Annexes

Annex A: Supporting data quality improvements in England and Wales

Figure A1 sets out our understanding of the policing crime data landscape. It captures the bodies and groups whose remit covers crime data quality, including:

• Home Office-coordinated governance groups responsible for ensuring consistency in crime recording in England and Wales (explained in the main report).
• National policing coordination bodies and groups that aim to improve the quality of policing data to meet the operational needs of police forces.
• Police force-led coordination groups.

The relationships between the different bodies and groups are complex, and in general, there is little information in the public domain about them. Note that the figure does not include all bodies and groups whose remit covers data.

Figure A1. Governance groups and policing coordination bodies and groups that focus on crime data

National policing coordination bodies and groups

There are several national policing coordination bodies and groups that are driving improvements to data quality to meet the operational data needs of police forces. Home Office and ONS analysts are not involved in this improvement work.

As these groups focus on operational data quality, their work is likely to have limited direct impact on the quality of police recorded crime statistics. However, it complements improvements to crime recording and recorded crime data quality.

The National Police Data and Analytics Board (NPDAB) was set up by the National Police Chiefs’ Council (NPCC) to provide coordination, collaboration and communication on data across policing in England and Wales. It reports to the NPCC’s Digital, Data and Technology Committee.

NPDAB has five workstreams. The most relevant streams for data quality are the Data Foundations stream and the Centre for Data and Analytics in Policing.

The Data Foundations stream aims to improve data quality, data definitions and flows, data ownership and data standards across policing. The stream has been reviewing the current data quality management practices in a subset of police forces to identify gaps and propose steps for enhancing data quality and data governance. It found that data quality is a priority for most forces, that forces had many similar issues and that forces are keen to share and adopt best practice. The stream is making a number of recommendations to address existing problems and improve the consistency of data quality management across forces, including adopting a national data quality framework, maintaining a national data quality register and developing a model for data quality targets.

The Centre for Data and Analytics in Policing (CDAP) aims to improve the quality and use of data and develop national analytical capability in policing. It commissions and coordinates work across the other four streams. CDAP recently carried out a review of the data analytics landscape in policing to understand which platforms and analytics capabilities exist across police forces and the current challenges to data quality.

The Police Digital Service (PDS), a partner body funded by Home Office and police forces, is responsible for coordinating, developing, delivering and managing digital services and solutions in policing across England and Wales. It was set up to deliver the National Policing Digital Strategy. PDS staff is involved in both the Data Foundations stream work and CDAP.

PDS, on behalf of the NPDAB, is currently undertaking a programme of data maturity assessments across police forces, which complements the work carried out by the Data Foundations stream. The assessment asks officers and staff at different levels in the police force, ranging from chief officers to analysts, to rate the maturity of their force’s data. This information is then fed back to the forces, with advice and recommendations for the forces about where they should focus improvements.

Police force coordination groups

There are also a couple of coordination groups run by police forces which focus on improving the consistency of crime recording and data quality.

The regional force crime registrar groups, attended by force crime registrars from geographically co-located forces, act as forums for discussing and agreeing on the interpretation of the Counting Rules. These regional groups ensure consistency in how the rules are applied, which leads to more-consistent recorded crime data.

The All Forces Performance Group is a forum for performance analysts and officers from police forces across the UK. It is used to share good practice in managing data, improving data quality and producing analysis.

The Knowledge Hub is a platform created and managed by the Police Digital Service. It enables police forces and other national bodies and partner agencies to share information, discuss ideas and support closer working between forces. This knowledge sharing and collaboration allows forces to see what works well and how they may improve the quality of their data.

Annex B: Wider initiatives for data quality improvement

Our report focuses on the quality of the data about crimes. We chose this as our focus because the police recorded crime statistics published by ONS and the Home Office mainly consist of breakdowns of the number of crimes by the type of offence and comparisons of the number of crimes over time.

When a crime is reported to the police, the police record a wealth of information about the incident. This includes the name, address and demographic characteristics (including age, sex and ethnicity) of the victim and suspect; where the crime occurred; and any objects involved in the crime. The College of Policing’s Authorised Professional Practice on the Management of Police Information advises forces to use the people, objects, locations and events (POLE) standards to categorise information.

