Available via license: CC BY-SA
Content may be subject to copyright.
Journal of Comparative Social Work 2013/2
!
Article:
“You child is just wonderful”: On
ethics and access in research with
children
by
Sara Eldén
PhD in Sociology, postdoctoral researcher
Department of Sociology,
Lund University, Sweden
Journal of Comparative Social Work 2013/2
!
!
2!
Abstract(
(
The need to conduct research not only on children, but also with children, has gained
almost complete acceptance in the social sciences in recent decades. The capturing
of the agentic voices of children has resulted in new perspectives on the social world,
as well as creating an awareness of the necessity of new ways of doing research.
Not least, children’s involvement in research has become fraught with ethical issues,
visible in the codes and guidelines of research organizations, in addition to being in
some contexts – such as Sweden – legally binding Ethical Review Acts. Within the
ethical guidelines, there is often a tension between enabling children’s voices to be
heard, while at the same time protecting them from harm. The “need-for-protection”
argument is most visible in the, in most ethical guidelines, recommendations – and
sometimes requirements – for informed consent from adult custodians: Access to
children’s voices is often dependent on an adult custodian’s approval.
This article discuss the consequences of the ambiguous view of children in the
ethical guidelines – the ambitions to “give voice” while also “protect” – with a point of
departure taken in the Swedish context, and in an actual research process of a
project on children and care. It shows how the regulation of informed consent through
the parent compromises the ideals of child-centred research; firstly, by limiting the
child’s possibility to opt in to research; secondly, by affecting the relationship
between researcher, child and parent in the research encounter; and thirdly, by
jeopardizing the child’s right to confidentiality. The author argues that we should view
not only the child’s but also the adult’s consent as a “continuous negotiation” and
discusses strategies for strengthening the child’s right to opt in and participate in
research on equal terms.
Journal of Comparative Social Work 2013/2
!
!
3!
Introduction
The need to conduct research not only on children, but also with children, has gained
almost complete acceptance in the social sciences in recent decades (Mayall,
2002:121). Together with broader societal calls for “the rights of the child”, the growth
and acceptance of what is often called the “New Childhood Studies” perspective
have all contributed to new views of children as socially competent actors with unique
and important voices that need to be listened to (CRC, 1989; Christensen and
James, 2008). At the same time, children’s involvement in research has become an
ethical issue fraught with difficulties, which is visible in the codes and guidelines
developed by research organizations, as well as in ethical review acts. However, the
ambition of “capturing the agentic voice” in New Childhood Studies is not necessarily
compatible with the more ambiguous aims of the ethical regulations, thus enabling
children’s voices to be heard, while at the same time protecting them from harm. The
“need-for-protection” argument is most obvious in the, in most ethical guidelines,
recommendations – and sometimes requirements – for informed consent from adult
custodians: Access to children’s voices is often dependent on an adult custodian’s –
most often parental – approval (Coyne, 2010). The ambiguous aims are especially
tricky in social science research on children in intimate contexts, such as the family,
in which views on the integrity of the “private sphere” are at work.
In this article, I discuss how the particular case of the Swedish Ethical Review Act
works in an actual research encounter with the aim of doing research with children.i
In contrast to other national contexts, in which ethical guidelines of different
organizations and funders can differ and are seen more as recommendations, the
Ethical Review Act of Sweden has been a binding law since 2003 that researchers
are obliged to follow. The Swedish case is also an act in which the ambiguity of
“giving voice” and “protect” is very apparent, particularly in its stressed requirement
for parental consent when doing research with children. While the need for parental
consent is not as strong in other national contexts, it is increasingly becoming a
recommendation and a practice (Davies, 2008). Thus, given its obligation, the
Swedish case and its consequences for research practice are of particular interest in
the international debates on ethics and child research. Furthermore, the differences
in the possibilities of doing research caused by differing ethical requirements need to
Journal of Comparative Social Work 2013/2
!
!
4!
be attended to when comparing studies from different national contexts, or when
doing a comparative study involving children.
I argue in the following that the contrasting views of children in research – the “giving
voice” and the “protect” arguments, and most importantly the requirement for
informed consent through adults, have consequences. Firstly, in relation to what
voices of children are accessed and thus present in research; and secondly, on the
actual research encounter and dilemmas facing the researcher in attempting to take
the incentive of taking the child’s perspective seriously, while at the same time
following the ethical rules and ensuring informed consent through adult custodians.
Lastly, I will illustrate my arguments using examples from a research project on
children’s narratives on care.
Contrasting ambitions: A children’s voice ethos and the ethical review acts
Research involving children has changed fundamentally in recent decades.
Longstanding ideas of children as either outside the scope of social science or (at
best) only interesting in their roles as “adults-in-the-making” have been challenged by
the developing “New Childhood Studies” framework, in which the child as a social
being in him/herself is in focus (Alanen, 1992; Brannen, 1999; James, 2007; Näsman
and others, 2008; Christensen and James, 2008; Mayall, 2002; Smart and others,
2001). This emerging view of children brings new theoretical, methodological and
ethical questions to the fore. It also represents a new epistemological interest in
children’s knowledge, “prioritizing the idea that children have subjective worlds worth
researching” (Wyness, 2006: 185), thereby assigning children a capability and
competence that was not noticed before (Alanen, 1992). The change is often
described as a shift from working “on” children to working “with” children (Mayall,
2002: 121). However, the idea of capturing children’s voices is by no means
unproblematic (e.g. James, 2007; Komulainen, 2007; Lewis, 2010; Spyrou, 2011;
Uprichard, 2010). As James (2007) argues, there is a tendency to slip into arguments
of childhood research as somehow representing “authentic” voices of children, which
risks simplifying and reducing the complexity of children as social actors. Moreover,
others have pointed towards the risk of using “children’s needs” and “children’s
competences” as excuses for political reforms (Lindgren and Halldén, 2001; Sandin,
Journal of Comparative Social Work 2013/2
!
