Research and analysis

Technical guide: Prison education in England: Educational background, characteristics, and criminogenic needs

Published 11 September 2025

Applies to England

1. Introduction

This technical guide accompanies the Ministry of Justice (MoJ) ad hoc statistics report Participation in prison education in England. It provides additional detail across the following areas:

• use of administrative data
• data sources
• definitions and cohorts included
• methodology
• data linkage and governance
• dimensions of statistical quality
• Users
• Contact details and feedback

1.1 Use of administrative data

The administrative data sources used in this publication have been collected for operational purposes. A common limitation of data from large-scale administrative record systems, is the possibility of mistakes with entering and processing the data. The quality of the information held in these operational databases, while generally high, does differ by data field, depending on the field’s frequency of use and importance in the day-to-day running of the courts, prisons and probation systems.

Details of some of the administrative data sources used in the production of this release can be found in the MoJ Statement of Administrative Sources.

2. Data sources

2.1 CURIOUS

CURIOUS is the operational database for capturing attendance and learner outcomes under the Prison Education Framework (PEF); it was launched in April 2019.

CURIOUS profiles are created for prisoners during their education induction, so not all prisoners appear on the system and multiple profiles may be created as prisoners transfer between establishments. CURIOUS records initial assessments in maths and English and details about courses undertaken that are delivered through PEF. It is used to provide information on participation and achievement in prison education.

It is a contract management system and has robust quality assurance checks in place. There is a strong degree of confidence in the quality of the data. However, coverage is not complete: private prisons and Welsh prisons are not mandated to use CURIOUS and recording of courses may be undercounted in these establishments. Education courses funded through other funding routes are not recorded.

2.2 MOJ and HMPPS data sources

2.2.1 National Delius (nDelius)

National Delius (nDelius) is the probation case management system. It captures information about offenders receiving probation supervision upon release from prison, as well as those serving community sentences. nDelius is used to define the cohort released under supervision and to provide information on prisoner characteristics and offences.

It is used within a number of National Statistics publications and measurement of key performance indicators (KPIs) as well as offender management.

2.2.2 Prison National Offender Management Information System (p-NOMIS)

The Prison National Offender Management Information System (Prison NOMIS or p-NOMIS) is the operational database used to hold information on prisoners, their movements and activity while in public prisons. It is used to provide a location history for custodial sentences to exclude offenders serving part of their sentence outside of the England adult public prison estate and to calculate length of custodial sentences including time on remand. It is also used within the Ministry of Justice-Department for Education (MOJ-DfE) data share for release dates from prison (see section 2.3).

It is widely used by MOJ for official statistics and measurement of KPIs as well as offender management within prisons.

2.2.3    Offender Assessment System (OASys)

The Offender Assessment System (OASys) is used by His Majesty’s Prison and Probation Service (HMPPS) in England and Wales to measure and record the risks and needs of offenders in custody or under supervision in the community. Offender assessments are conducted on individuals serving sentences in custody and in the community and include a wide range of information on offending-related risks and needs. 

Layer 3 assessments (‘full assessments’) contain detailed questions in relation to the risks and criminogenic needs related to an individual’s offending behaviour. Section 4 relates to education, training and employability. The layer 3 assessment closest in time to the conviction date is used as a proxy for needs at start of time in custody for individuals in cohort 1.

Relies on the prisoner’s disclosure and availability of supporting sources of evidence, as well as the capacity of operational staff to conduct the assessment.

Data from OASys is not regularly published but has been used in a number of research and analysis publications. Coverage of assessments is very high for offenders serving custodial sentences and under probation supervision on release from custody.

2.3 MOJ-DfE data share

2.3.1 Police National Computer (PNC)

Criminal Justice data originating from the PNC is not used in this report but identifiers from the PNC were used to create the MOJ-DfE data share.

The PNC is an administrative data system used by police to monitor recordable offences, those convicted or cautioned for them, and the outcomes received. All offences that can attract a custodial sentence are included so all prisoners released and under probation supervision are in scope.

2.3.2 National Pupil Database (NPD)

The National Pupil Database is a large, linked data resource created by DfE for analytical purposes which combines a number of datasets that hold information on educational performance and context for pupils in schools in England. It has been linked to the PNC and other MOJ datasets as part of the MOJ-DfE data share, which is managed by Data First.

The NPD includes information on educational attainment, exam results and progression for pupils at different key stages. It also holds information collected annually on pupil characteristics, such as their gender, free school meals status, exclusions and attendance.

The component datasets used in this report are:

• pupil level annual school census
• annual pupil referral unit (PRU) census
• children looked after dataset
• absence and exclusions dataset
• key stage 4 attainment data
• 16 to 18 (key stage 5) attainment data

This data is less up-to-date and not linked to all MOJ sources. Only younger prison leavers who attended school in England are included. Please see section 3.5 for details of matching. Indicators are selected for availability across multiple years. Please see section 3.4 for details.

3. Methodology and definitions

 3.1 Prison leavers included

Two cohorts are considered:

Cohort 1: Adult offenders released from 1 January 2024 to 31 December 2024, after serving sentences of at least 3 months, including time spent on remand, in public prisons in England.

