Understanding Public Attitudes to Hate: Developing and Testing a U.K. Version of the Hate Crime Beliefs Scale

Abstract

Against the backdrop of an increase in reported hate crimes, we present the development of a U.K.-focussed instrument designed to evaluate the nature of public beliefs about hate crime, legislation, offenders and victims. In Study 1, 438 participants completed an Anglicized version of the Hate Crime Beliefs Scale (HCBS). Factor analyses revealed three subfactors: Denial (high scores represent a denial of hate crime severity and need for legislation), Compassion (high score reflect compassion toward victims and affected communities) and Sentencing (higher scores reflect more punitive attitudes). In Study 2 (N = 134) we show that scores on Denial are positively associated with those on Right-Wing Authoritarianism (RWA) and Social Dominance Orientation (SDO), ideologies known to be associated with prejudice. Compassion was negatively associated with these ideologies. Mediation analyses showed that Big Five personality traits Openness to Experience and Conscientiousness predicted Denial and Compassion via RWA, whereas Agreeableness and Openness predicted scores via SDO, consistent with a dual-process motivation model of hate crime beliefs. Results are discussed in terms of the nature of hate crime beliefs and the importance of understanding public attitudes which may support undesirable social norms and influence jury decision making in trials of hate related offenses.

Keywords

hate crimes community violence cultural contexts

The term hate crime is used to describe a range of criminal behaviors where the perpetrator is motivated by hostility or prejudice toward protected characteristics of the victim. In the United Kingdom, these are currently disability, race, religion, sexual orientation, and transgender identity (College of Policing, 2014; for details on the history of hate crime in the United Kingdom and how these characteristics came to be protected, see Walters et al., 2017). During 2017 to 2018, police in England and Wales recorded 94,098 hate crime offenses, an increase of 17% compared with the previous year. This increase is thought to be largely driven by improvements in police recording, although there have been spikes in hate crime following events such as the European Union (EU) Referendum and the terrorist attacks in 2017 (Home Office, 2018). Race was identified as a motivating factor for more than three quarters of recorded cases (76%), followed by sexual orientation (12%), religion (9%), disability (8%), and transgender identity (2%). The British Crime Survey, which accounts for experiences of victimization not reported to the police, indicates that around 184,000 incidents took place in the same period (Home Office, 2018).

Targets of hate crimes are often emotionally affected, experiencing loss of confidence, vulnerability, fear, difficultly sleeping, anxiety or panic attacks, or depression (Home Office, 2018). The trauma of witnessing hate crime can also have damaging effects on family members, particularly children (Williams & Tregidga, 2014; Zempi & Chakraborti, 2014) and minorities who have not experienced hate crime restructure their daily lives to avoid putting themselves at risk. In the wider community, hate crime can damage social cohesion as groups distance themselves from one another. In recognition of these impacts, the Criminal Justice Act (2003) introduced specific sentencing provisions allowing for “uplifts” (e.g., longer prison sentences), for those found guilty of hate crimes.

In this research, we are concerned with public perceptions of hate crime and the associated legislation. Understanding this is vital, not least because of the damaging impacts of hate crime for individuals and for wider society, but also because members of the general public sit on juries when crimes, including hate crimes, are tried in court. The way that communities respond to hate crimes may also impact how such incidents are dealt with and police can also be influenced by their own opinions (Grattet & Jenness, 2008). Addressing hate crime requires combatting public beliefs and social norms that support the perpetuation of hate behaviors. Social norm theory describes how behavior is influenced by what individuals see or hear of others doing (Cialdini & Goldstein, 2004; Miller & Prentice, 2016; Wright et al., 1997). A wealth of research has shown that the perception of what is normal behavior within a given society or social group triggers and guides action (see Chung & Rimal, 2016, for review). Norm-based behavioral interventions are based in the assumption that people are unaware of social norms and their relationship to them. For instance, individuals are often consciously unaware of norms or are mistaken about their relation to them, frequently overestimating the prevalence of many undesirable behaviors. That perception is used as a standard against which to compare their own behaviors. Providing people with information about the behavior and attitudes of their peers is a strategy commonly employed as a means to changing behaviors considered harmful to individuals (e.g., health behaviors) or to society (e.g., environmentally friendly behaviors; e.g., Michie et al., 2011; Miller & Prentice, 2016). Receiving feedback as to whether they are in the majority or in the minority compared with peers can shape an individual’s perception of their social group and the evaluative significance of their behavior (Miller & Prentice, 2016). It is sometimes claimed that the introduction of antidiscrimination legislation and awareness of prejudice as less socially acceptable has meant that hate beliefs and bias are becoming increasingly covert, subtle, and difficult to detect (Hodson et al., 2005). However, the rising prevalence of hate crime defies this perception and suggests that prejudice against some groups is still embedded within social norms in the United Kingdom and elsewhere.

One powerful influence on social norms is media. The ever-growing prevalence of social media and 24-hr news has meant that the public are more informed about crime (including hate crime) than ever before, whether or not they have experienced it themselves. The U.K. press in particular gives significant attention to hate incidents and Warren-Gordon (2018) concluded that stringent laws regarding hate crimes might have contributed to the greater amount of print media coverage in this country. Chakraborty (2014) has also highlighted that differences in terms of reported hate crimes across Europe are partly because of the way in which hate crimes are defined and publicized in different countries. Media coverage of high-profile cases can have an immediate and sustained impact on the prevalent public values, attitudes, and behaviors (McCombs, 2005; Shah et al., 1996). Newspapers consistently reflect the culture and societal values of a community in their reportage, and, consequently, the media representation of social problems has a direct impact on society’s perception of them. What is more, the nature of media accounts may in itself inflame public attitudes toward particular groups. For instance, in the United Kingdom, religious-based offenses (often intersecting with racially motivated behaviors) are the most highly reported form of hate crime and Muslim adults are the most likely to be a victim (Home Office, 2018). Responding to the release of these statistics, Dearden (2018) in The Independent described incidents occurring in Manchester, London Bridge, and Parsons Green as “Islamist atrocities.” A media which primarily views Muslims through a lens of terrorism and security risk can inflame an increased prevalence of anti-Muslim hate crimes in the immediate aftermath of terrorist incidents (Githens-Mazer & Lambert, 2010). Media articles rarely highlight sectarian incidents perpetrated by non-Muslims in the same language. Overall, this can lead to the increasing stigmatization and isolation of U.K. inhabitants who happen to be Muslim (Githens-Mazer & Lambert, 2010).

Media accounts also influence views of hate crime perpetrators. The impression often conveyed is of hate-fuelled extremists who target their victims in premeditated attacks. However, there is evidence that many perpetrators are ordinary people who offend in the context of their daily life (Iganski, 2008). In these cases, the offense is often driven by everyday frustrations, often at a situation rather than an individual, and where race, sex, disability, or other characteristics form an easy target for the venting of annoyance or anger and may not be perceived as hate speech by the perpetrator, who may not understand the impact of their words. Mason (2005) further highlighted how many perpetrators of hate crime are already known to the victim, as neighbors, colleagues, or customers, and that a lack of consideration of this context obscures the everyday nature of hate for many individuals.