!
5!
2012), not the least of which is problematic in light of the cutbacks in the Welfare
State (Wihstutz, 2011). While well aware of these problems, I still want to argue for
the necessity of continuing the inclusion of “children’s voices” in research, but doing
so with a critical awareness of the production of voices in the research process and
the situatedness and complexity of “voices” – be it adult’s or children’s voices (which
I have discussed in other writings, e.g. Eldén, 2013; Eldén, in press; see also Alanen,
1992; Spyrou, 2011; Smart, 2009).
The voices of children – how to best “capture” them, what methods to use and how to
interpret them – have thus been a central focus in the new childhood studies
framework. The view of children as competent agents has also contributed to the
acknowledgement of children’s ability to make their own decisions on participating in
research. The ethical guidelines of various research and childhood organizations –
such as the BSA (British Sociological Association) and the NCB (National Children’s
Bureau) – have incorporated the “child competence” argument by arguing for the
necessity of informed consent from children and for the child’s right to opt out of the
research. However, in guidelines the “competent child” assumption is combined with
one of “children in need of protection” through recommendations for adult guardian
consent as “good practice”, in addition to the child’s consent (Balen and others, 2006;
Davies, 2008).
In the case of Sweden, the ambiguity of “give voice” versus “protect” is highly
apparent. Here, all research involving people has been regulated by the Ethical
Review Act since 2003 (2003:460). The Act stipulates which research has to be
examined by ethical review committees (which is almost all research involving
children), but also sets out more general regulations on informed consent,
confidentiality and so on. For research involving minors, the Ethical Review Act
states that information about the research projectii and consent to participate must be
obtained from “the subject’s guardians”. However, the Act also stresses the
importance of informing participating children about the research “as far as possible”
and – importantly – giving them the right to opt out:
In [...] cases when the subject of the research has not attained the age of 18,
the subject’s guardians are to be informed and their consent is to be acquired
Journal of Comparative Social Work 2013/2
!
!
6!
[…]. As far as possible, however, the research persons themselves are to be
informed about the research. Even if the consent of guardians has been
obtained, research may not be carried out if a person who is the subject of the
research is younger than 15 years of age, understands what it entails for his or
her part and objects to it being carried out. (EPL 18§)
Childhood researchers usually take the appeal for informed consent from children
very seriously, and there is an ongoing discussion as to how this consent is acquired
(see e.g. Alderson, 1995; Christensen and Prout, 2002; Cocks, 2006; Eriksson and
Näsman, 2012; Gallagher and others, 2009). Strategies to make information about
the research project accessible (pictures, drawings, etc.) have been developed; it has
further been argued that consent from children should not only be obtained when
instigating the research, but should be an ongoing negotiation throughout the
research encounter, e.g. by being attentive to the child and giving him/her several
“opportunities” to opt out during the research process, such as suggesting a break or
asking whether the child would rather do something else (Eriksson and Näsman,
2012). The methods chosen can also themselves give the child opportunities to opt
out or to drop a sensitive subject (Eldén, 2013).
However, while there has been considerable discussion on the importance of giving
children ways to opt out of research, there have been fewer discussions on how to
enable children to opt in to research. The most obvious – and not uncommon –
situation where this question arises is when a child expresses interest in taking part
in a particular research project, but the adult custodian – most often the parent –
declines consent. Thus, following the ethical review acts or guidelines, the researcher
must (or should) in effect deny the child the possibility of participation (cf. Mason and
Falloon, 2001; Smart and others, 2001; Coyne, 2010). According to Wyness (2006),
these instances reveal the enduring adultcentrism of the ethical review acts and
guidelines. Here, children are on the one hand “ontologically established as full
members of society” as “agents who can choose to take part in research”, and whose
“status as agents assures them of a place within […] a ‘hierarchy of gatekeepers’”,
but are simultaneously “located firmly on the bottom rung” within this hierarchy, with
parents (and sometimes other adults) being the ones granting access (Wyness,
2006:195). An unavoidable effect of this is a breach of the “child’s right to a voice”, as
Journal of Comparative Social Work 2013/2
!
!
7!
not all children have the same possibility to make themselves heard in research.
Moreover, this calls for caution in how researchers choose to represent “children’s
voices” in research: It is important not to forget those voices that for different reasons
– the absence of adult consent being one – are not heard.
In addition, with adult custodians being the gatekeepers for children’s participation, I
would argue that discussion of the “work” of the researcher in ensuring the consent of
the child should be complemented by discussions of the work of ensuring the
consent of the adult custodian. This is an often neglected but (as in the case with the
child) ongoing and sometimes delicate negotiation, a pursuit that can get in the way
of doing child-centred research.iii Furthermore, the requirement for adult consent
prescribed in the ethical review acts, together with the social situation of the child as
typically dependent on adults, can jeopardize the child’s right to confidentiality in the
research process. This is a particularly difficult aspect when conducting research on
children on – and in – intimate settings such as the family.