Cohort 2: Adult offenders released from June 2019 to December 2020, after serving sentences in public prisons in England, and for whom DfE data has been linked to provide information on education experiences while in school. Due to restrictions around data linking analysis for this cohort does include time on remand.

Information (for both cohorts) relates to prison leavers: 

• who were released in the time period and were subsequently under the supervision of the probation service;

• who were aged 18 or over at their release date;

• who were previously held in adult public prisons (directly managed by HMPPS) in England for all of their sentence (or released from these prisons in cohort 2);

• who had been in prison custody for at least 3 months before their release (to allow time for participation in courses) including time on remand for cohort 1. Findings on sentence length discuss participation among prisoners leaving prison after less than 3 months.

It does not include: 

• prisoners under the age of 18 or those held in the Youth Prison Estate;

• prisoners serving their sentence in Wales or in private prisons in England (or released from these prisons in cohort 2);

• prisoners held on remand only;

• those who did not receive probation supervision.

16,810 offenders who had served their whole sentence in public adult prisons in England were released from prison between 1 January and 31 December 2024 (an additional 9,773 were released after serving sentences of less than 3 months). This is considerably fewer than the total number of prisoners released over the timeframe due to the levels of movement between English, Welsh, public and private prisons.

The make-up of prison leaver cohorts may differ from the full prison population at a single point in time. By focusing on those being released from prison, offenders serving longer sentences will be under-represented in the analysis, while those serving very short sentences (i.e. less than 3 months) are excluded from the majority of the statistics.

Cohort 2 (DfE-linked) only:

Information about this cohort relates to prisoners:

• aged 18 to 35;
• who attended state-funded schools in England since 2001/02;
• who were released from English adult public prisons between June 2019 and December 2020 (they may have spent time in Welsh or private prisons during their sentence).

In addition to the different timeframe the make-up of the DfE-linked cohort differs from cohort 1 because it does not include:

• older offenders;
• offenders who grew up outside of England;
• offenders who solely attended independent schools or were homeschooled.

3.2 Prison education courses

3.2.1 Accredited and unaccredited courses

CURIOUS includes both accredited and unaccredited education courses. To be included as an accredited course, the course code must have been successfully matched to the Learning Aim Reference Service (LARS). Most aims lead to recognised qualifications regulated by Ofqual. Some learning aims refer to stand-alone course units rather than to full multi-unit qualifications.

Unaccredited courses include a diverse range of learning activities that are generally less formal and do not lead to a recognised qualification. They may comprise stand-alone course units from larger qualifications, or locally developed materials and one-off training sessions. Unaccredited courses are excluded from these statistics. For context though, off the prison leaver cohort, 48% participated in unaccredited courses, including 22% who participated only in unaccredited courses.

The term accredited courses used in this publication is distinct from ‘Accredited Programmes’, which are rehabilitative interventions developed to target the risks and needs for different types of offending behaviour, which are not covered in this release.

3.2.2 Participation and achievement

An offender counts as participating in a course when they have registered to attend, and the course has a start date between their arrival in prison (and following the introduction of CURIOUS in April 2019) and their release date (on or before 31st December 2024). This does not consider the level of engagement or session attendance.

An individual counts as achieving in a course when they have completed all the work and have been awarded a passing grade or level. Partial achievement is not counted and reasons for course withdrawal or non-completion are not considered.

3.2.3 Qualification groups and subjects

Functional skills courses in English, maths, Information and Communication Technology (ICT) and English for Speakers of Other Languages (ESOL) form a core part of the curriculum offer in prisons. These are intended to provide the fundamental skills that people need for their working and personal lives. Prisoners study for the qualifications in practical ways and apply core skills to real-life situations.

HMPPS refers to all education courses that are not English or mathematics as vocational courses. This category includes ICT and ESOL courses, as well as a small number of more traditionally academic subjects and qualifications (such as science GCSEs) alongside employment-focused Subject Sector Areas such as Construction, Planning and the Built Environment or Health, Public Services and Care.

3.2.4 Course levels

Most qualifications have a difficulty level. The higher the level, the more difficult the qualification is. There are 9 qualification levels in England (and in Wales and Northern Ireland) which are summarised in table 1 below.

Table 1: Qualification levels and what they mean

Level	Example qualifications	Indicative skill level[footnote 1]
Entry level (1,2 and 3)	Entry level functional skills Entry level English for speakers of other languages (ESOL)	Equivalent to literacy levels spanning ages 5-11. Adults with skills below Entry Level 3 may not be able to understand labels on pre-packaged food or understand household bills.