Eliminating hate crime in society is contingent on combatting public beliefs and attitudes that support social norms and hence the perpetuation of such crimes. Accordingly, a growing body of research aims to understand public attitudes concerning hate offenses and associated legislation, especially in the United States (e.g., Cramer et al., 2013; Mallett et al., 2011; Saucier et al., 2017). To this end, Cabeldue et al. (2018) developed the HCBS, a 40-item psychometric measure which assesses public beliefs across four subscales: Negative Views (i.e., higher scores reflect negative views of legislation and minority group protection), Offender Punishment (i.e., higher scores suggest endorsement of greater punishment), Deterrence (i.e., greater scores denote support for hate crime legislation as a deterrent of more violence), and Victim Harm (i.e., higher scores reflect provictim attitudes). The Negative Views subscale displayed predictive utility, such that more negative views of legislation/minority group protection were associated with elevated victim blame, as well as lower perpetrator blame and sentencing recommendations.

However, although being a valuable contribution to the research toolkit, some HCBS items are culturally specific to the United States where the scale was developed, for instance, several items refer to African Americans and one refers to the First Amendment. Cabeldue et al. (2018) acknowledge this limitation and state the need for further developments of the questionnaire outside the U.S. context. Furthermore, none of the items address the issue of religious intolerance against Muslims which has received a good deal of press coverage in the United Kingdom. In this article, we present the Hate Crime Beliefs Scale–U.K. version (HCBS-UK), an Anglicized version of Cabeldue et al.’s measure with additional items specifically addressing beliefs about Muslims. In Study 1, we discuss the development of the scale, the factor structure, and subscales. In addition, Cabeldue et al. presented evidence that liberal political beliefs were positively associated with provictim/legislation hate crime attitudes. In terms of the English political system, we therefore predicted that more left-wing political orientation would be positively associated with provictim/legislation beliefs. In Study 2, we test the construct validity of the scale by examining the relationship between scores and those on measures of Social Dominance Orientation (SDO) and Right-Wing Authoritarianism (RWA), factors consistently found to influence social prejudice (see Hodson & Dhont, 2015, for a review).

Study 1

Method

Participants

Four hundred and thirty-eight participants completed the study. Undergraduate students took part in return for course credit (N = 211; M_age = 20.79, SD = 4.14, range = 18–39), of which 186 (82%) were female, 22 male, three nonbinary, and one female-to-male transgender. One hundred and seventy-six (83%) described themselves as heterosexual, five as gay/lesbian, 24 as bisexual, and six as other. The majority were White, 205 (96%), three Black, two Asian, and one reported other. Most (172; 82%) described themselves having no religion, a further 29 declared themselves Christian, one Muslim, one Buddhist, and one reported other. Students’ home locations were spread across the United Kingdom, though the majority were from the Southwest of England (167; 79%) and a further 23 (10%) were from London/Southeast of England. In assessing Social–Economic Status (SES), we used the five-category Social Grade model, an occupation-based classification produced by the U.K. Office for National Statistics and which is used widely for market research in the United Kingdom. We asked that participants aged below 30 classify the home where they spent most of their childhood (i.e., their family background), whereas participants aged more than 30 rated their present household. One hundred and two (48%) students rated their background as Category A or B (higher and intermediate managerial, administrative, professional occupations), 38 (18%) from Category C (supervisory, clerical, junior managerial, skilled manual workers), and a similar number, 37, Category D (semi-skilled and unskilled manual occupations). Sixteen (7.6%) declared their background as Category E (long-term unemployed for whatever reason). In terms of political views, 100 (47%) described themselves as either slightly, moderately, or very left wing, with just 19 (9%) describing their views as right wing. Eighty-two students (39%) stated having no interest in politics and hence no political affiliation. Most students (150; 71%) had no experience of hate crime, whereas 12 (6%) perceived themselves to have been a victim and 49 (23%) knew someone who had been a victim.

General public (N = 227; M_age = 34.74, SD = 9.96, range = 18–68) were recruited through Prolific, a U.K.-based research participation website and paid £3. None of the participants in this sample were currently students. One hundred and seventy-two (76%) were female, 54 male, and one described themselves as nonbinary. The majority, 199 (88%), described themselves as heterosexual, six as homosexual/gay/lesbian, 14 as bisexual, two as asexual, and six as other. Two hundred (88%) defined themselves as White, five as Black, 11 as Asian, five as mixed race, and six as other. One hundred and thirty-two (58%) described themselves as having no religious affiliation, a further 75 (33%) described themselves as Christian, 10 Muslim, one Jewish, one Hindu, two Buddhist, and six as other. Participants were based across the United Kingdom, with the largest grouping in London/Southeast of England (58; 26%), followed by West/East Midlands and Eastern England regions. The least represented English region was the South West (13; 6%). Six participants were from Wales and five from Northern Ireland. In terms of SES, 112 (49%) of these participants rated themselves as being in Classes A and B, with 70 (31%) in Category C, 23 in Category D, and 22 from Category E. One hundred and twenty-five participants (55%) described their political stance as either slightly, moderately, or very left wing, whereas 61 (27%) defined themselves as right wing. None of this sample declared having no political interest at all. The majority (168; 74%) had no personal experience of hate crime, 24 (11%) perceived themselves to have been a victim, and 34 knew someone who had been a victim.

Development of HCBS-UK items

Cabeldue et al.’s (2018) original HCBS and their factor loadings for each item are shown in Table 2. For items which refer to African Americans, we retained the original wording other than to amend the term African American to Black, for example, their Item 24 becomes Offenders who target Black people based on their race deserve a longer prison sentence. Item 40 refers to the First Amendment. Here we changed the wording to I do not believe hate crime violates the right to freedom of speech or religion. Finally, Item 32 referred to prosecutors pursuing hate crime. We changed this to The police spend too much time pursuing hate crimes, to better reflect the U.K. law enforcement system. In addition, we added four new items with structures based in original items but, addressing beliefs about those of Muslim faith specifically, these were Offenders who target Muslims based on their religion deserve a harsher sentence; Offenders who target Muslims based on their race deserve a harsher sentence; Crimes against Muslims receive too much media attention; and Hate crime law protection of Muslims is unnecessary. In addition, Cabeldue et al.’s Items 12 and 29 state that hate crime law is unnecessary with regard to transgender and Black people, respectively. We added two further items mirroring this wording but with reference to other characteristics which seemed to be underrepresented in the original scale: Hate crime law protection of Jewish people is unnecessary and Hate crime law protection of people with disabilities is unnecessary. Overall, therefore, our final scale, the HCBS-UK, had 46 items. We used the Excel RAND function to generate a new randomized order for presenting these items to participants. All participants received items in this same order and also completed a demographics questionnaire.

Analysis

Confirmatory factor analysis (CFA) used the lavaan package (Rosseel, 2012) and semTools (Pornprasertmanit et al., 2016) within R 3.2.4 (R Core Team, 2015). We defined good fit as comparative fit index (CFI) and Tucker–Lewis index (TLI) > .90, standardized root mean square residual (SRMR) < .08, and root mean square error of approximation (RMSEA) < .05, and acceptable fit as approximating these levels (e.g., RMSEA < .10). We examined modification indices to check for items that were substantially cross-loading and see if correlating error terms within subscales further improved the model fit. We stopped when acceptable fit was obtained. We provide the data and analysis scripts at https://github.com/jon-may/HateCrime.

