Sage Journals: Discover world-class research

Abstract

The last 20 years have witnessed a growth in the recognition of the profound negative impacts of hate crime. As a result, many criminal justice systems have adopted sentencing approaches that hold the perpetrator accountable for the additional harm these offences cause to victims. While this added sentencing penalty may be supported through a retributive lens, and the wider ‘signalling’ element of these offences, there is comparatively little robust empirical evidence which demonstrates that hate crimes hurt victims more than similar non-hate crimes. This lack of evidence leaves open the possibility that the harms of hate crime could be challenged and the sentencing approaches threatened. Using 8 years (2012/13-2019/20) of the nationally representative Crime Survey for England and Wales, hierarchical linear modelling was used to disentangle the unique emotional impact of hate-motivated violence (from non-hate-motivated violence) and hate-motivated vandalism (from non-hate-motivated vandalism). The results support the notion that hate hurts more, particularly for racist and homophobic violence and gender and disability-related vandalism. However, victims of religiously motivated vandalism did not report greater emotional impacts than those who experienced non-religiously motivated vandalism. The limitations and implications of these unique results are discussed.

Keywords

Crime Survey for England and Wales emotional impact hate crime victimisation violent offences

Introduction

Background

The last 20 years have seen a notable expansion in the field of ‘hate studies’ within academia occurring in parallel with a growth in recognition of the importance of hate crime among the general public, criminal justice system and third sector agencies (Diaz-Faes and Pareda, 2022; Wilkin, 2020). However, this growth has not necessarily fostered consensus about the definitions or appropriate societal and/or criminal justice responses to hate crime (e.g., see the case of Harry Miller; Glass, 2021; Sodha, 2022). In fact, recently a number of UK politicians and commentators have questioned the legitimacy of identity politics and the validity of the very notion of hate crime itself (Hardy and Chakraborti, 2020; Gallagher, 2023). This has the potential to threaten the post-Macpherson consensus that the voice of the victim should be prioritised when deciding whether to record a crime as hate-related or not (Hall, 2013). If, however, it can be established that hate victimisation really does ‘hurt more’ than the non-hate equivalent form of victimisation, this may provide a foundation by which to combat this threat. The aim of the current study is to use multiple years of the largest and most representative victimisation survey in England and Wales (the Crime Survey for England and Wales [CSEW]), to evaluate whether hate, does indeed, hurt more.

Hate crime and the criminal justice system

According to the CSEW, crime victims reported 5.6 million offences in the year ending March 2020, including 1.2 million incidents of violence (Office for National Statistics [ONS], 2020).¹ The personal effects of this victimisation have been well documented, with Shapland and Hall (2007) detailing seven direct effects of victimisation. These included feeling shocked, a loss of faith/trust in society, feeling guilt at having become a victim, physical injury, financial loss (direct loss, such as property stolen and indirect loss, such as time off work), and psychological effects including fear, anger and depression. All of these may persist past the immediate event and result in changes to social behaviours, such as changing one's lifestyle and consequential effects or changes in the perceived risk of future victimisation.

One particular category of offences, that of hate crimes,² has been noted to be especially harmful for victims and their wider identity communities (see, e.g., Iganski, 2001; Levin, 2009; Smith et al., 2012). While the definition of hate crime is still the subject of academic debate, in England and Wales the criminal justice system does have an agreed working definition:

Any criminal offence which is perceived by the victim or any other person, to be motivated by hostility or prejudice, based on a person's disability or perceived disability; race or perceived race; or religion or perceived religion; or sexual orientation or perceived sexual orientation or a person who is transgender or perceived to be transgender. (Crown Prosecution Service, 2016: 2)

In England and Wales there were over 145,000 hate crimes officially recorded by the police for the year ending March 2023 (Home Office, 2023), while in the most recent analysis of the CSEW, it was estimated that there were 190,000 hate crimes reported per year. Hate crimes were more likely to be violence without injury, violence with injury and robbery when compared to other CSEW crimes (Home Office, 2020: 22).

There is a suite of hate crime legislation covering hate offences and hate speech. Racially aggravated offences were covered via the Crime and Disorder Act 1998, including those for assault, criminal damage, harassment and stalking, as well as public order offences (the Act was later amended to incorporate religiously aggravated offences via the Anti-terrorism, Crime and Security Act 2001). In addition, as part of the Criminal Justice Act 2003 (and now incorporated into s66 of the Sentencing Act 2020), new sentencing provisions were, at various stages, introduced for those convicted of hate-related offences. These allowed for enhanced penalties for those found guilty of offences committed with hostility towards the victim's actual or presumed race, religion, disability, sexual orientation or transgender status. In addition, hate speech, in the form of legislation which prohibits incitement to hatred, covers (to varying degrees) racist and religiously motivated speech (Parts 3 and 3A of the Public Order Act 1986, respectively), and that designed to incite hatred against sexual orientation (Part 3A of the Public Order Act 1986).

There are a number of theoretical justifications for the enhanced sentencing provisions for hate crimes. For example, an increase in the severity of punishment may increase general deterrence. That is, individuals may be less likely to engage in hate crime because the increased magnitude of punishment influences their decision to perpetrate such an offence (e.g., Nagin, 2013). Also, the retributionist desert theory (e.g., Von Hirsch, 2007), which proposes offenders should receive a sentence proportionate to the harm they have caused, would suggest that the enhanced sentencing of hate crime is justified because hate is more harmful than the equivalent non-hate offence.

There are two broad (and not mutually exclusive) ways in which hate offences may hurt more. First, hate crimes are proposed to contain a hostile ‘message’ to others in the victim's broader identity community. These harmful ‘ripples’ spread increased levels of anxiety, fear, anger and sadness in a way that it is argued non-hate crimes do not (Perry and Alvi, 2012; Paterson et al., 2019a). These hate offences may also be more likely to lead to behavioural changes for those from the victim's identity community, such as avoidance of certain locations or withdrawal (Bell and Perry, 2015). Inter-group emotions theory (IET) suggests that when group identities are salient, individuals redefine themselves as group members as opposed to individuals and therefore think, feel, act and emotionally respond as a member of the group (Mackie and Smith, 2015). This theory was empirically tested using a sample of 120 LGBTQ+ participants recruited from the Brighton Pride festival by Paterson et al. (2019b). Participants were randomly assigned to one of four conditions where they read a fictious (but presented as real), newspaper article about an attack in which a man's leg was broken and the motivation (Homophobic Hate vs. Non-Hate) and number of perpetrators (single man vs. six men) was varied. The results demonstrated that the indirect experience of a hate crime targeted at one's identity increased the perception of threat and anger compared to the non-hate condition.

The form and magnitude of the ripple effects of hate offences are clearly important and deserving of additional study. These may, in part, justify enhanced penalties, but these are not the topic of the current paper.

