Bringing data to life - Crime webinar Accessible.pptx

For our eighth webinar, we explored what crime statistics are and how we measure them. We also answered some complex questions on crime statistics, like whether crime is going up or down, or whether there is a 'best' measure to understand trends in overall crime.

1. Bringing data to life

2. Before we start Recording Today’s webinar will be recorded. The recording will be available on YouTube. We’ll share a link after the event.  Questions The MS Teams chat function has been disabled. Please submit questions to Slido #BDTLCrime Be social Share the ‘Bringing data to life’ webinar series by using #StatisticsForThePublicGood on LinkedIn and X. Feedback At the end of the webinar, you will be invited to take part in a short survey.

3. The story of crime in numbers Nick Stripe, Joint Head of Crime Statistics, Office for National Statistics Nick Evans, Deputy Police and Crime Commissioner for Gloucestershire

4. Unpicking the story of crime Nick Stripe, Joint Head of Crime Statistics Office for National Statistics

5. What do you think has happened to crime in the Country as a whole over the past few years?

6. And what do you think has happened to cime in your area over the past few years?

7. Crime is a serious (and emotive) issue

8. Are we experiencing a crime wave? (Or not…?)

9. What data can we use to answer this question? Police Recorded Crime Crime Survey for England and Wales Charities and support services Hospital admissions Industry research

10. What do you think are the challenges of measuring crime data?

11. Crime Survey for England and Wales (CSEW) An in-depth household survey providing estimates of crime, independently from what gets reported to the police.

12. Percentage of CSEW respondents who think crime is going up nationally and locally

13. Are people worried about different crimes? And do they have confidence in the police?

14. Trend in CSEW Headline Crime estimates

15. Trends in estimates by broad crime type

16. Trends in estimates of sensitive crimes and experiences

17. Is the CSEW Headline Crime trend realistic? Northern Irish and Scottish Crime Surveys also both available from 2008/09

18. Police Recorded Crime (PRC) data Includes crimes that have been both reported to (or discovered by) the police AND recorded by them.

19. PRC data since 2002/03 Introduction of current National Crime Recording Standards

20. What’s been going on in PRC since mid-2010s? Crime recording improvements

21. Is the PRC trend realistic?

22. PRC best for lower-volume higher-harm crime

23. Revisiting those headlines… Is theft really “so high”? Is gun crime “soaring”? Is there a “knife-crime epidemic”?

24. Revisiting those headlines… (continued) Is there a shoplifting “epidemic”? Does vehicle theft mean the “harsh reality of crime continues to impact drivers”?

25. And what was one of the five charts showing that Britain is already broken? Hint: You might recognise it…

26. Final thoughts… Understand your data sources, your topic, and its context Nuance doesn’t sell or encourage clicks, but bad news does Think about where a commentator might be coming from SHOCKING! EXCLUSIVE! REVEALED! = we want you emoted! Is data being used selectively? Why are certain time periods being used? Why doesn’t a chart scale start at 0? If you don’t measure it, you can’t understand it or change it If invited, please take part in our surveys

27. How Police and Crime Commissioners use ONS data Nick Evans Deputy Police and Crime Commissioner for Gloucestershire

28. Overview What does a Police and Crime Commissioner (PCC) do? Monitoring police performance in Gloucestershire How we use ONS and police data

29. What do you think a Police and Crime Commissioner does?

30. What does a Police and Crime Commissioner do?

31. Monitoring performance in Gloucestershire PCCs are supported by their offices (OPCCs) who consist of a range of support staff, including data analysts In Gloucestershire, our analysts run a monthly process where they look at over 100 metrics, looking at the data from both a local and national perspective: Local perspective – analysing trends and looking for signals to understand any problems National perspective – understanding whether problems are unique to Gloucestershire or affecting other police forces This is boosted by further analysis to better understand the issues and how we can put measures in place to address them

32. How do you think using data can help to effectively reduce crime rates?

33. Monitoring performance – local perspective We establish a level or limit of crime that we would expect to see for Gloucestershire We look for other signals such as: - consecutive monthly falls - a number of months close to the limits - a number of months above or below average This helps us spot problems quickly Local Normal

34. Monitoring performance – national perspective We establish an average for the forces most similar to Gloucestershire and calculate levels that would be ‘normal’ for that group Where Gloucestershire is outside of these levels this is an ‘outlier’ which may result in further analysis We produce these charts for the latest month/quarter, the year-on-year percentage change and a rolling 12-month total/average National Normal

35. Example of how we use police data – shoplifting Local Trend Analysis (Police.uk) Police.uk data for October 2024 showed shoplifting levels in Gloucestershire were abnormally high, with an increasing trend Comparisons to the most similar group average shows similar shoplifting trend Source: Police.uk Gloucestershire vs Most Similar Group Average