Research with children in intimate settings
Smart and others (2001) argue that the dominant views of children have previously
precluded us from letting children speak for themselves on matters concerning
family. Parents were presumed to be able to talk for the child on issues concerning
family life, and talking to children was viewed as inappropriate, as an intrusion into a
private sphere of life (Smart and others, 2001:10; cf. Wyness, 2006:187). Smart and
others argue that there is an ethical concern on the part of the family unit at work
here, that “speaking to children about personal family matters will somehow unravel,
undermine or subvert the family and damage its integrity” (2001: 10). This, they
argue, has contributed to the marginalization of children in family sociology and, as
Wyness (2006) puts it, resulted in the rather peculiar imbalance in research on
children. While the family remains a dominant setting for children in Western
societies, given the professional institutions’ greater openness to public gaze, “we
probably know much more about children in schools than children within the home”
(Wyness, 2006: 187).
Nonetheless, parallel to the development of childhood studies, there have been
Journal of Comparative Social Work 2013/2
!
!
8!
important and groundbreaking changes in the field of family sociology. The move
away from the functionalist paradigm, with its fixed institutional definitions of “the
family”, towards a focus on family as “doing”, as “fluid networks of personal
relationships and practices” (Smart and others, 2001: 17; Morgan, 1996; Silva and
Smart, 1999), has allowed change and diversity to enter the paradigm. Most
importantly, moving away from the assumption of a family as an integrated unit
enabled an interest in the perspectives of individual family members – including
children (Brannen and others, 2000; Mason and Tipper, 2008): .
Within this formulation, then, children need no longer be invisible; they emerge
as fully fledged family members, actively engaged in negotiating their own
family practices and relationships. They no longer just belong to families; as
reflexive agents of their own lives, they are part of the creation of families.
(Smart and others, 2001: 18)
Together, the new developments in both childhood and family studies have
stimulated studies on children’s own experiences of family life and related areas.
However, this does not mean that ideas about children not being able to correctly
report on family matters have disappeared, nor has the concern of parents about
letting children speak to researchers. Moreover, following the “good practice” – or in
the case of Sweden, the law – of informed consent from parents, these are matters
researchers have to deal with in gaining access to children’s voices. To further
explore this, I will discuss below my ongoing project on children’s narratives of care
in relation to the following three aspects. Firstly, the question of children’s
possibilities of opting in; secondly, the “juggling act” performed by the researcher in
trying to ensure consent through the parent and the child in the research encounter;
and thirdly, the problems of ensuring children the right to confidentiality.
Research in practice
The aim of the current research project is to capture relations of care around children
from the child’s point of view (“Relations of care are beyond the ‘family’”; Swedish
Council for Working Life and Social Research, FAS 2010-0505). In line with recent
developments in family studies, the project seeks to decentre the previous
dominance of the nuclear family imaginary (cf. van Every, 1999; Roseneil and
Journal of Comparative Social Work 2013/2
!
!
9!
Budgeon, 2004; Smart, 2007). It starts with an open approach to where, in what
relationships and under what circumstances care for children takes place. Thus,
instead of assuming in advance the precedence of the nuclear family and the
parental relationship as the hub of care, the project sets this out as an empirical
question to test (for a presentation of the results of the study, see Eldén, in press).
The qualitative methods developed for the project have therefore been carefully
designed so as not to prioritize discussions about family or parents at the outset, but
rather to allow the child a greater freedom when reflecting upon who the important
actors are in caring for and about him or her. In short, I conduct an interview with the
child using two drawing methods. The first concentrates on the “doing” of the care,
with the child’s everyday activities and the everyday caring situations in focus in a
“draw-your-day” exercise; the second focuses on relationships, in which the child is
asked to put people that care for and about him/her and/or that he/she cares for and
about in “concentric circles of closeness” (the methods developed for the project and
their potential for giving “voice” to children is discussed in Eldén, 2013).
The aim of the project was to get at a diversity of experiences, to investigate and
show different ways of “doing” care in children’s lives, as well as different ways of
narrating care on the part of the children. Twenty-three children between the ages 5
to 12 years, and from different socioeconomic, ethnic and family backgrounds,iv took
part in the project. Contact with the children was established through their parents,v
who in turn were primarily contacted using the snowball technique (friends of friends
and one participating parent or child suggesting another).
In contacting the parents, the project was presented both verbally (over the
telephone or through personal contact) and in writing as a project on care for children
where I was interested in the many different ways in which children were cared for by
parents, “but also by grandparents, siblings, neighbours and nannies”, and that the
aim of the study was to “let children tell about care in their everyday lives”, since
there are a lot of research on adult’s views on care but few instances in which
“children themselves have had a say”. vi I also stressed the child’s right to
confidentiality, including in relation to the parents, verbally and in the informed
Journal of Comparative Social Work 2013/2
!
!
10!
consent form signed by the parents. The children were informed after the parents,
and offered both verbal and written information about the project (see discussion
below), and their consent to participate was ensured before beginning the interview
as well as continuously during the research encounter.
The choice to use the snowball technique in the project was in fact less of a choice
than a necessity that came out of the failure of the initial attempts to establish contact
with parents through schools, day care centres and care organizations. vii The
reasons for this failure could of course be many, though my impression (given the
experience of the contacts with the parents of the children who actually participated,
which is further discussed below) was that the topic of the research – care for
children – might have been seen as a sensitive issue by some parents. Therefore, to
grant permission for their children to participate, a close contact with me as a
researcher seemed to be required in order to establish trust. However, as I will show
below, the snowball technique turned out to be crucial in gaining access to children
whose voices would not have been otherwise heard.