The second common justification for more severe sentences is that hate offences hurt the victim more than the equivalent offences without this bias element (see, inter alia, Chakraborti and Garland, 2015; Hall, 2013; Perry, 2001). Such crimes can, it is suggested, damage the victim's sense of belonging, affect their sense of wellbeing and confidence, damage their sense of self-worth and heighten psychological distress more than non-hate offences (Awan and Zempi, 2020; Walters, 2022).

However, it has seemed at times as though this assertion that ‘hate crimes hurt more’ has been ‘taken as read’ by many working within and around the field of hate studies. As Iganski and Lagou asserted over a decade ago, ‘much of the evidence for the supposed greater harm caused by hate crime has been equivocal’ (2009: 2). The explanation for this is these studies have tended to include non-representative samples, lack control or comparison groups, treat victims as homogeneous entities and use different tools to measure the impact of victimisation (e.g., as described by Diaz-Faes and Pareda, 2022). While contextual explorations (e.g., Hardy and Chakraborti, 2020; Wilkin, 2020) are essential for eliciting experiences that too often go unheard, these select samples are not well placed to tell us whether being the victim of a hate crime is more negatively impactful than being the victim of a similar valence non-hate offence.

Assessing the harms of hate crime

During the last 20 years, the field of hate studies has focused primarily upon the forms, frequency and impact of hate victimisation. In one of the first quantitative investigations of this type, Barnes and Ephross (1994) used a purposive sampling approach to identify 59 individuals in the United States who had experienced a hate crime. Results from their survey suggested that the most prevalent emotional response to hate crime was anger at the perpetrator (68%), followed by fear for oneself and family (51%). Feeling sadness and powerlessness were the next most common emotional responses. Over one-third of the participants reported behavioural changes as a result of the victimisation, including moving out of the neighbourhood, decreased social participation and purchasing a gun.

There are many other important investigations which provide critical detail about the profound negative impact of hate crime from the perspective of those victimised. These include the Leicester Hate Crime Project, which explored the emotional and physical impacts of acts of hate on over 1100 victims (Chakraborti et al., 2014), the All Wales Hate Crime Project (Williams and Tregidga, 2013) which undertook a survey examining the experiences of 564 victims of hate, and the Sussex Hate Crime Project (Paterson et al., 2018; Walters et al., 2019), which revealed the traumatising effect of hate crimes upon members of lesbian, gay, bisexual, and transgender (LGBT) groups and Muslim communities. Similarly, Herek et al. (1997) and then Herek, Gillis and Cogan (1999, in an expanded version of the 1997 study) noted that the symptoms of damage to the mental health of lesbians and gay men (including post-traumatic stress, anxiety and depression) after they were victims of bias crimes were notably more significant than those of victims of non-bias crimes. Williams and de Reya's (2019) analysis of online hate speech noted impacts on victims including shame, embarrassment, anger and shock, and loss of self-esteem. Marais et al.’s (2021) analysis of 409 cases of hate crime in South Africa found that victims were more susceptible to emotional consequences if the victim identified as Black African, if they knew the offender or if there were economic consequences of being a victim.

This research is clear in demonstrating that the impact of hate crime can be profound and include negative emotional reactions, as well as negative cognitions and behaviours. The valence of these reactions appears to vary depending on certain characteristics, such as why the person perceives that they were targeted and whether they knew the victim. While it is critically important that the impacts of hate crime continue to be investigated and articulated, it would be disingenuous to fail to acknowledge that being the victim of non-hate crimes could have quite similarly profound negative impacts (Shapland and Hall, 2007).

Studies which use a control or comparison group to contrast the effects of hate with non-hate crimes on victims are, however, relatively rare (as noted by Diaz-Faes and Pareda, 2022). Fetzer and Pezzella (2019) used the 2013 National Crime Victimisation Survey (NCVS) in the United States to evaluate the physical and psychological trauma resulting from both hate and non-hate violence. The data included self-reported details of over 4,600,000 violent victimisations (rape, sexual assault, robbery, aggravated, assault, and simple assault), of which 6.5% were perceived by the victim to be motivated by hostility towards their race, ethnicity, religion, sexual orientation, gender or gender identity, disability, association, or perceived characteristics. The results suggested that hate offences were significantly more likely than non-hate offences to result in serious injury, controlling for the type of weapon used, the number of co-offenders, and whether the offender was under the influence of drugs (OR = 1.2).³ Similarly, hate crime was also associated with significantly greater psychological trauma (OR = 2.5; i.e., feeling worried or anxious, angry, sad or depressed, vulnerable, violated, unsafe, or distrustful). Interestingly, the psychological impact was significantly greater if the offender was perceived to be under the influence of drugs or alcohol (OR = 1.8), and significantly less if the offender was a stranger (OR = 0.6).

This study clearly demonstrates the additional harm of hate crime victimisation; however, there are some limitations to consider. First, despite an allusion to the contrary, this study does not actually allow for the isolation of the unique emotional impact of being the victim of a hate offence. That is, the results showed that hate crimes were on average more vicious, and also that they resulted in greater psychological trauma, but this was very likely as a result of the increased viciousness. It makes sense that more vicious offences result in greater consequences, but this does not provide insight into the added detrimental impact of a hate crime. If both a non-hate and hate offence were of equal seriousness, would the hate offence result in greater emotional consequences? Also, while a number of important offence-related characteristics were controlled in this study (e.g., offence type), no characteristics of the victim were controlled. It is well known that the emotional impacts of crime are greater for females, older individuals and those of certain ethnicities (Shapland and Hall, 2007). It is important to attempt to disentangle the unique psychological impact of hate crime with these individual characteristics considered.

It may be that hate crimes are on average more vicious because they are more likely to include more than one perpetrator than non-hate offences. Lantz and Kim (2019) examined whether hate crimes were more violent because of the bias element of the incident or because they involved co-offenders. Analysing over 88,000 violent bias and no-bias incidents from 2003 to 2012 contained in the National Incident-Based Reporting System and focusing on physical harm, Lantz and Kim found that hate crimes were positively associated with serious injury but that this was partially driven by the co-offender element. Incidents involving both bias and co-offenders were especially violent.

The difficulty disentangling the seriousness of the offence from the hate element of the offence, and essentially equating these in explaining the impact of hate crime, is a common gap in much past research in this area (e.g., Mellgren et al., 2021). Similarly, the most recent results from the CSEW in England and Wales also suggest that hate crime victims were more likely to report being affected compared to the victims of all CSEW offences (Home Office, 2020: 28). That is, more than twice as many hate crime victims said they had suffered a loss of confidence or had felt vulnerable after the crime compared with all victims of CSEW crimes. Hate crime victims were also more likely to report experiencing fear, difficulty sleeping, anxiety or panic attacks, depression or crying. Importantly (and as mentioned above), hate offences were much more likely to be serious, such as violence without injury, violence with injury and robbery, than all other CSEW crimes reported (Home Office, 2020: 22), leading to the possibility that the increased negative emotional consequences of hate crime are a result of the increased seriousness of the particular offence rather than the unique emotional impact of being the victim of a hate crime.