Level 1	GCSE grades 3 to 1 or grades D to G Level 1 functional skills Level 1 award or certificate	Adults with skills below Level 1 may not be able to read bus or train timetables or understand their pay slip.
Level 2	GCSE grades 9 to 4 or grades A* to C Level 2 functional skills Level 2 ESOL	Adults with skills below Level 2 may not have the skills to spot fake news or bias in the media.
Level 3	A-level (Advanced Level) Level 3 BTEC	Advanced skills
Levels 4-8	University degree Higher National Diploma (HND)	Higher level skills

3.2.5 Prison education out of scope

The following courses delivered in prisons are not included:

• courses pre-dating the introduction of CURIOUS in April 2019 or delivered under Offenders’ Learning and Skills Service (OLASS) funding when prison education was led by DfE;
• unaccredited courses recorded on CURIOUS (and not matched to a Learning Aims Reference Service course code);
• local offers funded through the Prison Education Dynamic Purchasing System (DPS): an alternative, more flexible commissioning route, which allows prison governors to deliver courses tailored to local requirements.

The statistics also do not cover wider education-related activity within prisons in England such as:

• Participation in apprenticeships^{[footnote 2]} or outside courses through Release on Temporary Licence (ROTL)^{[footnote 3]} during their sentence.
• Collaborative arrangements with third sector organisations. For example:
  • Prisoners’ Education Trust (PET) which receives an annual grant from the MOJ and funds distance learning courses in every prison in England and Wales, including Open University modules.^{[footnote 4]}
  • Shannon Trust which offers reading and numeracy programmes that are available in all prisons across England and Wales (and also Northern Ireland).^{[footnote 5]}
  • Innovative employment-focused partnerships in individual prisons such as the Clink^{[footnote 6]} or Redemption Roasters.^{[footnote 7]}
• Embedded learning in work areas across the regime (e.g. workshops, kitchens/serveries and wing work)
• Accredited programmes (behavioural interventions and similar programmes developed to target the risks and needs for distinct types of offending behaviour).

3.3 OASys data processing and indicators 

Data from the Offender Assessment System (OASys), a tool used by HMPPS to assess offender risks and needs and support them during custody and in the community, have been matched to cohort 1.

• Layer 3 assessments (‘full assessments’) have been used. These contain detailed questions in relation to the risks and criminogenic needs related to an individual’s offending behaviour. Section 4 of this template asks detailed questions relating to education, training and employability.

• Only assessments completed within 2 years of their conviction date are considered.

• The assessment with a completion date closest in time to the offender’s conviction date is selected.^{[footnote 5]} This is chosen to provide a ‘baseline’ picture of the period closest to their engagement with prison education services; however it does not take into account learning during previous spells in custody or that prisoners may have been assessed after beginning to undertake educational activities.

• 96% of prisoners in cohort 1 had an OASys layer 3 assessment matched to their record. This very high level of coverage reflects the fact that the cohort is restricted to offenders who have both served a custodial sentence of at least 3 months and been subsequently supervised by probation services: a group that would be expected to be assessed.

3.3.1 Criminogenic needs

Criminogenic needs are dynamic risk factors that are directly linked to criminal behaviour. There are eight criminogenic needs linked to offending behaviour which are assessed using OASys: Accommodation, Employment, Relationships, Lifestyle & associates, Drug misuse, Alcohol misuse, Thinking & behaviour, and Attitudes.

Those needs are determined using a system of scored questions. Each question is scored on a scale from 0-2 (some being binary scored 0 or 2, and others 0, 1 or 2). 0 denotes ‘no need’, 1 denotes ‘some need’ and 2 denotes ‘significant need’. The scores within each set of questions are summed, and a need is identified if the score equals or exceeds a threshold or cut-off value for that set of questions (see Table 2).

Table 2: Criminogenic needs and relevant scored questions in OASys

Criminogenic needs	Scored questions	Scale range	Cut-off
Accommodation	3.3, 3.4, 3.5, 3.6	0-8	2+
Employment	4.2, 4.3, 4.4, 4.5	0-8	3+
Relationships	6.1, 6.3, 6.6	0-6	2+
Lifestyle & associates	7.2, 7.3, 7.5	0-6	2+
Drug misuse	8.4, 8.5, 8.6, 8.8, 8.9	0-10	2+
Alcohol misuse	9.1, 9.2, 9.3, 9.5	0-8	4+
Thinking & behaviour	11.5, 11.6, 11.7, 11.9	0-8	4+
Attitudes	12.1, 12.4, 12.5, 12.8	0-8	2+

3.3.2 Learning disability and challenges screening tool

Alongside criminogenic needs, additional ‘responsivity measures’ are used to aid practitioners in how an offender may respond to support. A learning screening tool combines responses to 7 questions from across the OASys assessment to identify those who may be more likely to have a learning disability and/or learning challenges and may benefit from support or reasonable adjustments, as shown in table 3. Offenders scoring above the cut off are considered for referral to receive specialised assessment and support.

A learning disability is an impairment of intellectual abilities (often signified by an IQ of less than 70) together with significant problems with social and practical, everyday functioning, both of which are present before the age of 18. People with learning challenges may not meet the formal diagnosis of a learning disability, but will still struggle with some cognitive, social and practical skills.

