Someone recently bought our

students are currently browsing our notes.


Pathways In And Out Of Crime Risk Factors And Desistance Notes

Law Notes > Criminology Notes

Updates Available  

A more recent version of these Pathways In And Out Of Crime Risk Factors And Desistance notes – written by Oxford students – is available here.

The following is a more accessble plain text extract of the PDF sample above, taken from our Criminology Notes. Due to the challenges of extracting text from PDFs, it will have odd formatting:

CRITICAL DEBATES IN DEVELOPMENTAL AND LIFE-COURSE CRIMINOLOGY (MCARA AND MCVIE, 2012) What is developmental criminology?-Field of study that "explores age-based changes in individual offending behaviour" o Focus on crime and anti-social behaviour across the individual's lifespan
? i.e. tracks the same individuals o Uses longitudinal studies to examine changes over time Originated in the mid-nineteenth century o Based on the assumption that knowledge is "progressive, perfectible and universalisable" o Seeks to measure/quantify human behaviour and identify regularities in conduct Bio-psycho-social approach to studying individuals o Identified several predictors of criminality based on longitudinal research o E.g. Boston research by Glueck (1930s) identified early onset of delinquency, family relationships and social deprivation as important factors

AGE-CRIME CURVE-Usually shows a peak in offending in the late teens/early twenties, followed by a steep decline into adulthood First recorded illustration was by Quetelet (1831) o Has been replicated in subsequent research projects Peak can be affected by a number of factors (but it will usually be present) o Convictions data might show a later peak than self-reported offences o Peak has moved to a younger age over time o Peaks are generally shallower for females Possible reasons for the curve o Quetelet: peaks when strength and passion reached maximum levels but ability to reason had not yet developed enough to control those o Gottfredson and Hirschi (1990): propensity for crime manifested during teenage years and could later be tempered by self-control Farrington (1986): the age-crime curve for individuals didn't resemble the aggregate curve o The peak was largely due to changes in participation (more offenders)

THEORIES EXPLAINING THE AGE-CRIME CURVE DUAL TAXONOMY THEORYProposed by Moffitt (1993, 2003) 2 forms of anti-social behaviour o Life-course persistent (LCP): could begin in early childhood and the form of behaviour changed according to age
? Causal factors underpinning their anti-social behaviour are present throughout their lives, so their anti-social behaviour is constant
? Due to neuropsychological deficit and family-adversity risk factors o Adolescence limited (AL): temporary anti-social behaviour
? Starts in early adolescence and de-escalates after a peak
? Due to external risk factors and rebellion


Age-crime curve peaks due to the surge of offending amongst AL offenders, masking the consistent offending of LCP offenders

AGE- GRADED THEORY OF INFORMAL SOCIAL CONTROLProposed by Sampson and Laub (1993) Offending is closely related to social bonds and type/strength of relationship o Emphasises human agency rather than structural background Offending during teenage years peaks as a result of strong attachment to delinquent peers, which changes in adulthood due to positive relationships resulting from various factors (including marriage and employment)

INTERACTIONAL THEORYProposed by Thornberry (1987, 2005) Delinquency is a result of weak bonds to society These bonds can be influenced by structural variables o E.g. social class position, residential area o Delinquency can result from weak bonds but can also cause weakening


Proposed by Farrington (2003, 2008) Anti-Social Potential (AP): potential to engage in crime o Depends on impulsiveness, strain, modelling, socialisation process and life events in the long term (varies between individuals) o Short-term AP depends on motivational and situational factors (varies within the same individual over time) Cognition: decision-making process AP peaks in teenage years due to a change in risk factors o E.g. reduced influence of parents, increasing influence of peers o Also temporarily increased due to short-term factors and availability of criminal opportunities

NEGOTIATED ORDER THEORYProposed by McAra (2005, 2012) Individual identities are built up based on formal and informal regulatory orders o Formal regulatory orders: e.g. schools, police, courts o Informal regulatory orders: e.g. friends and family Exclusion from formal orders might push the individual to seek inclusion in negative informal groups

METHODOLOGICAL ISSUESSelf-reporting vs official statistics o Official Statistics: limited data
? Doesn't cover all offences since most crime doesn't come to police attention
? Distorted picture of offending since more serious offences are more likely to be reported o Self-reporting (more widely used in recent years)
? Presents a wider set of data

