Background Quality Report
Last updated: December 2025
Background Quality Assessment
This section provides information about the quality of the data used to produce this publication, and any statistics derived.
Introduction
Context for the quality report
The publication entitled Adult and Youth Reoffending In Northern Ireland is produced annually by the Analytical Services Group (ASG) a branch of statisticians from the Northern Ireland Statistics and Research Agency (NISRA) located within the Department of Justice (DoJ), and has been done so since 2010/11. The reports are available to download from the DoJ website (opens in a new window) and are published in line with the ASG Customer Service and Engagement Statement, Policy 6 in the Code of Practice for Statistics – Policies and Statements PDF (536 KB)(opens in a new window).
This report presents data on the reoffending rates for adults and youths in Northern Ireland along with trend information from 2010/11.
The statistics are produced using administrative data sourced from the Causeway Data Sharing Mechanism, an interconnected information system, launched as a joint undertaking by the Criminal Justice Organisations in Northern Ireland. They were extracted primarily based on records contained on the Criminal Records Viewer (CRV), held on Causeway. CRV utilises data which originated in the Police Service of Northern Ireland (PSNI), Public Prosecution Service (PPS) and from Northern Ireland Courts and Tribunals Service (NICTS).
Information is also provided from the Northern Ireland Prison Service (NIPS) case management system (PRISM) and the Youth Justice Agency (YJA) regarding releases from custody. Information on these offenders is matched to information taken from CRV.
Relevance
The degree to which the statistical product meets user needs in both coverage and content.
The information in the publication is used to inform policy decisions within DoJ, as well as to inform requests from other Government organisations and to answer NI Assembly questions and queries from the general public.
User needs are identified through biennial online customer surveys, the most recent undertaken in February 2025, Customer Surveys (opens in a new window).
Analysis is also provided in relation to reoffending interval, number of proven reoffences, offending history, gender, age, disposal, baseline offence and specified and serious offences. Some comparisons with figures from 2010/11 are also included.
The cohort is made up of all adults and youths who have been given a non-custodial disposal at court, a diversionary disposal or who have been released from custody during the financial year of interest. An adult is defined as someone aged 18 or over at the time of community disposal or release from custody and a youth is defined as anyone aged 17 or under at this point.
The baseline date is the date of entry into the cohort. This is the date that a non-custodial sentence is given at court, a diversionary disposal imposed or an individual is released from custody. This date forms the starting point for the observation period. The baseline offence is the principal offence associated with this baseline date. Following consultation, it was agreed that an offender would be counted once within each relevant disposal category, including the initial baseline disposal associated with their entry into the cohort and also disposal categories associated within any subsequent proven reoffences.
The observation period is the window of time over which a person is observed following receipt of a diversionary disposal or non-custodial disposal at court or release from custody. A balance needs to be struck between the need for timely information and the need to make the reoffending rate as informative and meaningful as possible. For one year reoffending, the observation period is one year and the follow up period is six months.
Under one year proven reoffending methodology, an offence is counted as a reoffence if it -
Occurs within the one year observation period;
Has been committed within Northern Ireland;
Is prosecuted via the PSNI, the NCA, the Airport Constabulary or Harbour Police. Prosecutions brought by government departments, public bodies and private individuals are not included, as their prosecution is beyond the remit of the DoJ;
Is not a breach offence (for example breach of a probation order); and
Has been ‘proven’, meaning that a court conviction or diversionary disposal has been imposed within the observation year or by the end of the 6 month follow up period.
Accuracy and reliability
The proximity between an estimate and the unknown true value.
These statistics are sourced from administrative systems that encompass data originating from PSNI, PPS, NICTS, NIPS and YJA. Staff within each of these organisations, are required to accurately record this information as it is used to manage both day-to-day business and also to communicate important information between organisations. The nature of the information is factual and the scope for subjectivity is minimal. It is not a sample but a complete data set.
There is no means of verifying if information has been entered incorrectly, or not at all. However, as it is the main source of information used to manage day to day business within the organisations and also to communicate important information between them it needs to be highly accurate. Each organisation therefore has its own set of checks and alerts for their own administrative data systems and run regular validation checks to ensure all key pieces are up-to-date and accurate. There is also a degree of logic checking once this information is drawn together.
Numerous validation checks are carried out by a number of parties, including NISRA statisticians, to ensure the data is fit for purpose. These checks have been put in place to ensure complete and accurate linkage between each of the administrative sources used in the production of the reoffending outputs. In particular, an element of manual validation is required to link data from NIPS discharges from custody to DoJ’s prosecutions and convictions information (which incorporates PSNI, PPS and NICTS data). This is undertaken to ensure the custody releases cohort includes only those released from a custodial sentence, and does not include, for example, those released on bail who then go on to receive a non-custodial sentence or no conviction at court. Given the ongoing issues with bail and remand in NIPS, and increasing numbers of individuals released on bail, the number of manual checks has increased substantially year-on-year with over 2,000 checks required for the most recent cohort; work is ongoing to determine ways to reduce this. In relation to potential sources of bias and error, further details can also be found in the QAAD report (opens in a new window).
Users should however bear in mind that the statistics originate from various administrative data sources which have different purposes, aims and objectives and are kept for non statistical purposes, for example reviewing and vetting individuals’ criminal records.