Table 3: LDC tool and scored questions in OASys

Responsivity need	Scored questions	Scale range	Cut-off
LDC tool	3.3*, 4.4, 4.7, 4.8, 4.9	0-11	3+ / 5+

*question 3.3 asks whether the offender has no fixed abode and ‘Yes’ (no fixed abode) is scored as 1.

 The four responses shown as outcomes from the tool are:

1. This individual is not likely to have a learning disability and/or learning challenges. (score of 0 to 2)

2. This individual may have some learning challenges. Further assessment may be needed to determine the support required. (score of 3 or 4)

3. This individual may have a learning disability and/or learning challenges. Further assessment may be needed to determine the support required. Consideration for referral for specialised assessment should be given, if appropriate. (score of >=5)

1. Not enough items have been scored to identify whether this individual has a learning disability and/or learning challenges. (fewer than 4 relevant questions answered)

This is not a diagnostic tool, and it is not expected that it perfectly identifies the group of prisoners who would be clinically considered to have a learning disability. In development the cut-off scores chosen were designed to balance the numbers of false positives and false negatives identified so that it can be used to target further assessment and intervention. HMPPS has published research behind the development of the tool.^{[footnote 8]}

3.3.3 OASys education questions

Section 4 of the layer 3 assessment collects information on offenders’ education, training and employability needs. The questions included in this report are shown in Table 4:

Table 4: OASys layer 3 questions

Question number	Question	Responses
question_4_3	Indicates if the offender has any historical issues with employment	Significant problems, some problems, no problems
question_4_4	Indicates if the offender has any issues with work related skills	Significant problems, some problems, no problems
Question_4_5	Indicates if the offender has any attitude issues with regards to employment	Significant problems, some problems, no problems
question_4_6	Indicates if the offender has any issues with school attendance	Significant problems, some problems, no problems
question_4_7	Indicates if the offender has any issues with reading, writing or numeracy	Significant problems, some problems, no problems
question_4_7_1_numeracy	Indicates if the offender has any issues with numeracy	checkbox
question_4_7_1_reading	Indicates if the offender has any issues with reading	checkbox
question_4_7_1_writing	Indicates if the offender has any issues with writing	checkbox
question_4_8	Indicates if the offender has any learning difficulties	Significant problems, some problems, no problems
question_4_9	Indicates if the offender has any educational or formal professional / vocational qualifications	Any qualifications, No qualifications
question_4_10	Indicates if the offender has any attitude issues with regards to education / training	Significant problems, some problems, no problems
question_4_96	Indicates if there are any education / training / employability issues linked to risk of serious harm, risks to the individual & other risks	Yes / No
question_4_98	Indicates if there are any education / training / employability issues linked to offending behaviour	Yes / No

3.4 National Pupil Database (NPD) indicators

The MOJ-DfE data share includes a range of indicators calculated by DfE from across NPD datasets, including:

• the annual school census
• key stage 4 data (GCSE results)
• absence and exclusions data
• children looked after (in care)

The prisoners included have been matched to at least one NPD dataset between the academic years 2001/02 and 2019/20. NPD indicators change frequently, especially those associated with pupil attainment, exam results and school performance indicators. The indicators selected are restricted to those that are available or can be calculated across all years. In most cases measures reflect whether a learner was ever recorded as meeting a definition, across any year, regardless of the number of years in which the learner appeared in the NPD.

Offenders are present in the relevant NPD datasets for varying numbers of years, and this has not been controlled for in calculating the indicators. Prisoners matched to fewer years of NPD data therefore have a lower chance of being counted as meeting the definition.

Older prisoners (who were already partway through their schooling in 2001/02) appear in fewer years, as do individuals who only spent some of their school years in state-funded schools in England. These groups may have met the definition in years not recorded.

Persistent and severe absence

Whether the individual ever missed more than 10% (persistent) or 50% (severe) of available sessions in a single academic year that they were recorded on the school census.

Ever permanently excluded

Whether the individual was ever recorded as permanently excluded (expelled) from a state-funded school in England.

Ever eligible for free school meals (FSM)

Whether the individual was recorded as eligible for free school meals in any year they were recorded on the school census.

This means both that free school meals were applied for on their behalf, and that they met the applicable criteria in that year through household receipt of certain benefits (such as income-based Jobseeker’s Allowance, Child Tax Credit or low-income element of Universal Credit).^{[footnote 9]} This does not include eligibility for universal infant free school meals.

Ever resident in 20% most deprived areas (IDACI)

Whether the individual was recorded in any school census as usually resident in an area ranked in the top 20% most deprived areas in England using the Income Deprivation Affecting Children Index (IDACI). IDACI is a supplementary index created as part of the English indices of deprivation.^{[footnote 10]} These measure relative deprivation in small areas in England called lower-layer super output areas (LSOAs).

The Income Deprivation Affecting Children Index is the proportion of all children aged 0 to 15 in that small area who are living in income-deprived families that receive certain benefits (a wider range of working household benefits are considered compared with free school meal eligibility). This captures children who may not themselves be living in a low-income family but are growing up in an area with a concentration of low-income households.

