The education background of looked-after children who interact with the criminal justice system: December 2022

• More than half (52%) of looked-after children born in the academic year ending 1994 who attended school in England had a criminal conviction by the age of 24 years compared with 13% of children who had not been in care.

• Imprisonment was a relatively unusual outcome, however, 1 in 7 (15%) looked-after children had received an immediate custodial sentence by the age of 24 years; that is over 10 times the proportion of children who had not been in care.

• A further 37% of looked-after children had received a non-custodial conviction, such as a fine or caution; this is compared with 12% of children who had not been in care.

• 1 in 4 (24%) looked-after children who were male had received an immediate custodial sentence compared with 2% of male children who had not been in care.

• Looked-after children were more likely to have been identified as having special educational needs (SEN); among looked-after children who received an immediate custodial sentence by age 24 years, 9 in 10 (92%) had been identified with SEN.

Looked-after children, also known as children in care, were more at risk of interacting with the criminal justice system during early adulthood than their peers. Among looked-after children, over half (52%) had been convicted of a criminal offence by the academic year they turned age 24 years.

Custodial sentences are an unusual outcome for all young people, including looked-after children; 1 in 7 (15%) looked-after children received an immediate custodial sentence. This means they were found guilty of an offence and sentenced to time in a prison, young offender institution, or other secure settings.

A further 2 in 5 (37%) had convictions that did not result in custodial sentences, instead receiving sentences such as a fine or community service.

By comparison, 1% of children who had not been in care received an immediate custodial sentence by age 24 years.

It is important to note that there are separate sentencing guidelines for children and young people aged under 18 years, and that these vary according to the age of the young person and the nature of the offence.

Of the group of people who received custodial sentences, 18% had been classed as looked-after children. Further detail of the education and demographic background of all young people who interacted with the justice system has been published by the Office for National Statistics (ONS).

Looked-after children typically received their first custodial sentence at a younger age than their peers who had not been in care.

On average, looked-after children received their first custodial sentence in the academic year they turned 18 years. For children who had not been in care, this was in the academic year they turned 20 years.

Around 30% of looked-after children who received an immediate custodial sentence by age 24 years had received one by the academic year they turned 16 years. For children who had not been in care, the figure was 9%.

The most common offences leading to custodial sentences by the age of 24 years were theft offences and summary non-motoring offences; 29% of custodial sentences received by looked-after children were for theft offences and 20% were for summary non-motoring offences. For people who had not been in care, 20% received sentences for theft and 15% for summary non-motoring offences.

Summary offences include cases such as minor criminal damage or common assault. They are tried by Magistrates’ Court, who cannot impose a custodial sentence of more than six months. These offence types typically carry lighter sentences and receiving an immediate custodial sentence for these generally indicates a pattern of offending.

The offences that looked-after children committed that led to an immediate custodial sentence were more likely to be for theft and summary non-motoring offences, but less likely to be for violence, drugs or weapons. It is important to note that this figure is the percentage share of offences committed rather than the total number of offences. This does not mean that children who did not experience care committed fewer offences of these types than looked-after children, rather that the pattern of offence type varies between the two groups.

On average, looked-after children who received an immediate custodial sentence by age 24 years committed twice (1.9 times) as many offences leading to immediate custodial sentences as children who had not experienced care.

Looked-after children were more likely to have been identified with special educational needs (SEN) during their schooling. Among looked-after children, 4 in 5 (82%) had been identified with SEN, compared with 1 in 3 (34%) of those who had not been in care.

Looked-after children were more likely to have greater levels of SEN than children who had not been in care. During this time, there were three levels of SEN provision: School Action, School Action Plus and SEN Statement. Around 1 in 3 (33%) looked-after children had been recorded with a SEN statement, the highest level of need, compared with 1 in 20 (4%) children who had not been in care.

Children who interacted with the criminal justice system were more likely to have SEN and more likely to have more severe SEN than children who did not interact with the criminal justice system.

This was especially true for looked-after children. Almost all (92%) looked-after children in the data who received an immediate custodial sentence by age 24 years had been identified with SEN and 40% of looked-after children who received a custodial sentence by age 24 years had a SEN statement during their schooling.

The SEN type that looked-after children were most likely to be recorded with was behavioural, emotional and social difficulties (BESD), with 1 in 2 (50%) looked-after children being recorded with BESD as their primary or secondary SEN type.

BESD was more prevalent among young people who had interacted with the criminal justice system, especially among looked-after children. Three-quarters (76%) of looked-after children who received an immediate custodial sentence by age 24 years had been identified with BESD, compared with half (51%) of children who had not been in care.

This shows that looked-after children were more likely to be identified with behavioural issues at school. For example, looked-after children who interacted with the criminal justice system were more likely to have been suspended or expelled from school.

Among looked-after children who received an immediate custodial sentence by age 24 years, 1 in 5 (18%) had been expelled and 4 in 5 (81%) had been suspended. This compares with an expulsion rate of 1 in 60 (2%) for looked-after children who had not interacted with the criminal justice system, and a 1 in 4 (24%) rate of suspension.

Looked-after children had lower rates of educational attainment at both key stage 2 and key stage 4. However, the relationship between offending and attainment differs by key stage.

