1. About this Quality and Methodology Information report

Title of output: Crime in England and Wales, quarterly bulletin

Designation: National Statistics

Geographic coverage: England and Wales

Contact details: crimestatistics@ons.gsi.gov.uk

This quality and methodology report contains information on the quality characteristics of the data (including the five European Statistical System dimensions of quality) as well as the methods used to create it.

The information in this report will help you to:

  • understand the strengths and limitations of the data

  • learn about existing uses and user of the data

  • reduce the risk of misusing data

  • help you to decide suitable uses for the data

  • understand the methods used to create the data

Back to table of contents

2. Quality summary

Output details


 National Statistic   
 Frequency  Quarterly
 How compiled  Combination of survey data and administrative data
 Geographic coverage  England and Wales
 Sample Size  The Crime Survey has a sample size of approximately 35,000 households per year. Police recorded crime is a full count of notifiable crimes reported to and recorded by the police
 Last revised  19 July 2018

Important points

  • The Crime in England and Wales statistical bulletin provides detailed information on different types of crime on a quarterly basis using two main sources of crime data: the Crime Survey for England and Wales (CSEW) and police recorded crime.

  • The CSEW is a face-to-face victimisation survey asking residents of households about their experiences of a range of crimes in the 12 months prior to the interview.

  • The main aim of the CSEW is to provide robust trends for the types of crime and the population it covers; it does not cover crimes against businesses or those not resident in households, and also excludes homicide and crimes often termed as “victimless” (for example, possession of drugs).

  • Police recorded crime figures are supplied to us via the Home Office, from the 43 territorial police forces of England and Wales, plus the British Transport Police.

  • The coverage of police recorded crime figures is defined by the Notifiable Offence List; they do not include “non-notifiable” offences, such as anti-social behaviour and parking offences.

  • Due to concerns over the quality and consistency of crime recording practice, police recorded crime data were assessed against the Code of Practice for Statistics and found not to meet the required standard for designation as National Statistics1; however, the National Statistics status of statistics about unlawful deaths based on the Homicide Index2 was restored in December 2016.

  • For offences that are well recorded by the police, police figures provide a useful supplement to the CSEW and provide insight into areas that the survey does not cover well, such as some of the more harmful crimes that occur in relatively low volumes.

Overview of the data sources

Crime in England and Wales has two main data sources:

Together they provide a more comprehensive picture than could be obtained from either series alone. However, neither the CSEW nor police recorded crime provide complete counts of crime, and there are exclusions from both series. The bulletin also uses a range of supplementary sources to provide a more complete picture of crime, such as:

  • Commercial Victimisation Survey

  • fraud statistics from Action Fraud and the National Fraud Intelligence Bureau (NFIB)

  • anti-social behaviour incidents recorded by the police

  • recorded non-notifiable offences (such as motoring offences, or prosecutions by TV licensing)

Uses and users

We publish statistics on crime levels and trends in England and Wales on a quarterly basis. There is significant interest in crime statistics from a range of users. These include elected national and local representatives (such as MPs, Police and Crime Commissioners and local councillors), Home Office and other government agencies, police forces, those delivering support or services to victims of crime, lobby groups, journalists, academic researchers, teachers and students.

For example, crime statistics produced from CSEW data are used for a variety of purposes, including the development and monitoring of crime and justice policy; public safety campaigns; raising awareness of particular forms of crime; and academic research. They also help to ensure that information on trends in different crime types in England and Wales are available to help inform the choices and decisions of the general public. Further breakdowns, such as offence type and victim characteristics, allow a greater depth of understanding about crime.

Table 1 shows the main user groups of crime statistics and how they use data provided in crime statistics The column on the left lists the classes of use identified by the UK Statistics Authority in their monitoring brief, The Use Made of Official Statistics (PDF, 125.6KB). The right column provides more detail on how crime statistics fits that class of use.

Our other recent publications that make use of the CSEW (and police recorded) data include:

In addition to the range of outputs using data from the CSEW, there are also publications outlining the quality and methodology of the survey. You can find out more about methodology and information reports for crime statistics in England and Wales on our crime statistics methodology page.

Assessment of user needs and perceptions

Any proposals for the future dissemination of crime statistics are made with input from those that read and use crime statistics outputs. Users are invited to share their views on the proposals with us using the consultation feedback form, and comments will help to shape the final proposals.

One example of this was a consultation to review the methodology for addressing high-frequency repeat victimisation in 2016.

Strengths and limitations of the data

Crime Survey for England and Wales

  • The CSEW is our most reliable indicator for long-term trends, particularly for the more common types of crime experienced by the general population, as it is unaffected by changes in reporting rates or police activity, and it includes unreported crimes.

  • The CSEW does not cover crimes against businesses or those not resident in households (for example, people living in institutions or short-term visitors), and also excludes homicide and crimes often termed as “victimless” (for example, possession of drugs).

  • As the CSEW sample size is relatively small in terms of very low volume crimes, estimates of less frequently-occurring crime types can be subject to substantial variability (or “noise”) making it difficult to detect short-term trends.