Results

We began by fitting the structure identified by Cabeldue et al. (2018) to the original 40 items. Although this model was better fitting than a unifactorial model, it was of borderline acceptability (see Table 1: four-factor model, 40 items). Modification indices suggested that nine items might fit better on different factors, but these changes compromised the factors’ identity. We therefore combined the Offender Punishment, Victim Harm, and Deterrence factors and seven Negative Beliefs items into two new factors, Compassion and Sentencing, and renamed the remaining Negative Beliefs factor Denial. With this structure, fit was improved and met criteria on all indices except TLI (Table 1: three-factor model, 40 items). The six novel items were then added, two to Sentencing and four to Denial, and, with a total of 12 pairs of error covariances correlating, fit met the SRMR and RMSEA criteria (Table 1: three-factor model, 36 items).

Table 1.

Robust Fit Indices.

Model	χ²	df	CFI	TLI	SRMR	RMSEA	AIC
Unifactorial 40 items (14)	2,047	726	.773	.756	.090	.070	40,024
Four-factor model, 40 items (4)	1,729	729	.828	.816	.082	.061	39,644
Three-factor model, 40 items (11)	1,218	726	.916	.909	.053	.042	39,052
Three-factor model, 46 items (9)	1,892	977	.886	.879	.053	.051	43,867
Three-factor model, 20 items (7)	299	159	.963	.955	.054	.052	17,993

Note. CFI = comparative fit index; TLI = Tucker–Lewis index; SRMR = standardized root mean square residual; RMSEA = root mean square error of approximation; AIC = Akaike’s information criterion.

To produce a shorter 20-item scale, we retained five Sentencing items, five Compassion items, and 10 Denial items with the highest item-subscale correlations. Fit criteria remained strong (Table 1: three-factor model, 20 items) and the subscales each had strong alpha coefficients (.90 for Denial, .76 for Compassion, and .89 for Sentencing). Table 2 presents the final 20 items and their factor loadings from Cabeldue et al. (2018) and from our analyses above.

Table 2.

The Final 20-Item HCBS-UK and Descriptive Statistics.

Item	Original HCBS Factor	Revised HCBS-UK Factor	ITC	M	SD
Offenders who target Muslims based on their race deserve a harsher sentence.	Offender Punishment	Sentencing	.77	3.4	0.97
A hate crime offender should receive a lengthier prison sentence.	Offender Punishment	Sentencing	.73	3.5	0.94
Offenders who target Muslims based on their religion deserve a harsher sentence.	Offender Punishment	Sentencing	.72	3.5	0.95
Offenders who target Black people based on their race deserve a longer prison sentence.	Offender Punishment	Sentencing	.70	3.4	0.97
Evidence of bias motivation in a crime should be an aggravating factor in sentencing.	Negative Beliefs	Sentencing	.66	3.6	0.83
Hate crime victims receive too much attention.	Negative Beliefs	Denial	.78	1.9	0.9
Hate crime law protection of Muslims is unnecessary.	Negative Beliefs	Denial	.75	1.6	0.89
The police spend too much time pursuing hate crimes.	Negative Beliefs	Denial	.73	2	0.89
Crimes against Muslims receive too much media attention.	Negative Beliefs	Denial	.72	2.1	1.02
Hate crime law protection of Jewish people is unnecessary.	Negative Beliefs	Denial	.72	1.7	0.85
Hate crime law protection of Black people is unnecessary.	Negative Beliefs	Denial	.70	1.6	0.84
Having to report crimes against transgender people is unnecessary.	Negative Beliefs	Denial	.70	1.6	0.74
I believe hate crimes receive too much attention.	Negative Beliefs	Denial	.69	1.9	0.88
The media makes hate crimes into a bigger deal than they actually are.	Negative Beliefs	Denial	.67	2.2	0.96
Charging someone with a separate hate crime charge is excessive prosecution.	Negative Beliefs	Denial	.67	2	0.93
Legislation including people with disabilities will discourage crimes against this group of people.	Deterrence	Compassion	.60	3.4	0.9
Crimes against people that are physically impaired are threatening to the entire community of people with disabilities.	Negative Beliefs	Compassion	.60	4	0.87
Sexual orientation bias–motivated crimes are threatening to the lesbian, gay, bisexual, transgender community at large.	Negative Beliefs	Compassion	.58	3.9	0.87
Harsh punishments of hate crime offenders will decrease the likelihood of future hate crimes.	Deterrence	Compassion	.53	3.5	0.93
Hate crime perpetrators cause psychological trauma to their victims.	Negative Beliefs	Compassion	.46	4.4	0.73

Note. Factor loadings are shown for this and the original scale (Cabeldue et al., 2018; Table 1). ITC is the correlation between the item and its subscale total. HCBS-UK = Hate Crime Beliefs Scale–U.K. version.

We predicted a positive association between left-wing political orientation and provictim/legislation attitudes in line with that reported by Cabeldue et al. (2018). We coded political beliefs such that a lower score indicated more left-wing attitudes and a higher score more right-wing attitudes. Bivariate correlations indicated only one significant association, with our Compassion scale (r = −.21, p < .001) suggesting that, in line with predictions, those with more right-wing attitudes were less compassionate toward victims and the harm that hate crime may cause. Correlations with the other two subscales were not significant (−.08 for Sentencing and .07 for Denial). The vast majority of our sample (99%) identified as either male or female. Comparing scores across these two groups, males (M = 2.23, SD = 0.79) scored more highly than females (M = 1.81, SD = 0.63) on the Denial subscale, t(431) = 4.83, p < .001. Males and females scored comparatively on both Compassion and Sentencing (p > .2 in both cases). Age was positively associated with both Denial (r = .23) and Compassion (r = .46; p < .001 in both cases). The majority (93%) of participants reported having no religion (n = 305) or being Christian (n = 103). Comparing scores across these two groups, Christians scored most highly on Compassion, t(406) = 3.19, p = .002, and also on Denial, though this did not quite reach significance (p = .06). The groups showed no difference on Sentencing (p > .03). There were too few other religions reported to include in the analysis. Similarly, 93% (405) of the participants reported as White, with too few in any other ethnic group to make the analysis viable. Finally, we coded experience of hate crime on a scale where 1 = no experience, 2 = know someone who has experienced, and 3 = experienced personally, and observed a modest though significant positive association (r = .12, p = .01) suggesting that those who had personal experience of hate crime were most likely to endorse harsh punishment.

Discussion

Our development of a new HCBS-UK resulted in a shorter, three-factor instrument, compared with that previously developed by Cabeldue et al. (2018) in a North American context. Our Sentencing factor comprised six items. Five of the items fell within Cabeldue et al.’s original Offender Punishment scale. The sixth item previously fell within their Negative Beliefs factor, though examination of this item (Evidence of bias motivation in a crime should be an aggravating factor in sentencing) shows that the meaning is clearly in line with the others in Offender Punishment and is reflecting a belief that hate-influenced offenses should attract a harsher sentence. As such, a higher score on these six items reflects the support for harsher punishments. Cramer et al. (2013) showed that severity of sentencing increased when evidence of a hate crime was provided.

Our second factor, Denial, comprises 10 items, all of which are drawn from Cabeldue et al.’s Negative Beliefs factor. However, examination of item content, in our view, reflects a denial of the offense of hate crime itself, or of its importance, rather than simply a negative belief. A high score on the subscale reflects a higher level of denial. Although several published articles discuss issues such as holocaust denial in the context of hate crime (e.g., Bleich, 2011), relatively few have considered that for some people more everyday hate crime is simply not an issue worthy of concern. Perry (2010) described how, although students in her study indicated an awareness of such actions occurring, they did not see it as problematic, with some suggesting that researching hate crime and legislating for it was a waste of time. Perry cited a participant who stated that “undue attention” is itself to blame for creating a false sense of the disparities that might exist. Individuals who oppose legislation often believe that the attention received by hate crimes, in fact, sensationalizes the crime (e.g., Dunbar & Molina, 2004). This belief also reflects Ditomaso et al.’s (2003) claim that color blindness “allows whites to ignore, deny, or disregard any notion that race matters in people’s lives” (p. 197). Further research could usefully explore the nature of hate crime denial.

