Repeat violent victimisation: evidence review

3. Synthesis of the Evidence 

This section will synthesize the findings from the evidence base in regard to the research questions stated in the Introduction. 

The Extent and Prevalence of RVV

A range of studies have sought to establish the extent and prevalence of RVV using crime surveys. Crime surveys typically ask a representative sample of people which crimes, if any, have been committed against them over a fixed period. The rate of RVV can be estimated by calculating the percentage of incidents of violence that are repeated against the same persons over the specified period. 

Evidence from several crime surveys has shown that repeat victimisation represents a significant proportion of all violent offences. In Scotland, the most recent SCJS has shown that although fewer than 1 in every 100 adults suffered RVV in 2017/18, their experiences accounted for more than half (59%) of all violent crime (Scottish Government, 2019). These repeat victims are estimated to have experienced on average three violent crimes each during 2017/18, though a small proportion of the population (0.1%) were high frequency repeat victims experiencing five or more incidents. 

Similar results have been shown in previous sweeps of the SCJS, although the proportion of adults experiencing two or more and five or more incidents of violence were all lower in 2017/18 than in 2008/09 (Scottish Government, 2019). However, there have been fluctuations in these figures across this time period. For example, those experiencing two or more violent crimes has been below the 2008/09 baseline since 2010/11 with the exception of 2016/17 when the estimate was not significantly different to the 2008/09 figure. The return to a significant decrease in 2017/18 suggests that the 2016/17 figure may have been an outlier in an otherwise declining trend. On the other hand, the proportion of adults experiencing five or more violent crimes has shown a less consistent trend and has only been below the 2008/09 baseline figure in 2014/15 and 2017/18. As such, it will be important to monitor these findings into the future to see whether the lower victimisation rate seen in 2017/18 is maintained.

More detailed analysis of repeat victimisation in Scotland using crime survey data was conducted by Shaw and Pease (2000). Using the 1982, 1988, 1992 and 1996 sweeps of the then Scottish Crime Survey (SCS), Shaw and Pease (2000) demonstrated that a range of crimes are disproportionately suffered by repeat victims, with the probability of becoming a victim again increasing as the number of prior victimisations increase. Although Shaw and Pease (2000) did not focus specifically on violent crime, their analysis included ‘crime against the person’, covering assault, personal theft and robbery.  

Similar results are found elsewhere in the UK. The most recent Crime Survey for England and Wales (CSEW) has demonstrated that a disproportionately large number of violent crimes are suffered by a small number of victims experiencing RVV (ONS, 2019). The findings showed that in the year ending March 2018, 57% of violent incidents were experienced by repeat victims. 18% of all victims of violence were victimised twice, while 7% were victimised three times or more. Further, a study by Ignatans and Pease (2015) analysed data from a total of close to 600,000 respondents to the CSEW over a 30 year period (1982-2012), and found that although crime has declined in absolute terms, the proportion of crime accounted for by those most victimised has increased. Ignatans and Pease (2015) examined three general crime types (vehicle, property and personal), with this pattern applying to each. The personal crimes examined included wounding (where the incident results in severe or less serious injury), assault (where the incident results in minor injury), robbery and theft from the person^[2]. 

At the international level, the International Crime Victims Survey (ICVS) has also demonstrated that a high proportion of violent crimes occur against individuals that have already been victimised. The ICVS has been carried out six times over the period 1989–2010. Although national samples are relatively small, the ICVS is standardised and far-reaching, and has been conducted in more than 80 countries (Kesteren et al., 2013). In the ICVS, respondents who report victimisation of a particular type are asked how often they have been victimised by such a crime in the course of the last year. It is therefore possible to determine the proportion of crimes repeated against the same persons, across country and crime type, over the period of the last 12 months. The ICVS has consistently shown that across all included countries RVV constitutes a large proportion of all violent offences, ranging from 16% to 39% depending on the type of violence considered (Kesteren et al., 2013). 

However, while it is possible to assess the prevalence of RVV with good precision using crime surveys, there are limits to what this data can tell us about RVV. In particular, in order to reliably estimate trends of incidents of crime, crime surveys tend to cap the number of incidents which can be reported as a series, commonly at 5 (Farrell and Pease, 2007; Lauritsen et al., 2012). The cap ensures that survey estimates of incidence are not disproportionately affected by a very small number of respondents who report an extremely high number of incidents, which can be highly variable between survey years. This enhances the ability of the survey to monitor underlying trends consistently (Grant et al., 2016), but limits the influence of the relatively small number of victims who yield a high number of violent victimisations on incidence estimates. Nonetheless, even with a cap in place it is clear that experiences of crime are dominated by a group of repeat victims of violence. 

