Safeguarding Children in Research: Ensuring the Protection and Inclusion of Children with Disabilities

In April 2025, I was honoured to speak at UNICEF's webinar series Beyond Barriers: Advancing Research for Children with Disabilities, in the session on protecting children with disabilities from violence and abuse.

My message was simple. Child safeguarding in research involving children with disabilities should not be a formality or a checkbox. It is a fundamental pillar of responsible research.

Safeguarding must be proactive, meaningful and built into every stage of the research process: from design and data collection to analysis and dissemination. When it is done right, it does not only protect participants. It also strengthens the quality and credibility of the research.

That applies to all research with children. This guide sets out what it looks like in practice, stage by stage, with particular attention to children with disabilities.

What does safeguarding children in research mean?

Safeguarding children in research means making sure the research itself, and everyone involved in it, does no harm to the children taking part, and that any concern raised during the study is acted on. It is not the same as ethics approval: approval reviews a protocol, while safeguarding runs from design to dissemination.

An ethics committee reviews what a team says it will do. Safeguarding is what the team, and the organisation behind it, actually does. The Ethical Research Involving Children (ERIC) initiative, developed through an international collaboration involving UNICEF's Office of Research–Innocenti, the Childwatch International Research Network, Southern Cross University in Australia and the University of Otago in New Zealand, provides internationally recognised ethical guidance for research involving children. Its International Charter emphasises that ethical responsibility extends beyond formal approval processes and requires researchers to continually reflect on their practices, decisions and responsibilities throughout the research process.

UNICEF's own procedure on ethical standards in research (2015 edition) is explicit. It requires protection protocols that provide safe environments for data collection, safeguard children from abusive or incompetent researchers, respond to safety concerns, and refer children to local support during and after the research. That is child safeguarding: a system around the research, not a paragraph in the ethics application.

Stage 1: Design

• Is the research necessary? The ERIC Charter asks researchers to assess whether involving the individual child is justified. If adults or existing data can answer the question, why should children carry the burden?
• Assess the risks: A risk assessment should cover the topic, setting, methods and the children involved, including risks at home or in the community that come from taking part.
• Involve children: A children's advisory group can test whether questions are understandable and methods feel safe. Child participation makes research safer as well as better.
• Budget for it: Accessible materials, interpreters, trained facilitators and follow-up support cost money. If they are not in the budget, they will not happen.

Example: The importance of safeguarding risk assessments in research involving children

Consider a research project exploring children's experiences of bullying in schools. The original research design involved collecting detailed information about participants, including their ages, school locations and personal experiences of bullying. However, insufficient consideration had been given to the safeguarding risks associated with collecting potentially identifiable and sensitive information from children, particularly the risk of exposing vulnerable children to further harm or emotional distress.

Following a safeguarding risk assessment, the research design was revised to minimise unnecessary data collection. Identifying details about the children and their schools were removed or replaced with pseudonyms, and specific locations were generalised to reduce the risk of identification.

The assessment also highlighted the importance of obtaining informed parental consent and children's age-appropriate assent, ensuring that children understood the purpose of the research, that participation was voluntary, and that they could withdraw without negative consequences. Clear confidentiality boundaries were established, including explaining when information might need to be shared to protect a child from harm. Appropriate safeguarding procedures were also introduced to respond to any disclosures of bullying, abuse or other concerns during the research.

This example illustrates how integrating safeguarding risk assessments into the research process goes beyond meeting ethical and procedural requirements. It can actively shape research methodology, minimise unnecessary data collection, strengthen informed consent processes and ensure that children's safety, rights and wellbeing remain central throughout the research.

Stage 2: Recruitment, consent and assent

Consent is not a signature at the start. It is an ongoing conversation.

• Respect evolving capacities: Seek the child's own assent or consent, alongside parental consent where required, in formats suited to their age and abilities. The ESRC guidance is clear that researchers should not assume children are incapable of providing consent simply because of their age.
• Think about gatekeepers: Parents, teachers and institutions control access. They can protect, but they can also pressure a child to take part, or exclude the children who most need to be heard.
• Make withdrawal real: Children must know they can stop participating without negative consequences. The ERIC Charter emphasises that indications of dissent or withdrawal must be respected, including dissent that is not spoken.

Stage 3: Data collection

This is where children meet the research team, so it is where much of the risk sits.

