Ethical Research with Children
This page intentionally left blank
Ethical Research with
Children
Untold Narratives and Taboos
Sarah Richards, Jessica Clark and Allison Boggis
University Campus Suffolk, UK
© Sarah Richards, Jessica Clark and Allison Boggis 2015
Softcover reprint of the hardcover 1st edition 2015 978-1-137-35130-2
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6–10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified as the authors of this
work in accordance with the Copyright, Designs and Patents Act 1988.
First published 2015 by
PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,
175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies
and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,
the United Kingdom, Europe and other countries.
ISBN 978-1-349-55996-1 ISBN 978-1-137-35131-9 (eBook)
DOI 10.1007/978-1-137-35131-9
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
To our children
This page intentionally left blank
Contents
List of Figures viii
Introduction 1
1 Boundaries and Battlegrounds: Negotiating Formal
Ethical Approval for Research with Children and Young
People 14
2 Ethical Spaces and Places 32
3 The Rights of Participation and the Realities of Inclusion 62
4 The Illusion of Autonomy: From Agency to
Interdependency 78
5 Ramifications of Category Entitlement: In What Ways
Does Who We Are Determine What Others Will Say? 99
6 Privileging Voices 121
Conclusion 146
References 155
Index 176
vii
Figures
2.1 This is a story about where I was born in China 44
4.1 The place where I live now 91
4.2 First days in my family 92
viii
Introduction
This book builds on the legacy of work by Hobbs and May (1993) and
Bell and Newby (1977), some of the first researchers to publish collec-
tions of personal accounts of doing social research. Telling stories is a
natural part of the human condition (Barthes 1977). They help us to
understand our worlds, enable us to explore who we are and facilitate
our interactions with others. Indeed, such is the inevitability of nar-
ratives it is argued to be problematical only in a culture in which they
are absent (White 1980:5). Here, in this book, we tell some of our sto-
ries with the aim of shining a light on the ways in which research
with children is constructed, conducted and critiqued. Moving from
the proscriptive accounts of how research should be conducted, the
latter decades of the 20th century saw a rise in the explicit considera-
tion of research as far from straightforward (see, for example, Devine
and Heath 1999). This book emphasises that by acknowledging and
talking about these challenges and ethical dilemmas, students, aca-
demics, researchers and practitioners are better placed to engage in
empirical research with children.
The rise in telling stories across the social sciences to understand
the landscapes of our world (see Frank 2010; Oakley 2007; Plummer
2001) developed in parallel with an increased interest in the stories
of children and childhood (Cockburn 1998; James and Prout 1997;
Wall 2010). Telling stories is not new, exploring ethics in the research
process is not new and, as shall be seen in Chapter 1, research with
children certainly is not new either. In fact, research with children
is now well enough established to withstand reflexive critique that is
aimed at strengthening rather than diminishing the field (Gallacher
1
2 Ethical Research with Children
and Gallagher 2008). This book is a collection of stories that don’t
often get told: the messiness, the ambiguities and the parts of our
research that lay on the cutting room floor. A desire to present sani-
tised versions that document research means the presentation of any
narrative which focuses on the emotions or problems of research
is tainted with anxiety and as such the public availability of these
stories are limited (with notable exceptions such as Montgomery
2007). The presentation of our stories does not mean that we do not
feel this anxiety too. However, we believe the methodologies that
promote participatory research with children are strong enough to
bear this critically reflexive gaze. Research with children is mature
enough to have developed its own ‘cherished conceits, stubborn eva-
sions [and] persistent illusions’ (Segal 1999:118), produced in part
by the canonical narratives of these participatory methods, which
are often advocated as ‘best practice’, and too often go uninterro-
gated. We borrow Segal’s argument by suggesting that these stories
are what we ‘need to study not seek to evade’ (ibid.). This is our
attempt to shine a light in some of the dark places of research to
advance methodological conversations.
Traditionally, the voices of children, most especially young chil-
dren or those who are verbally less articulate, have largely been silent
within research. It is not that research has ignored children (because
there is much literature that documents childhood itself), it is sim-
ply that children in Western societies have been traditionally viewed
as objects of concern rather than persons with voice (Prout and
Hallett 2003). Until relatively recently, they have tended to be placed
within discrete sub-disciplinary boundaries such as the ‘sociology of
the family’ or the ‘sociology of education’. Indeed, the dominant
research approach has viewed childhood from adultist perspectives
on grand scales whereby performance or behaviour has been mea-
sured by standardised tests. However, since the 1970s, there has been
a growing sociological interest in children and childhood as distinct
social categories with a contribution to make beyond how we mea-
sure their development and educational attainment. The view that
children and young people should be viewed as ‘agents’ or ‘social
actors’ has to some extent alleviated their role as objects to be stud-
ied, emphasising that including their views and experiences could
provide important insights into the nature of childhood. This shift
in perspective fuelled a policy focus on children, and subsequently
Introduction 3
the 1989 United Conventions on the Rights of the Child (UNCRC)
was formed. The increasing enthusiasm for the concept and practice
of ‘child voice’ is outlined within Article 12 of the UNCRC and states
that children not only have the right to articulate their opinions with
regard to issues that affect them but also have a right to have these
opinions heard. Additionally, Article 13 declares that a child has the
right to seek, receive and impart information and ideas of all kinds.
Following on from this, a raft of policies including the Children Act
1989 and 2004; Children (Scotland) Act 1995; Children (Northern
Ireland) Order 1995; Every Child Matters (Department for Educa-
tion and Skills) 2003 were introduced in the United Kingdom, and
while they were designed primarily to assert and protect the inter-
ests and rights of children, they ostensibly promoted child voice by
implying that children should be encouraged to contribute to deci-
sions that affect them. Rather than traditional research methods of
‘looking down’ on childhood, the micro-level analysis of children’s
lives stimulated by the alternative rights approach advocates ‘look-
ing up’ at childhood in a more contextualised way. As a result, there
is now more recognition of their social agency and active participa-
tion, and such an approach has reached a point in its development
where it can sustain critical review. These assumptions, just like those
surrounding child development and unidirectional maturation that
previously dominated childhood studies, need themselves to be inter-
rogated. Such assumptions are themselves productive, shaping how
we view children and how we conduct research. Our stories here
question the extent to which these concepts are effectively applied
and understood in the context of children’s lives and everyday
experiences.
There are many standard ethical considerations associated with
researching with children and quite rightly so. The combination of
concerns about children’s capacity to understand what the research
entails giving informed consent; conflicts between parents and chil-
dren; and the avoidance of possible coercion by researchers, peers
or parents to participate in research produces something of an eth-
ical maze. It also creates tensions between researchers who seek to
empower children to participate and hear their opinions and those
who seek to regulate studies to protect children and their right to
privacy. While sensitive to the implications of rights to privacy, we
want children to remain active participants in research and therefore
4 Ethical Research with Children
advocate an approach that regards the two as complimentary rather
than oppositional.
Along with agency and participation however, often comes
the over-bureaucratisation of processes and, consequentially, the
privileging of voice. While the drive to hear and understand voices
and the assumptions of truth opens up innovative practices of col-
lecting authentic, spontaneous and realistic voices, it does little to
engage with the epistemological and methodological limits of voice.
The collection of chapters that follows presents a series of interpre-
tive transgressions that result from our need to complicate notions
of research with children. We were driven to gain a nuanced under-
standing of how these voices and experiences were sought. Indeed, as
researchers, we refused to follow the ‘too easy’ ethical pathway and
curiously examined the power relations that produce voices. In doing
so, we raise even more pertinent questions such as what constitutes
voice, who is allowed to speak and what does it mean to be classi-
fied as agentic? Why do we listen to some voices and not others?
How do we listen? And what journey does the voice take from utter-
ance to publication? And what are the ethical implications of these
questions? We do not admit to having all or any of the answers,
just an insatiable curiosity. Therefore, the stories embedded within
this text are products of our practice of scholarship which created
ethical narratives generated by real research dilemmas we wrestled
with in the field. It is our narratives, which usually remain unex-
posed by formalities of research, that we believe play pivotal roles in
our understandings of research with children and the construction of
knowledge.
Storytelling or narrative research has emerged as increasingly pop-
ular approach over the last 20 years (Andrews et al. 2008). Lawler
(2008:27) contends that the analysis of narratives is ‘embedded
within a hermeneutic tradition’ of enquiry and as such seeks to
understand the ways in which people make sense of their lives. Sto-
ries are produced through reference to other stories and texts in a
selective process which builds not only a story but a presentation
of self. There are numerous ways in which the term ‘narrative’ can
be understood and applied. Here we define it as being ordered ‘rep-
resentations’ (Andrews et al. 2008:12) of knowledge, in this case
about the ambiguities, dilemmas and revelations of doing research
with children. Our application of narrative is to use our stories to
Introduction 5
reveal particular interpretations and highlight multiple ‘layers of
meaning’ (Andrews et al. 2008:1) embedded in each story. Madison
(1988:99–100) argues that such experiences are meaningful ‘precisely’
because they can be told. We take isolated stories about our individ-
ual research experiences and weave something bigger (Alvesson and
Sköldberg 2009:129), a narrative which questions how we construct
children in participatory research and the implications of this social
positioning. We do not tell stories about how to conduct research
with children as others have done (see, for example, Greig et al.
2013). Nor do we provide advice about selecting a methodology or
method (see Lewis et al. 2004). This book is the one you reach for
when you discover that the ethical guidelines that you have based
your approach on fall short of helping you navigate the realities
and complexities of working with children and their families in the
settings where much research is conducted.
There are, of course, many interpretations of data and research
experiences that are possible depending on the research aim and the
researcher. Here we use reflexivity to move beyond primary interpre-
tations which focus on the research topic or aim. Instead, we explore
events in the field and what they can tell us about the practice
of research with children and how the knowledge about children’s
experiences is constructed. Madison (2005) claims that reflexivity is
underused in research, even where it is an important element such as
in ethnography. Perhaps such reflexion is restricted to the researcher’s
journal. As Aull Davis (2008:7) contends, ‘reflexivity in its “fullest
form” becomes destructive of the process of doing such research’.
Here we use it to provide a more transparent gaze on our own research
experiences. Telling our stories enables us to achieve this.
The relationship between reflexivity and ethics is long-standing.
Foucault (1972) presents ethics as the process undertaken by indi-
viduals as they reflect on and comply with or, indeed, resist the
moral codes which inform and organise social institutions such as
school and family. This ‘public discourse of morality’ (Zignon 2008 in
Laidlaw 2014:31) is argued to be an unreflective condition of our
‘everyday social life’ (ibid.) which is, on occasions, destabilised by
what Zignon refers to as ‘ethical moments’ when such an unreflective
habitus becomes untenable. The act where consideration of moral
codes has to take place is what Zignon (2008:260) refers to as ‘ethics’.
Reflecting on the moral codes that guide and shape research is thus an
6 Ethical Research with Children
ethical endeavour. Here we use this reflexivity to interrogate the nar-
ratives and moral codes, such as formal ethics guidelines, that shape
how, when, where and what research topics children participate in.
The tendency in research with children is to err on the side of
caution and to avoid so-called sensitive topics, to the extent that
certain domains and research questions become taboo and thus risk
becoming absent from research with children. Particular topics such
as intercountry adoption, disability or sex are drawn upon within
this book. We acknowledge that all are readily constructed as sen-
sitive, even contentious, topics and considered even more so when
it is children who are the intended research participants. These top-
ics often serve to destabilise the moral order (Zignon 2008) as they
cross the boundaries that adults have constructed around the norma-
tive themes of research with children. Constructing such significant
boundaries around children and childhood within research risks con-
tributing to the structural vulnerability of children and reproducing
the discursive motif of the vulnerable child. Philosophical assump-
tions about the nature of children and childhood are intimately
linked with the concept of innocence, equated with purity, irra-
tionality and a state of unknowingness, all of which characterise
the child as vulnerable (Kehily 2012). Indeed, as Robinson (2013)
argues, childhood as emotional capital is infused with sentimentality,
romanticism and nostalgia propped up by broad socio-cultural, legal
and political practices, which are in turn reinforced by discourses of
developmentalism. Transgressions from these cultural norms associ-
ated with childhood, particularly practices and activities which may
be perceived to contravene childhood innocence, foster significant
anxieties in society. These anxieties perpetuate stereotypes of child-
hood incompetence and myths and misconceptions about children’s
vulnerabilities. Thus research with children in so-called sensitive or
controversial arenas risks transgressing the discursive boundaries of
childhood innocence and is often considered to be ethically sensitive
and, as the title of this book suggests, taboo.
Taboos are rules of behaviour which restrict the human uses of
things and people, to avoid dirt and pollution and to manage ‘mat-
ter out of place’ (Douglas 1991:36). Taboos, Douglas argues, are held
in the eye of the beholder and have little to do with real dangers;
they are socially constructed and culturally specific but do function
to label certain things and activities as right or appropriate and others
Introduction 7
as abhorrent (Douglas 1979). Taboos do, however, result in the forma-
tion of systems of classifications and social rules; the aim of this being
to negate the potential disorder that breaking a taboo threatens.
Applying the above ideas to childhood and research with children,
certain topics are perceived as taboo, crossing some invisible yet pow-
erful boundaries which have been constructed around children and
childhood in part to protect the discursively constructed innocence
of childhood itself.
Innocence is the discursive essence of childhood, and one of the
strongest taboos in the contemporary world is the defilement of such
innocence. Therefore, asking children to share their views on topics
which are constructed as sensitive or outside the realms of childhood
risks unsettling the discursively constructed dominant social order of
childhood innocence and unknowingness. Such taboos and bound-
aries are powerful and shape the common-sense way in which we
view the world. These barriers are not arbitrary, and they serve to
support the structures that are located within and between. We thus
argue that such constructions of childhood innocence and the result-
ing topical taboos serve to reinforce the adult/child power relations
that structure contemporary Western societies, contributing to chil-
dren’s vulnerability and lack of voice. Such constructions shape what
we know about the worlds of children because they can only answer
the questions we are willing to ask.
What we argue throughout this book is that what exists in research
with children is a hegemony of voice whereby particular positions
and arguments privilege or marginalise both what is being said and
by who. This applies not just to the children being asked to par-
ticipate but also to those doing the questioning. Just as researchers
are often attributed the position of ‘expert’ on a topic (Danieli and
Woodhams 2005) by the very virtue of their status as questioner,
the positions they simultaneously occupy as mother, woman and
professional practitioner, to name but a few, can render them vulner-
able. This vulnerability results in the potential of having our voices
labelled as flawed because we reveal ourselves to not just (or at all)
being experts but rather subjects who are part of the co-construction
of knowledge. Moreover, we don’t display the objectivity still often
desired within social science research, having views which are shaped
by more than just the label of researcher or by deeming to speak
in ways that are at odds with dominant knowledge in our fields.
8 Ethical Research with Children
This hegemony of voice occurs within the diverse topics addressed
throughout this book: what we are allowed to research, who we are
allowed to speak to and then what it is that we are allowed to say.
Just as specific topics have been regarded as taboo in research with
children, certain conversations about the uncertainty and the messi-
ness of such research are also perceived as risky. Thus, here we talk
reflexively about institutional expectations, the spaces of research
encounters, the voices that dominate and the topics deemed ‘appro-
priate’ to be explored with children. Within these pages, our hope is
to promote a culture where, once ethics committees have been satis-
fied and when textbooks have been left on desks, the issues faced by
researchers working with children can be interrogated and shared.
We have deliberately chosen to use our first names, as authors,
within this book. This decision we believe is important in order to
emphasise the critically reflexive nature of our stories and to avoid
distancing ourselves from the conversations we’re having with read-
ers. To help readers navigate the content and discussions of this book,
we have briefly detailed below the research from which many of the
examples and themes are drawn.
Sarah’s research explored the contested topic of intercountry adop-
tion with families who live in England and have adopted daughters
born in China. Data was collected through semi-structured inter-
views with nine families and one social worker specialising in this
uncommon form of adoption. Nine mothers, three fathers and
eleven daughters aged between five and twelve years took part in
the interviews. Their narratives revealed complex belongings co-
constructed and managed by children and their parents alike. It also
highlighted the ways in which adoption discourse in general and
intercountry adoption in particular is perceived and subsequently
shapes the ways that these families construct their belonging. Sarah’s
research illustrated normative assumptions about childhood, such
as their agency, vulnerability and dependence, commonly ascribed
through the research process and instrumental in constructing roles
considered appropriate for them to play. Identities which move
beyond these ascribed categories are highlighted. Identity as fixed,
autonomous and integral to a concept of self is revealed rather
as being contingent, co-constructed and performed. (Some of the
names in transcripts from Sarah’s research have been changed for this
publication.)
Introduction 9
Allison researched with children and young people who have little
or no speech and who regularly used high-tech Augmentative and
Alternative Communication Systems (AACS) as part of their commu-
nication repertoire. The research focused primarily on voices – voices
that are young and voices that are disabled. Specifically, she con-
centrated on the ways in which voices are simultaneously facilitated
and inhibited. By including voices that are sometimes hard to hear,
Allison’s research built upon the moral perspective of respect for the
role and status of children and aimed to promote their entitlement
to being considered as persons of value and with rights. The partic-
ipants gave their consent for their words and gestures to be used in
this book, and Allison upheld their anonymity by using pseudonyms
to replace their real names.
Jessica’s research to date has focused on the discursive motifs of
childhood, notably in relation to gender, sex, sexuality and sexuali-
sation. Her work draws primarily from sociology and cultural studies
to evaluate constructions of childhood in policy, political debate and
media representations. Jessica has actively evaluated the dilemmas of
researching with children in virtual worlds and the topics deemed
appropriate for children to share their views upon. Her work ques-
tions the assumed vulnerability of children, considering children
as interdependent and gendered beings who are competent in dis-
cussing, or indeed refusing to discuss, the landscapes of their own
lives and experiences while simultaneously recognising how such
worlds are structured by adult ideas and expectations.
Chapter overview
Chapter 1 examines our experiences of seeking formal ethical review
of research with children and young people. Rather than focusing on
traditional conversations in this domain regarding whether research
should be subject to such review and to what extent (see, for exam-
ple, Dingwall 2008; Hammersley 2009), this chapter considers the
implications for researchers of negotiating what can often be the con-
tradictory demands of Research Ethics Committees (RECs). Here we
argue that within childhood studies there can be found an ethical
enthusiasm whose aim may ostensibly be to protect the well-being
of children but whose actions, which seem increasingly focused on
risk management, have detrimental emotional and practical costs
10 Ethical Research with Children
for researchers and children and young people alike. Philosophical
assumptions regarding the nature of childhood itself in the gov-
ernance of research with children results in overzealous attempts
to protect children and young people’s welfare. This can result in
the inappropriate inclusion of third parties in research and a lack
of acknowledgement of children’s competency to understand and
thus consent to a variety of research endeavours. Taboos constructed
around particular topics shape the conversations that children are
allowed to participate in, and this leads to suspicion regarding the
motives of researchers who dare to step beyond the normative bound-
aries of research. Such protectionism, as we argue, without critical
reflexivity, contributes to the structural vulnerability of children
and results in lacunae in knowledge where they, as participants,
and we, as researchers, are unable to share our experiences and
stories.
The places where such stories emerge are the basis for Chapter 2.