Interestingly, the Home Office annual hate crime statistical bulletin for 2020 (which contains the most recently published CSEW hate crime data) explicitly states that the increased emotional impact of hate crime ‘is probably not due to the type of hate crimes identified by the CSEW, as victims of violent crime tend to be less likely to say they were emotionally affected than victims of other crime types’ (Home Office, 2020: 28). This somewhat misleading statement was also made in a previous version of a Home Office hate crime statistical release (Corcoran and Smith, 2016). Evidence for this statement appears to be justified by comparing the dichotomous summary item ‘Respondent was emotionally affected’, between those who were victims of theft from a person/theft of personal property and those who were victims of all violence. This comparison shows that those who were the victims of violence were marginally less emotionally affected (ONS, 2020; Tables 3.4 and 7.5 cited in Home Office, 2020: 28). However, the same tables show that this is not true when victims of more serious violence or wounding (disproportionate amongst hate crime victims) were compared to the victims of theft from a person/theft of personal property.

The most relevant UK-based research which examined the impacts of hate on a scale equivalent to the present study were those undertaken by Iganski and Lagou (2009, 2014). In the first of these, the researchers assessed the behavioural changes of victims of racially and non-racially motivated crimes via an examination of data from three sweeps of what was then the British Crime Survey (2002/03, 2003/04 and 2004/05). The results showed that a significantly higher proportion of victims of racially motivated crimes, compared to victims of parallel crimes (similar crimes but without the hate element), reported feelings of shock, fear, depression, anxiety, and panic attacks; feelings of a loss of confidence and of vulnerability; difficulty sleeping; and crying, with feelings of fear manifesting the highest difference (Iganski and Lagou, 2009: 9).

In their later study, Iganski and Lagou (2014) analysed the impacts of four strands of hate crime – race, religion, disability and sexual orientation. Their analysis combined data from three sweeps of the CSEW (2009/10, 2010/11 and 2011/12). They concluded that hate crime victims were ‘more likely to report having an emotional reaction to the incident and with a greater intensity’ compared to victims of non-hate crimes (Iganski and Lagou, 2014: 41), with hate victims over twice as likely than those of non-hate crimes to state that they had been affected ‘very much’ (Iganski and Lagou, 2014: 41). A higher proportion of hate victims reported each type of emotional reaction than non-hate, while by crime type, the percentage of hate crime victims reporting being ‘very much’ affected was the highest for serious wounding but also higher across all crime types (Iganski and Lagou, 2014: 41–42).

The work of Iganski and Lagou (2009, 2014) is similar to that of the ONS (2020) and Fetzer and Pezzella (2019) in making an explicit attempt to compare and contrast the impact of hate crime to non-hate offences. However, and as noted, these contributions did not simultaneously control for potentially important details of the offender (e.g., whether known or not known), details of the victim (gender, age) and offence characteristics (e.g., location or timing of offence).

The current paper is the first of which we are aware that uses a multilevel modelling approach to simultaneously control for these and other important offender, victim and offence-related characteristics. Arguably, this approach provides the most accurate assessment of the unique, and potentially additive, effect of being the victim of a hate crime. This approach extends standard regression approaches by incorporating information about the complex structure of the dataset. Here, each victim is able to report on multiple offences, each with a potentially different perpetrator, unique circumstances (including whether or not it was hate-motivated) and varying emotional impacts. Rather than arbitrarily selecting one incident per victim and including details about the perpetrator and victim in a victim-focused analysis or retaining all incidents but only including details about the incidents and not the victim in an incident-focused analysis, multilevel modelling treats both sources of data as different ‘levels’ in the same model. Victims therefore constitute one level in the model, with details about the victim included on one level and model estimates and significance tests based on the total sample of victims. Offences constitute a second (lower) level, with each victim able to provide detailed information on up to six separate incidents during their interview. Offence and offender details were included at this level, with model estimates and significance tests based on this larger sample of incidents.⁴ Importantly, the approach also explicitly recognises that all incidents that are experienced by the same victim will not be independent of one another. Instead, the emotional reactions to each incident are likely to share a common basis, with estimates further adjusted to reflect this statistical dependency.

Method

We used data from the CSEW between 2012/13 and 2019/20 (ONS, 2021) to assess whether victims of violent crime and vandalism reported more substantial emotional impacts when the incidents were perceived to be motivated by hate. Distinguishing between the main types of hate crime,⁵ we estimated multilevel regression models to disentangle the emotional reactions attributable to the form of hate crime from other offences and victim characteristics commonly linked to adverse emotional effects. We also examined whether the emotional impacts of hate crime have changed during this 8-year window.

Data

The CSEW is a nationally representative victimisation survey fielded to approximately 35,000 residents of England and Wales each year with response rates between 70% and 75% since 2001 (Kantar Public, 2021). We restricted our focus to victims of vandalism (covering arson, damage to a vehicle or home, accidental/nuisance and attempted damage), a comparatively ‘low-level’ form of victimisation that is experienced by a comparatively large number of individuals, and violent crime (covering serious assault, common assault, other wounding, other assault, attempted assault, threat to kill/assault), where the victim was likely to be able to provide detail on the offenders’ characteristics.

Despite the large annual sample size, reported instances of violent victimisation and vandalism are still comparatively rare in the CSEW, and only approximately 15% of these are identified by the victim as being hate-motivated. To ensure a sufficient sample of hate crime victims for our empirical models, we therefore combined data from eight rounds of the survey, resulting in a final analytic sample of 7423 individuals that had experienced a total of 8290 violent victimisations, and 11,631 individuals that had experienced a total of 12,426 vandalism incidents.

Emotional impact of victimisation

To measure the emotional impact of victimisation, respondents were first asked: ‘Many people have emotional reactions after incidents in which they are victims of crime. Looking at this card did you PERSONALLY have any of these reactions after the incident? (Anger, Shock, Fear, Depression, Anxiety/panic attacks, Loss of confidence/feeling vulnerable, Difficulty sleeping, Crying/tears, Annoyance, Other)’, before being asked, ‘Overall, how much were you affected? (just a little, quite a lot, “very much”. We generated an ordinal scale distinguishing people that reported ‘not being affected emotionally’ (0), from those affected ‘just a little’ (1), ‘quite a lot’ (2), and ‘very much’ (3). We treated this as a linear measure in our reported empirical models. Results were consistent when modelled as ordinal (available on request). The type of reaction was further used to identify more serious (shock, fear, depression, anxiety/panic attacks, loss of confidence/feeling vulnerable, difficulty sleeping, crying/tears) and most severe (depression, panic attacks, loss of sleep) reactions.