  • The CSEW is subject to sampling error as it is based on a sample not a census, and also non-sampling error such as issues with respondents recalling past events.

Police recorded crime

  • For crimes that are well-reported and accurately recorded, police recorded crime can provide a good picture of the volume of this crime type.

  • Police recorded crime data also provide an insight into the demands being made on the police, and on where policing effort is being spent.

  • However, for many types of crime, police recorded crime statistics do not provide a reliable measure of levels or trends in crime; they only cover crimes that come to the attention of, and are recorded by, the police and can be affected by changes in policing activity and recording practice and by willingness of victims to report.

  • Due to concerns over the quality and consistency of crime recording practice, police recorded crime data were assessed against the Code of Practice for Statistics and found not to meet the required standard for designation as National Statistics3.

Recent improvements to the data

Information on recent improvements to the design, coverage and presentation of crime statistics can be found in Improving crime statistics for England and Wales.

Notes about quality summary

  1. The full assessment report can be found on the UK Statistics Authority website. Data from the Crime Survey for England and Wales (CSEW) continue to be badged as National Statistics.

  2. Police forces supply a more detailed statistical return for each homicide (murders, manslaughters and infanticides) recorded in their force area to the Home Office than the main police recorded crime series. These returns are used to populate the Home Office database called the Homicide Index.

  3. The full assessment report can be found on the UK Statistics Authority website. Data from the Crime Survey for England and Wales (CSEW) continue to be badged as National Statistics.

  4. Police forces supply a more detailed statistical return for each homicide (murders, manslaughters and infanticides) recorded in their force area to the Home Office than the main police recorded crime series. These returns are used to populate the Home Office database called the Homicide Index.

Back to table of contents

3. Quality characteristics of the data

The quality of a statistical product can be defined as the “fitness for purpose” of that product. More specifically, it is the fitness for purpose with regard to the European Statistical System (ESS) dimensions of quality. This section addresses these quality dimensions and other important quality characteristics, including:

  • relevance

  • accuracy and reliability

  • timeliness and punctuality

  • coherence and comparability

  • accessibility and clarity

More information on the use of these dimensions to measure statistical output quality can be found in the Guidelines for measuring statistical output quality.

Relevance

(The degree to which the statistical outputs meet users’ needs).

Crime in England and Wales has two main data sources:

  • Crime Survey for England and Wales

  • police recorded crime

Together they provide a more comprehensive picture than could be obtained from either series alone. However, neither the CSEW nor police recorded crime, provide complete counts of crime, and there are exclusions from both series.

The bulletin also uses a range of supplementary sources to provide a more complete picture of crime, such as:

  • Commercial Victimisation Survey

  • fraud statistics from Action Fraud and the National Fraud Intelligence Bureau (NFIB)

  • anti-social behaviour incidents recorded by the police

  • recorded non-notifiable offences (such as motoring offences, or prosecutions by TV licensing)

Crime Survey for England and Wales

The combination of the CSEW and police recorded crime generally provides good coverage of crime committed against the public, particularly for offences involving physical harm, loss or damage to property.

For the crime types and population it covers, the CSEW provides a better reflection of the true extent of crime experienced by the population resident in households in England and Wales than police recorded statistics. This is because the CSEW includes crimes that are not reported to, or recorded by, the police. It is also a better indicator of long-term trends than police recorded crime because it is unaffected by changes in counting rules and levels of reporting to the police or police recording practices. The survey is widely seen to operate as an independent check of the police figures. The public perception of the independence of the survey was further strengthened by the transfer of responsibility to us from the Home Office in April 2012.

The CSEW has necessary exclusions from its main estimate of crime (for example, homicide, crimes against businesses and other organisations, and drug possession). The survey also excludes sexual offences from its main crime estimate given the sensitivities around collecting this information in a face-to-face interview. However, at the end of the main interview there is a self-completion element (also via a tablet computer) where adults aged 16 to 74 years are asked about their experience of domestic and sexual violence; these results are reported separately. The upper age limit for the self-completion modules was increased from 59 years to 74 years from the start of the survey year in April 2017, and the first results for this extended age range will be published in due course. In 2009, the CSEW was extended to cover children aged 10 to 15 years.

Until recently, the CSEW did not cover fraud. However, in October 2015, new questions on fraud and computer misuse were added to the survey and in July 2016 sufficient data had been collected to publish experimental data based on these new questions. These estimates were incorporated into the headline CSEW estimates as Experimental Statistics in the Crime in England and Wales, year ending September 2016 bulletin, when a full year’s interview data were available. With two full years of estimates on fraud and computer misuse available from the CSEW since September 2017, we were able to release valid year-on-year comparisons for the first time in the Crime in England and Wales, year ending September 2017 statistical bulletin. Fraud and computer misuse statistics have now been formally assessed by the Office for Statistics Regulation against the Code of Practice for Statistics and were awarded National Statistics status in March 2018.

