The nature of violent crime in England and Wales: year ending March 2017

• The Crime Survey for England and Wales (CSEW) is the preferred measure of trends in the prevalence of violent crime since this is unaffected by changes in police activity, recording practices and propensity of victims to report such crimes.
• The estimated number of CSEW violent crime incidents was 1.2 million in the year ending March 2017, showing no statistically significant change from the previous year.
• Longer term there has been a statistically significant decrease of 68% from the peak of 3.8 million incidents in 1995.
• Victimisation rates have declined in line with this, with 1.7% of adults aged 16 and over being a victim of violent crime in the 12 months prior to interview in the year ending March 2017 CSEW, compared with 4.8% in 1995.
• Violence without injury accounted for around half (53%) of all CSEW violent incidents, while the higher harm crimes of wounding and assault with minor injury accounted for 24% and 23% respectively.
• The CSEW showed that more than half of victims (57%) did not report their experiences to the police.
• Improvements in crime recording processes by the police are thought to be the main driver of a rise in the number of violence against the person offences recorded by the police.

Violent crime covers a wide range of offences from minor assaults (such as pushing and shoving), harassment and abuse (that result in no physical harm), through to wounding and homicide.

This article includes information on violent crime from two main sources:

• the year ending March 2017 Crime Survey for England and Wales (CSEW)
• violent crimes recorded by the police in the year ending March 2017

CSEW violent crime

The Crime Survey for England and Wales (CSEW) is a face-to-face victimisation survey in which people resident in households in England and Wales are asked about their experiences of a selected range of offences in the 12 months prior to the interview. More information on the methodology can be found in the Crime Statistics Quality and Methodology Information report.

Violent crime statistics from the CSEW are presented under the overall category of “Violence”, which is then broken down into:

• violence with injury – consisting of wounding and assault with minor injury
• violence without injury – where the victim is punched, kicked, pushed or jostled with no resulting injury

Additional breakdowns are also provided based on the relationship between the victim and the perpetrator (domestic, acquaintance, and stranger). Both actual and attempted assaults are included in CSEW figures.

The CSEW covers crimes that are not reported to or recorded by the police, and for the violent crimes it covers, the survey provides a reliable estimate of trends as a consistent methodology has been used to measure these crimes since the survey began in 1981. However, not all victim-based crimes are covered by the CSEW.

The survey does not cover homicide as it is based on the responses of victims. It also does not cover the population living in group residences (for example, halls of residence), those not resident in households (for example, tourists) or crimes against the commercial or business sector. Information on violent crimes against businesses can be found in the 2016 Commercial Victimisation Survey (CVS)¹.

Following criticism of the methodology for handling high-frequency repeat victimisation in CSEW estimates, particularly with regard to violent crime, we commissioned an independent review of the current and alternative methods for handling high frequency repeat victimisation. Following this review, the current cap of five will be changed to a lighter cap to better reflect experiences of repeat victims in CSEW estimates. This change will not impact the estimated number of victims and repeat victims, but it will impact the number of incidents of crime. For more information, see Improving estimates of repeat victimisation derived from the Crime Survey for England and Wales.

All changes reported in this article, based on the CSEW, are statistically significant at the 5% level unless stated otherwise.

Police recorded violent crime

The other main source used in this bulletin is the number of crimes reported to and recorded by the police. These figures are principally a measure of the level of crime-demand on the police and are useful in assessing how caseload has changed both in volume and nature over time.

The coverage of police recorded violent crime is defined by the Notifiable Offence List², which includes a broad range of violent offences, ranging from threats to kill, harassment, and assault without injury, to offences such as homicide. There are some crimes which are not captured in this article, which may involve some degree of violence but which do not have a specific victim, for example, public order offences. See Appendix Table A4, year ending September 2017 for police recorded figures for public order offences.

Police recorded crime figures are supplied to ONS (via the Home Office) by the 43 territorial police forces of England and Wales, plus the British Transport Police³. As with the CSEW, both actual and attempted assaults are included in the figures.

The police recorded crime series covers a broader set of offences and a wider population than the CSEW (for example, residents of institutions, tourists and crimes against commercial bodies), and is a better source of data for high harm, but low volume, violent crimes that are not well measured by a sample survey. However, police recorded crime statistics are affected by changes in recording practices and police activity, and can’t provide a full count of such crime as not all crimes are reported to the police.

Police recorded crime data have been assessed against the Code of Practice for Official Statistics and found not to meet the required standard for designation as National Statistics⁴. However, the National Statistics status of statistics about unlawful deaths based on the Homicide Index⁵ was restored in December 2016.

A renewed focus on the quality of crime recording by the police in recent years is thought to have led to improved compliance with the National Crime Recording Standard (NCRS), leading to a greater proportion of reported crimes being recorded by the police⁶. This means caution should be taken when interpreting police recorded crime trends. Despite improvements made in the recording of violent crime in recent years, the latest inspection reports⁷ from Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) suggest that these offences are still significantly under-recorded by the police⁸. For more information see the Quality and methodology section in Crime in England and Wales, year ending March 2017.

Given the different factors affecting the reporting and recording of these offences, these data do not currently provide a reliable indication of current trends in violent offences and must be interpreted with caution. Although police recorded crime data cannot provide a reliable estimate of trends in the prevalence of violent crime, they do provide information about demands on the police in relation to these offences.