Our third factor, which we label Compassion, has five items. Three emerged from the reverse worded items within Cabeldue et al.’s Negative Beliefs subscale, whereas the remaining two came from their Deterrence scale. Compassion has been described as a felt response to suffering that involves caring and an authentic desire to ease distress (Goetz et al., 2010), a definition clearly reflected in the scale items. A higher score on this subscale indicates greater compassion for victims and the wider community and the desirability of prevention. Such notions are consistent with the literature showing support for increased punishment for hate crime offenders, along with increased perpetrator blame (e.g., Cramer et al., 2013). A previous study has shown that advocates of hate crime legislation believe that if hate crime offenders receive harsher sentences, others will be less likely to commit this type of crime in the future (Saucier et al., 2006).

In line with Cabeldue et al., we observed a modest though significant positive association with previous experience of hate crime, with those who had experienced it personally being more likely to endorse harsher punishment. Compassion was negatively associated with political orientation such that individuals with right-wing views reported less compassion. Males reported more denial than females, broadly in line with previous research on other types of crime. For instance, a meta-analysis by Anderson et al. (1997) showed more rape acceptance for men, whereas women with experience as, and/or exposure to, rape victims were associated with less rape acceptance. Men are also found to show more acceptance of myths and higher victim blame in cases of prostitution (e.g., Cotton et al., 2002), child sexual abuse (e.g., Cromer & Freyd, 2007), and human trafficking (Cunningham & Cromer, 2014). No research to date has examined myths about hate crime. Previous research has also suggested higher rates of compassion generally among women (e.g., Mercadillo et al., 2011). Age showed a positive correlation with both Denial and Compassion. Anderson et al. (1997) showed older people to show higher rape acceptance, but they are also generally found to be more compassionate of others. Future research should explore the degree to which these findings are explicable by cohort effects or developmental mechanisms.

Finally, it is interesting to note that none of the items we retained in our final 20-item measure came from Cabeldue et al.’s (2018) Victim Harm subscale, even though both Sentencing and Compassion items suggest an acknowledgment of the harm done to victims and communities. The failure to retain any of these items might be indicative of cultural differences, such as the United States being a more individualist culture, or in specific differences in the social construction of hate crime between the United States and United Kingdom. Such possibilities might be considered in future research.

Overall, the links with previous research and fit indices reported above suggest that the new HCBS-UK is a robust instrument for measuring hate crime attitudes in terms of sentencing, compassion for victims, and denial of the issue. In Study 2, we present a test of the construct validity of the HCBS-UK by examining the relationship between its subscale scores and personality traits previously documented to be associated with prejudice.

Study 2

When developing their original HCBS, Cabeldue et al. (2018) showed that their scale scores showed associations with measures of prejudice against various protected characteristics and social groups. This is a useful test of the validity of the questionnaire with their samples. Hate crimes can be thought of as an extreme demonstration of prejudice (Cramer et al., 2013). Although much research on prejudice has focused on the role of social and intergroup influences, over recent years interest in individual and personality factors has increased. Cabeldue et al. (2018) presented evidence that scores on all four of their HCBS subscales were related to some form of prejudice, with their Negative Beliefs scale showing positive associations and the other scales (Offender Punishment, Deterrence, and Victim Harm) presenting negative associations.

In Study 2, we investigate the relationship between attitudes to hate crime and personality traits found to influence prejudice, those inherent within right-wing ideologies, specifically RWA and SDO. Although these RWA and SDO are by no means the only drivers of hate crime, a significant amount of influential research has focused on them and they therefore form useful constructs by which to test the concurrent validity of our new scale. RWA—the sociocultural component of right-wing ideology—comprises a combination of conventionalism, authoritarian aggression, and authoritarian submission (Altemeyer, 1998) and has been found to predict a range of political, social, ideological, and intergroup behviours and attitudes. RWA is a consistent predictor of general prejudice and ethnocentrism (for review, see Sibley & Duckitt, 2008). SDO—the economic component of right-wing ideology—reflects an individual’s general attitude toward intergroup relations and whether they prefer such relations to be equal or hierarchical. Measures of SDO assess perceptions that one’s own in-group is superior to certain outgroups and therefore should dominate them (Pratto et al., 1994; Sibley & Duckitt, 2008). Other research has considered the role of more general personality traits, especially the Big Five: Openness to Experience (imaginative, preference for variety, openness to different value systems), Conscientiousness (impulse control, purposeful, well organized), extraversion (active, excitement seeking, highly sociable), Agreeableness (altruistic, empathic, helpful, trusting), and Neuroticism (prone to worry, anxiety, depression, angry hostility). The Big Five model of personality is arguably the most widely used in psychology and has been found to explain a wide range of social behviours. In terms of prejudice, the most consistent finding is that Openness to Experience is negatively associated with prejudice and stereotyping and that higher Openness predicts positive intergroup attitudes (Flynn, 2005; McCrae, 1996; Van Hiel & Mervielde, 2004). Both RWA and SDO correlate negatively with Openness, but RWA also correlates positively with Conscientiousness and SDO negatively with Agreeableness (Heaven & Bucci, 2001).

The dual-process motivation model of prejudice (e.g., Duckitt, 2001; Duckitt & Sibley, 2010) suggests that SDO and RWA are not personality traits in themselves, but rather dimensions of ideological attitudes that mediate the relationship between traits such as the Big Five and prejudice. They support individual goals or values regarding group-based dominance and superiority (SDO) and social cohesion and collective security (RWA). These two motivational goals are made salient for individuals by a combination of personality and socialization in certain social contexts (Duckitt, 2001; Duckitt et al., 2002; Duckitt & Sibley, 2010). Consistent effects are observed whereby Conscientiousness and (low) Openness predict prejudice via RWA, whereas (low) Agreeableness and low Openness predict prejudice via SDO

Relatively little work has focussed on individual determinants of hate crime other than social and economic motivations. Levin and McDevitt (1993) and McDevitt, Levin and Bennett (2002) discuss three types of hate crime offender: those who commit their crimes for the excitement or thrill, those who perceived themselves as defending their home or way of life, and those whose life’s mission is to rid the world of groups they consider evil or inferior. We can see how SDO and RWA may link to these motivations. The only extant study we are aware of which has examined Big Five traits in the context of hate crime is that of ElSherief et al. (2018) who found that online hate speech instigators presented openness scores associated with low emotional awareness and adventurousness, but a wild imagination. They also had low Conscientiousness scores, reflecting a tendency to disregard rules and obligations and to act impulsively, and lower levels of Agreeableness, associated with suspicious and antagonistic behaviors.

In Study 2, we test the integrity of the new HCBS-UK by examining the relationship between scores on its three subscales, RWA, SDO, and the Big Five traits. Assuming that attitudes to hate crime are associated with prejudice, construct validity of the scale will be indicated by positive associations between HCBS-UK Denial scores and measures of SDO and RWA, and a negative association with Openness and Agreeableness. Sentencing and Compassion however should present the opposite pattern, negatively associated with SDO and RWA, and positively with Openness and Agreeableness. In addition, we investigate whether the dual-process model of prejudice (Duckitt & Sibley, 2010) also applies to hate crime attitudes by testing for mediation of these Big Five effects by RWA and SDO, respectively.

