Monitoring Review on Overcoming Barriers to Trust in Crime Statistics ( England and Wales )

Note on the open discussion of issues raised in the interim report

An open meeting was held on 18 January 2010 at the Royal Statistical Society.

The interim report and the slides from the meeting are available on the Authority website.

Richard GarsideCentre for Crime and Justice Studies, KCL
Paul QuintonNational Policing Improvement Agency
Richard LauxUK Statistics Authority
David Blunt and Julian CornerHome Office
Iain BellMinistry of Justice
Allan BrimicombeCrime and Justice Statistics Network

The meeting was chaired by Partha Dasgupta, a non-executive member of the UK Statistics Authority. The speakers’ presentations were followed by a questions and answers session which included discussion of the interim recommendations.

Recommendation 1: the National Statistician to publish a regular commentary on trends and patterns in crime

This was generally welcomed by those present, although the resource implications were noted, and the fact that there isn’t always a single or simple story to tell on crime. It was suggested that the National Statistician might commission the work from external specialists. It was also suggested that case studies at a local level might be more instructive than trying to agree a narrative at the national level.

Several participants spoke of a wider problem of distrust in society (reinforced by recent events such as MP expenses) so there was no guarantee that attempts to improve trust in crime statistics would be successful. The National Statistician had convened a working group on the presentation and dissemination of statistics: it was possible that more standardisation across the Government Statistical Service would help to build trust.

Recommendation 2: consultation on developing the range of statistical publications on crime and the criminal justice system

This was welcomed, but with a strong plea for continuity in the data series. The Ministry of Justice has already started to introduce improvements and is planning a user guide to its statistics. More information is needed on flows through the criminal justice system.

Recommendation 3: independent oversight of the production of national crime statistics

This was also welcomed. No one present at the meeting argued in favour of transferring crime statistics from the Home Office, and it was pointed out that there have been improvements to statistical governance, both at the Home Office and more generally. It was noted, however, that although the user network was mostly against transferring the British Crime Survey from the Home Office, this was not a unanimous view. It might therefore be worth
carrying out a thorough cost-benefit analysis.

Recommendation 4: guide to preferred sources of data

This proposal was welcomed. There was less support for the idea of a protocol on the responsible use of crime statistics: in addition to the practical issues mentioned in the report, it might be interpreted as an attempt to stifle comment or discourage new ways of analysing the data. The quality of crime statistics has improved over the years, but confidence has declined. It would not be easy to reverse public perceptions and it would be difficult to influence politicians and the media.

Recommendation 5: guidance on the communication of statistical information at a local level

This was welcomed. The user network felt that all downloadable data should be accompanied by ISO compliant metadata and that there should be standards in place for the compilation of local data. Local statistics needed to be contextualised because they often reflect operational decisions.

Recommendation 6: improving online access to recorded crime and British Crime Survey data

Research carried out during the development of the national crime-mapping website suggests that the public are most likely to access this kind of data when they have a specific purpose in mind (such as moving house), but that on the whole they prefer to receive information via newsletters. The National Policing Improvement Agency has commissioned a randomised control trial, in order to examine what impact these different ways of providing information have on public perceptions.

Recommendation 7: conceptual framework for crime and criminal justice data

This was welcomed and deemed to be necessary, although it remains a long term goal.

Recommendation 8: consultation on supplementary indicators of crime

Some attendees felt this should be attempted, although it was noted that the task would be quite challenging, and that there had been limited feedback when the Home Office consulted on the subject. The work done to produce a crime indicator for the index of multiple deprivation might provide a starting point.

Other points made by attendees at the meeting

  1. The report was considered to be balanced and the interim recommendations were broadly supported.
  2. It was suggested that the report say more about how crime statistics can be used to benefit the public.
  3. The true picture of crime recorded by the police is likely to be distorted where there are local pressures not to record crimes that have a low possibility of detection.
  4. The number of definitions of ‘violent’ crime currently in use makes it easier for different interpretations and narratives to arise.
  5. It was noted that reducing the notifiable list and focusing the statistics on serious crime would weaken continuity and make it more difficult to plan the resources required to cope with demands on the criminal justice system.

Next steps

Those present were reminded that written comments on the interim report would be very welcome (by the end of January). These would be considered by the project board for the review and by the UK Statistics Authority.