The Home Office are continuing to implement an improved data collection system called the Data Hub. This allows the police to provide more detailed information to the Home Office, such as characteristics of victims and associated aggravating factors of crimes. The migration to the Data Hub is ongoing and for forces providing data via the Data Hub, a more in-depth analysis of police recorded violent crime is included in this article as Experimental Statistics in advance of all forces being able to do so.

Together, the CSEW and police recorded crime provide a fuller picture of violent crime than either source on its own; however, neither should be viewed in isolation. The two data sources should therefore be considered together to develop a picture of violent crime in England and Wales.

Notes for: How is violent crime defined and measured?

1. Results from the 2017 CVS will be published in May 2018.
2. The Notifiable Offence List includes all indictable and triable-either-way offences (which could be tried at a crown court) and a few additional closely related summary offences (which would be dealt with by a magistrate). For information on the classifications used for notifiable crimes recorded by the police, see Appendix 1 of the User Guide.
3. Police recorded crime figures are continually updated. At the time this release was published, the appendix tables published alongside Crime in England and Wales: year ending September 2017 provided the most up-to-date data on crime recorded by the police in the year ending March 2017.
4. 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.
5. 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.
6. The Crime-recording: making the victim count report, published by HMICFRS in late 2014, found that violent offences had been substantially under-recorded (by 33% nationally) and led to police forces reviewing and improving their recording processes.
7. These reports were published during 2016 and 2017, and the most recent reports were published on 28 November 2017.
8. Of the 17 published inspection reports, only four forces received a rating of “good”, with a further five rated as “requires improvement” and eight as “inadequate”.

The year ending March 2017 Crime Survey for England and Wales (CSEW) estimated that 1.7% of adults aged 16 and over were a victim of violent crime in the previous 12 months¹ (Appendix Table 3, year ending March 2017), equating to 1.2 million incidents of violence:

• 0.9% were a victim of violence without injury
• 0.4% a victim of wounding
• 0.4% a victim of assault with minor injury

Figures were similar to the previous year’s survey.

Just over half (53%) of all CSEW violent incidents in the last 12 months resulted in no injury to the victim (Figure 1).

Assaults with minor injury (such as scratches or bruises, 23%) and wounding (where the incident results in severe or more serious injury, 24%) each accounted for around one quarter of incidents.

The police recorded 1.2 million (1,167,998) violence against the person offences in the year ending March 2017², an increase of 18% compared with the year ending March 2016 (992,246 offences):

• violence without injury accounted for 41% (474,006 offences)
• violence with injury accounted for 40% (465,765 offences)
• stalking and harassment accounted for 19% (226,754 offences)³
• death or serious injury – unlawful driving accounted for 0.06% (749 offences)
• homicide accounted for 0.06% (724 offences)

This was the first year that violence against the person offences peaked over 1 million. While these figures are useful in giving an insight into the caseload of the police, they are not believed to provide a reliable measure of the prevalence of violent crime.

Over three-quarters of violence without injury offences⁴ recorded by the police in the year ending March 2017 were classified as assaults without injury⁵ (86%, 408,428 offences), with the remainder covering a range of offences, including:

• threats to kill (5%)
• assault without injury on a constable (3%)
• cruelty to children/young persons (3%)

The majority of violence with injury offences (93%) were classified as assaults with injury (434,737 offences). Assaults with intent to cause serious harm accounted for a further 5% of violence with injury offences (25,304 offences).

There were 724 homicides recorded, which accounted for 0.1% of the total violence against the person offences. This differs slightly from the figure reported in the Homicide in England and Wales: year ending March 2017 article (709), which uses more accurate data on homicide from the Home Office Homicide Index. Please see the Homicide in England and Wales: year ending March 2017 article for more information.

The majority of stalking and harassment offences (97%) were recorded as harassment (219,630 offences). The remaining offences of stalking and racially or religiously aggravated harassment and made up 2% (5,300 offences) and 1% (1,824 offences) respectively. In the latest year, harassment offences accounted for 37% of the overall increase in violence against the person offences.

Around three-quarters of death or serious injury – unlawful driving offences (75%, 564 offences) were classified as causing death or serious injury by dangerous driving. Causing death by dangerous or careless driving accounted for a further 18% of these offences (136 offences).

Violent crimes against children

The year ending March 2017 CSEW estimated that 5.6% of children aged 10 to 15 (359,000) were a victim of violent crime⁶ in the previous 12 months, showing no statistically significant change from the previous year’s survey. Children were more commonly victims of violence with injury than violence without injury (4.5% compared with 1.1%). Over three-quarters of children experiencing violence with injury (3.5% of children) were victims of assault with minor injury (Appendix Table 11, year ending March 2017).

The Home Office Data Hub provides data on child victims of violent crime recorded by the police. For the year ending March 2017 these data are available for 28 forces who supplied adequate data to the Data Hub. Children aged under 16 were proportionally less likely to be victims of violent crimes recorded by the police in the year ending March 2017; those under 16 accounted for 19% of the population, but only 13% of victims of violent crime. Further breakdowns by age can be seen in Figure 8 in the Which groups of people are most likely to be victims of violent crime? section of this article.

Notes for: What is the extent of violent crime?