An additional problem with crime surveys is that they are time-based, typically asking individuals about their experiences during a recall period of one year. This means that some victimisations will appear to be single incidents but may be repeats of crimes suffered in the previous year, or the first in a series extending into the next year. Finally, crime surveys rely on recall; it is possible that respondents may forget or exclude crimes they suffered (Shaw and Pease, 2000). These factors mean that even the large concentration of violent crime found in crime surveys is likely to be an underestimate (ONS, 2016a; Farrell and Pease, 1993).

A complementary measure of RVV comes from police recorded crime. A range of studies have examined the extent of RVV using this type of data, and have also demonstrated that a high proportion of violence is experienced by repeat victims (Lloyd et al., 1994; Hanmer et al., 1999; Shaw and Pease, 2000; Taylor, 2004; Matthews et al., 2001; Sampson and Phillips, 1992). Shaw and Pease’s (2000) research is particularly relevant, given that it was conducted in Scotland. Shaw and Pease obtained data from police recording systems in three divisions: Falkirk, Maryhill and Dundee. Although the authors did not focus specifically on violence, they did collect data on assault. They used data collected over the course of a year, and found that around 10 per cent of those suffering assault suffered two or more incidents during the period represented. 

However, this research is now dated. Moreover, there are limitations to identifying repeat victims using police data. In particular, it is well known that police data tend to underestimate the extent of crime, as not all incidents are reported or recorded. The issue of underreporting of crime to the police is compounded in the case of repeat victimisation, as repeat victims are less inclined to report crimes to the police than others (Weisel, 2005). Further, even if a crime is reported to the police, there are obstacles in its identification as a repetition of an earlier crime due to police recording practices (Shaw and Pease, 2000). 

One way to address the shortcomings with survey and police recorded data is to supplement the findings with qualitative research, which provides a more detailed insight into the nature of RVV. Shaw and Pease (2000) adopted such an approach in their investigation of repeat victimisation in Scotland, conducting interviews with repeat victims and offenders, which showed the familiar repeat victimisation phenomena: crime being disproportionately suffered by repeat victims. However, although relevant, their study was conducted nearly 20 years ago, and does not focus specifically on violence. Indeed, this is reflective of a general paucity of qualitative research in the literature on RVV, with most studies assessing the extent of RVV using quantitative data. In the absence of additional measures of RVV beyond the SCJS in Scotland, it is difficult to fully ascertain the current extent and nature of RVV.

What Types of Violence are Repeated?

The SCJS definition of violence is comprehensive, covering assault (including serious assault, attempted assault, minor assault with no-negligible and minor injury) and robbery (Scottish Government, 2019). However, at present, the small number of repeat victims in annual SCJS samples makes it difficult to look at their experiences in detail from any one sweep, including the types of violence which they experience (Grant et al., 2016)^[3]. Further, although Shaw and Pease (2000) analysed ‘assault’ as a distinct category in their analysis of police recorded crime, they did not individually examine the different types of violent crime which fell under the broad category of ‘crimes against the person’ in their analysis of the SCS. 

RVV is broken down by offence type in the CSEW. The latest CSEW found that RVV was most common for violence without injury (25% of victims), followed by assault with minor injury (23%) and violence with injury (21%) (ONS, 2019). In addition, RVV is also broken down by the relationship between the victim and the perpetrator in the CSEW – stranger, acquaintance or domestic. While domestic violence is not the focus of the present RER, evidence from the CSEW suggests that it is an important component of RVV, with the most recent results showing that a higher proportion of victims of domestic violence were repeat victims (34%) than victims of acquaintance violence (27%) and stranger violence (15%). However, these results come from the face-to-face interview of the CSEW, which are impacted by the cap on the number of incidents which can be reported as a series. In addition, domestic violence reported in this way is likely to be an underestimation because of the high level of underreporting of this type of violence in face-to-face interviews^[4] (ONS, 2019).