The collection of good-quality data about the nature and the circumstances of the crime is equally important for effective and efficient policing. For example, police forces must have the correct contact details for a victim if they need to follow up with them. Analysis of these data also helps police forces understand patterns of crime. For instance, it allows them to explore differences in crime by age and sex and to determine geographical areas where crime commonly occurs (crime ‘hotspots’).

A range of work aimed at improving data quality is happening within police forces and nationally, focusing on data about the nature and the circumstances of the crimes. We give a few examples below.

Reducing duplication of person records

Several police forces told us about data quality projects that aim to reduce the duplication of person records (‘nominals’). Duplication can be extensive and cause operational problems for forces. This issue is not unique to one police crime recording system; forces using different systems experience this issue.

The aim of these projects is to create a ‘golden nominal’ – a single record for every person. The work is often manual and labour-intensive, but forces told us it has been effective in reducing duplication and therefore improving the quality of person record data.

Improving the completeness and consistency of protected characteristics data

As part of all inspections that cover crime data integrity, HMICFRS assesses whether police forces have collected and recorded diversity information (protected characteristics) about victims of crime. It makes recommendations to improve experiences in support of Public Sector Equality Duty.

Inspection reports regularly highlight gaps in police forces’ protected characteristics data. HMICFRS’s thematic inspections have also commented on this aspect of the data. For instance, both the 2018 police response to hate crime inspection and the 2021 police response to violence against women and girls inspection found that forces frequently fail to record the ethnicity of victims. The Operation Soteria Year One report similarly found gaps in police forces’ data on victim and suspect sex and ethnicity in rape cases.

HMICFRS has called on forces to record protected characteristics information more accurately and consistently to support understanding of the scale and nature of crime and to ensure that forces provide a better service to victims.

In 2021, the Home Office added a requirement for police forces to provide the Home Office with data on the ethnicity of victims of racially or religiously aggravated offences. The Home Office now publishes statistics on the ethnicity of victims as part of its hate crime official statistics, published annually. Although there are issues with the completeness of the data, this requirement is providing insight on racial disparities in this type of crime.

In 2023, the Home Office added a new voluntary requirement to its Annual Data Requirement on cross-cutting demographic data. This requirement asks police forces to capture demographic data in a consistent way by aligning them with ONS’s 2021 England and Wales Census. This is expected to lead to more-complete and more-consistent protected characteristics data, but we do not know what impact it has had to date.

The National Police Chiefs’ Council (NPCC) has also carried out work to promote more-consistent recording of protected characteristics information across police forces. In 2023, it launched a new protected characteristics operational recording data standard which outlines the set of values that police forces should use to record each protected characteristic.

Monitoring data quality through the Police National Database

The Home Office manages the Police National Database (PND), a national information management system that allows police forces and selected law enforcement agencies to share intelligence and other operational information, including police recorded crime data.

Since 2019, the PND team has been providing police forces with a quarterly assessment of their crime data quality. The team runs a series of over 50 validation and logic checks across a range of quality dimensions, including accuracy, validity, completeness and uniqueness. These checks are then weighted and combined to produce an overall data quality score (out of 100). The PND process is different from the Data Hub’s reconciliation process as it focuses largely on person data and location data.

Forces receive a report that breaks down the results and explains the main data quality issues and how these can be addressed. Common errors include invalid postcodes and incomplete mandatory fields.

It is not compulsory for forces to engage with the reports, but the PND team told us that those forces that are actively using the reports and quality advice are seeing improvements in their data quality score. We encourage all police forces to engage with PND data quality reports to maximise data quality improvements.

Because the PND process focuses mostly on the data about the nature and circumstances of the crimes, it will have a limited direct impact on the quality of the police recorded crime statistics. However, it may still indirectly impact quality, for instance, if a force strengthens its entire quality assurance process as a result of engaging with the PND reports.

Several stakeholders, including police forces and the Home Office, told us that they think it would be helpful if HMICFRS considered the PND data quality reports in their crime data integrity inspections, to give a fuller picture of a force’s crime data quality. We support this idea. We consider that it would have additional benefits: it would encourage police forces to engage with the PND reports and support a more joined-up approach to monitoring crime data quality across police forces, HMICFRS and the Home Office.

Annex C: ONS crime statistics reliability ratings by offence type

Source: ONS’s User guide to crime statistics for England and Wales: March 2023