The questions were previously asked of half the survey sample to test for detrimental effects on the survey as a whole and help ensure that the historical time series was protected. From October 2017 onwards, fraud and computer misuse questions have been asked of the full survey sample.

There are acknowledged methodological limitations of the CSEW. As the survey is based on a sample not a census, there will be sampling errors in the estimates (as with any survey sample). Additionally, non-sampling errors in the estimates arise from a range of causes, including:

  • when respondents have recalled crimes in the reference period that actually occurred outside that period (“telescoping”)

  • crimes that did occur in the reference period may not have been mentioned at all, either because respondents failed to recall a fairly trivial incident or, conversely, because they did not want to disclose an incident, such as a domestic assault (although the self-completion section is used for collecting answers that may be uncomfortable for a respondent to share with an interviewer)

  • some respondents may have said they reported a crime to the police when they did not (a “socially desirable” response)

  • some incidents reported during the interview could be miscoded (“interviewer or coder error”)

Police recorded crime

Recorded crime figures are an important indicator of police workload. They can be used for local crime pattern analysis and provide a good measure of trends in well-reported and well-recorded crimes (in particular, homicide, which is not covered by the CSEW). There are some categories of crime (such as drug possession offences) where the volume of offences recorded is heavily influenced by police activities and priorities; in such cases, recorded crime figures may indicate police activity in this area rather than levels of criminality.

Unlike the CSEW, recorded crime figures do not include crimes that have not been reported to the police or incidents that the police decide not to record. It was estimated in the year ending March 20171 CSEW that around 41% of CSEW comparable crimes were reported by the public to the police, although this proportion varied considerably for individual offence types.

Recorded crime statistics are affected by changes in reporting and recording practices. Police recording practice is governed by Home Office Counting Rules (HOCR) and the National Crime Recording Standard (NCRS). These rules provide a national standard for the recording and classifying of notifiable offences by police forces in England and Wales.

Ongoing consultation on the formulation and development of the policy on crime recording is provided through working groups, comprising members of the Home Office, Office for National Statistics (ONS), police force regional representatives and representatives of the National Police Chiefs’ Council (NPCC), Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) and the Crown Prosecution Service (CPS).

In January 2014, the UK Statistics Authority published its assessment of ONS crime statistics. It found that statistics based on police recorded crime, having been assessed against the Code of Practice for Statistics, did not meet the required standard for designation as National Statistics. However, following a further assessment in December 2016 the Board of the Statistics Authority, on the advice of the Assessment team, restored the National Statistics status of statistics about unlawful deaths based on the Homicide Index. All other recorded crime statistics remain undesignated.

Other sources of crime data

The quarterly statistical bulletin also draws on data from other sources to provide a fuller picture. These include incidents of anti-social behaviour recorded by the police (which fall outside the coverage of notifiable offences), other non-notifiable crimes dealt with by the courts, like littering or parking offences (again outside the coverage of recorded crime or the CSEW), and the results of the Commercial Victimisation Survey (a nationally representative sample of business premises).

Accuracy and reliability

(Accuracy is the degree of closeness between an estimate and the true value. Reliability is the closeness of the initial element value to the subsequent estimated measure).

Crime Survey for England and Wales

Preliminary analysis indicates the response rate for the year ending March 2018 CSEW is 73%. Reasons for non-response were broadly similar to previous surveys. To reduce the risk of bias the survey has been successful at maintaining a high response rate, compared with other large-scale random probability surveys, with very little variation between years, ranging between 70% and 75% over the past decade.

Estimates based on a sample will differ from those that would have been achieved had the whole population been surveyed. This average expected difference is estimated as the sampling error, which is used to construct confidence intervals around the sample estimates. These intervals present the range of possible values that the true population would be expected to provide, with a stated degree of confidence.

Confidence intervals on the CSEW are based on standard errors around estimates, which reflect the stratified and semi-clustered design of the survey and are calculated using the SPSS Complex Samples Module. Where standard errors are calculated without the complex element, a design effect of 1.2 is applied to both the confidence interval and significance testing to allow for the fact that the survey design is not a simple random sample. More information can be found in Section 8.1 of the User guide.

Statistical significance for change in CSEW estimates for overall crime cannot be calculated in the same way as other CSEW estimates. This is because there is an extra stage of sampling used in the personal crime rate (selecting the adult respondent for interview) compared with the household crime rate (where the respondent represents the whole household). Technically these are estimates from two different, though obviously highly related, surveys. Our methodology experts have provided an approximation method to overcome this problem.

The approach involves producing population-weighted variances associated with two approximated estimates for overall crime. The first approximation is derived by apportioning household crime equally among adults within the household (in other words, converting households into adults), and second, by apportioning personal crimes to all household members (converting adults into households). The variances are calculated in the same way as for the standard household or personal crime rates (that is, taking into account the complex sample design). An average of the two estimates of the population-weighted variances is then taken. The resulting approximated variance is then used in the calculation of confidence intervals for the estimate of all CSEW crime and in the calculation of the sampling error.

