Introduction

The UN Convention on the Rights of the Child was ratified in 1989. One of the cornerstones of the convention is the child's right to participation. Article 12 of the Convention, also called the "democracy article" states that "States Parties shall assure to the child who is capable of forming his or her own views the right to express those views freely in all matters affecting the child, the views of the child being given due weight in accordance with the age and maturity of the child." This wording is often quoted in the literature on children's participation in research, as an argument in favour of children having the right to participate in research in order to present their views. In parallel with this are arguments regarding competence; that children have the ability to communicate adequate and credible knowledge, in addition to having important and unique information to provide about their lives. This knowledge provides a different picture of childhood and growing up than that provided by adults.

The research ethics issues that arise in relation to including children in research can be grouped under a few headlines: consent and competence to grant consent, benefit and harm associated with research, and the consideration of confidentiality. I will discuss these issues in the following.

Consent and competence to grant consent

Even though it is agreed that children make valuable contributions to research, it does not automatically follow from this that children have the competence to grant consent to participation in research on their own behalf. This primarily applies to legal competence, which children generally do not possess until they come of age. The general rule for minors is that their parents or guardians must grant consent, and that children and adolescents then agree or refuse to participate.

At the same time, legal competence cannot be seen as separate from the ability to understand what the research is about. The question regarding the ability of children and adolescents to grant informed consent to participation in research is therefore important, especially in relation to the focus in the Convention on the Rights of the Child on participation and the current more general emphasis on user participation. This means that we cannot avoid discussing how old children must be before they can grant consent on their own behalf. If parents or guardians grant consent, in principle there is no lower age limit for participation, as long as the consent is based on real information and the usual ethical weighing of benefits, harm and confidentiality has been conducted.

The issue of how old children should be – or how young they can be – is primarily related to the cognitive development of children. Knowledge development during the past few decades clearly shows that children's cognitive competence is greater than previously thought, and that they usually reason like adults from early in their teens. From this perspective, it should be possible for children to grant consent on their own behalf from the age of 13–14, for example. Depending on the subject of the project, children as young as 12 may grant consent on their own behalf, without the parents necessarily being informed.

However, children's participation in research can also be viewed as a matter of rights. It is then worth comparing the provisions that govern participation in research with other legislation on children's participation. The Patients' Rights Act of 1999 stipulates that children shall be heard in relation to treatment from the age of 12, while the Child Welfare Act stipulates that children shall be heard from the age of 7 on important matters of conflict like assumption of care and visitation. Children cannot make decisions, but they must be asked for their opinion, and these views will be assigned importance commensurate with their age and level of maturity. This does not necessarily mean that children will be given independent competence to grant consent to participation in research from the age of 7, for example. The provisions regarding participation can equally serve as arguments that the parents or guardians should be positive towards their children participating in research, and should encourage them to do this, even when they are very young.

In other words, it is a matter of a relationship between legal competence to grant consent, children's ability to grant consent on their own behalf and parents or guardians' understanding that children's participation in research can be important and useful. The fact that society today encourages more participation from children and adolescents both in general and in special situations is an argument in favour of children and adolescents also participating more in research. Research findings are also important to children and adolescents ' everyday lives. However, the emphasis on participation does not automatically mean that children should have independent competence to grant consent to participation in research. On the other hand, the new focus on participation means that parents or guardians should ask children whether they want to participate before they decide whether they want to grant consent on their children's behalf or not – and maybe also that parents or guardians will become more open to letting children decide whether they want to participate, even though the adults themselves do not see the immediate benefits of participation.

A third important aspect that must be brought into the discussion on children's options for granting consent on their own behalf is the view of children as a vulnerable group. NESH's Guidelines for research ethics in the social sciences, law and the humanities (NESH 2006) state, for example in article 12, that children are entitled to special protection that should be commensurate with their age and needs. The same point is made in the Declaration of Helsinki, which regulates medical research.

Weighing of benefit and harm associated with children's participation in research

There thus appears to be a contrast between the relatively strict rules regarding children's participation in research, linked to the requirement of parental consent, and the far more participation-oriented rules that govern children's participation in other areas of society, as well as important decisions regarding them. Is it more harmful for children to participate in research, so that they must be cared for particularly closely, or is it simply less important?

Benefit and harm must always be weighed in relation to the topic when children participate in research. In general, we can say that the more harmful or unpleasant the research, the greater the benefits the researcher must explain to the participant. Participation in social science research on children and adolescents is seldom very harmful, but doubts will nevertheless arise that may either indicate that the research should not be carried out as intended or that special measures should be implemented to limit any harm.

