Quality

Introduction

Quality means that the statistics and numerical information represent the best available estimate of what they aim to measure at a particular point in time and are not materially misleading.

Quality is analytical in nature and is a product of the professional judgements made in the specification, collection, aggregation, processing, analysis, and dissemination of data.

Findings

Focus of the police funding statistics

There are many different sources of police funding. The statistics specifically concentrate on police funding and not funding contributing to bodies related to policing, which include the National Crime Agency (NCA) and Her Majesty’s Inspectorate of Constabulary and Fire and Rescue Services (HMICFRS). These bodies are mentioned in the statistics because they are partly funded by the police funding grant.

The statistics are clear that the funding they highlight is not the whole budget for policing for either an individual police force area or a body like the NCA. For example, some PCCs receive funding from other sources, such as their financial reserves or charging for special police services, relating to football matches or concerts. The statistics aim to provide a snapshot of police funding as set out each year to the UK parliament as part of the annual police funding process.

Quality and methods documentation

While the contextual documents covered above provide a good level of insight about how the data are collected, the users we spoke to told us they considered it would be better if the information on sources was included in a single document about quality. For example, some users felt that while there is information about quality within the statistical bulletin, police funding is still difficult to understand and having a document covering the strengths and limitations would enhance understanding of the quality of the statistics for all types of user, including non-technical users. During the assessment process Home Office started developing a user guide to accompany the next publication which will be based on expanded versions of the current quality information provided. Home Office should apply our Quality Assurance of Administrative Data framework (QAAD) framework, to guide understanding about the limitations of all stages of the data process and assure itself and users of data quality. The framework will also be useful in helping the team to produce public information about quality.

The quality assurance process adopted by the data team is based on Office for Budget Responsibility best practice methods for publishing data about funding. The team has a strong working relationship with analytical teams in MHCLG, particularly around the use of the latter’s data on local authority contributions to police funding. It is good to see that the collaborative approach taken to assuring the data on police funding is robust. We encourage the team to include details of these relationships and best practice methods in the quality and methods documentation.

A quality and methods document would be a good place to provide greater detail about the funding formula used to calculate the funding grant and alert users about how changes to the formula may affect the statistics in the future and this is to be reflected in the proposed user guide. As highlighted in the Value chapter the reason why the funding statistics are published four months after the start of the financial year could also be covered.

Strong relationships with data suppliers and other collaborators

We found the relationship between the Home Office team that supplies and assures the quality of the data and the statistics team to be positive. For example, the teams have regular catchups about the timeliness of the data and the data supply team is closely involved in the production of the statistics, particularly around assuring the commentary and caveats in the data.

We were pleased to hear that the data team along with the publication team have started an improvement project, making the spreadsheet – currently large – and database more streamlined. The team should work closely with the data team during this project.

We consider that adopting Reproducible Analytical Pipeline (RAP) best practice will make streamlining the data process easier and eliminate mistakes that can often arise when using large spreadsheets. Though we are assured that no such mistakes have happened, implementing RAP will be an excellent way of future-proofing the quality of the statistics. Our recent review, Enhancing the use of Reproducible Analytical Pipelines, provides some useful case studies on how to implement RAP. Adopting all or part of RAP methodology will ensure the continued quality of the data used to produce these statistics.

Summary of Findings and Requirements