The Edinburgh Study of Youth Transitions and Crime

1. The Edinburgh Study of Youth Transitions and Crime Lesley McAra and Susan McVie University of Edinburgh

2. Overview of seminar • Programme aims and objectives • Current phase of Study • Key findings relating to youth justice • Policy implications

3. The Edinburgh Study Longitudinal study of pathways into and out of offending Funded by: ESRC, the Nuffield Foundation, and the Scottish Government Aims to understand: - Why some young people become heavily involved in crime and why most stop - Gender differences in offending - The influence of social and neighbourhood context - The impact of contact with agencies of control on subsequent behaviour 3

4. The cohort Target group: children in Edinburgh aged 12 in autumn 1998 Mainstream (all 23), special (9 out of 12) and independent schools (8 out of 14) Cohort size: 4,380 Response rate in participating schools: - 95% up to sweep 4 - 90% at sweep 5 - 81% at sweep 6 4

5. Data sources Self report questionnaires (6 annual sweeps) Semi-structured interviews (sweeps 2 and 6) School, social work, children’s hearings records (annual sweeps) Teacher questionnaires (1999) Police juvenile liaison officer records Scottish criminal records (conviction data up to age 22) Parent survey (2001) Geographic information system 5

6. Current phase • Funded by Nuffield Foundation and undertaken in collaboration with the Scottish Government • Aims: • to map the criminal justice careers of cohort members (from age 8, the age of criminal responsibility in Scotland, to 22) • to explore transitions from the juvenile to adult system • to assess the impact of these careers on desistance from criminal offending. • Follows up sub-sample of around 444 cohort members, including: - all those with offence referral to the children’s hearing system - two control groups (one matched to those with early history of referral; one matched to those who had a first referral at age 15).

7. Key Findings Relating to Youth Justice

8. Four ‘facts’ • Persistent serious offending is associated with victimisation and social adversity • Early identification of at-risk children is not a water-tight process and may be iatrogenic • Critical moments in the early teenage years are key to pathways out of offending • Diversionary strategies facilitate the desistence process.

9. On the basis of these facts….. Conundrum facing policy-makers: how to develop a system of youth justice which is holistic in orientation (intervention proportionate to need) AND which maximises diversion from criminal justice? Solution: age-graded services and support to include ‘universal targeting’ in the early years and more finely tuned individual targeting in the teenage years Social justice not criminal justice

10. Claim1: evidence Persistent serious offending is associated with victimisation and social adversity

11. % involvement in violent offending(Robbery, carrying weapon, 6+ incidents of assault in past year)

12. % North Edinburgh state school pupils involvement in violent offending(Robbery, carrying weapon, 6+ incidents of assault in past year)

13. Violence and vulnerability

14. Violence and vulnerability cont.

15. Violence and vulnerability cont.

16. Claim 2: evidence Early identification of at-risk children is not a water-tight process

17. Majority of serious and persistent offenders under the radar (based on self-report data) • - Serious offending: 6+ incidents of assault; robbery; weapon carrying; fire-raising • housebreaking; breaking into motor vehicle to steal; riding in stolen motor-vehicle; • - Chronic high level serious offenders: 11+ incidents at every study sweep • - Violence: 6+ incidents of assault; robbery; weapon carrying • Chronic violence: admitted to at least one violent offence every study sweep

18. Majority of serious and persistent offenders amongst north Edinburgh pupils under the radar (based on self-report data) • - Serious offending: 6+ incidents of assault; robbery; weapon carrying; fire-raising • housebreaking; breaking into motor vehicle to steal; riding in stolen motor-vehicle; • - Chronic high level serious offenders: 11+ incidents at every study sweep • - Violence: 6+ incidents of assault; robbery; weapon carrying • Chronic violence: admitted to at least one violent offence every study sweep

19. How soon can we tell? cont. 19

20. How soon can we tell? cont Inability to identify the vast majority of serious and persistent (self-reported) offenders from an early age Dunedin longitudinal study (see White et al. 1990) - 19% wrongly predicted by age 11 (around 1 in 5 false positive rate) - 35% wrongly predicted by age 15 (around 1 in 3 false positive rate) - Predictability declines in the mid teenage years as other influences become important “Due to the high rate of false positives among those children predicted to have antisocial outcomes, the usefulness of preschool behaviour predictors for selecting children for intensive early intervention efforts may be limited at present” (pp 523) 20

21. Claim 3: evidence Critical moments in the early teenage years are key to pathways out of offending

22. Conviction trajectories(McAra and McVie 2010 in press) 22

23. Conviction trajectories(McAra and McVie 2010 in press) 23

24. What distinguishes the people who were convicted early from each other at age 12? • Nothing … • … but, they are equally problematic! • They are similar on the following measures: • They live in the top 25% deprived neighbourhoods • They are entitled to free school meals • They come from broken families with poor supervision and high levels of conflict • They belong to delinquent peer groups • They hang around the streets most days • They are badly behaved at school • And they are impulsive, social alienated and morally accepting of offending behaviour … • and…