This method incorporates the effect of any covariance between household and personal crime. By taking an average of the two approximations, it also counteracts any possible effect on the estimates of differing response rates (and therefore calibration rates) by household size.

More information can be found in Chapter 8 of the User guide.

Police recorded crime

Recorded crime statistics are affected by changes in recording practices. To ensure consistency, police recording practice is governed by Home Office Counting Rules (HOCR) and the National Crime Recording Standards (NCRS). These rules provide a national standard for the recording and classifying of notifiable offences by police forces in England and Wales.

There have been two major changes to the recording of crimes in recent years. In April 1998, the HOCR for recorded crime were expanded to include certain additional summary offences and counts became more victim-based (the number of victims was counted rather than the number of offences). In April 2002, the NCRS was introduced across England and Wales, and although some forces adopted important elements of the standard earlier than others, compliance with the standard has continued to improve in the years following its formal introduction. The NCRS was devised by the Association of Chief Police Officers (ACPO) in collaboration with HO statisticians. It was designed to ensure greater consistency between forces in recording crime and to take a more victim-oriented approach to crime recording.

Both these changes resulted in an increase in the number of crimes recorded. Certain offences, such as the more minor violent crimes, were more affected by these changes than others. All of these factors need to be considered when looking at the trends in recorded crime. For these reasons, statistical bulletins present trends following the introduction of recording changes in police recorded crime, from the year ending March 2003.

Police recorded crime statistics, like any administrative data, will be affected by the rules governing the recording of data, what systems are in place and operational decisions in respect of the allocation of resources. More proactive policing in a given area could lead to an increase in crimes recorded without any real change in underlying crime trends.

The UK Statistics Authority (2010) and the National Statistician (2011) have highlighted concerns about the absence of periodic audits of police recorded crime data. A HMICFRS quality review in 2009 into the way police forces record the most serious violence (which at the time was part of a central government target) found some variation in recording. This was partly attributed to the lack of independent monitoring of crime records. In line with a recommendation by the National Statistician, HMICFRS carried out a review of police crime and incident reports in all forces in England and Wales during 2011.

In January 2014, the UK Statistics Authority published its assessment of our crime statistics (PDF, 221KB). The Authority noted “accumulating evidence that suggests the underlying data on crimes recorded by the police may not be reliable” . As a result, the Authority removed the National Statistics “badge” from the recorded crime data and set out a series of essential requirements to enable re-designation as National Statistics. A project is ongoing to address these requirements, supported by a Working Group and Strategic Board, both of which contain representation from important stakeholders. An update on the project’s progress is available.

Additionally, as part of an inquiry by the Public Administration Select Committee (PASC) in 2014 into crime statistics, allegations of under-recording of crime by the police have been made. As part of this PASC inquiry the Chief Inspector of Constabulary, Tom Winsor, outlined how HMICFRS would be undertaking an inspection of the integrity of police recorded crime during 2014.

HMICFRS published an inspection report, Crime-recording: making the victim count (PDF, 1.1MB), on 18 November 2014 based on inspections in all 43 territorial police forces.

The report highlighted that at the national level an estimated 4 in 5 (81%) offences that were brought to the attention of the police and should have been recorded as crimes, were actually recorded, with compliance for specific offence types as follows:

  • burglary – 89%

  • robbery – 86%

  • criminal damage and arson – 86%

  • other offences (excluding fraud) – 83%

  • sexual offences – 74%

  • violence with or without injury – 67%

Compliance varied by police force and separate crime inspection reports for each of the 43 police forces in England and Wales were published on 27 November 2014.

In November 2015, HMICFRS wrote to all Chief Constables advising that they would be commencing an unannounced programme of rolling inspections of crime recording on an ongoing basis. Reports on these inspections will be published on a rolling basis and can be found on the HMICFRS website. Whilst the reports released so far show some improvements have been made, there is still evidence of variation between police forces and need for further improvement in compliance with recording standards.

Most recently, HMICFRS have published their annual report, State of Policing: The Annual Assessment of Policing in England and Wales 2017. This summarises the progress of police forces to improve their recording practices, on the basis of the ongoing Crime Data Integrity programme.

We will continue to publish and provide commentary on police recorded crime data. Responses to a user engagement survey indicate that there remains substantial demand for published data on police recorded crime. There is a summary of findings from the user engagement exercise available: Response to survey of users of police recorded crime statistics (PDF, 159KB).

In addition to the audits described previously, the Home Office carries out regular internal quality assurance of the recorded crime data. Automated monthly variation checks are carried out with error reports being returned to forces for correction, if appropriate. Prior to the publication of any crime statistics bulletin, a verification exercise is carried out with all forces. The data held on the Home Office database are returned to individual forces asking for confirmation that the data is in-line with what is held on their own systems. Again, forces resubmit data if required.

There is more information on the collection and collation of police recorded crime in Chapter 3 of the User guide.

