Performance and Measurement

The website presents Police Performance Assessments for all forces in England and Wales.

The Police Performance Assessments allow you to view the performance of your local force and its performance across key areas of policing. Your force can also be viewed in the context of the national picture and against similar forces. A national overview is also available.

Police Performance Assessments are intended to measure, compare and assess the performance of police forces in England and Wales in an effective, fair and transparent way. They have been developed by the Home Office with support from Her Majesty’s Inspectorate of Constabulary (HMIC), Association of Police Authorities (APA) and Association of Chief Police Officers (ACPO).

Public services should meet the needs of users. The assessments reflect this aim by focusing on the ‘end results’ delivered by police forces (with partners) so that:

• local people and communities have information about the quality of local policing;
• police forces can manage and improve their own performance;
• police authorities can monitor local delivery and improvement;
• good practice can be identified and shared;
• problems can be addressed.

 

In 2006/07, seven headline assessments were made covering: Tackling Crime; Serious Crime and Public Protection; Protecting Vulnerable People; Satisfaction and Fairness;  Implementation of Neighbourhood Policing; Local Priorities; and Resources and Efficiency. The HMIC inspection assessment of Protecting Vulnerable People was composed of four components - child abuse investigations; domestic violence; missing persons; and public protection.

For 2004/05 and 2005/06, seven headline assessments were made covering Reducing Crime; Investigating Crime; Promoting Safety; Providing Assistance; Citizen Focus; Resource Use; and Local Policing.
For more information on the coverage of headline areas in a particular year, please refer to the relevant national publication.

Statistical techniques are used to assess performance on the basis of a set of performance indicators. As a complement to this  HMIC uses professional standards and judgement to assess other key areas of policing. These components are allocated to performance areas and aggregated to produce an assessment for each. The process used to aggregate grades may take account of:

• the quality of crime data provided by forces;
• any instances where forces have not provided data or where the data provided is inaccurate in some way;
• priorities set by the Home Secretary that year, for example in the National Community Safety Plan.

Two types of assessments are made for headline areas of performance. They may use one of either performance data and professional judgement, or a combination of the two.

The first assessment concerns the performance delivered by a force over the last year. Typically, this judgement is made by comparing the performance achieved by a force to that achieved by a group of similar forces (peers) in the first year of assesment (2004/05).

Police forces are not identical but this ‘like-for-like’ comparison helps identify any forces which are performing better or worse relative to their peers. For each performance area, four ‘delivery’ grades are possible:

• excellent - if a force performs significantly better than its peers;
• good  - if a force performs better than its peers;
• fair  - if force performance is similar to its peers; and
• poor  - if a force performs much worse than its peers.

A second assessment is made for some headline areas on direction by comparing the performance of a force in one year to that achieved by the same force in the previous year (for example, in the case of 2004/05, forces are compared to 2003/04). For each performance area, three ‘direction’ grades are possible:

• improved   - if performance is much better than previously;
• stable   - if performance is much the same; and .
• deteriorated  - if performance is much worse than previously

Delivery grades should be interpreted alongside direction grades (where available). For example, Tackling Crime might be graded as ‘good / stable’. ‘Good’ means the force is better than its peers were in the first year of assessment (2004/05). ‘Stable‘ means that its performance is much the same as the previous year.

Headline grades are aggregates of other component grades. As such, a force with an excellent grade may have some areas of relative weakness and a force with a poor grade may have areas of relative strength. Finally, assessments are retrospective, so do not necessarily indicate current performance, since this can change over time.

No. Due to the complexity of policing, it would be difficult to bring together assessments across the the broad range of activities involved to provide an overall assessment without hiding the detail of a force’s performance. Police performance assesments are still developing and there is a need to learn lessons before considering if and how to perform such an aggregation.

The current system of assesments was introduced in 2004/05. For the three years prior to this the Home Office published comparative data on a range of statutory performance indicators. HMIC also published grades following baseline assessment. In the recent assessments performance data and baseline grades have been brought together to form an explicit and integrated assessment of performance. For 2006/07 an interim assessment has been developed which brings together data with judgements from HMIC’s inspections.

Statutory Performance Indicators (SPIs) is the term used to describe the quantitative data used to generate Police Performance Assessments. The Home Secretary sets SPIs annually following consultation and police forces are subsequently required to supply data to be used in assessment. Police authorities must also publish force data locally.

• Please refer to further information for the full list of SPIs

Performance assessments can be used locally and nationally to identify good practice and address areas of weakness. This can help to:

• reduce crime and disorder;
• improve crime investigation;
• bring more offenders brought to justice;
• promote community confidence and engagement;
• increase satisfaction;
• reduce fear of crime;
• improve value-for-money; and
• ensure that services are delivered fairly and equitably.