1. “All violence” includes violence with injury (wounding, assault with minor injury), and violence without injury. For more information see Chapter 5.1 of the User Guide to Crime Statistics for England and Wales.
2. Recorded crime figures presented in this release are those notified to the Home Office and that were recorded in the Home Office database on 29 November 2017, covering the year ending March 2017 – see Appendix Table A4 for the full data table.
3. Stalking and harassment was previously counted under “violence without injury” in police recorded crime statistics, but has been separated out into its own category since June 2017.
4. Stalking and harassment offences are no longer included within the category of “violence without injury”.
5. Assaults without injury offences are those where at the most a feeling of touch or passing moment of pain is experienced by the victim.
6. These are based on a “preferred measure” that takes into account factors identified as important in determining the severity of an incident such as the relationship of the victim to the perpetrator and the level of injury to the victim. See Chapter 2 of the User Guide for further information.

For the population groups and offences it covers, the Crime Survey for England and Wales (CSEW) is the best source for assessing long-term trends in violent crime as the survey’s methodology has remained consistent over time.

CSEW violence peaked in 1995 and has fallen by more than two-thirds (68%) since (Figure 2). There was a sharp decline of 41% between the year ending December 1995 and the year ending March 2002 surveys (from 3.8 million to 2.3 million incidents). More gradual decreases have been recorded thereafter that were not large enough to be statistically significant year on year.

However, the cumulative effect of these changes has resulted in a statistically significant decrease of 41% between the year ending March 2007 and the year ending March 2017 surveys, indicating a gradual, long-term downward trend. The level of CSEW violence has remained fairly flat since the year ending March 2014, with no statistically significant change.

The longer-term reductions in violent crime shown by the CSEW are also reflected in the findings of research conducted by the Violence and Society Research Group at Cardiff University. Their annual survey, covering a sample of hospital emergency departments and walk-in centres in England and Wales shows that serious violence-related attendances in 2016 fell by 10% compared with 2015, continuing a generally long-term downward trend. In addition, the most recent admissions data for NHS hospitals in England show similar findings. Assault admissions for the year ending March 2017¹ (26,450) are 43% lower than the year ending March 2007 (46,763 admissions).

Victimisation rates for violent crime have fallen considerably since 1995 when crime was at its peak (Figure 3). The latest figure (1.7%) is almost a third of that estimated by the year ending December 1995 survey (4.8%). The largest decline over this period has been for assault with minor injury, which at 0.4% in the year ending March 2017 was around a fifth of the rate in the year ending December 1995 (1.9%).

In contrast to the recent flat trend shown by the CSEW, police recorded violent crime has increased by 94% between the year ending March 2013 and the year ending March 2017, with the 1,167, 998 violent offences recorded in the latest year being the highest in a 12-month period since the introduction of the National Crime Recording Standard (NCRS) in April 2002. However, while these figures provide useful insight into the demands on the police and how this is changing, they do not provide a reliable measure of recent trends in the occurrence of violent crime, as the increases are thought to be largely driven by improvements in recording in response to the findings of two Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) inspections.

Firstly the ‘Crime-recording: making the victim count’ report, published by HMICFRS in 2014, showed that violence against the person offences had the highest under-recording rates across police forces in England and Wales. For more information see the User Guide to Crime Statistics.

The 2015 HMICFRS report ‘Increasingly everyone’s business: A progress report on the police response to domestic abuse‘ then showed there had been an improved response by police to domestic abuse following an inspection in 2013.

Police recorded crime data also cannot provide information on longer-term trends in violent crime, as major changes such as the introduction of the NCRS, and the expansion of the Home Office Counting Rules (HOCR) in April 1998², mean that the volume of violent crime recorded before and after these dates is not comparable.

Notes for: What are the long-term trends in violent crime?

1. ‘Hospital Admitted Patient Care Activity, 2016-17’ and ‘Hospital Episode Statistics, Admitted Patient Care - England, 2006-07’ provided by NHS Digital. Assault admissions do not include sexual offences but include assault codes X85-Y04 and Y08 and Y09 from the dataset.
2. The HOCR for recorded crime were expanded to include certain additional summary offences.

The Crime Survey for England and Wales (CSEW) and police recorded crime have shown differing trends in recent years. Although this is largely because of improvements in police recording, it is also partly because of the increased willingness of victims to come forward and report these crimes to the police. It is thought that recording improvements are more likely to affect relatively lower harm violent offences, which explains the larger increase in violence without injury compared with violence with injury.

The different types of violence covered by the two sources may also explain the different trends. The CSEW estimates are dominated by high-volume but relatively low-harm violent crime, and the survey is limited in its ability to detect trends in lower volume offences. In contrast, it tends to be the lower volume but higher harm violent crime that comes to the attention of the police and these types of offences are thought to be generally well recorded. Evidence provided by Metropolitan Police¹ states that, whilst violence against the person crimes recorded by them increased by 5% in the year ending March 2017, the number of “calls for service”² relating to violent crime decreased by 2%. This, along with evidence from the CSEW, which shows no significant change in the prevalence of violent crime in recent years, suggests that the rise in police recorded violence against the person is largely due to process improvements rather than a genuine increase in violent crime.

CSEW violence

All types of CSEW violence³ have followed a similar trend to overall violence, with large reductions since the mid to late 1990s. Assaults with minor injury have shown the larger decrease (79%) since the peak in the year ending December 1995, while wounding and violence without injury decreased by 67% and 58% respectively.