Revisions

Crime Survey for England and Wales

The general principle applied to the CSEW will be that when data are found to be in error, both the data and any associated analysis that we have published will be revised in line with our revisions and corrections policy.

Police recorded crime

The police recorded crime figures are a by-product of a live administrative system that is continually updated with incidents that are logged as crimes and subsequently investigated. Some incidents initially recorded as crime may, on further investigation, be found not to be a crime (described as “transferred or cancelled records”). Some offences may change category, for example, from theft to robbery. The police return provisional figures to the Home Office on a monthly basis with any revised totals for months that have previously been supplied. The Home Office Statistics Unit undertake a series of validation checks of the data and query outliers with forces, who may then resubmit data.

Every three months the Home Office Statistics Unit takes a “snapshot” of the database, covering all crimes recorded by the police. Individual forces are sent their crime counts for the last eight quarters to quality assure. During this process, forces can update historic counts, which are sent back to the Home Office, who then supply final data to us for inclusion in the quarterly publication. As a result of this quality assurance, figures for a specific period in the current publication may differ slightly from those figures corresponding to the same period reported in earlier publications, reflecting the revisions that have subsequently been made by the forces. Figures are always correct at the time of publication and any revisions tend to be small. A table of the differences in data published between the current and preceding publications can be found within the quarterly data tables alongside each quarterly publication.

Timeliness and punctuality

(Timeliness refers to the lapse of time between publication and the period to which the data refer. Punctuality refers to the gap between planned and actual publication dates.)

Crime in England and Wales is published on a quarterly basis. Publication date is approximately 14 weeks after the quarter ends.

For more details on related releases, the official statistics release calendar is available online and provides 12 months’ advance notice of release dates. In the unlikely event of a change to the pre-announced release schedule, public attention will be drawn to the change and the reasons for the change will be explained fully at the same time, as set out in the Code of Practice for Statistics.

Coherence and comparability

(Coherence is the degree to which data derived from different sources or methods, but referring to the same topic, are similar. Comparability is the degree to which data can be compared over time and domain, for example, geographic level).

Between Crime Survey England and Wales (CSEW) and police recorded crime

The CSEW provides a measure of the level of crime committed against a population resident in households in England and Wales, while police recorded crime is a measure of crimes recorded by the police. The CSEW includes crimes that are not reported to or recorded by the police, but is limited to crimes against people resident in households and does not cover all crime types.

The CSEW is a better indicator of long-term trends, for the crime types and population it covers, than police recorded crime because it is unaffected by changes in levels of reporting to the police or police recording practices. The victimisation methodology and the crime types included in the main count of crime have remained comparable since the CSEW began in 1981.

The implementation of both the Home Office Counting Rules (HOCR) and the National Crime Recording Standard (NCRS) mean it is not possible to compare police recorded crime earlier than the year ending March 2003. Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) inspections will provide further information on variation between individual police forces in future.

However, by adjusting each series, comparisons can be made between the two series, which allows a better interpretation of overall crime trends. A comparable subset of crimes has been created for a set of offences that are covered by both measures. Over three-quarters of CSEW offences reported via interviews in recent years fall into categories that can be compared with crimes recorded by the police.

To aid comparability, the offence coding used in the CSEW has been designed to align as closely as possible with the definition of offences in police recorded crime.

There is more information in Chapter 4 of the User guide. A methodological note Analysis of variation in crime trends (PDF, 175.4KB) explored the issue of a possible divergence between the two series.

Between police recorded fraud (including Action Fraud) and other measures of fraud

From April 2011, Action Fraud, the UK’s national fraud and internet crime reporting centre, began to gradually take responsibility for recording fraud in all police forces areas in England and Wales, although it was not until March 2013 that responsibility was transferred from all police forces. In successive quarterly bulletins the proportion of fraud offences recorded by individual forces gradually diminished, and the proportion recorded by Action Fraud grew. The quarterly publication covering data for the year ending March 2015 (published in July 2015) was the first bulletin that includes two comparable years of recorded fraud, where all fraud was recorded by Action Fraud. Comparisons across earlier years should be made with caution, as the transition from local police forces to Action Fraud is thought to have impacted on trends.

Although Action Fraud receives reports of fraud from victims across the UK, data presented in this bulletin cover fraud offences where the victim resides in England or Wales only, based on the victim’s postcode. Experimental Statistics giving a police force area breakdown of crimes recorded by Action Fraud were published for the first time in July 2016.

In the year ending June 2015 quarterly publication, for the first time the police recorded crime series additionally incorporated fraud data, at England and Wales level, from two industry bodies, Cifas and Financial Fraud Action UK (FFA UK). Data from these bodies are fed through to the National Fraud Intelligence Bureau (NFIB) alongside the data from Action Fraud; in previous quarterly bulletins these have been presented separately at UK level only.

Further details on these industry bodies can be found in Section 5.4 of the User guide.

Concepts and definitions

(Concepts and definitions describe the legislation governing the output, and a description of the classifications used in the output).

