In this blog, ADR UK Research Fellow and ISER Senior Research Officer, Dr Paul Garcia Hinojosa introduces his project which is using linked administrative data to explore how socio-emotional characteristics observed in early childhood relate to offending behaviour during adolescence. The research aims to inform early interventions that could help prevent young people from becoming involved in crime. Read on for more information and early findings.
Early cognitive and socio-emotional development are key determinants of later life outcomes. My project explores how socio-emotional characteristics identified in early childhood are associated with offending behaviour during adolescence. I also examine the pathways through which these characteristics shape offending behaviour as children transition into adulthood.
To do this, I am using linked, de-identified administrative data from the Ministry of Justice and the Department for Education. This dataset provides information on socio-emotional assessments from the Early Years Foundation Stage Profile, recorded at the end of the school reception year when children are aged five, as well as criminal records from the Police National Computer for ages 10 to 17.
The data allows me to follow anonymous pupils across their full period of compulsory education. This makes it possible to account for the role that school-related experiences play in shaping the relationship between early socio-emotional characteristics and later offending.
Identifying these early characteristics and understanding their long-term impacts can help inform early interventions and strategies to prevent children’s involvement in crime.
My initial findings
Children who went on to offend show lower scores across all scales of early child development on the Early Years Foundation Stage Profile. My initial analysis identifies two key underlying dimensions: cognitive and socio-emotional difficulties.
- The cognitive dimension reflects challenges in language, communication, literacy, problem solving, reasoning, and numeracy.
- The socio-emotional dimension is primarily characterised by poor emotional regulation and difficulties in forming relationships.
After controlling for a range of background characteristics, I find a significant association between early developmental difficulties and the probability of being cautioned or convicted for an offence during adolescence.
The effects of cognitive difficulties on later offending appear to be largely mediated by school-related experiences such as persistent absenteeism, exclusions, and low academic attainment. Socio-emotional difficulties, however, remain significantly associated with a higher probability of being cautioned or convicted for an offence in adolescence – even after accounting for potential mediating pathways and group-level variability.
What comes next?
My next step is to examine whether early socio-emotional development can predict the type of offence children may become involved in during adolescence. I will follow existing literature on the classification of serious and non-serious violent offences.
I also plan to investigate how broader contextual factors – such as characteristics of the school or local authority – interact with early development in shaping offending outcomes. Previous research suggests that these wider contexts can play an important role in adolescent behaviour.
In particular, I will examine local authority-level variables such as spending on services that both directly support children’s socio-emotional development and indirectly promote positive behaviour and wellbeing. I plan to draw this data from publicly available sources.
Sharing what I’ve learned
I will be presenting my initial findings at the Stockholm Criminology Symposium and the ADR UK Conference 2025. I am also actively engaging with organisations that share similar research agendas, particularly those focused on early intervention, youth justice, and data-driven approaches to social policy.
If you are attending either event, or if this research resonates with your work, I would be very happy to connect. Feel free to get in touch!
Read more about Paul’s project.
Find out about how ADR UK provides secure access to deidentified data for researchers.