Aporophobic and Homeless Victimisation—the Case of Ghent

Aporophobia, the rejection of poor people, is a major social problem with known harmful consequences, especially for the most excluded in our society, i.e., homeless people. However, this phenomenon has been only rarely studied using testable means on the ground. This article reports on empirical research conducted on a sample of homeless people in Ghent, Belgium. The principal objective was to analyse aporophobic discriminatory incidents and hate crimes which were experienced by around 50% of the respondents. The most common offences were property crimes, particularly thefts and robberies, with the belief being that the perpetrators were motivated to perform these crimes as they saw this particular demographic group as more helpless and vulnerable. Moreover, the victims usually experienced more than one crime and different typologies. Unfortunately, underreporting was pronounced. It is therefore necessary to increase the visibility of these crimes and the reporting rate, as well as to reduce victimisation. To this end, it is essential to know the real scope of such crimes and their characteristics.

but does not address the absence of discrimination towards immigrants or foreigners when these people have assets, economic resources or other status (Cortina, 2017).
To date, only a few empirical studies have been conducted, mainly in Spain (Ávila & Garrido, 2019;García, 2019García, , 2023García, et al., n.d;López et al., 2021;Rais Fundación, 2015), and especially with regard to one of the most excluded groups in society: the homeless (García, 2020a, for empirical studies focusing on other aspects of victimisation in other countries see e.g. Gaetz, 2004, Molnar & Hashimoto, 2023and Tyler & Melander, 2009. Rais Fundación (2015), who interviewed a sample of 261 homeless people in four Spanish cities, carried out the first study. The principal observation was that 47,1% of the interviewees have experienced aporophobic discriminatory incidents and hate crimes. Similar results were found in later studies conducted in different cities of Spain (García, 2019;García et al., n.d;Muñoz et al., 2017).
Research on victimisation and homelessness has also been conducted at an international level. For example, the National Coalition for the Homeless (2020) has recorded hate crimes committed against homeless people in the USA over the last 20 years through different data sources. Additionally, Ellsworth's (2019) literature review on street crime victimisation and victimisation-prone lifestyle also yielded important insights. He found that assaults, theft and robberies are usually experienced by homeless people living on the street. Furthermore, Ellsworth claimed that certain characteristics increased the risk of victimisation, such as being old, alcohol and drug abuse or mental illness. He also stated that there is a lack of studies related to rough sleepers (Ellsworth, 2019). However, neither of these studies include the concept of aporophobia.
In criminology and criminal law, aporophobia is said to have two sides. On the one hand, the concept of aporophobia includes all processes or actions that contribute to the criminalisation of poverty. On the other hand, aporophobia also encompasses everything that contributes to the victimisation of the poor (Benito & Pérez, 2022). In most European countries, criminalisation of poverty can especially be found in administrative law provisions and municipal ordinances which directly or indirectly punish poverty and homelessness (Calatayud, 2017;De Verteuil et al., 2009;FEANTSA, 2021;May, 2014; National Law Center on Homelessness & Poverty, 2019; Peršak, 2017;Pleysier, 2017;Podoletz, 2017;Stuart, 2014). Punishment of the homeless is encompassed in a punitive expansion of criminalising anti-social or disorderly behaviour-partly explained by an increase in feelings of insecurity experienced in many European countries-such as Anti-Social Behaviour Orders (or ASBOs) introduced by the Crime and Disorder Act of 1998 in the UK (Di Ronco, 2016;Di Ronco & Peršak, 2014;Peršak, 2017). Examples of criminalisation of poverty are forbidding of sleeping or begging on the street and punishing to rummage in public street bins, even if it is for getting food. Moreover, it is relevant to highlight that the criminalisation of poverty relates to aporophobic motivation for victimisation (hate crimes), especially towards those forced to spend most of the day on the street, for whom the likelihood of experiencing victimisation in general (Puente, 2019;Sanders & Albanese, 2016;Walklate, 2000) and victimisation with an aporophobic motivation increases significantly (García, 2020a; RAIS Fundación, 2015). Again, Spain is the only country where official data on aporophobic discriminatory incidents and hate crimes are collected, analysed and published separately (García, 2020a;Hanek, 2021;López et al., 2021;Ministry of the Interior, 2022;Rais Fundación, 2015). The results of these studies show the existence and seriousness of the phenomenon and they have directly influenced the inclusion of aporophobic motivation in the general aggravating circumstance of the Spanish Criminal Code (2021). 1 However, other countries have also previously incorporated aggravation by the economic situation of the person into their Criminal Code, e.g., Belgium. However, despite the introduction of a comprehensive hate crime regulation in May 2007 2 and its significant involvement in the fight against homelessness (see projects such as "Housing First" 3 or "Roof Project" 4 ), there is no research on hate crime due to aporophobia in Belgium. As the Organization for Security and Cooperation in Europe (2020) stated that a lack of comprehensive approach to hate crimes leaves them invisible and unaddressed.