There has been no statistically significant change for any type of violence in the year ending March 2017 compared with the previous year, although the estimated number of incidents of wounding is the lowest since the survey began (Appendix Table 1, year ending March 2017).

Police recorded violence

Ongoing work by police forces over the last three years to improve crime recording practices make interpreting trends in police recorded violence against the person offences difficult. This is thought to have had a larger effect on relatively lower harm types of violent crime, but to have had a lesser impact on higher harm types of violence such as homicide.

Within the overall category of police recorded violence against the person, there were 465,765 offences of violence with injury in the year ending March 2017, an increase of 8% from the previous year. This follows a fluctuating trend since the year ending March 2003, where violence with injury peaked at 543,044 offences in the year ending March 2006, and then gradually declined before rising again from the year ending March 2014 (Figure 4).

Violence with injury

The majority of the increase in police recorded violence with injury between the year ending March 2017 and the previous year was made up of assaults with injury, which increased by 8% (a volume increase of 32,315 offences) (Appendix Table 4, year ending September 2017).

The increase also includes a 16% rise in the number of attempted murder offences (a volume increase of 109 offences) in the last year; this increase may have been influenced by improvements in crime recording. See Focus On Violent Crime and Sexual Offences: year ending March 2016 for more information.

Violence without injury

The trend in the number of police recorded violence without injury offences was relatively flat between the years ending March 2006 and March 2014 at around 240,000 offences per year. Following this there was an increase of 28% to 317,166 in the year ending March 2015, a 26% increase to 399,171 offences in the year ending March 2016, and a 19% increase to 474,006 offences in the year ending March 2017. The majority of this increase (87%) in the most recent year was made up of assault without injury, which increased by 65,233 offences.

Other offences within violence without injury that also saw large volume increases were:

• threats to kill, increasing by 4,449 offences (26%)
• modern slavery, increasing by 1,370 offences (158%)
• cruelty to children/young persons, increasing by 1,324 offences (10%) (Appendix Table 4, year ending September 2017)

Homicide

There were 724 homicides⁴ recorded in the year ending March 2017, an increase of 26% compared with the previous year. However, the 724 homicides recorded in the year ending March 2017 include the 96 cases of manslaughter that resulted from the events in Hillsborough in 1989. Excluding these cases, the number of total homicides increased by 9%; a 24% increase in cases of manslaughter and a 7% increase in cases of murder.

Although showing an increase in the number of offences, homicides have continued to make up only 0.1% of all violence against the person offences since the introduction of the National Crime Recording Standard (NCRS) in April 2002. Please see the Homicide in England and Wales: year ending March 2017 article for more information.

Stalking and harassment

The number of stalking and harassment offences has more than doubled in the last two years (up to 226,754 offences from 86,368 offences), contributing to 36% of the overall increase in violence against the person offences. It is likely that recording improvements are an important factor in this rise.

Within stalking and harassment, the offence of harassment has more than doubled in the last two years (from 81,616 offences to 219,630 offences), representing 98% of the total increase in stalking and harassment in this time period. The expansion of harassment to include two new offences that were previously non-notifiable – “Disclosure of private sexual photographs and films with the intent to cause distress or anxiety” and “Sending letters with intent to cause distress or anxiety” has been a significant contributor to the rise in violent crime. The continued increase will in part be a result of improved compliance in recording of these offences over time.

In the latest joint inspection⁵ conducted by Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) and HM Crown Prosecution Service Inspectorate (HMCPSI), it was found that stalking was not always recorded accurately by the police and in some of these cases, stalking was recorded as harassment. These findings suggest we cannot currently be confident about the accuracy of the recorded crime figures for the separate categories of stalking or harassment.

Notes for: What is happening to trends for different types of violent crime?

1. Metropolitan Police Service Recorded Crime Figures and Associated Data published on London Datastore.
2. Calls for service refer to emergency and non-emergency calls from members of the public or police to attend an incident or investigate a case.
3. See Chapter 5.1 of the User Guide for more information on the offences included in this breakdown.
4. This differs slightly from the figure reported in the Homicide in England and Wales: year ending March 2017 article (709), which uses more accurate data on homicide from the Home Office Homicide Index.
5. Living in fear – the police and CPS response to harassment and stalking inspection report was published in July 2017.

Crime Survey for England and Wales (CSEW) violence can be broken down by the relationship between the victim and the perpetrator – stranger, acquaintance, or domestic. Domestic violence here refers to incidents reported through the face-to-face interview that were perpetrated by a partner, ex-partner or a family member. Domestic violence reported in this way is prone to significant under-reporting and is impacted by the current methodology for handling repeat victimisation¹; therefore the measures of domestic violence reported in this section are an under-estimation of the true extent of this type of violence. As a result, the proportions of violence perpetrated by strangers² and acquaintances³ are over-estimated.

Although the face-to-face CSEW measure of domestic violence is impacted by the current method of handling high frequency repeat victimisation, even when the new methodology is implemented, estimates obtained from the face-to-face interview will still be a large under-estimation of the level of domestic violence experienced by adults in England and Wales because of the high level of under-reporting of this type of violence in face-to-face interviews. Of those aged 16 to 59 who reported being victims of physical domestic abuse in the CSEW self-completion⁴ module on intimate violence, only 14%⁵ also reported being a victim of domestic violence in the face-to-face interview. More information can be found in Chapter 2 of the User Guide for Crime Statistics.

