Youth And Gun Crime Criminology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Young people and 'knife' and 'gun crime' are at the forefront of public attention, and an almost constant focus of the media and policy makers in this country. Yet what exactly do people mean when referring to 'knife' or 'gun crime'?

'Gun crime' and 'knife crime' are not discrete entities, but encompass a variety of unlawful behaviors. Such weapons can in fact be used in actions which fall under a range of offences in the UK legal context, e.g.:


''Violence against the person


''Sexual offences and domestic violence

''Criminal damage

'Gun' and 'knife crime' are therefore an amalgam of different sanctioned behaviors. Interpreting such apparently simple, and misleadingly unifying, concepts is further complicated by the way offences are officially recorded. Police recorded data are generally limited in their reliability because of a number of factors, including: much criminal activity does not get reported to the police, is not detected, or does not get recorded by the police. Moreover, changes in recording practices (which may mean, for example, that more or new offences are included in the records) make it difficult to compare data historically. Trends in police data are also susceptible to changes in the way the police go about their activities, in Home Office requirements and in the way suspected crimes are recorded. Victimization surveys (like the British Crime Survey (BCS) in England and Wales and the Scottish Crime and Victimization Survey (SCVS)) were developed in part because of the recognition that police recorded crime data can only provide a partial picture of crime levels and trends. Most commentators agree that such surveys provide a more reliable estimate of the offences they cover than that given by police recorded data. However, the range of offences covered by the BCS is narrower than police data. It also tends to underestimate some of the offences it covers (e.g. domestic violence), and, importantly, persons under 16 have not so far formed part of the survey.

Official figures are also problematic specifically with regards to knives and guns related offences. In England and Wales, the recording of 'knife enabled' offences did not start until April 2007 (with the exception of London, where such records started to be collated in 2003), which makes it impossible to make any comparison with preceding periods (Eades et al 2007). Moreover, the Home Office category of 'firearm enabled' and 'knife enabled' crime excludes offences of possession (Squires 2008a). The fact that offences of illegal knife and gun possession are not routinely collected for the Home Office makes it difficult to establish the impact of strategies aimed at deterring young people from carrying weapons, including searches and detectors, public safety education campaigns, knife amnesties (Squires 2008b). Official records have until recently not distinguished the weapon by which an offence is committed, and the official definition of 'wounding' comprises acts where the skin is punctured or cut, which is likely to mostly involved stabbings by knives, but includes other means by which such injury may be caused. Official wounding figures show that 'knife crime' is nothing new. Knives/sharp objects/ stabbings have always been the primary means by which young men have killed one another in peace time in England, Wales and Scotland. The overall trend of killings from stabbings rose fairly consistently throughout the second half of the 20th century. It began to increase more rapidly from the mid-1980s and, then again, after 1998 (Home Office 2007).

Although guns are far less common, and far less easily accessible, than knives, the overall trend of gun-related offences follows a similar pattern to knife-related offences, increasing since the mid-1960s and then again in the late 1980s.

Despite current public concerns about 'kids killing kids', official figures show most young people under 16 are killed by their parents. In 2007/08, 62% of all homicide victims under 16 were killed by their parents, an increase from 51% in 2006/07.

The group most at risk overall for homicide are very young children, i.e. those under one year of age. In 2007/08 this group had a homicide rate of 36 per million populations (Povey et al 2009).

The second most at risk age group in 2007/08 (and consistent with previous years) were males aged between 16 and 29. Within this group, those most at risk are young males aged between 16 and 20 inclusive (Povey et al 2009).

Chapter 2

2.1 Who is possibly involved in violent crime?

Evidence derived from quantitative research represents the most extensive base of social scientific knowledge about criminal 'careers'. Quantitative research into criminal careers is often referred to as 'predictive'. Findings suggest that there are problem areas, known as 'risk factors', which can predict the likelihood of future violent criminal behavior among young people.