• Safer recruitment: Everyone with contact with children, including enumerators, interpreters and consultants, should undergo appropriate vetting and reference checks. The ESRC guidance makes clear that ensuring researchers are suitable is the employer's responsibility.
• A code of conduct: Clear, trained expectations: no unauthorised contact outside the study, no personal social media connections with participants and no photographs on personal phones.
• Visible, not alone: Interviews should take place where the child can be seen but not overheard, with arrangements adapted to the child's needs and the safeguarding risks.
• Online and remote research: The same safeguarding principles apply on video calls, messaging apps and online surveys. ERIC notes that online there may be no feasible way to trace a participant who raises a concern, so safety has to be designed in from the start.

Stage 4: Disclosure and referral

Every study with children must be ready for a disclosure of abuse, whatever its topic.

• Explain the limits of confidentiality up front: Before agreeing to take part, children and parents should know when information may need to be shared to protect a child from harm or meet legal reporting obligations.
• Name a safeguarding focal point: Every researcher should know exactly who to contact when a concern arises. Immediate safety concerns require urgent action, while other concerns should be reported promptly through established procedures.
• Map referral pathways before fieldwork: Who provides child protection, health and psychosocial support where the research takes place, and are those services safe and accessible? UNICEF's Beyond Barriers report found that referral plans and ethical protections are frequently absent, especially in low-resource or humanitarian contexts.

The research team is rarely the right body to investigate. Its job is to recognise, respond, refer and record, through a clear reporting and response procedure.

Stage 5: Data protection

• Collect only what you need, anonymise or pseudonymise early, and store data securely with access limited to authorised people.
• Treat recordings, photographs and case stories as sensitive. A quote with a school name and an unusual family detail can identify a child as surely as a name.
• Know the law that applies. For research in Europe, consider the requirements of the General Data Protection Regulation (GDPR), including provisions relevant to children's personal data.

Stage 6: Analysis and dissemination

• Check identifiability: UNICEF's procedure requires findings to be de-identified at the individual level, or aggregated, particularly where stigma or reprisals are possible. Small samples need special care.
• Images need their own consent: Agreeing to take part is not agreeing to appear in a report or campaign. Never use images that could identify a child who has disclosed harm.
• Share findings back with children, in a form they can understand.

Children with disabilities: The same standard, different tools

A 2022 systematic review in The Lancet Child & Adolescent Health, covering 98 studies, estimated that around one in three children with disabilities had experienced violence, and that they were about twice as likely to experience it as children without disabilities. Yet they are among the children most often left out of research.

Their rights are the same as every child's. Article 23 of the United Nations Convention on the Rights of the Child (UNCRC) recognises their right to a full and decent life and active participation in the community. Article 7 of the Convention on the Rights of Persons with Disabilities guarantees their right to express their views on all matters affecting them, on an equal basis with other children, with disability- and age-appropriate assistance.

In practice:

• Accessible communication: Easy Read materials, visual and tactile tools, sign language interpretation, alternative communication and time. Adapt consent processes; never skip them.
• Do not assume incapacity: Excluding a child because of their impairment or communication style is not protection. It produces evidence that does not reflect their lives.
• Caregivers as gatekeepers: Carers are often essential partners, but the Beyond Barriers report notes that unexamined gatekeeping can limit children's agency, and sometimes the adults closest to a child are the source of harm. Plan for both.
• Read non-verbal distress: Train researchers to recognise a child's signals and to stop.

The heightened risks experienced by children with disabilities should not be understood as an inherent consequence of disability. They are often shaped by environmental, institutional and communication barriers, unequal power relationships and failures to provide appropriate protection and support. Safeguarding should address these barriers without unnecessarily restricting children's rights to participate and be heard.

Why it makes research better

Safeguarding is sometimes treated as a cost, or as something that slows research down. In my experience, the opposite is true. Children who feel safe are better positioned to share their experiences openly. Researchers who know what to do when a child discloses harm are better equipped to navigate difficult topics. Accessible methods widen the evidence base instead of narrowing it. And research that has protected its participants is research that funders and communities can trust.

The example of research into children's experiences of bullying demonstrates this in practice. A safeguarding risk assessment did more than identify potential harm. It informed changes to the research methodology, reduced unnecessary data collection, strengthened consent procedures and established clearer arrangements for responding to safeguarding concerns. These changes strengthened both the protection of participating children and the integrity of the research.