In the year ending March 2017 CSEW, the most common perpetrators of violent crime were strangers (43%, 531,000 offences), with 37% of offences (462,000) perpetrated by an acquaintance, and the remaining 20% (246,000 offences) categorised as domestic violence (Figure 5). These figures have fluctuated over recent years, with acquaintance violence accounting for the largest proportion of offences in some years and stranger violence accounting for the largest proportion of offences in others. The year ending March 2017 CSEW showed that:

• incidents of stranger violence have fallen by 47% from the peak of 1 million offences in the year ending December 1995 to an estimated 531,000 offences in the year ending March 2017 (Figure 5)
• in recent years, stranger violence has shown the largest reduction in number of incidents, decreasing by 38% from 854,000 in the year ending March 2011
• CSEW acquaintance violence fell by 75% from the peak of 1.8 million offences in the year ending December 1995 to an estimated 462,000 offences in the year ending March 2017 (Figure 5), being an important driver of changes in overall violence
• there was no statistically significant change in the year ending March 2017 for domestic, stranger or acquaintance violence compared with the previous year; however the most recent figure is the lowest volume of acquaintance violence ever recorded by the survey
• there were 246,000 incidents of domestic violence measured by the year ending March 2017 CSEW, but this is an underestimation of the true extent of domestic violence (Figure 5)

Given the under-reporting of domestic violence in the face-to-face interview, the 246,000 incidents of domestic violence estimated by the face-to-face CSEW in the year ending March 2017 can only tell us that 20% of the violence measured by this source was domestic violence, and does not provide an indication of the true level of domestic violence. Almost one-third (32%) of violence recorded by the police in the year ending March 2017 was domestic abuse-related, which provides an indication of the underestimation of this type of violence by the face-to-face CSEW. See Domestic abuse: findings from the Crime Survey for England and Wales, year ending March 2017⁶ for information on domestic abuse using the self-completion module of the CSEW.

Relationship between victims and suspects from the Home Office Data Hub

Analysis of the relationship between victims and suspects of violent offences recorded by the police, based on 35 police forces for the year ending March 2017, shows that “other”⁷ relationships between the victim and suspect were most common, particularly for offences against men (76% compared with 47% against women). Further analysis, using a subset of forces⁸, identified that for relationships classified as “other” the victim and suspect were known acquaintances in 50% of offences, were strangers in 28% of offences, and had an unknown relationship (mainly because a suspect had not been identified) in 22% of offences (Appendix Table 10).

For female victims, over a third (37%) of suspects were intimate partners⁹ compared with 12% of male victims of violent offences (Figure 6). There was also a higher proportion of violence against the person offences against females where the suspect was another family member, than against males (16% compared with 12%).

Notes for: How are victims and perpetrators related?

1. See the ONS review commissioned in July 2016 and the update published in October 2017 for more information.
2. Stranger violence includes wounding and assaults in which the victim did not have any information about the perpetrator(s), or did not know and had never seen the perpetrator(s) before.
3. Acquaintance violence is comprised of wounding and assaults in which the victim knew one or more of the perpetrators at least by sight. It does not include domestic violence.
4. “Self-completion” means that the respondent reads the questions themselves and records their answers directly onto a laptop.
5. In previous years, this analysis has included victims of any sexual assault by a partner or family member but these victims have now been excluded to be more comparable with the definition of domestic violence used in the face-to-face CSEW interview.
6. The self-completion module uses a broader definition of domestic abuse, including emotional or financial abuse and threats to hurt the respondent or someone close to them, rather than just physical violence.
7. “Other” relationship comprises other sexual relationships such as casual sexual partners, acquaintances, strangers and cases where the relationship status has not been identified or is unknown.
8. Based on 10 forces who provided detailed relationship data via the Data Hub at a disaggregated level.
9. “Intimate partner” refers to all partner and ex-partner relationships, not just where the couple were married or in a civil partnership, but also including co-habiting partners and those considered in a relationship.

The most common type of violence to be experienced on a repeated basis is domestic violence, which is not well measured by the face-to-face interview of the Crime Survey for England and Wales (CSEW) (Annual Trend and Demographic Table D6). See the How are victims and perpetrators related? section of this article for more information. Therefore, the analysis can only be used to draw conclusions about the repeat victimisation of the violence measured by this source, and cannot be used to draw any conclusions about the overall level of repeat victimisation experienced, or the repeat victimisation of victims of domestic violence.

In the year ending March 2017 survey, the CSEW showed that 73% of victims of violence were victimised once, while 27% were repeat victims in the previous 12 months (12% victimised twice, 15% victimised three times or more) (Annual Trend and Demographic Table D5). The level of repeat victimisation shows no significant change from the previous year’s survey, but has significantly decreased by 12 percentage points since 1995 (Appendix Table 8).

The latest estimates showed that 55% of violent incidents were experienced by repeat victims, compared with 69% in the year ending December 1995 (when CSEW violence peaked) (Annual Trend and Demographic Table D7).