Method

Participants

One hundred and thirty-four members of the general public took part. They were recruited through the Prolific online research recruitment platform and paid £2.50 for their time. None had taken part in Study 1 and none declared themselves to be currently students. Mean age was 36.44 (SD = 11.32). Twenty-nine were male, 102 female, and three reported as gender fluid. The majority were White (123; 92%), two Black, four Asian, four mixed race, and one other ethnicity. One hundred and seventeen (87%) reported themselves to be heterosexual, four homosexual, nine bisexual, and four other. Eighty participants (60%) reported having no religion, a further 45 (34%) reported as Christian, three Muslim, one Hindu, one Sikh, two Buddhist, and two other religion. The largest proportion came from London/South East England (29; 22%) or South West England (20; 15%). A further 19 (14%) came from North West England. The reminder came from locations across the United Kingdom, including Scotland (11; 8%) and two from Northern Ireland. In terms of SES, nine reported being from the higher managerial/professional category, 46 (34%) from intermediate managerial/professional, 36 (27%) from supervisory, clerical, junior management, 18 (13%) from skilled manual, eight (6%) from unskilled manual, and 17 (13%) from long-term unemployed for whatever reason. Finally, 40 (30%) reported extremely left-wing views, 79 (59%) reported central political views, and just three people (2%) reported extreme right-wing beliefs.

Procedures

Participants completed the following measures:

HCBS-UK. This is the final 20-item version as developed in Study 1 above. We observed very good reliability in this study (Denial α = .92, Compassion α = .75, and Sentencing α = .83).

Social Dominance Orientation (SDO) scale (Pratto et al., 1994). This 16-item scale presents items such as some groups of people are just more worthy than others and we must increase social equality (reverse scored). Participants respond on a seven-point scale where 1 = do not agree at all and 7 = agree completely and scores reflect the mean response; hence, the highest possible score is 7. We observed excellent reliability with the present sample (α = .91).

Right Wing Authoritarianism (RWA) scale (Altemeyer, 1998). This 12-item scale presents statements such as There are many radical, immoral people in our country today who are trying to ruin it for their godless purposes, whom the authorities should put out of action. Participants respond on a seven-point scale where 1 = strongly disagree and 7 = strongly agree. Score is calculated as the mean response, so the maximum possible score is 7. We observed very good reliability with this sample (α = .85).

Big Five Inventory (BFI; John et al., 1991, 2008) yields scores for each of the Big Five trait dimensions. It lists 44 attributes, for example, I am someone who is . . . talkative, (Extraversion, eight items), helpful and unselfish with others (Agreeableness, nine items), perseveres until the task is finished (Conscientiousness, nine items), worries a lot (Neuroticism, eight items), and curious about many different things (Openness, 10 items). Participants indicate how much each attribute reflects themselves on a scale where 1 = disagree strongly and 5 = agree strongly. We observed very good reliability (Extraversion α = .89, Openness α = .82, Agreeableness α = .79, Conscientiousness α = .83, and Neuroticism α = .87).

Results

Table 3 presents descriptive statistics for the three hate crime attitude factors and other measures.

Table 3.

Descriptive Statistics for Measures in Study 2.

Mean/SD	Compassion	Denial	Sentencing	RWA	SDO	O	E	A	C	N
M	3.88	1.88	3.25	5.33	5.75	3.31	2.76	3.74	3.69	3.22
SD	0.63	0.76	0.92	3.15	2.1	0.66	0.83	0.62	0.68	0.86

Note. RWA = Right-Wing Authoritarianism; SDO = Social Dominance Orientation; O = Openness to Experience; E = Extraversion; A = Agreeableness; C = Conscientiousness; N = Neuroticism.

Table 4 shows correlations between measures. Compassion is negatively associated with both RWA and SDO, and positively with Openness to Experience. Denial was positively associated with both RWA and SDO and negatively with Openness. Sentencing was negatively related to SDO but presented no significant relationship with any other variable.

Table 4.

Correlations Between All Measures in Study 2.

Variable	Denial	Sentencing	RWA	SDO	O	E	A	C	N
Compassion	−.60**	.54**	−.30**	−.42**	.22**	−.11	−.003	−.12	.10
Denial		−.43**	.38**	.61**	−.20*	.06	−.05	.08	−.07
Sentencing			−.10	−.28**	.11	−.003	.01	−.08	.10
RWA				.41**	−.30**	.18*	.27**	.42**	−.36**
SDO					−.19*	.01	−.13	.08	−.12
O						.01	.03	−.05	−.01
E							.36**	.25**	−.36**
A								.44**	−.56**
C									−.49**

Note. RWA = Right-Wing Authoritarianism; SDO = Social Dominance Orientation; O = Openness to Experience; E = Extraversion; A = Agreeableness; C = Conscientiousness; N = Neuroticism. * significant at p =.05; ** significant at p = .01.

We conducted six regression analyses, two on each of the three hate crime factors. Results are shown in Table 5. In each case, we entered the Big Five at Stage 1 and either RWA or SDO at Stage 2. The results are shown in Table 5. For Denial, the only independent predictor at Stage 1 was Openness. When RWA was added at Stage 2, the model was significantly improved, ΔR² = .11, F(1, 127) = 17.27, p < .001, and RWA appeared to mediate the effect of Openness. When we repeated this process with SDO entered at Stage 2, we found a similar result with SDO mediating the effects of Openness in Model 2, ΔR² = .32, F(1, 127) = 66.84, p < .001. With Compassion, similar effects were observed in terms of RWA, ΔR² = .03, F(1, 127) = 3.78, p = .05, and SDO, ΔR² = .13, F(1, 127) = 21.41, p < .001. Finally, as indicated in Table 5, for attitudes regarding Sentencing no significant effects were observed in Stage 1 of the regression which accounted for negligible variance. When RWA was added, this made little difference, ΔR² = .001, F(1, 127) = 0.18, p = .67. When SDO was added at Stage 2, it showed a significant independent effect on Sentencing, but no other significant effects were observed, ΔR² = .06, F(1, 127) = 8.04, p = .01.

Table 5.

Regression Analyses on Denial, Compassion, and Sentencing.

