 2009 [Persons Found Guilty for Offences. Final Statistics for 2009]. Stencil. Stockholm: Brottsförebyggande rådet. Brå (2011a). Available at: http://www.bra.se/ extra/pod/?action¼pod_show&id¼16& module_instance¼11 (Accessed 27.06.2011). Brå (2011b). Available at: http://www.bra.se/ extra/pod/?action¼pod_;show&id¼17& module_instance¼11 (Accessed 27.06.2011). ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 213 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 http://www.bra.se/extra/faq/?module_instance=2&action=question_show&id=349&category_id=0 http://www.bra.se/extra/faq/?module_instance=2&action=question_show&id=349&category_id=0 http://www.bra.se/extra/faq/?module_instance=2&action=question_show&id=349&category_id=0 http://www.bra.se/extra/faq/?module_instance=2&action=question_show&id=349&category_id=0 http://www.bra.se/extra/pod/?action=pod_show&id=16&module_instance=11 http://www.bra.se/extra/pod/?action=pod_show&id=16&module_instance=11 http://www.bra.se/extra/pod/?action=pod_show&id=16&module_instance=11 http://www.bra.se/extra/pod/?action=pod_show&id=17&module_instance=11 http://www.bra.se/extra/pod/?action=pod_show&id=17&module_instance=11 http://www.bra.se/extra/pod/?action=pod_show&id=17&module_instance=11 Craig KM (2003). Examining hate-motivated aggression: a review of the social psycho- logical literature on hate crimes as a distinct form of aggression. In: Perry B (ed.). Hate and Bias Crime. A Reader. pp. 117 – 130. New York: Routledge. Dagens Nyheter (2008). Available at: http:// www.dn.se/sthlm/man-utsatta-for-hatbrott-1. 635846 (accessed 29 September 2009). Dahl U (2005). Det viktigaste är inte vad extremisterna tycker utan vad den stora majoriteten gör. En kunskapsinventering av forskning om homofobi och heteronormativi- tet [The Most Important Thing is not What the Extremists Think, It’s What the Vast Majority Do. A Knowledge Inventory of Research on Homophobia and Heteronormativity]. Stock- holm: Forum för levande historia. Herek GM (2009). Hate Crimes and Stigma- related Experiences among Sexual Minority Adults in the United States. Prevalence Estimates From a National Probability Sample. Journal of Interpersonal Violence 24(1):54 – 74. von Hofer H (2006). Svensk våldtäktsstatistik i internationell belysning [Swedish rape stat- istics in the international lighting]. In: Roxell L, Tiby E (eds). Frågor, fält och filter. Kriminologisk metodbok. Lund: Studentlit- teratur. Iganski P (2003). Hate crimes hurt more. In: Perry B (ed.). Hate and Bias Crime. A Reader. pp. 131 – 138. New York: Routledge. Jenness V (2001). The Hate Crime Canon and Beyond: A Critical Assessment. Law and Critique 12:279 – 308. Kyvsgaard B (1998). Den kriminelle karriere [The Criminal Career]. Köpenhamn: Jurist- og Økonomforbundets Forlag. Lander I (1996). Dödligt våld mot homosex- uella män—en registerstudie [Lethal violence against gay men—a register study]. In: Tiby E, Lander I (eds). Hat, hot, våld—utsatta homosexuella kvinnor och män. En pilotstudie i Stockholm. Report 1996:84. pp. 103 – 140. Stockholm: Folkhälsoinstitu- tet. Levin J, McDevitt J (2002). Hate Crimes Revisited. America’s War on Those Who Are Different. Colorado: Westview. Loeber R, Farrington D (eds) (1998). Serious and Violent Juvenile Offenders. Risk Factors and Successful Interventions. Thousand Oaks: SAGE. Mason G (2005a). Hate Crime and the Image of the Stranger. British Journal of Criminology 45:837 – 859. Mason G (2005b). Being Hated: Stranger or Familiar?. Social & Legal Studies 14(4): 585 – 605. McDevitt J, Levin J, Bennett S (2003). Hate crime offenders: an expanded typology. In: Perry B (ed.). Hate and Bias Crime. A Reader. pp. 109 – 116. New York: Routle- dge. McDevitt J, Levin J, Nolan J, Bennett S (2010). Hate crime offenders. In: Chakra- borti N (ed.). Hate Crime. Concepts, Policy, Future Directions. Devon: Willan Publishing. Perry B (2003). Introduction. Causes and consequences. In: Perry B (ed.). Hate and Bias Crime. A Reader. pp. 93 – 108. New York: Routledge. Ring J (2006). Intolerans. Att mäta ungas attityder till och upplevelser av antisemitism, homofobi, islamofobi och invandrarfien- tlighet [Intolerance. Measuring young persons’ attitudes to and experiences of anti-Semistism, homophobia, Islamophobia, and anti-immigrantism]. In: Roxell L, Tiby E (eds). Frågor, fält och filter. Kriminologisk metodbok. Lund: Studentlitteratur. Ring J, Morgentau S (2004). Intolerans. Anti- semitiska, homofobiska, islamofobiska och invandrarfientliga tendenser bland unga ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention214 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 http://www.dn.se/sthlm/man-utsatta-for-hatbrott-1.635846 http://www.dn.se/sthlm/man-utsatta-for-hatbrott-1.635846 http://www.dn.se/sthlm/man-utsatta-for-hatbrott-1.635846 [Intolerance. Anti-Semitic, Homophobic, Islamophobic and Xenophobic Tendencies among Young]. Stockholm: Brottsförebyg- gande rådet och Forum för levande historia. Sullaway M (2004). Psychological Perspectives on Hate Crime Laws. Psychology, Public Policy and Law 10(3):250 – 292. Tiby E (1999). Hatbrott? Homosexuella kvin- nors och mäns berättelser om utsatthet för brott [Hate Crime? Homosexual Women and Men’s Narratives of Victimization]. Stock- holm: Kriminologiska institutionens avhan- dlingsserie nr 1, Stockholms universitet. Tiby E (2010). Hatbrott. Monster, mönster & möjligheter [Hate crimes. Monster, patterns and possibilities]. In: Deland M, Hertzberg F, Hvitfeldt T (eds). Det vita fältet. Samtida forskning om högerextremism. Uppsala: Opuscula Historica Upsaliensis 41. Värmlands Folkblad (2009). Available at: http://www.vf.se/Nyheter/Varmland/ Rasistiska-motiv-kan-ligga-bakom- branderna-090915.aspx (accessed 29 Sep- tember 2009). Wikström P-OH (1987). Patterns of Crime in a Birth Cohort. Age, Sex, and Social Class Differences. Project Metropolitan no. 24. Stockholm: Department of Sociology, University of Stockholm. LENA ROXELL Department of Criminology Stockholm University 106 91 Stockholm SWEDEN Email: lena.roxell@criminology.su.se ROXELL: REGISTER STUDY OF PERSONS SUSPECTED OF HATE CRIME Journal of Scandinavian Studies in Criminology and Crime Prevention 215 D ow nl oa de d by [ U ni ve rs it y of S to ck ho lm ] at 0 5: 12 1 9 D ec em be r 20 11 http://www.vf.se/Nyheter/Varmland/Rasistiska-motiv-kan-ligga-bakom-branderna-090915.aspx http://www.vf.se/Nyheter/Varmland/Rasistiska-motiv-kan-ligga-bakom-branderna-090915.aspx http://www.vf.se/Nyheter/Varmland/Rasistiska-motiv-kan-ligga-bakom-branderna-090915.aspx