Repeat victimisation was most common for assault with minor injury (28% of victims), followed by violence without injury (24%) and wounding (21%). Repeat victimisation for these offences has fluctuated from year to year, but only violence without injury has decreased significantly since 1995 (by 15 percentage points) (Appendix Table 8).

In the year ending March 2017, a higher proportion of victims of domestic violence measured by the CSEW were repeat victims (39%) than victims of acquaintance violence (29%) and stranger violence (20%). However, these proportions fluctuate considerably from year to year and this does not provide a complete picture of repeat victimisation suffered by victims of domestic abuse.

The CSEW self-completion module provides a more complete measure and a greater level of detail on domestic abuse. New questions to measure frequency of abuse (and therefore repeat victimisation) have been implemented into the self-completion module from April 2017 (see the Crime and justice methodology page to see the questionnaires).

The year ending March 2017 Crime Survey for England and Wales (CSEW) showed that victimisation of violent crime varied by certain personal and household characteristics (see Appendix Tables 1 to 4 for a full breakdown). Many of the characteristics are closely associated with each other, so caution is needed in the interpretation of these different characteristics when viewed in isolation. While the CSEW provides good estimates of most crime types, it is known that the main face-to-face survey underestimates the number of domestic violence incidents. See Domestic abuse: findings from the Crime Survey for England and Wales, year ending March 2017 for a breakdown of characteristics of victims of domestic abuse taken from the self-completion module of the CSEW.

Sex

Men were more likely to be victims of CSEW violent crime than women (2.1% of males compared with 1.3% of females¹, Figure 9). This was true for all types of violence, with the exception of acquaintance violence which showed no significant difference and domestic violence which showed the reverse trend (0.4% of females were victims compared to 0.2% of males). The year ending March 2017 CSEW showed that:

• stranger violence showed the largest difference in victimisation between men and women (1.3% compared with 0.4%)
• around twice as many men (1.2%) as women (0.6%) experienced violence without injury

Data from the Home Office Data Hub shows that in the year ending March 2017, more violence against the person offences recorded by the police had female victims (53%) than male victims (47%). This is notably different from the CSEW for the same period, which estimated that 61% of victims of violence were male, with 39% being female (data not shown). A likely reason for the difference between the sources is that females are more commonly the victims in cases of domestic abuse, which accounts for around one-third of violence recorded by the police. While the CSEW provides good estimates of most crime types, the main face-to-face interview under-estimates the number of domestic violence incidents (see the How are victims and perpetrators related? section of this article for more information).

The breakdown of male and female victims differed by the type of violence, with females accounting for a larger proportion of victims of violence without injury than males (57% compared with 43%), but males accounting for a larger proportion of victims of violence with injury (53% compared with 47%, Figure 7).

Figures from the Home Office Homicide Index for the year ending March 2017 show that 74% of homicide victims were male and 26% were female. More information on homicide offences can be found in Homicide in England and Wales: year ending March 2017.

Age

Adults aged 16 to 24 were more likely to be victims of CSEW violence (4.1%) than those in older age groups (Figure 9; Appendix Tables 1 and 3). This was particularly pronounced for stranger violence (2.1%) and acquaintance violence (1.5%).

Information from the Home Office Data Hub on the age of victims of police recorded violence from 28 forces shows that younger adults were more likely to be victims of violent crimes recorded by the police in the year ending March 2017 (Figure 8). For example, while those aged 20 to 29 made up around 13% of the population², they were victims in 27% of violence against the person offences recorded by the police. Older people and the very young were less likely to be victims of violent crime. For example, those aged 80 or older were victims in less than 1% of violent offences but made up around 5% of the population. Those aged under 10 accounted for 12% of the population but only 4% of victims.

Figure 8: Proportion of violence against the person offences, by age and sex of victim, Home Office Data Hub (28 forces)

England and Wales, year ending March 2017

Source: Home Office Data Hub

Notes:

1. Police recorded crime data are not designated as National Statistics.
2. Police recorded crime data based on 28 forces that provided data via the Home Office Data Hub or in manual return. Data have not been reconciled with forces.
3. Population figures are based on the ONS population estimates for England and Wales.

Download this image Figure 8: Proportion of violence against the person offences, by age and sex of victim, Home Office Data Hub (28 forces)

.png (20.6 kB) .xls (29.7 kB)

Other characteristics

• Those who were single (3.2%) were more likely than adults of other marital statuses to be victims, except for those who were separated (2.6%) where there was no significant difference.
• Adults who were unemployed (4.4%) were more likely to be victims of violent crime than those who were employed (1.8%).
• Adults living in the 20% most deprived output areas were more likely to be victims (2.2%) than those living in the 20% least deprived output areas (1.2%) and other output areas (1.6%).
• Single adults with children (3.7%) were more likely to be victims of violence than those living with other adults and children (2.0%) and adults living with no children (1.4%).
• Renters (2.8% social and 2.3% private) were more likely to be victims than home owners (1.2%, Appendix Table 2).

For children aged 10 to 15, boys were more likely than girls to have experienced violent crime (7.5% compared with 3.5%). See Tables D3 and D4 of the Annual Trend and Demographic Tables, year ending March 2017 for more information.

Notes for: Which groups of people are most likely to be victims of violent crime?

1. This pattern is different for domestic abuse and sexual assault. For more information please see “Domestic abuse: findings from the Crime Survey for England and Wales, year ending March 2017” and “Sexual offences in England and Wales: year ending March 2017”.
2. Based on the Office for National Statistics England and Wales population estimates.