Hence, the self-completion module tends to capture higher levels of domestic violence than the face-to-face module. This module also employs a broader definition of domestic ‘abuse’, covering non-physical abuse, physical abuse (including threats or force), sexual assault and stalking, as opposed to just physical violence (ONS, 2016b). The proportion of victims of domestic abuse experiencing repeat victimisation in the CSEW was last reported on in 2016, which showed that over 30% of domestic abuse victims suffered more than one victimisation (15% experienced 2 victimisations and 16% experienced 3 or more victimisations) within the 12 month reference period (ONS, 2016b). The 30% of victims of domestic abuse who experienced more than one victimisation amounted to over half (60%) of domestic abuse incidents estimated by the CSEW (ONS, 2016b). However, nearly 70% of respondents opted not to provide an answer to this question, reducing the estimate of the proportion of victims who were victimised more than once (ONS, 2016b). 

Similarly, the SCJS partner abuse self-completion module is reported on biannually. ‘Partner abuse’ is defined in the SCJS as any form of physical, non-physical or sexual abuse, which takes place within the context of a close relationship, committed either in the home or elsewhere (Scottish Government, 2019). The latest report showed that of those who reported partner abuse within the 12 months prior to interview, 12% experienced two incidents, 6% experienced three incidents and 5% experienced four or more incidents (Scottish Government, 2019). A further 13% said that there were too many incidents to count. Overall, around three in five respondents (61%) had experienced more than one incident^[5]. In addition, just under two-thirds (63%) of those who reported an incident of partner abuse in the 12 months prior to interview also reported at least one incident prior to this period.

The ICVS has also broken down repeat victimisation by type of crime (Farrell et al., 2005). The ICVS covers 11 crime types, including three types of violence: assault and threats, robbery and personal theft. RVV was most prevalent for assault and threats, with 39% of victims being repeat victims in the last 12 months, followed by robbery (22% of victims) and personal theft (16% of victims).  Repeat victimisation is more common for these types of crime than others, such as property crime (Kesteren et al., 2013). This is reflected in evidence from other studies, which have also shown that although repeat victimisation occurs for all types of crime, rates of repeat victimisation are on average highest for violence (Graham-Kevan et al., 2015). Despite the lack of Scottish-specific evidence, therefore, the CSEW and the ICVS suggest that the types of violent crime which are repeated include assault, threats, robbery and theft. In addition, domestic violence is an important component of RVV. 

Moreover, while this RER is focused primarily on repeat victimisation, research not covered in this review has highlighted the overlapping nature of this phenomenon with multiple victimisation. As above discussed, repeat victimisation refers to those who are victims of the same type of crime more than once, whereas multiple victimisation refers to those who are victims of multiple types of crime (Scottish Government, 2019). However, it has been highlighted that repeat victims often become more likely to be victimised more than once by another crime type (e.g. Hope et al., 2001). Indeed, Shaw and Pease (2000) examined cross-crime type sequences and demonstrated that while assaults best predicts future victimisation of the same type, victimisation by violence follows repeat housebreaking three times more often than one might expect.

Who are the Victims of RVV?

It is well established that the likelihood of experiencing violent crime is not evenly spread across the population, and varies across characteristics. For example, the most recent CSEW has shown that men, younger people, those who are single, those who are unemployed and those living in deprived areas are most likely to be victims of violent crime (ONS, 2019). In Scotland, the most recent SCJS showed that younger groups and those living in the 15% most deprived areas were more likely to experience violence than others (Scottish Government, 2019). 

In the context of RVV, the evidence shows that repeat victims of violence also tend to have particular characteristics. For example, Jansson et al. (2007) analysed the 2006/07 sweep of the then British Crime Survey (BCS), covering England and Wales, and found that repeat victims of violence are often young, male, non-white and from deprived socioeconomic backgrounds. In Australia, analysis of the nationally representative National Crime and Safety Survey (NCSS) showed that men and younger people were more likely to be repeat victims of personal crimes (including assault and robbery) than others (Mukherjee and Carach, 1993). More recently, a review of 106 studies on repeat victimisation found that the risk factors for being a repeat victim of violence include being male, younger, not married, on a low income and unemployed (van Reemst et al., 2013). 

There is also research from the US which provides some insight into the characteristics typically associated with experiencing RVV. For example, in their analysis of victims of violence who were admitted to hospital over a four-year period in Ohio, the US, Buss and Abdu (1995) found that compared to one-time victims, repeat victims were more likely to be non-white and from deprived socioeconomic backgrounds. Similarly, Cooper et al. (2000) conducted a case-control study of persons admitted to a trauma centre due to violent assault in Baltimore, the US, over a 16-month period. They found that victims of RVV were more likely than single victims to be African American, male, unemployed and on a low income.