36. Example of how we use ONS data – shoplifting Latest Quarter Latest 12 Months ONS data showed that the previous quarter to Sept-24 Gloucestershire was just below average and not abnormal compared to the most similar group of forces ONS data showed that the previous 12 months to Sept-24 Gloucestershire was just about average but not abnormal compared to the most similar group of forces Source: Police recorded crime and outcomes open data tables – GOV.UK (www.gov.uk)

37. What the data tells us Rising crime figures aren’t necessarily a concern – improvements in recording practices and more targeted, proactive policing operations are examples of why police recorded crime might be going up At first glance the data suggests that shoplifting is increasing, however this is in line with the average for similar forces so is not a trend unique to Gloucestershire ONS data helps us understand that while the data for the month of October 2024 was particularly high, the previous quarter was below average and levels across the year were within what would be expected Further analysis of our own local data suggested a link to cost-of-living increases

38. Our plans Using ONS data helps us put together reports to understand crime levels across the country We intend to produce reports for other PCCs so that they can monitor performance in a more consistent way All of this helps us understand where interventions might be needed

39. Questions [email protected]

40. Q and A word cloud

41. Our upcoming webinars Prices (houses and private rentals) – May Population – June Migration – July GDP – September

42. Find out more Attend other events Follow us on LinkedIn, Instagram, Threads and X Respond to surveys if selected Respond to our consultations and engagement questionnaires [email protected]

43. Annex: FAQ

44. 1. Crime measurements and trends How do we distinguish between real increases in crime and increased public awareness or reporting, measure emerging crime types like cybercrime, and address the challenges in crime statistics reliability and evolving policing priorities? Recorded crime is dependent on what gets reported to or discovered by the police. Increased reporting of crimes to the police will result in an increase in recorded crime, all other things being equal. For example, shoplifting offences are under-reported to the police. In a scenario where the number of shoplifting offences being committed remains constant, but there is an increase in the proportion of those crimes being reported to the police, this will result in an increase in the number of crimes recorded. Police priorities may also focus on certain types of crime at different times, such as street crime, town centre crime, or weapons carrying. An example of this was evident during the height of the covid pandemic. With very extensive social restrictions in place, far fewer people were out of their homes and both the day- and night-time economies were severely curtailed resulting in much less related crime. With resources freed up, many police forces focussed on drug-related crime, leading to a noticeable increase in the recording of these offences. Public awareness, or societal changes and cultural shifts can also impact recorded crime figures. It can be difficult to disentangle changes in the number of crimes recorded that coincide with such societal shifts. For example, sexual offences recorded by the police have increased over the last decade, but this also coincides with much greater public awareness of, and public campaigns related to current and historical sexual abuse. One factor in the increase in recorded sexual offences may have been that more victims were willing to report these crimes. Recorded crime is also dependent on what the police record when crimes are reported to them. We know, from audits by HMICFRS (His Majesty’s Inspectorate of Constabulary and Fire and Rescue Services) and assessments by the OSR (Office for Statistics Regulation) that the police in England and Wales have made notable improvements to the recording of crimes they are made aware of over the last decade. This particularly relates to violence and conduct-related crimes. For these reasons, we do not recommend the use of police recorded crime data as an accurate reflection of long-term trends in crime, particularly those crimes that are more affected by potential changes in public awareness and reporting, and in police recording.

45. 1. Crime measurements and trends (continued) These are important reasons for carrying out the Crime Survey for England and Wales (CSEW). For the crimes that it measures and the population that it surveys, the CSEW provides a long-running time series of crime estimates unaffected by the public’s willingness to report their experiences to the police or the police’s recording of those experiences. Questions are designed to extract information that indicates whether respondents have been the victim of crimes. The core methodology of the CSEW has remained unchanged for decades with estimates of violence, theft and criminal damage going back to the start of the 1980’s. The CSEW has been expanded in recent decades to cover more crimes and more sensitive crime-related experiences. Self-completion modules encompassing domestic abuse experiences, sexual offences, and stalking were introduced from the 2004/05 survey year. We know, from answers to these modules, that the main interviewer-administered parts of the questionnaire do not accurately pick up incidents of sexual offending and domestic violence. We use prevalence measures (the proportion of people affected by these experiences) from the self-completion modules to track changes over time. We are currently updating these modules to reflect changes in legislation and to bring them up to date with emerging awareness of new and different forms of victimisation, for example cyber-related or methods of controlling and coercive behaviour. These updates will result in a break in the time series of prevalence measures related to these experiences. Experiences of fraud and computer misuse were added to the CSEW in the 2016/17 survey year. These crime types increasingly reflect cyber-related victimisation.

46. 2. Criminal justice system and accountability How have changes in crime reporting, police accountability, and prison numbers impacted the reliability of crime statistics, and how do these factors correlate with actual court convictions?   Changes in crime reporting are covered in the answer to the previous question. As is the impact of greater scrutiny of the completeness and accuracy of crime recording by the police over the last decade. Were the police to be held accountable for changes in the levels of crime that they record, this would have the potential to introduce unintended consequences. There may be pressure to reduce the number of crimes being recorded to hit targets. There may be pressure not to focus police resources and priorities on certain types of crime to avoid scrutiny of the likely result – rising numbers of those crimes being discovered and recorded. This would be extremely counter-productive and is again a key reason for having an independent estimate of crime available from the CSEW. Prison numbers have no impact on the reliability of crime statistics. Statistics related to prisons and the courts are produced by the Ministry of Justice. The ONS occasionally compares crimes estimated by the CSEW to those recorded by the police, those that proceed through the courts, and those that result in convictions. A recent example can be found in our latest domestic abuse compendia .

