Image Alt

Ethical Research Involving Children

Child protection and confidentiality: Surveying children’s experiences of violence, abuse and neglect. By Lorraine Radford

You can download this ERIC case study as a pdf in English, français, español, 한국어, Türkçe and Bahasa Indonesia.

In 2008 the National Society for the Prevention of Cruelty to Children (NSPCC) decided to set up a comprehensive UK-wide study of the prevalence and impact of violence towards children and young people at home, in school and in the community. The study was the first ever in the UK to ask children and young people directly about all forms of violence experienced during childhood and within the past year. A UK-wide household survey was conducted in 2009 with 6196 participants, of whom 2160 were parents/carers of children under 11 years, 2275 were children and young people aged 11 to 17 years and 1761 were young adults aged 18 to 24 years. See www.nspcc.org.uk/childstudy for further details.

The ethical challenge:
Most ethical guidelines advise researchers to explain to participants that confidentiality is limited by child protection concerns. In the NSPCC study, age appropriate Computer Assisted Self Interviewing (CASI) methods were to be used to ask children and young people privately about their experiences. This meant that parents would not see the questions asked nor the answers given by their child. In CASI interviews, the interviewer also does not see the participant’s answers and will not know at the time whether a person has disclosed experiences of abuse. Adults interviewed this way are generally offered full anonymity. The UK does not have a mandatory approach to reporting child abuse. Would it be ethical to conduct a survey giving full anonymity to children reporting their own experiences of violence and abuse? What would happen if a child was in immediate danger of harm? In what circumstances would it be right to breach confidentiality? If confidentiality was to be limited by child protection responsibilities, how would the purpose of the survey and the limits of confidentiality be explained to children and young people and their parents without causing alarm or upset?

The research was conducted in a high income country with relatively well developed child protection and support and advice services. It had been agreed that all participants in the survey would be given a de-brief booklet that gave information on, and contact details for, relevant services so that parents, children and young people could access help or advice themselves if they wished to do so. But what if a child or young person was upset or wanted access to further help? How would they be able to tell us? What help could be offered directly to children and young people who wanted it and how could this be arranged safely in situations where the perpetrator of the violence might be a parent or another person living within the home?

Choices made:
Consultations were held with child protection and research experts, with parents and with children and young people, including young people who were known to be survivors of child abuse. These confirmed our view that offering complete anonymity would be unacceptable. Research experts recommended limiting confidentiality and developing within the survey a system to identify children likely to be in immediate danger, as well as those wanting further help or support. Young survivors thought it was important that the young person involved should be given a choice of what to do. If the young person did not want to report abuse to authorities at that time then they should not be pressurised or have action taken against their will. Questions were therefore included in the survey at key points to ask if the young person had felt upset or wanted to talk in confidence with a person who knew about keeping children safe.

An alert red flag system and review process was developed so that if a young person asked for support or if there was a combination of answers in the survey indicating a possible immediate risk, a flag was activated when the interview was uploaded onto the data system. This meant that the interviewer and any other person in the household would not know at the time an alert had been activated, only the NSPCC research team would be told about the alert. In case further action was required, in red flag cases participants’ contact details were not disaggregated from survey answers until after review. Strict protocols were agreed with child protection experts for managing the alert, review and referral processes. All interviews that were red-flagged, or where the respondent had expressed the wish to talk to a professional, were sent to the NSPCC on a daily basis and reviewed by two members of the research team. Cases requiring further action were passed on the very same day to ChildLine counsellors or to the NSPCC child protection services to follow-up.

The thresholds for raising red flag reviews were agreed with child protection experts within ChildLine. Some key issues considered included: the severity and frequency of the abuse, whether or not the child had suffered life threatening injury or rape, whether the perpetrator was likely to be abusing other children; any self-harming or suicidal intent; whether the child or young person already had access to help and support; the child or young person’s wishes and feelings; potential for a referral against the child’s wishes to help or to pose a further threat to the child’s safety.

Parents and children were told before agreeing to the interview that the survey was about child safety and victimization. Following advice from consultations, a show card listing the topics covered, including sensitive issues such as sexual abuse, was given to participants when negotiating consent. Participants were also told that their answers would be anonymous unless they gave information that suggested a child was in immediate danger. Children and young people were reminded on screen during the survey that they could choose not to answer and skip questions if they wanted to. Telling parents about the limits of confidentiality did not lower the response rate for the survey. It was 60.4% which compares favourably with other household surveys at the time conducted in the UK. In total 85 of the 191 reviewed cases were referred on, 35 to ChildLine, 44 to an independent counsellor and six to the NSPCC Helpline. In all but four cases the information was passed on with the agreement of the research participant.

Participants were asked during the interview how they felt about taking part. Of the red-flagged participants, 103 (over 54%) said that taking part in the survey had been very or extremely worthwhile; 38 (just under 20%) said they had been upset by the survey (33 young people and five caregivers). Of the 33 young people who reported being upset, 27 also said that taking part in the survey had been at least quite worthwhile. Our experience of conducting this research confirms that young people want to take part and want to express a view about their own experiences of violence, even when this may be upsetting.

Reflexive questions/considerations:

  • What might be the benefits as well the risks of harm in asking children and young people about their experiences of violence and abuse in the context in which you plan to conduct your research?
  • How could you best use the expertise of young survivors and child protection professionals to inform your approach to child protection and confidentiality in research?
  • This research was done in the context of a high income country with relatively well developed child protection services. How would you go about identifying appropriate sources of support and help for children and young people when conducting similar research in contexts where services are limited and public attitudes to victims of violence and abuse are hostile?
  • If sources of support are very limited, should funders of research be expected to provide some additional resources to help children directly?
  • Contributed by: Lorraine Radford, Professor of Social Policy and Social Work, University of Central Lancashire, Preston, UK.

Post a Comment