The Crime in England and Wales quarterly bulletin uses a wide range of classifications.

Geographical

A classification of residential neighbourhoods (ACORN):

ACORN classifies households into 1 of 62 types according to demographic, employment and housing characteristics of the surrounding neighbourhood. It remains a part of the CSEW dataset, but the National Statistics Output Area Classification (OAC, described later in this section) is now used in the standard demographic tables released as part of the National Statistics outputs.

Community safety partnerships (CSPs):

In nearly all cases these have the same boundaries as (are coterminous with) local authority areas. Recorded crime figures for headline offences for each CSP are published.

Regions:

Government office regions (GORs) were established in 1994, in 1996 they became the primary classification for the presentation of regional statistics. From 1 April 2011, the areas covered are referred to as “regions” for statistical purposes.

Indices of deprivation:

Local area deprivation is measured in the quarterly bulletin using the English Indices of Deprivation 2015. There are seven domains of deprivation:

  • income

  • employment

  • health and disability

  • education

  • skills and training

  • barriers to housing and services

  • living environment

  • crime

There are a number of indicators of deprivation within each of these domains, such as level of employment and incapacity benefit claimants, which are combined into a single deprivation score for each local area on that domain. The analysis in this report uses the employment deprivation indicator.

Output Area Classification:

Used to group together geographic areas according to characteristics common to the population in that grouping. These groupings are called clusters and are derived using 2011 Census population data.

Physical disorder:

Used in the CSEW to describe a measure based on the interviewer’s assessment of the level of:

  • vandalism, graffiti and deliberate damage to property

  • rubbish and litter

  • homes in poor condition in the area

The interviewer has to make an assessment as to whether each of these problems is very common, fairly common, not very common or not at all common.

Rural and urban areas:

There are two approaches – the first is our official definition of rural or urban, which was introduced in 2004 to define the rurality of Output Areas. The second is the Rural or Urban Local Authority Classification, which differentiates between rural and urban for those statistics that are only available at local authority level.

Super Output Areas (SOAs):

These are aggregated geographical areas built from Output Areas designed for the collection and publication of small area statistics and, as the building blocks for all National Statistics on a geographical basis.

Household

Household accommodation type:

The CSEW uses the definition of the household’s accommodation, based on the National Statistics harmonised classification, more information can be found in Section 7.2 of the User guide.

Household reference person (HRP):

The HRP is the member of the household in whose name the accommodation is owned or rented, or is otherwise responsible for the accommodation.

Household structure:

The classification of households in the CSEW is based on the number and combination of adults and children living within a household.

Household income:

Total household income is the combined income of all members of the household.

Tenure:

Used by the CSEW and based on the National Statistics harmonised classification.

Personal

Ethnicity:

CSEW respondents are asked to make a choice from a card to identify their ethnic background using a standard National Statistics harmonised classification.

Marital status:

The CSEW uses a classification based on the National Statistics harmonised classification.

Employment status:

The CSEW uses a classification based on the National Statistics harmonised classification, though provides further breakdowns for those in the “economically inactive” category.

Occupation (National Statistics socio-economic classification (NS-SEC)):

An occupationally based classification, but provides coverage of the whole adult population. The NS-SEC aims to differentiate positions within labour markets and production units in terms of their typical “employment relations”.

Chapter 7 in the User guide contains full details on the classifications used in the quarterly bulletin.

Changes to recorded crime classifications

On occasion, the structure of the classifications used to compile recorded crime may change. For example, in April 2012 the number of crime classifications reduced from 148 to 126. The User guide has further details (Chapter 3).

Accessibility and clarity

(Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of the release details, illustrations and accompanying advice.)

Our recommended format for accessible content is a combination of HTML web pages for narrative, charts and graphs, with data being provided in usable formats such as CSV and Excel. Our website also offers users the option to download the narrative in PDF format. In some instances other software may be used, or may be available on request. Available formats for content published on our website, but not produced by us, or referenced on our website but stored elsewhere, may vary. For further information, contact Meghan Elkin via email at crimestatistics@ons.gov.uk.

For information regarding conditions of access to data:

In addition to this, quality information relevant to each release is available in the Quality and methodology section of the relevant statistical bulletin.

Why you can trust our data

Office for National Statistics (ONS) is the UK’s National Statistical Institute and the largest producer of official statistics in the UK. It is the Executive Office of the UK Statistics Authority and is independent of ministers, reporting through the Authority to the UK Parliament and the devolved legislatures.

ONS statistics are crucial for effective debate and decision making in government, industry, academia or by private individuals. In line with the Statistics and Registration Service Act 2007 and the supporting Code of Practice for Statistics, the statistics we produce are designed to meet the wider public good as well as the needs of government.

The Code of Practice for Statistics requires us to continuously improve our engagement with customers on statistical quality by ensuring customers understand the quality of our statistics and by establishing how far the quality meets their needs. Additional government security requirements and the continued concern about data assurance requires us to ensure our confidentiality commitments are met and that the security of our statistics is maintained. Establishing a climate of continuous quality improvement will also reduce cost and enhance value.

