Do a quarter of sexual offences go unrecorded?

"Police officers fail to record a quarter of sexual offences—including rapes—and one-third of violent attacks, a damning report by the police watchdog has concluded." - The Independent

The Independent is not alone in drawing out this statistic from today's HM Inspectorate of Constabulary's report on crime recording in England and Wales. The figure is accurately reported but it's worth bearing in mind that the police can only record crimes that are reported to them.

Ten times as many sexual offences are estimated to have occurred by the Crime Survey as appeared in police records.

The sensitivity of the topic may still lead to under-reporting of sexual offences in the Crime Survey. Either way this is a significant gap and suggests that under-reporting of sexual offences to the police is a problem.

So considerably more than a quarter of all sexual offences will not be recorded by the police in the first place. The headlines don't make clear that although roughly a quarter of all claimed sexual offences aren't recorded, the proportion of rape claims that go unrecorded is around 1 in 10.

How crime recording works

The way claims are currently supposed to be recorded presumes that the person reporting an offence is to be believed until there is evidence to the contrary.

The National Crime Recording Standard instructs officers to record a crime if:

"on the balance of probability: the circumstances as reported amount to a crime defined by law...
and there is no credible evidence to the contrary.

Once recorded, a crime will remain recorded unless there is credible evidence to disprove that it has occurred."

So if the person reporting a crime refuses to then support the police investigation or decides not to pursue an allegation, the crime will still be recorded.

In 60% of the rape claims that were incorrectly left unrecorded by the police, the person making the report had described events that amount to a crime under the law. But in these cases, the police have decided that a crime didn't happen without adequately explaining why. These could be cases where the officer thinks it is highly unlikely that events unfolded as described, but has no evidence to the contrary.

Easy to record, hard to remove?

"Even when crimes are correctly recorded by the police, too many are removed or cancelled as recorded crimes for no good reason, including more than 200 rapes" — The Guardian

A crime is recorded when the officer thinks that the reported event describes a crime and there is no evidence to the contrary. To remove a recorded crime ('no crime-ing' it), the police must prove that no crime took place. The incorrect removal of a recorded crime does not necessarily mean that a claim hasn't been investigated. Equally, the recording of a crime does not necessarily mean that a crime has occurred.

One in five of the 'no crime' decisions on rape were found to be incorrect. While some sexual offence claims that were 'no crimed' were not investigated as fully as they should have been, others were investigated and found to lack evidence sufficient to show that a crime had occurred. However, this is not sufficient for 'no crime-ing' a report. In the absence of conclusive evidence to the contrary, these reports should have remained on the records.

Improvements have been made

Despite the issues found by the Inspectorate's report, the number of sexual crimes recorded by the police has increased significantly over the past two years. This effect has been attributed to increased willingness to report sexual offences to the police, and more consistent recording of complaints.