Research with children: ethics, safety and avoiding harm

Topics: Safeguarding and child protection

What to consider when conducting research involving children

Doing research with children can help us understand what children think about the issues that are affecting them.

But any research with children must balance the aims of the research with the safety and wellbeing of the participants.

By providing the right support and knowing when to take appropriate action, researchers can ensure that children feel respected and can participate safely.

In particular, researchers need to consider:

  • how to obtain informed consent
  • how to manage the risk of harm to participants
  • what to do with the information gathered during the research.

Ethical guidance for researchers

We believe that it's vital to find out whether what we do makes a difference for families and children and we're committed to doing that in a way that avoids upsetting children or families. To ensure that all our research is ethical, we have detailed guidance and all our studies have to be approved by our Research Ethics Committee.

Our ethics committee is made up of experienced researchers from outside of the NSPCC who are able to review research proposals and give advice on reducing the risk of harm as much as possible. Read our Insights blog Embedding ethics into the research process to find out the benefits of having our own ethics committee.

> Download our Research Ethics Committee guidance (PDF)

Types of research methods involving children

There are 3 main ways of gathering information about children and the issues that affect their lives:

  • Asking children about their feelings, opinions and experiences
    Either in face-to-face interviews with children or by questionnaire. Data can also be collected by asking parents about their child's experiences.
  • Observing children's behaviour
    Using monitored experiments or activities or observing children in an uncontrolled environment to see how they react during specific situations.
  • Analysing information contained in files about children
    Reviewing information held in documents like social care case records, case reviews or school files.

Research needs to be based on clear ethical guidelines and robust measures must be in place to protect any children who are involved.

Recognising children's rights and value in the research process

Consent process

Consent is possibly the largest and most complicated issue for researchers hoping to involve children in a study. It's best practice to ensure that all research participants fully understand the research so that they can give their permission to be part of it - known as informed consent.

If a child's under 16 then they can't legally give consent themselves and a researcher should ask a parent or guardian for consent. Wherever possible, consent should also be sought from any children involved in the research. This is part of engaging children so they have opportunities to make their own decisions as well as ensuring they're willing participants.

Consent isn't a one-off process, but continues for as long as anyone is involved in the research. This means that a child who agrees to be part of a study (or their parent or guardian) can withdraw consent at any time.

Researchers should be aware of signs that a child doesn't want to participate. They may "say no" or "show no" through "non-response", "pulling away" or "ignoring" (Skånfors, 2009).

With adults, it's normal to use a consent form, or to seek verbal consent before the research starts. If the adult has the capacity to give consent and the research has been adequately explained, then ethically it's fine to proceed.

Obtaining retrospective consent

For some studies, most often observational studies, obtaining consent from all participants before you start may compromise the quality of the research.

In these cases, researchers need to have clear reason:

  • why prior consent wasn't possible
  • why the research still needs to go ahead
  • how consent will be sought after data collection.


You may want to thank participants for their time by offering some form of appreciation such as vouchers. However, it's important that this isn't set at a level which would risk skewing the results because people are taking part solely for the reward.

The level of incentive will vary depending on the individual circumstances, and will be a matter of judgement for researchers. One-off incentives must not be dependent upon completing participation in the research, so participants would still receive it even if they were to withdraw early from the study.

Being inclusive

Not every study can include a complete cross-section of society. But there are simple actions that can open up a research project to wider social and cultural group and improve the quality of the study. This might include translating research tools and supporting documents into other languages or providing an easy read version.

Researchers should also consider how, where and when the research takes place to ensure that specific groups aren't excluded from taking part.

It's good practice to have a version of the final report written specifically for the children who participated on the research. This will help show the outcomes of their participation and create a more positive and inclusive experience.

Identifying potential risks of harm

Managing risks and avoiding harm to participants and researchers

4 key things can influence whether a participant suffers harm during a research project:

  • individuals can find participating in research stressful, especially if they are vulnerable
  • hidden or suppressed feelings or memories may be uncovered
  • additional concerns may come up
  • participants may worry about what they have shared.

Researchers must access whether the research or the inclusion of child participants is justified. Research for the sake of research is not ethically sound especially when it may expose children to harm.

Unexpected topics

Qualitative research will often go into more depth than a qualitative approach, so there's more opportunity to discuss issues that the researcher or participant wasn't expecting. The interview schedule should be structured so that difficult topics are given enough time and aren't crammed in at the end. 

Interviewing friends and family

It can become difficult to remain impartial when interviewing different family members (Gorin et al, 2008). But researchers can't discuss the content from interviews with other family members without ethically compromising the research. Revealing information given in confidence could not only affect the participation in the research but could also place participants at risk.

Past experiences

When selecting participants, it's important to take personal histories into account. Researchers should consider how participants are likely to cope with being asked to talk about their past experiences. Being sensitive to past experiences can help minimise distress for example, avoiding interviewing about past abuse in the house where that abuse occurred.

Children who have been abused can be particularly vulnerable and researchers must take extra measures to protect both the children and themselves.

There must be a balance between the needs of the researcher and the need to protect children from any further harm. Having a clear ethics statement at the beginning of a project can help researchers with assessing potential risks to a child or stopping if abuse is disclosed (Gorin et al, 2008).

Non-disclosure of identity and personal information

A participant's personal information and their identity should remain confidential unless a child is at risk of harm.

Complaints procedure

It's good practice to have a complaints procedure when conducting any research. If children are involved, this should include a way for them to be able to make a complaint and be adequately represented. You should make the complaints procedure available when obtaining consent. 

What to do if a child discloses abuse

It's essential to have clear procedures to follow if a child says anything that indicates they or another child may be at risk of harm.

The researcher must include a confidentiality policy that clearly sets out the circumstances when a researcher can or should break confidentiality. The procedures should also include places where a researcher or child can access further support.

If a researcher suspects that a child might be at risk of harm then the research must be stopped until that child's safety is secured.


Gorin, S et al. (2008) Ethical challenges in conducting research with hard to reach families. Child Abuse Review, 17(4): 275-287.

Skånfors, L. (2009) Ethics in child research: children's agency and researchers' 'ethical radar' (PDF). Childhoods Today, 3(1): 1-22.