Aporophobia and Homelessness in Belgium
Since 2007, the Belgian Criminal Code has provided for aggravating circumstances for a wide number of specific crimes when committed with a bias motive 5 related to a number of specific discrimination criteria, 6 including "wealth". According to Unia-Belgium's independent public institution that fights discrimination and promotes equal opportunities-"wealth" as a discrimination criterion is mainly applied in cases where people are refused the rent of an apartment or house on the basis of their income (Unia, n.d.). Moreover, official Belgian statistics of hate crimes do not provide data on discrimination on the grounds of "wealth" separately, but include them in a general category called "hate crimes committed against other groups". No other statistics are available. A search in the judgments included in Unia's database 7 only revealed one case (first instance and appeal) about a homeless person being victimised out of aporophobic motivation. 8 The appeal 1 Through the "Organic Law 8/2021, of 4th June, on the comprehensive protection of children and adolescents against violence" ("Ley Orgánica 8/2021, de 4 de junio, de protección integral a la infancia y la adolescencia frente a la violencia"). 2 Through the "Law against certain forms of discrimination" ("Wet ter bestrijding van bepaalde vormen van discriminatie"). 3 More information available at: http:// www. housi ngfir stbel gium. be/ en/ 4 More information available at: http:// www. roofn etwork. eu/ rooft opeu2 022 5 These are: voyeurism, indecent assault and rape (art. 377-bis); murder, intentional injury and manslaughter (art. 405-quater); non-assistance to a person in danger (art. 422-quarter); violation of personal liberty and trespass (art. 438-bis); libel and defamation (art. 453-bis); arson (art. 514-bis); and destruction of personal possessions or property (art. 532-bis) (Micheletto, 2020). 6 The 19 specific discrimination criteria are the following: race, colour, descent, national or ethnic origin, nationality, sex, sexual orientation, marital status, birth, age, wealth, religious beliefs or philosophy of life, current and future state of health, disability, language, political beliefs, trade union, physical or genetic characteristics, or social origin (Micheletto, 2020). 7 Despite the fact that sentences are not public in Belgium, Unia has developed a database where it collects the sentences that deal with incidents and crimes committed for discriminatory reasons. As there is a law that obliges all national Criminal Courts to report sentences to Unia, it can be thought that it includes a high percentage. In fact, it contains more than 900 sentences from 1983 to the present. 8 Brussels Court of First Instance of 17 February 2016 and Brussels Court of Appeal of 27 January 2021. judgment states that the assault "was carried out with a motive of hatred, contempt or hostility towards the victim, in particular because of his national or ethnic origin, his wealth status, his social origin or his political convictions" (own translation), 9 and the aggravating circumstance was applied.
While the victimisation of homeless people and aporophobia has not been studied in Belgium, there is data on homelessness available. In 2021, the European Federation of National Organisations Working with the Homeless (FEANTSA, 2021) pointed out that there were more than 700,000 roofless people in the European Union. According to the European Typology on Homelessness and Housing Exclusion Light (ETHOS Light), 10 the category "roofless" refers to people sleeping in the street or in emergency accommodation. Belgian census research shows that sleeping in the street is a reality in Belgium (King Baudouin Foundation, 2021). In Ghent, a city in the Flemish region of Belgium of about 250,000 inhabitants, 124 people living rough and 113 in emergency accommodation were reported daily in 2020 (Hermans et al., 2021).