Assessments will also be used by the Home Office and others, for example, to help decide if forces need additional support or whether or not to relax inspection protocols in the light of good performance.

Police performance assessments do not influence the amount of general police grant funding for forces. However, police funding has been revised to better align it with the performance areas within the assessment.

Chief officer pay arrangements include an element of performance pay. Decisions on this should take account of assessment data where appropriate.

Police forces operate in different environments and it is reasonable to expect that performance will vary as these environments differ in complexity. For this reason, it would be misleading and unfair to assess all 43 police forces in England and Wales against each other. For example, it would not be fair or accurate to compare the performance of the Metropolitan Police Service with that of Devon & Cornwall as they each face very different challenges.

Consequently, we compare forces which are similar to each other (peer comparison). This means that for each force a list of other forces that are the most similar to it has been identified. Decisions on which forces are the most similar to each other are made using a range of geographic, demographic, and socio-economic information. As a result, there are 43 ‘most similar groups’: one for each force. We believe this is a more appropriate and useful way of measuring, comparing and assessing performance. The Home Office website shows the performance of each force in comparison to its peers. These most similar forces have not changed between the 2004/05, 2005/06 and 2006/07 assesments.

In most cases, where assessments are made on delivery, this is achieved by comparing the performance of a force to the average performance of its most similar group.  Therefore it should not be assumed that if Force B is in Force A’s most similar group that Force A would be in Force B’s most similar group, as shown is the diagram below.

While the Police Performance Assessments were still under development, an interim approach was used for the ‘Local Policing’ performance area in the 2004/05 and 2005/06 assessments.
That approach was to choose two national indicators that have a bearing on local policing to measure performance and to be used as a benchmark for future years.

The 2006/07 assessment includes, for the first time, locally selected indicators reflecting priorities set by forces and police authorities. These locally selected indicators are designed to ensure that the assessments recognise the results of efforts that forces and authorities make to deal with those issues that are of particular concern to local citizens.

Forces and authorities work with local people to discuss and identify specific issues of concern, and these are then reflected in improvement targets which are measured and monitored locally by the force and authority in question.

For the “Local Priorities” assessment, these local indicators are assessed, along with local perceptions measures.

There were 29 SPIs in 2006/07 and 27 have been assessed. The two SPIs that were not assessed are:

• SPI 3d) Parity of detection for violence against the person: This is because violence in the night time economy is mainly committed by young white men, and when it is seen by police it is easily detected, our focus on reducing alcohol fuelled violence has meant we have inflated the detection rate for such violence and unbalanced the measure, we are considering how to improve it; and
• SPI 8a) Domestic Violence Arrest Rate: This is covered by the detailed HMIC inspection into Protecting Vulnerable People.

There were 32 SPIs in 2005/06 and 29 were assesed. Three SPIs were not assesed because either a robust statistical assessment could not be made, or to avoid perceived double-counting. 

There were 36 SPIs in 2004/05 and 32 were assessed. Four SPIs were not assessed because the type and amount of data concerned meant that a robust statistical assessment could not be made.

There were 27 baseline areas in 2004/05 and 26 were assessed. The baseline area ‘Professional Standards’ was not assessed since HMIC were undertaking a full thematic inspection of this area in 2005/06. All baselines were assessed in 2005/06. HMIC replaced baseline assessments with a more detailed inspection regime in 2006/07.

The following SPIs changed between the 2005/06 and 2006/07 assessments.

•  SPI 3b (Comparative satisfaction of minority ethnic groups) has changed for 2006/07. Previously satisfaction levels of ethnic groups were calculated based on a pooled sample of all victim groups. Under the new definition, the SPI is calculated by taking the mean satisfaction of the 5 victim groups: road traffic casualties, burglary, vehicle crime, violent crime and racist incidents. Each user group therefore makes a 20% contribution to overall satisfaction. This change has been made to eliminate the bias that could arise if forces over-sample any particular user group.
•  SPI 5b (Violent Crime Rate): Following the Smith Review of crime statistics the Home Office has made a number of changes to the way offences are categorised. In particular, Bigamy and Concealing an infant death close to birth have been removed from the categories which constitute violent crime. The number of such offences is small and the effect on the SPI value is negligible.
•  SPI 13a and 13b (Officer and staff sickness absence) have changed to measure the percentage of available hours lost to sickness instead of the sickness hours lost per officer or staff member.

The following SPIs changed between the 2004/05 and 2005/06 assessments.

•  SPI 5b (violent crime rate): Forces are assessed on their violent crime rate not counting Penalty Notices for Disorder (PNDs). PNDs were introduced to tackle low-level anti-social behaviour and to reduce police bureaucracy in dealing with these types of incidents. They can also be issued for behaviour which causes harassment, alarm or distress, but not in cases where (i) the nature of the offence is too serious and/or involves any aggravating circumstances, or (ii) where there has been any injury to any person or realistic threat or risk of injury to any person.
•  SPI 8a (domestic violence arrest rate):  Data and definition changed as a result of a legal change in arrest powers on 1 January 2006 (see Serious Organised Crime and Police Act 2005, section 110).
•  SPI 11a (front-line policing): The way this measure is calculated has changed to account for probationers’ contribution and to incorporate an alteration in the way sick absence is included.