1) Whose voice: Opting in – and being opted out
As stated above, the Swedish Ethical Review Act required consent from parents for a
child’s participation. In initiating the study, I interpreted this to mean that I had to talk
to the parents before approaching the child. As I subsequently learned, this is not
always the case, as children are sometimes approached and asked for consent
before their parents are contacted. In many ways, this latter approach could be
argued as strengthening the child’s position to opt into research: If the child states his
or her will to opt in, it might make it harder for the parent to refuse consent.
Nevertheless, there are pros and cons to both procedures.
On the negative side, the approach I chose gave parents complete control over their
child’s access to anything to do with the research project. In contacting parents, I was
firstly dependent on their interest in the project. Some parents declined access as
early as at the outset, stating that they did not think their child would be interested.
Secondly, I was dependent on their passing on the information to the child.
Information leaflets were sent to parents declaring an interest in the project, one
Journal of Comparative Social Work 2013/2
!
!
11!
directed to the parent and one (an easier version) to the child. My dependence on the
parent was still total at this stage: I needed the parent to pass on the information (the
leaflet, or in the case of smaller children by reading the leaflet) to the child, and of
course I could not influence the way the parent presented the project – as a
potentially “fun” enterprise for the child, or as something the parents did not like or
found interesting or important. When some parents told me that their child had not
expressed any interest, I had to accept their reason (although I sometimes suspected
that the child had not been asked).
On the positive side, the approach I chose did take protecting the child from possible
harm more seriously than the one that asked the child first. Asking the child first
would have given me an opportunity to present the project in a way that might have
made the child more interested in participating. However, in contrast, I could have put
the child in a position where he or she wanted to participate and then had to deal with
the refusal of the parent. Apart from disappointing the child, this could also place the
child in a position where he/she had to defend his or her willingness to take part to a
parent who declines to let the child participate. For example, in a project on children’s
conceptions of child abuse (Mason and Falloon, 2001), the ask-the-child-first
approach was used; the result was that some children who opted in were not allowed
to participate due to their guardian’s refusal of consent. I would argue that projects
that deal with family relationships – and particularly parent-child relationships –
indeed put children in the potentially difficult position of having to defend his/her
willingness to opt in to the parent, thereby reflecting the child’s subordinate position
in the family and society at large. And as a researcher, it is practically impossible to
know the consequences of a child’s expression of willingness to opt in when one is
no longer present.
Whether one chooses to ask the child or the parents first, there is an unavoidable
selection of voices, with the parents being the ultimate gatekeepers. However, I
would argue that there are ways in which researchers can strengthen the possibilities
of children’s agency in relation to opting into research that go beyond the simple
dichotomy of “first asking the child” or “first asking the parent”. This awareness came
Journal of Comparative Social Work 2013/2
!
!
12!
to me through an unexpected episode that occurred when I conducted interviews for
the projectviii :
Astor is a seven-year-old boy whose parents I did not contact about the project
– Astor contacted me. I was conducting an interview with another child – a girl
whom I had come into contact with through the snowball technique (I knew a
friend of her mother) – when the girl suddenly remembered that her friend also
wanted to participate. After the interview, the girl reminded her mother that
Astor wanted to take part in the project, and her mother informed me on how to
contact Astor. The mother explained that some days earlier they had spent time
with friends – among them Astor and his parents – and had come to talk about
my project. Astor became curious and declared to the entire group of adults and
children that he also wanted to participate, and he made the girl’s mother
promise that she would pass on the information to me.
The interview with Astor is one in which many difficult issues of care emerge: his
mixed feelings about his parents, his feelings of a lack of care regarding significant
adults and his longing for more caring persons in his everyday life. However, Astor’s
narrative also demonstrates his competence and ability to initiate and form new
caring relationships, an agency that one could argue is visible in the way he
approached me.
What I could learn from the story of Astor is that to enable and strengthen children’s
opportunities to opt into research, we might need to think in different ways to develop
new and less formal ways of approaching children. Snowball techniques are
sometimes frowned upon in research, being viewed as the “easy way out”, risking a
skewed representation of the research population. However, in this project, the
snowballing by the research participants themselves resulted in voices being raised
and heard that might not have been otherwise audible. Of course, we do not know
whether or not Astor’s parents would have consented to his participation had I asked
them directly. However, what was apparent through Astor’s-, the girl’s- and her
mother’s story was that the “contextual talk” about the project – the fact that there
had been many children and adults discussing it – had made Astor voice his
willingness to take part, which made it harder for the parents to reject his wish.
Making a research project into “a topic to talk about” within a context, especially
contexts breaching the isolated family and the adult-child relationship (e.g. by
Journal of Comparative Social Work 2013/2
!
!
13!
presenting the project in social activities for children, to groups of peers, in parent-
child activities in school and in afterschool care or sports activities), and informing
many different actors and on many levels at the same time, enhances the child’s
possibility to see the research as something he or she can react to (opt in or out of).
This makes the parental consent less of a “big issue” compared to the more formal
“ask-the-child-first” or “ask-the-adult-first” approaches.