Recognising and Studying Homeless and Aporophobic Victimisation
Traditionally, hate crime victims are believed to come from marginalised minority groups. Members of majority communities were therefore not understood to be victims of hate crimes as this is the preserve of historically disadvantaged minorities (Chakraborti & Garland, 2012). Although the definition and focus of hate crimes have developed further to include more groups and open clauses (OSCE & ODHIR, 2005:12), homeless people have hardly been considered and studied as victims of aporophobic hate crimes. This traditional restrictive focus on hate crimes may explain the lack of attention to homeless people. The fact that governments often turn a blind eye to homelessness and even tend to criminalise poverty (Drilling, 2020) may be additional reasons why homeless and aporophobic hate crimes remain rather unnoticed, unpunished and understudied.
According to Perry (2001), hate crimes are acts of violence and intimidation. As such, they are a mechanism of power and oppression to maintain the established social order. In this sense, Expósito (2015) developed the term of "institutional aporophobia", which was reconceptualised by García (2023) as a network of formal and informal, structural economic, political, social and cultural practises that produce and reproduce both social exclusion and aporophobia. This definition points out the systematic violence experienced by groups excluded from society, especially if they are seen as undesirables, such as the homeless.
Victims of hate crimes might, however, be targeted not only to reinforce social structure but also because they are perceived as vulnerable and "easy" victims. The risk of victimisation indeed increases some factors, such as sleeping rough. In this sense, Puente (2023) found that homeless who engaged in a broader range of risky behaviours experienced 9 Original text "de travail et portés avec un mobile de haine, de mépris ou d'hostilité envers la victime en raison notamment de son origine nationale ou ethnique ou de sa nationalité, de son état d'infortune, de son origine sociale ou de ses convictions politiques (prévention C)". Brussels Court of First Instance of 17 February 2016, p.4. 10 As homelessness is not a static phenomenon and covers different stages, FEANTSA (n.d) developed the European Typology on Homelessness and Housing Exclusion (ETHOS) with two versions: ETHOS and ETHOS Light. This is the reference classification at European level to use a common language and, consequently, to be able to make comparative analyses. victimisation events to a greater extent. Examples include being arrested multiple times, reporting recent drug use, or having previously received a non-prison sentence. Additionally, it has been pointed out that perpetrators may be in an economical disadvantaged situation too (Chakraborti & Garland, 2012).
For these reasons, this paper explicitly recognises homeless people as an important but hidden group of victims of hate crimes. The perspective taken is eminently victimological, studying incidents of discriminations and crimes committed because of the situation of extreme poverty as experienced by homeless people. It allows us to highlight experiences of a vulnerable group and intends to add to the need for more inclusive approaches by focusing on the victim's perspective when studying hate crimes (Chakraborti & Garland, 2012). This paper reports on a study into homeless people in Ghent with a special focus on victimisation through aporophobic discriminatory incidents and hate crimes. The objectives are: to get an estimate of the occurrence of these incidents and crimes; examine the relationship between being a victim and sociodemographic characteristics, as well as victimisation and variables related to crime; and to gain an insight into the experiences of victims.

Sample
Recruitment of respondents was done through homeless shelters in Ghent, with a special focus on those shelters that participated in the 2020 census (Hermans et al., 2021). They were contacted by phone and/or email, explaining purpose of the research, with the aim to obtain their collaboration. More than 10 centres were contacted, but only 5 agreed to collaborate with the research. Thus, the sample was obtained from 5 centres for homeless people in the city of Ghent from 22 nd April until 9 th May 2022. The shelters only provided the possibility to inform and reach out to the people in their centre. They did not interfere with the selection of the sample.
An anonymous questionnaire that was designed, peer-reviewed and pilot tested in Spain was applied. Also, it was validated through the judgment of 10 experts (Fleiss Kappa statistic was applied) and was, along with the entire set-up of the study, approved by the Ethics Committee of the University of Salamanca.
For this specific study in Ghent, the questionnaire was translated into English and Dutch by two native speakers. The researchers went to the homeless shelters in order to get the sample. There, they explained the objective and content of the research to all users, as well as the selection criterion: to have lived or slept on the street at some point during their homelessness. All homeless people who fulfilled the condition had the opportunity to participate, although it was stressed as voluntary with no rewards given. Thus, the respondents were not compensated. Furthermore, the researchers, one of whom was born in Belgium, communicated with foreign participants in their native language if it was Dutch. Alternatively, they used English, but only if the participants demonstrated fluency. Both researchers are proficient in English.
The questionnaire was administered on paper to those who agreed to participate. Although the questionnaire itself included the instructions for completing the questionnaire, it was also explained to everyone when it was distributed. The researchers stayed in the same room while the respondents completed the questionnaire and, when needed, they helped to resolve any doubts related to the questions in order to achieve a higher accuracy in the responses. A total number of 32 homeless people participated in this study; 13 were in a shower service, 9 in a drug rehabilitation day centre, 8 in a night shelter and 2 in a long-term shelter.
Related to ethical and methodological considerations, the homeless were treated with great care, dignity and fairness and were fully informed on the details of the research. The questionnaire included explicit information about the name and affiliation of the participating researchers; the objective of the research; voluntariness, confidentiality, anonymity and lack of compensation; ethical recommendations and research contact details. Moreover, we stressed that they could leave any answer blank and withdraw from the research at any time, without their decision affecting them in any way (Runnels et al., 2009).
Additionally, with the aim of addressing possible harms or distress in the respondents (Runnels et al., 2009), the questionnaire contained a list of institutions that can be contacted in case they need help, such as: The Centre for General Welfare Work (Centrum Algemeen Welzijnswerk-CAW), which helps people with their questions and problems related to well-being; Unia, related to discrimination, hate speech and support related to these issues; services dealing with sexual violence; and the police. For all the institutions, we included the following: type of service, phone number, website and languages.

