Exploring the feasibility of a survey measuring child abuse in the UK: June 2024

Child abuse can have long-lasting adverse individual, familial, and society-level effects. It is recognised by The World Health Organisation that accurate prevalence data is needed to monitor trends over time, develop effective prevention strategies, and appropriately support those who have experienced abuse. 

We have been assessing the feasibility of a survey to measure the prevalence of child abuse in the UK as there is currently no single source of accurate data.  

We are currently in phase two of the project, which includes developing and conducting a pilot survey with children and young people aged 11 to 25 years. (For more information on phase one, please see previous progress updates published in January 2021 and April 2022).  

In November 2022, we commissioned the University of Edinburgh and the University of Greenwich to design two questionnaires and a safeguarding procedure. Alongside this work, we have been developing the methodological and operational design for the survey.  

Based on research to date, a survey should: 

• cover 10 types of violence or abuse which includes the addition of community violence, corporal punishment, peer violence and bullying and dating violence to our original proposal (neglect, physical abuse, emotional abuse, sexual abuse, child exploitation and exposure to domestic violence or abuse) 

• cover demographics, risk factors, protective factors and outcomes 

• be conducted independently online for young people aged 16 to 25 years 

• sample schools from publicly available school lists across the devolved nations 

• be collected completely anonymously with a wide range of opt-in support options across respondents, parents, and school staff   

• include tailored safeguarding messages as a result of survey responses, encouraging respondents to seek support 

• balance children's rights alongside safeguarding, allowing children aged 13 years and over to solely consent for themselves to participate

We have also agreed accessibility functions to be built into the data collection process to increase inclusion of disabled children and children with additional needs. 

These recommendations are in addition to our previous proposals which were that a survey should: 

• be framed as a survey of child safety 

• ask about lifetime and past year experiences of abuse 

• be self-completed electronically 

• take place in schools for children aged 11 to 15 years 

• be relatively large, with a minimum of 45,000 respondents needed to produce useful estimates 

Before a survey can be piloted, further research activities are required including:  

• validating existing legal advice on the proposals surrounding consent, anonymity, and safeguarding with an independent advisor  

• conducting further school engagement to gain insight on our current proposals and how these could work operationally 

• conducting further parental engagement on the design of the materials, consent proposals and information sessions 

• agreeing the sampling frame for the online survey of 16-to-25-year-olds  

• establishing appropriate data protection arrangements for data collection and storage 

Next, we'll be completing the outstanding research activities while planning the pilot.

The aims of the contract with the University of Edinburgh and the University of Greenwich were to:

• design two questionnaires, one for 11-to-16-year-olds, to be completed in schools and one for 16-to-25-year-olds to be completed independently either online or with an interviewer present

• design an appropriate safeguarding procedure ensuring the participants are supported whilst they are completing the survey and afterwards

Their research involved:

• online Delphi surveys and focus groups with violence research and child protection professionals

• focus groups with adults who have experienced abuse as a child

• individual and group cognitive interviews, participatory group research sessions and qualitative interviews with children and young people aged 11 to 27 years who have experienced abuse as a child

• individual and group cognitive interviews, participatory group research sessions and qualitative interviews with children and young people aged 13 to 26 years who have no known experiences of abuse as a child

• group consultations with our Child Abuse Statistics Steering Group (CASSG) (see Section 7: Further information on the Child Abuse Statistics Steering Group)

Their research also included individual consultations with: 

• government departments 

• researchers from other nations experienced with their own national prevalence studies of child abuse  

• safeguarding professionals and researchers 

• education personnel 

• ethics board members 

• special educational and/or support needs professionals 

• abuse/violence support, prevention, and research organisations

We have also been developing the methodological and operational design for the survey including considering ethical implications of decisions.

This progress report will outline the main findings from the phase two research to date, and how it has impacted our previous proposals. The University of Edinburgh and the University of Greenwich aim to publish more detailed findings from their research in journal articles later in the year.

The survey should cover 10 types of violence of abuse

We previously recommended that the survey covers six types of abuse:

• neglect

• physical abuse

• emotional abuse

• sexual abuse

• exploitation

• exposure to domestic violence and abuse

Further research, including consultation with the Child Abuse Statistics Steering Group, has outlined the importance to also include:

• community violence

• corporal punishment

• peer violence and bullying

• dating violence

The survey should cover demographics, risk factors, protective factors and outcomes

Including these additional questions hugely improves the utility of the data for policy makers and service providers.