Here we consider the implications of space, notably the home, the
school and the virtual, for research encounters and experiences. Such
places shape the data generated and present particular ethical dilem-
mas. These dilemmas are not absent in other spaces, but particular
places where research with children is common can serve to magnify
and bring to the forefront particular ethical issues and concerns that
must be addressed. What is clear from our experiences of the spaces
of research with children is that the discourses of power that flow
through research encounters actively shape the relationships that
develop and data that emerges. Ideas that adults are powerful and
children are powerless are both exacerbated and dismantled in such
spaces. The relationships between researcher, participants and others
involved (such as parents or teachers) are more ambiguous than the
expected, normative relationships between researchers and respon-
dents. We thus had to manage multiple expectations and roles such
as professional or guest alongside that of researcher. These intimate,
institutional and political arenas should thus be conceptualised as
producing ethical dilemmas such as those relating to confidential-
ity, informed consent, reciprocity or privacy. Our stories reveal the
complexities of such social dynamics that require reflexive ethical
consideration. These stories themselves inform us about how chil-
dren navigate the social orders that structure the spaces we attribute
to childhood.
Introduction 11
The stories told in Chapter 3 consider ways in which researchers
could be prepared to step beyond rigid guidelines and boundaries
advocated within traditional research methods in order to reflect
upon capturing the stories of marginalised populations. The dis-
cussions explore the adoption of a bespoke approach to gaining
informed consent and, in doing so, highlight some of the reali-
ties of upholding children’s rights of participation while including
them in research. The discussions expose a number of real-life ethi-
cal issues experienced ‘in the field’ and explore how ethical processes
can take on a more heuristic and realistic pathway, rather than fol-
lowing the tried and tested, step-by-step research guides that do not
prepare researchers for the messiness, fluidity and diversity needed
while researching with children and young people.
Chapter 4 critically examines some of the key concepts, such
as agency, autonomy and active participation, which have charac-
terised and dominated contemporary social research with children.
The usefulness of such terms for the study of childhood are actively
interrogated, questioning the efficacy of attributing ideas of agency or
autonomy to children in order to understand their everyday experi-
ences. These values emerged within Enlightenment thought and are
characteristic of traditional models of citizenship, which are often
argued to be based on an individualist and male approach to partici-
pation in public life. These dominant ideas of agency and autonomy
construct an individualist rights position which is used to inform
ethical frameworks such as regulations surrounding informed con-
sent. These qualities are problematic because of the assumptions we
make about the discursive construction of childhood itself, whereby
children are not attributed the very attributes and skills that ethical
research with children is premised upon. We question the appropri-
ateness of these concepts and ask if alternative values might be more
appropriate to reflect the roles of children as negotiated, relational
and interdependent. We argue that rather than the individualist
notions of agency, we should instead consider relational agency and
interdependency as concepts which better reflect how children (and,
indeed, adults) are socially embedded in the navigation of their
worlds.
The aim of Chapter 5 is to acknowledge the presence of ourselves
in our research and to explore the ways in which who we are shapes,
what we find and why we find it. We use a specific focus on the
12 Ethical Research with Children
concept of category entitlement to illuminate the situated and con-
nected category memberships of those involved in our research. Here
the purpose is to extend the debate that the researcher is influen-
tial, to go further by examining how this influence is established and
articulated. This chapter explores the processes of production and
interpretation of data through distinct concepts including category
membership, self-disclosure and insider status. These concepts are all
related to the co-construction of relations and the establishment of
authority to speak. Our stories here evaluate the connections cre-
ated and negotiated within this research context and consider how
such relations build an entitlement for the researcher to ask and the
participant to respond. Particular membership of categories is often
the very reason for the participation of children in research, such
as pupil, young carer, unaccompanied asylum seeker, street child or
sibling. This membership has not only predicated inclusion within
research itself but also carries with it particular traits, characteristics
and behavioural expectations. These memberships are not neces-
sarily the categories that children have established for themselves
but are commonly attributed to them. Nonetheless, we argue that
children use category entitlement as effectively as adults in their
relationships in order to make decisions about issues such as partici-
pation. We demonstrate here that their use of this ‘tool’ to affirm or
indeed resist such classifications enables them to successfully nav-
igate research relationships. What is particularly pertinent are the
ethical implications that are brought to bear by acknowledgement
of these connections between the researcher and the researched.
Chapter 6 challenges the ways in which some children’s voices are
privileged while others are overwritten by adults who talk at them,
about them and on their behalf. While acknowledging that includ-
ing children with unconventional or quiet voices often presents
researchers with significant challenges and that facilitating their
inclusion requires a considerable investment in time resources and
the development of new skills, we use our research stories to argue
that it is unacceptable to exclude voices on the basis of incompetency
or because they pose challenges to traditional research methods. The
narratives contained within this chapter highlight the choices that
we, as researchers, make about the voices we listen to, how we lis-
ten to them and why we consider some to be more authentic than
others. Here we also examine how we use our own voices to share
Introduction 13
our experiences. The ethical dilemmas associated with the dissemi-
nation of research are considered including the gatekeeping practices
(Brooks et al. 2014) associated with both topics to be explored and
ways of writing about and presenting data and ideas. Arguments sur-
rounding the ‘spiral of silence’ (Noelle-Neuman 1984) suggest that
ways of speaking are constrained by dominant discourses surround-
ing both research and childhood itself, and as such particular voices
are rendered illegitimate and therefore silenced.
According to White (1980) narratives are only problematic by their
absence, however we contend that stories about doing research with
children are not absent but far less prevalent than accounts where
children are claimed to be in some way active, empowered or agen-
tic by virtue of a methodology, method or particular value stance of
a researcher. Thus, we attempt to address the scarcity of such stories
within this domain (with notable exceptions such as Montgomery
2007; Pole 2007; Punch 2002) by adding our stories to this slowly
burgeoning field. The structure of this book aims to take you on a
journey. We start in Chapter 1 where we explore how one might enter
the field, as well as consider the regulation of access to particular
voices, speaking in particular arenas. Chapters 2 to 5 focus upon expe-
riences on the ground, where we tell stories of the co-construction of
data and research relations. Our final chapter then considers what
happens to these voices when we attempt to share them, how and
where they can be heard and who listens. Throughout, we attempt to
reflect upon the moral orders of research with children, questioning
some of the persistent illusions and cherished conceits (Segal 1999)
of the institution of childhood itself. We argue that such interroga-
tion is what the practice of ‘doing ethics’ is all about and that these
stories are what we should engage with rather than evade. This book
then has become our ethical moment.
1
Boundaries and Battlegrounds:
Negotiating Formal Ethical
Approval for Research with
Children and Young People
The Nuremberg Code (1947) and the Declaration of Helsinki (1964)
are both heralded as the result of what are now widely recognised
as notorious examples of unethical research including the Tuskegee
syphilis study from the 1930s to the 1970s and Stanley Milgram’s
obedience research in the 1960s, not to mention the atrocities that
took place during the Holocaust. These formal protocols and frame-
works, developed in the United States and Western Europe, aim
to guide the ethical conduct of human research and have seeped
from governing medical research into other disciplines including
the social sciences. These trends have led to the establishment of
Research Ethics Committees (RECs) (known internationally by differ-
ent terms: for example, Institutional Review Boards (IRBs) in North
America) in a range of institutions including the United Kingdom’s
National Health Service, research councils such as the Economic
and Social Research Council (ESRC) and universities and Higher
Education Institutions (HEIs) around the world.
The aim of RECs is broadly to guide and support sound ethi-
cal research practices and to safeguard participants and researchers
alike. Both the committees themselves and the guidelines that they
may make use of (for example, the ESRC Framework for Research
Ethics 2012, British Educational Research Association (BERA) Ethical
Guidelines 2011 or British Sociological Association (BSA) Statement
of Ethical Practice 2002) have been criticised across a range of
areas including becoming increasingly bureaucratised, drawing upon
14
Boundaries and Battlegrounds 15
a medical model that is inappropriate for the social sciences and
restricting potentially ‘risky’ research which is viewed as important
and legitimate – for example, research on so-called sensitive top-
ics or with participant groups who are perceived as vulnerable (see
arguments, for example, in Dingwall 2008, 2011; Hammersley 2009;
Hammersley and Traianou 2011). The discussions here are purpose-
fully focused on the governance of ethical research in the United
Kingdom as this is the context where our research has been con-
ducted. However, there is a need to examine the development of
ethical regulation more broadly across the globe, outside of domi-
nant Western perspectives which run the risk of being imposed on a
range of cultures in somewhat imperialist terms (Benatar 2002).
The aim of this chapter is not to engage in overarching arguments
surrounding the degree to which social science research with (or,
indeed, without) children should be subject to ethical regulation or
the form that this should take. Rather, here we will consider how
individual researchers negotiate existing formal ethical procedures
and the impact of this on all the individuals involved. We will focus
on three specific issues that we ourselves have experienced during
ethical approval processes and that are growing to be increasingly
common sights in the literature (see, for example, Monaghan et al.
2012; Skelton 2008): first, the positioning of children as providing
assent rather than consent in the research process; second, the often
imposed existence of third parties within research interactions; and
finally, the governance of the kinds of individuals ‘allowed’ to do
research with children and the topics that are and are not acceptable
to talk about. Sikes and Piper (2010) highlight the importance of lis-
tening to researcher’s stories. Narratives of such processes shine an
important light on the often contradictory priorities of childhood
researchers and on the RECs and ethical guidelines which govern
their endeavours. We will explicitly consider our experiences of mak-
ing applications to RECs for our own research and as dissertation
supervisors supporting students in their first forays into primary
research. In exploring these narratives, we will argue that fundamen-
tal philosophical assumptions about the nature of childhood and
the competencies of children position children and young people
as a homogenously ‘vulnerable’ group with certain ‘sensitive’ topics
constructed as perpetually ‘inappropriate’. We also draw attention to
the emotional costs to researchers and supervisors of engaging with
16 Ethical Research with Children
risk-averse RECs and the impact on the creation of knowledge in par-
ticular arenas, which are constructed as taboo and ‘out of bounds’.
Such positions foreground an overly enthusiastic principle of protec-
tion over and above that of participation whereby we risk denying
the right of people to participate in research through ‘unnecessarily
protective and paternalistic measures’ (Huxley et al. 2005:59).
Children’s assent versus informed consent
Informed consent is closely associated with the ethical principles
of ensuring respect for persons (Brooks et al. 2014). This includes
broadly three key principles: first, that adequate knowledge is pro-
vided to prospective participants; second, that consent is voluntary
at the outset and throughout the research process; and third, that
such decisions are made by competent individuals choosing freely
(Brooks et al. 2014). This model of consent has been critiqued by
particular scholars for being based on a neo-liberal, individualist and
masculine model of personhood (Hammersley and Traianou 2012).
It is further complicated when considering the philosophical posi-
tioning of children in contemporary Western societies as vulnerable,
dependent and irrational (Archard 2004). Such discursive construc-
tions of childhood have historically made it very difficult for adults to
consider children as capable of offering such consent. Recent changes
in the positioning of children heralded by the new social studies of
childhood (James et al. 1998) and the United Nations Convention
on the Rights of the Child (UNCRC) mean that assumptions of chil-
dren’s inability to participate meaningfully in decision making have
been questioned. What concerns around informed consent reveal is
a focus on the way in which power relations play out in research
relationships (Thorne 1980) and, as is increasingly being consid-
ered in contemporary research contexts, the desire of institutions
that govern research to protect themselves from legal consequences
arising from research (Edwards and Mauthner 2012). The experi-
ence detailed below briefly articulates what happened to one of our
dissertation students in their negotiations with ethics committees
whereby the issue of children providing informed consent was far
from straightforward.
As members of academic teaching staff, we all regularly act
as undergraduate dissertation supervisors, and recently, Jessica
Boundaries and Battlegrounds 17
supervised a student, Rebecca, who was proposing to conduct
research in a setting that was also governed by an additional regional
REC external and separate to the university REC. In this section, we
specifically consider Jessica’s experiences of supervising this student
as they engaged in a series of negotiations between these two RECs.
Rebecca wished to explore children’s perspectives on healthy eating,
using informal focus groups and drawings with children aged 3–4
years old in an early years setting. Within our institution, Rebecca’s
research gained ethical approval on the basis that her participant-
led, child-centred methodology and documentation were considered
ethically sound by the university REC. Due to the location of the
setting, Rebecca also had to apply for ethical approval with an addi-
tional REC external to the university. This REC, on considering the
proposal, reached a conditional decision whereby a series of require-
ments had to be met in order for ethical approval to be granted. The
first of these conditions was a change of wording throughout the
proposal and all documentation. This change entailed a shift from
referring to children’s ‘informed consent’ to their ‘assent’. We use
this example to consider the contradictory dilemmas that using the
terms ‘assent’ and ‘consent’ in research can create.
Assent is defined as the child’s permission or affirmation of agree-
ment to participate in research (Broome and Richards 1998) and is
regarded by some researchers as a substitute for consent so that chil-
dren do not consent in their own right but assent to their parent’s
consent (Powell and Smith 2010). Assent is not a legally mandated
process (Twycross 2009), and, in addition to cases where parental
consent may itself be inappropriate or harmful, we question whether
the upholding of dignity, highlighted in many ethical guidelines
(BERA 2011; BSA 2002; ESRC 2012) and in key human rights instru-
ments (Bell 2008), can be maintained while positioning children’s
agreement as lesser than that of adults within the research process,
with their views and desire to participate (or not) as having reduced
value. This ‘fails to accord them the same rights as adults in terms of
what their consent means’ (Skelton 2008:23). Examples of this can be
found in other research stories; for example, Scott (2000) documents
that there was a desire in their research to use a cash incentive with
children completing their version of the British Household Panel
Survey (BHPS) to mirror that which was provided to adults. They,
as researchers, had an inherent desire to place the same value on
18 Ethical Research with Children
children’s contributions as that of adults. This ideological commit-
ment was somewhat constrained by financial budgets which resulted
in the young people being paid a sum, but that was less than that of
their fellow adult participants.
This issue is an example of how social research is ‘a compli-
cated balancing between many opposing options and minefields’
(Hallowell et al. 2005:142); it also highlights the continued ambiguity
about children’s voices and the problems that still remain in attempt-
ing to empower children and young people in the research process.
As Lukes (2005) argues, caution is particularly necessary in relation
to the least visible, and therefore most potent forms of power and
ethics is partly, in our view, about avoiding abusing power discrep-
ancies between researchers and participants (Alderson and Morrow
2006). In considering the positioning of children as unable to provide
full informed consent (see, in addition, discussions in Chapter 4), we
were presented with a motif of children as irrational beings with-
out the desired capacity to participate fully in social life. Drawing
on Foucault’s concept of the ‘examination’, ethics review processes
can be seen as disciplinary strategies that combine techniques of an
observing hierarchy and those of normalising judgements (Guta et al.
2013) functioning as a ‘normalising gaze, a surveillance that makes
it possible to qualify, to classify and to punish’ (Foucault 1995:184).
The classification of children as incapable subjects reinforces the per-
petual ‘othering’ of children within the research process (Lahman
2008), and such practices contribute to the more general powerless-
ness of children under an adult gaze. Frankel (2012) argues that a
barrier to children’s engagement in social and moral discourses has
been and remains based on the extent to which children are seen
as competent to engage with moral issues (see further discussions in
Chapter 5 on children’s autonomy). Philosophers such as Aristotle
and Locke saw moral reasoning as linked to full ‘adult’ humans,
with children viewed as on the journey to such desired capacity
(see Archard 2004). These philosophies provided the backdrop for
stage-based psychological theories regarding children’s development
which emphasise the development of children and their ability to
attach moral meanings to their actions. The risk here is that chil-
dren become objectified as passive social objects, conceptualised as
lacking the rationality and reasoning to make social or moral deci-
sions (Clark 2014). This includes the ability to process information
Boundaries and Battlegrounds 19
about and consent to participating in social research. However, we
argue that positions such as these, those that use the language of
assent rather than consent, fail to recognise the extent to which even
very young children engage in decision making, whereby morality is
part of the everyday life of all human beings (see Mayall 2002; Short
1999).
It has been argued that this can be attributed to the fact that
traditionally RECs have not been developed from a child-centred per-
spective (Skelton 2008), and many organisations still lack explicit
guidelines for child-centred research (Hendrick 2000). The principles
and regulations that guide REC decision making thus appear to exist
in contradiction to many of the approaches and ontological assump-
tions used in contemporary research with children. As Boden et al.
(2009) point out:
Social science has a long and creditable history of praxis on ethical
issues, both in terms of research conduct and consideration of how
the resulting understandings may be deployed.
The development of such work is evident in some ethical guide-
lines, for example, the BSA (2002), which are the product of long
traditions of scholarly work on ethics and power in research (see,
for example, Hollway and Jefferson 2000; Mauthner et al. 2002;
Silverman and Gubrium 1989). These guidelines attempt to be widely
drawn and recognise the need for researchers to make rapid and
pragmatic decisions in the field (Boden et al. 2009). They do not,
however, always offer researchers a useful tool for guiding or support-
ing decision making. As Flewitt (2005) states, she found little practical
support in ethical guidelines to support the translation of general
principles into actioning the specific issues that arose in the course
of her fieldwork. In addition, such guidelines often do not explic-
itly take into account the methodological and ethical issues that may
arise in relation to research with children or the development and
role of children’s rights, both in the United Kingdom and globally,
on such research philosophies and practices.
Research with children, as advocated by the ‘new’ social studies of
childhood (James et al. 1998), views children and young people as
agentic social actors who actively contribute to their social worlds
and, as such, should be viewed as co-creators of knowledge within
20 Ethical Research with Children
the research process. The United Nations Convention of the Rights
of the Child (UNCRC 1989) Articles 13 states that children ‘shall
have the right to freedom of expression; this right shall include free-
dom to seek, receive and impart information and ideas of all kinds,
regardless of frontiers’. However, in the case of Jessica’s student, the
position adopted by that of the additional REC whereby there was
an explicit desire that the student used the language of assent rather
than consent in her documentation and research proposal seemed to
be in direct contradiction to the approaches to research with children
articulated in literature on child-centred research practices, taught by
supervisors, promoted by the university REC and believed in by the
student researcher. As Boden et al. (2009:742) argue:
[B]y categorising people as vulnerable and therefore unable to
decide for themselves whether or not to participate in research,
ethics committees are accreting to themselves these people’s power
of exercising voice.
Third-party involvement
Another area of contention in published stories of research with
children, and indeed in the experience of our undergraduate stu-
dent negotiating the contradictory demands of two RECs, involved
the role of third parties within the research process. This relates to
both additional ‘researchers’ from within the setting or appointed
to surveil the research being undertaken and the storage and han-
dling of the resulting data. Rebecca’s research proposal stated clearly
the plan to use tape recordings of the children’s conversations while
they undertook a free drawing activity if they wished. These drawings
could be kept by the child, given to the researcher or photocopied
depending on the children’s wishes. This, along with the issue of
the tape recorder, was addressed in the information letter and con-
sent documentation for the children, the parents and the gatekeepers
but was also to be negotiated verbally and ongoing throughout the
research. The student had carefully explained the secure, confiden-
tial and anonymous storage of the research data within her proposal.
However, the additional REC stipulated that the data should not
leave the setting and that other members of staff within the setting
would be able to have access to the data, notably the manager who
Boundaries and Battlegrounds 21
held the key to the locked cabinet within which it was to be stored.
There was a fundamental difference here in perceptions of owner-
ship of the ‘data’ between the university REC and the external REC.
