Crime and Security

Measuring crime is important yet answering questions on the amount of crime and how it is changing is not easy. We have seen a steady improvement in the way statistical producers report what is happening to crime, although there is still more to do. The COVID-19 pandemic has presented a very real challenge to the continued provision of some of the data sources used in crime statistics, and producers are having to think creatively about how they can address the data quality issues that are arising during this time.

Overview

Crime is a devolved matter in Scotland and Northern Ireland. It follows that there are separate statistics about crime published for each of England and Wales, Scotland and Northern Ireland.

In each reporting area, statistics that measure the prevalence, type and impact of crime experienced by individuals, households and society[1] come primarily from a household survey and administrative data from police forces. The table below details the key statistics in each area. Until March 2020, the three household surveys listed were face-to-face surveys: however, these had to pause because of COVID-19. In each reporting area, the face-to-face survey has been replaced with a telephone survey. In August 2020, OSR reviewed the new telephone-operated Crime Survey for England and Wales.  

Police recorded crime in England and Wales

Most of the primary statistics used for reporting on crime experienced by the general public are designated as National Statistics. This means that we have assessed them as meeting the standards of Trustworthiness, Quality and Value set out in the Code of Practice for Statistics. The exception to this are the police recorded crime (PRC) statistics for England and Wales, which lost their designation in 2014 as part of a wider assessment of Office for National Statistics (ONS) crime statistics.

Following a further assessment in 2016, the National Statistic status of statistics about unlawful deaths based on the Homicide Index was restored, but the rest of ONS PRC data remain as official, rather than National Statistics. During 2020, we will be considering whether re-assessment of these statistics would be appropriate in 2021.

Reporting crime statistics

It is not always easy to say what is happening to crime. Crime can be hard to define and difficult to measure: we have previously written a blog that expands on these issues. We have been pleased to see some significant improvements in the way that statistical producers are reporting crime statistics, which is informing better public debate about crime. These include:

• Producers are telling more coherent stories with crime data and statistics. Rather than reporting on individual data sources, producers are more often presenting data from multiple sources side by side, enabling them to give a more insightful picture into crime overall. The most recent Crime in England and Wales statistical bulletins, produced by ONS, are a very good example of this. The ONS has also worked collaboratively with several other official statistics producers to enable the production of new, influential statistics on topics such as child abuse and the criminal justice system and modern slavery in the UK. These releases bring together data sources to tell more-coherent stories about individuals’ journeys through the whole justice system and provide insight into important societal issues, where previously information has been scarce.
• In all three reporting areas, more information about the quality of the data and statistics, including their strengths and limitations, is being made available to users of the statistics. This makes it easier for users to interpret changes in the statistics and to know how they can and cannot be used.
• There has been considerable investment in improving data quality. To give two examples: there has been ongoing work to better understand and to improve the quality of police recorded crime data in England and Wales and in Scotland, and; there has been financial investment in the NI Safe Community Survey, which enabled the survey sample size to increase, making a wider range of analysis possible.

Of course, there is still more that should be done to ensure crime statistics better serve society’s needs. Our recent regulatory work, which includes a public intervention on the misuse of ONS crime statistics during the 2019 General Election, a compliance check of Recorded Crime in Scotland, and a series of compliance checks on the NI Safe Community Survey, has highlighted ways in which all these statistics can be strengthened. We will continue to highlight areas for improvement in future regulatory work, with the overarching aim of improving the trustworthiness, quality and value of crime statistics.

Changes to crime statistics due to COVID-19

Since March 2020, the COVID-19 pandemic has presented a real challenge to the continued provision of some of the data sources used in crime statistics: most obviously, those statistics that previously relied on face-to-face survey data, but the quality of administrative data from police records could also be affected. Statistical producers in the ONS, Scottish Government and the Northern Ireland Statistics and Research Agency are working hard to overcome quality issues that are likely to arise: this includes instigating telephone surveys to replace face-to-face surveys and planning how to clearly present discontinuities in data series. Since there is always a time-lag between data collection and reporting, the effects of COVID-19 on crime statistics are only now starting to become apparent in statistical publications.

In this situation, we expect producers to explain the impact of changes in the circumstances and context of data sources on the statistics. We will continue to communicate regularly with statistical producers to offer guidance and support on how this can best be achieved. Where substantial changes are made to a data collection, or to the content or presentation of crime statistics, we will undertake appropriate regulatory work to ensure the principles of the Code are upheld, and that the National Statistics status remains appropriate where it is currently applied.

[1] This statement discusses statistics about crime experienced by the public. It does not cover crime against businesses.