Making it easier for children to “opt in” into research also puts a great responsibility
on the researcher to match up to the child’s expectations, both in regard to the
research encounter and to the output of the research. This is particularly the case
when the researcher invites the child to reflect upon and share his/her personal
stories.ix The ability to explain the purpose and implication of research in general,
and the research project at hand in particular, to the child is crucial (see e.g. Cocks,
2006), and just as in the case of consent, a continuous process during the research
encounter. Very few children are familiar with research, not least the amount of time
that passes from the interview to the actual reporting of the project, and that the
“outcome” of the research might be quite inaccessible to children. However, there
has been considerable efforts made recently in research projects involving children
and young people on developing new ways of “giving back” to the participants and
reporting in ways other than conventional academic writing, e.g. in leaflets with
accessible writings and images, or in websites.x
2) Juggling parents and children
Even so, the questions surrounding consent do not end when the parent gives
permission for the researcher to meet the child. On the contrary, I would argue that
parental consent is a continuous negotiation – just as with the children. Moreover, it
is a negotiation that often has to take place simultaneously as one negotiates the
child’s informed consent. And importantly, negotiating the consent of the parent can
hinder the aim of a child-centred research encounter.
This becomes most apparent in the time I spend with the child and parents before an
interview. This is a crucial time for every research encounter, and especially so when
Journal of Comparative Social Work 2013/2
!
!
14!
interviewing children (cf. Danby and Farrell, 2005). While often (and hopefully) being
a relaxed chat over coffee, this is the time when the child and the parents are
simultaneously checking me out: Am I OK, is the project OK, does the child really
want to take part and am I sufficiently trustworthy to talk with the child – preferably
out of earshot of the parents? However, given the requirement of the parent’s
consent prior to starting the actual interview with the child – preferably in the form of
a signed consent form – I have found that the juggling of my attention between the
child and the parent too often lands in favour of the latter.
It is remarkably easy to slip into an adult way of talking and acting at this stage.
Aiming for child-centred research means becoming aware of how often adults –
myself included – talk to each other as though the children were not there. In the
research encounter, I have felt obliged to develop strategies to counter this, e.g. by
trying to answer in a way that includes the child in the conversation while listening to
and confirming the talk of the parent. In many cases, however, this is difficult,
especially when the parent discusses “sensitive issues” such as the child’s problems
in school, the lack of involvement of an absent parent or the parent’s personal
problems. Some children take measures to deal with this on their own. For instance,
Amy, a lively and talkative eight-year-old living with her mother, makes sure that she
gets her space in my fieldwork:
I arrive at Amy’s and her mother’s place late on a Tuesday afternoon, the same
time that they themselves arrive home from afterschool care and work. Amy’s
mother starts cooking dinner and I help out. She starts to tell me about her and
Amy’s life, about the difficult situation they have been in for some time, about
her being made redundant, about the difficulty making ends meet and problems
with her ex-husband and the problems of solving the everyday care situation for
Amy. I listen and respond to the mother’s story, while simultaneously noting that
Amy is circling around me in the kitchen, trying to get my attention. I make some
reluctant attempts to include Amy in the conversation. After a while, Amy goes
to her room and returns with a board game, insisting that I stop helping her
mother and play with her instead. We start playing, still in the kitchen, and I
continue to listen to her mother, while also trying to direct my attention to Amy
and the board game. The mother then realizes that they have run out of milk
and asks Amy to run down to the local store. Amy looks at me inquiringly, and I
get the message – I offer to go with her. On our walk to the store I sense how
she is carefully checking me out, asking me questions about my life situation.
Journal of Comparative Social Work 2013/2
!
!
15!
She wants to hear my opinion on different things, and she soon starts to tell me
a little bit about herself, her situation and her concerns.
Of course, not all children are as proactive as Amy in making contact and space for
themselves, and in ensuring that their voices are heard. At work here is the very
understandable concern on the part of the parents: They are letting me spend time
alone with their child and want to assure themselves that I am OK. Of course, there is
no guarantee of researchers being “good” and responsible, as trust must always be
earned by the researcher. In addition, the parents’ worry needs to be seen in context.
Researchers have – historically and maybe even so today – been allied with
controlling institutions and authorities, and have had the power to intervene and
disrupt families, particularly in relation to families diverging from the white middle-
class “norm” (Donzelot, 1979; Skeggs, 1997; Sandin, 2012). With this in mind, the
parents’ scepticism towards the researcher is indeed well grounded.xi However, the
argument I want to make here is that the delicate task of assuring the parents that
“I’m OK” often means diverting from the child-centred ideal of attempting to show the
child that I’m there for them, to listen to their narratives irrespective of the views of
the parents. And also, I do think that we find elements of the adult-centric
assumptions of the parent’s right to control and to speak for their children on family
matters at work here. There is a worry that the child might say something that would
show them in a bad light, which indeed potentially challenges the privacy of the
home.
This latter point is very obvious in the juggling I sometimes have to engage in after
the interview. A fairly common reaction from the parents when the child and I have
finished is, “So…how did it go?”, which is followed by nervous laughter and
sometimes even, “Are you going to call the social services now?” Despite the fact
that I have stressed the child’s right to be anonymous (and that the parents have
signed an informed consent form stating this, in which I specifically point out that this
also applies in relation to parents), they quite often try to tease out something from
me. As a way of trying to assure them that nothing “bad” came up, my answer is
always a placatory one, “Oh no, we had a wonderful talk, your child is just wonderful”
(which is of course true), and then I try to direct the talk away from the interview.
Journal of Comparative Social Work 2013/2
!
!
16!