Variables
The variables were chosen based on previous research (García et al., n.d; National Coalition for the Homeless, 2020; Rais Fundación, 2015) in order to meet the objectives of this study (see Appendix Table 4 for the variables under study and their correspondence to the questions asked in the survey). A database with 72 variables was created. These are briefly presented in Table 1. The first variables are centred around sociodemographic characteristics, homelessness duration and others related to alcohol, drugs and the criminal justice system. Subsequently, there are variables that refer to the period of their life in which they were living on the street. Finally, some variables are about victimisation and types of crimes, aporophobic victimisation, a homeless person being the offender, reporting and reasons of not reporting. The categories of the majority of them are a Frequency Likert Scale-FLS from here on-with five categories: (1) very often; (2) often; (3) sometimes; (4) rarely; and (5) never, with the exception of the last question which contains an Agreement Likert Scale-ALS from here on-with five categories: (1) strongly agree, (2) agree, (3) neither agree nor disagree, (4) disagree and (5) strongly disagree.

Methods
The method was a semi-structured victimisation survey with a combination of semi-closed and Likert-Scale questions. Moreover, the last question of the questionnaire was open and invited homeless people to add any incidents or hate crimes experienced due to their poverty. The first author and a colleague distributed the questionnaire in shelters where informal discussions were held with workers, volunteers and users. Accordingly, field notes were taken.  (1) less than 3 months (2) between 3 and 12 months, (3) between 1 and 3 years, (4) between 3 and 5 years and (5) more than 5 years Living on the street/frequency living on the street NDS/O In order to know about the period in which they lived on the street, including sleeping on the street or in the night shelter according to categories 1 and 2 of ETHOS Light. There are two variables. For the first one the answers are (1) yes and (2) no, and for the second one the same five categories of the previous variable are applied Currently living on the street/frequency living on the street NDS/O It refers to their current situation and the duration of the last time they were living on the street. As well as the last variable, it includes sleeping on the street or in the night shelter according to categories 1 and 2 of ETHOS Light. The first variable has two categories: (1) yes and (2)  The questions ask about the frequency of the following groups of crimes: (1) insults and humiliations, (2) threats, (3) damage to their belongings, (4) thefts or robberies and (5) physical violence or assault (including sexual assault), with one variable for each group, with the FLS. Based on these, another variable was created referring to if they have ever suffered one of these crimes. The answers were (1) yes and (2)  The questions ask about the frequency of filing a report for the following groups of offences: (1) insults and humiliations, (2) threats, (3) damage to their belongings, (4) thefts or robberies and (5) physical violence or assault (including sexual assault), with one variable for each group, with the FLS. Based on these, another variable was created referring to if they had ever filed a report of one of these crimes. The answers were (1) yes and (2) no Reasons of not reporting NP In order to know the reasons for which they had not filed a report the question contained an ALS. The reasons presented were: (1) I did not consider it serious enough; (2) I solved it by myself (I knew the offender); (3) I reported it to the emergency centre, shelter, soup kitchen, or another recourse that I went to; (4) lack of evidence; (5) believe nothing would be done about it; (6) mistrust of the police; (7) fear of reprisals; (8) I do not know how to report it; and (9) others-free text Statistical analysis was performed using IBM® SPSS Statistics®, version 26. We used descriptive analysis and Fisher's exact test 11 in order to study the relationship between variables by crosstabs. The significance criterion was a p-value lower than 0.05. We also calculated adjusted standardised residuals in which values over ± 1.96 evince statistically significant relations since they deviate from zero with a probability greater than 0.95. The relationship is determined by the positive or negative sign. Likewise, the higher the absolute value of the residual, the greater the relationship between each pair of categories. The effect size was calculated with Cramer's V statistic. The benchmarks were the following: 0.10-0.30 weak, 0.30-0.50 medium and higher than 0.50 strong (Coe, 2002;Cohen, 1988;Ellis, 2010;Weisburd & Britt, 2014). All the variables related to victimisation and were subsequently compared by gender, age and nationality.