Important risk factors for young people committing violent crimes include individual characteristics (e.g. whether they are male or female), their relationship with parents and family, their behavior and their performance in school, whether they have friends who carry out violent offences, whether they carry out other types of crimes, drug and alcohol use and also whether they have been exposed to violence, either as a victim or as a bystander, witnessing violence against other people.

Risk factors are cumulative: the more a young person has, the greater the likelihood of violence. A body of research identifies factors which can have a 'protective' role against delinquency (see Glossary), i.e. which help young people avoid offending and entering deviant lifestyles. Although this research is less developed than risk factor research, protective factors appear to include good academic achievement, non-delinquent peers, good relations with parents and caring and supportive adults present in the young people's lives. Social crime prevention programs have sought to tackle the 'causes of crime' by removing or neutralizing risk factors impinging upon young people's lives and/or bolstering their protective factors - for example, by attempting to reduce poverty, enhancing educational and training opportunities or offering family support.

Predictive methods are not infallible: they may over- or under-predict what young people are likely to become deviant or violent (what researchers call 'false positives' or 'false negatives') and should be therefore used with caution. An over-deterministic reading of predictive research may lead to labeling children and young people as potentially criminal, and this may actually impact negatively on their outcomes. Although there is considerable research on youth violence and the factors which are likely to engender or restrain it, much less is available on specific forms such as gun and knife crime. However, the social scientific knowledge produced by predictive research provides an important backdrop to understanding the phenomena of weapon carrying and use by young people.

2.2 Youth offence risk factor research

2.2.1 Protective factors

Many researchers working in the field of prediction are interested in what is often referred to as a 'risk-factor' approach to preventing crime. However, most children and young people at risk of violence or crime do not actually manifest such behavior, even though they may well have been exposed to a variety of risk factors. This has led to attempts to identify those factors in young people's lives which reduce or negate the impact of risk factors. These factors are referred to as 'protective factors' and their contribution to young people staying out of trouble is known as 'resilience'.

Protective factors include positive relationship with parents, high academic achievement, positive friendships with non-delinquent peers, extracurricular school activities, belonging to smaller (in terms of numbers of children) families, good problem solving skills and empathetic skills. Some protective factors can be seen as being the opposite of risk factors - for example, young people not being exposed to violence compared with young people who are exposed.

While many risk factors have been replicated in numerous studies, more research is needed around the relative importance of protective factors and how they can be introduced into the lives of young people at risk of violent behavior (Farrington 2002, Loeber and Farrington 1998).

Meta-analyses of relevant studies into predictors of violent crime delineate the most significant factors in serious and violent offending among young people. For the purpose of this chapter we looked at two such meta-analyses, which combined examined the results of 92 high quality studies into the genesis of youth violent behavior (Lipsey and Derzon 1998, Derzon 2001).

Research evidence shows that the major predictors of young people committing violent crimes relate to:

'Individual characteristics (e.g. gender, age, age at first offence, impulsivity, academic performance or IQ: that is to say, being male and/or low performing at school, and/or emotionally unstable, and/or with a low IQ are factors which increase the likelihood of violence during adolescence and into adulthood)

Their relationship with parents and family

Their behavior and their performance in school

Whether they have friends who carry out violent offences

Whether they carry out other types of crimes

Drug and alcohol use

Whether they have been exposed to violence, either as a victim or as a bystander

Witnessing violence against other people.

Risk factors are cumulative: the more a young person has, the greater the likelihood of violence.

Table 1 below thematically summarizes the key risk factors for serious and/or violent offending among young people, as revealed by the literature.

Table 1: Key risk factors for serious and/or violent offending among young people

(Source: Howells 1997)

2.2.2 Risk factors for various age groups

Risk factors change over time, with different age groups being affected by the same factors to varying degrees: a risk factor that predicts a 7 year old will be violent later in life may not be a good predictor when that individual is 12. As an example, the most significant predictive factors of serious and violent offending among young people aged 12-14 (as established by Lipsey and Derzon's meta-analysis) are shown in Table 2 overleaf. These factors, when present among young people aged 12-14, are likely to predict (further) violent offending when they are aged 15-25. The factors are ranked according to their statistical strength, with group 1 containing the most reliable predictors and group 5 the least reliable. (Within each group, the factors are also ranked according to statistical strength, e.g. 'weak social ties' being a relatively stronger predictor than 'antisocial peers'.)

