Crown Prosecution Service Annual Report and Resource Accounts 2007 - 2008

Annex F: Data Systems

PSA 1:

Improve the delivery of justice by increasing the number of crimes for which an offender is brought to justice to 1.25 million by 2007-08.

An offence is said to have been brought to justice when a recorded crime results in an offender being convicted; cautioned; issued with a penalty notice for disorder; given a formal warning for the possession of cannabis or having an offence taken into consideration. The MoJ collects these data from the police and the courts. They constitute National Statistics and are published annually in 'Criminal Statistics in England and Wales'. The total offences brought to justice in the year to December 2007 is comprised of the following:

PSA 2

Reassure the public, reducing the fear of crime and anti-social behaviour, and building confidence in the CJS without compromising fairness.

The Citizenship Survey (formerly the HOCS)

- is a household survey of adults (age 16 and over) carried out by Communities and Local Government (CLG). It covers a range of topics, including perceptions of racial discrimination by public service organisations, and is used to measure performance against PSA targets for CLG, the Home Office, the OCJR and the Office of the Third Sector. The survey has previously been carried out in 2001, 2003 and 2005, providing performance data every two years. From April 2007, the survey has run on a continuous basis. Headline findings on the PSA measures will be available quarterly, with the more detailed 2007-08 annual research reports available in autumn 2008. Data from April-December 2007 are included in this report.

British Crime Survey

The BCS is undertaken continuously, and figures for rolling 12-month periods are available quarterly. Although data are available quarterly, quarter-on-quarter comparisons need to be interpreted carefully as the data sets overlap. It should be noted that the BCS does not measure crimes against people living in group residences, under 16s, or against businesses.

Statistical significance

Statistics produced from surveys are most often estimates of the real figure for the population under study and therefore they may differ from the figures that would have been obtained if the whole population had been interviewed; this difference is known as sampling error. Because of the sampling error, differences in the figures may occur by chance rather than as a result of a real difference. Tests of statistical significance are used to identify which differences are unlikely to have occurred by chance. In tests that use a five per cent significance level, there is a one in 20 chance of an observed difference being solely due to chance.

Confidence intervals

Surveys produce statistics that are estimates of the real figure for the population under study. These estimates are always surrounded by a margin of error of plus or minus a given range. This margin of error or confidence interval is the range of values between which the population parameter is estimated to lie. For example, at the 95 per cent confidence level (used in most surveys), over many repeats of a survey under the same conditions one would expect that these confidence intervals would contain the true population value in 95 per cent of cases

Performance Assessments

These targets are directional (to achieve an increase or decrease) and are measured using survey data. In these cases the survey data must register at least a statistically significant change to be reasonably sure that the measured change is due to an actual change rather than a statistical aberration. In these cases, where interim trends are moving in the right direction but a statistically significant change has not yet been achieved, have been assessed as 'on course'.

Where data trends are moving in the wrong direction or too slowly, it is assessed as 'slippage'.