IDACI scores and ranks for LSOAs were calculated using:

• 2001 LSOA boundaries (32,844 areas) from 2002 to 2015
• 2011 LSOA boundaries (32,844 areas) from 2016

IDACI 2015 rankings were used from 2016 to 2019 and IDACI 2019 for 2020.

Each LSOA contains similar numbers of households and people so around 20% of the population live in the most deprived 20% of LSOAs. Department for Education data show that 23% of key stage 4 pupils lived in these areas in 2023/24.^{[footnote 11]}

Ever a child looked after

Whether the individual has ever been recorded as looked after by the local authority (in care) while aged 4 to 15.

This information is only recorded from 2005/06 and so periods in care prior to 2005/06 are not counted. Individuals not recorded on a school census from this date are excluded, however, as with other measures, the number of available years of data varies for each individual. Prison leavers who were older in 2005/06 are less likely to have periods in care recorded.

A child is looked after by a local authority if they are provided with accommodation for a continuous period of more than 24 hours; are subject to a care order or are subject to a placement order. It excludes children looked after under an agreed series of short placements and also those who were on remand/committed for trial or sentence and accommodated by the local authority.

Ever recorded as having Special Educational Needs (SEN)

Whether the individual has ever been recorded on a school census as having a Special Educational Need (SEN) status^{[footnote 12]} of any of:

• Statement of SEN (pre-2014)
• Education, health and care (EHC) plan (from 2014)
• School Action or School Action Plus (pre-2014)
• SEN support (from 2014)
• SEN recorded at stages 1-5 of the 1994 Code of Practice on the Identification and Assessment of Special Educational Needs^{[footnote 13]} (in 2001/02 only. A new SEN Code of Practice replaced this in January 2002.)

A second indicator identifies whether a Statement of SEN or an Education Health and Care (EHC) plan has ever been recorded. This indicates a more official recognition of needs and plan for support. (SEN stage 5 in 2001/02 is included.)

Key stage 4 achievement (at end of year 11 or age 16)

Whether the individual was recorded as achieving 5 or more GCSEs or equivalents at grade C or above (or grade 4 or above) at the end of key stage 4 (end of year 11, usually aged 16).

Passes at these grades are considered to be level 2 qualifications, and 5 GCSEs is sometimes considered a ‘full level 2 qualification’. This does not necessarily include a pass in English or maths at these grades. Almost all of the cohort would have entered old-style GCSEs graded A* to G (or equivalent qualifications included in those years).

Whether the individual was recorded as achieving any passes in GCSEs or other qualifications at the end of key stage 4. Passes at any grade or level are included.

Only 85% of cohort 2 were ever included in national key stage 4 results, and the two key stage 4 measures only includes these pupils in the comparison. Those living in England but not registered with a school (for example due to exclusion) are excluded, as well as those not resident in England.

Table 5: latest year of key stage 4 results

Academic year	Number of individuals	Percentage of individuals
2001/2002	541	3%
2002/2003	869	5%
2003/2004	925	5%
2004/2005	989	5%
2005/2006	1,051	6%
2006/2007	1,164	6%
2007/2008	1,110	6%
2008/2009	1,127	6%
2009/2010	1,305	7%
2010/2011	1,219	7%
2011/2012	1,178	6%
2012/2013	1,234	7%
2013/2014	1,101	6%
2014/2015	1,027	5%
2015/2016	840	4%
2016/2017	516	3%
2017/2018 or later	171	1%
Not included in any key stage 4 results	2,358	13%

Achieving two or more A levels when aged 16 to 18

Whether the individual was both included in key stage 5 (or 16 to 18) results for a school or college in England and achieved at least two A levels in this period. This is a stretching measure capturing those known to have both remained in education post-16 and to hold common level 3 (advanced) qualifications. Two A levels passes are sometimes considered to be a ‘full level 3 qualification’.

Prison leavers not captured by this measure may still hold equivalent level 3 qualifications or A levels gained through adult further education.

3.5 Data linkage

3.5.1 Linking MOJ sources

Data has been linked deterministically at person level using the NOMIS identifier and its alias on CURIOUS, the Learner Reference Number. As the publication focuses on offenders serving custodial sentences, this identifier exists for this group in each of the MOJ sources used and considered to be of high quality.

In some instances, offenders may receive more than one NOMIS profile on repeat periods in custody, but this is not expected to affect this analysis, which focuses on the latest custodial sentence.

Offenders often have multiple profiles on CURIOUS, and these are linked via their NOMIS ID. Data from OASys assessments were linked on NOMIS ID. Section 3.3. describes the assessment selected.

3.5.2 Linkage in the MOJ-DfE data share

Data linking in the main MOJ-DfE data share was carried out by DfE using a deterministic approach, developing matching rules using common variables between the different sources. Matching rules included combinations of at least an exact match on three of the five variables available (forename, surname, date of birth, gender, postcode, plus an additional derived variable: full name) as well as applying ‘fuzzy matching’ techniques to names. Further matching to CURIOUS was carried out within a secure environment at MOJ using NOMIS ID.