Demographics would help understand variation in prevalence by characteristics. Risk and protective factors (such as household dysfunction, and school culture or resource) would give an understanding into the impact on prevalence and future preventative strategies.

Outcomes would include health risk behaviours (such as drug and alcohol use) and mental health questions, which would provide insight into the impact of abuse.

Young people aged 16 to 25 years should complete the survey online

We have considered two methodologies, Computer-Assisted Self-Interviewing (CASI) with an interviewer present or independent online collection.

Our previous research with young people with experiences of abuse found they would prefer to complete the survey online. They felt this would give them a greater sense of privacy and control. Further research with professionals and children/young people, raised concerns about providing appropriate safeguarding, particularly on the distress of participants answering such sensitive questions. However, research has shown that "participant distress is infrequent and transitory, (and) that researchers can satisfy ethical requirements towards participants".

We have carefully considered the potential risk for harm and embedded ethical considerations throughout the design of an online pilot. The National Statistician's Data Ethics Advisory Committee (NSDEC) in July 2023 was assured by our suggested approach, which is discussed in Section 5: Ethical findings.

Accessibility functions and considerations should be built into the data collection process

Accessibility has been a key consideration in the design of the questionnaires and the process to ensure that as many children and young people as possible can complete the survey independently, and in particular to increase inclusion of disabled children and children with additional needs.

In the questionnaires there would be:

• simple language

• the ability to increase the text size

• a magnification tool

• an accessible colour scheme with the ability to make the screen greyscale

• audio recordings for each question (headphones would be provided in schools)

• alignment with W3C Accessibility Guidelines 

• hover over function to give definitions of important terms

It is essential that every child with a disability or additional need is considered individually by the research team and the school. Firstly, their capacity to give informed consent and understand the survey would be considered to inform decisions on their participation. If it is agreed that a child can take part, it would be discussed if the child can complete the survey independently. Non-sensitive practice questions would also be included as part of the process, which could be completed with support. For pupils who have devices with readily programmed accessibility features, an online copy of the questionnaire would be available for use on their own device.

In a small number of cases, disabled children and children with additional needs would need assistance to complete the survey (such as with a school special educational need and disabilities (SEND) professional). To protect the anonymity of the child, the proposal is that both the assistant and the child complete the survey opposite each other so that the assistant is looking at the same questions and would be able to provide support to the child without seeing their answers. It would be explained to these participants beforehand, that if they disclose anything to their assistant during the survey, confidentiality may not be possible and onward referral may be required.

The sample for 11-to-16-year-olds should be drawn from publicly available school lists

We had originally planned to draw a sample from school censuses across the UK. We now plan to use the publicly available school lists. These are easier to access and are more up to date, as well as containing sufficient information to achieve a non-biased, representative, and generalisable sample (England and Wales, Scotland, and Northern Ireland).

Children aged 13 years and over should be able to solely consent for themselves to participate

Previously we suggested that the survey should adopt a procedure that allows parents to withdraw their child from participation. However, concerns raised in our public consultation, qualitative research with children and young people, and review of related legislations subsequently led to a review of this approach.

The key arguments for allowing children aged 13 years and over to provide their own consent without requiring parental permission to take part in the survey are the:

• legislation on the importance of children's rights such as articles 12 and 13 of the United Nations Convention on the Rights of the Child
• lowering of the age of sole consent in other countries researching child abuse, such as to the age of 12 years in Norway
• evidence from other research, which has shown children below the age of 11 years are cognitively developed to provide valuable information on their own health
• NSPCC Research Ethics Principles which outlines that children aged 16 to 17 years are considered able to consent for themselves, and those aged 8 to 15 years can also consent for themselves in a situation the researcher sees appropriate
• positive impact on the accuracy of the data, as parents who are perpetrators of abuse would not be able to remove their child from participation
• UK law around child comprehension to make decisions

The UK law we considered were:

• the premise in UK GDPR that children aged 13 years and over may lawfully provide their own consent for the processing of their personal data through an online information society service.
• The Data Protection Act 2018 which states that 'a person aged 12 or over is to be presumed to be of sufficient age and maturity to have such understanding, unless the contrary is shown', in relation to children in Scotland giving consent for the purposes of data protection legislation.
• the age of criminalisation, which is 10 in England and Wales, and 12 in Scotland, and means children of this age can be arrested or charged with a crime.