Outlaw et al. (2002) also demonstrated significant individual-level predictors for RVV in their analysis of around 5,000 residents of Seattle, the US. In particular, being non-white and having a low income were associated with experiencing more than one violent victimisation in the last two years. However, these studies are based on relatively small, non-representative samples in the US, making it difficult to determine the extent to which they apply to Scotland. Indeed, it is likely that a similar study would produce different results in Scotland, given the different ethnic composition of the US. 

Although most research finds that men are more likely to experience RVV than women, the exception to this is in cases of domestic violence. As above discussed, while not explored extensively in this RER, evidence from the CSEW has shown that when RVV is broken down by the relationship between the victim and the prepetrator, domestic violence emerges as an important component of RVV (ONS, 2019). However, in the case of domestic violence, research has shown that women tend to compose the majority of victims (Walby and Towers, 2017). 

In summary, while it is possible to infer from the existing literature who the victims of RVV are likely to be, at present there is no Scotland-specific research that has sought to identify the characteristics of repeat victims of violence. Although the SCJS provides useful estimates of the proportion of adults who have been a victim of violence more than once, more detailed analysis would be required to establish the exact characteristics of these individuals. The small sample size of repeat victims of violence in the SCJS also means that any difference found in characteristics between single and repeat victims may lack statistical significance. It is therefore unclear if the characteristics associated with being a single victim of violence are the same for repeat victims in the Scottish context.  

Why Does RVV Occur?

The research suggests two general explanations for RVV (Clay-Warner et al., 2016; Tseloni and Pease, 2004). The first perspective views subsequent victimisation as the direct consequence of prior victimisation, which has come to be known as the ‘state dependence’ perspective. This perspective suggests that being victimised once can change individuals or their social circumstances in ways that heighten or ‘boost’ the risk of future victimisation (Clay-Warner et al., 2016). In particular, it has been argued that the experience of victimisation puts into motion a ‘victim labelling’ process that enhances a person’s risk of future victimisation. That is, when a person experiences an initial victimisation it can make them appear vulnerable, which leads potential offenders to view them as targets (Ousey et al., 2008).

It is also argued that experiences with victimisation can result in changes in people’s activities and lifestyle that elevate their probability of future victimisation (Ousey et al., 2008). For example, in response to a previous victimisation experience, individuals may withdraw from pre-existing pro-social attachments and commitments and become socially marginalised (Schreck et al., 2006). Victims who withdraw may find themselves isolated from the conventional social ties that ordinarily constrain their involvement in high-risk activities, such as using drugs or associating with delinquent peers, which can in turn increase exposure to offenders (Schreck et al., 2006). Victims of violence may also retaliate, which increases victimisation risk via counter retaliation (Jacobs and Wright, 2006). 

State dependence effects can also result from knowledge gained by the perpetrator after the first offence is committed. For example, taking the example of domestic violence, if during the first incident the police are not called, neighbours do not intervene, or the victim’s family and friends do not become aware, the perceived risks for the offender are lower on subsequent victimisations, and they are more likely to perpetrate again. According to this perspective, therefore, prior victimisation causally impacts subsequent crime (Ousey et al., 2008). 

In contrast, the population heterogeneity perspective proposes that the relationship between victimisation and repeat victimisation is spurious, because the factors that increase one’s risk of being victimised the first time are the same factors that lead to subsequent victimisation. There are a range of factors which can make an individual more vulnerable to violence. For example, an individual’s occupation can make them vulnerable to being repeatedly victimised, as some jobs have higher risks of violence than others (e.g. the police, security guards, nurses, care workers and public transport workers) (Health and Safety Executive, 2018). People who spend time in particular places, such as bars and nightclubs and on public transport, are also at greater risk of violence (Finney, 2004; Gerrell, 2018). In addition, some areas have higher rates of violent crime than others; living in such areas keeps one vulnerable to RVV (Morenoff et al., 2001). 

Demographic characteristics also play a role in the population heterogeneity perspective; as above discussed men, younger people and those living in deprived areas are more likely to be victimised by violence than others (ONS, 2019). The exception to this is domestic violence, where research has shown that women are more likely to be victims than men (Walby and Towers, 2017). Possessing such characteristics, therefore, makes one susceptible to repeated victimisation. The population heterogeneity perspective therefore argues that victimisation has no independent effect on repeat victimisation; rather an underlying stable factor determines both victimisation and repeat victimisation (Turanovic and Ogle, 2017). 