Table 2: Predictors for serious and violent offending for young people aged 12-14, in descending order of statistical reliability (Source: Lipsey and Derzon 1998)

Lipsey and Derzon (1998) found that, for children aged 6-11, the strength of the relationships between risk factors and violent offending were different. For example, having anti-social friends for 6-11 year olds was the weakest predictive factor of all those identified. For 12-14 year olds however, having anti-social friends was one of the strongest predictors of serious and violent offending. This means that we must be careful to relate research evidence to young people according to their age: what helps prevent violence in younger children may not do so in older children or teenagers. Between ages 14 to 16, gang membership is a significant predictor of involvement in violence (Hawkins et al 1998). Having friends who commit crimes and acts of violence is also a predictor of the likelihood of gang involvement (Thornberry 1998).

2.2.3 Other Risk Factors

As well as the risk factors mentioned above, other experiences in young people's lives increase their likelihood of themselves becoming violent. These experiences often occur for young people who live in the most deprived areas with high levels of social exclusion.

Living in a neighborhood where there is violence makes it likely that at least some young people will actually see a violent act or will be victims of violence.

Both these types of experience can have long lasting effects. Experiencing violence - either as a victim or witness - in fact increases the likelihood that young people will experience mental health problems such as depression, abuse drugs and alcohol or perform poorly at school. It also increases the likelihood of them being violent and carrying weapons (Patchin et al 2006, Sieger et al 2004).

Research has also highlighted that young people who experience multiple victimization as traumatic often end up seeing the world as fundamentally unsafe (Cicchetti and Toth 1997) and perceive existence as lacking meaning. They may come to think that their own life could end at any time and therefore disregard how current actions will influence their future (Reebye et al 2000).

Research in the United States into the impact of adverse childhood experiences demonstrates the cumulative impact in adult life of experiences which include emotional, physical and sexual abuse, neglect, witnessing parental violence and/or drug abuse, family mental illness, parental separation or divorce or a member of a household being sent to prison (ACE 2009).

Different forms of parenting may act as risk or protective factors. Research in the US has found that low parental control and monitoring are associated with increased likelihood of gang involvement (Thornberry 1998). Better quality parenting, however, has been identified as a protective factor that reduces the risk of young people exposed to violence becoming violent themselves (Gorman-Smith et al 2004). By contrast, some American research has shown how parents may 'condone' their children resolving disputes violently because this accords with certain values associated with toughness and masculinity (Solomon et al 2008).

2.2.4 Risk factors as interacting

Risk factors do not operate in isolation, but interact with one another in a complex set of social, economic and cultural circumstances. They therefore need to be understood in their relationships with other factors. For example, gangs have been shown to be associated with an increased risk of criminal behavior (e.g. Thornberry 1998). However, membership of a gang in itself does not necessarily cause a young person to commit crime. Apart from the fact that 'gang' is an all encompassing term which covers all sorts of groupings and behavior (see below), it is the involvement in certain activities, as well as the sharing of certain beliefs and values held by gang members, which may dispose them toward offending.

The more we can develop an understanding of the way risk factors interact, the better we may be able to understand how we can intervene to prevent violent offending using guns and knives.

2.3 Knives and gun crimes

Whilst it is relatively straightforward for researchers to identify the risk factors which are likely to play a role in future offending by young people, difficulties arise in predicting specific types of offending. Predicting the likelihood of young people committing some sort of offence is much easier than being able to say what that offence will be.

In any attempt at predicting behavior, a key issue is the extent of such behavior in the population. If certain behavior was carried out by 90% of the population, it would be easy to predict, since nearly everybody would be behaving in that way.

Violent offending is relatively rare and therefore is likely to be harder to predict.