Limitations
Although we overcame many obstacles, some limitations need to be pointed out. Firstly, it is a convenience sample due to its difficult accessibility and because some of the users of the centres could not participate owing to a language barrier. Secondly, one of the centres was for drug and alcohol rehabilitation, so alcohol and drug-related variables could be affected, and the sample of women was low. Thirdly, the variable "mental health problem" was not defined in the questionnaire and it may cover a wide number of situations. Similarly, the aporophobic motivation is difficult to measure and the crimes could be also related to the situation of vulnerability of the victim (Chakraborti & Garland, 2012). This has an impact on the interpretation of the results. Fourthly, related to the research technique used, victimisation surveys may overestimate the rate of certain crimes and inflate the rate of others (Serrano, 2022). Therefore, the results presented here are not generalisable, but rather represent an exploratory study. Also, some respondents may have normalised violence and not identified themselves as victims. Moreover, some of them may not have been able to identify the motivation of the perpetrator and, conversely, may have identified a different motivation for the crime. Consequently, aporophobic crimes might be overestimated or under-estimated. Finally, the Likert Scale was applied because of its simplicity and ease. However, it also has its disadvantages. The difficulty of not being able to go deeper into the experiences of discrimination and victimisation was overcome with the open-ended questions. Nonetheless, there are other factors that should be considered when interpreting results, such as response set bias or acquiescent response set (United Nations, 2010).

Sociodemographic Characteristics, Homelessness Duration and Others
Related to sociodemographic characteristics (see Table 2), most of the sample was male. The average age was around 42 years old. The predominant nationality was Belgian, especially women-with more than 85%-while the most frequent foreign nationalities were Tunisian, Turkish-both with dual nationality-and Polish with a frequency of two. The rest of the sample population was Dutch, Algerian, Kenyan, English-with dual nationality-and Syrian. Foreign people were in most cases younger than Belgian people. Likewise, more than half of the sample had completed secondary education, which was the 11 Chi-Square test was not applied because of the sample size. category with the highest percentage, but it is highlighted that four of them had university education-three men and one woman. Despite the fact that 30% of them did have serious mental health illnesses-in fact, one of them was waiting for a place in the psychiatric centre, 9.4% of them preferred not to say. On this variable, there were no gender differences.
Regarding homelessness duration (see Fig. 1), most of the sample had been homeless for more than 5 years in their lifetime. However, the time they had spent living on the street was lower. Similarly, 58.1% of the sample were currently living on the street, compared to 41.9% who were no longer living on the street when they filled in the questionnaire. The most frequent duration of this last period was more than 3 months and less than a year.
Prior to their homelessness, 59.4% of the sample had alcohol or drug abuse problems. Meanwhile, the question about current consumption of alcohol was most frequently answered as "sometimes", "very often" and often-28.1%, 21.9% and 15.6% respectively. The use of drugs was lower-50.0% of the sample, with the most frequent responses being "never" and "rarely"-25.0% each one. Furthermore, more than half of the sample had been fined, arrested or detained by the police, and a similar percentage had been in prison. Nevertheless, the homeless sample pointed to higher frequencies of problems with the police than being in prison. In relation to the police, one male respondent told us that he had been detained on numerous occasions-spending the night in a cell each time-and that if you added them all up it would be about a month. Another woman was fined for illegal dumping of trash. Also, two employees of the shelter said that the police had kept their identification documents after releasing them. One of them added that some police officers took money from homeless people to allow them to beg-because it is forbidden-and sometimes also committed abuse and intimidation. On the other hand, in regard to prison, three respondents included the duration: 10 years, 6 months and 14 days. Finally, a social worker told us that many of the users had been traumatised by their time in prison, which was corroborated by two homeless people.

