Youth transitions into and out of crime – and what we’re missing

Research using administrative data can offer unprecedented insights into youth experiences of the criminal justice system – but this comes with some challenges. In this blog, Professor Alex Sutherland explores some of the complexities around using this data. Alex is a co-Investigator on the ADR England Research Community Catalyst: Youth Transitions project.

Most young people may never commit a crime, and even if they do, they may not be caught. For some of those who break the law, and who are caught and sanctioned, it can be the start of repeated contact with the justice system throughout their adolescence. For adults, as well as children aged 10 or older, criminal offences, arrests, and most criminal justice outcomes are recorded on the police national computer (PNC). However, the wealth of information held in PNC is not widely used for secure research.

This is partly because criminal records are sensitive personal data, and access is tightly controlled by both police forces (as data owners) and the Home Office. In addition, the processes for accessing the PNC can be complex and not always fully transparent to external researchers. This limits the extent to which researchers working on public policy topics like criminal justice, can use the data for public benefits. One notable exception is the de-identified Ministry of Justice & Department for Education linked dataset – England, which allows accredited researchers working on approved projects to analyse links between childhood characteristics, educational outcomes and re-offending.

Another challenge is that not every criminal justice outcome appears in the PNC. For example, some diversion programmes – which focus on addressing the root causes of offending and divert individuals away from formal prosecution – are not recorded. These programmes, recorded in police systems as ‘Outcome 22’, form an increasingly large part of early ‘disposal’ schemes for under-18s. But one consequence of this is that a key transition into the criminal justice system is being omitted from administrative data. This may have contributed to the number of First Time Entrants to the youth justice system in 2024 being the lowest recorded to date. Previous work using population-level data on first time entrants has shown that the nature of youth justice cohorts was changing as the overall numbers in the system reduced.

By not having an ‘Outcome 22’ code recorded officially, we are in the dark about who is getting these sanctions and whether they are being applied equitably or not. However, Outcome 22 may soon be reclassified as a recordable outcome in police data following a push from the police to make that change. This is partly because some young people would qualify for additional support or different treatment following subsequent arrest if this outcome was recorded.

This change, if it comes, is not intended to criminalise children. Diversion schemes are not formal sanctions and do not appear on criminal records. But we know from previous research that the earlier a child has contact with the criminal justice system, and in particular commits their first offence (whether sanctioned or not), is a strong predictor of later crime or justice involvement. This is particularly the case for persistent offending, where 22% of persistent young offenders were convicted prior to aged 14. Identifying those who may become persistent young offenders earlier is important because about 5% of young offenders end up committing around 50% of all youth crime. We can also use information about the type of offence first committed to help us understand possible future trajectories - although this sort of work is complicated by the fact that early, serious offending is unlikely to be followed by more offending for many children (as has been reported by gov.uk and ADR UK).

Police data is much richer than the PNC, but the actual crimes that people are suspected of and the outcomes associated with them are crucial parts of understanding the crime problems we face now and may face in the future. Furthermore, police data is only one part of the puzzle in youth transitions. When we can combine data from different sources, we increase what we can do with it. For example, in recent work with colleagues, we linked offending with educational data, we looked at how school exclusion relates to future offending and sanctions.

By increasing access to administrative data about the criminal justice system and enabling the secure linking of this data to other sources for research purposes, we can build a much more complete picture of young people’s movement through public services and think about ways to improve those services. This could provide real benefits to young people – for example, providing a long-lasting, single point of contact for children working with several agencies, to support them as they move through education, justice, social services and their adolescent and young adult lives.