For the 2006/07 assessment, headline assessments of direction of travel have not been possible in some areas, reflecting the new inspection regime adopted by HMIC, and the inclusion of locally selected priorities. A direction assessment has not been made for certain components, as these areas are either new or have undergone significant revision for the relevant assessments.  It would therefore not be accurate to compare the results for these components against the achievements from the previous year.  A direction assessment for these components will be shown once it becomes possible.

Data is obtained from a variety of sources.  This may be recorded crime data, results from User Satisfaction and British Crime Surveys or other sources.  Data for road traffic safety is only available on a calendar rather than financial year basis (for example, 2004 calendar year data is used for the 2004/05 assesments).

Yes.  The data for the SPIs are of differing types, and thus cannot be assessed on the same basis.  However, every indicator has the following information in common, regardless of data type. 

• Data relating to the forces’ performance;
• Comparable data from other police forces are used or, for the direction of travel indicator, comparison uses appropriate historical data from the force itself; 
• A score comparing the force and comparator data;
• The score thresholds required to achieve a particular grade.

For the majority of SPIs, to make an assessment on delivery, a force is compared to the aggregate performance of its peers in the first year of assessment (2004/05). For this assessment, the peer average includes the force itself.

A national target, the same for all forces, has been set for the SPIs relating to front-line policing (11a), female officer representation (12c in 2004/05, 3g from 2005/06 onwards) and sickness absence (13a&b). For these indicators the national average is used as the comparator in place of peer comparison. For the SPI relating to asset recovery (8c) a force specific target has been set and this is used as a comparator.

Some indicators require a force to compare different elements of their own data.  This could be comparing the satisfaction of White and Visible Minority Ethnic (VME) victims of crime with respect to the overall service provided, or comparing the arrest rates from Police and Criminal Evidence Act (PACE) searches between White and VME.

The first year of assessment has been made a benchmark year to which later years are compared in order to make any change in the national picture of policing standards as clear as possible.

The City of London does not have a peer group for comparison purposes because of its relatively unique nature when compared to other forces.  Comparison is therefore made against the national average for indicators where it is sensible to do so.  These include user satisfaction, Offences Brought to Justice (OBTJ), sanction detection measures and road traffic safety.

The City of London provides policing services to a far larger population than those who are resident in the area. Assessment of the recorded crime indicators which are per 1,000 residents is not meaningful. The BCS does not distinguish responses from the City of London to those of the MPS and so grading of these measures is not possible.

Overall this means the City of London is assessed on fewer indicators, and so is proportionately more reliant on HMIC assessments.

Not directly. In the processes of refining assessments, headline assessment areas have changed; a small number of components have changed; and, for a few other components, the methods of assessment have changed. Each set of assessments are best seen as standalone to avoid “comparing apples and pears”.

See "Can I compare assessments in different years?". The direction grade is determined from the change in performance of the components used in the year assessed. These may not be the same as the components of a previous year and therefore directly comparing across years in this way may lead to apparently counter intuitive results.

• Please refer to further information for a detailed explanations of how the SPIs are graded in the various years.

• Please refer to further information for a detailed explanations for the method of aggregating grades in the various years.  (Worked examples are included).

The police use of resources (PURE) scores represent auditor judgements on three of the five themes. The scores for each of the five themes are defined as:

• Level 1 - below minimum requirements - inadequate performance
• Level 2 - only at minimum requirements - adequate performance
• Level 3 - consistently above minimum requirements - performing well
• Level 4 - well above minimum requirements - performing strongly.

Yes, a Welsh version of the 2006/07 publication and a Welsh version of the 2005/06 national publication are available.

Assessments and data

• Methods for Grading SPIs (2004/05)  (PDF download)
• Methods for Grading SPIs (2005/06)  (PDF download)
• Methods for Grading SPIs (2006/07)  (PDF download)
• Method for Aggregating Grades (2004/05) (PDF download)
• Method for Aggregating Grades (2005/06) (PDF download)
• Method for Aggregating Grades (2006/07) (PDF download)
• Monitors Report 2003/04
• Explanation of terms used in publication (PDF download)

Statutory Performance Indicators

• Statutory Instruments, Technical Guidance and Survey Guidance

HMIC Inspection

• List of baseline assessments (2004/05) 
• List of baseline assessments(2005/06)
• Frameworks
• 2006/07 Inspections (available after 9th October 2007)

Crime statistics

• HO bulletin