As both a supervisor and a researcher, Jessica considered the voices
of these children to belong first and foremost to the children them-
selves and, upon gaining their permission, the data could be used by
the researcher whose duty was to keep it confidential, anonymous
and to represent such voices fairly and honestly. The university REC
considered that it is the student researcher’s duty for safe and con-
fidential data storage and handling. In contradiction, the external
REC appeared to indicate that the setting itself and the other staff
within also had a claim to this data, to be able to access and view
it and to be responsible for its safe storage. We were puzzled as to
how the student was to work with her data if it was not able to
leave the setting. We also considered that the manager of the setting
and the other staff had no right to see the raw data where children
and their views would certainly have been identifiable to staff who
work with them and know them well. Jessica considered that this
violated the claims to confidentiality and anonymity, which the stu-
dent researcher had made to those participating in the research, in
the proposal, information letters and consent forms (with notable
exceptions for safeguarding issues). This is not an unusual occur-
rence in research which takes place in institutions of education and
care for children whereby researchers have to grapple with varying
demands regarding the confidentiality of data. As Barker and Weller
(2003) articulate, it is not unusual for teachers to want to hear or see
interview data considering they have a right to do so, given the set-
ting of the research in their classroom or their role as a gatekeeper
in the researcher’s access to the child participants. The management
of such pressurised situations is not easy (Masson 2000) as refusal to
capitulate directly contradicts the mechanisms of surveillance that
dominate such institutions (Barker and Weller 2003). This is partic-
ularly the case for student or new researchers engaging in their first
forays into research with children. This is also particularly difficult if
such researchers have an existing relationship with the setting as a
volunteer or employee, for example.
As supervisors, we began to reflect here on the insistence
and appropriateness of involving others in the research process.
Monaghan et al. (2012) also document that their REC requested an
22 Ethical Research with Children
independent third party should be in the room during their research
when interviewing the children who had agreed to participate. In
Monaghan et al. (2012) detail the experience of Maria O’Dwyer, one
of the co-authors, who defended her research from the inclusion of
additional adults. Maria’s decision was based on the argument that
the inclusion of multiple adults in interviews with children could
in fact contribute to the creation of a more unethical situation where
the children involved felt pressured and surrounded by a multitude of
‘powerful’ adults. The external REC, having considered the proposal,
listed by name the other staff in the setting and wanted their role
in the research clarified. Our student had made it clear that she was
the principal and only researcher on the project. No others had been
identified, and it was through separate investigation that the external
REC identified other members of staff in the setting and made a series
of assumptions about their involvement.
The implications of such stipulations by RECs may appear at first
to be small or easily managed, but they relate to wider assumptions
about the ownership of data and the voices of children. Concern was
not articulated here by the external REC that the children’s experi-
ences, what they had chosen to share with the researcher of their
own volition and about confidentiality. As adults in the lives of
these children, the employees and leaders in the setting in which the
research was undertaken were considered to have a right to see or, at
the very least, be tasked with ‘holding’ the data. The implications
of adult/child power relations in particular settings is considered
in much more detail in Chapter 2; however, as with the previous
assent/consent discussion, what this experience reveals is a funda-
mental questioning of children’s voices as legitimate and owned by
themselves. They are somehow undeserving of the same degree of
confidentiality afforded to adults by virtue of philosophical attributes
of childhood itself (discussed in the following section) which deem
them to be irrational, incompetent and vulnerable.
Ethical enthusiasm in childhood studies: Taboos and
boundaries
Ethical enthusiasm has been noted in childhood studies (Hammersley
2009) where Graue and Walsh suggest that ‘one should enter the field
as though on one’s knees . . . not merely an entry ploy but a posture
Boundaries and Battlegrounds 23
that one maintains throughout the entire research’ (1998:57). The
existence of ethical guidelines and their deployment by RECs are of
clear benefit in prohibiting some of the unethical practices seen pre-
viously in both medical and social research. We do not advocate here
that research should not be assessed for its ethical integrity, but we do
share the concerns of other prolific writers in this field (Hammersley
2009; Hammersley and Traianou 2011) that such regulation risks clos-
ing down both the topics explored and groups with which research
can be conducted. As articulated previously, overprotection in eth-
ical regulation contributes towards the structural vulnerability of
children reiterating children’s lack of power and status within soci-
etal structures (Lansdown 1994). We argue that true protection of
children requires protection of their rights, including that of partici-
pation (see also Miller 2000; Powell and Smith 2010; Sandbaek 1999).
The argument put forward by Graue and Walsh (1998) above
impacts not only on the positioning, value and rights of children
within the research process but also has costs for researchers. The lit-
erature that does exist where others have shared their stories speaks
to a mistrust of childhood researchers or indeed any researchers
engaging in topics deemed problematic, suspicious or ‘other’ (see,
for example, Goode 2010). Goode (2010) devotes a chapter in her
book solely to the difficulties involved not only in gaining ethical
approval to complete her research with self-defined ‘minor attracted
adults’ but in the risks of such a project to her academic reputa-
tion. Similarly, Monaghan et al.’s (2012) account of Maria O’Dwyer’s
experiences seeking university REC’s approval for a qualitative soci-
ological study of childhood asthma draws upon Hochschild’s (1983)
discussions of emotional labour, with the aim of forewarning and
forearming other social researchers entering this field; they in fact
dub their account ‘a cautionary tale’ (2012:66). What is evident is
that the ethical governance of research with children has real emo-
tional costs for researchers who, like Goode (2010), are often left
feeling vilified, uncertain and devalued in their endeavours (see also
Chapter 6). Fraser et al. (2010) posit a relational approach to emotions
whereby they emerge between subjects and between subjects and
society. The consequence of this is that they play a part in the social
construction of personhood and to feelings of belonging and legit-
imacy (Monaghan et al. 2012). Researchers are subject to demands
of including third parties, questions about their competencies and
24 Ethical Research with Children
characteristics, and find themselves stressfully caught between alter-
native and competing agendas of various RECs and institutions. Thus
they run the risk of what Stanley and Wise (2010: 6.9) describe
as ‘crying in the dark’. The costs of such interactions, Monaghan
et al. (2012) argue, emerge slowly and accumulate over time dur-
ing multiple micro-interactions. The dramaturgical stress of such
contradictory demands on emotionally expressive bodies (Freund
1990) has real material consequences for social actors. Emotions are
central when dealing with competing demands not just between
different RECs but also between individual REC requirements and
individual’s own ethical conscience. In this sense, researchers can
be conceptualised, as Boden et al. (2009:243) argue drawing upon
Foucault’s work, as docile bodies ‘self-regulating and self-disciplining
their own actions against particular ideas of what is standard and
good’. This risks a reduction in the researcher’s imagination and
individual’s ability to engage in innovative and creative work as
effort and energy are channelled in ‘easier’ directions, and research
intentions are channelled to fit a particular set of norms man-
ifest in the relevant bureaucratic apparatus (Boden and Epstein
2006).
Foucault’s concept of governmentality is useful here in highlight-
ing how ‘capillary forms of power insert themselves into actions,
attitudes and discourses’ (Guta et al. 2013:302), promoting the desir-
able behaviour of researchers (engaging with particular groups about
particular issues) through a myriad of institutional techniques. This
perhaps highlights the divide that some scholars have pointed to –
between procedural ethics and ethics in practice that run the risk of
sapping our emotional energy (Collins 2004). Although we promote
in all of our work increased conversations regarding what constitutes
‘ethical’ research more generally and specifically, that with children,
the increasing discourse of ‘risk’ adopted by RECs under increasingly
bureaucratic conditions does run the risk of stifling researchers keen
to embark on exploring often under-researched aspects of children
and young people’s everyday lives.
We too have shared disappointing and emotional experiences, like
that described by Goode (2010) and Monaghan et al. (2012), in
seeking ethical approval for our own research. Sarah’s research with
children and families regarding intercountry adoption was subject
to significant university ethical review. This process saw her REC
Boundaries and Battlegrounds 25
application viewed at a number of levels in the institution as various
groups labelled the area a ‘sensitive one’ in need of rigorous ethi-
cal evaluation. What appeared to be overlooked during this process
was that ‘being adopted’ for these children was not a special topic or
issue, or one that necessarily rendered them vulnerable, but rather an
integrated and embedded part of their everyday life.
Allison’s research proposal was greeted with what she describes as
‘horror’. The title of her REC application included the words ‘dis-
abled’ and ‘children’, both of which sent the members of the ethics
committee into something akin to panic. It conspired that this partic-
ular school within the institution had never received a call to approve
a proposal such as this before, as they were usually dealt within
the school of medicine or health. Similar to Sarah, Allison’s appli-
cation was heavily scrutinised, and, as a consequence, the process of
approval took 18 months to complete.
Jessica was actively discouraged by a variety of colleagues and supe-
riors from engaging in online research with young people under a
certain age boundary (which lay anywhere from 16 to 18 years of
age). The reasoning for such recommended restrictions was to avoid
having to grapple with legal issues of informed consent for partici-
pants and parents, problems with affirmation of identity in virtual
spaces and concerns regarding the well-being of children in online
spaces more generally (see Chapter 2 for further discussion). In plan-
ning to conduct research with girls online about their feminist beliefs
and political activism, Jessica was additionally advised to consider
the prospect of entering online spaces but not actually engaging
in conversations with the girls themselves. This was considered by
many a less ‘risky’ approach than attempting to contact the girls
directly, whereby there was no guarantee of appropriate adult con-
sent. Little consideration appeared to be given to the reiteration of
unequal power relations between adults and children by an adult
researcher lurking in online spaces created and owned by the girls
themselves. Additionally, by the possibility of misinterpretation by
an adult researcher of online content created, shared and reconfig-
ured for a variety of purposes and with a multitude of intentions by
these young people. In other arenas where Jessica has presented her
work on sexualisation and childhood, she has been approached after-
wards by fellow academics who have dubbed her brave for embarking
on discussions of such controversial topics. In addition, Jessica’s
26 Ethical Research with Children
workshops for practitioners on considering the positive elements
of young people’s sexual lives were highlighted by attendees as the
only offering they had yet to experience as part of their professional
development which did not focus on unplanned pregnancies, sexu-
ally transmitted infections and sexual exploitation and wasn’t that a
brave approach to take. In fact, sex and sexuality is an area which is
considered particularly difficult for researchers to engage in with chil-
dren and young people and is somewhat emblematic of areas which
remain under-researched for particular groups and where particular
voices remain predominantly absent. The result has been a distinct
lack of children and young people’s voices in such debates around
sexualities, sexual cultures and sexualisation (Clark 2013a) – with the
notable exception of Renold’s (2013) research with 10–12 year olds.
As stated previously, this runs the risk of generating lacunae in fields
of knowledge (Clark 2014).
The ‘difficult’ topics highlighted above are examples of topics con-
sidered ‘sensitive’ in all research but particularly so in research with
children. The tendency in research with children is to err on the side
of caution, to avoid so-called sensitive topics, to the extent that cer-
tain domains, certain research questions become taboo and thus risk
becoming entirely absent. Particular topics such as adoption, disabil-
ity or sex, some of those drawn upon within this book, are all readily
constructed as sensitive topics and considered even more so when it
is children who are the intended research participants.
Philosophical assumptions about the nature of children and child-
hood are intimately linked with the concept of innocence, equated
with purity, irrationality and a state of unknowingness, all of
which characterise the child as vulnerable (Kehily 2012). Indeed, as
Robinson (2013) argues, childhood as emotional capital is infused
with sentimentality, romanticism and nostalgia propped up by broad
socio-cultural, legal and political practices, which are in turn rein-
forced by discourses of developmentalism. Transgressions from these
cultural norms associated with childhood, particularly practices and
activities, such as particular sensitive research topics, which may be
perceived to contravene childhood innocence foster significant anxi-
eties in society. The phrase ‘sensitive topic’ is, we argue, significantly
overused and underevaluated, and rolled out to cover particular areas
of research as if it were self-explanatory (Lee 1999). Sieber and Stanley
Boundaries and Battlegrounds 27
(1988) define sensitive topics as those which are socially sensitive
and dependent on the individuals involved, the topic or the context
of the investigation. The advantages of this definition, as Lee (1999)
highlights, are that they point us to topics which might not ordinar-
ily be sensitive, but which in particular contexts are. This is often the
case in research with children whereby discursive constructions of
childhood, manifest in structural regulations, render particular top-
ics problematic. In Sarah’s research she recognised adoption as being
a stigmatised family formation (Fisher 2003); intercountry adoption
perhaps even more so (Johnson 2012). Defined in this way, adop-
tion becomes a culturally sensitive topic (Hydén 2008) and triggers
notions of protection. Nevertheless, being adopted is a visible and
lived reality for the children whose stories she wished to capture.
So we are forced to ask for whom is this a sensitive topic, and what
role do such assumptions play in preventing children talking about
their lived experiences when they are termed ‘sensitive’? It cannot
be assumed that this topic is universally sensitive for all families and
children involved in adoption. Perhaps, as argued above, sensitivity
might more effectively be perceived as relational and context-specific
(Hydén 2008).
Despite extensive literature and research examples which promote
the empowerment of the role of children in the research process,
there remain boundaries and taboos about topics which children
must be protected from, even if the topic relates to their own life
experience. The field of research with children remains populated
with children’s perceptions of friendship, education and normative
kinship relations. Far less research exists on children’s views of con-
tentious or, indeed, topics simply defined as adult (Uprichard 2010).
What is particularly dangerous about such classifications is that the
designation of a topic as taboo or sensitive is often simply assumed
and rarely interrogated, an unreflected moral order in the ways in
which we research with children.
Sensitive topics can be found in the diverse work of Farberow
(1963), drawing on psychoanalytic and anthropological research,
who labels those topics which inspire feelings of awe or dread as
sensitive or taboo. Extending upon these anthropological leanings,
explorations of taboo by Douglas are defined, like dirt, as that which
can be thought of as ‘matter out of place’ (1991:36). In its original
28 Ethical Research with Children
Polynesian manifestation (tabu) it refers to a ban or a prohibition
(originally, a religious restriction), but it’s entrance into European
languages can be perhaps better thought of as
a ban on touching or eating or speaking or seeing. Its breach
will unleash dangers, while keeping the rules would amount to
avoiding dangers and sickness.
(Douglas 1979:75)
Taboos are thus rules of behaviour which restrict the human uses of
things and people. As with dirt, taboo for Douglas is held in the eye of
the beholder and has little to do with real dangers, to well-being for
example. Nonetheless taboos do result in the formation of systems
of classifications and social rules; for example, in this current discus-
sion of ethics committee. The aim of such rules and classifications
being to negate the potential disorder that breaking a taboo threat-
ens. Douglas’s (1991) anthropological exploration of global societies
in relation to religion and the supernatural points to the dreaded
effects (for example, famine, drought or epidemic) that are perceived
to be the result of breach of taboos. Taboos thus become broken up
and extended to form classes and rules, and, as part of this pro-
cess, we come to label things and activities as right or appropriate
and others as abhorrent (Douglas 1979). Applying the above ideas to
childhood and research with children, it can be seen that certain top-
ics, for example sex, are perceived as taboo, crossing some invisible
yet powerful boundaries which have been constructed around chil-
dren and childhood to protect the discursively constructed innocence
of childhood itself. The potential implications of this for research
with children are that particular groups perceived as ‘vulnerable’,
particular situations or contexts thought of as ‘inappropriate’ and
particular topics constructed as taboo or sensitive, risk becoming
marginalised by both researchers and the groups and committees that
govern research for the very risks to the social order that they pose.
In Purity and Danger (1991), extending on explorations of taboos,
Douglas theorises that the difference between purity and impu-
rity is a question, like that with taboo and dirt, of matter out of
place. The classification of impure is applied to an object or subject,
which transgresses the overarching social system. For example, the
researcher, who dares to tread paths into exploring sensitive topics,
Boundaries and Battlegrounds 29
or the child who is given the opportunity to speak about topics
that we as adults find uncomfortable. As a result of human need for
order, ‘our pollution behaviour is the reaction which condemns any
object or idea likely to confuse or contradict classifications’ (1991:36).
As Duchinksy highlights, purity and impurity are best thought of as
‘an assessment of particular phenomena in terms of what is imputed
as their relative self identity’ (2013:66). That is to say, the further
a particular phenomenon is from its socially constructed discursive
essence, the more impure or taboo it can be considered to be. Draw-
ing upon this discursive construction of the essential truth of an
object or subject and classification of something as impure as evi-
dence of distance from this truth allows us to explain the fear and
anxiety that research with children often evokes. If innocence can
be thought of as the imputed discursive essence of childhood then
asking children to share their views of topics which are constructed
as taboo or sensitive (by some as inappropriate) thus risks unset-
tling the discursively constructed dominant social order of childhood
innocence and unknowingness.
If, indeed, we are to aspire to Foucault’s concept of egarement –
‘straying afield or distancing oneself from normative definitions
of morality . . . to attempt to gain a critical, usefully knowledgeable
perspective’ (Boden et al. 2009:728), we need to challenge such
boundaries of acceptability. To be ethical in Foucault’s terms is thus
to reflect upon such classifications of normative and taboo and, as a
result, either to comply or to resist the moral order. However, cross-
ing these boundaries leads us to dangerous territory. The experiences
of academics such as Goode (2010) demonstrate that there can be
both professional career and personal emotional costs to working in
these contested landscapes (see also Channel 4 Press 2014). The expe-
riences of a fellow academic, Ken Plummer, who was lambasted in the
press for exploring at an academic conference the social construc-
tion of paedophilia are also indicative of this (Gilligan 2014). In the
reporting, there was an expression of disgust that public money for
social research would be spent, supporting such events which were so
clearly outside of the normative moral order of accepted knowledge.
The impact of this can be felt amongst bodies of knowledge as top-
ics become sanitised, inhibited and censored. This impact is not just
amongst the academics who may come to fear they cannot ask but
also amongst particular groups who never get the chance to answer.
30 Ethical Research with Children
Conclusion
The problems identified in this chapter coalesce to provide a picture
of ethical approval in research with children as far from straightfor-
ward and, perhaps more pertinently, inherently contradictory. These
contradictions exist in terms of what research can be undertaken with
children, who has the authority to undertake such projects and how
children are viewed within these processes. We have also highlighted
that, despite calls for increased multi-disciplinary approaches to eth-
ical regulation (Alderson and Morrow 2006), individual RECs often
act at cross-purposes prioritising particular philosophies, discourses
and bodies of knowledge. This in itself poses serious challenges for
all researchers as they attempt to appease competing agendas. It runs
the risk of closing down research with children, particularly for new
or student researchers without the resources, time or power to nego-
tiate such contradictions and to challenge associated decisions and
decision-making processes. It can be argued that this accumulation
of power by institutions and their management is the product of
the rise of managerialism and declining trust in professionals (Boden
et al. 2009) in the context of the risk society (Beck 1992; Giddens
1991a). It certainly seems this way when we consider the prominence
of requests for third-party involvement in researcher’s interaction
with children and/or the research data, the positioning of children’s
consent as diminished in comparison to adults and the labelling of
particular topics as sensitive and thus to some extent prohibited or
at least marginalised. Our experiences with our own university REC
have been far more positive than that described by Monaghan et al.
(2012) or Skelton (2008); however, there are vast differences in the
requirements and philosophies of such committees as the student
experience, articulated above, demonstrates. This can create signifi-
cant problems for even the most experienced of researchers as they
attempt to please and appease competing agendas and pose a risk for
new or student researchers by closing down research with children
altogether where a middle ground between agendas and priorities is
not easy to find.
By reflecting critically upon such research governance, it is pos-
sible to explicitly consider the wider philosophies, classifications
and moral orders which underpin such decision making. Using
Foucault’s concept of the ‘examination’ to conceptualise ethical
Boundaries and Battlegrounds 31
review processes allows us to see them as disciplinary strategies that
combine techniques of an observing hierarchy and those of normal-
ising judgements. It is a normalising gaze, a surveillance that makes
it possible to qualify, to classify and to punish’ (Foucault 1995:184).