Hate-motivated crime

Victims of crime were asked a series of follow-up items designed to determine whether they perceived the incident to have been motivated by their identity. That is, towards their race/ethnicity, religion or religious beliefs, sexuality or sexual orientation, age, sex, or disability.⁶ We created a series of binary indicators distinguishing those offences that people thought were hate-motivated (where a respondent was not sure, these were coded as non-hate-motivated). Only a small proportion of violent and vandalism incidents were identified by victims as having a hate motivation across the study window – typically fewer than 5% of incidents each year, with race, age and gender the most common forms of hate motivator reported.

Offence characteristics

We included a categorical indicator to distinguish between the range of different incident types and levels of seriousness included under each crime classification (vandalism or violence). The number of incidents reported by each victim was also identified, as well as whether the incident is believed to be part of a series of related events, providing a direct estimate of the potential cumulative impact of repeated victimisation (Farrell, 1995). For violent offences we also recorded details of the timing of the incident (distinguishing between an offence occurring at night, dusk and during the day), where it occurred and whether a weapon was used.

Offender details

Victims of violence were also asked to provide details of the perpetrators involved in each incident. We recorded both the number of perpetrators involved (distinguishing one, two, three, four or more) as well as their connection to the victim (stranger, acquaintance, well-known). In addition, we also flag those incidents that involved perpetrators that were believed to be acting under the influence of drugs and/or alcohol, as these are frequently connected to heightened emotional reactions from victims (Fetzer and Pezzella, 2019).

Victim characteristics

Finally, we also included basic socio-demographic information about each victim. This covered the victim's gender, race/ethnicity (White, Black, Asian, Mixed, Other), age (in years) and the number of years they had lived at their address.

Analytic strategy

We estimate multilevel regression models to correctly account for the nested structure of our victimisation data – with some victims reporting more than one victimisation (mean number of incidents 1.1, ranging from 1 to 3). The full model has the following form.

y_{i j} = β_{0} + H a t e_{i j}^{'} β + H a t e_{i j}^{'} * Y e a r_{j} β + x_{i j}^{'} β + z_{j}^{'} β + u_{j} + e_{i j}

where $y_{i j}$ is the reported emotional impact of incident i on victim j, with victims and incidents assumed independent and identically distributed with variances $σ_{u}^{2}$ and $σ_{e}^{2}$ . We include a vector of binary hate indicators ( $H a t e_{i j}^{'} β$ – covering race/ethnicity, religion, sexuality, age, gender, and disability) enabling us to estimate the differential emotional impact of hate-specific incidents. Each hate indicator was further interacted with year, capturing any linear change in the emotional impacts of hate incidents over the duration of our study window. $x_{i j}^{'} β$ is a vector of offender and offence details and $z_{j}^{'} β$ is the list of victim characteristics, allowing us to disentangle the emotional effects of hate crime from other salient features of the incident and the victim that may influence their emotional reaction. Models were also estimated separately for each form of hate crime with (and without) the full list of included control⁷ variables. Our substantive findings remained unchanged when using these similar model specifications; therefore, we focus below on the results from the full models.

Sufficient numbers were available to examine the emotional impact of being the victim of two hate offences, vandalism and violence. For both forms of victimisation, three models were estimated. The first included all negative emotional reactions (see Table 1, Models 1 and 4); the second was re-estimated restricted to those individuals who reported experiencing a more serious emotional reaction (omitting those who only felt anger or annoyance; Table 1, Models 2 and 5); and the third was re-estimated for those reporting the most severe form of emotional reaction (covering depression, panic attacks, or loss of sleep; Table 1, Models 3 and 6).

Table 1.

Full results from multilevel models examining harms of hate crime compared to equivalent non-hate crimes.