25. …they reported exactly the same levels of serious offending

26. What changed between age 12 and 15? • The group that went on to have a pattern of ‘chronic’ convictions got worse in terms of: • truancy from school • exclusion from school • getting into trouble with the police • receiving statutory supervision from CHS NOTE: the groups did not differ on any of these things at age 12

27. What about those young people who were not convicted until later? • Those convicted later showed problematic signs at age 12: • Parental conflict, family breakdown and poor supervision • Alcohol use and impulsive tendencies • Weak attachment to school and peer involvement in offending • But they were more advantaged than others at age 12: • Less socially deprived • Less involvement in offending, hanging around and drug use • Better behaved at school, less truancy and exclusion • Less contact with the police or the CHS • Better social skills and a better moral attitude

28. So what changed to lead to their conviction between age 12 and 15? • On some measures they were similar to early onset groups at age 12 but got worse at age 15: • Family breakdown and parental supervision • Involvement with delinquent peers • On some measures they were better than early onset groups at age 12, but became the same as them by age 15: • More likely to be living in an area of social deprivation • More likely to hang out on the streets, take drugs and get involved in serious offending • On other measures they were better than early onset groups at age 12 and got worse by age 15, but so did the early onset chronics: • More likely to truant from school and get excluded • More likely to get into trouble with the police

29. Claim 4: evidence Diversionary strategies facilitate the desistence process

30. The usual suspects(McAra and McVie 2005, 2007a) • Working cultures of police and Reporter mean that certain categories of youngsters are constantly recycled into system whilst other equally serious/vulnerable offenders escape tutelage of agencies altogether • Children with ‘previous form’ - 7 times more likely to be formally charged by police - 4 times more likely to be referred by police to Reporter - 3 times more likely to be brought to a hearing by Reporter (the ‘usual suspects’ are mostly boys, from socially deprived areas living in single parent households)

31. Damaging features of system contact (McAra and McVie 2007a) • Compulsory measures of care appear to inhibit the normal process of desistance from serious offending that is evident from around age 14 in the cohort • Conversely police warnings/charges (but no further action) associated with a significant reduction in serious offending one year later • Edinburgh Study findings in tune with other international comparative research e.g. Denver/Bremen longitudinal studies (Huizinga et al. 2003)

32. Impact of agency contact • We looked at 3 levels of agency contact at age 15:- • Being ‘charged’ by the police • Being referred to the Reporter, but no action • Being referred to the Reporter, and brought to a hearing • To make sure we were comparing like with like we ‘matched’ young people on the basis of their offending, background characteristics, lifestyles, risk factors and family backgrounds. • We then compared each set of matched pairs to see how their offending changed (intervention group v comparison group)

33. % Involvement in serious offending one year later

34. Within group % change in serious offending from age 15 to age 16

35. Supervision requirements

36. Regular one-to-one contact with a social worker • Regular one-to-one contact with social worker age 15: statistically significant decline (p <.000) in serious offending over the next year • Lack of regular one-to-one contact with social worker age 15: no statistically significant change in serious offending

37. Youth to adult criminal justice transitions: up-tariffing the vulnerable(McAra and McVie 2007b, 2010a and 2010b)

38. Youth to adult criminal justice transitions: up-tariffing the vulnerable(McAra and McVie 2007b, 2010a) • Key factors predicting transition from children’s hearings to adult system are: - Excluded from school by 3rd year of secondary school - Early history of police warning/charges - Being male - ***Assessed as most ‘needy’ in reporter files***

39. The revolving door(McAra and McVie 2010b)

40. Longer term impact of school exclusion

41. Lessons from the Scottish Case

42. Core messages • Persistent serious offending linked to victimisation and social adversity • Early identification difficult and risk of labelling (creating a self-fulfilling prophecy) • Critical moments in early teenage years key to pathways out of offending • Diversionary strategies effective Key question: how to develop a youth justice policy which is both holistic (intervention proportionate to need) and maximises diversion from criminal justice?

43. 43

44. Themes and questions for discussion • The disjuncture between self-reported offending and institutional contacts • Vulnerable transition points: - primary to secondary education - leaving institutional care - exiting the children’s hearing system - leaving prison • Given the ‘facts’ about youth crime and justice what are the key gaps in current service provision and how could existing services be made more effective? • The challenges posed by the current economic context and the likelihood of a change in Scottish government in 2011…..

45. References:McAra and McVie, 2005, The Usual Suspects? Young People, Street Life and the Police, Criminology and Criminal Justice, 5 (1):1-36McAra and McVie, 2007a, Youth Justice? The Impact of System Contact on Patterns of Desistance from Offending, European Journal of Criminology, 4 (3)McAra and McVie, 2007b, Criminal Justice Transitions, Research Digest, no. 14McAra and McVie, 2010a, Youth Crime and Justice: Key messages from the Edinburgh Study of Youth Transitions and Crime, Criminology and Criminal Justice, 10 (2): 179-209. McAra and McVie, 2010b (forthcoming), Criminal Justice Pathways: Key findings from the Edinburgh Study of Youth Transitions and Crime, Research Digest, no. 15www.law.ed.ac.uk/cls/esytc/