Crime Survey for England and Wales consultation update: August 2025

We held a consultation on the redesign of the Crime Survey for England and Wales (CSEW) in 2022. The consultation aimed to update CSEW users on the planned survey redesign and give them the opportunity to comment on the design and future content of the survey. The consultation covered a wide range of topics, including:

• the potential transition to a longitudinal panel design, involving respondents being re-interviewed annually over several waves

• the development of a multi-modal survey, enabling the survey to operate through face-to-face, telephone and online modes of delivery

• a review of the CSEW offence coding classification system

• developments in domestic abuse, sexual victimisation, and stalking estimates

• redevelopment of the design and operation of the Children's Crime Survey for England and Wales (CCSEW)

• development of the Safety During Childhood survey

Following the consultation, users' views and our response were published alongside a planned programme of work in our Consultation response document (PDF, 558KB). Several user events have since taken place and we have published regular updates on work within this programme, alongside our main crime statistical release.

This article updates users on progress made across the entire CSEW redesign programme.

The Crime Survey for England and Wales (CSEW) has always operated as a cross-sectional victimisation survey, with interviews taking place face-to-face in respondents' homes. The transition to a panel design has aimed to enhance the survey by:

• enabling more granular estimates by producing an increased sample size

• modernise data collection and respond to current and future challenges with survey response rates

• improve the flexibility of the survey, so that new questions are developed and implemented at pace

• enable additional data collection that is cost effective

The transformation programme has maintained the established approach to the first interview, ensuring current estimates are unaffected, while experimenting with a panel design. We have not changed how we initially contact respondents, or the mode, conduct, and content of the initial interview. Changes have only taken place following this interview, with respondents being asked to take part in further interviews annually over the phone. Telephone interviews taking place at second and subsequent interviews are a temporary option, while we carry out development work on online interviewing. This is discussed further in Section 3: Update on the development of a multimodal survey. We envisage that respondents will eventually be offered a fully multimodal survey, operating face-to-face, online or over the phone.

We started experimenting with second interviews in October 2022. The first full year of Wave 2 telephone data were available in March 2024. First results were not as expected. There were substantial differences in the estimates of crime provided by the initial face-to-face interviews (Wave 1), compared with second interviews (Wave 2) conducted over the phone. We have been investigating a wide range of possible reasons for these differences, including:

• panel design effects

• attrition between waves

• telescoping effects

• interviewer effects

• modal effects

The discrepancy between waves in the year to March 2024 data could have been caused by any number of reasons. We needed to consider whether such differences would persist over time. As a result, we decided to continue the panel design experiment for a further year to the end of March 2025. This would enable us to see whether differences were consistent over time and would provide more granular analysis based on two years of data, rather than one. We will publish the results of the work conducted in relation to the experiment with panel designs later in 2025, using both the year to March 2024 and year to March 2025 data.

Transitioning the Crime Survey for England and Wales (CSEW) from a face-to-face survey to a multimodal design is complex. We are currently assessing its feasibility and are prioritising online development.

Online development is based on self-completion. This is considered the more challenging mode of delivery, since it operates without the support of interviewer assistance when completing the questionnaire.

Our development work for a multimodal survey involves three main strands:

• development of online screener and victimisation modules

• testing sampling and contact approaches

• development of the "rest of survey" content

Development of online screener and victimisation modules

The main parts of the survey - the screener and victimisation modules - measure the incidence and prevalence of crime. The screener module asks if a respondent has experienced any of 31 offence types (including crimes against their residence, vehicles, and person, and fraud) in the previous 12 months. Further questions identify which incidents will be asked about in detail in a following victimisation module. Answers in the victim forms (which sits within the victimisation module) are used to assess which offence code is allocated.

Currently, the interviewer plays an important role in reducing double counting and further issues, such as the chances of multiple victim forms being generated for the same incident. For this reason, several changes have been suggested to the online screener module ahead of "cogability" (cognitive and usability) testing.

Changes to the online screener module

• Multi-feature incidents, which involve more than one crime happening at the same time, are now identified before the victimisation module.

• Prioritisation of the incidents to be taken forward to a victim form has been improved to more closely align with the Home Office Crime Recording Rules (HOCR) and CSEW offence coding manual.