Allows for research into "the characteristics, background, and behaviours of the offender in order to test aetiological theories of crime"
? Possible issue: validity
? Need to ensure that respondents report their offending accurately
? Surveys must be designed without bias and administered well Longitudinal vs cross-sectional o Greenberg (1991): longitudinal studies are required to examine the causal effects of age on crime, especially since effects of events can be timelagged o Rutter (1995): could confuse age, cohort and period effects
? Age effects: variations in risk depending on the age
? Cohort effects: variations in risk that apply to all individuals in a certain group or that shared the same experience
? Period effects: variations in risk that apply to a whole generation o Farrington (1986): better to combine both longitudinal and cross-sectional studies in order to track changes across ages Semi-parametric group-based modelling o Identify different groups of individuals based on statistical algorithms which can be distinguished based on their trajectories o Nagin and Land (1993) identified 4 different classes from the Cambridge Study
? Long-term, high-rate trajectory (LCP offenders)
? Short-term, late-onset trajectory (AL offenders)
? Low-rate chronic offenders (lower IQ offenders)
? Individuals who were never convicted o ISSUES
? Many different classes might be identified from different studies
? The groups identified may not represent actual groups in society Random effects growth-curve modelling o Identifies patterns in individual trajectories o Used to show overall gradual change, but not small-scale changes Other modelling techniques o Growth mixture modelling o Non-parametric growth mixture modelling Kreuter and Muthen (2006): Semi-parametric group-based modelling and growth mixture modelling best fit results Osgood (2005): Growth-curve modelling is preferable o Trajectories may vary o Danger of being treated as real groups o Labelling them high-rate chronic offenders may result in more punitive action by the authorities--


Sutton (2004): Support from the Start o Primary intervention: universal support targeting whole communities o Secondary intervention: support and services for families and schools where children have been identified as being at risk of offending o Tertiary intervention: aimed at preventing re-offending Aos (2008): secondary intervention has good cost-benefits o BUT universal targeting can also be cost-effective in the long term

Especially since many self-reported offenders never have criminal convictions (in Cambridge Study, half of all self-reported offenders were still unconvicted at age 50), so they wouldn't be detected
? Edinburgh Study: 76% of violent young offenders had no convictions and were unknown to agencies (i.e. only universal targeting would reach them) Helping families o Elmira Study (USA)
? Home visits by nurses (pre- and post-natal) to support development and care for the children
? Much lower levels of child abuse/neglect and fewer arrests later in the child's life for the group with intervention o Farrington (2003): proper parenting management can minimise anti-social behaviour and delinquency by 20%
Programmes for schools o Aimed at enhancing educational attainment and fostering attachment o Seattle Social Development Project
? Combined parenting techniques, teacher training and skills training
? Tracked children from ages 6 to 12 and 18
? Randomly selected experimental group was less likely to be involved in delinquency/drug use/violence/alcohol abuse Communities that Care (CTC) o Involves the whole community in preventing crime o Introduced in the UK in 1998 o Combines various community groups to identify local prevention needs and implement intervention activities NOTE: most trial programmes have been run in the USA o May not be directly transferable to a different community ISSUE: developmental criminology targets individual needs o Doesn't seek to address structural needs (e.g. for children from disadvantaged backgrounds) o Targeted intervention might even exacerbate the problem by labelling and stigmatising the potential offender early on
? Involvement with the criminal justice system usually inhibits desistance--FARRINGTON, CHILDHOOD RISK FACTORS AND RISK-FOCUSED PREVENTION (2007) MAIN PREVENTION STRATEGIES Developmental preventionPrevent the development of criminal potential in individuals Targets risk and protective factors identified in studies

Community preventionInterventions that change social conditions and institutions Aimed at reducing offending in a residential community

Situational preventionAimed at reducing opportunities for crime and increasing the risk/difficulty of offending

Criminal Justice preventionTraditional deterrent, incapacitative and rehabilitative strategies Implemented by law-enforcement and criminal justice agencies

Study of developmental factors in offendingDevelopment of offending and antisocial behaviour Risk factors (determined at different ages) Effects of life events on the course of development

Studies of within-individual changes in criminal activity over time"offending is often part of a larger syndrome of antisocial behaviour that arises in childhood and tends to persist into adulthood" Behavioural manifestations can change over time even if the underlying measure of antisocial behaviour is stable o Might be affected by opportunities (e.g. only students can play truant, only adults can abuse spouses) and individual capabilities

RISK FACTORSFactors that "increase the risk of occurrence of the onset, frequency, persistence, or duration of offending" Similar results are obtained on self-reported offenders as well as officially recorded data Many risk factors are highly replicable o Farrington (1999): comparison of longitudinal surveys in London and Pittsburgh
? Similar predictors of impulsivity, low educational attainment, attention problems, parental problems, low family income, etc.