Children are generally expected to attain level 2 by the end of key stage 1 and level 4 by the end of key stage 2.

Looked-after children that did not receive an immediate custodial sentence by age 24 years were the least likely to have attained the key stage 2 benchmark.

Among looked-after children, over half (57%) of those that did not interact with the criminal justice system had reached the level 2 benchmark in English and maths. Attainment was, however, higher among looked-after children who did go on to interact with the criminal justice system. Three-quarters (71%) of looked-after children who received an immediate custodial sentence by age 24 had reached this benchmark.

This pattern changes by the time young people reach GCSE years. There are less data available on GCSE results compared with key stage 1 and 2 tests. However, the results show that pupils with low attainment at key stage 4 were more likely to interact with the criminal justice system by age 24 years. Pupils with low attainment at key stage 4 were more likely to interact with the criminal justice system by age 24 years. Looked-after children who received an immediate custodial sentence by age 24 years generally had lower attainment at key stage 4 than those who did not interact with the criminal justice system, or who received a non-custodial sentence or caution.

Around 1 in 6 (16%) looked-after children who did not interact with the criminal justice system by age 24 years passed five GCSEs A* to C including English and maths. This compares with 1 in 50 (2%) looked-after children who received an immediate custodial sentence by age 24 years passing five GCSEs A* to C.

The vast majority (86%) of looked-after children who received an immediate custodial sentence were male, compared with 94% of children who had not been in care. However, looked-after children who received a non-custodial sentence or caution by age 24 years were equally likely to be male or female. In comparison, 30% of children who had not been in care and received a non-custodial sentence or caution were female.

Around 1 in 4 (24%) looked-after children who were male received an immediate custodial sentence compared with 2% of male children who had not been in care. For females, 1 in 20 (5%) looked-after children received an immediate custodial sentence compared with less than 1 in 500 children who had not been in care. This is 25 times higher.

Looked-after children were less likely to be White or Asian and more likely to be Black or Mixed ethnicity compared with children who had not been in care.

Black (19%) and Mixed (21%) looked-after children had higher rates of imprisonment than White (15%) or Asian (7%) looked-after children. White looked-after children were most likely to receive a non-custodial sentence or caution (40%).

Because of small numbers in some ethnic groups, this analysis reports statistics for aggregated ethnic groups. Readers should be aware that outcomes can vary between different ethnic groups within the aggregated group.

Looked-after children (CLA)

In England, a child looked after is one that is either provided accommodation by, or in the care of, a local authority. The term “looked-after children” includes unaccompanied asylum-seeking children, children in friends and family placements, and those children where the agency has authority to place the child for adoption. It does not include those who have been permanently adopted or who are on a special guardianship order.

A child in custody is automatically considered looked after. For the purpose of this analysis, incidences of being looked after because of being in custody have been filtered out to avoid misleading the reader.

Custodial sentence

A custodial sentence is one where the offender is imprisoned or held in another institution; for example, a Young Offender’s Institution or secure hospital. Custodial sentences are generally reserved for the most serious offences. For the purpose of this analysis, “imprisonment” refers to an immediate (non-suspended) custodial sentence.

Non-custodial sentence

A sentence where the offender is not imprisoned or held in an institution. Examples of non-custodial sentences include community sentences such as unpaid work or a curfew, or fines. They are used as punishment for a range of crimes, including motor offences and theft.

Special educational needs

A child or young person has special educational needs (SEN) if they have a learning difficulty or disability that calls for special educational provision to be made for them, that is, educational or training provision that is additional to or different from that made generally for their peers.

From 2002 until 2014, there were three levels of SEN provision: School Action, School Action Plus and SEN Statement. These levels reflected a graduated approach to support, with SEN statements generally reserved for those requiring a level of support beyond what most mainstream schools could typically provide, such as access to certain therapies. This analysis looks at whether an individual has ever been classed as having SEN since the academic year ending 2003.

From 2015, the School Action and School Action Plus categories were combined to form one category of SEN support. Following the introduction of education, health and care (EHC) plans in September 2014, statements of SEN and EHC plans were grouped together within the data. A pupil has an EHC plan when a formal assessment has been made.

Behavioural, emotional and social difficulties (BESD)

A type of special educational need. Children and young people with BESD experience challenges with personal, social and emotional development. This can have an impact on behaviour and emotional responses, and their relationships with self. Since 2014, “Social, emotional and mental health" (SEMH) has replaced the term BESD and emphasises that behaviour is a symptom of the needs that underlie it.

Ministry of Justice – Department for Education data share

This analysis uses Ministry of Justice (MoJ) and Department for Education (DfE) data that have been linked as part of the Data First project. The data share includes data from prison, courts, Police National Computer (PNC), National Pupil Database (NPD), looked-after children (CLA) and children in need (CIN). It covers variables such as attainment, demographics, offending data, school exclusions and children in care.