Based on this research, we are currently recommending that children aged 13 years and above should be able to consent for themselves without parental permission. Parents of children under the age of 13 years would be able to withdraw their child from taking part. To ensure the safeguarding of all participants is prioritised, there are numerous support options recommended (see the following subsection on anonymity). This recommendation received support from our internal legal team, NSPCC and National Statistician's Data Ethics Advisory Committee (NSDEC) in November 2023.

The survey should be collected completely anonymously

Previously, we concluded that a partially anonymous survey should be adopted. That is, that mandatory referrals would be made for respondents who were flagged as a safeguarding concern based on their responses. However, further research has shown that this would not be favourable or practical across most stakeholders. This is particularly true for children and young people with experiences of abuse, whose views we feel should be of upmost priority.

We are now recommending the survey should be collected completely anonymously with numerous opt-in support options available to all respondents before, throughout and after the survey.

There would also be a flagging safeguarding procedure, where support messages will be displayed if a response suggests that the child or young person is at risk (for example ongoing abuse, self-harm, and suicidality). These messages are discussed in the next subsection of this report.

A completely anonymous data collection would mean that:

• no personal identifiable information is collected in the survey

• no safeguarding referrals would be made on the basis of survey responses

• all respondents would be made aware of all available support options

• safeguarding action would only be taken when the child opts-in to safeguarding support

• onward referral may occur if abuse disclosures are made during follow-up, which children would be made aware of beforehand

This new recommendation received support and approval from our internal legal team and NSDEC in February 2024. The recent qualitative research found that the majority of child abuse professionals and children and young people with experiences of abuse recommended an anonymous survey (see Figure 1).

They thought non-anonymity and mandatory referrals would:

• create a sense of further loss of control for those who have experienced abuse

• cause distress and further negative consequences such as outcomes that are not best for children, or a child self-harming as a result of a referral

• risk the accuracy of responses with children being dishonest to avoid referrals

• reduce participation by those who have experienced abuse, again biasing the data

• overburden already strained systems, which would result in increased waiting times for those seeking support

Tailored safeguarding messages should be triggered as a result of survey responses

Although referrals would not be made based on survey responses, if a response does raise a safeguarding concern, the survey would prompt respondents to seek help through opt-in support options (discussed next in this report).

In the qualitative research, children and young people with experiences of childhood abuse and professionals said that if a child's responses indicate abuse, they should have the support options made clear to them. This is particularly important if this is the first time a child realises they have experienced abuse.

Safeguarding support messages would be shown if a child's responses indicate:

• current or historic abuse of any type (except community violence, criminal exploitation, corporal punishment, and sexual harassment)

• current self-harm, or historic or ongoing suicidality

• that they do not feel safe at school, work or home

• that they "don't feel comfortable to say" or are "not sure or don't remember" on any abuse question

The safeguarding message would:

• acknowledge and thank them for sharing their experience

• remind them that the survey is anonymous, and no action would be taken based on their responses

• be tailored depending on what they have reported in the survey

• encourage them to engage with a support organisation via a number, website or QR code

• encourage them to view other support options available to them (discussed in the next subsection)

• remind them that depending on the support option they choose, their ongoing discussion may not be anonymous

Respondents, parents and school staff should be offered a wide range of support

This section explains the current recommendation for support for pupils, young adults, school staff and parents or carers. Within the pilot we would develop and refine these proposals further. The aim is that these would be delivered through partnership with a child support organisation. 

Support for respondents

A pre-survey information session would be delivered giving detail on the survey, particularly on anonymity, an overview of topics and support options available.

During the survey there would be three buttons displayed continuously throughout; "Exit", "Pause My Survey" and "View Support Options". 

"Exit" would take the child away from the survey, first to the support options page then to an activity. "Pause My Survey" would do the same but prompt a return to the survey after three and five minutes.