The MOJ-DfE data share comprises data on those released up to and including December 2020 and NPD datasets include people who attended state-funded schools in England from the 2001/02 academic year. Matching to the NPD has therefore only been attempted for offenders between the ages of 10 and 35 as of December 2020 (only adult prison leavers are considered in this publication).

4. Limitations and caveats 

There are limitations and caveats that need to be considered when interpreting the findings in this report, as follows:

Reasons for differences in participation

These statistics describe which groups of prisoners participated in education courses during the period considered but do not attempt to explain the reasons for this. The analysis is descriptive and does not model the impact of characteristics while controlling for other factors.

Both the opportunity and motivation to participate in education affect the findings.

There is a strong element of self-selection bias in who undertakes courses. Prisoners may differ in both their general motivation and interest in education and in their fit to the learning offer available. While these figures provide some context for prisoners’ prior learning, they do not measure current aims or aspirations.

The courses on offer may vary by establishment or over time, and prisons must balance logistical considerations in delivering education to a population with complex needs in a difficult operational context. A very strong factor showing a relationship with participation is sentence length, reflecting the importance of the practical opportunity to enrol and attend classes. Other outside factors may limit certain prisoners’ ability to participate such as health needs or scheduling conflicts (for example with requirements to attend alcohol treatment or accredited programmes).

Coverage of prison population and establishments

These statistics do not cover all prisoners, or all prison establishments. The cohort was chosen to understand education participation throughout an entire custodial sentence and to make fair comparisons about levels of participation between different prisoner groups.

Prisoners serving very short sentences of less than 3 months have been excluded to avoid disproportionate influence on the statistics for prisoner groups more likely to receive such sentences. Only 23% of prison leavers completing sentences of less than 3 months participated in education.

Prisoners serving sentences in Welsh or private prisons, and Youth establishments, have been excluded because it is not mandatory for those institutions to record information about education courses on the CURIOUS system and data may be unavailable or less complete. Education delivery and patterns of participation in these establishments may differ, as education policy is devolved in Wales and specific policies are in place regarding the education of offenders aged under 18.

For cohort 2 only prisoners aged 18 to 35 were including in the matching exercise, and only those who had attended school in England will have been successfully matched, meaning that the cohort is not representative of the full prison population and patterns may not reflect those of the wider prison community.

Coverage of prison education activities

Not all education activities in prisons are covered in these statistics, because of the data source used (CURIOUS, the contract management system for PEF). Prison leavers who have only taken part in activities outside of this scope may be counted as not having participated in education.

Courses undertaken prior to the introduction of CURIOUS in 2019 are not included. This will affect the completeness of recording for those in prison before this date. Cohort 1 were released from prison during the calendar year 2024 so those serving longer sentences of 4 or more years are affected. Only 6% were convicted prior to 2020. The impact is much larger on cohort 2, of whom all were convicted prior to 2020.

Prisons also provide a more flexible local offer through a Dynamic Purchasing System (DPS) which allows prison governors to commission a wider range of courses and training. The statistics also do not cover wider education-related activity within prisons in England.

Timeframe of datasets and Covid-19 pandemic

Covid caused widespread disruption in the delivery of prison education in the years 2020 to 2022. These statistics do not attempt to quantify the impact covid had on prison education, but the impact was substantial. At points during these years a single instance of covid infection could lead to the lockdown of an entire prison wing, preventing access to education classes. While some prisons had the capacity for prisoners to continue education within their cells while on lockdown, this would have affected learning, engagement and achievement. Education inductions were also unable to go ahead as intended under these conditions.

Cohort 1 were released from prison during the calendar year 2024 so it is mainly those serving longer sentences of 2 or more years who would have been affected by covid restrictions while in prison. The impact is much larger on cohort 2, of whom a large proportion would have been completing education courses during 2020.

There is also a much longer time lag than usual between the collection of data and the release of the statistics from the MOJ-DfE data share.

Regions and prison categories

This analysis looks at prison education nationally for England, for all public adult prisons under the PEF. It does not provide insight into differences in provision and delivery of prison education by geographical region or prison category. Some prisons are more focused on preparing prisoners for release into the community and have a greater specialism in education. The courses on offer, as well as the availability and quality of teaching, may vary between establishments. Prison governors may vary the curriculum offer depending on local and regional employment opportunities.

Men and women are also held in separate establishments and differences in course offer and delivery between these setting may contribute to differences by gender.

Interpreting findings on prisoner cohorts

Where this analysis discusses the make-up of the prison leaver cohort in terms of their characteristics and education background it is important to note that, although a large percentage of those released from prison might belong to a certain group or share certain experiences, only a small minority of the population ever serve a custodial sentence, and most people with that experience or characteristic will not have spent any time in prison. For example, although more than half of the prison leaver cohort matched to DfE data were eligible for free school meals (FSM), the majority of FSM recipients have never received a custodial sentence.^{[footnote 14]}

The current statistics only report descriptive findings about the educational needs and experiences of prison leavers and their education while in prison. Many prisoners have complex and intersecting needs which are not fully explored here. This does not imply that these factors are causally related either to receiving a custodial sentence, or to subsequent participation in prison education programmes.