Model		Denial		95% CI		Adj. R²	Compassion		95% CI		Adj. R²	Sentencing		95% CI		Adj. R²
Model		β	Significance	Lower	Upper	Adj. R²	β	Significance	Lower	Upper	Adj. R²	β	Significance	Lower	Upper	Adj. R²
1	O	−.20	.02	−.43	−.03		.22	.01	.05	.37		.11	.23	−.10	.39
	E	.06	.49	−.11	.23		−.10	.30	−.21	.07		.03	.75	−.18	.24
	A	−.15	.17	−.45	.08		.11	.30	−.10	.33		.10	.36	−.17	.47
	C	.08	.46	−.14	.31		−.09	.39	−.27	.10		−.05	.61	−.35	.20
	N	−.10	.38	−.28	.11	.03	.08	.42	−.10	.23	.04	.15	.20	−.08	.39	−.01
2	O	−.08	.33	−.29	.10		.16	.07	−.01	.32		.06	.49	−.16	.32
	E	.06	.54	−.11	.21		−.09	.32	−.21	.07		.03	.72	−.17	.24
	A	−.17	.10	−.45	.04		.12	.26	−.09	.34		.03	.82	−.29	.36
	C	−.04	.71	−.26	.18		−.03	.76	−.22	.16		−.03	.75	−.31	.23
	N	−.02	.83	−.21	.17		.05	.64	−.12	.19		.08	.45	−.15	.33
	RWA	.40	<.001	.17	.48	.14	−.19	.05	−.26	.002	.06	−.25	.01	−.44	−.08	−.01
2	O	−.09	.19	−.27	.06		.15	.07	−.01	.29		.06	.49	−.16	.32
	E	.06	.45	−.09	.19		−.09	.29	−.20	.06		.03	.71	−.17	.24
	A	.03	.75	−.18	.26		−.002	.98	−.21	.21		.03	.82	−.29	.36
	C	.03	.76	−.16	.21		−.05	.56	−.22	.12		−.03	.75	−.31	.23
	N	.05	.63	−.12	.20		−.003	.98	−.15	.15		.08	.45	−.15	.33
	SDO	.60	<.001	.39	.63	.36	−.39	<.001	−.38	−.16	.17	−.25	.01	−.44	−.08	.05

Note. CI = confidence interval; RWA = Right-Wing Authoritarianism; SDO = Social Dominance Orientation.

The dual-process approach (Duckitt, 2001; Duckitt & Sibly, 2007, 2010) suggests that low Openness and Conscientiousness influence prejudice via RWA and that low Openness and Agreeableness predict prejudice via SDO. We tested for these specific mediating effects on Denial and Compassion hate crime beliefs using the PROCESS procedure (Hayes, 2018). The model on the left-hand side of Figure 1 shows significant mediating effects suggesting that lower levels of Openness will result in higher Denial via RWA (β = −.13, 95% confidence interval [CI] = [−.27, −.05]), and also that higher levels of Conscientiousness result in Denial via RWA (β = .13, 95% CI = [.05, .24]). No significant mediating effects on Compassionate beliefs were observed. On the right-hand side of Figure 1, analyses of the effects of traits Openness and Agreeableness via SDO are presented. Significant indirect effects via SDO are presented between both traits and Denial (Openness β = −.12, 95% CI = [−.27, −.01]; Agreeableness β = −.22, 95% CI = [−.43, −.05]). Significant mediating effects of SDO on Compassion were also observed (Openness β = .07, 95% CI = [.01, .17]; Agreeableness β = .12, 95% CI = [.03, .27]).

Figure 1.

Mediation models showing indirect effects via RWA and SDO on Denial (Den) and Compassion (Com) beliefs.

Discussion

The first aim of Study 2 was to test the construct validity of the new HCBS-UK by examining the relationship between scores on the three factors and those on measures of RWA and SDO, both traits known to influence prejudice, and hence we assumed also hate crime. Our analysis supported this aim. Both RWA and SDO were positively associated with the Denial subfactor which assesses negative attitudes toward hate crime legislation and designated groups. The other factors, Compassion and Sentencing, present measures of positive attitudes toward designated groups and are supportive of legislation to control hate crime. Compassion was negatively associated with both RWA and SDO as predicted, whereas Sentencing only presented this relationship with SDO.

Our second aim was to examine whether previous findings relating to the Big Five traits and prejudice would be replicated with regard to hate crime. We found no significant direct effects of Consciousness or Agreeableness on hate crime attitudes; however, the Denial factor was associated with low Openness, whereas Compassion was associated with high Openness. This supported our predictions and earlier research on prejudice which has consistently found that higher Openness predicts positive intergroup attitudes (e.g., Flynn, 2005; McCrae, 1996; Van Hiel & Mervielde, 2004). Furthermore, in line with the dual-process motivation theory of prejudice, low Openness predicted hate crime Denial indirectly via both RWA and SDO. Conscientiousness influenced Denial via RWA, but not Compassion. Finally, Agreeableness influenced both Denial and Compassion via SDO. For Sentencing-related attitudes, no significant effects were observed. Overall, we can conclude that the dual-process motivation hypothesis (Duckitt, 2001; Duckitt & Sibley, 2010) applies to some aspects of hate crime attitudes in a similar way to how it explains prejudice in general. Further application of this model to hate crime might be fruitful.

Furthermore, we suggest that RWA and SDO should exert subtly different effects on hate crime attitudes as a function of the specific groups under scrutiny. High RWAs, who are motivated to conform to and protect the status quo as a reaction to the “dangerous” world they perceive, are particularly prejudiced against groups who they perceive threaten existing social structures. Importantly, this might be seen to include some groups protected by hate crime legislation; for instance, Muslims are frequently portrayed in both mainstream and social media as threatening “British” values and traditions (El-Farra, 1996). RWAs may therefore be more supportive of hate crimes against groups who they see as most threatening to the social order (e.g., radical as opposed to moderate Muslims). Conversely, SDOs’ belief in life as a “competitive jungle” where groups must compete to survive may be motivated to derogate groups they perceive as low-status groups (e.g., people with disabilities). We advocate a future direction in hate crime research that considers specific social targets, or hate crimes with specific underlying motivations (cf. Levin & McDevitt, 1993, 2002).

General Discussion

Understanding public attitudes toward hate crime is important to consider in the context of community demands on police, jury decision making in trials of hate crime offenses and also because such attitudes feed into social norms regarding behaviors toward different social groups. These studies aimed to develop and validate a U.K. version of Cabeldue et al.’s (2018) HCBS. This new scale was found to comprise three robust factors, Denial, Compassion, and Sentencing (as opposed to four factors in the original version). These subscales represent denial of the severity of hate crime and the harm it causes, compassion toward victims and affected communities, and punitive attitudes toward sentencing. Scores on these factor scales presented associations with those on traits known to be associated with prejudice in line with previous research. Scores on the Denial factor are positively associated with those on both RWA and SDO, widely accepted to be indicators of prejudice, but negatively associated with Openness to Experience which is usually linked to tolerance. Compassion scores were negatively associated with RWA and SDO and positively with Openness. Furthermore, Duckitt and Sibley’s dual-process model of prejudice whereby Big Five traits Openness to Experience, Consciousness, and Agreeableness account for attitudes via either RWA or SDO also applies to hate crime Denial and Compassion beliefs as measured with our scale. Altogether, our results suggest that the HCBS-UK can be recommended for use in future research in a U.K. context as a robust measure of three aspects of hate crime attitudes.