One example of such considerations is questions to children and adolescents that lead them to give so-called "self-incriminating" answers. With questionnaires, one can for example ask the participants whether they have committed crimes or used illegal intoxicants. The participants might find it troubling to answer these questions, even though no one else will know who answered in the affirmative. One option, which can also be used if someone feels uncomfortable because they are asked about violence and abuse, is to contact the school health service and ask them to be on standby or to provide the phone number of for example the Norwegian Red Cross hotline on the questionnaire. But is asking such questions too ethically problematic?

And what happens if researchers receive this type of information when interviewing children and adolescents? The information may be given because a relationship of trust has arisen between the researcher and the research subject, and the latter might want help with his/her problems. Then it may be an advantage that the researcher knows how to get assistance, and can help the child or adolescent receive it.

Research on children normally does not trigger this type of ethical dilemma. It is rather a matter of potential discomfort because children and adolescents are answering questionnaires, or are interviewed about problems they may have. The discomfort may then be due to the problem coming to the fore, in a setting in which the person who receives the information has no responsibility for doing anything about these problems. I believe that the best solution must be to try to minimise the discomfort by knowing where the participant can go or encourage them to contact the support services or adults that they trust. Perhaps it is best not to expose children and adolescents to this risk?

So far I have mainly discussed harm and children's participation in research. Surely there are benefits associated with children and adolescents' participation? A dilemma in social science research is that the results seldom benefit the participants directly. If for example a school study leads to an improved learning environment, the beneficiaries will primarily be those who come later. A number of children and adolescents who participate in research also point out that this can be considered a benefit. However, in line with the rights perspective I discussed initially, an argument can be made for participation in research being a way to exercise participation rights – and it being useful because it can reinforce the feeling of being an individual worth listening to. When children and adolescents are asked about their view on participation, they generally agree that they want to speak their minds and be heard, even though they understand that they are not in charge (Alderson and Morrow, 2004). These perspectives can strengthen the position of the benefit aspect of the assessment that must be made, without providing a false sense of security for researchers.

Confidentiality

The consideration of participants' confidentiality is a central principle of research ethics, regardless of age. Here I will briefly touch on the matter of protecting children and adolescents' confidentiality in relation to their parents. There may be ethical challenges associated with this issue. On the one hand, the parents may want a 'report' on what their child has answered. On the other hand, the child may not want their parents to know anything. In that event, greatest consideration must be given to the child as informant. If the researcher has spoken to the child or adolescent on their own, it may be a good idea to talk to him or her in advance about what information should be passed on, and to respect their wishes.

But what if the researcher receives information about violence and abuse, and considers this information so serious that the researcher's obligation to notify the child welfare service is triggered? Discussing the matter with the parents may at worst lead to further abuse of the child or that the child is pressured into retracting his/her statements if the child welfare service contacts the family. In this case, there is a risk of the research causing additional harm to the child as a result of having to set aside the consideration of confidentiality. On the other hand, neglecting to file a report with the authorities if the researcher has reason to believe that a child is a victim of serious neglect or abuse is a large ethical responsibility. Occasionally researchers attempt to resolve the dilemma by giving parents and children advance warning that certain situations trigger an obligation to notify the child welfare service – particularly when the topic of research is violence and abuse. It is worse if the researcher receives this information unexpectedly. Even then there is an obligation to notify the authorities in certain cases.

In this context it is interesting to learn what children and adolescents think about the obligation of researchers to report serious situations. US researcher Celia Fisher (1994) studied this phenomenon, and asked teenagers if they thought researchers should notify the school, the parents or the support services of e.g. attempted suicide, drug/alcohol abuse or neglect. She found that the older the adolescents, the more they wanted to resolve matters themselves, and the matter had to be more serious before they found it acceptable that the researcher notified anyone at all. In other words, the adolescents felt, albeit with varying premises, that adults should be able to notify others in order to provide help to those who need it.

Conclusion

This article begins and ends with children's rights, not least because several authors see ethical considerations associated with research on children and children's rights as interrelated. Nancy Bell (2008) points out that children's rights exist when research interests and children's everyday lives intersect, and they continue throughout the research process and beyond. An ethical imperative to treat children as moral actors with their own rights requires that researchers acknowledge children as participants in research, and consequently also with rights as participants in the same research. This is why rethinking the ethical principles that govern research with children and adolescents may be a way to go; to develop guidelines where the consideration of human rights is integral, and to promote a rights-based approach to research on children.

This article has been translated from Norwegian by Cristina Pulido Ulvang, Akasie språktjenester AS.