• The current approach to "series" crime has been modified to define a series as two or three incidents of the same offence type that a respondent considers to be "related"; four or more incidents of the same offence type will now be automatically treated as a series.

• Question wording and ordering has been simplified to better align with respondent mental models and to compensate for the absence of an interviewer.

• Where relevant, more screeners ask about attempted crime than the existing CSEW to improve the completeness and accuracy of data quality.

• The fraud questions have been revised so they no longer ask whether fraud occurred following a traditional crime; they also more methodically route out respondents who are not the Specific Intended Victim (SIV) before the victimisation module.

More information and details of next steps in this programme of research and development are provided in our CSEW Transformation - Discovery Part 3: Redesign of the screener module article. A prototype of the online screener specification is currently being programmed (using Blaise 5 software) in preparation for cogability testing later this year.

Testing sampling and contact approaches

In addition to the development of the screener and victimisation modules, we have begun work to test other aspects of online data collection. This work focuses mainly on operational aspects of an online survey, including:

• the effectiveness of respondent selection approaches

• incentive strategy

• survey response rates

 A field test is planned for autumn 2025, with results published in 2026.

Development of the "rest of survey" content

The CSEW covers a broad range of topics beyond estimates of crime that any multimodal survey instrument will need to incorporate. We have begun work with Verian on the development of the rest of the survey for online completion.

Cognitive testing of revised questions is planned for autumn 2025 and spring 2026.

The Crime Survey for England and Wales (CSEW) offence coding classification process aims to replicate how offences would be recorded by the police, based on the Home Office Counting Rules (HOCR). Updates to the HOCR have introduced differences between the prioritisation order of offences used in the CSEW and the HOCR. These differences largely affect the prioritisation of violence, burglary, criminal damage, and theft. 

We plan to update the CSEW prioritisation order to align more closely with the current HOCR priority order. The proposed changes to the priority order are feasible using the current content of the questionnaire, and do not require making any changes to the survey instrument itself.

We have been reviewing impacts of proposed changes to the priority order have on the published CSEW estimates for the year ending March 2024. This review will be published later this year.

The way domestic abuse manifests is constantly changing. Because of this and legislative changes, over the past few years we have undertaken a programme of user engagement, research and testing to ensure our statistics continue to provide the most accurate information and meet the needs of users. Through this work, we developed a new set of survey questions to measure domestic abuse for the Crime Survey for England and Wales (CSEW).

We have evaluated these new questions and concluded that they provide a better measure of domestic abuse. We decided to use the new survey questions in the CSEW to measure domestic abuse from April 2025. The full evidence behind our decision is provided in our Evaluating a new measure of domestic abuse article.

As the new domestic abuse questions collect information on domestic stalking, from April 2025 the stalking module on the CSEW has been updated. Questions on domestic stalking have been removed and the questions on non-domestic stalking have been updated. As such, the stalking measure for the year ending March 2026 onwards will combine data on domestic stalking from new domestic abuse questions with data on non-domestic stalking from the stalking module.

Alongside the development of new survey questions on domestic abuse, we have also been conducting research to redevelop the survey questions on sexual victimisation. We have considered legislative changes and the evolving nature of sexual crimes to ensure the new questions meet the needs of users. The current questions on sexual victimisation were removed from the CSEW in April 2025, and we aim to launch new questions in October 2025.

The changes to the survey questions on domestic abuse, sexual victimisation and stalking mean that data on these topics will, in the short-term, not be updated on a quarterly basis. The next data to be published will be for year ending March 2026.