This examination can be situated within the wider framework of gov-
ernance and governmentality whereby ethics review processes are
conceptualised as strategies of power. The positioning of children as
vulnerable subjects, incapable of providing full consent to take full
ownership of their voices or to comment on particular areas of social
life or their own experiences, reinforces the discursive power of the
labels ‘adult’ and ‘child’ within wider society. Even if we take the posi-
tion that children are perpetually othered within the research process
as a result of inherent differences between ‘adult’ and ‘child’ (Lahman
2008), it is still possible to view the disciplinary processes that gov-
ern ethics review as contributing to the powerlessness of children
and, indeed, researchers. When children are denied a voice about a
significant aspect of their lives, they are denied the ability to rep-
resent themselves as they would want to be seen (Holland 1992).
By relying on terms such as ‘assent’, insistence on often inappro-
priate third-party involvement and positioning certain topics as out
of bounds, sensitive or taboo can be viewed as disciplinary strate-
gies that reify children’s subordinated and vulnerable status, creating
research which may be considered as less ‘risky’ but is simultaneously
potentially less ‘ethical’.
2
Ethical Spaces and Places
Introduction
The discipline of children’s geographies reached critical mass in the
late 1990s (Holloway and Valentine 2000) with the dedicated Chil-
dren’s Geographies journal launching in 2003, now issued six times
per year. This signalled an increasing interest by social scientists in
the spaces and places of childhood and the role of space and place
in children’s lives. Given this interest, it is no surprise that children’s
geographers and other childhood researchers have, in recent years,
begun to pay increasingly explicit attention to the role of space and
place in the conducting of research with children (Barker and Weller
2003; Pole 2007; Wang 2012).
The concept of place is an abstract one, but this has been defined
by both geographers and, more broadly, social scientists as a combi-
nation of ‘physical place, understandings of place . . . and processes
of place making’ (Clark 2013b:273). Following the work of Pink
(2012:24), place can be considered as ‘actual experienced environ-
ments and the practices that form a part of these’. Given these
definitions then it is imperative that the principles and practices
of research with children should pay significant attention to the
places and spaces within which such research is conducted as signifi-
cantly influential on all aspects of the research process. This includes
everything from the development of guidelines for ethical research
with children as a product of specific cultural, geographical and
temporal thinking (Abebe and Bessell 2014); the provision of infor-
mation to potential participants (David et al. 2001); the identity and
32
Ethical Spaces and Places 33
positionality of researchers in particular fields such as the child’s bed-
room (Pole 2007); the conditions under which informed consent is
given freely or perhaps otherwise (Heath et al. 2007); and the gath-
ering and dissemination of data (Corsaro 1997). This chapter aims to
critically consider some of the ethical issues that we as researchers
have encountered when planning and conducting research across
particular spaces and places. Here we position research encoun-
ters as socially and spatially embedded, notably in the home, the
educational institution and within virtual spaces. As Herz (1996)
argues research is context-bound and the associated ethical mine-
fields should be addressed in a situational way. We thus argue that
different places pose a range of situated opportunities and challenges
in research with children. These are often not (with some notable
exceptions such as Young and Barrett 2001; Barker and Weller 2003)
explicitly evaluated or subject to critical reflexivity for their impact
on the participants, researchers or data as well as their interrelations
with wider methodological and ethical principles of research with
children.
Research in the home
Ethical decisions are situated and contextualised (Simons and Usher
2002; see also Plummer 2001) and are taken in the moment
(Wall 2010; Zignon 2008). As such Guillemin and Gillam (2004)
have argued that research guidelines are insufficient in preparing
researchers for the complexity of the field. In consequence, some
such as Yee and Andrews (2006) (see also Jordan 2006) call for more
reflexive debate and literature to explore the situatedness and com-
plexity of these ethical encounters. In this section, we discuss some of
the challenges we encountered when the research field is someone’s
home, and we as researchers are cast and required to perform at least
two identity roles: guest and researcher. We use our research examples
with adults and children to discuss issues related to the management
of self; building a rapport in the field, and the ethics of reciprocity;
and negotiating the social conventions through which interactions
between researcher and respondents in their home are organised.
Children are situated in specific geographical places where the
institution of childhood is socially organised and reproduced. As
such many children can be argued to gravitate between two specific
34 Ethical Research with Children
worlds: home and school (Jenks 2005). It is in these places where chil-
dren are assumed to belong and are often perceived as being out of
place when in other locations (ibid.). It is therefore unsurprising that
these places are the spaces where much of the research undertaken
with children occurs (Livingstone 2002; Zelizer 1985).
Idealised notions of the home underpin our social understanding
of this private space. It is perceived to offer respite from the tribu-
lations of the public sphere, and, in the face of evidence to the
contrary, the home continues to be socially constructed as a harmo-
nious idyll – a place of safety, security, love and support that we each
aspire to be part of (Allan and Crow 2001). This private space is inex-
tricably linked to the belonging associated with being a member of
another equally idealised institution: the ‘family’ (McKie and Callan
2012). Long-standing research into both these institutions has high-
lighted that each can frequently mask complex social relations where
differentiated practice and unequal power relations, based on age,
gender, ability and income, exist (Crompton 2001). Such inequal-
ity and abuse of individuals in the home has, historically, been easy
to overlook in the face of the idealisations and assumptions about
the home and the family relations therein. As Aarsan and Aronsson
(2009:499) argue, ‘families are political bodies’, the home is thus a
political and politically contested arena.
Researching with children in environments familiar to them has
been part of the journey that research methods have taken in a
post-positivistic period (Guba and Lincoln 1994) where the rise of
qualitative methods which value the expertise and subjectivities of
children has emerged as being not only the most ethically appro-
priate, but also effective in highlighting aspects of children’s lives
beyond the traditionally dominant subjects of education and devel-
opment. The home as an assumed place of familiarity and security
for children has perhaps not surprisingly become a common loca-
tion for this research. Environments which are assumed to be familiar
to children – nursery, home, school being typical examples – have
been argued to promote more effective participation as a result of
that familiarity (Harden et al. 2000; Lewis et al. 2004). Hill (2005; see
also Mayall 2000a) has argued that the informality of home is also
more conducive to children feeling in control. Such empowerment
is representative of a ‘canonical narrative’ (Bruner 2004) in contem-
porary research with children but perhaps is more representative of a
Ethical Spaces and Places 35
‘persistent illusion’ (Segal 1999:118) in many children’s lived experi-
ences. Power relations that shape children’s lived realities are rarely
that unambiguous, and it is these nuanced power relations within
each home that is a pivotal and recurring theme of this discussion.
The living room as a research field
Research in the home is not without dilemmas, both social and
ethical, and has been described as a ‘triangle of conventions and
negotiations’, where the researcher is also a guest and obliged to
comply with particular social rules (Mayall 2000b:116; 2008). Yee
and Andrews (2006) also emphasise this guest role and argue that
this identity is inextricably linked to the ways that the researcher
is perceived and the social conventions that the researcher follows.
It therefore cannot be divorced from the knowledge constructed.
As Jordan (2006:172) argues these challenges are not unique or partic-
ular to the home setting but can be amplified by the more ‘intimate
and private setting’. Where in the home the research takes place also
brings with it dilemmas and constraints (Pole 2007).
In Sarah’s research the interviews with parents mostly took place
in living rooms or at kitchen tables with coffee and biscuits. Her
preconceived ideas about the homes where interviews would take
place occasionally became explicit. On one occasion, she googled the
address for directions and found distant images of the rooftops of a
large and impressive farmhouse. Sarah accordingly wore clothes that
were perhaps more smart than usual and certainly less practical. In
reality, the large farmhouse was in part derelict and under extensive
renovation with only a few rooms habitable. Sarah’s shoes sank into
deep, winter mud as she approached the front door on a very cold
February day. She shared coffee in the kitchen with her participant
and a rather boisterous hen, and the interview itself took place in
front of a small fire in an otherwise unheated building. The farm cat
shared Sarah’s space on the sofa and periodically climbed onto her lap
for affection. Sarah’s main concern throughout the interview was not
to let the interviewee see that her teeth were constantly chattering,
thinking that it might make the interviewee uncomfortable.
Some parents wished to be interviewed with their children at home
but without their children being present. They set up a variety of
diversions for their children. One mother set out toys on the patio
36 Ethical Research with Children
and had invited a friend of her daughter’s to play. Their play was
observed by the two of us through French windows as the interview
progressed. Throughout this time, the girls knocked on the windows
with a variety of requests and came in for refreshments. The mother
apologised and responded in various ways. Each time Sarah suggested
that it was fine and thanked the mother once again for taking part.
Such ongoing negotiation of her presence in this home was a sig-
nificant feature in how the research was conducted. As Lull (1985)
has argued, the presence of the researcher in a home with a small
number of inhabitants can be highly conspicuous. Before the inter-
view ended, the girls came back into the living room, and the final
discussions were more clipped and less effusive with some responses
from the mother portrayed by gesture more than verbal content. This
also inhibited Sarah’s enquiries as she followed the mother’s lead and
avoided any inference that would draw the attention of the children
in the room. Researchers in the home can exert little control over
who is there and what impact they may have, but requesting alter-
native arrangements is not straightforward as a researcher seeking to
maintain good relations, or appropriate, as a guest where a balance
must be struck between the two roles (Jordan 2006).
One mother put on a DVD at one end of the room as the inter-
view took place at the other. Hushed voices were used so that the
children would not hear. This level of volume was not discussed by
the interviewee, but when she responded to the first question in this
way Sarah did the same in the following interactions. This was so
successful that on occasion the digital recorder picked up the DVD
sound over the quiet conversation. Jordan (2006) argues that par-
ticipants can determine the quality and the climate through which
these interviews take place. On some occasions, young children in
the house who were not respondents wanted to play with Sarah who
found herself variously being a guest at a tea party, a patient in a game
of doctors and nurses, required to cuddle a loved toy and admire a
painting or two. On one occasion, a younger sibling sat on Sarah’s
lap throughout the interview with her elder sister. Interspersed with
the conversations for the interview are long-detailed descriptions of
favourite colours, television programmes, teachers at school and best
friends from her respondent’s younger sibling. Her mother tried to
entice her away on a couple of occasions, but she refused each time.
Asking her to leave was not possible or appropriate, but her presence
Ethical Spaces and Places 37
made the desired conversations with Sarah’s respondent much harder
to achieve and record.
Sarah also had to respond to enquiries about her presence in homes
which sounds simple enough but complimenting information that
the parents of these young children may have said about her pres-
ence needed some careful consideration. Sarah took part in these
unscheduled activities in the interests of maintaining good guest rela-
tions as well as good research relations. Jordan (2006) suggests that
the relationship the researcher develops in the home environment is
less conventional than other interviewer/interviewee relationships.
First and foremost, Sarah was cast as a guest in these homes, and this
ensured that Sarah performed that role even at the expense of her
researcher role. Stepping out of the roles assigned to her, and into the
one she wished to perform in a space that Jordan (2006:174) describes
as ‘families own “turf” ’, deserves consideration and is, according to
Coffey (1999), underexplored.
The management of our identities
Narratives which emerge from fieldwork are situated, contingent and
performed based upon the category entitlements (see also Chapter 5)
of those present and the particular identities that the narrators select
to perform (Mishler 1986). As Lawler (2014) contends, identities
can be understood as being produced through the stories we tell.
We argue that the location where the narratives are told influences
the identity performances and subsequent stories which emerge.
Such cultural behaviour can be perceived as a performance which
has a long-standing tradition in critical ethnography (see, for exam-
ple, Conquergood 1991; Denzin 2003). Carlson (2004:73) claims that
performance theory is premised upon the recognition that
our lives are structured according to repeated and socially sanc-
tioned modes of behaviour [which] raises the possibility that all
human activity could potentially be considered as ‘performance’.
The performance of a researcher identity as a mode of behaviour
in the home has been identified as underexplored and complex to
manage (Coffey 1999; Yee and Andrews 2006). Maintaining a social
distance is a research skill that is arguably required to reduce the
38 Ethical Research with Children
potential of contaminating data. Yet establishing a rapport with
respondents in their homes are also necessary skills for successful
research endeavours. They are also social skills where reciprocity,
self-disclosure and the building of an identity through narrative con-
versation are culturally expected. Holmes (1997) claims a gendered
element to this reciprocity arguing that female listeners are more
likely to contribute in this way. Gilligan (1993) links the kinship
work of maintaining relations to social constructions of femininity
contending that the capacity to be autonomous is, for women, com-
monly linked with an emotional responsibility for, and attachment
with, others. These emotional responsibilities are not suspended dur-
ing the interview process but are inextricably linked to the need to
establish and manage a rapport between researcher and interviewee.
Cerulo (1997) claims that subjective identity causes us to address
two key questions: ‘Who am I . . . and how should I act’ (Alvesson et al.
2008:6). Researchers are required to perform or act an explicit iden-
tity role that is not always compliant with the position respondents
put them in as a guest in their home. These competing identi-
ties within the intimacy of the home can make remaining at what
many researchers regard as an appropriate research distance, chal-
lenging to negotiate. For example, respondents often asked questions
about Sarah’s own adoption experiences having shared theirs in the
interviews. Oakley (1981) refers to these situations as, when ‘ques-
tions are asked back’. Maintaining good and guest relations within
a semi-structured interview made it awkward at times not to con-
tribute stories upon request and risk damaging the rapport needed
for effective data collection. Examples of such self-disclosures are pro-
vided in Chapter 5 on category entitlement in this volume. Sarah
gained access to the field in part through her status as an adoptive
mother. Withholding this aspect of her self is not something she
necessarily felt empowered to do in certain interviews though what
was revealed was always the minimum amount that relations as a
good guest would allow. Coffey (1999) argues that diverse values and
expectations will be used by participants during such interviews and
perceiving the researcher as a guest is often a role that interviewees
have difficulty relinquishing (Bird 2003). Sarah’s interviews often
lasted for a number of hours and she frequently found herself being
invited for lunch or dinner. On three occasions the original arrange-
ment included a meal at the insistence of the respondents. It was
Ethical Spaces and Places 39
therefore constructed at the outset as a social occasion by these
respondents.
Coffey (1999) and Luff (1999) both argue that participants can
frequently be active in determining the nature of the interaction
between the researcher and the researched. First and foremost Sarah
was treated as a guest in these homes and this ensured that she per-
formed that role even at the expense of her researcher role. This
negotiation often used category entitlements (discussed in Chapter
5) and competing roles of researcher and guest which were hierarchi-
cally organised and dynamic within each interview relationship and
social context.
Maintaining good relations is instrumental in the construction of
the following narrative where a professional social worker talks to
Sarah about people like Sarah. Intercountry adoption is uncommon
in the UK in part due to professional anxieties concerning how chil-
dren can acquire the necessary skills about managing racism from
parents who have or are assumed to have little or no experience of it
(see, for example, Barn and Kirton 2012). Other concerns which high-
light how such adoptions are regulated and exploited internationally
(see, for example, Smolin 2006), also ensure that these adoptions,
though possible, are unusual in the UK. Treated as a private endeav-
our the financial costs of such adoptions are high and this cost is
met by the applicants. This fosters a general assumption of wealth
attached to those who adopt this way. Such topics are not easy to
discuss when the researcher asking the questions is known to be
an adopter and the person responding to these questions is a social
worker. These dynamics are only further complicated by the conven-
tions of being a guest. In this case, the interview was taking place in
Sarah’s home at the request of the interviewee.
In the following account, the social worker rather neatly turns the
negative ‘consumer’ category frequently attributed to intercountry
adopters into autonomous and neo-liberal aspirations that become
seemingly complementary to those who engage in such adoption
applications.
Anne: ‘The commitment is huge, the effort is huge. People who
are used to negotiating their lives who make decisions and have
successful careers, they are also used to getting their own way
and having a level of success. They are therefore more confident
40 Ethical Research with Children
because that is their experience. If they work hard they get to
where they want to be and undertaking ICA [intercountry adop-
tion] is a mammoth operation. You wouldn’t do it if you were a
shy retiring individual. I suppose that they pay for the process
and perhaps recognise that they are purchasing a service and
are used to getting what they pay for. In my experience they
are not as tentative about being assessed as domestic adopters
can be. They are also less intimidated by the professionals in
the process. That is not to say that confident people do not
adopt domestically because they do. ICA’s by and large are
more confident about what they are doing and have done their
research.’
The narrative above performs numerous roles. It uses the negative
attributes commonly ascribed to intercountry adopters and yet man-
ages not only to complement those involved but suggests that the
adoption process actually requires these attributes (Richards 2013).
Anne is aware that she is speaking to an intercountry adopter and
this knowledge shapes what she says and how she says it. These narra-
tives then are nuanced cultural performances (Denzin 2003). Anne’s
narrative constructed these identity traits as she sat at Sarah’s kitchen
table. The extent to which guest relations influenced Anne’s narrative
is not possible to determine without a further contribution from her.
Jordan (2006:182) calls these influences ‘tangible but immeasurable’.
Rarely, if at all, have the attributes that Anne identified been regarded
as beneficial qualities for potential adopters to display. More usually
they form part of a critical consumer narrative about intercountry
adopters who neglect the best interests of children in their pursuit of
a child (Richards 2014).
Sarah’s interviews with children took place in a variety of locations
throughout their homes. On one occasion she was invited to use the
table in a conservatory with a plastic roof. The draw and write activ-
ity began well enough but a very heavy rainstorm ensured that the
digital recorder recorded the sound of heavy rain on the plastic roof
for about 15 minutes, making the conversations almost inaudible.
Putting the host in a position of finding an alternative space was not
something Sarah felt able to do and so sections of this participant’s
conversations were lost to the rain and the maintenance good guest
relations.
Ethical Spaces and Places 41
All the interviews with the children took place either within
earshot or with parents remaining very close by in another room.
Most parents suggested that they would remain out of sight but be
available should they be needed. This is of course understandable
with young children and a researcher that they are not familiar with
however it did raise concerns about confidentiality that were never
really resolved. In one interview, a mother remained in the room
with her daughters when they did the draw and write activity, she
made suggestions about what they might write in the various pages
of the life book activity and prompted them about the information
that they might verbally contribute. At times, she attempted to dom-
inate their interviews. However, it became apparent that the elder girl
(aged 12 years) was not acknowledging her mother’s contributions or
including them in her life book, despite her mother’s growing frustra-
tion. What Sarah gradually became aware of was the performance of
a rather typical mother–daughter relationship where power relations
were being tried and tested with Sarah’s research being one of the
battlegrounds. The more the mother wanted her daughter to speak,
the more her daughter refused (see Chapter 4 for further discussion),
while filling other pages in the life book with her dreams, aspirations
and friendships. In contemporary discourses surrounding children
ways to articulate their voices are given paramount focus with lit-
tle consideration of how their silences are used or what they might
mean. The following section taken from Sarah’s PhD thesis expands
on this use of silence in family relations within the home (Richards
2013; Christensen and Prout 2002; Kellett et al. 2004).
Silences as relational strategies
Haavind (2005:142) turns the contemporary notion of the child
having the power to speak somewhat on its head by questioning
the ‘potential power for children in saying nothing’ in situations
where independent opinions on shared matters are not possible to
enunciate. Her argument here is that silence becomes a strategy of
participation. My data suggests it to be a strategy that some of the
girls use in family relations and in a research context. In the fol-
lowing exchange, the younger sibling demands her elder sibling’s
compliance to the colour order of the crayons that she has devised
and which they are using to complete their books with me.
42 Ethical Research with Children
Rosie: ‘May, can you put them back in order please or then I have
to sort them out again.’