	Vandalism			Violence
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)
Constant	0.052	0.516***	0.887***	0.634***	0.693***	0.944***
	(0.082)	(0.103)	(0.144)	(0.076)	(0.076)	(0.102)
Racially motivated hate	0.490***	0.251***	0.139	0.332***	0.328***	0.308***
	(0.068)	(0.079)	(0.107)	(0.051)	(0.051)	(0.077)
Religiously motivated hate	−0.13	−0.232	−0.472**	−0.141	0.011	0.097
	(0.122)	(0.151)	(0.189)	(0.086)	(0.083)	(0.130)
Sexual orientation-motivated hate	0.385	0.416	0.312	0.422***	0.300***	0.239**
	(0.226)	(0.223)	(0.231)	(0.092)	(0.082)	(0.105)
Age-motivated hate	0.308***	0.101	0.061	0.096	0.06	−0.024
	(0.079)	(0.092)	(0.109)	(0.059)	(0.059)	(0.083)
Gender-motivated hate	0.559***	0.293***	0.301**	0.159***	0.088	0.07
	(0.095)	(0.094)	(0.117)	(0.055)	(0.051)	(0.072)
Disability-motivated hate	0.662***	0.382***	0.190	0.259***	0.098	−0.086
	(0.099)	(0.102)	(0.113)	(0.081)	(0.073)	(0.083)
Year	0.008**	0.018***	0.026***	0.010**	0.001	−0.003
	(0.003)	(0.006)	(0.009)	(0.005)	(0.005)	(0.007)
Racially motivated × year	−0.028	−0.070**	−0.047	−0.025	−0.031	−0.061*
	(0.027)	(0.031)	(0.039)	(0.021)	(0.020)	(0.032)
Religiously motivated × year	0.025	0.019	0.058	0.005	0.025	0.082
	(0.054)	(0.069)	(0.086)	(0.036)	(0.034)	(0.052)
Sexual orientation motivated × year	−0.001	−0.043	−0.061	−0.023	0.003	0.026
	(0.093)	(0.092)	(0.092)	(0.037)	(0.032)	(0.042)
Age-motivated × year	0.004	0.049	0.044	−0.036	−0.058**	−0.051
	(0.037)	(0.043)	(0.049)	(0.025)	(0.026)	(0.036)
Gender-motivated × year	−0.035	−0.043	−0.073	0.029	0.045**	−0.014
	(0.041)	(0.040)	(0.052)	(0.023)	(0.021)	(0.031)
Disability-motivated × year	0.015	−0.022	−0.042	0.065*	0.046	0.032
	(0.042)	(0.043)	(0.046)	(0.035)	(0.032)	(0.036)
Female	0.103***	0.118***	0.133***	0.361***	0.227***	0.216***
	(0.016)	(0.028)	(0.045)	(0.022)	(0.023)	(0.035)
Ethnicity (ref: White)
Mixed ethnicity	0.03	0.054	0.038	0.149	0.099	0.082
	(0.067)	(0.103)	(0.151)	(0.079)	(0.077)	(0.115)
Asian	0.032	0.166***	0.195**	0.354***	0.146***	0.178**
	(0.037)	(0.058)	(0.095)	(0.055)	(0.051)	(0.077)
Black	0.237***	0.086	0.148	0.152**	0.061	0.013
	(0.055)	(0.076)	(0.134)	(0.065)	(0.063)	(0.093)
Chinese/other	0.094	0.283**	0.264	−0.035	−0.024	0.008
	(0.086)	(0.133)	(0.200)	(0.110)	(0.111)	(0.157)
Age (10-year intervals)	0.031***	0.040***	0.023	0.064***	0.069***	0.046***
	(0.005)	(0.009)	(0.015)	(0.008)	(0.008)	(0.012)
Length of residence	−0.012*	−0.012	−0.015	−0.033***	−0.019**	−0.00002
	(0.006)	(0.011)	(0.016)	(0.008)	(0.008)	(0.011)
Number of incidents	0.064***	0.069***	0.055***	−0.017	0.067***	0.057**
	(0.008)	(0.012)	(0.017)	(0.018)	(0.021)	(0.027)
Serious incident	0.213***	0.157***	0.083*	0.098***	0.122***	0.098***
	(0.020)	(0.031)	(0.043)	(0.024)	(0.025)	(0.033)
Offence (ref: Arson)
Damage to vehicle (<£20)	−0.583***	−0.462***	−0.286
	(0.082)	(0.116)	(0.179)
Damage to vehicle (>£20)	−0.295***	−0.216**	−0.188
	(0.076)	(0.088)	(0.118)
Damage to home (<£20)	−0.574***	−0.378***	−0.304**
	(0.080)	(0.100)	(0.136)
Damage to home (>£20)	−0.193**	−0.142	−0.137
	(0.078)	(0.090)	(0.119)
Other (<£20)	−0.415***	−0.470***	−0.314
	(0.117)	(0.155)	(0.242)
Other (>£20)	−0.123	−0.11	0.024
	(0.095)	(0.114)	(0.158)
Accidental damage/nuisance	−0.532***	−0.323***	−0.213
	(0.077)	(0.092)	(0.125)
Attempted damage	−0.309*	0.084	0.269
	(0.159)	(0.206)	(0.286)
Offence (ref: Serious assault)
Other wounding				−0.411***	−0.295***	−0.214***
				(0.058)	(0.054)	(0.066)
Common assault				−0.737***	−0.499***	−0.361***
				(0.052)	(0.049)	(0.061)
Other assault				−0.736***	−0.270	−0.186
				(0.126)	(0.138)	(0.233)
Attempted assault				−0.765***	−0.572***	−0.404***
				(0.063)	(0.061)	(0.083)
Threat to kill				−0.664***	−0.490***	−0.362***
				(0.052)	(0.049)	(0.060)
Time of day (ref: Day)
Dark				−0.037	−0.019	−0.002
				(0.023)	(0.024)	(0.033)
Dawn/Dusk				0.004	−0.016	−0.005
				(0.047)	(0.048)	(0.066)
Offence location (ref: Home)
Work				−0.521***	−0.377***	−0.256***
				(0.030)	(0.031)	(0.048)
Public				−0.204***	−0.178***	−0.090***
				(0.025)	(0.025)	(0.034)
Number of offenders (ref: 1)
2				0.120***	0.147***	0.095
				(0.033)	(0.035)	(0.051)
3				0.047	0.077	0.191**
				(0.050)	(0.053)	(0.084)
4 or more				0.133***	0.155***	0.122**
				(0.035)	(0.037)	(0.055)
Relationship to offender (ref: Stranger)
Casual acquaintance				0.124***	0.164***	0.099**
				(0.026)	(0.028)	(0.043)
Knew well				0.304***	0.336***	0.204***
				(0.028)	(0.030)	(0.045)
Offender drunk (ref: Yes)
No				0.059**	0.021	−0.048
				(0.026)	(0.027)	(0.037)
Unsure				0.078**	0.014	−0.07
				(0.034)	(0.035)	(0.051)
Offender on drugs (ref: Yes)
No				−0.131***	−0.130***	−0.062
				(0.028)	(0.029)	(0.040)
Unsure				−0.052*	−0.070**	0.026
				(0.030)	(0.031)	(0.045)
Offender had a weapon				0.176***	0.128***	0.095*
				(0.036)	(0.036)	(0.052)

Incidents	12,426	3282	1305	8290	5154	2226
Victims	11,631	3142	1243	7423	4785	2082
Incident variance	0.360	0.278	0.239	0.271	0.221	0.397
Victim variance	0.353	0.276	0.269	0.256	0.256	0.407

Note: *p < .05, ** p < .05, *** p < .01.

Results

The main results for vandalism can be seen in Figure 1 (tabular results in Table 1, Models 1–3). Here the vertical dashed line represents the average emotional reaction to an incident, controlling for the full list of victim, incident and offender characteristics, with the dots (and associated 95% confidence intervals) indicating the difference in emotional reaction for a particular group. Those effects where the confidence intervals do not cross the dashed vertical line were statistically significant. For example, when considering all emotional reactions (black dots), victims who perceived the incident to be racially motivated had a stronger negative emotional response than those who did not perceive the incident as racially motivated (coefficient value of 0.490 and statistically significant at the p < .001 level, see Table 1). This suggests that those who experienced racially motivated vandalism were significantly more likely to score higher on the measure of emotional distress controlling for victim characteristics (e.g., age, gender, ethnicity) and offence characteristics (e.g., type of vandalism, seriousness, number of incidents). This remained true when restricted to those who experienced more serious emotional reactions (dark grey dots), but not when limited to those who experienced the most severe emotional reactions (light grey dots). However, the magnitude of the B-value of this analysis suggests that the lack of significant finding may be the result of a small sample size rather than a lack of effect.

Figure 1.

The coefficients for the estimates of the emotional impacts of hate vandalism compared to non-hate vandalism.

There was strong evidence that those who experienced gender-motivated vandalism scored significantly higher on the measures of emotional reaction (at all levels) than those who experienced a similar non-hate-related offence. This was also the case for disability-motivated vandalism.⁸ Age-motivated vandalism was associated with significantly higher scores on the measure of overall emotional reaction only (Model 1), and whilst the scores were also higher for sexual orientation-motivated vandalism, the small sample of victims meant this did not reach conventional levels of significance.

Interestingly, religiously motivated vandalism was not associated with significantly higher scores on the measure of emotional reaction. In fact, those who experienced religiously motivated vandalism had significantly reduced severe emotional reactions when compared to those who experienced a similar non-hate-related vandalism incident (Model 3).