5. Data governance

The BOLD programme has established procedures for the effective governance of data it uses across all pilot projects. You can find BOLD’s Privacy Notice on GOV.UK.

5.1.1 Governance

Analysis and research using data collected in operational systems across HMPPS is covered by privacy notices and the MOJ Personal Information Charter, as well as internal Data Protection Impact Assessments (DPIA). This work is covered under the research and analysis purpose.

A DPIA has been carried out to consider the privacy implications of the linkage of CURIOUS data to other sources and their use in this project.

A Data Sharing Agreement between Department for Education and MOJ governs the use of data in the MoJ-DfE data share. In line with this agreement, MOJ acts as the controller for the matched data in its possession.

5.1.2 Confidentiality

This statement sets out the arrangements in place for protecting persons’ confidential data when statistics are published or otherwise released into the public domain. The Code of Practice for Statistics states that:  

“Organisations should look after people’s information securely and manage data in ways that are consistent with relevant legislation and serve the public good.” ^{[footnote 15]}(https://www.gov.uk/government/statistics/estimates-of-children-with-a-parent-in-prison/technical-guide-estimates-of-children-with-a-parent-in-prison#fn:10) 

To comply with this and with the Data Protection Act of 2018 and to maintain the trust and co-operation of those who use these Official Statistics in Development, the following provisions have been put in place:  

• Private information collected by MOJ is stored in line with our data security policies.
• Electronic data is held on password-protected networks.
• All new staff undergo security vetting before receiving access to data systems and all staff undertake mandatory training on information responsibility annually. 

Some counts may have been removed for Statistical Disclosure Control purposes. In line with MOJ and GSS guidance, assessment of the risk of disclosure considers the following: 

• Level of aggregation (including geographic level) of the data; 
• Size of the population; 
• Likelihood of an attempt to identify; and  
• Consequences of disclosure.

Statistics provide aggregate information for the purposes of understanding offenders and prison education; analysis is not used to look at information about individual prisoners’ education or employment pathways or make decisions that affect specific individuals. Identifying information about individuals has only been used in data linkage steps and access has been restricted to those needing access for this purpose.

5.1.3 Engaging the public

Public trust around how data is shared is critical for BOLD, and we partnered with the Centre for Data Ethics & Innovation (CDEI), and the research company Britain Thinks, to undertake extensive engagement with affected groups, trusted intermediaries, and the general public. The results of this exercise, and what we have learnt from listening to the public, have tangibly informed the design of the BOLD programme and has been published by the CDEI.

6. Dimensions of statistical quality

6.1 Statistical quality at Ministry of Justice

The MOJ aims to provide a high quality and transparent statistical service covering the whole of the justice system to promote understanding and trust.

Producing ad hoc statistics is an important part of ensuring that our statistics provide relevant information for users and cover areas of the justice system which are not yet regularly reported on.

Statistical quality is defined as meeting users’ needs with particular reference to the relevance, accuracy, timeliness, accessibility, comparability and coherence of the statistics collected, analysed and reported.

This framework is based around European Statistical Systems (ESS) Dimensions of Quality and forms part of the Code of Practice for Official Statistics and the Government Statistics Services Quality Strategy.

Ministry of Justice aim to deliver a service in line with our four core objectives:

• Provision of data which are accessible, consistent and fully documented.
• Production of statistics which clearly communicate the story and meet users’ needs.
• Provision of analysis which is timely and based on robust methodology.
• Building capacity, capability and engagement.

You can find out more in our statistical quality strategy on gov.uk

At this stage we would welcome the input of statistics users and topic experts from the sector on the value of these statistics and their further development.

6.2 Dimensions of statistical quality

6.2.1 Relevance

We encourage feedback from users on whether the topics covered in these statistics, the coverage, and methodology used, meet their needs.

We recognise that coverage and timing of data are two key factors limiting the relevance of these statistics.

• We are not able to provide a holistic picture of education activity within prisons, and the offenders reported as ‘not participating’ may be engaging in other forms of education provided through other funding routes such as tailored local training courses, literacy support, or distance learning.

• Data on participation in PEF courses only provide course starts and completions. This provides a limited understanding of engagement in learning, or achievement and progress, that might contribute to later outcomes. We cannot tell whether learners attend classes regularly, or improve their skills and knowledge, separately from success in a recognised qualification.

• The statistics do not include all prisoners, or all courses delivered in a particular timeframe, due to the definition of the cohort (prioritising identification of a group for whom consistent and complete information is available).

• The data cover a limited timeframe, and their availability overlaps with the interruption of prison education due to Covid-19 disruption. They therefore include an atypical period of course delivery.

6.2.2 Accuracy

We have worked closely with policy, operational and commercial colleagues to understand how the data is collected and used. As the prison education data is used to monitor commercial contracts, we are confident that the recording quality is high for the prisons and courses funded through the PEF. Further information on the data sources used is provided in this guide.

A quality assurance plan has been followed to peer review the sources used, code development and appropriateness of analysis.