May: [silence, no response or acknowledgment that she has heard]
Rosie: ‘What colour did you just use?’ [she shuffles the pens
between them at this point]
May: ‘Skin colour.’
Rosie: ‘Skin colour and grey so that goes just there [places them
back in her order] ‘Black and red.’ [then says] ‘Remember to put
them in order, red goes there remember?’
May: [silence but does not comply]
Rosie: ‘Oh May I told you to put them in order, why do you have
to keep doing that?’
May: [silence but smiles]
Rosie: [says in a raised voice] ‘It’s not funny.’
May: [silence]
Rosie: ‘Put them back in order.’
May: [ignores sister but to me asks] ‘Do I have to colour in all the
skin colour now?’
I respond, ‘No.’
May only uses two words to respond to her sister’s remonstrations
of the colour order in this exchange. Silence seems to be used as a
powerful strategy developed in this sibling relationship. This silence
is not about a non-response but is actually more indicative of a way of
responding that neither agrees, legitimises nor exacerbates her sister’s
demands. However, the smile may indicate that she is aware that she
is aggravating her sister (and not finding it too onerous to do so).
I contend that silence here is a strategy tool that this sibling is using
with good effect.
Current research with children promotes their empowerment to
speak rather than a right to silence (Lewis 2010). It may have become
more of an obligation and expectation than a right for children.
There is recognition in the above exchange that silence is neces-
sarily neither neutral nor empty of meaning and, therefore, that
ethically informed interpretation of it should be reflexively included
in analysis. Borland et al. (1998) argue that children are able to
reveal or withhold aspects of themselves in context-specific situa-
tions, again revealing a limitation to the term ‘participation’ where
children are concerned. Silence may be the compromise position
assumed by some family members between being submissive and
Ethical Spaces and Places 43
an open rejection which would potentially destabilise the family
relations.
Guest, researcher or teacher in the home?
Despite much research being conducted in the homes of children to
enhance empowerment, informed consent and familiarity (Harden
et al. 2000; Hill 2005; Lewis et al. 2004), we question the extent to
which the role of researcher is dramatically different from that of a
teacher in an educational setting. Waller and Bitou (2011) claim that
current methods often used in participatory research with children
induce pedagogical relations. Sarah’s use of a draw-and-write activ-
ity was not intended to emulate school work, yet inevitably it did
though this varied in each collection context. In one instance, the
journal page said, ‘This is a story about where I was born in China.’
One participant read this, went to a bookshelf, looked up her birth
province in the index of a tourist guide and copied out what was
written about geographical location, proximity to the sea, popula-
tion size and historical aspects to be seen if visiting. This was done
in silent concentration and far more indicative of a set homework
activity than the telling of a story (Figure 2.1).
Wim (aged 8 years), another participant in Sarah’s research, con-
firmed her perception of the similarity between Sarah and a teacher
throughout the interview by her continued compliance and deferral
to her in decision making, and also through requests for guidance in
spelling and for what I wanted her to write in order to address cor-
rectly the pages in the book. The expectation of being a compliant
and good pupil often ascribed to girls is evoked in her transcript and
her journal entries. Waller (2010:530) argues that school can embody
and reproduce ‘particular forms of masculinity and femininity’. Sarah
writes ‘I am uncomfortably aware that I selected a method in which
this identity trait was profitable for me as a researcher even though
my research took place in the children’s homes’ and supposedly was
a place where they are argued to feel more in control.
As Sarah reflects on her approach in her thesis (Richards 2013:157)
A researcher’s implicit or explicit use of power is to their advan-
tage whether it is acknowledged or not, making such methods
and this methodology less innovative and more reflective of pre-
vious approaches than some would hope. Jones (2001 cited in
Gallacher and Gallagher 2008:505) argues that these methods
44
Figure 2.1 This is a story about where I was born in China
Ethical Spaces and Places 45
have less to do with emancipation of children’s voices and are
more representative of the ‘colonization’ of childhood by adults.
However, I argue that the socially constructed space of childhood
is already adult space and that this further incursion by adults rep-
resents a new way of exploring how children manage, negotiate or
challenge the space that we have constructed and ascribed them.
I have reached an uncomfortable conclusion that, as a researcher
I temporarily occupied the individual spaces of these girls, using
a method which reflected their pupil role, justifying this incur-
sion in part by the use of the language of inclusion and child
centredness, and an assumed insider status as an adoptive mother.
I also remain uncertain as to whether I have elevated the voices of
those less able to speak about intercountry adoption in the spirit
of ethnography, or participated in othering this group even fur-
ther. The production of knowledge is not removed from notions
of power (Foucault 1995), and it can be argued that for chil-
dren to participate in knowledge construction about their lives
is to ‘encourage them to take part in the processes used to reg-
ulate them’ (Gallacher and Gallagher 2008:504). Such tension
concerning power cannot simply be resolved through the use of
‘child centred’ participatory methods where it is suggested that
the children are actively involved rather than passively recipient.
Educational institutions
Christensen and James (2001) argue that the spatial-temporal order-
ing of school shapes children’s everyday experiences and Pike (2008)
highlights the regulation of children in institutions through a range
of socio-spatial strategies. Issues of power and control are central to
an analysis of childhood space (Smith and Barker 2000), and edu-
cational institutions are key organisations where children’s lives are
constructed by adults and where children seek to carve out space and
time of their own. School spaces thus should be critically consid-
ered not just as environments for children but as part of wider social
structures where power is articulated and as places where meaning is
negotiated (Clark 2013b). Critically considering issues of power and
agency as articulated within and through the institutional spaces of
childhood (the school, the early years setting, the children’s centre)
46 Ethical Research with Children
has resulted in a plethora of critical research which examines not
just ‘traditional’ educational and institutional issues but all aspects of
children’s lives, including, for example, the role of different forms of
childcare on shaping expectations of good parenting (McDowell et al.
2005) and the school dining room as a site of intervention regarding
children’s nutrition and well-being (Daniel and Gustafsson 2010).
This section of the chapter argues that the organisation of time and
space and the distribution of power in educational institutions play
an important and unavoidable role in research conducted within,
through and about such spaces, discourses and cultures. Here we
will consider the impact of institutional pressures on the provision
of informed consent, the positions of power avowed to adults in
such settings and children’s negotiation of these assumptions and the
role of gatekeepers in selecting, silencing and empowering children
within such settings.
The issue of informed consent, discussed previously in Chapter 1,
dominates discussions of all ethical research and is considered par-
ticularly important and complex in terms of research with children.
The basis of the assumption of the importance of informed consent
is one by which we should respect the autonomy of individual deci-
sions to participate. The assumption here is that children have the
competency to consent and that this conception of them as agentic
beings is universal (see further discussions of assent and consent in
Chapter 1). Aside from the fact that this model of consent is based
on a neo-liberal individualist model whereby human nature is con-
ceptualised as fundamentally Western and masculine (Hammersley
and Traianou 2012), as discussed in the previous chapter, it is also
clear that the ability to meaningfully participate in decision mak-
ing is shaped by the environment in which information giving and
consent seeking takes place.
It is well documented that children in institutional settings such
as schools may feel pressure to consent to participate in research
whether this is to be seen as cooperative by those in powerful posi-
tions, ‘pleasing’ a teacher or following school rules by being polite
to visitors (Heath et al. 2007). Indeed, high response rates achieved
in school-based research is argued by some to be rooted in this hid-
den pressure (Denscombe and Aubrook 1992). David et al. (2001)
discuss the information sessions they provided to the young peo-
ple who would potentially participate in their research. They chose
Ethical Spaces and Places 47
whole-group sessions as the young people would be familiar with
such ways of interacting as a result of their schooling experience,
to adhere to their philosophies of assuming competence and as part
of considering informed consent negotiations as ongoing (Alderson
1995). Despite this the position of the researchers within these infor-
mation sessions was one of ‘quasi-teacher’ and as such like many
researchers in schools they were able to make effective use of exist-
ing adult/child power relations. The spatial-temporal regulation of
the school day itself allows for such whole group sessions to take
place as children are ordered (by age rather than educational com-
petence David et al. 2001) and adults in such spaces are positioned
as authoritative knowers rather than genuine questioners (Graue and
Walsh 1998). We all therefore as childhood researchers benefit from
accessing such spaces where children and young people’s lives are
structured by existing power relations.
Such socio-spatial regulations do not always provide benefits to
the researcher, particularly where they actively prohibit the voices
that can be heard in general or indeed in specific spatial or tem-
poral environments. Allison found in her research that a complex
interplay of power operated within specific times and spaces within
the school environment. She concluded that the young participant’s
voices were both controlled and restricted by adults. For example, she
observed that during play time and lunchtime the children packed
(or were helped to pack) their Augmentative and Alternative Com-
munication Systems (AACS) away. While the children were strongly
encouraged to use the AACS’s in educational spaces such as the class-
room, Allison discovered that school policies and health and safety
regulations imposed formal restrictions on voices in social spaces
such as the playground, the dinner hall or on the school bus. She also
noted that during the day, some children’s physical access to their
AACS was restricted. For example, one participant’s AACS was kept in
a rucksack on the handlebars of her wheelchair. She could not access
what she called ‘her voice’ without help and consequently had to rely
on other forms of communication to make herself heard and under-
stood. Indeed, the programming of many of the AACS’s, completed
by adults, focuses almost entirely on curriculum based language. Here
we can see that adult perceptions of the need for children to commu-
nicate, to have a literal voice, are dominated by a focus on education
and learning at a cost to the perceived importance of voices in the
48 Ethical Research with Children
social spaces of school life. This can be seen elsewhere in research
by Harden (2012) who documents the regulation of movement of
children in a primary school classroom as entirely restricted by essen-
tial tasks for learning rather than for social interaction or emotional
expression. In addition, Allison noted that the practicalities of setting
up the AACS served to restrict its use:
Sharon: ‘Yeah, Justin should have it [the communication aid] with
him all the time, but his chair is heavy enough anyway without
the communication aid weighing it down . . . ’
Allison: ‘It is quite big, isn’t it? And does it take a lot of effort to
get it on the tray?’
Sharon: ‘Yeah and they’re not easy to use when the sun is shining,
so you think, aghhhh, is it really worth the investment?’
This relates to Priestley’s (1998:219–220) suggestion that the reg-
ularisation of socially structuring disabled children as dependent,
vulnerable and passive has served to ‘de-sensitize us from their
agency as social actors’. Allison concluded that external structures
of power such as those illustrated above regulate and restrict chil-
dren and young people’s voices and simultaneously highlight how
oppression can be routinised in the daily denial of agency (Watson
2003). The justification of this as part of school policy or health and
safety regulations demonstrates the power of such institutions over
their members and in turn, their ability to participate meaningfully
in all aspects of school life. Such restrictions have radical implications
for the construction of data, the establishment and maintenance of
research relationships, and crucially for the discussion here, for the
giving of informed consent.
In another example from Allison’s research (discussed in detail in
Chapter 3), she actively aimed to respond to the differing needs of
the range of children and young people who she sought to partic-
ipate in her research. As such the ways she managed the giving of
information and the negotiation of informed consent was not part
of a whole group session that aimed to tap into wider commonly
used and thus familiar pedagogical techniques but instead differed
for each of the young people she approached. Despite childhood
researchers advocating the importance of participatory methods and
child-centred research to empower participants through all parts of
Ethical Spaces and Places 49
the research process such approaches are by no means exempt from
concerns regarding the potential problems associated with power and
authority (Veale 2005 in Heath et al. 2007). Sarah’s previous example
earlier in this chapter regarding her ‘write and draw’ activity illus-
trates this well. Despite a research philosophy committed to listening
to the voices of children and the provision of outlets by which such
communication could take place the activity she designed did inad-
vertently replicate the kind of work children are expected to engage
with at school. As such Sarah discussed some of the girls whom she
researched with asking her questions regarding spelling, presenta-
tion, clarification on the tasks. In this sense these girls placed Sarah
in the role of a ‘quasi teacher’ in the much the same way as the roles
David et al. (2001) played in their whole group information sessions.
What the previous example demonstrates, alongside the manifes-
tation of adult/child power relations in research interactions with
children, is the fluidity of positionality and power. Despite concerns
regarding the power of adults in educational institutions where social
spatial regulations are constructed to rely upon, maintain and rein-
force adult/child power relations what Sarah’s example demonstrates
is that children hold a great deal of power in these situations also,
much of which goes unrecognised. As Barker and Smith (2001) high-
light in an example from their research that despite what may seem
to be the inevitability of adult power over children within such spaces
this is not necessarily the case. When Barker decides not to play foot-
ball with the boys as they are excluding girls from their activities
‘some of the boys withdrew . . . their consent for the research’ (Barker
and Smith 2001:144). In such spaces, it is important to recognise
that children are the experts and engage in multiple performances
as they negotiate what is required of them within such hierarchies of
power. We enter as researchers occupy positions of power as adults
but often lack the expertise of children, when significant proportions
of their life are spent within such spaces, and they have developed
and work to maintain in a diversity of ways. As Foucault (1998)
argues, power is not fixed but slippery and exercised from innu-
merable points. The previous discussions of the potential influence
of such spaces on the provision of informed consent highlight that
power can be and is often structured in expected ways; that is, adult
authority in schools. What is clear from the other examples pre-
sented is that power is more diffuse than simply this one-directional
50 Ethical Research with Children
process; it is everywhere and can produce contradictory outcomes
(Foucault 1998). Sarah sought to minimise unequal power differences
between herself and the children she sought to research with but they
positioned her as a powerful adult with answers to their questions.
Indeed, when Sarah made it clear to one participant that she was in
control, the girl in question simply laughed. With this simple act this
participant made it clear that she was well aware of her position as a
child and Sarah’s as an adult and the implications of these positions
for their interactions and relationship.
Despite the dominance of age based assumptions about children’s
competence research has documented that spaces of education and
of care are interpreted and utilised by even very young children to
manage power relations very effectively. Gallacher (2005) highlights
how two–year-old children make active use of the spaces and adult
imposed rules of a nursery setting in attempting to appropriate space
and time for themselves. Her conceptualisation of the toddler room
as a site of adult control for the ordering of children is entangled
with recognition of the active role of toddlers as social beings in
appropriating such spaces, highlighting the ‘underlife’ of the setting.
What we must critically reflect upon here is our desire as childhood
researchers to establish equality in our research relationships with
children. In addition to questioning whether this is indeed possi-
ble given existing socio-spatial constructions of ‘adult’ and ‘child’,
we should also ask whether this is desirable for children. Does posi-
tioning adult researchers as powerful actually make negotiating and
managing research encounters easier for children? Researchers and
the researched are positioned simultaneously in such spaces by a
number of fields of power which intersect and overlap. Such spaces
and the power relations that flow within them thus require criti-
cal evaluation to consider explicitly their multifaceted roles in the
research process and production of data. This requires us not just
to talk in accepted ways about the empowering of children but to
consider the ‘reality’ of power relations embedded within particular
spaces. Here we consider not just whether participatory methods can
deliver all that they promise (Gallacher and Gallagher 2008) but also
whether all that they promise is indeed all we should desire. Assump-
tions regarding what children want from a research encounter and
what is ‘best’ for children to enable them to meaningfully partici-
pate regard equality as the ultimate achievement. This is not to say
Ethical Spaces and Places 51
that inequality in research relationships is something to be desired
but rather to encourage active consideration of how children actively
manage such situations beyond simplistic declarations of ‘agency’
and our need to recognise this. Seeking to minimise power differ-
ences as influential for all participants within the research encounter
may blind us to reflecting upon children’s ability to use the fluidity
of power to their own advantage.
Power within research in educational spaces does not just flow
between the researcher and the researched. There are a variety of
other individuals that are directly and indirectly involved within
research which takes place within such institutions. These include
a diverse range of parties including parents, relatives, head teach-
ers, local governors and local education authorities all of whom play
a part in the ‘politics of access’ (Barker and Weller 2003:213) – the
decisions regarding whether researchers can work with children. Dis-
cussions in Chapter 3 illustrate Allison’s experience of these complex
chains of consent which somewhat ironically place children as the
last group to be consulted as to their participation. In our experiences
the role of gatekeepers plays a significant part in the negotiation
of access to particular spaces, the exploration of particular topics
and the participation of particular individuals and groups. Parnell
and Patsarika (2011) in their discussion of children’s participation in
England’s since cancelled Building Schools for the Future Programme
are explicit in highlighting the role of adults in choosing those chil-
dren deemed suitable to participate in the research and consultation
thus mediating the range of voices which could be heard.
The need for gatekeepers is indicative of the institutionalisation
of children and young people’s lives (Qvortrup et al. 1994) whereby
increasing amounts of children’s time is spent in institutions for
children and young people separate from the worlds of adults. Inad-
vertently, this often leads to children and young people not only
being marginalised from the spaces of adulthood but their concerns
and voices can be sidelined too (Heath et al. 2007). Gatekeepers hold
the keys to give or withhold access to such institutions. This is not
just a yes or no exercise but gatekeepers can also shape the direc-
tion of the research, the topics explored and the ways in which such
explorations take place. For example, Dockett et al. (2009) explored
family experiences of school transitions and one community partner
withheld families with a history of violence on the grounds that the
52 Ethical Research with Children
research would take place within the home. In addition, DePalma’s
(2010) explorations within primary schools of attitudes to lesbian,
gay, bisexual, transgender (LGBT) individuals and groups was medi-
ated in its use of videos due to concerns regarding parental reaction
and their impact on school life beyond the research. In Allison’s
research, she clearly documents the positioning of particular par-
ticipants by gatekeepers within the setting, including teaching and
support staff, as able to (or, of course, not) contribute in meaning-
ful ways to her research (for further discussions, see Chapter 6). She
was told that Lizzie was ‘difficult’ and would probably not communi-
cate with her. She was also told that she was likely to ‘kick off’ at any
time. Here Allison was forced to negotiate maintaining a rapport with
the adult gatekeepers in a setting while at the same time adhering to
her beliefs in the value of all children and young people’s voices.
As with Sarah’s discussion earlier in this chapter of rapport within
the home and Jessica’s explorations in Chapter 1 of an employee
researching within their own workplace, maintaining relations in a
group and/or setting that the researcher is familiar with within their
personal and/or professional life and adhering to the philosophies of
child-centred research is not an easy endeavour.
The role of gatekeepers although frequently problematised (as seen
in conversations throughout this book), notably in relation to their
role in gaining informed consent within institutions, also offer
insight into worlds that researchers may be unfamiliar with. As
Flewitt (2005) points out in her research, in an early years setting staff
were able to offer a unique insight on occasions into the preferences
of very young children. This did not necessarily entail speaking for
such individuals by offering views on their likes or dislikes but in one
example they were able to notify the researcher of signs in pre-verbal
children of dissent thus potentially contributing to the management
of consent with very young children which is certainly challenging.
In addition in Jessica’s explorations of the ethics of engaging with
children and young people in research in online spaces discussed
momentarily, those children and young people who create their own
online spaces in the forms of blogs, forums and zines (see, for exam-
ple, Harris 2003) are able to act as their own gatekeepers on behalf
of themselves and their peers in negotiating researcher access to such
spaces and the data contained within. This is something that is a
potentially empowering experience although fraught with its own
Ethical Spaces and Places 53
concerns regarding negotiating consent in the absence of ‘corpo-
real’ confirmation of identity, age and adult/parental involvement:
concerns which are not so problematic in educational spaces.
Virtual worlds
Virtual worlds, as briefly highlighted at the end of the previous
section, offer both exciting opportunities and new challenges for
research with children and young people. In fact, online spaces, in
what has been dubbed by some as ‘information age’ (Holloway and
Valentine 2000), exist as a series of juxtapositions. While parents are
encouraged to develop children’s use of technology for the purposes
of education and for later success in the labour market, they are also
supposed to watch carefully and rigorously for the potential risks.