The one factor that did not fit with all our expectations was attitudes to sentencing, which presented only a negative association with SDO. That individuals with the most punitive attitudes seem to be those with lower levels of SDO suggests a generally magnanimous character, possibly with wider societal concerns as opposed to personal grudges. However, Gerber and Jackson (2016) suggested that punitive sentiment was related to RWA and arises out of conformity, adherence to conservative moral values, and concerns about collective security and cohesion. They did not consider SDO specifically; however, their results suggest that punitiveness as measured in our scale may differ from the form they discuss. We observed no association between sentencing beliefs and any demographic variables or Big Five traits; however, Sentencing scores were possibly related to personal experience of hate crime, a variable not associated with either of the other subscales. One possibility is that experience of hate crime fuels a harsher attitude toward sentencing, though recent work has suggested that exposure to crime does not make for more punitive attitudes (Kleck & Jackson, 2017) and that ideology and media influences are more important. Further investigation into the association between scores on this Sentencing subscale, general attitudes toward crime, and experience of it is worthy of further study. A more fine-grained examination of the Big Five traits, encompassing subfacet scores, may also reveal more about this type of belief. It may be that scores on this subscale reflect generally punitive attitudes toward crime. As such, the factor still applies to hate crime beliefs and is a useful measure, but we might not necessarily expect to see much relationship between scores and those on traits known to influence beliefs about prejudice and/or hate crime specifically. Overall, however, it is important to note that Study 1 clearly indicated a distinct factor and questionnaire items evidently reflect beliefs about harsher sentencing for hate-related offenses. As such, the Sentencing subfactor clearly applies to hate crime beliefs and is a useful measure alongside the other two factors. Future research might benefit from considering the role of other ideology-relevant constructs which are relevant to prejudice more distally. For instance, individuals high in system justification (Sibley, 2010) may be less inclined to use (hate crime) legislation to address societal inequities, whereas those higher in just-world belief may be more supportive of hate crime uplifts as a means of redressing the extra harm to victims (Gromet, 2012).

These studies are not without limitations. First, although a strength of the research is that we employed general public participants (as opposed to, say, student populations), our samples comprised mostly individuals who identified as White and female (as is typical of volunteer participant samples). Although research on hate crime perpetrators in the United Kingdom has tended to find that the majority are White (e.g., Iganski & Smith, 2011; Wilcox et al., 2010), they also tend to be male. Controlling for demographic factors appeared to have little influence on our results; however, the relatively few males in the sample may have led to a degree of bias and this should be acknowledged. In this respect, our studies share the limitations acknowledged by Cabeldue et al. (2018) in that participants were not particularly diverse. Whereas our results provide valuable insight into how majority members—who by definition are most likely to be jurors and law enforcers—perceive hate crime, a more diverse sample may further enhance the breadth of this promising research by expanding understanding of how factors such as sex, race, and personal experience of hate crime may influence beliefs. Second, although Study 2 presents useful evidence for concurrent validity of the HCBS-UK, it is important to acknowledge that RWA and SDO are not the only drivers of prejudice and hate crime. It will be fruitful to investigate how attitudes measured with this new scale are associated with other sociopersonality factors such as left-wing antifascism, or the pressures associated with strain theory (Agnew, 2006), whereby hate offenses are assumed as a response to perceived social threats by minority groups (e.g., competition for jobs or housing).

In conclusion, in these studies, we extend the work of Cabeldue et al. (2018) in presenting a new measure of beliefs about hate crime developed within a U.K. context. We further provide support for the dual-process motivation model of prejudice while extending its potential utility to include the explanation of attitudes toward hate crime. Importantly, the new HCBS-UK appears a robust instrument for the measurement of beliefs associated with denial of hate crime severity, compassion toward victims, and the belief in harsher sentencing. Research is now required to further validate the scale with new populations, establish predictive validity through the use of behavioral measures, and to obtain greater understanding of the processes by which these beliefs influence real-world behavior.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iDs

Alison M. Bacon

Jon May

Author Biographies

Alison M. Bacon, PhD, is an associate professor of Psychology at University of Plymouth, UK. Her research interests include personality and dispositional influences on attitudes and life outcomes in crime, health, and well-being.

Jon May, PhD, is professor of Psychology at University of Plymouth, UK. His research interests are in cognition and emotion, and the role of mental imagery in motivation and behavior change.

Jaysan J. Charlesford, PhD, is a lecturer in Psychology at University of Plymouth, UK. His research interests are in dispositional predictors of prejudice, prejudice reduction, and intergroup climate.

References

Agnew

(2006). Pressured into crime: An overview of general strain theory. Oxford University Press.

Altemeyer

(1998). The other “authoritarian personality.” Advances in Experimental Social Psychology, 30, 47–92.

Anderson

K. B.

Cooper

Okamura

(1997). Individual differences and attitudes toward rape: A meta-analytic review. Personality and Social Psychology Bulletin, 23, 295–315.

Bleich

(2011). The rise of hate speech and hate crime laws in liberal democracies. Journal of Ethnic and Migration Studies, 37, 917–934.

Cabeldue

M. K.

Cramer

R. J.

Kehn

Crosby

J. W.

Anastasi

J. S.

(2018). Measuring attitudes about hate: Development of the Hate Crime Beliefs Scale. Journal of Interpersonal Violence, 33, 3656–3685.

Chakraborty

(2014). Re-thinking hate crime: Fresh challenges for policy and practice. Journal of Interpersonal Violence, 30, 1738–1754.

Chung

Rimal

R. N.

(2016). Social norms: A review. Review of Communication Research, 4, 1–28.

Cialdini

R. B.

Goldstein

N. J.

(2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55, 591–622.

College of Policing. (2014). National policing hate crime strategy. http://library.college.police.uk/docs/college-of-policing/National-Policing-Hate-Crime-strategy.pdf

10.

Cotton

Farley

Baron

(2002). Attitudes toward prostitution and acceptance of rape myths. Journal of Applied Social Psychology, 32, 1790–1796.

11.

Cramer

R. J.

Miller

A. K.

Amacker

A. M.

Burks

A. C.

(2013). Openness, rightwing authoritarianism, and antigay prejudice in college students: A mediation model. Journal of Counseling Psychology, 60, 64–71.

12.

Criminal Justice Act (2003). Revtrieved from https://www.legislation.gov.uk/ukpga/2003/44/contents

13.

Cromer

L. D.

Freyd

J. J.

(2007). What influences believing child sexual abuse disclosures? The roles of depicted memory persistence, participant gender, trauma history, and sexism. Psychology of Women Quarterly, 31, 13–22.

14.

Cunningham

K. C.

Cromer

L. D.

(2014). Attitudes about human trafficking: Individual differences related to belief and victim blame. Journal of Interpersonal Violence, 31, 228–244.

15.

Dearden

(2018). Religious hate crime rises 40% in England and Wales—With more than half directed at Muslims. The Independent. https://www.independent.co.uk/news/uk/crime/uk-hate-crime-religious-muslims-islamophobia-police-racism-a8585846.html

16.

Ditomaso

Parks-Yancy

Post

(2003). White views of civil rights: Color blindness and equal opportunity. In Doane

Bonilla-Silva

(Eds.), White out: The continuing significance of racism (pp. 189–198). Routledge.

17.

Duckitt

(2001). A dual-process cognitive-motivational theory of ideology and prejudice. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 33, pp. 41–113). Academic Press.

18.

Duckitt

Sibley

C. G.

(2007). Right wing authoritarianism, social dominance orientation and the dimensions of generalized prejudice. European Journal of Personality, 21, 113–130.

19.

Duckitt

Sibley

C. G.

(2010). Personality, ideology, prejudice, and politics: A dual-process motivational model. Journal of Personality, 78, 1861–1894.

20.

Duckitt

Wagner

du Plessis

Birum

(2002). The psychological bases of ideology and prejudice: Testing a dual process model. Journal of Personality and Social Psychology, 83, 75–93.

21.

Dunbar

Molina

(2004). Opposition to the legitimacy of hate crime laws: The role of argument acceptance, knowledge, individual differences, and peer influence. Analyses of Social Issues and Public Policy, 4, 91–113.

22.

El-Farra

(1996). Arabs and the media. Journal of Media Psychology, 1(2), 1-7.

23.