This difficulty in prediction is likely to be worsened for girls and young women, where the occurrence of violence tends to be lower than that for boys and young men (Monahan et al 2001).

Knife and gun crimes form specific categories of violent crime and are therefore rarer than occurrences of violence generally. This means that such complex actions are still less prevalent in society and therefore more difficult to predict.

Some predictive research has however been carried out in the US, where it was found that strong predictors of weapon carrying include having been shot at, threatened or injured, being involved in drug dealing and having the belief that their parents did not care about them (Kingery et al 1999).

Research also shows that those who carry weapons are likely to become victims themselves. In fact, many of the risk factors for being in a gang and carrying a gun were also risk factors for being shot (see e.g. Loeber, Kalb and Huizinga 2001). In the UK, a study of shootings in Manchester (Bullock and Tilley 2002) found that both perpetrators and victims of gun crime tended to be black males, with criminal records and often a previous history of involvement in gun crime.

2.3.1 Gangs

Discussion of serious, weapon-involved violence raises the question of the role of gangs as facilitators of criminal involvement and escalators of violence. A range of American research confirms that gang membership is itself a significant risk factor for increasing violence (Hagedorn 1988, Thornberry et al 1993, Klein 1995, Decker and Van Winkle 1996, Zimring 1998, Huff 2002, Braga 2003).

When considering gang related issues, however, it is important to bear in mind that both the general understanding of the concept of 'gangs', as well as their criminal significance in the British context, remains a contested issue.

Understandings around what gangs are and what they do vary greatly, from simple groups of young people congregating in groups, to conducting minor nuisance behavior, to involvement with organised criminal activities (see e.g. Hallsworth and Young 2008, Alexander 2008).

It is also important to remember that 'gang activities' are not the only reason behind gun and knife crime. Violence and injuries can be inflicted for a number of other reasons, including the perpetrator's mental health state or, in the context of public disorder incidents, especially in connection with the heavy consumption of alcohol in city centres at weekends.

2.4 Instructive note

Agencies working in criminal justice use predictive tools which assess how likely somebody is to offend. Usually these tools are scales based on factors identified as having a statistical relationship with offending, such as those discussed above.

Such predictive tools score a person according to how many and which risk factors are present in their lives.

However, there are problems with relying on statistical risk prediction alone, as they can lead to people being dealt with unfairly or can lead to potentially dangerous people being ignored or over looked. In reality, someone identified as likely to offend May not in fact actually do so; on the other hand, some people identified as low risk may go on to carry out a violent crime.

It is important to bear in mind that 'risk factors' are just that - factors which put some young people at 'risk' of carrying out violent behavior. This does not mean that they will they inevitably do so. Derzon's findings exemplify the need for caution when trying to predict violence. Derzon (2001) carried out a meta-analysis combining and analyzing the results of 60 prospective studies of juvenile violence.

He examined whether it was possible to establish a link between early anti-social behavior and later violence. Whilst predicting some violent crime, his findings did not support the notion of an underlying trait in individuals that would make them bound to commit violence in later life. Many of the young violent offenders in the studies had not previously exhibited any of the general risk factors for crime.

Predictive methods had failed to identify 66% of those who later became violent.

This calls into question the notion that there is an immutable progression toward violent behavior for many young people who engage in antisocial behavior.

Caution needs to be used, therefore, in trying to apply these types of predictive tools to groups of young people. Treating all those who have a high number of risk factors as potential criminals may actually draw more of them into the criminal justice system - what is referred to as 'net widening'. There is no evidence that early intervention by the criminal justice system has a positive impact upon young people and, indeed, the Edinburgh Study of Youth Transitions and Crime found that early contact with the criminal justice system was a predictor of later, more serious contact (McAra and McVie 2007).

Chapter 3

3.1 The impact of where children and young people live on their involvement in violent crime

Research shows that high rates of crime and violence mostly affect disadvantaged areas. People living in deprived areas do not just experience higher levels of crime: they suffer from other problems, including poverty, low social capital and limited social mobility. The impact of multiple problems is cumulative: the more social problems are encountered by families and individuals, the more likely they are to remain in poverty.