Table 1 (first three columns) shows that, of all forms of vandalism, arson resulted in the greatest emotional impact (particularly when compared to minor damage to home and accidental damage), and an increased number of incidents were related to greater emotional impacts. There was also evidence that females and those who were older experienced greater emotional impacts. Those of Asian, Black and Chinese (other) ethnicity also reported greater emotional impacts, although the comparatively small sample sizes mean these effects were only significant in some of the models.

There was also evidence of a small, but significant increase in emotional impacts over the years, for both hate and non-hate vandalism (‘Year’ in Table 1). However, the interaction between hate-motivated vandalism (racially motivated, religiously motivated, sexual orientation motivated, age motivated, gender motivated, disability motivated) and year was not significant. This suggests that the emotional impact of these offences had not changed (compared to non-hate offences) over the years.⁹

Figure 2 shows the results for the emotional impacts of violence victimisation (tabular results in Table 1, Models 4–6). For racially motivated and sexual orientation violence, there was evidence of a significantly increased emotional impact across all three models. For gender- and disability-motivated violence, this was the case for overall emotional impacts (Model 4).

Figure 2.

The coefficients for the estimates of the emotional impact of hate violence compared to non-hate violence.

There was some limited evidence that the overall emotional impact of violence victimisation (Table 1, Model 4, ‘Years’) had increased over the years, but the interaction effects between hate-motivated violence victimisation and year produced null (religiously motivated, sexual orientation motivated), significantly positive (gender motivated) and significantly negative (age motivated) results. This suggests that the negative emotional impacts of some forms of hate victimisation have remained constant, others have worsened, and others have improved from 2012 to 2020.

Experiencing serious violence resulted in greater negative emotional impacts for all models compared to other forms of violence (other wounding, common assault, other assault, attempted assault and threat to kill), as did experiencing an increased number of incidents (this was true both for series incidents that were related to the identified violent crime, and for other, unrelated incidents).

As with vandalism, females scored higher on the measure of emotional impact of violent victimisation, as did those who were older and those of Asian ethnicity. There was also evidence that those who were of Mixed and Black ethnicity scored higher on overall emotional impact to violent victimisation when compared to White victims.

There was no relationship evident between the time of day of the violent victimisation and the emotional impact, but being violently victimised in the home, and being victimised by more than one person resulted in greater emotional impacts. Also, being violently victimised by a casual acquaintance or by a perpetrator who the victim knew well resulted in a greater emotional impact compared to being victimised by a stranger. Not surprisingly, if the perpetrator had a weapon, greater emotional impacts were evident.

There were inconsistent findings about the relationship between the emotional impact of violence victimisation and whether the perpetrator was intoxicated. If the victim thought that offender was not drunk (or was unsure), this was associated with greater emotional impacts overall (Model 4). However, if the victim thought the perpetrator was on drugs (or was unsure), this was associated with greater emotional impacts.

Discussion

It is, of course, essential that the results of this research study are interpreted appropriately. This research utilised 8 years of nationally representative data and a robust methodology to isolate the unique impact of hate vandalism and violence in the aggregate. This can be illustrated by using one of the more eye-catching findings, namely that those who experienced religiously motivated vandalism hate offences did not ‘hurt more’ than those who experienced the equivalent crime without the bias element; in fact, if anything, they hurt less. While this indicates that, on average, those who experienced religiously motivated hate-related vandalism were emotionally better off (and this difference cannot be explained by unique characteristics of the offence, perpetrator or situation), this certainly does not mean that any individual who experiences this form of hate offence will not be significantly traumatised and require significant psychological support. Nor does it mean that these results are necessarily replicated across other types of hate offences, such as the so-called lower-level ones that are known to cause significant emotional trauma but are not part of the current investigation (Hall, 2013; Paterson et al., 2019a; Williams and Tregidga, 2014).

This finding may be explained by the fact that certain religious communities may have longer-standing networks and organisations that have developed robust and nuanced ways of supporting victims over long periods of time (such as the Community Security Trust, which has been established for decades and which focuses on supporting victims of antisemitism). These religiously oriented networks and organisations’ long experience of helping hate victims may help mitigate the impact of victimisation for their communities in a way that victim groups with less established support mechanisms may struggle to replicate. In addition, it may be that the rise in Islamophobia post-9/11 and 7/7 and the long history of antisemitism may have caused the victimisation of those religious communities to become more normalised, building up higher levels of resilience within them. This is not to say that hate victimisation has not become normalised among other communities too, nor that these communities do not have significant resilience within them. However, the results suggest that the emotional reaction of those who experienced religious hate crime appeared different.

The results also provide a firmer empirical grounding to assess existing assumptions about the emotional impacts of different hate crimes on individuals. That is, some hate offences really do hurt more, and this is particularly the case when considering racist and homophobic violence and gender- and disability-related vandalism. This is important in the context of the challenge to the notion of hate crime evident in the last few years (Williams et al., 2023). The results of this study support the allocation of significant resources to tackling hate by the criminal justice system – and that such incidents and crimes have greater negative emotional impacts than their non-hate equivalents – even though they make up a relatively small overall proportion of offences. The results also undermine arguments of those in certain sections of the press and of ‘anti-hate crime’ campaigners – that the harms of hate are exaggerated and do not justify the specific policies and practices surrounding them.

It is important to note, however, in some instances the additional magnitude of this hurt is quite modest. In particular, when attention is turned to the most severe forms of emotional reactions to crime (including depression, panic attacks, or loss of sleep), hate victims are often indistinguishable from victims of crimes that were not judged to be hate-motivated. Instead, it was a person's gender, age and (in some cases) their race/ethnicity that was associated with increased emotional reactions to crime. This distinction is important. Women experienced greater emotional impacts of victimisation irrespective of whether they perceived the incident itself to be motivated by their gender (or any other protected category). Similarly, older people and those from certain minority groups experienced greater emotional distress, independent of any impacts associated with the incident being defined as a hate crime. This important result may go some way to explaining the prior inconsistent findings and highlights the importance of adequately matching victims of hate crime with equivalent victims of crimes not motivated by hate. This is because when account is properly taken of individual characteristics, some of the harms previously attributed to the hate motivation itself might better be linked with other prior-held vulnerabilities.

It is perhaps not surprising that those possessing protected characteristics have more negative emotional reactions to victimisation (both hate and non-hate). These individuals inhabit a space in society replete with structural discrimination and daily invalidating experiences (Ching et al., 2018). Meyer (2003) introduced Minority Stress Theory, which suggests that those with protected characteristics suffer ongoing forms of discrimination and racism which are maintained by cultural norms and stable societal structures. This fosters both the expectation that these experiences will likely continue and the internalisation of the prevailing prejudice discrimination and stereotypes which have significant negative impacts on self-esteem and psychological health (Ching et al., 2018).