Risks of children’s engagement with online worlds vary from reli-
gious and political extremism to pornography to paedophilia and
grooming and on to cyberbullying (Kitchin 2000). Such risks tend to
dominate conversations about children’s online social interactions,
running the risk of being technologically determinist in their per-
spective (Bryson and de Castells 1994) where technology is divorced
from a wider normative context of social practice.
Children can thus be viewed as simultaneously both technologi-
cally competent and at significant risk from these skills as they use
them to interact online, which is commonly viewed as a public space
(Valentine et al. 2000). Operationalised here are popular motifs of
the Apollonian and Dionysian child which exist in parallel and have
been identified as underpinning diverse policy orientations towards
children (Stainton Rogers 2001). The Apollonian child is concep-
tualised as ‘angelic, innocent and untainted by the world’ (Jenks
1996:73) and can be seen here in the invocation of images of the
vulnerable child at risk in the online, public sphere who must be
protected so that such spaces can be used as effective learning tools.
In contrast, the Dionysian child is considered to enter the world with
a bias towards evil ‘drawn to self-gratification and pleasure, lack-
ing sensitivity and social control’ (Murphy n.d.:6), prone to access
inappropriate imagery or place themselves at risk by conversing
with strangers and whose online behaviour should be surveilled or
restricted in the name of protection, both for the child and for wider
society. It is thus no surprise that it is therefore fears and problems
54 Ethical Research with Children
with online spaces which tend to dominate conversations about
how to conduct research in this arena: for example, how to assure
informed consent (Kanuka and Anderson 2007) or the verification of
identity (Pittenger 2003).
There is no doubt that children are increasingly using virtual
worlds and technologies which allow access to such spaces as part
of their everyday lives. Survey data suggests that half of all chil-
dren aged 5 to 10 own their own laptop, more than 40% own a
mobile phone and nearly all 7–16-year-old children access the inter-
net readily within their own homes (Childwise 2013). These spaces
thus become important landscapes within which to engage with chil-
dren and young people about themselves and their lives. This section
of the chapter draws on Jessica’s experiences planning a project to
engage with girls about their online political activism and the debates
that such planning invoked. Here we will consider how online spaces
can highlight discursive constructions of childhood itself (as with the
innocence and evil discourses highlighted above) while simultane-
ously serving to unsettle traditional dichotomies of public/private
spheres and adult/child relations. In addition, we will examine the
ways in which virtual worlds offer both challenges and opportuni-
ties for the conduct of research with children and for the explicit
consideration of ethics and ethical issues within such domains.
The internet has long been conceptualised by social scientists
engaged in exploring the lives of girls (indeed, all children and young
people) as having some emancipatory potential and for facilitating
knowledge about how girls express their beliefs, engage in relation-
ships and manage in complex ways their own embodiment and wider
cultural trends and social change (see, for example, Aapola et al. 2005;
Harris 2003; Ringrose and Barajas 2011). Jessica sought to expand
upon these explorations in a project to examine the practice of fem-
inist politics in online spaces by girls and young women. The aim
being to explore how girls who self-identify as feminists use online
spaces to present their identities, communicate their beliefs and prac-
tise their politics. Virtual ethnography was the planned adopted
method for this project underpinned by methodological principles
which draw upon both a range of feminist perspectives and the
importance of listening to the voices of children. This was in response
to methodological concerns that orthodox quantitative approaches
dominate studies surrounding young people and politics (Bhavnani
Ethical Spaces and Places 55
1991), which don’t necessarily allow girls and young women to
articulate their own meanings surrounding feminist political partici-
pation and/or activism. At the very outset, this research posed a series
of ethical dilemmas, most of which centred around the online spaces
within which the project was to be undertaken. These dilemmas and
the wider ethical considerations they inspired are discussed here.
Jessica’s initial choices regarding how to engage with girls and
young women, generating their own blogs and zines regarding their
feminisms, centred on the use of virtual ethnography. This would
involve the use of participant observation with conversations with
participants to examine and interrogate the meanings they them-
selves attributed to their online content, politics and identities. This
choice was a deliberate one in order to avoid accusations of ‘lurk-
ing’ (Richman 2007) in online spaces, which is viewed as akin to
covert non-participant observation in face-to-face research. Lurking
is widely conceptualised as entering virtual worlds and not declaring
your presence, as a researcher or otherwise. There are assertions by
some researchers (Smyres 1999) that these environments constitute
public space and are, as such, fair game for the researcher, who can
remain anonymous and silent. This is not an uncommon perspective
and the issue of the dilemmas between public and private texts has
been subject to wider discussion within the literature (see, for exam-
ple, Madge 2007; Spriggs 2010). Some researchers take the view that
such texts when freely available online are open to use by researchers
without the seeking of consent as they are published resources which
are publically available (Pittenger 2003). However, evidence from
online communities suggested that some have expressed disdain at
the invisible presence of researchers who lurk (Hudson and Bruckman
2004), and they have a duty to make their presence known. Indeed
we argue that debate can become problematic when for example,
researchers attempt to disseminate the results of such endeavours run
up against reviewers who consider their use of such data as ethically
unsound. It is not just the views of researchers and academics that
differ in how they conceptualise such data but also young people
themselves, with some research suggesting that young people view
online ‘stuff’ as public (Stern 2004) and some as private (Spriggs
2010).
An interesting way to conceptualise these arguments is to consider
whether we define online content as simply data to be harvested
56 Ethical Research with Children
or whether we conflate such texts with the author themselves as
a human subject (Bassett and O’Riordan 2002). This highlights an
important point regarding not only the use of online spaces for iden-
tity construction but also the situatedness of online research with
children and young people. For example, a young person’s blog post,
fashioned in a similar way to a newspaper article, might be consid-
ered to be ‘text’, but yet a social networking site such as a Facebook
or Bebo page or an Instagram account could be constructed as an
extension to the human subject. However researchers may choose to
respond to such debates, what is key here is an acknowledgement
of the potential, if we are not careful, to homogenise online spaces.
As we have argued earlier in this chapter, the spaces within research
are conducted is crucial for understanding research relationships and
the co-construction of data – this is just as much the case with online
spaces, not all of which are the same. Virtual spaces are a heteroge-
neous category where the virtual geography or architecture is just as
important as in physical spaces when considering the construct of
affective relations (Papcharissi 2009; Ringrose 2011).
We do not attempt to resolve what is now a long-running debate
regarding public and private data online by offering answers here
regarding what researchers venturing into online spaces should
do. As Alleyne (2015:170) points out, in research endeavours it is
researchers who must act as their own ‘ethical monitors’. However,
we do suggest that a ‘free for all’ (Madge 2007) in the virtual worlds
of children and young people (and indeed adults) is potentially
dangerous and ignores many of the debates that online research
has the potential to inspire. Here, it is argued, we should pay
attention to the cultural context of online communities and the
potential for such spaces to inspire new ways of thinking about
ethics. Madge (2007) proposes a hybrid relational model of ethics
that incorporates text, space and bodies and involves a rethink-
ing of procedural ethics which dominate contemporary institu-
tional ethical review (as discussed in Chapter 1) destabilising estab-
lished dichotomies of public/private, embodied/disembodied, writ-
ing/speech and online/offline. As Hine (2005) argues online research
is marked as a special category in research ethics governance and thus
requires new ways of thinking about what constitutes ethical enquiry.
After considering the fierce debates regarding the issue of public
and private in online spaces, Jessica chose not to engage in practices
Ethical Spaces and Places 57
of ‘lurking’ as it was felt that there was a need to involve partici-
pants as fully as possible in the research itself, and lurking covertly
involved a level of deceit she was uncomfortable with. She considered
that the girls and young women owned the data they had themselves
created. In addition, as the internet as a series of spaces can still be
conceptualised as emerging, and experimented with by researchers
and children alike, assumptions regarding public availability remain
inherently ethically problematic for all. In addition, Jessica felt that
as an adult researcher she had the responsibility to declare her pres-
ence in the online spaces she entered and to seek permission and
informed consent from those who contributed to the architecture
of such worlds. As Livingstone (2006) argues, children deeply value
privacy and have made increasing use of online spaces to carve out
places and spaces of their own. They do this as a respite from other-
wise increasingly restrained lives (Livingstone 2006) in order to feel
in control when much of their lives are subject to the surveillance of
an adult gaze. A fundamental reason for Jessica’s decision related to
this issue of privacy was that she felt there was the potential, if she
did not make her presence known, to perpetuate adult/youth power
inequalities where girls’ spaces are subject to an (invisible) adult gaze
(Smyres 1999). This position, however, posed different problems of
its own.
After making her decision not only to declare her presence in the
online worlds of the young people she sought to research with, but
also to engage them directly in conversations, Jessica found that iron-
ically she was now going to encounter more ethical issues than had
she taken the alternative route. Not seeking informed consent to har-
vest publically available data seemed to pose a range of problems,
according to ethics documentation and senior colleagues. This was
despite a commitment, like fellow researchers, to maintain an ethi-
cal commitment to uncover the meanings of digital communication
as part of working with young people (Ringrose and Barajas 2011).
However, the desire to approach the girls themselves, to listen to
their stories and to understand the meaning they attributed to the
online worlds they had created, and contributed to, seemed to gen-
erate a range of issues relating to informed consent, parental consent
and age and identity verification. Jessica was guided on a number of
occasions to abandon the idea of speaking to the girls themselves and
instead revert to a form of content or discourse analysis, using only
58 Ethical Research with Children
the openly available online data. This guidance was premised on the
thinking that her research would be easier or more straightforward
if she in some way bypassed the ethical issues at hand rather than
took the time to grapple with them. Such advice is not uncommon,
particularly in the case of research with children, and anecdotal evi-
dence suggests that inexperienced researchers and those seeking to
work with children in new spaces (such as virtual landscapes) are fre-
quently advised to avoid such endeavours. They are too complex,
too time-consuming and too potentially problematic to warrant time
and attention. What unfortunately is lost here is the potential for
such research endeavours to enable us as researchers to reflexively
engage with wider ethical issues and discursive constructs relating to
research methodologies, online research in particular, or childhood
itself.
The most problematic issue Jessica went on to encounter in the
planning of this project was that of informed consent. This ethical
principle is subject to much discussion and debate in methodolog-
ical literature, as discussed extensively in Chapters 1 and 3, and
remains so in virtual landscapes. Broadly, problems of identity confir-
mation are at the heart of informed consent being so problematic for
online research. There is the potential that it is easier for participants
to deceive the researcher, regarding, for example, their age, than is
the case for offline research. However, this practice is rarely docu-
mented (Hewson et al. 2003). It does pose issues relating to gaining
the informed consent of children and young people (for example,
Stern 2004) and notably to gaining authentic parental consent, con-
sidered a must for research with children. Potential advice in this
arena seems to be dominated by a need to affirm consent (Bruckman
2002) or indeed make contact with participants in offline domains
first (Mann and Stewart 2000). This is a way of managing such issues
and does address the tendency of some scholars to divorce online and
offline worlds by not generating a binary between digital and face-to-
face identities, worlds and interactions (Ringrose and Barajas 2011).
It does, however, remove some of the advantages that the very nature
of online research offers us; for example, the ability to access diverse
populations and to overcome geographical barriers (Madge 2007).
Such a model of informed consent, as already discussed in Chapter 1,
remains based on a neo-liberal model of the rational, individual
self, charged with minimising risk. Children are not comfortably
Ethical Spaces and Places 59
assimilated into such a model when they are constructed within com-
mon tropes of vulnerable, irrational and dependent, and as such they
are not viewed as competent beings capable of providing informed
consent in offline or online spaces.
A desire to hear the voices of girls themselves rather than har-
vest data thus posed a significant problem. Perhaps some of the
anxiety which children’s use of and presence within online spaces
invokes can be attributed to the role that such landscapes play in
opening up the dynamics of traditional power relations between
children and adults. Children and young people have previously
been conceptualised as ‘digital natives’ (Prentsky 2001) as a result
of their superior knowledge and understanding of information tech-
nologies (Livingstone 2006), with adults comparatively characterised
as ‘digital immigrants’. For the purposes of considering research with
children and young people, such discursive constructions of chil-
dren as knowledgeable, yet perpetually at risk, highlight how online
spaces unsettle some of the traditional dichotomies and bound-
aries that dominate not only conversations about childhood but
also research with children such as adult/child or online/offline
or risk/opportunity. As Livingstone (2008) says, all activities are
both risky and opportunitistic, and the reification of dichotomies
of risk/opportunity is not a useful dynamic from which to consider
children’s participation in online worlds or the ethics of engaging
with them in research within and about such landscapes. As Ringrose
and Barajas (2011) highlight girls report being upset by pornographic
material they encounter online (Livingstone and Haddon 2009) but
also revise and negotiate porno discourses in complex ways thus
affording new creative opportunities for sexual subjectivity online
and discursive formations of femininity.
We offer no answers regarding how best to approach to gaining
of informed consent online, the debates regarding public or private
data or the other mechanics which exist regarding how to engage in
ethical, virtual research with children. As Tapscott (1998) points out
children and young people are relatively relaxed about the dangers of
online activity and consider that adults overreact about such issues,
the suggestion here is one of being sensible about risk and consid-
ering it as both relational and situated. While we have no definitive
answers, we do suggest that an unsettling of boundaries is happening
as more researchers engage in such endeavours. This relates not only
60 Ethical Research with Children
to methodological issues pertinent to online research such as issues of
privacy, ownership and consent but also to wider discourses informed
by ‘the virtual’. For example, binaries of adult predator and vulner-
able child in conventional media risk frameworks (Byron 2008) and
gendered binaries of female victim and male predator in sexualisation
discourses (Duchinsky and Clark unpublished) can be unsettled if we
continue to question what it means to engage ethically with children
and young people online.
Conclusion
This chapter has critically considered some of the ethical issues
we as researchers have encountered when planning and conduct-
ing research across particular spaces and places. We have explicitly
positioned research encounters as socially and spatially embedded,
notably in the home, the educational institution and within virtual
spaces. As Herz (1996) argues research is context-bound and the asso-
ciated ethical minefields should be addressed in a situational way.
We have thus argued that different places pose a range of situated
opportunities and challenges in research with children. These are
often not (with some notable exceptions such as Barker and Weller
2003; Young and Barrett 2001) explicitly evaluated or subject to criti-
cal reflexivity for their impact on the participants, researchers or data
as well as their interrelations with wider methodological and ethical
principles of research with children.
The home, the school and the virtual all pose ethical dilemmas
to researchers who enter such territories often unaware of the power
dynamics embedded within each social context and which shape the
social relations that ensue. In the home, researchers must manage
expectations of what it means to be a good guest versus a good ‘eth-
ical’ researcher. Similarly, professionals and non-professionals alike
have to manage power dynamics associated with ascribed roles in
schools and educational institutions. Such dynamics shape the rela-
tionships which develop during the research process, the data that is
generated and pose a range of unique ethical, situational dilemmas.
Virtual spaces also offer a range of opportunities and challenges for
considering the ethics of research with children. What we have sug-
gested here is that such ‘new’ spaces, rather than being sites of risk
Ethical Spaces and Places 61
and fear for children and researchers alike, actually have the poten-
tial to challenge some of the existing assumptions we make regarding
children’s competencies and about the governance of ethics more
generally regarding, for example, the issues of privacy or the notion
of a ‘public’ domain.
3
The Rights of Participation and
the Realities of Inclusion
Introduction
As suggested in the introductory chapter, there has been a growth
in academic literature that documents children’s increasing inclu-
sion in research and highlights their agency in terms of compe-
tency and knowledge (see, for example, Corsaro 1997; Haugen 2008;
MacNaughton et al. 2007). Studies have also illustrated ways in which
children are quite capable of making informed decisions about own
lives: their medical treatment (Alderson and Montgomery 1996), liv-
ing with terminal illness (Bluebond-Langner 1978), responding to
their parent’s divorce (Smart et al. 2001) and talking about their own
experiences of disability (Stalker and Connors 2003). While much
emphasis has been placed on children’s rights of participation in
research, we believe that there is little evidence that scrutinises the
ways in which their inclusion becomes a reality. This chapter presents
the reader with personal ethical stories that makes the messy business
of inclusion and gaining informed consent more transparent.
In Chapter 1, we considered how researchers negotiate existing
formal ethical procedures and pondered on the potential impact of
this on the individuals involved. The discussions in this chapter will
argue that while informed consent is an established and undisputed
part of ethical research processes, it is a research term which is com-
monly adopted and understood, but rarely interrogated. Indeed, the
paperwork proving that participants have been given appropriate
information and a physical space to articulate written confirmation
make up a significant proportion of ethics review documentation.
62
Rights of Participation and Realities of Inclusion 63
However, if we take meaningful informed consent to be more than
just a name on a page, then as researchers our management of it offers
us an opportunity to be reflexive in acknowledging the situatedness
of research and the diversities of ourselves and our participants.
While researchers increasingly acknowledge children and young
people as social actors and agents, there is a need to recognise diver-
gent understandings of agency (see Chapter 4); for clearly, some
children are attributed more agency than others. Indeed, Mayall’s
(2002:21) distinction between ‘actor’ and ‘agency’ can be used
instructively here, with the former implying that children are of
the social world (beings rather than becomings) and the latter tak-
ing ‘action’ forward, implying that children make a difference and
that their views should be taken seriously. Agency should therefore be
taken not simply to imply the liberation of children, but as an oppor-
tunity for opening up possibilities for hearing all children, consulting
with them and creating new spaces for their voices. However, given
the emphasis of articulation, rationality and strategy on agency,
those who cannot give informed consent in conventional ways are
generally excluded, and we believe that this negates their rights of
participation. Arguably, research discussions that fail to incorporate
a critical embodied and engendered account of agency in childhood
studies serve only to reinstate a model in which conventional voices
are privileged, prioritised and accorded more agency than those who
have unconventional voices.
Tried and tested, step-by-step research guides suggest that pre-
scribed methods produce findings, keep to tight schedules and meet
standards of publications and accreditation. However, we found that
this somewhat static, prescriptive approach did not allow for the
modification, fluidity or diversity needed when researching with
children and young people. Therefore, in an attempt to uphold chil-
dren and young people’s rights of participation, and suggest ways
in which marginalised voices could be included in research, the
discussions within this chapter advocate a more bespoke approach
to gaining informed consent and acknowledge the use of con-
ceptual tools sculpted and adapted from relational and specific
practices.
The discussions in this chapter draw predominantly on Allison’s
research conversations with disabled children and young people who
have little or no speech. The narratives recount personal stories of
64 Ethical Research with Children
daily experiences of using high-tech Augmentative and Alternative
Communication Systems (AACS), and the stories collected exposed
a number of real-life ethical issues relating to disabled children’s
rights of participation, illustrating the reality of their inclusion in
research. However, as a point to note, these conversations easily relate
more broadly to the ways in which informed consent with children is
negotiated per se. They also illustrate the often tokenistic paths taken
towards the participation and inclusion of children in research.
While advocating what some might describe as a non-conventional
approach to researching with disabled children and young people,
it is important to emphasise here that traditional research methods
have not been ignored. While it is true that dogmatic responses to
research templates were sidestepped and intellectual common sense
were applied to ethical guidelines, the processes used for seeking
and gaining informed consent have not only retained credibility but
become heuristic and realistic rather than procedural. By exposing
and addressing these and other ethical dilemmas, the ways in which
we, as researchers, can become more critical, considered and reflexive
when seeking informed consent will, we hope, be informative.