Violence in such areas is a complex product of the way opportunities and lives are shaped for people living there, and the way that people respond to their situation and to their environment.

Violence causes fear and stress: it can result in various forms of mental illness and even lead to suicide. Being exposed to violence - as a victim or by seeing someone else being victimized - makes people more predisposed to commit violence themselves and to carrying weapons. Although research is still limited in this area, where young people live seems to affect whether they get involved in crime, and whether (or how difficult) it is to stop. Where neighborhoods are threatening, weapon carrying may make young people feel safer. However, the presence of weapons may escalate conflicts and increase the likelihood of (serious) injuries.

Research has endeavored to identify why and how certain (mainly urban) areas are associated with higher levels of crime than others. Studies typically compare measures of disadvantage (e.g. census data and unemployment statistics) in a certain area with local crime rates in order to establish a statistical relationship between the two. Some research also tries to link these measures to individual factors, in order to establish the ways in which social and economic disadvantage can impact on young people and make them more likely to offend.

Such research confirms that areas with higher levels of social disadvantage (for example, run down or poor quality housing, high levels of unemployment, drug and alcohol abuse, low rates of educational achievement and high levels of excluded pupils and truancy) also have higher rates of crime and violence than better off areas in the same city or locality. However, in these studies the link between social disadvantage and violent crime is well established without being necessarily adequately explained.

Perhaps one reason for this is the nature of quantitative research. Variables in such research routinely draw on sources such as income levels or welfare take-up to use as proxy measures of socio-economic deprivation, or they utilize complex formulae to estimate the effects of living in a deprived area (Geronimus 2006).

However, these are aggregate and abstract data and provide little insight into the lives of those individuals living in deprived neighborhoods.

We have also not encountered research that specifically examines any causal relation between the nature of certain areas and the carrying and use of guns and knives amongst young people living there.

3.2 Associations among 'Neighborhoods' and violent behavior

We use the term 'neighborhood' to refer to an area that people generally tend to recognize as an entity this may well be an estate or a block or blocks of flats.

(Some researchers talk about the 'environment' where people live, others may refer to a 'community'.) For young people, their locality - their 'endz', as some refer them to as, may be mapped out around a shared conception of what constitutes 'their' territory (Chandiramani 2008). But, however defined, the area which surrounds the places in which people live provides 'a transactional setting that influences individual behavior and development both directly and indirectly' (Elliot et al 1996).

There have been concerns about high levels of crime in certain areas of towns and cities for over a hundred years (Mayhew 1865, Thrasher 1927, Shaw and McKay 1942, Morris 1957, Cloward and Ohlin 1960, Bottoms et al 1987, Davis 1990, Campbell 1993, Taylor et al 1996, Walklate and Evans 1999, Hayward 2004, St. Jean 2007). The high number of negative outcomes associated with areas with concentrated disadvantage alerts us to the impact that such deprivation has on the well being of those residents (Sampson 2004).

People living in deprived areas do not just experience higher levels of crime, but also suffer from other problems, including poverty, low social capital and limited social mobility. The impact of multiple problems is complex and cumulative: the more social problems are encountered by families and individuals, the more likely they are to remain in poverty (Department for Work and Pensions 2004). Many areas of concentrated deprivation comprise housing stock that was owned and administered by local councils - 'council estates'. Although originally intended to improve the lives of working class people, it has been argued that an unintended effect was 'the virtual ghettoization of some estates' (Ravetz 2001).

3.2.1 Neighborhoods, economic conditions and social identities

The ways localities respond to economic downturn and recession differ from place to place (Taylor et al 1996). Research has mainly focused on urban areas characterized by industrial decline and loss of industry - e.g. Hagedorn (1988) describes the consequences of the loss in the late twentieth century of semi- and unskilled labor markets in what are now referred to as the 'Rustbelt' cities in the US. Conditions in those areas which no longer have access to employment have frequently deteriorated, with rising levels of social problems, increased crime and disorder and, more recently, the emergence of gangs and rising levels of violence.