Another potential explanation for the observed results may be related to the concept of victim sensitivity. Victim sensitivity can be defined as an individual characteristic in which individuals harbour a latent fear of being exploited and are chronically hypersensitive to cues associated with untrustworthiness (e.g., Gollwitzer et al., 2015). As hate crime was defined by the victim, it could be that those with higher victim sensitivity may be both more likely to perceive an offence as a hate crime and experience stronger emotional impacts of victimisation. This may mean that the reason that hate crimes appear to hurt more in this study is because those who perceive victimisation as hurting more are also more likely to perceive their victimisation as motivated by hate.

Not surprisingly, the severity and circumstances of the incident are also crucial. Our results identified consistently greater emotional impacts when more serious incidents of vandalism and violence were considered and when victims were subject to multiple incidents in a comparatively short timeframe. Past research (e.g., Fetzer and Pezzella, 2019) which shows that hate crimes tend to be more harmful has tended to package this result as the impact of a hate offence, when it appears to be independent.

There was also evidence that the emotional impacts of gender- and disability-motivated violence may be increasing over time. At the same time, the added emotional impacts of race- and age-related violence may actually be subsiding. By contrast, whilst we observed a modest increase in the impacts of vandalism over time, there were no consistent differences in time trends across hate and non-hate victims. These effects were all modest in scale, and more focused research is needed to definitively verify this change.

Limitations

This study benefits from its scale and coverage, but it is not without its limitations. Difficulties surrounding such a ‘siloed’ analysis of the emotional impacts of hate crime, where different groups are analysed separately from one another in a way that underplays the importance of intersectionality between those groups, need to be noted and examined in future research. Additionally, even combining 8 years of survey data, we are still limited in our ability to accurately identify the effects of those forms of hate crime that are only experienced by a small proportion of the population (in particular, the effects of hate offences motivated by sexual orientation and religion were measured with greater uncertainty). The small samples are represented in the wide confidence intervals of some of our key findings, meaning that replication of this work is needed.

In addition, many of the most vulnerable in society, including those in managed care and the homeless, may not be included in the sampling frame (Garside, 2015). Given that these same individuals are also subject to the forms of hate-motivated offending that this study engages with, we must expect that our results may represent a lower-bound estimate of the true impacts of these forms of hate.

The small sample sizes also ruled out the possibility of examining the emotional impacts of anti-transgender hate crimes. We are mindful of the importance of examining the emotional impact of victimisation on this minoritised group in particular, but perhaps as was the case with other protected groups, qualitative approaches would be best to begin to further develop the evidence base here.

Finally, for practical reasons we limited our focus to two forms of victimisation, namely violence and vandalism, where the survey respondents were able to provide sufficient information about potential hate motivation. The extent to which these trends would be observed for other crime categories remains to be seen. Similarly, we are also unable to say anything concrete about the many forms of non-crime hate incidents that are likely to be the primary reference point for many vulnerable members of society subjected to hate. However, by definition, there is no suitable comparison group to adequately examine this, as only those with protected characteristics can experience these incidents. This poses a complex methodological challenge and makes it difficult to envisage how future studies might usefully address this omission without falling into the same trap as past hate crime studies. That is, failing to effectively separate the effects of the incident itself from prior-held vulnerabilities or other features of the incident.

Conclusion

It appears that hate crime does hurt more, but not always and not equally amongst those with certain protected characteristics. However, these methodologically robust results are a single contribution, with the results used to inform, but not alter, how practitioners and legal advocates support those victimised. In particular, we have shown that the assumption that hate crime really does hurt more is far from axiomatic. Whilst there is some evidence that some forms of hate crime do elicit a stronger emotional reaction from victims than equivalent offences not motivated by hate, in other instances the emotional response remains unchanged, and in some circumstances, victims of hate offences actually appeared to cope better. Future research should examine this assertion, and if so, aim to uncover the mechanisms as to why those who experience religiously motivated hate offences appear to be less affected.

Ultimately, it seems that the harms of hate do exist, but these are nuanced. Much more research is needed to understand the complexities associated with these harms with the ultimate aim of mitigating their potential considerable negative consequence.

Footnotes

Funding

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

ORCID iDs

Ian Brunton-Smith

Darrick Jolliffe

Jon Garland

Notes

References

Awan

Zempi

(2020) “You all look the same”: Non-Muslim men who suffer Islamophobic hate crime in the post-Brexit era. European Journal of Criminology 17(5): 585–602.

Barnes

Ephross

(1994) The impact of hate violence on victims: Emotional and behavioral responses to attacks. Social Work 39(3): 247–251.

Bell

Perry

(2015) Outside looking in: The community impacts of anti-lesbian, gay, and bisexual hate crime. Journal of Homosexuality 62(1): 98–120.

Chakraborti

Garland

(2015) Hate Crime: Impact, Causes and Consequences, 2nd ed. London: Sage.

Chakraborti

Garland

Hardy

(2014) Leicester Hate Crime Project: Findings and Conclusions. Leicester: University of Leicester.

Ching

THW

Lee SY Chen

, et al. (2018) A model of intersectional stress and trauma in Asian American sexual and gender minorities. Psychology of Violence 8(6): 657–668.

Corcoran

Smith

(2016) Hate Crime, England and Wales, 2015/16, Statistical Bulletin 11/16. London: Home Office.

Crown Prosecution Service (2016) Hate Crime: What It Is and How to Support Victims and Witnesses. London: CPS.

Díaz-Faes

Pereda

(2022) Is there such a thing as a hate crime paradigm? An integrative review of bias-motivated violent victimization and offending, its effects and underlying mechanisms. Trauma, Violence, & Abuse, 23(3): 938–952.

10.

Farrell

(1995) Preventing repeat victimization. Crime and Justice 19: 469–534.

11.

Fetzer

Pezzella

(2019) The nature of bias crime injuries: A comparative analysis of physical and psychological victimization effects. Journal of Interpersonal Violence 34(18): 3864–3887.

12.

Gallagher

(2023) A Year of Hate: Anti-Drag Mobilisation Efforts Targeting LGBTQ+ People in the UK. London: Institute for Strategic Dialogue.

13.

Garside

(2015) Crime Is Up. Crime Is Down. What’s Going On? London: Centre for Crime and Justice Studies.

14.

Glass

(2021) Former police officer wins appeal over ‘hate incident’ guidance, The Independent, 20 December 2021. Available at: https://www.independent.co.uk/news/uk/crime/humberside-police-college-of-policing-court-of-appeal-high-court-lincolnshire-b1979455.html (accessed 29 July 2022).

15.

Gollwitzer

Süssenbach

Hannuschke

(2015) Victimization experiences and the stabilization of victim sensitivity. Frontiers in Psychology 6: 439.

16.

Hall

(2013) Hate Crime, 2nd ed. London: Routledge.

17.

Hardy

Chakraborti

(2020) Blood, Threats and Fears: The Hidden Worlds of Hate Crime Victims. Basingstoke: Palgrave Macmillan.