Victims of Crime Survey for England and Wales (CSEW) violent crime were able to provide some detail about the perpetrator(s) for 99% of incidents (Nature of Crime Table 3.1). The following profiles are based on the victims' recollection and perception of the perpetrator(s):

• perpetrators were most likely to be male, being reported to be the perpetrator in around three-quarters of violent incidents (78%)
• perpetrators were most likely to be aged between 25 and 39, with the perpetrator believed to belong to this age group in 37% of violent incidents
• in 76% of violent incidents a sole perpetrator was reported to have been responsible; for incidents with more than one perpetrator, victims most commonly reported that four or more perpetrators (11% of incidents) or two perpetrators (9% of incidents) were involved
• only 5% of domestic violence incidents involved more than one perpetrator, compared with 24% of incidents of acquaintance violence and 33% of incidents of stranger violence
• incidents involving four or more perpetrators accounted for 12% of acquaintance violence and 15% of stranger violence, but 1% of domestic violence

Victims aged 10 to 15 were able to say something about the perpetrator in 93% of violent incidents in the year ending March 2017 CSEW. Incidents of violence against children were most likely to be committed by someone known well to the victim (53% of incidents), with a small proportion of incidents being committed by strangers (11%). The perpetrator was a pupil at the victim’s school in 72% of violent incidents, and was a friend (including boyfriend or girlfriend) in 18% of incidents. The perpetrator was most likely to be male (72% of incidents) and aged between 10 and 15 (78%) (Nature of Crime Table 4.3).

Involvement of alcohol in violent crime

Victims believed the perpetrator(s) to be under the influence of alcohol in 40% (464,000) of violent incidents¹. In 18% (207,000) of violent incidents, the victim believed the perpetrator(s) to be under the influence of drugs (Nature of Crime Table 3.11). Characteristics of perpetrators of alcohol-related violent crime can be found in Appendix Table 7.

Information on the involvement of alcohol as a factor in violent crime is also available from the Home Office Data Hub, which contains a field where police forces can identify whether the offence was “alcohol-related”. It was not mandatory for forces to populate this field in the year ending March 2017, but a standard definition for usage was introduced in April 2016.

Analysis for alcohol-related violent offences is based on 31 forces providing data using the alcohol-related aggravating factor flag in the Data Hub. These forces accounted for around 77% of violence against the person offences in England and Wales in the year ending March 2017, and include the Metropolitan Police, who alone recorded 16% of violence against the person offences in England and Wales in this year.

The analysis may not be representative of all forces in England and Wales and data have not been reconciled with forces and are therefore subject to revision. The Home Office continue to work with police forces to ensure the consistency and comparability of the victim information they supply to the Home Office.

In the year ending March 2017, 16% of violence against the person offences were flagged by the police as alcohol-related. The corresponding figure (as published last year) for the year ending March 2016 was also 16%, however, the two years are not directly comparable as a different set of police forces were used in each year.

The offence of “assault without injury on a constable” had the highest proportion of offences that were alcohol-related (35%), although it only accounted for 1% of all violent offences. “Assault with injury” (22% of which were flagged as alcohol-related) and “assault without injury” (16%) were more common offences and respectively accounted for 37% and 35% of all police recorded violence. Violent offences most likely to be flagged as alcohol-related are shown in Figure 10^2,3.

Notes for: What do we know about perpetrators of violent crimes?

1. Questions were asked if the victim was able to say something about the perpetrator(s), which they could do in nearly all (99%) incidents. If there was more than one perpetrator, victims were asked if any of the perpetrators were perceived to be under the influence. Questions were not asked if any perpetrator was perceived to be under 10 years.
2. There are some violent offences with a high proportion of alcohol-related offences which have been excluded from this analysis due to very low volumes.
3. Selected violence against the person offences were those offences with the highest proportion of the alcohol-related flag.

In the year ending March 2017, the Crime Survey for England and Wales (CSEW) estimated that 43% of violent incidents were reported to the police (Table D8, year ending March 2017). This is a statistically significant decrease from the previous year (52%). However, this is a return to levels seen in the year ending March 2013 and all previous years in which this measure has been recorded. This latest reporting rate for violence compares with a reporting rate for all CSEW crime of 40%.

As in previous years, reporting rates continue to vary by type of violence, with 51% of wounding incidents in the latest survey year being reported to the police compared with 40% of incidents of assault with minor injury or no injury (Figure 11). This indicates respondents are more likely to report more serious incidents to the police.

In the year ending March 2017, the CSEW showed that 16% of violent incidents experienced by children aged 10 to 15 were reported to the police (Nature of Crime Table 4.5). This figure is likely to reflect the relative low severity of violent incidents experienced by children.

More than half of violent incidents (55%) occurred in the evening or during the night (Nature of Crime Table 3.3). This level has remained fairly stable over the previous three years, showing no significant change. However, there has been a significant decrease since the year ending March 2007, indicating a longer-term shift in the pattern.