The "View Support Options" page would take them to a page listing all available support options which include:

• the option to attend a drop-in support room in the school which is staffed by specialists from a support organisation

• an anonymous online chat staffed by specialists from a support organisation

• the option to talk in person with staff from the support organisation or a member of the school safeguarding team before the end of the day

• a callback request where they can input their name and request a call from a support organisation

• helpline support, an ONS specific website for the survey, and additional resources, which could be sent through to their phone or email

The dedicated drop-in room would be open on the day of the information session, the day before the survey, the day of the survey and two days following. Students could go before, during or after the survey should they require support, experience any distress, or need to take time before returning to normal classes. The anonymous chat and support helpline would be available 24 hours a day, 7 days a week, from the first information session until 60 days after survey fieldwork ends. The call back service would be bookable in 30-minute slots from 8am to 10pm seven days a week, until 14 days after survey fieldwork ends.

Upon completion of the survey, participants would be thanked for their participation and taken to the support options page.

Generic posters would be put around the school and the backs of toilet doors, two weeks before the survey takes place, and for six months after the survey, outlining child support services, support services numbers and links (including QR codes). Leaflets with the same information would also be provided to pupils.

The support provided to young people aged 16 to 25 years completing the survey online would be similar but exclude school specific support such as the drop-in room and in-person appointments.

Support for school staff and parents

An information session would be held separately for school staff and parents. This would provide information on the survey, anonymity, support options, how the data would be used and consent. An option for teachers to sign up for additional safeguarding training would also be provided.

The support helpline would also be available for school staff and parents with the same operating times mentioned in the previous subsection 'Support for respondents'.

Please see Figures 2 and 3 outlining how the whole safeguarding process would work in schools. A more detailed flowchart can be provided on request through emailing crimestatistics@ons.gov.uk.

Figure 2: School safeguarding procedure recommendation for before the survey takes place

Source: ONS Centre for Crime and Justice from the Office for National Statistics

Download this image Figure 2: School safeguarding procedure recommendation for before the survey takes place

.png (36.1 kB)

Figure 3: School safeguarding procedure recommendation for during the survey

Source: ONS Centre for Crime and Justice from the Office for National Statistics

Download this image Figure 3: School safeguarding procedure recommendation for during the survey

.png (46.5 kB)

There are further activities we will conduct before piloting the survey, mainly:

• validating existing legal advice on the proposals surrounding consent, anonymity, and safeguarding with an independent advisor

• conducting further school engagement to confirm our design and process, and how the current proposal would work operationally

• conducting further parental engagement on the design of the materials and information sessions, including their views around our consent proposals

• agreeing the sampling frame for the online survey of 16-to-25-year-olds through continuing engagement on acquisition, and use, of the Personal Demographics Service database (and equivalent across the devolved) for sampling purposes

We would also need to establish appropriate data protection arrangements for data collection and storage, which would comply with the Code of Practice for Statistics. Strict disclosure control methods would be applied when producing any data outputs and if the data is shared. This would involve strict access management internally and with external accredited researchers.

Next, we’ll be completing the outstanding research activities while planning the pilot.

The pilot would involve running the full survey in a small number of schools and with a small online sample. Post-survey qualitative interviews with respondents and non-respondents, parents and school staff would take place to get feedback on the survey.

The pilot would inform the outcome of the feasibility study, and if deemed successful, next steps regarding launching the full survey at a national level.

In November 2022, the Child Abuse Statistics Steering Group (CASSG) was established to provide advice and guidance on ONS child abuse statistics, with the initial focus on the child abuse prevalence survey feasibility study. This group includes a range of organisations and academic researchers. The organisations in the steering group are:

• Barnardo's

• Child Sexual Abuse Centre

• Children's Commissioner for England

• Department for Education (DFE)

• Department of Health and Social Care (DHSC)

• Department for Science, Innovation & Technology (DSIT)

• Home Office

• National Association for People Abused in Childhood (NAPAC)

• National Police Chief's Council, Norfolk Police

• National Society for the Prevention of Cruelty to Children (NSPCC)

• Northern Ireland Department for Health

• Ofsted

• Scottish Government

• Welsh Government