3) Confidentiality and eavesdropping
The post-interview juggling discussed above highlights the problem of ensuring and
maintaining the confidentiality of children participating in research. The absolute
standpoint of never breaching confidentiality is often rather difficult for the researcher
conducting research in family settings, particularly when parents are pushy and one
still needs them to be “positive” to grant the child’s continued participation.
The problems of granting the child confidentiality also emerge during the interview
with the child in the not uncommon eavesdropping by the parent. In theory, this is
straightforward: One simply states that one wishes to talk to the child in private,
behind closed doors, e.g. in the child’s room. In reality, one often has to compromise,
as it is sometimes not possible to conduct the interview out of the earshot of the
parents – because of space limitations (small flats, no doors), because the child does
not want to or because the parents choose to stay close by. As Wyness argues,
adults’ control of space means that they often have the power to regulate the child’s
privacy in the encounter with the researcher – “[p]rivacy is a common good to which
adults have access” and children have to negotiate access to this space (Wyness,
2006: 193). Again, as a researcher I have to take into account the very real concern
of the parent of me being “OK”: if I emphasize too strongly the wish to be alone with
the child, there is a risk of the parent withdrawing the child from the research. The
only solution in such cases is to compromise on the issue of privacy and carry out the
research, knowing that the parents are listening and therefore also compromising the
child’s right to confidentiality:
Anna, a 10-year-old girl, lives with her father in a three-bedroom flat. The rooms
are quite small, and to help make ends meet, Anna and her father have a
lodger who rents one of the rooms. During the interview with Anna, both the
father and the lodger are at home (the lodger is in her room). When getting
ready to start the interview after some initial chitchat, Anna’s father suggests
that we sit in the living room. Anna’s own room is very small and “very untidy”,
the father says with a knowing look. Since this seems to be Anna’s wish too
(she sits down on the couch straight away), I agree to this arrangement. During
the interview, the father moves around in the flat. At the beginning, he is in his
own bedroom, but after a while he sits in the kitchen, reading the newspaper.
There is no door between the kitchen and the living room, and I am not sure
Journal of Comparative Social Work 2013/2
!
!
17!
how much of Anna’s and my conversation he can hear, which makes me feel
uneasy. Nor am I sure whether Anna is as aware as I am that he might be
within earshot. Eventually, my worries are confirmed: When Anna tells me
rather late in the interview about her absent mother, the father appears from the
kitchen to “help out” and explain to me her whereabouts.
This incident highlights the difficulty of implementing the child’s right to confidentiality
in the research encounter, illustrating children’s lack of control over space and
privacy, and how this lack of control “spills over” onto the researcher in the research
encounter. I want to argue that as a researcher (especially when one is interviewing
children on potentially sensitive matters), one needs to develop strategies to deal
with this lack of control. When a private space for the research encounter is not
possible for some reason, the researcher needs to be extra attentive to the child’s
situation. If one suspects that a parent is listening in, the researcher can draw the
child’s attention to this (e.g. by saying something to the parent such as asking for
more coffee, or by directly pointing out that “mum’s in the next room”), thus
attempting to ensure that the child will not tell you something that will get them into
trouble afterwards. Of course, this means that some questions might not be
addressed in the interview and that the child might not share all of his or her
thoughts. But it is a necessary compromise, given that the safety of the child is
always paramount.xii
In addition, problems of confidentiality emerge in relation to the unavoidable giving-
away of the child as a result of the requirement for informed consent through adult
guardians. Adults participating in research can in most cases control who (if anyone)
knows of his/her participation, hence minimizing the risk that someone will recognize
their narratives. This is not the case for children. The fact that the parents will always
know that their child is participating in a certain research project means a huge
undertaking for the researcher to ensure that parents cannot identify the utterances
of their children. When the research – as in my case – is about family and personal
life, and furthermore is qualitative and seeks to reveal children’s complex narratives
of their everyday life, this is almost impossible. When deciding whether a certain
child’s narrative should be presented in the research publication, the researcher not
only has to change names, places and circumstances, but must also try to imagine
Journal of Comparative Social Work 2013/2
!
!
18!
whether the narrative could be identified by the parents and – if there is a risk of that
happening – if the narrative could in any way result in harm to the child if the parent
read it.
Conclusion
Including children’s voices in research is a very important and pressing – although
not uncomplicated – task facing the social sciences. It is also a task that raises new
questions, not least ethical ones. Taking the argument for child-centred research
seriously means critically examining whether our ways of carrying out research really
measure up to this ideal. In this article, I have shown how there are conflicting
assumptions about “the child” in the ethical guidelines and review acts – on the one
hand of children as agentic, and on the other in need of protection from – and thus
subordinated to – adult guardians. All of this impacts the possibilities of conducting
child-centred research. By looking at the case of Sweden, where the Ethical Review
Act clearly encompasses the ambiguity of “voice” versus “protection” through its
requirement of informed consent from both children and legal custodians, I have
highlighted the consequences of the requirement for parental consent in relation to
children’s possibilities of opting into research, as well as in the actual research
encounter – the “juggling” work of getting consent from both parents and children –
and also the (im)possibilities of granting children full confidentiality. What becomes
obvious in this discussion is that the protective aim of an ethical review act can
impede both the researcher’s and children’s actions in the research encounter. In
turn, this can affect the quality of the research produced. I am not arguing for a
removal of the requirement for parental consent in the case of Sweden. As has been
argued, there are good reasons for this requirement and no guarantee of all research
(or all researchers) being “good”. Nevertheless, I do think we need to critically
evaluate and discuss its consequences. Firstly, we need to ask: Who is this
requirement protecting? In some cases – and often in research in intimate settings
such as the family – the protected subject might actually be the integrity of the
“private sphere” rather than the child. Secondly, we need to look critically at how this
requirement is interfering with the aim of conducting child-centred research under the
ethical rules and/or guidelines of different national contexts. In so doing we can
become aware of the limits of our own research (e.g. whose voice is represented,
Journal of Comparative Social Work 2013/2
!