Although the state dependence and population heterogeneity perspectives are often treated as competing, most research finds that neither state dependence nor population heterogeneity alone can explain RVV. Instead, both contribute to risk of RVV (Clay-Warner et al., 2016; Daigle et al., 2008; Everson, 2003; Tseloni and Pease, 2003). For example, using propensity score matching with longitudinal data from the National Crime Victimisation Survey (NCVS) in the US, Clay-Warner et al. (2016) found that only those with high levels of underlying propensity for violent victimisation (due to being younger, male, non-white and non-married) experienced a ‘boost’ in risk from a victimisation event. Their results indicate that the same factors that predispose individuals to violent victimisation also amplify the effects of a victimisation event on odds of future victimisation.

A competing perspective within the literature posits that victimisation can actually reduce the risk of future victimisation, as experiencing victimisation makes people aware of risk, which motivates them to take self-protective action to prevent future incidents (Averdijk, 2011, cited in Clay-Warner et al. 2016). However, there is little empirical support for this perspective. Indeed, in demonstrating that victimisation increases the likelihood of future victimisation, several studies have directly refuted this argument (Ousey et al., 2008; Lauritsen and Davis Quinet, 1995; Wilcox et al., 2006).

When Does RVV Occur?

In the wider literature on repeat victimisation, it has been demonstrated that after the initial incident, repeat offences tend to occur quickly (Weisel, 2005). The evidence shows that the risk of a repeat occurring is greatest in the period immediately after victimisation, with many taking place within a week of the initial offence (Farrell and Pease, 1993). After this period of heightened risk, the chance of a repeat declines rapidly until the victim has around the same risk of victimisation as those who have not been victimised. This pattern of a relatively short high-risk period, followed by a decline and levelling off of risk, is referred to as the ‘time course’ of repeat victimisation, and has been demonstrated consistently for crimes such as burglary and car crime (min Park and Eck, 2013; Weisel, 2005). 

There is also a smaller body of evidence which suggests that this time course is applicable to violence. For example, in their study of the RVV of female college students in the US, Daigle and Fisher (2008) found that there was an elevated risk of repeat violence in a short time. In particular, the elevated risk was greatest within the same month. Shaw and Pease (2000) also demonstrated a similar pattern for assault using police recorded crime data. Studies focusing on domestic violence have also shown that subsequent victimisations tend to occur soon after the initial offence. For example, Lloyd et al. (1994) examined domestic violence incidents reported to the police on Merseyside, and found that 15% of repeat offences occurred within a day. However, it is unclear from the existing evidence whether this holds true for all forms of violent crime. 

RVV and Repeat Offending: The Overlap

The correlation between victimisation and offending is a well-established empirical finding. Various studies have demonstrated that one of the most reliable predictors of violent victimisation is offending (Berg et al., 2012; Jennings et al., 2012; Lauritsen and Laub, 2007; Schreck et al., 2008), which has come to be known as the ‘victim-offender overlap’. Given the durability of this finding, it has been argued that criminal behaviour has the potential to be an important risk factor for experiencing RVV (Tillyer, 2013). Indeed, Tillyer (2013) proposes various reasons why offenders run an enhanced risk of being repeat victims of violence. For example, from the lifestyles-routine activity perspective, a criminal lifestyle might repeatedly present opportunities for victimisation. 

There is evidence to support the notion that offending may be an important risk factor for RVV. For example, using data from the Youth Lifestyles Survey, Deadman and MacDonald (2004) found that when compared to non-offenders, violent offenders were more likely to be repeat victims of assault and theft. Similarly, in their analysis of two waves of data from the US National Longitudinal Study of Adolescent to Adult Health, Farrell and Zimmerman (2017) found that exposure to violence (including both being a victim of violence and witnessing violence) increases violent offending risk, with the strongest effects found for multiple victimisation and repeat exposure to violence. 

However, Tillyer (2013) argues that despite the fact that criminal behaviour is likely to be a significant risk factor for violent victimisation, many RVV studies fail to account for it. At present, therefore, the extent to which offending contributes to RVV risk appears to be a gap in the evidence base. Tillyer (2013) highlights that any research seeking to establish the characteristics associated with RVV should therefore consider the important overlap between victimisation and offending. This is particularly important given research which suggests that victims who are involved in criminal behaviours are less likely to report their victimisation to authorities (Berg et al., 2012), making it unlikely they will receive any sort of victim services. Tillyer (2013) argues that offering victim support and prevention services to individuals in correctional settings is one way to reach a victim population that is at high risk for subsequent victimisation.