Vazsonyi (2001): similar association between self-control and antisocial behaviour in Hungary, Netherlands, Switzerland and USA)

Risk factors may not necessarily be causal (might be indicative rather than "leading" people to criminal careers)Some might be both indicative and causal (e.g. unemployment, alcoholism)

IMPLICATIONS-Since risk factors might be underlying issues that arise early on, crime prevention should start at an earlier age Can use risk factors to identify potential offenders and also address issues

ISSUE: it is hard to determine which of these are causal factors and not just indicative o E.g. alcoholism
? Indicative: variations between individuals of antisocial tendency may be reflected by variations in alcohol consumption
? Causal: variations of offending within an individual might be caused by alcohol consumption (more antisocial behaviour during heavier drinking) o Easier to take measures if the factors are shown to be causal of withinindividual variations
? E.g. unemployment
? Unemployed people are more likely to be offenders than employed people (West and Farrington, 1977)
? Unemployment is also a risk factor within individuals who are more likely to offend while they are unemployed ISSUE: prospective prediction is poorer than retrospective prediction o i.e. not all high-risk children become offenders, but a large percentage of offenders were high-risk children

RISK-FOCUSED PREVENTION Farrington: risk-focused prevention is the most cost-effectiveEspecially multi-component programmes (Seattle programme by Hawkins) and educational pre-school programmes

Nagin: investing in police officers would be effective in reducing crime rates, apprehension rates (and hence imprisonment rates)ISSUE: difficult to implement due to politics and possible negative repercussions (Tonry) Mauer: to have a significant economic impact, would have to close entire prisons Sherman: increased investment in hotspot policing could have significant effects

McNeill: investing in one-to-one treatment (which can be revised based on constant review mechanisms) would be the most effective at promoting desistance Overall, most criminologists agree that prison has limited deterrent effectMight even create more problems by labelling/stigmatizing offenders


This method was imported from medicine and public health Since different outcomes might share the same risk factors, tackling the factors would have a wider positive effect NOTE: some risk factors (e.g. race, gender) cannot be changed Would be more effective if the targeted risk factors are causal o BUT since this is difficult to determine, might use interventions that target multiple factors (blunderbuss approach)

KEY INDIVIDUAL RISK FACTORS Low intelligence and attainment--Can be measured early on in life and acts as a good predictor Cambridge Study in Delinquent Development o Longitudinal study of 400 males from age 8 to 48 o Boys who scored 90 or less on an intelligence test at age 8-10 were twice as likely to be convicted as juveniles Can successfully predict both convictions and self-reported delinquency o i.e. the link isn't due to the less intelligent boys being more likely to get caught o Worked to predict offending independently of other risk factors (e.g. family income and family size) Possible explanation: Ability to manipulate abstract concepts o i.e. inability to foresee the consequences of their offences and appreciate the feelings of their victims Possible explanation: Might be related to deficiency in "executive functions" of the brain o Montreal longitudinal study (Seguin, 1995): measure of executive functioning at age 14 was the strongest "neuropsychological discriminator of violent and non-violent boys" Possible explanation: low IQ predicts school failure, which leads to delinquency


Empirical evidence for this risk factor is not as clear Jolliffe and Farrington (2004) o Low cognitive empathy (understanding/appreciating others' feelings) has a strong relation to offending o Low emotional empathy (feeling with others) has less relation to offending o Relationship between low empathy and offending is severely reduced after controlling for other factors (intelligence or socio-economic status)
? Might be less important than these other factors Kaukiainen (1999) and Luengo (1994) found there was a negative correlation between empathy and offending/aggression Jolliffe and Farrington (2007): low empathy was related to self-reported offending


ISSUE: too many different manifestations of a poor ability to control behaviour o E.g. inability to delay gratification, hyperactivity, restlessness o All of these appear to be related to offending Copenhagen Study (Brennan, 1993): hyperactivity at ages 11-13 significantly predicts arrests for violence up to age 22, especially for boys with delivery complications Cambridge Study: based on peer/teacher/parent nominations, boys who were the most impulsive tended to become offenders

Buy the full version of these notes or essay plans and more in our Criminology Notes.