There are some data limitations to be aware of as is expected with data collected for administrative, rather than statistical, purposes. Overall, the MoJ and the DfE assesses the match rate between the PNC and the NPD as good (PDF, 2.02MB). The data were linked using a rules-based approach and the data linking accuracy therefore varies. For example, it is harder to determine that records belong to one person if they have used different names and moved address often. As expected, a share of offenders is unmatched to the NPD as they did not attend education in England. The MoJ and the DfE report that the matched dataset has similar characteristics to the unmatched dataset in terms of gender and age, but that there are some noted differences for ethnicity.

Further details of the linking process and data quality are available in section one of the MoJ-DfE Technical Note (PDF, 2.02MB) and in the ADR-UK Data quality report (PDF, 3.05MB).

Methodology

The methodology for this analysis is largely identical to the methodology used in the previous publication looking at all children. It is important to note that the figures may not be directly comparable with other analysis because of coverage and linkage rates.

Coverage and context

This analysis explores the education and social care background of care-experienced young people who were imprisoned at any point up to the age of 24 years. It looks at a single cohort of individuals who were born in 1993 to 1994 and undertook at least part of their schooling in England and so appear in the National Pupil Database. This analysis focuses on those that were classified as “looked-after children” at any point from academic year ending 2006 to the academic year ending 2013.

The cohort is split into three groups:

• those that had received a custodial sentence in the UK (such as imprisonment or time served in another institution)
• those that had been convicted of a crime and received a non-custodial sentence, for example, a fine
• those who did not have any criminal convictions

In total, 687,101 individuals are included in the analysis, of which 12,420 were classed as looked-after children.

As is common with administrative data, coverage varies over time. Years of coverage are reported on all figures and the cohort was selected to maximise available data as far as possible.

Because of the sensitivity of data relating to looked-after children, counts have been rounded to the nearest 10, and counts smaller than 6 have been suppressed. As a result, counts may not sum to 12,420 in every instance. For disclosure control, all percentages are rounded to the nearest number.

Missingness

Where valid data are not available for the entire cohort, the number missing are reported in the notes accompanying each figure. Relative proportions (percentages) are calculated from the valid data only.

Not all variables have a “missing” category because of how they are recorded. For example, a person either has recorded exclusions or they do not. In contrast, if a person is missing attainment data at key stage 4, this does not mean they failed to attain the qualification, but that the data are not included in the National Pupil Database.

Education data

To maximise coverage, the analysis includes all individuals who have a record in the National Pupil Database (NPD). This means the data include some who did not appear in the records for all stages of education. The advantage of this approach is that it does not exclude children who appear in the data at one point but missed formal full-time education (PDF, 1.32KB) at another time, some of whom will be among the most vulnerable. There will also be instances of children and young people who moved away from England or the rest of the UK. This is an important difference between this work and other studies, such as the recent Ministry of Justice and Department for Education article, which examined only young people who had key stage 4 records (PDF, 2.02MB).

The limitation of this approach to coverage is that there are not the same amount of data available for every individual included in the cohort. For this reason, this article focuses on whether someone has ever met certain criteria, rather than looking at a single stage of schooling.

Coverage of independent school attendees is limited to key stage 4 attainment data. As many children attend state schools for the earlier stages of education before moving to an independent school, characteristics have been taken from the School Census where this is the case. Because of this increased coverage, there are more records covering key stage 4 than the other stages of schooling.

Demographic data

All demographics reported in this analysis are taken from the School Census, with the most recently recorded response used where possible, including refusals. As previous, demographic data are limited for independent school attendees.

Justice system outcomes

Individuals in the study were assigned to one of the three outcome groups depending on their interactions with the justice system. These outcomes are based on data from the Police National Computer (PNC), a database used by law enforcement organisations across the UK.

First, the data were filtered for those that had been found guilty of an offence or had an outcome classed as “Caution/Warning/Reprimand”. This removed individuals who had been investigated or tried but were not found to have committed an offence.

Next, the details of the sentence were used to determine whether someone had received a custodial or non-custodial sentence for an offence they were guilty of. For example, “Imprisonment” or “Return to Prison” denote custodial sentences, whereas “Community Order” or “Driving Licence Revoked” are two of the non-custodial sentences included.

Individuals were then assigned a group according to the highest level of sentence received, either the “imprisoned” group for those with a custodial sentence, or “non-custodial” for those without.

Finally, those that did not appear in the Police National Computer, or who were filtered out because they had not been found guilty, were assigned to the “no convictions” group.

All convictions from the age of criminal responsibility and above are within scope. In England and Wales this is age 10 years. However, it is important to recognise that sentencing guidance varies according to age.

The authors welcome feedback and the opportunity to learn from best practices. This section details the planned developments.

This article builds upon previous work looking at the education and social care background of young people who interacted with the justice system, which gave an overview of outcomes for all young people regardless of whether they had interacted with children’s social care.

Additional analysis in collaboration with the Office for Standards in Education, Children's Services and Skills (Ofsted) will be published in early 2023. This will model the predicted probabilities of interaction with the justice system for all children and include pupil- and school-level controls.

The next stage of this project focusing on looked-after children will develop the analysis in this article to incorporate qualitative insights.

Given the richness of the dataset, the team is also interested in developing future projects analysing the links between education and the justice system, particularly where it supports an improved understanding of societal inequalities and how they could be tackled by policy development.