Aporophobic Discrimination and Victimisation
The results showed that 21 homeless people felt discriminated against for living on the street and 29 indicated that they had suffered at least one crime while they were living on the street. In this regard, one respondent said that "this is the first time that anyone has asked me about it". Some respondents who had not experienced criminal offences said that they tried not to appear "homeless" in order to not to be treated differently and not be victimised and another respondent pointed out that, although he had never been a victim, he knew that his homeless friends had suffered from crimes. In the same way, all employees and volunteers stated that homeless people often suffered from criminal offences. One of them asserted that "when you get the trust of the people, they tell you a lot". Moreover, victimisation sometimes occurs in shelters, and according to the workers, this is one of the reasons why some people, especially women, do not want to go there. The offence which happened the most often was insults and humiliations, and the least often was physical violence or assault (see Table 3). The frequency in which the crimes occur is similar, as "sometimes" was the answer used most frequently in all of them-excluding never. However, a small difference could be highlighted. While threats, 12 damage of their belongings and thefts/robberies were more  Fig. 1 Homelessness duration of the sample (%) 12 One user added the following sentence "a lot of them was from the police". frequent, especially the last mentioned, 13 insults and humiliations and physical violence or assault were less common (see Fig. 2). Similarly, one professional stated that verbal abuse was much more common than physical abuse, and that robbery and theft were the most frequent crimes-"items such as phones, wallets and cigarettes are often stolen", he said. In this sense, people who got drunk, especially in shelters, could be targeted by others-this information obtained by a worker was also confirmed by one respondent.

Aporophobic Victimisation
Regarding aporophobic motivation, 14 homeless people-43.75% of total sample-who had been victimised stated that the reason for some of these crimes was their poverty situation (without gender differences, but being more Belgian-66.7%-than foreign); 28.1% answered no (one of them said that the reason was her addiction); and 17.9% of the sample did not know. Victims of aporophobic crimes indicated that victimisation occurred frequently, especially property crimes (see Fig. 3). All of them had experienced thefts or robberies and almost 80% had suffered damage to their belongings (see Table 3). Additionally, victims were asked why they believed the motivation was aporophobia, giving four reasons. 69.2% of them agreed that "the aggressors saw me as more helpless and vulnerable" and 50% that "the aggressors insulted me by referring to their homelessness". With lower percentages, 42.9% and 28.6%, the following reasons were found: "this kind of thing only happens to people living on the street" and "the aggressors said so explicitly". The abovementioned results with regard to aporophobic victimisation have to be interpreted with caution. Strictly speaking, only the motivations "the aggressors insulted me by referring to my homelessness" and "the aggressors said so explicitly" (see Table 4 victimisation (these items were pointed out by 7 homeless people). The other explicit motivations "this kind of thing only happens to people who live on the streets" and "the aggressors saw me as more helpless and vulnerable" could also refer to crimes committed due to the vulnerability of the homeless person and not out of "hatred" in the strict sense (Teijón, 2022). In these results, this distinction was not always made. Differences between whether the crime is committed because of the victim's vulnerability and because of the underlying hatred are not easy to assess, let alone for victims to report on it. Moreover, aporophobic motivation and the vulnerability of the victim, especially of rough sleepers, overlap and are considered to be intrinsically linked (García, 2023). By using Fisher's exact tests no significant differences were found between experiencing aporophobic victimisation (including discrimination) and sociodemographic characteristics or prior alcohol or drug abuse (p > 0.05), nor between types of aporophobic victimisations and the variables mentioned (p > 0.05).

Offender
The offender was perceived as a homeless person at least once in 60% of the victimised sample, similar to the percentage found in the aporophobia sample (61.5%). Physical violence or assault was very often reported, with thefts/robberies and damage to their belongings also occasionally occurring. There was no relationship found between the offender's homelessness and sociodemographic characteristics, homelessness duration, living on the street duration and prior alcohol or drug abuse, or types of victimisation (p > 0.05). However, the category "physical violence and assault" was never perpetrated by homeless people in more than 50% of cases, and in all other categories it was less than 30%. Nonetheless, threats were commonly committed by homeless people, even though the most frequent answers given by respondents in the questionnaire were "rarely" or "sometimes". Last but not least, damage to their belongings was very often or sometimes done by the homeless people themselves. This was the case in 70% of the aporophobic victimisation sample.