18.

Herek

Gillis

Cogan

(1999) Psychological sequelae of hate-crime victimization among lesbian, gay, and bisexual adults. Journal of Consulting and Clinical Psychology 67(6): 945–951.

19.

Herek

Gillis

Cogan

Glunt

(1997) Hate crime victimization among lesbian, gay and bisexual adults: Prevalence, psychological correlates and methodological issues. Journal of Interpersonal Violence 12: 19–21.

20.

Home Office (2020) Hate Crime, England and Wales, 2019/20. London: Home Office.

21.

Home Office (2023) Hate Crime, England and Wales, 2022 to 2023. London: Home Office at Hate Crime. Available at: https://www.gov.uk/ (accessed 29 July 2024).

22.

Iganski

(2001) Hate crimes hurt more. American Behavioural Scientist 145: 626–638.

23.

Iganski

Lagou

(2009) How hate crimes hurt more: Evidence from the British Crime Survey. In: Iganski

(ed) The Consequences of Hate Crime. Westport, CT: Praeger, 1–13.

24.

Iganski

Lagou

(2014) The personal injuries of “hate crime”. In: Hall

Corb

Giannasi

Grieve

JGD

(eds) The Routledge International Handbook on Hate Crime. Abingdon: Routledge, 34–46.

25.

Kantar Public (2021) The Gold Standard in the Measurement of Crime: The Crime Survey for England and Wales. Available at: https://www.kantarpublic.com/our-work/national-statistics-crime-survey (accessed 2 September 2022).

26.

Lantz

Kim

(2019) Separating the influence of co-offending and bias on hate-motivated physical injury. Criminal Justice and Behavior 46(3): 437–456.

27.

Levin

(2009) The long arc of justice: Race, violence, and the emergence of hate crime law. In: Levin

(ed) Hate Crimes Volume 1: Understanding and Defining Hate Crime. Westport, CT: Praeger, 1–23.

28.

Mackie

Smith

(2015) Intergroup emotions. In: APA Handbook of Personality and Social Psychology. vol. 2 , Group processes. American Psychological Association, 263–293.

29.

Marais

Juan

Nel

, etal (2021) Emotional consequences of hate incidents: Experiences of a South African cohort. South African Journal of Psychology 52: 1–13.

30.

Mason-Bish

(2013) Examining the boundaries of hate crime policy: Considering age and gender. Criminal Justice Policy Review 24(3): 297–316.

31.

Mellgren

Andersson

Ivert

(2021) For whom does hate crime hurt more? A comparison of consequences of victimization across motives and crime types. Journal of Interpersonal Violence 36 (3–4): NP1512–NP1536.

32.

Meyer

(2003) Prejudice as stress: Conceptual and measurement problems. American Journal of Public Health 93(2): 262–265.

33.

Nagin

(2013) Deterrence: A review of the evidence by a criminologist for economists. The Annual Review of Economics 5: 83–105.

34.

ONS (2020) Crime in England and Wales: Year ending March 2020. Available at: https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/bulletins/crimeinenglandandwales/yearendingmarch2020 (accessed 2 March 2024).

35.

ONS (2021) Available at: https://www.gov.uk/government/statistics/hate-crime-england-and-wales-2020-to-2021/hate-crime-england-and-wales-2020-to-2021 (accessed 02 01 2022).

36.

ONS (2022) Crime in England and Wales: Year ending March 2022. Available at: https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/bulletins/crimeinenglandandwales/yearendingmarch2022 (accessed 2 March 2024).

37.

Paterson

Rupert

Walters

(2019b) Feeling for and as a group member: Understanding LGBT victimization via group-based empathy and intergroup emotions. British Journal of Social Psychology 58(1): 211–224.

38.

Paterson

Walters

Brown

(2019a) “Your pain is my pain”: Examining the community impacts of Islamophobic hate crimes. In: Awan

Zempi

(eds) The Routledge International Handbook of Islamophobia. London: Routledge, 84–95.

39.

Paterson

Walters

Brown

Fearn

(2018) The Sussex Hate Crime Project. Brighton: University of Sussex.

40.

Perry

(2001) In the Name of Hate: Understanding Hate Crimes. New York: Routledge.

41.

Perry

Alvi

(2012) ‘We are all vulnerable’: The In Terrorem effects of hate crimes. International Review of Victimology 18(1): 57–72.

42.

Shapland

Hall

(2007) What do we know about the effects of crime on victims? International Review of Victimology 14: 175–217.

43.

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: Home Office.

44.

Sodha

(2022) It’s Not the Police’s Job to Shut Down Political Debate. They Should Stick to Solving Crime. Available at: https://www.theguardian.com/commentisfree/2022/jan/02/policing-sex-and-gender-debate-worrying-freedom-of-expression?CMP=Share_iOSApp_Other (accessed 29 July 2022).

45.

von Hirsch

(2007) The “desert” model for sentencing: Its influence, prospects, and alternatives. Social Research: An International Quarterly 74(2): 413–434.

46.

Walters

(2022) Criminalising Hate Crime: Law as Social Justice Liberalism. Basingstoke: Palgrave Macmillan.

47.

Walters

Paterson

McDonnell

, et al. (2019) Group identity, empathy and shared suffering: Understanding the ‘community’ impacts of anti-LGBT and Islamophobic hate crimes. International Review of Victimology 26(2): 143–162.

48.

Wilkin

(2020) Disability Hate Crime: Experiences of Everyday Hostility on Public Transport. Basingstoke: Palgrave Macmillan.

49.

Williams

de Reya

(2019) Hate Behind the Screens: A Report on the Rise of Online Hate Speech. Cardiff: Cardiff University/Mishcon Academy.

50.

Williams

Tregidga

(2013) All Wales Hate Crime Research Project: Research Overview and Executive Summary. Cardiff: Cardiff University.

51.

Williams

Tregidga

(2014) Hate crime victimization in Wales: Psychological and physical impacts across seven hate crime victim types. British Journal of Criminology 54(5): 946–967.

52.

Williams

Sutherland

Roy-Chowdhury

Loke

Cullen

, et al. (2023) The effect of the Brexit vote on the variation in race and religious hate crimes in England, Wales, Scotland and Northern Ireland. The British Journal of Criminology 63(4): 1003–1023.

53.

Zempi

Smith

(eds) (2021) Misogyny as Hate Crime. London: Routledge.

How does hate hurt more? National evidence for the varying emotional impacts of hate crime

Abstract

Keywords

Introduction

Background

Hate crime and the criminal justice system

Assessing the harms of hate crime

Method

Data

Emotional impact of victimisation

Hate-motivated crime

Offence characteristics

Offender details

Victim characteristics

Analytic strategy

Results

Discussion

Limitations

Conclusion

Footnotes

Funding

ORCID iDs

Notes

References