Preventing RVV

Given the evidence regarding the extent and prevalence of RVV, it has been argued that focusing on RVV may be an effective and efficient means of preventing and reducing violent crime, because it focuses resources on where violence is most concentrated (Grove et al., 2012). Indeed, Grove et al. (2012) conducted a systematic review of the effects of initiatives to prevent repeat victimisation, and found that appropriately targeted measures can significantly reduce repeat crimes. Although their review primarily found evaluations relating to burglary, they also identified evaluations relating to domestic violence (e.g. Robinson, 2006; Morgan, 2004).  

In terms of the kinds of approaches and measures to implement, it has been argued that safety and protection from re-victimisation is an important way to reduce the likelihood of RVV (Callanan et al., 2012). There are a range of practical measures which can enhance the safety and protection of victims, including improved home security, panic alarms, heightened police awareness and refuges for victims of domestic violence (Callanan et al., 2012). These protect the victim by preventing future opportunities for violence, as well as increasing the actual or perceived risks of apprehension for offenders (Grove et al., 2012).

Several measures of this kind have been implemented, particularly in relation to domestic violence (David et al., 2006; Lloyd et al., 1994; Hanmer et al., 1999; Sampson and Phillips, 1995). For example, Lloyd et al. (1994) developed a preventive strategy to reduce domestic violence on Merseyside, involving: the provision of quick response pendant alarms to domestic violence victims at risk; additional support and information for victims of domestic violence; more complete transfer of injunction details from courts to police; and heightened police awareness. This strategy had preventive effects on domestic violence in the area (Lloyd et al., 1994). 

Moreover, in Scotland, Shaw and Pease (2000) have also provided examples of how RVV can be successfully addressed. They highlight a case where a victim of attempted murder and rape had been subjected to further threats from the offender. The individual was given home security advice and issued with a mobile phone and panic alarm linked to the police control room. No further incidents or threats were subsequently reported. Similarly, a female victim of domestic violence had been assaulted by her husband and was issued a mobile phone linked to the police control room. The man subsequently subjected her to threats. She used the mobile phone and officers attended and arrested the man. After this, no further incidents were reported (Shaw and Pease, 2000). 

Findings regarding the time course of repeat victimisation suggest that prevention measures such as those discussed here should be implemented quickly. As above mentioned, there is evidence to suggest that RVV often occurs soon after the initial victimisation. There then exists a heightened risk period for re-victimisation, which declines with time (Weisel, 2005). While more research is required to determine if this time course is applicable to all types of violent crime, the research indicates that temporary prevention measures which provide protection during the high risk period after victimisation could be an effective and efficient means of reducing violence (Farrell and Pease, 1993). 

Beyond measures for enhancing victim safety and protection, it has also been argued that meeting the practical needs of victims in the aftermath of an offence may be an important way to reduce the risk of repeated episodes of victimisation (Callanan et al., 2012). The main reason for this is that it ensures individuals have the resources and support available to them to prevent further violence occuring. This can include:

• Advocacy to participate in the criminal justice system to ensure offenders are prosecuted
• Help to access or communicate with a range of services and organisations (e.g. victim support networks, local councils regarding housing issues, employers relating to time off work)
• Help with housing, to ensure it meets safety needs 
• Help with childcare, to enable victims with caring responsibilities to engage with service providers.

In addition, it has been argued that effective partnership working between agencies is important to reduce the risks of repeated episodes of victimisations (Callanan et al., 2012). For example, joint working between the police and support services to improve the safety of victims of RVV.

However, some practical difficulties with implementing a violence prevention/reduction strategy based on RVV have been highlighted. In particular, identifying appropriate safety and protection measures is difficult for violent crime, and where known prevention measures exist, victims are often difficult to contact (Farrell, 2005). Further, when contacted, some victims do not want, or do not have the resources, to adopt such measures. Moreover, while Shaw and Pease (2000) provide useful illustrations of problem-solving policing using the concept of RVV, there are few examples of this concept being embedded into practice, in Scotland or elsewhere, with most focusing on reducing repeat burglaries rather than RVV (Forrester et al., 1990). This has led to the argument that there is a need for a broader agenda in relation to the many types of crime that might be fruitfully addressed by the repeat victimisation approach (Farrell, 2005), with violent crime yet to have been systematically addressed using such a strategy. 

Objections have also been raised around focusing crime prevention efforts on RVV. Namely, such a strategy is designed to act after the crime has taken place, meaning that it fails to address the root causes of violent crime (Farrell and Pease, 1993). In addition, concerns have been raised that the method in some way blames the victim, by implying that if they had behaved otherwise the crime would not have taken place (Farrell and Pease, 1993).