6.2.3 Timeliness

Reliable data pipelines provide regularly updated data on core prison and probation sources, while two one-off snapshots of CURIOUS prison education data have been used in this report. The data used in combination with MOJ sources covers to the end of calendar year 2024.

Information from the MOJ-DfE data share follows a more complicated series of processing steps. The share itself is updated approximately annually, but there is a lag between recording information, agreeing what will be included, and sharing and matching data between departments. There are then further steps to request and match additional information (in this case an extract of CURIOUS) to the data for further analysis by MOJ teams for a specific purpose.

As this is the first use of this matched data the extract was initially prepared for internal analysis. There is therefore a longer than usual lag between preparation of the linked data and publication, meaning there is a gap of 5 years between offenders’ release from prison and this publication.

We anticipate that this lag could be reduced considerably in future by including the matching of CURIOUS in the initial data share and if publication were on a routine basis, however, it is still likely to be at least 18 months after offenders’ release from prison due to the governance and processing steps required.

6.2.4 Accessibility

These statistics have been released on gov.uk to provide open and transparent access to new areas of analysis. We have aimed to make the commentary clear and informative, and include guidance on the data sources, coverage, data limitations and other relevant information to enable users to interpret and the statistics and assess the relevance of the data to their needs.

6.2.5 Comparability and coherence

Information is presented on two different cohorts of offenders which contain different subsets of the prison population and cover different time periods. This is due to more limited availability of Department for Education data.

Where possible definitions and terminology (such as participation and achievement in prison education) have been aligned with existing MOJ Official Statistics, particularly Prison Education and Accredited Programme Statistics.

It is not intended that reported figures can be directly compared, primarily due to differences in coverage and timing:

• Official statistics report on all prison learners and courses within a given year; these statistics report on courses recorded at any point for offenders released from prison in the period of interest.
• Official statistics only report on offenders recorded on CURIOUS; these statistics contextualise with information about the wider (released from) prison population.
• Official statistics use information about offender characteristics (such as age, learning difficulties and disabilities (LLD)) as recorded on CURIOUS; these statistics draw on other sources such as probation (nDelius) and the national pupil database (NPD).

Further information on quality and methodology for the Official Statistics publication, including history and revisions to data, is available in the Technical guide accompanying the series on gov.uk.

Releases from prison are reported quarterly in the Offender management statistics quarterly publication:

• Official statistics report on releases within a given quarter and include each release from prison within the time period; these statistics report on a longer period of interest and report only the latest release for each prisoner.
• Official statistics on releases include all prison leavers who have finished serving their custodial sentence in England and Wales; these statistics exclude offenders who spent time in establishments in Wales or private prisons, as well as those not supervised by the probation service on release.

Other minor differences may arise from separate processing and validation of the data sources used.

Published findings from the MOJ-DfE data-share have so far generally focused on education, children’s social care and offending and comparisons between young offenders and others who attended school in the same timeframe. It is expected that the education background of prison leavers will differ as they are a subset of mainly more serious or repeat offenders and include a wider age range.

Office for National Statistics (ONS) published statistics on The education and social care background of young people who interact with the criminal justice system in 2022.

7. Users

The contents of this report will be of interest to government policy makers, the agencies responsible for offender management at both national and local levels, prison governors, providers, practitioners, as well as others who want to understand more on participation in prison education.

Government policymakers may also use these statistics to inform key elements of government policies. Prison governors, prison education providers and offender management agencies more broadly may use these statistics to gain a better understanding of the needs of offenders who engage in prison education. Key agencies include: HMPPS, private and voluntary sector providers of prison and probation services.

8. Contact details and feedback

These statistics have been produced by the Ministry of Justice: Better Outcomes through Linked Data (BOLD) - GOV.UK (www.gov.uk).

Any enquiries and feedback on these statistics can be sent to MoJ at RR-pilot-BOLD@justice.gov.uk.

---

1. What do adult literacy levels mean? | National Literacy Trust ↩

2. Apprenticeships for prisoners to cut crime - GOV.UK ↩

3. Release on temporary licence policy framework ↩

4. Our work in prisons - Prisoners’ Education Trust ↩

5. Our work in prisons ↩ ↩²

6. Charity – The Clink Charity ↩

7. Social Impact – Redemption Roasters ↩

8. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/740303/development-screen-identify-individuals-oasys-report.pdf ↩

9. Apply for free school meals - GOV.UK ↩

10. English indices of deprivation - GOV.UK ↩

11. IDACI decile and degree of rurality of pupil residence data, Data set from Key stage 4 performance - Explore education statistics - GOV.UK ↩

12. Children with special educational needs and disabilities (SEND): Overview - GOV.UK ↩

13. [ARCHIVED CONTENT] Special Educational Needs In England: January 2002 ↩

14. A previous Office for National Statistics report on The education and social care background of young people who interact with the criminal justice system provides contextual information about the proportion of young people with characteristics recorded at school who went on to enter prison custody. ↩

15. https://code.statisticsauthority.gov.uk/the-code/trustworthiness/t6-data-governance/ ↩