47. 3. Victim and demographics How do victim demographics, geographic variations, and the prioritisation of certain groups influence crime reporting and monitoring, and how do crimes affecting specific groups—such as tourists, children, or rural communities—get tracked and analysed ? We provide annual estimates from the CSEW of victim demographics , both personal and household. These include breakdowns by age, sex, ethnicity, religion, marital status, occupation status, sexual orientation, disability status, preponderance to visit pubs and nightclubs, household structure and income, tenure, accommodation type, area classification, including urban and rural areas, and region. Police recorded crime data is broken down geographically for all 43 Police Force Areas and for approximately 300 Community Safety Partnerships, which largely match local authority areas. Further breakdowns by victim characteristics are not available. The CSEW does not include interviews with tourists, but where crimes affecting tourists have been reported to the police, they will be included in recorded crime data. The same is currently true for crimes where children are victims. Until March of this year (2025) the ONS has also run a Children’s Crime Survey, but due to falling response rates, we are in the process of replacing it with a new online survey . We hope to roll this out over the course of the next year. Demographics from the Children’s Survey can be found in the same link as above.

48. 4. Media, politics, and perception How does media influence public perception of crime, and how is political discourse shaped by crime data? The media, and increasingly social media, influences the public perception of virtually all areas of life, including which issues are of political or cultural salience. Crime has been a popular topic in the media for centuries, with stories seen to drive circulation and engagement. Reports, both accurate and inaccurate, tend to focus on more serious crimes, often with a particular focus on certain types of victims and offenders. With almost wall-to-wall coverage available for those who want it, the public perception of crime can often bear little resemblance to the actual level of crime and whether it is rising or falling. The CSEW, for example, shows a persistently large gap between people’s perceptions of crime in their local area and crime at the more abstract national level. As our event hopefully indicated, the reality of crime is usually much more nuanced. And the level of crime experienced today is much lower than it was 30 years ago for many crime types. With different data sources available, different narratives can be constructed and often are. It’s important to understand the strengths, limitations, and context of each data source, which each ONS crime statistics publication always strives to explain. When talking about public perception, it’s also important however to distinguish between the impact that a certain crime type may have on residents in a particular geography and what Police Forces describe as ‘feelings of safety’. For example, were a knife enabled crime to occur in an urban city centre where there is a (statistically proven) higher frequency of this type of crime, residents might be concerned about the crime, but it might not cause them to fundamentally change the way they go about their day or lead them to feel considerably less safe where they live. However, the same crime, carried out in exactly the same way in a rural village will be more likely to have a much greater impact on feelings of safety to other residents in that area because it is perceived to be a ‘safe place’, and residents will be much more likely to come to the authorities seeking action out of fear about what may happen in future, or whether this is the start of a trend or pattern of behaviour. This perception, or calls for action, are similarly more likely to result in media attention because of how rare such an event or crime is in such an area, therefore beginning to perpetuate those wider fears.

49. 5. Methodology and data collection How does the Crime Survey for England and Wales (CSEW) deal with emerging crime types and granular data gaps? What are the key limitations of these surveys?  The CSEW is a wide-ranging survey covering a large range of topics, experiences, and perceptions. Survey content is reassessed on an annual basis. Key considerations every year include a strong desire for comparability over time, and hence consistency of content, versus a desire to include content in response to current issues, legislative changes, or emerging methods and types of crime. The CSEW has evolved over time. This has included new modules covering domestic abuse and sexual offences (which are currently in the process of being revised and brought further up to date), stalking and harassment, fraud, and computer misuse. In an ideal world a larger sample size would enable estimates to be provided at greater levels of geographical and demographic granularity. Any desire for more content and more sample size must be traded off against the practicalities of maximum acceptable survey length and a broader environment of falling response rates and rising costs. Key limitations to be aware of include: crimes against businesses and crimes against the state, such as drug offences, are not covered the representative population of interest does not include non-household populations, which make up approximately 2% of the total population as a victim-based survey, the CSEW does not include rare crimes where victims are unable to be interviewed due to death or serious injury as fraud and computer offences were first included from survey year 2016/17, comparisons of trends earlier than this cannot include them as with all household surveys, the precision and accuracy of the CSEW is affected by its sample size and shape. victim experiences of many crime types generally have a low prevalence rate meaning relatively large changes must be observed to be statistically significant annual data includes information collected from interviews carried out during the previous year, and each interview asks respondents about the 12-months prior to interview, meaning that each set of estimates includes experiences which can encompass the previous two years