To continue meeting user needs and adapt to changes in respondent behaviour following the coronavirus (COVID-19) pandemic, we have been undertaking redesign work of the Children's Crime Survey for England and Wales (CCSEW). We are decoupling the survey from the adult Crime Survey for England and Wales and re-branding it the Young Person's Safety Survey (YPSS). The YPSS will rely on administrative data to sample children aged between 10 and 15 years of age and invite them to take part in the survey online. This transformation will offer many long-term benefits, including: 

• a larger, more representative sample taken directly from administrative data, with potential for including children aged 16 and 17 years 

• improved response rates 

• improved quality and granularity of estimates 

• reduced cost of collection 

• improved sustainability of the survey 

• opportunities for longitudinal data collection 

• improved capability to expand survey content where required 

Since our initial stakeholder engagement, we have undertaken extensive qualitative research exploring the use of an online survey to measure crime among children and young people. The findings from this research are outlined in our Research on Transforming the CSEW report (PDF, 2.3MB). These findings have shaped the development of the survey. For more information, see our Transforming children's crime statistics for England and Wales: March 2023 article.  

An initial small-scale online feasibility pilot was successfully carried out by Verian in early 2023. A second, larger-scale online statistical pilot of children sampled from the English School Census was delivered between February and April 2024. The survey was redesigned for online collection using Office for National Statistics (ONS) survey infrastructure. It achieved an excellent response rate, with results clearly demonstrating the feasibility of a children's online survey of this nature.

The iterative approach we have taken to transformation has enabled us to learn and adapt the survey design in response to feedback from children and parents. This has resulted in clear and engaging survey materials and a short, well-developed questionnaire that we can continue to improve and develop for a full-scale online survey.

Our Children's Crime Survey Steering Group, made up of key stakeholders and experts in this topic area, has been invaluable in providing advice and guidance to help shape the transformation and ensure the flexibility, inclusivity, and resilience of the survey.

Our current plan is to launch an initial module of questions in September 2025. This will cover:

• online activity

• sending and receiving sexual messages

• exposure to harmful content online

• perceptions of safety

• personal well-being

A module of questions that measure children's experience of victimisation and bullying will be added later, following further redevelopment to make them suitable for independent completion online.

The current CCSEW, run by Verian, stopped at the end of March 2025. There will be no data collection from children until the transformed survey is launched. This transformed survey will provide the continuation of data needed for evidence-informed government decisions and the framing of policy and legislative initiatives that support and protect children.

We are still carrying out our feasibility study to determine whether a national survey could provide an effective source of data on the current scale and nature of child abuse. From late 2022 to early 2024, Edinburgh Innovations and the University of Greenwich were contracted to develop a suitable questionnaire and an accompanying safeguarding procedure.

The survey has two components - a survey delivered within a classroom setting for children aged 12 to 16 years and an online survey for young people aged 16 to 25 years who have left school.

The Child Abuse Statistics Steering Group was formed with topic experts from support organisations, academia, government, and education. The steering group helped to agree the abuse types to be included in the survey, and the initial wording for questionnaire testing. We carried out numerous rounds of participatory sessions, followed by cognitive testing, to finalise the wording of the recommended questionnaire.

Alongside the questionnaire, a recommended safeguarding procedure was also developed. It was decided that the survey would be completed anonymously, with a wide range support options available for respondents, parents, and school staff. We published a progress update on this in our Exploring the feasibility of a survey measuring child abuse in the UK: June 2024 article.

We developed materials for the school survey in 2024 and agreed a draft end-to-end operational design for the school pilot. This was discussed with parents of children and schools to understand their perspectives on the end-to-end survey design and to collect feedback on how it could be improved before piloting in schools. The findings were evaluated and used to further improve the survey design.

We agreed an official collaboration with the National Society for the Prevention of Cruelty to Children (NSPCC) at the end of 2024. This involves co-branding all materials, working together on ethics approvals, and getting advice on materials and survey implementation.

The National Statistician's Data Ethics Committee and the NSPCC Research Ethics Committee have granted ethical approval for carrying out a pilot in a single school. We will soon be returning to both committees for approval for piloting in more schools and for the online survey.

We have taken an iterative approach to end-to-end survey design. This has enabled us to learn and adapt the survey design in response to feedback from children and young people, schools, parents, ethics committees, support organisations, and government departments. We plan to pilot both components of the survey this year. The pilots will be evaluated, and the feasibility study will be concluded.

While the Safety During Childhood Survey is still in the feasibility stage, the best source of data on child abuse is from the "abuse during childhood" module on the CSEW. An updated question set was added to the survey in April 2023 for the year ending March 2024 survey year. We plan to publish this data later in 2025.