ElSherief

Kulkarni

Nguyen

Wang

W. Y.

Belding

(2018). Hate Lingo: A target-based linguistic analysis of hate speech in social media. Twelfth International AAAI Conference on Web and Social Media (ICWSM 2018). https://aaai.org/ocs/index.php/ICWSM/ICWSM18/paper/view/17910

24.

Flynn

F. J.

(2005). Having an open mind: The impact of openness to experience on interracial attitudes and impression formation. Journal of Personality and Social Psychology, 88, 816–826.

25.

Gerber

M. M.

Jackson

(2016). Authority and punishment: On the ideological basis of punitive attitudes towards criminals. Psychiatry, Psychology and Law, 23, 113–134.

26.

Githens-Mazer

Lambert

(2010). Islamophobia and anti-Muslim hate crime: A London case study. European Muslim Research Centre, University of Exeter. http://library.college.police.uk/docs/Hate-crime-report-London.pdf

27.

Goetz

J. L.

Keltner

Simon-Thomas

(2010). Compassion: An evolutionary analysis and empirical review. Psychological Bulletin, 136, 351–374.

28.

Grattet

Jenness

(2008). Transforming symbolic law into organizational action: Hate crime policy and law enforcement practice. Social Forces, 87, 501–527.

29.

Gromet

D. M.

(2012). Restoring the victim: Emotional reactions, justice beliefs, and support for reparation and punishment. Critical Criminology, 20, 9–23.

30.

Hayes

A. F.

(2018). An Introduction to Mediation, Moderation and Conditional Process Analysis (2nd ed.). Guilford Press.

31.

Heaven

P. C. L.

Bucci

(2001). Right-wing authoritarianism, social dominance orientation and personality: An analysis using the IPIP measure. European Journal of Personality, 15, 49–56.

32.

Hodson

Dhont

(2015). The person-based nature of prejudice: Individual difference predictors of intergroup negativity. European Review of Social Psychology, 26, 1–42.

33.

Hodson

Hooper

Dovidio

J. F.

Gaertner

S. L.

(2005). Aversive racism in Britain: The use of inadmissible evidence in legal decisions. European Journal of Social Psychology, 35, 437–448.

34.

Home Office. (2018). Hate crime, England and Wales, 2017/18 (Statistical Bulletin 20/18). https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/748598/hate-crime-1718-hosb2018.pdf

35.

Iganski

(2008). Hate crime in the city. Policy Press.

36.

Iganski

Smith

(2011). Rehabilitation of hate crime perpetrators: Research report. Equality and Human Rights Commission.

37.

John

O. P.

Donahue

E. M.

Kentle

R. L.

(1991). The Big Five Inventory–Versions 4a and 54. Institute of Personality and Social Research, University of California, Berkeley.

38.

John

O. P.

Naumann

L. P.

Soto

C. J.

(2008). Paradigm shift to the integrative Big Five trait taxonomy: History, measurement, and conceptual issues. In John

O. P.

Robins

R. W.

Pervin

L. A.

(Eds.), Handbook of personality: Theory and research (pp. 114–158). Guilford Press.

39.

Kleck

Jackson

D. B.

(2017). Does crime cause punitiveness? Crime & Delinquency, 63, 1572–1599.

40.

Levin

McDevitt

(1993). Hate crimes: The rising tide of bigotry and bloodshed. Plenum.

41.

Mallett

R. K.

Huntsinger

J. R.

Swim

J. K.

(2011). The role of system justification motivation, group status and system threat in directing support for hate crimes legislation. Journal of Experimental Social Psychology, 47, 384–390.

42.

Mason

(2005). Hate crime and the image of the stranger. British Journal of Criminology, 45, 837–859.

43.

McCombs

(2005). The agenda setting function of the press. In Overholser

Jamieson

K. H.

(Eds.), Institutions of American democracy: The press (pp. 156–168). Oxford University Press.

44.

McCrae

R. R.

(1996). Social consequences of experiential openness. Psychological Bulletin, 120, 323–337.

45.

McDevitt

Levin

Bennett

(2002). Hate crime offenders: An expanded typology. Journal of Social Issues, 58, 303–317.

46.

Mercadillo

R. E.

Díaz

J. L.

Pasaye

E. H.

Barrios

F. A.

(2011). Perception of suffering and compassion experience: Brain gender disparities. Brain and Cognition, 76, 5–14.

47.

Michie

van Stralen

M. M.

West

(2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science. https://implementationscience.biomedcentral.com/articles/10.1186/1748-5908-6-42

48.

Miller

D. T.

Prentice

D. A.

(2016). Changing norms to change behavior. Annual Review of Psychology, 67, 339–361.

49.

Perry

(2010). “No biggie”: The denial of oppression on campus. Education, Citizenship, and Social Justice, 5, 265–279.

50.

Pornprasertmanit

Miller

Schoemann

Rosseel

Quick

Garnier-Villarreal

…Longo

. 2016. Package ‘semTools’, Version 0.4-13. https://github.com/simsem/semTools/wiki

51.

Pratto

Sidanius

Stallworth

L. M.

Malle

B. F.

(1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741–763.

52.

R Core Team (2015). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org

53.

Rosseel

(2012). Lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software, 48, 1-36.

54.

Saucier

D. A.

Brown

T. L.

Mitchell

R. C.

Cawman

A. J.

(2006). Effects of victims’ characteristics on attitudes toward hate crimes. Journal of Interpersonal Violence, 21, 890–909.

55.

Saucier

D. A.

Hockett

J. M.

O’Dea

C. J.

Miller

S. S.

(2017). The racism justification hypothesis and attitudes toward hate crime legislation. In Dunbar

Blanco

Crevecoeur-MacPhail

D. A.

(Eds.), The psychology of hate crimes and domestic terrorism (pp. 283–386). Praeger.

56.

Shah

D. V.

Domke

Wackman

D. B.

(1996). “To thine own self be true”: Values, framing, and voter decision-making strategies. Communication Research, 23, 509–560.

57.

Sibley

C. G.

(2010). The dark duo of post-colonial ideology: A model of symbolic exclusion and historical negation. International Journal of Conflict and Violence (IJCV), 4, 106–123.

58.

Sibley

C. G.

Duckitt

(2008). Personality and prejudice: A meta-analytic review. Personality and Social Psychology Bulletin, 12, 248–279.

59.

Van Hiel

Mervielde

(2004). Openness to experience and boundaries in the mind: Relationships with cultural and economic conservative beliefs. Journal of Personality, 72, 659–686.

60.

Walters

Wiedlitzka

Owusu-Bempah

Goodall

(2017). Hate crime and the legal process: Options for law reform. University of Sussex.

61.

Warren-Gordon

(2018). Perceiving hate crimes: A comparative analysis of major newspaper coverage of hate crimes in the United States and the United Kingdom. International Journal of Criminal Justice Sciences, 13, 172–180.

62.

Wilcox

Smithson

Christmann

Monchuk

Wong

(2010). Racially motivated offending and targeted interventions. Youth Justice Board.

63.

Williams

M. L.

Tregidga

(2014). Hate crime victimization in Wales: Psychological and physical impacts across seven hate crime victim types. The British Journal of Criminology, 54, I946–I967.

64.

Wright

S. C.

Aron

McLaughlin-Volpe

Ropp

S. A.

(1997). The extended contact effect: Knowledge of cross-group friendships and prejudice. Journal of Personality and Social Psychology, 73, 73–90.

65.

Zempi

Chakraborti

(2014). Islamophobia, victimisation and the veil. Palgrave Macmillan.