There continues to be significant public discussion – by both politicians and the media – about crime rates in England and Wales and the reliability of the crime statistics produced by the Office for National Statistics (ONS).
As the UK’s independent statistics regulator, the Office for Statistics Regulation (OSR)’s role is to ensure that official statistics serve the public good. This includes promoting the appropriate use of statistics in public debate and challenging the misuse of statistics where it risks undermining public confidence in statistics.
Recently, some claims have questioned the reliability of estimates from the Crime Survey for England and Wales (CSEW). We have also seen incorrect claims that police recorded crime statistics are more accurate than CSEW estimates because they represent recorded crimes. Given the importance of crime statistics to public understanding and policy debate, OSR has prepared this statement to clarify several common misunderstandings about the CSEW and explain why it remains a vital source for understanding long-term trends in crime.
Misunderstanding 1: National CSEW estimates are unreliable because they do not reflect individuals’ experiences or perceptions of crime
We recognise that individuals’ lived experiences of crime vary widely across England and Wales, and that crime can have profound effects on people and communities. However, the national estimates from the CSEW cannot mirror every individual’s experience of crime. By their nature, aggregate estimates describe trends across a whole population. This is the case for all household surveys, whether they measure crime, health or income. Simply because they differ from personal experiences does not mean that the national CSEW estimates are unreliable or incorrect.
The gap between crime rates and public perceptions of crime is a well-documented phenomenon in criminology: often, when crime rates go down or remain stable, the public’s ‘fear of crime’ increases. The CSEW consistently finds that most respondents believe crime has increased in recent years, even when the survey shows long-term declines in many types of crime. This mismatch is influenced by a range of factors including media reporting, political rhetoric and demographic differences. Again, differing perceptions do not invalidate the CSEW estimates.
Misunderstanding 2: The CSEW sample is not representative
Claims that the CSEW undersamples cities and oversamples rural areas are incorrect. ONS samples randomly selected households that are representative of England and Wales as a whole. This is recommended practice in household surveys. Because the population is larger in urban areas, more households in those areas are selected. In some cases, the CSEW sample is even slightly boosted in higher-crime areas to improve the precision of the estimates.
To account for differences in response rates and potential non-response bias across different geographical areas, ONS applies a standard statistical method called weighting. Weighting ensures that the final survey estimates reflect the true population structure and are representative. These well-established methods underpin confidence in the CSEW estimates.
Misunderstanding 3: Police recorded crime figures are inherently more accurate than CSEW estimates
Police recorded crime and the CSEW measure different aspects of crime. Police recorded crime figures reflect crimes that are reported to and recorded by the police. They are affected by changes in police recording practices, policing priorities and levels of crime reporting.
In contrast, the CSEW captures people’s experiences of crime whether or not those incidents were reported to the police. As a result, for the crime types it covers, the CSEW is the best source for measuring long-term trends in crime.
Using the most appropriate data source
There are certain crime types for which the CSEW is not the preferred data source. These include lower-volume, higher-harm crimes that are not captured well by the survey such as knife-enabled crime and offences involving firearms, and crimes which are not covered by the survey such as homicide and crimes against businesses. For these crime types, police recorded crime provides the most reliable measure of trends in crime.
ONS has set out its preferred data source for each crime type. Users should draw on both the CSEW and police recorded crime to get the most complete picture of crime in England and Wales.
ONS has examined the relationship between the CSEW estimates and police recorded crime figures.
Known limitations versus weaknesses
All data sources have limitations, and transparency about these limitations is essential. The limitations of the CSEW – such as the exclusion of crimes against businesses – are understood and well-documented by ONS.
It is important to distinguish between limitations, which are natural boundaries of the survey, and weaknesses, which are deficiencies or errors that undermine the quality of the estimates. Many criticisms of the CSEW that we have seen conflate these two terms by framing known limitations as weaknesses.
In 2024, we reinstated the accredited official statistics status of the CSEW estimates. Our judgement, informed by ONS’s data quality review, was that the quality of the current estimates is broadly similar to the pre-pandemic estimates. In other words, we have confidence in the quality of the CSEW estimates.
Conclusion
Taken together, these points emphasise the importance of using crime statistics carefully and responsibly. Public debate benefits from scrutiny, but such scrutiny must be based on evidence rather than assumptions and misunderstandings. The CSEW continues to be a robust source for understanding long-term trends in crime in England and Wales.
We will continue to challenge the misuse of statistics and uphold the standards to ensure that crime official statistics serve the public good.
Notes for editors
There are two main sources of official statistics on crime in England and Wales: statistics from the Crime Survey for England and Wales (CSEW) and the police recorded crime statistics. These statistics are published quarterly by the Office for National Statistics (ONS) in the Crime in England and Wales statistical bulletins.
Our 2024 review of the CSEW was undertaken following the temporary suspension of the accreditation in July 2022 due to quality concerns, including a drop in the survey response rate.