The Crime in England and Wales quarterly releases are produced in partnership with the Home Office who collate and quality assure the police recorded crime data presented in the bulletins. Home Office colleagues also quality assure the overall content of the bulletin.

National Statistics are produced to high professional standards set out in the Code of Practice for Statistics. They undergo regular quality assurance reviews to ensure that they meet customer needs. They are produced free from any political interference.

The UK Statistics Authority has designated this statistical bulletin as a National Statistics output, in accordance with the Statistics and Registration Service Act 2007 and signifying compliance with the Code of Practice for Statistics.

However, statistics based on police recorded crime data have been assessed against the Code of Practice for Statistics and found not to meet the required standard for designation as National Statistics. The full assessment report can be found on the UK Statistics Authority website. However, the National Statistics status of statistics about unlawful deaths based on the Homicide Index2 was restored in December 2016 (PDF, 162.3KB).

Notes about quality characteristics of the data

  1. Reporting rates for the year ending March 2017 are the latest data available. Updates for the year ending March 2018 have been pushed back to October 2018 while we implement a change in our method for handling repeat victimisation.

  2. Police forces supply a more detailed statistical return for each homicide (murders, manslaughters and infanticides) recorded in their force area to the Home Office than the main police recorded crime series. These returns are used to populate the Home Office database called the Homicide Index.

Back to table of contents

4. Methods used to produce the data

Roles and responsibilities

We work with the Home Office (HO) to produce crime statistics.

We are responsible for the contract management and processing of the Crime Survey for England and Wales (CSEW), as well as the compilation of results from the police recorded crime statistics. The HO Statistics Unit has responsibility for the collection and validation of the police recorded data from forces and supply the data to us for compilation and publication as official statistics.

At the time of supplying the police recorded crime data, the Home Office provide a document (using an agreed template) detailing all checks and quality assurance procedures that have been carried out as part of quarterly data delivery to us. The Home Office ensure that any corrections to the police recorded crime data made after delivery of the information are communicated as soon as possible, regardless of how minor the correction.

Once a draft bulletin has been compiled, we invite a small number of Home Office analyst colleagues, with expertise in the data, to quality assure the draft. This is to make sure that we have not made any mistakes in compiling the data or errors of interpretation. Home Office comments are considered by our production team and final editorial responsibility lies with us. Our responsible statistician will sign off the final draft before publication.

How we collect, process and analyse the data

Crime Survey for England and Wales

The CSEW is a face-to-face survey in which people resident in households in England and Wales are asked about their experiences of crime in the 12 months prior to the interview. The CSEW was first conducted in 1982 (covering crime in 1981) and ran at mostly two-year intervals until 2001, when it became a continuous survey. Although there have been changes to the survey over time, the wording of the questions used to derive headline estimates have remained unchanged. The core sample is designed to be representative of the population of households in England and Wales and people living in those households. The sample of the population is drawn from the small users’ Postcode Address File (PAF), which is widely accepted as the best general population sampling frame in England and Wales.

The target sample size for the CSEW in the year ending March 2018 was 35,000 (this was reduced from 46,000 in the year ending March 2012). The driver for the reduction in the sample size was both the abolition of central targets for police forces and the outcome of the Government’s Comprehensive Spending Review (CSR). A new requirement for a minimum sample of 650 interviews in each police force area (PFA) was introduced in 2012, replacing the previous design of 1,000 interviews in each PFA.

The sample design was reviewed in the year ending March 2013. The prior sample design employed between 2008 and 2012 had been a mixed design, differing between three types of area, each type defined by the spatial density of its addresses. The new methodology is outlined in Survey Methodology Bulletin, No. 71: September 2012.

Details of experiences of crime are recorded in a series of victim modules. The first three victim modules include detailed questions relating to each incident; the last three victim modules are shorter modules, designed to be much quicker to complete to avoid respondent fatigue during the interview.

Most incidents reported are one-off, single occurrences, but in a minority of cases respondents may have been victimised a number of times. In these cases respondents are asked whether they consider these incidents to be a “series”, that is, “the same thing, done under the same circumstances and probably by the same people”. Where incidents are determined to be in a series, the number of incidents is recorded, but with only one victim module being completed, based on the most recent incident. CSEW estimates include only the first five incidents in a “series” of victimisations in the count of crime.

Overall, each adult respondent can have a maximum of 30 incidents contained in the estimate of crime; a maximum of six victim modules with a maximum of five incidents on each victim module. In practice, most adult respondents have far fewer than this. The restriction to the first five incidents in a series has been applied since the CSEW began. This is to make sure that estimates are not affected by a small number of respondents who report an extremely high number of incidents, which are highly variable between survey years. The inclusion of such victims could undermine the ability to measure trends consistently.