It is also a matter of organisational accountability. Organisations that commission or carry out research involving children have a responsibility to ensure that appropriate safeguarding arrangements are established, implemented and monitored throughout the research process. This includes clear safeguarding policies and procedures, defined roles and responsibilities, appropriate training and oversight, and effective mechanisms for reporting and responding to concerns. Safeguarding should therefore be embedded within organisational governance, research agreements and operational practices, rather than being left solely to individual researchers.

A checklist for research teams

Before fieldwork begins, can you answer yes to each of these?

• Involving children is necessary and justified, and we have a written risk assessment.
• Children have helped review our methods and materials where appropriate.
• Consent and assent materials are accessible to every child we plan to include.
• Everyone with contact with children is appropriately vetted, trained and bound by a code of conduct.
• Participants are told the limits of confidentiality before they agree.
• A named safeguarding focal point is available throughout fieldwork.
• Local referral pathways are mapped and written down.
• Data will be minimised, appropriately anonymised or pseudonymised, and stored securely.
• Images and case stories have separate consent.
• We have a plan to share findings back with children.
• Safeguarding responsibilities are clearly established across the organisations involved in the research.

Safeguarding must be embedded, not added on

Safeguarding is not an additional requirement to be addressed once a research methodology has been developed. It is a fundamental consideration that should inform the decisions made throughout the research process.

For children with disabilities, this means ensuring that research is both safe and inclusive, that appropriate support and accommodations are provided, and that safeguarding measures do not become unnecessary barriers to participation. Ultimately, the measure of responsible research is not only the quality of the evidence it produces, but also how children are treated throughout the process.

Safeguarding must be understood as an organisational responsibility, a continuous practice and a fundamental condition for ethical, inclusive and credible research.

For organisations seeking to embed these principles into their research practice, child safeguarding policies, organisational frameworks and risk assessment processes provide an important foundation. The same considerations are increasingly relevant to emerging areas of research and innovation, including children's participation in artificial intelligence design and the use of child rights impact assessments.

Frequently asked questions

Is ethics approval enough to safeguard children in research?

No. Ethics approval reviews a research protocol before a study begins. Safeguarding is the ongoing system of vetting, conduct rules, disclosure procedures, referral pathways and data protection that protects children throughout the study and after it.

What should a researcher do if a child discloses abuse?

Listen calmly, do not promise secrecy and do not investigate. Follow the established safeguarding reporting procedure, record relevant information accurately and arrange appropriate referral or protective action. Where a child is in immediate danger, urgent action is required in accordance with applicable procedures and legal obligations.

Do researchers working with children need a criminal record check?

Vetting rules differ by country and role. In the UK, the ESRC expects Disclosure and Barring Service clearance in relevant circumstances. A check is one layer of safer recruitment, alongside references, interviews, training and supervision.

Can children with disabilities consent to take part in research?

Many can, with the right support. Researchers should not assume incapacity; they should adapt consent and assent processes with accessible formats and communication support, respect children's evolving capacities and recognise non-verbal signs of dissent.

Who is responsible for safeguarding in commissioned research?

Both commissioning and implementing organisations have responsibilities. The commissioner should establish safeguarding expectations in the contract and exercise appropriate oversight, while the research team applies the necessary safeguards in practice.

Sources

1. Beyond Barriers: Advancing Research for Children with Disabilities, UNICEF Innocenti, June 2025 (report of the April to May 2025 webinar series; Annex B covers the 24 April 2025 webinar).
2. Ethical Research Involving Children (ERIC) compendium, Graham, Powell, Taylor, Anderson and Fitzgerald, UNICEF Office of Research, Innocenti, Florence, 2013.
3. Ethical Research Involving Children (ERIC), Southern Cross University, UNICEF Innocenti, Childwatch International and University of Otago, accessed September 2026.
4. UNICEF Procedure for Ethical Standards in Research, Evaluation, Data Collection and Analysis, UNICEF, CF/PD/DRP/2015-001, effective 1 April 2015, section viii.
5. Research ethics: research with children and young people, Economic and Social Research Council (UKRI), last updated 12 May 2025.
6. Convention on the Rights of the Child, United Nations, 1989, Articles 3, 12 and 23.
7. Convention on the Rights of Persons with Disabilities, Article 7: Children with disabilities, United Nations, 2006.
8. Global estimates of violence against children with disabilities: an updated systematic review and meta-analysis, Fang, Cerna-Turoff, Zhang, Lu, Lachman and Barlow, The Lancet Child & Adolescent Health, 6(5), 2022.

Last reviewed November 6, 2026. Guidance cited here should be read alongside the ethics and safeguarding requirements of your own institution, funder and country.