!
19!
and under what circumstances it is expressed), and how different ethical standards
affect the conditions under which the research is carried out. We can also – cross
nationally – develop and share ideas and strategies on how to strengthen children’s
rights in research, to opt in and participate, to be at the centre of the research
encounter and to be granted the same right to confidentiality as adult research
subjects.
References:
The Act concerning the Ethical Review of Research Involving Humans, Sweden
2003:460 (EPL).
http://www.epn.se/media/45159/the_etical_review_act.pdf [accessed 30 June 2012].
Alanen, L. (1992). Modern childhood? Exploring the ‘child question’ in Sociology.
University of Jyväskylä: Jyväskylä.
Alderson, P. (1995). Listening to Children: Children, Ethics, and Social Research.
Barnardos: London.
Balen, R., Blyth, E., Calabretto, H., Fraser, C., Horrocks, C., & Manby, M. (2006).
Involving Children in Health and Social Research: ‘Human Becomings’ or ‘Active
Beings’? Childhood 13 (1): 29-48.
Brannen, J. (1999). Reconsidering Children and Childhood: Sociological and Policy
Perspectives. In The New Family? E. Silva & C. Smart (Eds.), Sage: London.
Brannen, J., Hepinstall, E., & Bhopal, K. (2000). Connecting children: Care and
family life in later childhood. Routledge: London.
British Sociological Association (BSA) (2002).
http://www.britsoc.co.uk/about/equality/statement-of-ethical-practice.aspx [accessed
16 November 2012].
Journal of Comparative Social Work 2013/2
!
!
20!
The Convention of the Rights of the Child (CRC) (summary) (1989).
http://www.unicef.org/crc/files/Rights_overview.pdf [accessed 30 June 2012].
Christensen, P., & James, A. (2008). Introduction: Researching children and
childhood cultures of communication. In Research with Children: Perspectives and
Practices, P. Christensen & A. James (Eds.), Falmer Press: London.
Christensen, P., & Prout, A. (2002). Working with ethical symmetry in social research
with children. Childhood 9: 477-497.
Cocks, A. J. (2006). The ethical maze: Finding an inclusive path towards gaining
children’s agreement to research participation. Childhood 13: 247-266.
Coyne, I. (2009). Research with Children and Young People: The Issue of Parental
(Proxy) Consent. Children & Society 24: 227-237.
Danby, S., & Farrell, A. (2005). Opening the research conversation. In Ethical
Research with Children, A. Farrell (Ed.), Open University Press: Maidenhead,
Berkshire.
Davies, H. (2008). Reflexivity in Research Practice: Informed Consent with Children
at School and at Home. Sociological Research Online 13(4)5.
http://www.socresonline.org.uk/13/4/5.html
Donzelot, J. (1979/1997). The Policing of Families. John Hopkins University Press:
Baltimore.
Eldén, S. (2013). Inviting the messy: Drawing methods and childrens’ ‘voices’.
Childhood 20(1): 66–81.
Eldén, S. in press. Bortom kärnfamiljen? Omsorgsgörande från barns perspektiv.
Kvinder, køn og forskning.
Journal of Comparative Social Work 2013/2
!
!
21!
Eriksson, M., & Näsman, E. (2012). Interviews with Children Exposed to Violence.
Children & Society 26: 63-73.
Gallagher, M., Haywood, S., Jones, M., & Milne, S. (2010). Negotiating Informed
Consent with Children in School-Based Research: A Critical Review. Children &
Society 24: 471‐82.
James, A. (2007). Giving voice to children’s voices: Practices and problems, pitfalls
and potentials. American Anthropologist 109(2): 261-272.
Johansson, T. (2007). Experthysteri: Kompetenta barn, curlingföräldrar och
supernannies. Atlas: Stockholm.
Komulainen, S. (2007). The ambiguity of the child’s ‘voice’ in social research.
Childhood 14(1):
11–28.
Lewis, A. (2010). Silence in the context of ‘child voice’. Children and Society 24(1):
14–23.
Lindgren, A-L., & Halldén, G. (2001). Individuella rättigheter; autonomi och beroende.
Olika synsätt på barn i relation till FN:s barnkonvention. In Utbildning & Demokrati
10(2), 65-80.
Mason, J., & Fallon, J. (2001). Some Sydney Children Define Abuse: Implications for
Agency in Childhood. In Conceptualizing Child-Adult Relations, L. Alanen & B. Mayall
(Eds.), Routledge: London.
Mason, J., & Tipper B. (2008). Being Related: How Children Define and Create
Kinship. Childhood 15(4): 441-460.
Journal of Comparative Social Work 2013/2
!
!
22!
Mayall, B. (Ed.) (1994). Children’s Childhoods Observed and Experienced. Falmer
Press: London.
Mayall, B. (2002). Towards a Sociology of Childhood: Thinking from Children’s Lives.
Open University: Buckingham.
