How to check a claim that crime is going up or down
England and Wales have two official measures of crime. The Crime Survey for England and Wales (CSEW) asks people about crimes they experienced, whether reported or not, and is the ONS's preferred measure of long-term trends. Police recorded crime counts offences recorded by forces and is better for rare, serious crimes such as homicide and knife crime, but is affected by recording changes. To check a trend claim, find out which measure and crime type it uses, and whether the change is statistically significant.
The two measures
The ONS bulletin Crime in England and Wales: year ending March 2026 , published on 23 July 2026, explains both:
The Crime Survey is "an interviewer-administered face-to-face survey" of people aged 16 and over living in households. It "is unaffected by police reporting or recording changes" and its methods "have remained comparable since the CSEW began in 1981". It excludes crimes against businesses, people in institutions, and crimes without a specific victim, such as drug possession.
Police recorded crime covers a wider range of offences and people, but only crimes "reported to and recorded by the police". The ONS says "Changes in recording practices have affected the reliability of these figures, particularly for violent crime. This makes them better indicators of police activity than crime trends."
Police recorded crime lost its accredited status in January 2014, when the Office for Statistics Regulation found "evidence that the quality and consistency of the underlying data may not be reliable". The Crime Survey figures are accredited official statistics. Scotland and Northern Ireland publish their own figures; see crime in Northern Ireland .
Source: Crime in England and Wales: year ending March 2026
Step 1: find which measure the claim uses
Claims of rising crime often use police recorded totals; claims of falling crime often use the survey. Both can cite the same bulletin. In the year ending March 2026:
the police recorded "around 5.2 million crimes (excluding fraud and computer misuse), a 1% decrease compared with last year (5.3 million), but up from 3.6 million in YE March 2015";
the ONS adds that "Increases in volume over the last 10 years have been largely influenced by improvements in recording standards".
So "recorded crime is up by nearly half since 2015" matches the police figures, but the ONS's own reading is that this mainly reflects better recording, not more crime.
Step 2: check significance in the survey
The Crime Survey is a sample, so small changes may be noise. The bulletin flags whether each change is statistically significant. In the year ending March 2026 the survey estimated:
9.6 million incidents of CSEW headline crime, "no statistically significant change" from around 9.4 million a year earlier;
4.5 million fraud incidents, also no significant change, although the number of fraud victims rose by 10% to 3.8 million.
So "crime rose by 200,000 incidents" is arithmetically right but not a change the ONS treats as real. See statistical significance and confidence intervals .
Step 3: match the crime type to the better source
The bulletin says police data "can give more insight into lower-volume, higher-harm offences" such as homicide, weapons offences and robbery. It also names offences where police figures are "not a good indicator of trends over time": violence, domestic abuse, sexual offences and stalking and harassment.
Worked example: four claims about the same year
"Homicides are at their lowest level in over 20 years." The bulletin: 499 homicides in the year ending March 2026, "the lowest figure since current police recording practices began in YE March 2003". The rate was 8.1 per million people. The claim matches, provided "over 20 years" means since that series began.
"Knife crime has fallen." Offences involving knives or sharp instruments "fell by 8% (to 48,774 offences)". The claim matches police recorded data for the year to March 2026. See knife crime .
"Robbery is up 10%." Overall robbery rose 10% to 87,362, "mainly because of improvements to recording practices for robbery of business property", while robbery of personal property fell 9%. The figure is accurate, but the bulletin attributes most of it to recording.
"Anti-social behaviour has soared." The survey found 40.9% experienced or witnessed ASB, up from 35.5%, but the ONS says the rise "was inflated because of changes to how questions in the CSEW were asked from April 2025".
In each case the useful answer adds the measure, the period and the ONS's own caveat.
Step 4: check the period and the next release
Crime statistics are published quarterly, each covering a rolling 12 months. The bulletin above lists its next release as 22 October 2026. A claim made after that date should use the newer figures. Also watch for a claim that compares a 12-month period with a single quarter, or with a calendar year from a different source. For local figures see how to use police.uk crime data .
Checklist
Measure identified (Crime Survey or police recorded crime)
Crime type matched to the better source
Statistical significance checked for survey changes
Recording changes checked for police figures
12-month period and release date noted
Geography checked (England and Wales only)
ONS caveat included alongside the figure
Official sources used in this guide
Sources: ONS: Crime in England and Wales, year ending March 2026, ONS: police force area data tables, OSR: accredited official statistics, data.police.uk: about the data
Questions
Is crime going up or down in England and Wales?
See our answer , which gives the latest ONS figures. The ONS says crime against individuals and households "has generally decreased over the last 10 years with some notable exceptions, such as fraud".
Source: ONS: Crime in England and Wales, year ending March 2026
Why do police figures go up when the survey goes down?
Mainly because of improvements in recording practices, and changes in police activity and public reporting.
Source: ONS: Crime in England and Wales, year ending March 2026
How much crime is reported to the police?
The ONS estimated that 36% of CSEW headline crime incidents (excluding fraud and computer misuse) were reported to the police in the year ending March 2026.
Source: ONS: Crime in England and Wales, year ending March 2026
Does the survey cover children?
The main estimates cover people aged 16 and over living in households.
Source: ONS: Crime in England and Wales, year ending March 2026