Poverty and inequality have become entrenched in areas where, in the absence of employment or meaningful material assistance, involvement in various forms of crime may be one of the few ways to actually make a living (Elliott et al 1996). The lack of economic opportunities for young people has stimulated the growth in both the US and the UK of illegal economies around drugs, stolen goods and protection (Klein 1995, Braga 2003).

A UK study of convicted gun offenders concluded that Illegal drug markets appear to 'significantly underpin the criminal economy [representing] the single most important theme in relation to the illegal use of firearms.' Violence levels connected to drug markets operations appeared to increase significantly 'towards the street (retail) end of the market' (Hales et al 2006).

Moreover, the masculine identities traditionally associated with jobs in heavy manufacturing or iron, steel or coal production become for some young people only primarily achievable through crime (Campbell 1993, Newburn and Stanko 1994, Hall 1997, Hall et al 2008). In communities suffering from 'capital disinvestment' and lack of social capital, young people are more likely to drift into 'cultural adaptations' that bring short term status and material benefits, but which longer term consequences include diminished life chances (Hagan 1994; see also James 1995 and Kramer 2000).

The 2006 UN Report on Violence against Children noted:

…physical violence between peers tends to be more common in urban areas characterized by lack of employment, education and social amenities and low standards of housing, where youthful and rapidly growing populations express frustration, anger and pent-up tension in fights and anti-social behavior. Much of the violence involves personal disputes between friends and acquaintances, and is strongly associated with the use of drugs and alcohol. Where guns and other weapons are available, fights often lead to severe injuries and death. (United Nations 2006: 20)

3.2.2 Territory

Young people frequently complain of 'boredom', frustration and a lack of age appropriate and affordable social and leisure facilities. They often describe a distrust of adults and authority figures. While neighborhoods can generate loyalty and a sense of belonging (as in postcodes gang identifications), a recent study of young people in British cities has shown how territoriality can both compound disadvantage and social exclusion as well as promote conflict and violence. Thus, territoriality has a potential 'to block access to opportunities, to foment violence and to act as an escalator to more serious forms of crime, including involvement in criminal gangs' (Kintrea et al 2008).

Fear of leaving one's estate can also fundamentally restrict a young person's social and geographical horizons. A telling assessment of a young person's view of neighborhood came from a young black man from the Prince's Trust being interviewed by members of the House of Commons Home Affairs Select Committee (at a seminar organized by the Committee in November 2008): 'These estates are like cages', he said.

For young people, being born into such areas is likely to offer them a narrower set of life-opportunities than for young people in more affluent areas. With limited options, few resources and little support, residents of such areas - especially young people, may feel that they have little hope of living a more comfortable life, like that perceived to be enjoyed by people outside their area or portrayed in the media.

There may be few recreational facilities locally and young people who go outside their area may be treated with suspicion, distrust or hostility by other young people they encounter there. Especially when they 'hang out' in groups, they may find themselves the object of police attention (Anderson et al 1994, Loader 1996, Measor and Squires 2000). To some extent this has now become part of official policy, with the introduction of dispersal orders, primarily targeted at young people gathering in public locations (Crawford and Lister 2007).

3.2.3 The effects of violence on a neighborhood

Violence, along with all the other negative aspects of living in deprived areas, tends to erode what researchers call 'collective efficacy' (Sampson and Raudenbush 1997), i.e. the capacity of a neighborhood to manage itself. In a neighborhood with 'collective capacity' people are more inclined to act in a collectively minded way: e.g. to stand up for others, to report crime to the police and to let wrongdoers know that someone is watching. In areas of concentrated deprivation, such informal mechanisms of control may be weak or absent. Research found that, controlling for a wide range of individual and neighborhood characteristics, collective efficacy directly predicted lower rates of violence (Sampson and Raudenbush, op cit).