Report
Despite the fact that more than half of the sample-66.67%-reported at least one of the crimes suffered, the mode in all types of crimes was "never". The most reported crimes were thefts and robberies, representing more than half of them-54.20%. The rest of the offences were less reported. On the one hand, insults and humiliations, damage to their property and threats were sometimes and rarely reported. On the other hand, thefts/robberies and, especially, physical violence or assault (including sexual assault) were more frequently reported. Nevertheless, one respondent who reported thefts three times said "The police do not care about it"; similarly, another added "we do not exist for official institutions". Regarding aporophobia, the report of at least one crime experienced due to their poverty situation was made by around 65% of the sample. Of the types of offences which they reported, the least common were damage to their belongings, and the most common were physical violence or assault (including sexual) and thefts/robberies, both with a percentage higher than 50%. However, it stands out that physical violence or assault (including sexual assault) did not have moderate frequency values-people who reported these crimes stated to report them "very often", "often" or "rarely" with the following percentages 15.4%, 15.4% and 21.4%. Furthermore, "sometimes" was the mode for insults and humiliations, as well as thefts/robberies-excluding the category "never"-and "rarely" for threats.
The reasons for not reporting the crime were several. The highest level of agreement among respondents was with "nothing would be done about it" and "mistrust of the police". 14 In fact, around 75% of the victimisation sample agreed with both. With a lesser percentage of agreement, around 50%, were "I solved it by myself", "lack of evidence" and "fear of reprisals". For the rest of the reasons, the percentage of agreement was lower, around 30%. In relation to reasons for underreporting of the aporophobia sample, 92.3% and 76.9% of them strongly agreed or agreed with "nothing would be done about it" and "mistrust of the police", while none disagreed with the first reason mentioned and only one disagreed with the second. In spite of the fact that more than half of the sample agreed with "fear of reprisals", "lack of evidence" and "solved it by themselves", for the rest of the reasons, i.e., "not considering serious enough", "reporting to other places" and "not knowing how to report it", the degree of agreement varied between all the categories. In addition, it is highlighted that 38.5% of the sample victimised because of aporophobia did not know how to file a report.
Finally, Fisher's exact tests indicated that there was no relationship between reporting or typologies of crimes reported and sociodemographic characteristics, homelessness duration, living on the street duration and prior alcohol or drug abuse, neither between "nothing would be done about it" and "mistrust of the police", nor for all the categories mentioned (p > 0.05).

Experienced Crimes and Other Comments
Respondents informed us of crimes experienced, some of them because of aporophobia. In relation to intentional injuries, one of the most serious crimes was an attempted homicide. While a man was sleeping in a destroyed building, four males set fire to the building and blocked the emergency door. He claimed that they had seen him and stated his belief of an aporophobic motivation. However, he did not report the crime because he mistrusted the police and thought that nothing would have been done about it. Another male had a similar incident but stated that the perpetrator was a police officer. Furthermore, we knew about sexual criminal offences. In this sense, one respondent experienced an attempted rape while he was sleeping in the street. The offender, who was a male, drugged him and tried to commit the crime. Fortunately, he woke up when he was going to rape him and fought back. Other incidents were related to theft in their sleep-e.g., a boy who was sleeping with his phone in his underwear when someone stole it. In addition, one girl pointed out "if you are homeless, people see you as a minor group. They can do anything what they want with you". Also, there was a special story of a refugee from Syria who had been beaten many times by Turkish police. Last but not least, some respondents added acknowledgments, such as "thank you for the attention" or "no thanks". Some shelter workers were also very appreciative of the research and thanked us for researching the victimisation of homeless people on the grounds that no one had done it before.