However, where victimisation is prone to be in a series rather than an isolated incident, such as with domestic violence, this current method does not adequately handle the very real occurrence of repeat victimisation. ONS is committed to selecting the best method possible and, as a result, we commissioned a review of the current and alternative methods for handling high-frequency repeat victimisation (PDF, 1MB). The review, which evaluated a range of methods, was completed in spring 2016 and a consultation on the recommendations of this review was published in July 2016. Based on the results of our consultation and advice from the National Statistician’s Crime Statistics Advisory Committee, we are changing the way we estimate repeat victimisation. The 98th percentile of victim incident counts for each crime type will be used as the maximum number of repeat incidents that are included within the calculation on headline CSEW estimates. We currently plan to implement the new methodology and revise the annual back series by Autumn 2018.

Based on information collected and processed from the adult and child victim modules, specially trained coders determine whether what has been reported constitutes a crime and if so, what offence code should be assigned to the crime. The full list of CSEW offence codes is shown in Appendix 2 of the User guide to crime statistics for England and Wales. CSEW crime statistics are produced from these data and presented as incidence or prevalence rates, based on weighted counts of incidents or victims.

To make sure the CSEW sample is representative two types of weighting are used. First, the raw data are weighted to compensate for unequal probabilities of selection. This includes the individuals’ chance of participation being inversely proportional to the number of adults living in the household and the over-sampling of smaller police force areas and the selection of multi-household addresses. Second, calibration weighting to known population totals is used to adjust for differential non-response. All CSEW percentages and rates presented in figures and tables are based on weighted data.

There is more information on CSEW methodology in the User guide to crime statistics for England and Wales and the 2016/17 Crime Survey for England and Wales Technical Report Volume One.

Police recorded crime

Police recorded crime data are supplied to us by the Home Office, who are responsible for the collation of recorded crime data supplied by the 43 territorial police forces of England and Wales, plus the British Transport Police. These data, which are supplied to the Home Office on a monthly basis, are then aggregated and supplied to us for final preparation and publication.

How we quality assure the data

Crime Survey for England and Wales

Several methods are used to ensure the quality of the data collection operation by both Kantar Public, the survey contractor, and ONS.

Kantar Public, the survey contractor, has robust quality management systems in place, which are formally accredited, and are endorsed and supported at a corporate level. This is supplemented by specific working practices and protocols covering all aspects of social research practice – from scripting through to interviewing, coding of offences and data checking. More information can be found in the annual technical reports.

ONS also have quality management systems in place to further quality assure the data when it comes to us for final preparation and publication. Any errors identified through these checks are returned to Kantar Public for validation or correction. Checks are also carried out within the team throughout the data production process before final publication.

Police recorded crime

Prior to submitting data to us, the Home Office Police Data Collection Section (PDCS) and Home Office Statistics Unit carry out internal quality assurance of the recorded crime data. Any anomalies or errors identified through these checks result in a report being returned to the relevant force for validation or correction. Prior to publication of any crime statistics verification checks are also carried out, asking individual forces for confirmation that the data accords with that held on their own systems. For more information, see Chapter 3 of the User guide.

How we disseminate the data

All crime statistics and analysis produced by ONS from both the CSEW and other sources are published on the ONS website under the terms of the Open Government Licence. These are based on aggregate data only and are available to the public free of charge.

Publication dates are planned in advance and pre-announced on the statistics release calendar on both the GOV.UK and ONS websites at least six to eight weeks before the agreed date.

How we review the data

Crime Survey for England and Wales

Every year we undertake a review of the CSEW questionnaire. Among other things it ensures that the questions remain relevant, and allows us to remove questions no longer needed and add questions that address current issues. Relevant stakeholders are contacted for review during the process and decisions about questions are made in consultation with stakeholders. Kantar Public undertake cognitive testing on new questions to ensure the questions are answered in the way intended. Final decisions about changes to the questions are made by ONS.

Police recorded crime

Police recorded crime data are subject to changes in police recording practices. The Home Office inform us each quarter about changes to the recording of crimes as defined in under the Home Office Counting Rules for Recorded Crime documentation.

Back to table of contents

5. Other information

Useful links

The Crime and Justice homepage has links to our full range of publications and data relating to crime statistics.

International standards

There are currently no recognised international standards for crime recording and international comparisons are limited due to the differing legal systems, which underpin crime statistics and processes for collecting and recording crime.

A number of countries run their own national victimisation surveys and all broadly follow a similar process to the Crime Survey for England and Wales (CSEW). While these surveys have a similar objective they are not conducted using a standard methodology – sampling, reference periods and questions vary widely. Therefore, it is extremely difficult to make valid comparisons between the surveys.

The system for recording crime in England and Wales by the police is widely recognised by international standards to be one of the best in the world. Few other jurisdictions have attempted to develop such a standardised approach to crime recording and some of those that have based their approach on the England and Wales model (for example, Australia and Northern Ireland). Therefore, it is difficult to make international comparisons of levels of recorded crime because of the lack of consistency in definitions, legal systems and police or criminal justice recording practices.

Back to table of contents

Contact details for this Methodology