More than half of violent incidents (57%) occurred during the week, with 43% occurring during the shorter period of the weekend (Table 1.1). This pattern is the same as last year, when the majority of violent offences took place during the week, but is different compared with the years ending March 2014 and March 2015. There is some fluctuation in this trend, and prior to this, the majority of violent offences took place during the week. It is therefore too early to conclude whether the changes seen in the years ending March 2016 and March 2017 are the beginning of a new pattern. However, the differences between week and weekend in the last two years have been larger than they were previously. Timing of alcohol-related violent crime can be found in Appendix Table 5.

In the year ending March 2017, 94% of incidents of violence against children aged 10 to 15 occurred during the week and 6% of incidents occurred at the weekend. This highlights the different lifestyles of children compared with adults, and also reflects that a large proportion of violent incidents against children (72%) occurred in or around school (Nature of Crime Tables 4.1 and 4.2).

Similar to previous years, the location of incidents of Crime Survey for England and Wales (CSEW) violent crime varied by the victim-perpetrator relationship.

As expected, the large majority of incidents of domestic violence¹ occurred around the home (79%), whereas incidents of stranger violence were most likely to occur either in pubs and clubs (27%) or around work (21%). Incidents of acquaintance violence were most likely to take place at work² (40%, Figure 12). Information on the location of alcohol-related violent crime can be found in Appendix Table 6.

Looking at the location of these incidents split by type of violence (Nature of Crime Table 3.2), incidents of wounding were most likely to occur around the home (38%).

For assault with minor injury and violence without injury incidents of violence were more likely to happen at work (27% and 26% respectively).

Notes for: Where do violent crimes occur?

1. As measured in the face-to-face section of the CSEW.
2. For more information see Health Safety Executive ‘Violence at work’ report using CSEW data here.

A weapon was used in 19% of violent incidents according to the year ending March 2017 Crime Survey for England and Wales (CSEW) (Nature of Crime Table 3.8). Weapons were used in a higher proportion of incidents of violence without injury (22%) than incidents of violence with injury (14%). The most commonly used weapons fell under the category of “other” (used in 7% of violent incidents) which includes axes, swords, cleavers, dogs, and other weapons. Knives and other stabbing implements¹ were used in 6% of violent incidents.

The most common types of weapon used in domestic violence were knives and stabbing implements (used in 6% of violent incidents). These weapons were also the most common for stranger violence (8%). An “other” weapon² was the most common (11%) in acquaintance violence.

A glass or bottle was the most common weapon used for wounding (5%) and an “other” weapon³ was most common for both assault with minor injury (9%) and for violence without injury (7%).

As offences involving the use of weapons are relatively low in volume, the CSEW is not able to provide a reliable measure of such incidents and can only tell us that 19% of the violence measured by this source involved a weapon. This is where police recorded crime can provide additional insight as these offences are believed to be relatively well recorded by the police. The past three years have seen a rise in the number of offences involving weapons recorded by the police. For more information on these offences, see Offences involving the use of weapons: year ending March 2017 data tables, which contain information on crime involving knives and firearms.

Knife-carrying

A self-completion⁴ module of the CSEW asks 10- to 15-year-olds and 16- to 29-year-olds about knife-carrying⁵.

The year ending March 2017 survey estimated that 5.7% of 10- to 15-year-olds knew someone who carried a knife for their own protection, a percentage that has not showed much variation over time (Appendix Table 9). A similar percentage (4.8%) of 16- to 29-year-olds knew someone who carried a knife.

Less than 1% of respondents (0.6% of 10- to 15-year-olds, 0.5% of 16- to 29-year-olds) personally carried a knife. Again, the changes in this series over time have not showed much variation.

Notes for: Use of weapons in violent crime

1. Includes screwdrivers and other stabbing implements.
2. Includes axes, swords, cleavers, dogs, and other weapons.
3. Includes axes, swords, cleavers, dogs, and other weapons.
4. “Self-completion” means that the respondent reads the questions themselves and records their answers directly onto a laptop.
5. The questions were first asked in the year ending March 2012 survey for 10- to 15-year-olds but were not asked of 16- to 29-year-olds until the year ending March 2014.

Victims sustained a physical injury in 47% of incidents of violence in the year ending March 2017 Crime Survey for England and Wales (CSEW) (Nature of Crime Table 3.7). This varied by perpetrator, with 63% of incidents of domestic violence resulting in physical injury compared with 51% of acquaintance violence and 36% of stranger violence incidents. The figure for domestic violence should be interpreted with caution as the level of impact on the victim could affect the likelihood of them reporting in the face-to-face interview.

The most common type of injury experienced was minor bruising or black eye (29% of violent incidents). More serious injuries such as concussion or loss of consciousness (2%) or broken bones (1%) were less common (Nature of Crime Table 3.7).

Incidents of violent crime experienced by children were more likely to result in injury, with victims aged 10 to 15 sustaining an injury in 81% of incidents in the year ending March 2017 CSEW. Victims received some form of medical attention as a result of the violence in 24% of incidents. Of those incidents where the victim aged 10 to 15 sustained an injury, minor bruising/black eye was the most common injury (65%), followed by scratches (19%) and severe bruising (15%). In 8% of violent incidents where the victim aged 10 to 15 sustained an injury, this was a serious injury¹ (Nature of Crime Table 4.6).

Notes for: Injuries resulting from violent crime

1. Serious injury includes facial/head injuries, broken nose, concussion, broken bones and eye/facial injuries caused by acid, paint, sand, etc thrown in face.