Discussion and Conclusions
The main aim of the empirical research carried out was to measure the victimisation of homeless people and their victimisation because of their extreme poverty situation, i.e., aporophobia, in Ghent. The main profile was male, adult and Belgian, very similar to the Ghent homeless Census (Hermans et al., 2021) and to other foreign studies such as Rais Fundación's (2015) research. Aporophobic discriminatory incidents and hate crimes were experienced by around 50% of our sample, with a frequency ranging from "very often" to "rarely". This is in line with earlier studies in Spain (Ávila & Garrido, 2019;García, 2019;García et al., n.d;López et al., 2021;Muñoz et al., 2017;Rais Fundación, 2015). In fact, we found that it is common for them to experience more than one offence, as does the Rais Fundación (2015) research. Moreover, our results support earlier findings on the particular impact of hate crimes on homeless people (National Coalition for the Homeless, 2020; Puente, 2019). The rate of types of victimisations was higher in our sample than others, especially for crimes related to property. Most of them believed that the aporophobic motivation was due to offenders seeing them as more helpless and vulnerable, among other reasons, such as offenders explicitly saying so. These findings are similar to those found by Rais Fundación (2015). Although vulnerability and aporophobic crimes are very closely linked (García, 2023), these results must be interpreted with caution. Some crimes may be committed because of the vulnerability of the victim (only) and not out of hatred. Therefore, strictly speaking, they might not be defined as aporophobic. However, this could also be related to the vulnerability and the situation of helplessness of the victim due to the fact they have no home and are forced to live on the street. Likewise, we did not find a significant relationship between sex and victimisation (including sexual assault). This differs from earlier findings which discovered that women were more victimised (Ávila & Garrido, 2019;Rais Fundación, 2015), especially with regard to sexual assault, and considering that approximately 10% of the homeless population are women (García, 2020a). On the contrary, García et al. (n.d) identified that being male was associated with a higher likelihood of being a victim. Furthermore, our findings showed that most perpetrators of aporophobic crimes came from their own group, i.e., people experiencing homelessness, mainly among female victims, as well as the police-though in a much lower proportion. These results are similar to what other studies had highlighted. Nonetheless, underreporting was much less pronounced in our study compared to the 10-20% reported in the literature review. Finally, the most frequent response in our research across all crimes for reporting was "never", with the majority believing that nothing would be done. Both results are consistent with findings from other studies (Ávila & Garrido, 2019;García, 2020a;Rais Fundación, 2015).
This study provides an empirical contribution to discussions about conceptualization, discourses and policies on hate crimes. It shows that homeless people and aporophobic discrimination deserve more attention. Indeed, victimisation of homeless people does occur and is sometimes linked to their situation of extreme poverty, as well as risk behaviours they engage in (Puente, 2023). More empirical studies, in other countries and locations, could add to sharpen our view and thoughts on aporophobic discriminatory incidents and hate crimes, with an ultimate view to do justice to victims and reduce these kinds of victimisation (Bustos, 2020;García, 2020a).
For policymakers, in Ghent, Belgium and beyond, this study could raise their awareness of the existence, shapes and forms of (aporophobic) victimisation experienced by homeless people. Official data collection activities and victimisation surveys could be made more susceptible to catch the specifics of such crimes and consider strategies for effective inclusion of aporophobic experiences of homeless people. Criminal justice actors could be more open to detect, prosecute and punish those responsible for (aporophobic) crimes against the homeless. As far as prevention is concerned, there are choices for the society to be made. Instead of ignoring homelessness or even criminalising poverty, inclusive approaches towards homeless people could be taken with policies and measures that address both the situation of homeless people and society's stance towards their vulnerability. The recognition of aporophobia as a serious social problem with known harmful consequences, is a first step to make. A step to which this paper hopes to contribute. Currently living on the street/Frequency living on the street Prior to your homelessness, did you have any alcohol or drug abuse related problems?

Appendix 1
Prior alcohol/drugs abuse While you were living on the street, how often did you drink alcohol or take drugs?

Alcohol/drugs
While you were living on the street, how often have you been fined, arrested or detained by the police?
Issues with the police While you were living on the street, how often have you been in prison?
Prison Table 4 (continued) Question Variable Clusters While you were living and sleeping on the street (as a homeless person), how often did you feel discriminated for living on the street, that is, for your situation of poverty?

Discrimination Discrimination and victimisation
While you were living and sleeping on the street (as a homeless person), how often: -Have you been insulted or humiliated? -Have you been threatened? -Has someone damaged your belongings? -Has someone stolen your money, belongings, documentation (passport, ID card…) or other objects?
-Have you experienced physical violence or assault (including sexual assault)?

Types of crimes/ Victimisation
If you have suffered any of the incidents and crimes mentioned in the last question, do you think that the reason was because you were living on the street, that is, because of your situation of poverty?
If you answer yes, how often did you think that the reason was your situation of poverty for the following incidents and crimes: (1) insults and humiliations, (2) threats, (3) damage to their belongings, (4) thefts or robberies and (5) physical violence or assault (including sexual assault)

Motivation/ Types of crimes
Why do you think that the reason was that you were sleeping and living on the street, i.e., your poverty situation? Please indicate the frequency of the following statement: (1) the aggressors said so explicitly (2) they insulted me by referring to my homelessness (3) this kind of thing only happens to people who live on the street; (4) they saw me more helpless and vulnerable; and (5) others

Motivation
How often the offender was a homeless person? This question asked about each crime individually: (1) insults and humiliations, (2) threats, (3) damage to their belongings, (4) thefts or robberies and (5)  How often you have filed a report for the following behaviours: (1) insults and humiliations, (2) threats, (3) damage to their belongings, (4) thefts or robberies and (5) physical violence or assault (including sexual assault)

Report
If you did not file a report, what was the reason for not reporting it/them to the police? Please indicate the degree of agreement with the following statements: (1) "I did not consider it serious enough"; (2) solved it by myself (I knew the offender); (3) "I reported it to the emergency center, shelter, soup kitchen, or another recourse that I went to"; (4) lack of evidence"; (5) "believe nothing would be done about it"; (6) mistrust of the police; (7) fear of reprisals; (8) "I do not know how to report it"; and (9) others Reasons of not reporting