Jon Simmons

1
Jon Simmons

• Head of Research Statistics
• Crime Reduction and Community
• Safety Group,
• Home Office

Annual meeting of the Crime and Criminal Justice
Statistics User Group (2007 ) 19th March 2007
2
Crime Statistics

• 2000 Simmons Review
• Home Office Review of Crime Statistics
• 2006 Statistics Commission
• Crime Statistics User perspectives
• 2006 Adrian Smith Group
• Increasing Transparency and Understanding

3
The path were on

• 2000 BCS technical review
• BCS became a continuous annual survey, increased
  its sample size to almost 50k respondents, and
  moved to a rolling recall period
• 2001 Changing performance environment
• Publication of police force level performance
  assessments and quarterly national updates
• 2002 Introduction of the new National Crime
  Recording Standard (NCRS) (note previous 1998
  changes also)
• 2002 Offending Crime and Justice Survey
  (lt16s)
• 2003 2nd Commercial Victimisation Survey

4
What were the issues in 2005?

• No concerns over technical quality of the survey
  design or statistical production
• Some perceived confusion over the way in which
  crime statistics were presented
• Significant issues around trust in government and
  perceptions of spin

5

• It is important for our statistics to be good.
  But it is just as important for the statistics to
  be trusted.
• Tony Blair, September 2002
• Public confidence is complex There was
  considerable confidence in the methodologies used
  and the accuracy of outputs BUT There was a
  perception among participants that the Government
  manipulated official statistics AND that the
  media misrepresented official statistics.
• ONS research on Public Confidence in British
  Official Statistics, 28 Feb 2005
• I have been concerned for some time that Home
  Office crime statistics have been questioned and
  challenged to the point that most people seem
  confused about what is happening to crime. This
  is why I have established an independent
  cross-party  group of experts to look at this
  issue. I have asked the group to feel free to
  advise me in whatever way they feel appropriate
  to help us increase public confidence in our
  measures of crime.
• Charles Clarke, Home Secretary, 26 January 2006

6
So, what has been proposed?

• The National perspective
• Expand the scope of crime surveys
• - Under 16s, commercial victims and hard-to-reach
  populations
• Build confidence through independent technical
  assurance
• Maintain the quality of the recorded crime data
• Publish monthly operational data, and more
  subject-specific analyses
• An annual report on the state of the nation
  bringing together data sources (as in the Crime
  in England and Wales statistical reports)
• Revise the statistical classifications,
  especially on violent crime
• Consider UK harmonisation issues
• The Local perspective
• New focus on the neighbourhood (cf. Lyons report,
  Neighbourhood Policing programme, partnership
  focus and Local Area Agreements)
• Provide more detailed local area crime statistics
  for the public
• Distinguish between operational data and national
  outcome measures
• Integrity and Independence
• Governance arrangements actual and perceived
  separation
• Publication arrangements at arms length

7
And what are we doing?

• The Crime Statistics programme Board has met
  twice
• (the RSSs Paul Kiff represents statistics users)
• We will soon be appointing a Technical Advisory
  group
• We will be commissioning some work to look at
  the issues and options in relation to the survey
  recommendations
• (Alison Walker will touch on these issues later
  today)
• We already publish a great deal of additional
  analyses on crime-related issues

8

• Some examples of Home Office research studies
  based on crime survey data
• Young People and Crime Findings from the 2004
  OCJS (Nov 2005)
• Perceptions and experience of anti-social
  behaviour Findings from the 2004/05 British
  Crime Survey OLR 21/06
• Domestic violence, sexual assault and stalking
  (British Crime Survey) HORS 276
• Rape and sexual assault of women the extent and
  nature of the problem. HORS 237
• Crime, Policing and Justice the Experience of
  Ethnic Minorities Findings from the 2000 British
  Crime Survey HORS 223
• Fraud and technology crimes Findings from the
  2003/04 British Crime Survey, the 2004 Offending,
  Crime and Justice Survey, and from administrative
  source
• Alcohol-related assault findings from the
  British Crime Survey OLR 35/03
• Homicides, Firearm Offences and Intimate Violence
  2005/06 HOSB 02/07
• Drug Misuse Declared Findings from the 2005/06
  British Crime Survey HOSB 15/06
• Rural Crime, England and Wales HOSB 1/02
• Crime in England and Wales 2005/06 Supplementary
  Tables Nature of burglary, vehicle and violent
  crime
• Violence at Work New Findings from the 2000
  British Crime Survey
• Two national Commercial Victimisation Surveys
  have been carried out in 1994 and in 2002
• And we also commission technical reports, such
  as
• Crime and Justice Survey General Population
  Feasibility Study (RDS, 2005)
• Crime and Justice Survey Communal Establishments
  Feasibility Study (RDS, 2005)

9

• We are continuing the collection of record-level
  crime data through the development of the NMIS
  Data hub
• (Chris Kershaw will speak about this later)
• We are developing a system to provide easier
  web-based access to the datasets we hold
• We are continuing our engagement with
  stakeholders over the counting rules and
  improving statistical definitions
• We will continue in the next annual volume to
  consider the presentation of crime trends,
  including the pros and cons of a weighted index
  (more later)

10
Crime is a complex social phenomenon

• The measurement of crime can vary according to
• Seriousness
• Injury or degree of harm
• The impact on the victim
• Value of property lost
• Whether demand-led or a product of policing
  activity
• The degree to which they are reported to the
  police
• Police recording practice and rules
• The legal framework
• The authority responsible for dealing with them
  (and therefore recording them)
• and so on

11
What do we mean when we talk about crime?

• The inclusion of a wide range of offences in the
  notifiable offences list means the national crime
  statistics are difficult to interpret, have
  less stable trends, include a number of crimes
  which have variable reporting and recording rates
  - or are the product of pro-active policing which
  varies over time - and are particularly
  susceptible to changes in counting rules or
  recording practices
• (Adrian Smith report)
• The Report went on to say that more serious
  crimes were generally more likely to be reported
  and recorded by the police and so act as a better
  proxy for the overall state of crime.

12
An illustration