Morgan, D. (1996). Family Connections. An Introduction to Family Studies.
Cambridge: Polity Press.
National Children’s Bureau (NCB) (2011).
http://www.ncb.org.uk/media/434791/guidelines_for_research_with_cyp.pdf
[accessed 16 November 2012].
Näsman, E., Källström Carter, Å., & Eriksson, M. (2008). Perspektiv på barns röster
om våld. In Barns röster om våld, M. Eriksson (Ed.), Malmö: Gleerups.
Roseneil, S., & Budgeon, S. (2004). Cultures of Intimacy and Care beyond ‘the
Family’. Current Sociology 52:2, 135-159.
Sandin, B. (2012). Children and the Swedish Welfare State: From Different to Similar.
In Reinventing Childhood After World War II, P. Fass & M. Grossberg (Eds.),
University of Pennsylvania Press: Pennsylvania.
Skeggs, B. (1997). Formations of Class and Gender. Sage: London.
Smart, C., & Silva, E. B. (Eds.) (1999). The New Family? Sage: London.
Smart, C., Neal, B., & Wade, A. (2001). The changing experience of childhood.
Blackwell Publisher: Malden.
Journal of Comparative Social Work 2013/2
!
!
23!
Smart, C. (2007). Personal Life. New Directions in Sociological Thinking. Polity
Press: Cambridge.
Smart, C. (2009). Shifting horizons: Reflections on qualitative methods. Feminist
Theory 10(3): 295–308.
Spyrou, S. (2011). The limits of children’s voice: From authenticity to critical, reflexive
representation. Childhood 18(2): 151–165.
Uprichard, E. (2010). Questioning research with children: Discrepancy between
theory and practice. Children and Society 24: 3–13.
van Every, J. (1999). From Modern Nuclear Family Household to Postmodern
Diversity? The Sociological Construction of ‘Families’. In Changing Family Values, G.
Jagger & C. Wright (Eds.), Routledge: London.
Wihstutz, A. (2011). Working vulnerability: Agency of caring children and children’s
rights. Childhood 18(4): 447-459.
Wyness, M. (2006). Childhood and Society: An Introduction to Sociology of
Childhood. Palgrave Macmillan: London.
Young People Creating Belonging: Spaces, Sounds and Sight” – project (ESRC
RES-061-25-0501; 2010-2013, Dr. Sarah Wilson): http://www.esrc.ac.uk/my-
esrc/grants/RES-061-25-0501/read
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
i!The research for this article was enabled by funding from the Swedish Council for Working Life and Social
Research (FAS 2010-0505). I am very grateful to the two anonymous reviewers, as well as the editors at JCSW for
their helpful comments on improving the text. I would also like to thank the participants at the ESA RN20 Midterm
conference in Lund 2012, in addition to the participants at the CRFR International Conference Researching
Families and Relationships in Edinburgh 2013 for their comments and questions, as well as David Wästerfors and
Terese Anving, who commented on earlier drafts of this text. !
ii This includes the overall plan for the research, its purpose, the methods that will be used, the consequences and
risks of the research, the identity of the responsible research body, the fact that participation in the research is
voluntary and the right of the research subject to cease participating at any time.
iii An important exception is Coyne’s article (2010) on parental consent requirements in research, where a number of
problems of parental consent are discussed (e.g. assumptions of children’s incapacity to understand, the parent’s
ability to assess risk and the parental consent requirement as blocking children’s participation). However, Coyne is
not discussing the particular problems that arise in research in intimate settings such as the family.
iv The participating children came from family situations with heterosexual and homosexual parents living together,
separated or single parents and children living in co-housing.
v In all cases, the legal custodians of the participating children were parents of the child. In the following, I will
therefore refer to “parents” when talking about legal custodians.
Journal of Comparative Social Work 2013/2
!
!
24!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
vi Quote from the information leaflet and informed consent form of the project, translation from Swedish by the
author.
vii Two of the participating children had parents who contacted me after I had posted information of the project on a
Single Parent Association website. To get a diverse sample, I also unsuccessfully attempted to contact parents through
LGBT websites, but in the end, the participating children from LGBT families were recruited through use of the
snowball technique.
viii The data presented below are retrospective reflections based on the field notes I gathered during and after every
research encounter. All the names and some of the circumstances of the children and parents have been changed to
avoid identification.
ix The researcher also needs to be prepared for the possibility that the child shares information that might indicate
that the child is at risk, information that might also make it necessary to breach the promise of confidentiality and
report the family to the authorities. Researchers are not required to report suspected cases of children “at risk”
(compared to e.g. social workers and teachers), but I argue that for ethical reasons any research involving children
needs to be prepared for the possibility of this occurring. In this particular project, a social worker has been acting as
a “background consultant” in the project. The social worker was informed of the aims and methods of the project
before initiating it, and could be asked for advice on how to handle cases “at risk”.
x See, e.g. the ”Young People Creating Belonging: Spaces, Sounds and Sight”- project (Dr. Sarah Wilson;
http://www.esrc.ac.uk/my-esrc/grants/RES-061-25-0501/read). Similar ways of reporting back will be used in this
project.
xi Interestingly, in this project, the “worrying” parents were more likely to be middle class, something that might be
explained by new ideals in parenting, see e.g. Johansson, 2007.
xii However, some researchers have argued that eavesdropping in interviews with children could actually contribute to
better data by initiating discussions among family members on the topic at hand (Mason and Tipper, 2008).