Rights of participation
It is necessary to critically examine the construction of disabled
childhoods in order that a more nuanced understanding of the appli-
cation and practice of their rights within research can be gained.
However, childhood itself is a complex notion. There is never only
one version of it, and this multivarious concept has been created
through history and culture, interactions with others and by images
of children and childhood. In addition, children tend to be united
under one term: ‘child’, grouped together and unified, dismissing
and disregarding their individuality and agency. The term ‘disabled
child’ therefore not only describes and signifies the young per-
son’s developmental position within their life-course but also indi-
cates the potential membership of a collective category (James and
James 2004). Such collectivisation is problematic because assumed
notions of dependency, incompetence and vulnerability embedded
in conceptions of disability not only serve to shape disabled chil-
dren’s agency in general but also homogenise their childhoods.
In terms of seeking and gaining informed consent from children and
Rights of Participation and Realities of Inclusion 65
young people who have differing communication requirements, this
homogenisation is not sufficient. The customary, standard written
consent form not only excludes those with unconventional meth-
ods of communicating or understanding, it negates their rights of
participation and marginalises their voices. Therefore, in order that
universally accepted rights of participation apply to all children, pro-
cesses that respect individual abilities should be used in order to seek
and gain informed consent.
While acknowledging the complexities that surround the ways in
which disabled childhoods are socially constructed, we also need to
take into account the many definitions of rights afforded to children
and childhood which draw on broad conventions such as the United
Conventions on the Rights of the Child (UNCRC). Within this Con-
vention, children are assigned a set of economic, social, cultural, civil
and political rights spanning across three broad areas: provision, pro-
tection and participation. When challenging researchers to look more
closely at the messy business of gaining informed consent, the real-
ity of disabled children’s participatory rights in terms of application
must be critically examined, and it is important that particular atten-
tion is paid to one of UNICEF’s core principles: respect for the views
of the child. This principle is concerned with the rights of all children
to participate, to have their views heard and respected.
Article 12 of the UNCRC clearly articulates children’s rights to
express their views, and promotes their capacity to participate in and
influence decisions about their lives. In addition, the guidance and
regulations dealing specifically with disabled children (Department
of Health 1991) makes it clear that, if a child has complex needs,
severe learning difficulties or communication difficulties, arrange-
ments must be made to establish their views. While a disabled child
should not be assumed to be incapable of sharing in decision mak-
ing, it is argued that the caveat within the UNCRC (UN 1989) is
potentially damaging to the general affordance of agency of disabled
children: ‘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’ (Article 12.1, empha-
sis added). It is not surprising that this caveat serves to negate the
voices of disabled children and young people who are often assumed
incapable of forming and expressing their own views.
66 Ethical Research with Children
We recognise that disabled children’s rights have been upheld and
they have participated in research and expressed their views (see Lewis
2001; Marchant and Crisp, 2001; Shakespeare 2006; Ytterhus 2012)
but Allison found that gaining access to disabled, young research par-
ticipants to be something of a complex, multi-tiered process. Indeed,
adults created a protective ‘fire-wall’ that shielded the children and
young people and while the young participant’s views and experi-
ences were considered central to the study, she found it curious that
they could not be consulted until the latter stages of the process.
Typically, gatekeepers and parents/carers’ informed consent should
be gained before the children themselves are approached. The pro-
tection of vulnerable participants takes precedence and subsequently
the adult’s voice is prioritised. This should not be surprising how-
ever, for the competence of disabled children to make decisions for
themselves has long been debated within the context of research,
medical care and in law. Indeed, while we should not forget that
children’s voices should be given due weight in accordance with the
age and maturity of the child, we should remember that it is adults
who ultimately decide whether the child is competent to make those
decisions.
When challenging the application of children’s rights and examin-
ing the realities of participation, we should also be mindful that some
adults will not always support other adults who try to gain access to
children’s knowledge. It is as if they are fearful of in-balancing the
status quo. Indeed, as mentioned previously, we should not forget
that it is adults who ultimately hold the power to make final deci-
sions in terms of allowing researchers to access children: what topics
are deemed to suitable to ask children about and whether the chil-
dren are competent to be included in research. While Allison was
keen to find out about the lives of children and young people who
used high tech augmentative and alternative communication aids on
a regular basis, and the focus on the young participants as competent
social agents was drawn from the discourse of the sociology of child-
hood, she found the data illustrated that mostly their agency was
situated and constructed in terms of dependence and competency.
Having concluded that the voices of disabled children and young
people were inextricably interlinked with other voices and influenced
by larger social structures, Allison began to consider the duality of
the relationships between the structure of disabled childhoods and
Rights of Participation and Realities of Inclusion 67
disabled children’s agency as central to her work. She then turned
to structuration theory to explore the interaction between human
agency and social structure and in doing so, became intrigued by
Giddens’ interest in structuration in relation to ‘the dynamic pro-
cess whereby structures come into being’ (1979:121) for he felt that
anyone who was keen on establishing the origin of structures would
also be interested in how they became structurised in the first place.
Consequently, he set out to link temporal levels of analysis of day-
to-day experiences with the relatively longer duration of invariant
structures that stretch over longer periods of time. He suggested that
structuration emphasises the processes of how human actors repro-
duce and transform social systems. He also believed that the more day
to day experiences are repeated and engrained into daily routines, the
more resistant they are to being manipulated or changed by individ-
ual agents. Similarly, the less they are repeated over shorter spaces of
time, the less likely they are to be fixed in daily life and social struc-
tures. Allison’s data offered a significant insight into the processes
of the structuration of voice. Her observations relating to the nature
and context of disabled children and young people’s lives demon-
strated how they contribute to social production and reproduction
on a day-to-day basis. In accordance with this and central to Giddens’
(1984) view of structuration, is the reflexivity of individuals as they
contribute to social production and while disabled children’s partic-
ipation is represented by their lack of social status, Giddens argues
that all agents can choose to use their knowledge and understand-
ing of rules in a reflexive way to either conform or act otherwise.
In these terms, it is important to consider that, within an environ-
ment that encourages participation, disabled children can contribute
to the processes of childhood change. Indeed, as will be discussed
in more depth later in this chapter, the disabled children and young
people of this study as agents in focus, demonstrated that they had a
fairly clear sense of their own power. At times, they felt empowered
to speak out and at others, when they acknowledged an equivalent
sense of the power of others within the same context, they chose
to keep quiet. Giddens (1984) would argue that this understanding
relates to the ‘virtual’ sense of power within relations which enabled
the disabled children and young people to act knowledgeably. In a
similar vein, Gilligan suggests that the choice to articulate or not is
a contextual judgment and one which is ‘bound to the particulars
68 Ethical Research with Children
of time and place’ (1993:58). However, while Giddens’ arguments
highlight the possibility of voice, Gilligan relates more to the pro-
cesses of production and interpretation. It is at this point we should
stop and reconsider the power imbalances at work here. While we
argue that including disabled children’s voices in research upholds
their rights of participation and goes some way towards re-evaluating
their status and position, if their voices are thwarted from the outset
by those who are more powerful, the risk of their exclusion is very
real. In a similar vein, Stones (2005:8) argues that analysis should
be directed at the reality of particular social processes and events in
particular times and places. While also emphasising the centrality
of the duality of structure and agency, he calls for ‘ontology in-situ’
where human agents are viewed as being linked within dynamic
networks of position practices. Indeed, disabled children and young
people often find themselves within environments of pre-existing
structures where social sanctions constrain agency and in terms of
Allison’s research, this indicates that while the nature of agency is
fluid and contextual, actors’ voices can also be shaped by pre-existing
contexts.
It is important for researchers to build a realistic but critical under-
standing of such pre-existing contexts in the guise of the lengthy eth-
ical processes involved when wishing to gain access to disabled chil-
dren. Perhaps this, in itself offers an explanation as to why some dis-
abled children and young people’s participatory rights are thwarted
and their voices are neither sought nor heard in research. Academic
research journeys often begin by seeking ethics committee’s approval
(see Chapter 1). Generally, lengthy research proposals needed to
be written, and ethical considerations defended. Once institutional
approval has been gained, the consent of the gatekeepers (for exam-
ple, schools) can then be sought. In Allison’s case, the gatekeepers
acted as intermediaries (see in addition Chapter 2) – sending the
research information to the parents/carers on the researcher’s behalf
(because supplying prospective participants’ contact details directly is
not deemed to be ethical practice). The chain of access then became
even more complicated and intricate: the school secretary forwarded
introductory letters to teachers who passed them to the pupils to give
to their parent/carers, who offered their consent (or not). The reverse
system then operated when parents sent the consent forms back to
the school via the children. Ironically, the voices that Allison was the
Rights of Participation and Realities of Inclusion 69
most interested in (those of the disabled children and young people)
were last in this complex chain of consent. Their participation, their
voices and consequently their own stories were only to be heard once
adults’ agreement and consent had been negotiated. Sarah’s research
journey had a different but equally long chain which also involved
parents actively shaping the data collection process (see Chapter 4).
This is also mirrored in Jessica’s experiences with concerns of not
being able to guarantee traditional written and signed informed con-
sent, including parental consent, in research encounters in virtual
worlds. Jessica also found when supervising student research that the
governance of research with children varies drastically in the impor-
tance it places on children’s informed consent (see Chapter 1). All
of these examples illustrate the challenges involved in retaining chil-
dren’s voices as foremost in research despite intentions to make them
central.
Seeking informed consent
As highlighted earlier, the introduction of legislation and interna-
tional conventions designed to assert and protect the interests and
rights of children (Children Act, 1989; Children (Scotland) Act, 1995;
Children (Northern Ireland) Order, 1995; United Conventions on
the Rights of the Child (United Nations 1989)) imply that as chil-
dren are active participants of their lives, they have the right to be
heard and to contribute to decisions that affect them. Article 13
(1) of the UN Convention of the Rights of the Child (United Nations
1989) outlines the child’s rights to freedom of expression and the
right to seek, receive and impart information. The implications of
this for research with children is that it becomes beholden on the
researchers to provide appropriate individualised information and
mechanisms of consent beyond the usual and often tokenistic forms
of written consent. This will also include recognising the diversity of
ways in which children provide their consent or dissent, for example,
through verbal and embodied indicators, as discussed in Chapter 1.
In addition, Article 10 of the UK Human Rights Act (1998) incor-
porating provisions of the European Convention on Human Rights
and Fundamental Freedoms (Council of Europe 1950) and applying
both to adults and to children, is equally concerned with freedom
of expression, echoing the importance of the rights expressed by the
70 Ethical Research with Children
United Conventions of the Rights of the Child (United Nations 1989)
to receive and impart information and ideas without interference.
Universal legislation and conventions recognise in principle that
children are active participants and have the right to be heard
and to contribute to decisions that affect them. However, tensions
between the social processes of seeking consent stem from the con-
tinuing debate regarding whether or not children themselves are
able to consent to participation in research, or whether the par-
ents should consent on their behalf. The National Children’s Bureau
(2003) reflects the ambiguities of obtaining consent, believing on
the one hand in the necessity of children consenting to participate
while on the other, recognising the necessity of seeking parents or
‘gatekeepers’ permission to approach the child. A consequence of
this is that if a child or young person consented to participate in the
research but the parent refuses, researchers are obliged to respect the
parent’s decision (Butler and Williamson 1994). Indeed, as Allison
discovered, if any of the gatekeepers – the school or the parents –
refused consent and exercised their responsibility in this way it may,
as Lloyd Smith and Tarr (2000) suggest, have impinged on the rights
of the child to be heard, to express their views on matters of concern
to them, be able to exercise choice and take action independently
of adults. While it is beyond the remit of this chapter, further and
more in-depth discussions of both the status afforded to children’s
informed consent and the ongoing negotiations that surround it can
be found in Chapters 1 and 4.
Realities of inclusion
Wright (2008) acknowledges the potential challenges of including
disabled children and young people in research but argues that it
is crucial that methods for fostering voices are sought. He contends
that findings will justify research with pupils with multiple and com-
plex needs as worthwhile, for not only will it provide authenticity,
it will also empower those whose views and opinions are not usu-
ally heard. Indeed, researchers (for example, Badham 2004; Cavet
and Sloper 2004; DeSchauwer et al. 2009; Morris 2003; Franklin and
Sloper 2009; Watson et al. 1999) who have championed the need
for disabled children’s inclusion in research and actively promoted
opportunities for their participation found that learning to listen
Rights of Participation and Realities of Inclusion 71
and consult with young people provided supporting evidence which
demonstrated that that they can hold and express views about their
own lives in terms of their school, their families, the additional sup-
port they need and the services they access. As a point to note,
Lewis and Porter (2004) produced a set of guidelines for critical self-
evaluation for those actively engaged in systematically collecting the
views of disabled children and young people, and others such as
Stalker and Connors (2003) and Rabiee et al. (2005) discuss meth-
ods used in seeking disabled children’s accounts of their everyday
lives. Ward (1997) and Morris (1998) produced practical ‘how to’
guides in order that disabled children and young people could be
more actively involved in research and development projects and
Stalker (2012) went as far as to edit and publish a special issue of
the peer-reviewed journal, Children and Society to increase aware-
ness of disabled children and young people’s views and experiences
and their involvement in research, offering a range of ways of seek
their opinions. While this positive inclusion of disabled children
and young people’s voices is to be applauded what should also be
acknowledged is the complex relationship between voice, power and
empowerment. Simply facilitating particular individuals or groups
to speak does not necessarily change the power dynamics of their
societal position or relations within the research process. Even when
particular voices are heard it remains possible for researchers, in their
purported positions as ‘experts’ to use such voices tokenistically or to
engage in interpretation based on their own positions and under-
standings rather than those of their participants (Oliver 1992) (in
addition see Chapter 6).
Cameron and Murphey (2006) warn researchers that they should
consider ethical issues carefully when establishing whether or not
participants can give consent to be involved in research in the first
place. Their study with adults with learning disabilities illustrated
that including specific adaptations enabled some participants with
comprehension difficulties to understand the nature of their involve-
ment and give their own consent but they also acknowledged that
those whose comprehension was poorer presented researchers with
specific challenges. Balancing the need to protect people assumed to
be vulnerable from being coerced or including them without consent
while simultaneously ensuring that they were not being excluded
is clearly challenging, but in order to challenge objectification and
72 Ethical Research with Children
silence, disabled children and young people need to be recognised as
active participants and their input should be prioritised.
Gaining informed consent
While earlier discussions outline the complexities of the process of
seeking consent, in terms of recognising agency and respecting rights
of participation, Mahon et al. (1996) believe that the process of
gaining consent is of particular importance when researching with
children and young people. Indeed, Kellet (2005) argues that it is
not simply a question of informing the participants of the nature
of the research and asking them to sign a consent form but sug-
gests there is a need to carefully outline the aims of the research,
explain how the data will be collected from them and how it will
be used. Where complex debates relating to children’s competencies,
age and maturity, and credibility of their statements have effec-
tively questioned whether their voices can be taken seriously, Rodgers
(1999) argues that given the appropriate information and mecha-
nisms, many disabled people can understand and make their own
decisions. Acknowledging that many disabled children and young
people require additional support in order to participate in research
that seeks their views and experiences, Stalker (2012) upholds the
need of flexibility and inclusivity within the research process and
argues that a range of ways are needed to seek their opinions. Clearly,
the ‘one size fits all’ approach to gaining informed consent is insuffi-
cient and including participants who have little or no speech requires
improvisation and adaptation.
When researching with children and young people who used high-
tech Augmented and Alternative Communications Systems (AACS),
individual abilities were respected whereby Allison produced bespoke
information for each prospective participant. Having sought advice
from staff and an older pupil within the school (also a high-tech
communication aid user), information sheets and letters seeking con-
sent were produced that explained the research aims and the extent
of their involvement in a participant friendly manner, using simple
sentences, picture exchange mechanisms and illustrations to convey
the information. For the participants that had difficulties in reading
or had visual impairments, parents/carers or staff were asked to read
the information leaflet to the participants. However, individualising
information in this way proved to be costly, time consuming and
Rights of Participation and Realities of Inclusion 73
challenging. Overcoming barriers, financial implications and time
constrictions offer further explanations as to why disabled children
and young people are often excluded from research, for it seems that
those whose voices are more conventional and easier to gather are
more readily accepted, included and, as a result are privileged.
Despite parent/carers consenting to their child’s participation in
research, gaining the children’s consent from the outset is of vital
importance, for as Taylor (1998) suggests, this takes seriously the need
for upholding their rights and including them in decision-making
processes. However, having researched with adults with learning
disabilities, Simons et al. (1989) warn that disabled people have a
tendency towards acquiescence because so much of their life is con-
trolled by others. The same consideration can, however, be applied
to many institutional contexts. We argue that different contexts pro-
duce specific responses and influence voices in particular ways. There
is the potential for less adult-controlled social settings such as a
playground to provide different responses to a particular issue than
a highly authoritarian setting such as school classroom (for exam-
ple, Punch 2002; Westcott and Littleton 2005). However, we are also
aware of the power that children can exercise within adult-controlled
spaces in ways that may not often be recognised or that are given
credit for, such as the use of silence, as discussed in Chapter 2. Criti-
cal exploration of both physical and virtual space allows for a greater
understanding of the power relations that flow within the various
institutions of childhood both in general and research encounters
and relationships.
Indeed, while Allison questioned whether those who were helped
to complete the consent form did actually fully make the deci-
sion to participate independently, she also found that self-advocacy
was problematic with non-verbal children who gave her informed
consent independently, for it is often the case that children with
little or no speech were taught to respond to promptings rather
than initiate communication themselves. The following extract of a
research conversation with a teaching assistant offers an enlightening
perspective:
Matthew is very good at initiating conversations, but he doesn’t
do it very often. I think a lot of students are so used to having a
passive role in life. Life just happens to them and they sit and it
74 Ethical Research with Children
passes them by. They lose the habit of initiating things. Matthew
can do it very well if he chooses to, and he chooses to quite often,
but not as often as we would like
This reminds us that there is also a vital distinction between
prompted and spontaneous communication whereby prompted com-
munication allows participant’s to respond to agendas and concerns
of others, whereas spontaneous communication allows free expres-
sion which stems from the things that the children themselves feel
are most important. Indeed, Potter and Whittaker (2001) found that
prompt-based approaches encouraged passive responses from actors
rather than active responses from agents who express their own
needs and exert control over their environments as self-determined
individuals. In order to address the ebb and flow of the partici-
pants understanding within Allison’s study, their continuing consent
was sought throughout the fieldwork in consideration for the sen-
sitive recognition of the issues of power, voice and representation.
This, she believed, resulted in the maintenance of respectful research
relationships based on informed consent.
While issues of prompting and passivity are particularly explicit
in Allison’s research, they are however, pivotal ethical considerations
in all research with children. This is because children cannot and
should not be abstracted from the social context within which they
are embedded with all the associated disciplinary regimes, power rela-
tions, opportunities and constraints that come with it. With this
in mind, we argue that hearing the ‘pure’ voice of the child is an
inherently problematic and flawed ideal.
Critical reflexivity and children in research
While upholding rights of participation and pondering on how best
to include children in research, Spyrou (2011) suggests that critical,
reflexive researchers should reflect on the processes which influence
children’s positionality. He stresses that when we include children in
research, we must not only be mindful of issues of representation, but
also of the power imbalances that shape their voices and the contexts
which inform them. As suggested previously, when we include their
voices, we should also move beyond claims of authenticity by explor-
ing the messy, multi-layered complexities that make up their voices.
