Methodological note: Why do the two data sources show differing trends?

Over the last few years there have been some apparently conflicting trends between our two main sources on crime. Whilst the Crime Survey for England and Wales (CSEW) has generally shown reductions in crime, the police recorded crime series has continued to rise.

There are a number of possible factors that explain such discrepancy between the two series, including:

• differences in the time periods that each source refers to
• differences in population and offence coverage of each source
• improvements to police recording practices and processes

This methodological note outlines each of these factors and their potential effect on the differences observed.

The police record crimes at the point at which they are reported to or identified by them. Therefore, non-recent crimes may appear in the latest year in the recorded crime series but not in the Crime Survey for England and Wales (CSEW).

In contrast, the CSEW asks respondents about crime experienced in the 12 months immediately prior to interview. Therefore, each survey year actually covers a reference period of two years (see section 2.4 of the User guide for more detail).

In addition to the time period covered by the CSEW lagging a little behind the police figures, being a sample survey, estimates are also subject to sampling error¹. Although one of the strengths of the CSEW is to chart significant changes in long-term trends in crime, its ability to detect short-term and smaller volume changes in crime can be limited.

Notes for: Reference periods

1. For more information on confidence intervals see section 8.1 of the User Guide and the User Guide tables.

The Crime Survey for England and Wales (CSEW) is designed to cover the experiences of crime amongst the population permanently resident in households and therefore will not pick up crime committed against businesses and other organisations, those living in communal establishments, the homeless or visitors from abroad. In contrast, all of these victim groups will be covered by the police figures.

The police recorded crime series also covers a number of offences that are not currently included in the comparable headline estimates from the CSEW. This includes offences such as homicide, sexual offences, shoplifting, harassment, public order, drug and weapon possession offences.

Some of the increase in police recorded crime is among victims and offences not covered by the CSEW.

In January 2013, the UK Statistics Authority withdrew the National Statistics designation from the police recorded crime statistics in the wake of accumulating evidence that the underlying data were not reliable. This was followed in 2014 by a national inspection by Her Majesty’s Inspectorate of Constabulary and Fire and Rescue Services (HMICFRS) of the integrity of police recorded crime data. This found a significant volume of reports of crime from victims were not being recorded in line with the Home Office Counting Rules (HOCR) for recorded crime. HMICFRS found that levels of under-recording were highest for offences involving violence against the person and sexual offences.

The renewed focus on the quality of police recorded crime has led to forces taking action to improve their recording practices. Such changes are thought to have been a significant factor in the volume rise of sexual offences, violence against the person and related public order offences in recent years.

The 2014 HMICFRS inspection report also made recommendations for changes to the HOCR with respect to reports of crime received from professional third parties, such as local authority social services or the medical profession. The existing HOCR required forces to record a crime only when confirmation had been received by the victim. In April 2015, the HOCR were changed so that forces were required to record a report from a professional third party as a crime. This is thought to have led to some increases in crimes against vulnerable people such as victims of child abuse, domestic abuse and elder abuse.

Many of the victims related to third party reports may not be covered by the Crime Survey for England and Wales (CSEW) because they will be resident in care homes (both children and older people) or victims of domestic abuse who may not be willing to disclose such experiences in the context of a face-to-face interview.

Evidence that such improvements in recording practices and processes has had an effect on crime figures can be found by examining trends for a comparable sub-set of crimes included in both sources. Between the year ending March 2005 and the year ending March 2012, this showed a growing divergence between the two series, suggesting a gradual erosion of police compliance with the HOCR over time. Over the last four years, we have seen a reversal of that trend¹.

It has been questioned whether improvements in crime recording practices can continue to be a plausible explanation of the rise in recorded crime. However, since 2014 there has been ongoing work by forces to address recording. This has been given impetus by the announcement by HMICFRS in 2015 of their intention to embark on a rolling programme of unannounced inspections of forces to ensure all were re-visited.

While we think changes to recording practices and processes explain a large part of the rise in the relatively less-harmful categories of violent crime, such factors are thought to be a less-significant driver behind recent increases in more-harmful violent crimes such as homicide and those involving knives and guns. Improved recording is also not thought to be a major influence behind increases in burglary and vehicle theft, as these were offences that HMICFRS generally found to have better levels of recording.

This raises the question as to why increases in burglary have not been reflected in the survey estimates. As outlined previously, as a sample survey the CSEW can be limited in its ability to detect shorter-term trends, but if the volume increases for domestic burglary seen in the police recorded crime series continue, we would expect these to be reflected in the survey in due course. We will continue to monitor these trends and investigate the factors driving any changes.

Notes for: Improvements to police recording practices and processes

1. Further details of this analysis are provided in Section 4.1 of the User Guide.