The ability to compare and discuss trends in reoffending is important to its usefulness as a performance target within government. However, differences in the offending related characteristics of those included in each cohort make comparing reoffending rates problematic, across both time and jurisdictions. Prior to the 2017/2018 cohort, reoffending figures were provided alongside adjusted reoffending rates for adults and the overall cohort, as an estimate of change in reoffending. Following consultation with key users, the decision was taken to exclude from future publications to avoid confusion in the interpretation of findings. We will continue to explore statistical techniques that could be employed to control for differences within the cohorts, meanwhile, care should be taken to understand the wider context within which offending and reoffending has occurred.
In addition, reoffending rates should not be used to measure the comparative success of different disposal types; no adjustments have been made to control for offender characteristics or factors relating to variations in sentencing and so such comparisons would be misleading.
Since 2015, there has been a consistent decline in the number of cases prosecuted and also in corresponding guilty findings. This has resulted in a further reduction in the size of the reoffending cohort.
Users should note that some figures may not add to the totals due to rounding.
Timeliness and punctuality
Timeliness refers to the time gap between publication and the reference period. Punctuality refers to the gap between planned and actual publication dates.
The annual datasets cover the financial year (1st April to 31st March). The statisticians producing the report continually look for ways that the processes can be streamlined while at the same time maintaining or improving the accuracy of the data. A balance needs to be struck between the need for timely information and the need to make the reoffending rate as informative and meaningful as possible. For one year reoffending, the observation period (that is the window of time over which a person is observed following receipt of a diversionary disposal or non-custodial disposal at court or release from custody) is one year and the follow up period is six months. This accounts for the gap between the reference date and the publication date. Publication dates for the last five years are provided below.
| Report Year | Publication Date |
|---|---|
| 2022/23 | 16/10/25 |
| 2021/22 | 17/10/24 |
| 2020/21 | 19/10/23 |
| 2019/20 | 10/11/22 |
| 2018/19 | 25/11/21 |
Accessibility and clarity
Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of metadata, illustrations and accompanying advice.
Accessibility to the statistics for this publication are in line with those detailed in the ASG Customer Service and Engagement Statement, Policy 6 in the Code of Practice for Statistics – Policies and Statements PDF (536 KB).
No issues relating to accessibility were identified during a Customer Survey undertaken in February 2025 PDF (946 KB) (opens in a new window). There was one suggestion relating to clarity which is currently being explored.
The report contains contact details for further information and is available to download from the DoJ website, along with reports from previous years.
The data presented in the report are also available in Open Document Spreadsheet (ODS) format on the DoJ website.
Explanatory information including sources, discontinuities and missing data have been included.
Coherence and comparability
Coherence is the degree to which data that are derived from different sources or methods, but refer to the same topic, are similar. Comparability is the degree to which data can be compared over time and domain.
In 2013, ASG embarked on a project to revise the methodology used to calculate recidivism rates within Northern Ireland, bringing it more in line with established methodology in England and Wales.
However, differences in the offending related characteristics of those included within each cohort, such as gender, age and criminal history, make comparing reoffending rates problematic across both time and jurisdictions.
Trade-offs between output and quality components
No trade-offs applied.
Assessment of user needs and perceptions
The process for finding out about users and uses, and their views on the statistical products.
The ASG Customer Service and Engagement Statement details ASGs customer engagement policy and can be found at Policy 6 in the Code of Practice for Statistics – Policies and Statements PDF (536 KB).
An online customer survey was undertaken in February 2025 PDF (946 KB) (opens in a new window). The vast majority of respondents (88%) worked for The Northern Ireland Civil Service or other Agency; the remainder were from academia and other non-NICS public sector employees. The most frequently stated reasons for using the publication were policy/performance/resource related, media/public interest related, and to facilitate academic research.
The vast majority (91%) of respondents said that they were satisfied or very satisfied with the publication overall; the remaining respondents were neutral. Almost all respondents (91%) said that the statistics fully or mostly met their needs; remaining respondents said it partially met their needs. Suggested improvements are documented in the customer survey report PDF (946 KB).
Regular meetings also take place with key users within the DoJ.
Performance, cost and respondent burden
The effectiveness, efficiency and economy of the statistical output.
There is no respondent burden, since the data are held on an administrative system, and data on new cases are automatically collected as part of the Criminal Justice process.
The annual operational cost (staff time) of producing the report is approximately 65 days.
Confidentiality, transparency and security
The procedures and policy used to ensure sound confidentiality, security and transparent practices.
The ASG Confidentiality and Protection Arrangements statement, Policy 2 in ASGs Policies and Statements PDF (536 KB), sets out ASG’s arrangement for maintaining the confidentiality of the statistical data used in this publication.
All staff involved are trained on the protocols for protecting and maintaining the confidentiality of the data. NISRA follows the Privacy and data confidentiality methods: a Data and Analysis Method Review (DAMR) – Government Analysis Function in the collection and dissemination of this report.
Data are held on a network that is only accessible to the few statisticians who need access. Printouts containing individual records or small cell sizes are locked away and shredded as soon as possible.
Standard disclosure control methodology is applied to the data. This ensures that information attributable to an individual is not identifiable in any published outputs and that the outputs are only seen by authorised staff prior to their publication.
The pre-release access list for the report is reviewed on an annual basis. The named individuals are checked to ensure that they are the correct contact and that they are available on the day before the release of the report (if they are not then they can nominate a deputy). A guidance document is also sent to those on the revised list explaining to them their obligations about data disclosure prior to the publication of the report.