3.3 The effects of neighborhoods on young people's offending

There is limited research evidence on the relationship between living in deprived areas and individuals' violence. The major reason for this would seem to be a polarization of research focus: 'researchers interested in neighborhood influences have generally not adequately measured individual and family influences, just as researchers interested in individual and family influences have generally not adequately measured neighborhood influences' (Farrington 1993). Another reason lies in the difficulty in demonstrating that a deprived neighborhood 'causes' those living in it to commit crime; it is extremely difficult to isolate the various elements that combined together constitute environmental facilitators to offending. In any area, young people will have different risk or protective factors (outlined in Chapter 2), different life experiences and different perceptions of their possible choices, actions and futures. These differences are likely to affect and mediate - positively or negatively - the experience of severe disadvantage.

Wikström and Loeber (2000) found evidence of the differential impact of neighborhood effects in an analysis of data from the Pittsburgh Youth Survey.

The research found that, for those young people who had a high number of risk factors, there was no discernible neighborhood effect upon serious offending. However, young people who should have been better placed to stay out of trouble (because of protective factors, or of risk factors being balanced by protective factors, in their lives), nonetheless tended to become involved in serious offending, although starting at a later age: this was found to be a result of neighborhood effects. This suggests that even for young people with many positive aspects in their lives, prolonged exposure to the various risk factors that are present in areas of concentrated deprivation may ultimately have a negative impact on behavior. Such findings are echoed in a study of delinquency in Edinburgh, where the main factor associated with whether a young person continued to offend or not was found to be the neighborhood where the young person lived. 'Continuing to offend was more common in deprived neighborhoods, whereas desistance was more common in advantaged ones. Also, desistance was less likely in neighborhoods perceived to be disorderly, and where residents were dissatisfied with the neighborhood' (Smith 2006:4).

Other indications that the urban environment can influence individuals' behavior, in ways not reducible to individual characteristics, come from a large-scale longitudinal research project carried out in the United States (Liberman 2007). The study involved repeated interviews with more than 6,000 young people and their caregivers, combined with a survey of almost 9,000 residents and systematic observation of social and physical disorder in 80 neighborhoods. The main findings from this research were that young people living in disadvantaged neighborhoods were more likely to carry firearms illegally and those who had been exposed to violence were more likely to commit violence themselves.

Importantly, the study also found that differences in offending patterns between different racial/ethnic groups were largely explicable in terms of neighborhood and family: young people of different race/ethnicity who lived in similar neighborhoods had similar patterns of offending (Liberman 2007).

3.3.1 'Moving to Opportunity'?

There is other evidence to suggest that crime may vary by area, due to individual responses to the areas in which they live. In The United States, the 'Moving to Opportunity' experiment enabled families in the experimental group to relocate to less disadvantaged areas. Researchers found that violence amongst young men and women aged between15-25 who had been relocated as part of the experiment between 4 and 7 years previously initially fell. However, longer term results showed that the reduction in violence amongst young men was offset by an increase in property crime. For young women however, the results were more positive, showing a decrease in crime and also other outcomes linked to education, substance abuse and mental and physical health (Kling et al 2004).

These findings are an indication that the relationship between neighborhood context and crime is considerably more complicated than it seems. Moving from one area to another does not guarantee that the move will necessarily result in better opportunities than before, nor that the inclination to commit crime among those moving will diminish - although in this case, the type of crime committed by young males, while increasing in frequency, changed from violence to property crime.

The different outcomes for young women in 'Moving to Opportunity' are not unlike those achieved following the adoption of educational integrationist 'bussing' policies in some US cities. Whilst some of the African American girls prospered, the experiment seemed to reinforce the gang affiliations of many of (early teenage) African-American males, who clubbed together for self-protection against white student majorities or began to avoid school attendance altogether. Moving the young people to new areas and new schools in different communities fractured the young people's links with their own families and communities (thereby weakening important protection factors) and left them in schools where they were perceived, at best, as unwelcome outsiders. At the same time these policies also contrived to increase substantially the role of their own peer groups in their lives, a widely acknowledged youth delinquency risk factor (Hagedorn 1988).