Indeed, Allison found that in order to include disabled children with
little or no speech in her research she needed to deconstruct the
Rights of Participation and Realities of Inclusion 75
popular notion of voice and pay particular attention to the ambiguity
of it. The young participants adopted a vast range of communica-
tion methods, not necessarily associated with the common notion
of ‘voice’. For example, information leaflets and consent forms most
commonly include a statement that outlines participant’s right to
withdraw from the research at any time. However, Allison was mind-
ful that some participants may have forgotten this or they might not
have fully understood it. In addition, while some of the participants
were able to communicate their desire to withdraw, others did not
have the vocabulary to enable them to do so. Therefore, she found
that by considering voice as social and co-constructed instead of
straightforward or linear. Allison looked for ‘clues’ as to how the par-
ticipants were feeling and judge if the questions, observations or her
very presence became intrusive. When trying to capture one young
person’s story, an incident occurred whereby pupils started filing past
the classroom, distracting the participant. The sound of a distant bell
indicated that it was break time. The participant indicated that he
wished to terminate the interview:
Justin: ‘Off’ [pre-programmed response]
Allison: ‘Off? You want me to turn this off?’ [pointing to the tape
recorder]
Justin: ‘Yes’ [and nods]
Allison: ‘Ok’ [I switch the machine off]
Pause
Justin: ‘Bag’ [pre-programmed response]
Allison: ‘You want me to put it in my bag?’
Justin: ‘Yes. Thank you’ [pre-programmed response]
Allison: ‘No worries’ [I put the recorder back into the bag]
Justin: ‘Classroom’ [pre-programmed response]
Allison: ‘You want to go back to your classroom?’
Justin: ‘Yes’ [pre-programmed response; and nodded]
Allison: ‘Ok, give me a mo . . . ’ [I put my notepad and pen into
my bag] . . . ‘Thank you for taking part in the interview, Justin.’
[after confirming that he wanted me to, I took him back to the
classroom]
Later and before the interview was transcribed, Allison reflected
on the ambiguity of the situation. Did the participant want to
76 Ethical Research with Children
withdraw completely from the research, withdraw momentarily or
withdraw his consent to use the material? After much deliberation,
she approached the young participant and asked him if he wanted
to withdraw from the study. He was clearly amused. Their ensuing
conversations revealed that being interviewed during break time was
much less important to him than joining his friends for their usual
game of wheelchair football! He confirmed that he had not intended
to withdraw; he just wanted to ‘go out to play’. On reflection, Allison
realised that while the new orthodoxy of children’s participation
advocates providing opportunities for children and young people to
be included, to express their views and explore the options open to
them. Such perspectives are however, merely the beginning of hear-
ing the voices of children and as discussed momentarily in Chapter 4,
notions of child-centred research remain under-interrogated. While
we do advocate the importance of listening, research encounters
are rarely this simple. Voices are shaped not only by our views as
researchers (for example, to listen carefully to the views of children)
but crucially by the social context, the positioning of the individuals
involved in the encounters and the fluidity of power relations that
produce such conversations and as a consequence co-construct such
knowledge.
Conclusion
If we are truly interested in childhoods and wish to gain mean-
ingful insights into children’s lives in an ethical way, we should
position ourselves as adults who wish to seek the knowledge of chil-
dren, who want to learn what children know and show them that
we genuinely want to learn from them (Mayall 2000b). However,
upholding the participatory rights of disabled children and young
people and including them in research is a tricky business. It calls
for a unique and carefully considered approach for we believe that
children’s voices are multi-layered and it takes time to collect and
understand them. Rushing in and out of children’s worlds to quickly
collect data and then shape it to fit and illustrate researchers own
agendas simply won’t do.
While we acknowledge that the discussions in this chapter draw
heavily on research with disabled children and young people, we
argue that the key issues raised are also of particular relevance
Rights of Participation and Realities of Inclusion 77
to those wishing to give serious and critical ethical consideration
to children’s rights to participate. Undoubtedly, while the notion
of universal rights offers a possible basis for cohesion in the face
of fragmentation and diversity, in reality the plausibility of child
agents actively engaging with them is minimal given their literal and
metaphoric lack of voice. Indeed, as Morris argues, research with dis-
abled children should place considerable significance on the study of
the social arrangements surrounding rights, the way in which they
‘function in relation to social structures and institutions, the way
in which rights are accrued in particular areas by particular social
groups and the contexts in which they are implemented and given
meaning’ (2006:243). In addition, real consequences for individuals
require further consideration of rights issues in terms of how they are
embedded within a society and its institutions (Stones 2005). There-
fore, it is concluded that the principles and practices of rights in situ
relate directly to opportunities made available to give individuals a
voice and include them in research.
4
The Illusion of Autonomy: From
Agency to Interdependency
Introduction
This chapter explores the influence on research that the contem-
porary emphasis on children’s agency and autonomy has had and
evaluates the usefulness of such terms in research-based contexts.
We begin by outlining how agency and autonomy have become
part of the social construction of childhood. We then consider the
development of participant-led approaches to research with chil-
dren, before evaluating the attribution of agency and autonomy to
children, characteristics traditionally assumed to be the properties
of adulthood (Lee 1999:458) yet currently promoted and pursued
extensively in research with children. In this chapter, we question
the appropriateness of the emphasis on these concepts and ask if
alternative values might be more appropriate to reflect the roles of
children in research as negotiated, relational and interdependent. If,
as Parry (1991:180) contends, ‘it is not possible to abstract from our-
selves all our natural and social bonds, allegiances and commitments
and still understand ourselves’ then are concepts such as autonomy
and agency require an independent selfhood relevant in research
endeavours which seek to understand more about the social worlds of
children? We identify and discuss ways in which children negotiate
power relations within these social contexts which rely less on the
skills of autonomy and more on establishing and negotiating social
relations.
This chapter uses data examples to illustrate children as
being socially embedded. We argue that greater recognition of
78
The Illusion of Autonomy 79
interdependency in research relationships would enable a shift in
ethical approaches from the current emphasis on agency of the indi-
vidual to a focus on research relationships, emphasising nurturance
and an ethic of care. These conversations begin with an outline of
how agency and autonomy came to be so integral to contemporary
childhood.
Constructing the child as being in social research
The sociology of childhood initially emerged to challenge assumed
notions of passivity and objectification of children. The social con-
struction of childhood was premised on dependence and vulnerabil-
ity, and the prevailing emphasis was on a ‘becoming’ child yet to
acquire the skills associated with adulthood. In contrast, the soci-
ology of childhood emphasised the agentic and ‘being’ child, and
these ideas transformed how childhood could be perceived and how
children might be regarded (James and James 2004; James et al.
1998; Jenks 1996; Qvortrup et al. 1994). The concept of the being
child challenged notions of passivity and recognised instead the
agency in children rather than the necessity of maturation for agentic
development to occur.
The impact of this paradigm shift should not be underestimated,
and in the field of social research into childhood it would be hard
to overstate. However, more broadly, the emphasis on the ‘being’
child to the exclusion of other states has now been questioned.
Gallacher and Gallagher (2008:510) argue that the notion of the
child as ‘becoming’ has fallen prey to currently dominant ideas about
the child as ‘being’ as a more appropriate discourse through which
to reveal the realities of children’s lives. As they put it, the idea
of the child as ‘becoming’ has ‘almost been vilified’ (Gallacher and
Gallagher 2008:510). Uprichard (2010) contends that the current
emphasis on the child constructed as ‘being’, neglects a reality of the
child as ‘becoming’, when children themselves frequently demon-
strate knowledge and awareness of this state. Ignoring this aspect of
childhood experience, she argues, places generated knowledge on a
par with that created through the previously dominant construction
of the child as ‘becoming’. It offers a partial rather than complete
picture, and Uprichard (2010) contends that both concepts are neces-
sary to reveal childhood as lived by children. Gallacher and Gallagher
80 Ethical Research with Children
(2008:510) posit that the term ‘becoming’ itself needs to be redefined.
They claim that the term need not be inextricably linked to prede-
termined developmental stages of childhood. But rather, that it is
preferable to see childhood and adulthood in a ‘present and contin-
uous tense’, where both adult and child are always emerging, unfin-
ished and therefore becoming subjects. It is as part of this emerging
critical evaluation of the perspectives held within the sociology of
childhood that we situate this chapter where we use examples of
our research experiences to review the dependence/independence
dichotomy with particular reference to the concept of autonomy, its
acquisition and its role in the research process. Here, we argue that
the self of child and adult emerges through social relations in social
contexts including our research endeavours.
Scientific and political interests in childhood and children have
been evident for some considerable time, linked to modernisation of
the nation state through industrialisation and the changing social
and political need (Parton 2006; Prout 2005). Punch (2002) argues
that perceptions about childhood and the status of children influence
the ways in which research occurs. The contingent normative ideas
of the time saw increasing perception of childhoods and their differ-
entiation from adulthood (Qvortrup 1994). The social and political
interest in this ‘othered’ state is inextricably linked to the method-
ological approach and research themes pursued. Educational and
developmental exploration from medical and psychological disci-
plines has historically dominated research on children and into child-
hood. Such perspectives were aimed at gaining knowledge about the
‘normal’ developing child. Though mostly positivistic in approach,
in aspiring to objectivity and neutrality, values and assumptions
about childhood are nevertheless evident. Childhood as natural and
becoming, following gender-specific roles and behavioural character-
istics have been dominant features of research. Until recently, these
assumptions have tended to be overlooked in academic literature
(James and James 2004). Research, at times intrusive and alienating,
focused mainly on the progression of the child towards rationality
and physical and cognitive maturation (Elbers 2004), the develop-
mental path towards the autonomous individual. Knowledge in how
to assist this progression coupled with enquiry about the factors that
might inhibit developmental progress has represented the majority
of disciplines surrounding childhood. These interests are related to,
The Illusion of Autonomy 81
and driven by, the social and political need to control and construct
childhoods to produce an effective adult citizenry sufficient to the
changing needs of the state (Cockburn 1998; Garrett 2009; Lange and
Mierendorff 2009).
The objectification and homogenisation of childhood in these tra-
ditional stances, arguably neglected to acknowledge the child as a
social actor with expertise in their social and cultural worlds, pre-
ferring instead the social construction of passive, dependent and
vulnerable children, frequently categorised children as objects of con-
cern or threat (Hendrick 2005). Also neglected were the social and
cultural worlds of children (Christensen and Prout 2002). Recent
emphasis on the rights of children, informed by the UNCRC legis-
lation and academic discussions surrounding what is termed ‘new
sociology of childhood’ (James and James 2004; Prout 2005), has
played a significant role in ensuring children’s individual subjectivity
be recognised. Rights legislation has required that children’s voices
be heard in decisions made about their lives. Such changing ideas
about children have caused a significant shift in how children and
their childhoods are researched and have produced a diverse and
changing methodological landscape. It has also shifted the role of
individual children within research (Kellett et al. 2004). Complex and
diverse social realities of children have emerged through alternative
research methods. This methodological approach has matured and is
now firmly established as an ethical way to conduct social research
with children. The time for critically evaluative debate which moves
the methodology forward has clearly arrived. Part of this emerging
debate involves a greater interrogation of concepts embedded in this
methodology such as agency and autonomy. Emirbayer and Mische
(1998:962) have previously identified that not only the concept of
agency is increasingly ‘contradictory’ and ‘strained’, but the term
itself remains ‘elusive and vague’. Yet agency, its pursuit and practice
in childhood remains canonical with, as yet, little attempt to interro-
gate its meaning or appropriateness in children’s lives, or indeed our
research activities with them.
Autonomy and its place in ethics
Autonomy in adulthood can be defined as being an individual who is
‘self-sufficient, self-reliant and self-realizing’, motivated by personal
82 Ethical Research with Children
gain whose independence is threatened by other self-serving individ-
uals (Walter and Ross 2014:16). This notion of autonomy is termed
as the ‘in-control model’ (Walter and Ross 2014). Agency is intri-
cately linked with the capacity to be autonomous and is associated
with ‘selfhood, motivation, will, purposivenss, intentionality, choice,
initiative, freedom and creativity’ (Emirbayer and Mische 1998:962).
The ‘in-control’ model assumes and values rationality, sensible deci-
sion making and self-reliance, rather than a susceptibility towards the
influence of others, referred to by Emirbayer and Mische (1998:965)
as being ‘freedom and rational self-interest’. There is an explicit
emphasis in autonomy and agency on individual rights and self-
protection (Greenfield 1994). Rose (1999:91) argues that this ‘self’
is ‘an object of knowledge and autonomy’ developed through a
continuous process of self-improvement achieved by applying and
privileging ‘rational knowledge’, the emerging autonomous self. This
process enables individuals to constitute themselves as the ‘self’
required by contemporary society and government. Autonomy is
dominant in a hierarchy of characteristics evident within an adult
rights holder in advanced liberal societies and is regarded as a nec-
essary component of economic citizenry (Tronto 2009:164). Unsur-
prising then that it is a characteristic that we expect children to
acquire.
Autonomy has also been a dominant principle of ethics through
the promotion of rights, protection of the individual and their capac-
ity to make decisions and to engage in activities that individual
actors find meaningful (Mullin 2014; see also Dworkin 1988). Its rela-
tionship with individualism is explicit. Code (1991) argues that this
definition of autonomy has become closely aligned with a neo-liberal
model of individualism where its virtues are privileged (Walter and
Ross 2014) and inform ethical processes surrounding research. How-
ever, this independent individual is not a universal truth but is, as
Greenfield (1994) reminds us, a particular cultural script for the devel-
opment of the ideal person. There has been, according to Emirbayer
and Mische (1998:963), a ‘consistent failure’ to recognise agency as
temporally situated and emerging in social relations through social
interaction. We argue that this failure to account for social inter-
action has had specific impact on ethical discourses concerning
research with children, where the privileging of individual rights and
selfhood remain pivotal in methods which promote participatory
The Illusion of Autonomy 83
methodologies. The embedded contradiction here ensures that the
ways in which children exert agency in research contexts is over-
looked, even discounted as it risks undermining certain ‘cherished
conceits’ (Segal 1999) in research such as objectivity and neutrality
and the importance of a distanced researcher.
Significant contemporary focus is placed upon the rights of chil-
dren. To be heard, to participate and to exert agency are all part of
the discourse of contemporary childhood. The agency being fostered
within the social institutions which surround children reflect ethi-
cal principles found in human rights legislation and are emphasised
in ethical frameworks through which social research with children is
conducted. These codes of behaviour are based upon the Cartesian
subject where the rights of selfhood and individualism are integral.
Children are assumed to have not yet developed characteristics asso-
ciated with the ‘in control model’ but are encouraged to acquire
these skills as they mature. This goal of future autonomy (Feinberg
1980) influences the expectation that children will now participate
in meaningful decision making in their own lives and the lives of
their communities. Indeed, agency – as being the ability to act inde-
pendently – is a concept which has become a canonical narrative, a
‘cherished conceit’ (Segal 1999) if you will, in the social construction
of childhood. Reflecting the need to acquire the skills associated with
autonomy, Greenfield (1994) has argued that schools in neo-liberal
and individualistic societies are structured around competitive mod-
els where individualism is valued and fostered. Within institutions
such as the school and the family, an assumption of the dependent
child being encouraged to move towards independent adulthood
predominates. The role of parents to recognise the need for this
developmental process (Killen and Wainryb 2000) and facilitate the
acquisition of autonomy for their children is part of the cultural
script. Individualism, agency and autonomy are required in contem-
porary adult citizenship; children are therefore expected to acquire
the skills ready for adulthood. Research participation is just one
arena for those skills to be practised. Yet participation constructed
as a right of children can perhaps too easily become a responsi-
bility. Within research with children terms such as autonomy and
agency are used interchangeably and are taken to be a reflection of
the more insightful and ethical approach to the social positioning
of children in contemporary research. Such terms populate research
84 Ethical Research with Children
texts generally but are rarely interrogated to determine their appro-
priateness. Yet, despite this emphasis on agency, children remain
commonly consigned to the margins of ethical debates. Children’s
assumed incapacity to demonstrate autonomy is used as justification
for this marginalisation (Richards 2012a). Indeed, many argue that
child-centred research continues to be premised on long-standing
assumptions of dependence and vulnerability (see, for example, dis-
cussions of assent versus informed consent in Chapter 2 of this
volume).
Societies that value independence in adulthood assume the devel-
opmental journey to it begins with dependence in childhood. Self-
protection as part of how independence and autonomy are perceived
(Greenfield 1994) ensures that protecting not yet independent or
autonomous children is an imperative aspect of the cultural script
and consequently explicit in ethical processes. This presence of both
protection and autonomy in ethical procedures creates another con-
tradictory tension where child-centred methodologies call for an
independent and agentic child to be at the forefront, while an
autonomous rights discourse emphasises the need to protect those
who lack the skills needed to protect themselves. In consequence,
both characteristics are required and somewhat clumsily attributed to
children by adults at various points throughout the research process.
Sarah’s ethics application is a case in point.
Methodologically, Sarah’s research was premised on elevating
the voices of young transracial intercountry adoptees, frequently
excluded from debates about their adopted status (Richards 2013).
Her first hurdle was getting ethical approval from the university
to conduct the research. This took six months and a number of
amendments to the proposal before eventually gaining approval
from the faculty ethics committee. Sarah was requested to articu-
late her ability to manage researching a sensitive topic, and care for
the vulnerable children involved. In the end, the research proposal
emphasised children’s agency and participation and their capacity to
speak about life experiences while her ethical section and university
application form spoke of the care of vulnerable children involved
in a sensitive topic. This experience demonstrates that despite the
impact that child-centred approaches to research are said to have
had the combination of the need to both protect children and
ensure their participation can still generate anxiety and inevitably
The Illusion of Autonomy 85
produce contradictory constructions of childhood in locations where
we generate knowledge.
In addition to producing contradictory expectations, the auton-
omy embedded within our dominant cultural script and ascribed to
children through individual rights discourses, impedes recognition of
relational or social agency achieved within the interdependency and
inter-subjectivity of children’s lives. Despite the plethora of discus-
sions of child-centred research (Alderson 1995; Alderson and Goodey
1996), and newly emerging critical consideration of such methods,
we argue that for the most part these approaches fall short because
they too frequently seek agency and the voices of children in the
‘wrong space’ (Cockburn 2005). Child-centred methodologies can
often focus on a decontextualised child within an autonomous and
individualised space: the autonomy of selfhood (Walkerdine 1999).
Greenfield (1994) argues that the individualism fostered in this cul-
tural script is to the detriment of cooperation and community. We
argue that it is also an inappropriate perspective to view the agency in
children’s social lives. MacIntyre (1981 in Parry 1991:182) contends
that
an essential element in the story of one’s life is the community
into which one is born, or is brought up . . . these give a person’s life
its own moral particularity . . . the story of my life is always embed-
ded in the story of those communities from which I derive my
identity.
As social individuals we require a society which allows us opportuni-
ties to cultivate our social competencies (Twine 1994). Competencies
such as agency, autonomy, and morality, should therefore be seen
as an expression of the social (Frankel 2012) rather than the indi-
vidualised self. The autonomous self depicted by Rose (1999:91) as
emerging through a continuous process of self-improvement is, we
argue as Twine (1994) does, achieved through knowledge acquired
in social relations. A dynamic self who constantly re-shapes and
repositions their ‘self’ depending on the social context they are in
and the power they can exert in it (Frankel 2012). A social self,
emerging from our social relationships through a dynamic of ‘choice
and constraint’ and thus never fully autonomous nor fully deter-
mined (Twine 1994:11), the being and becoming self. It is this social
86 Ethical Research with Children
and relational agency in children’s research experiences that we
explore next.
Children’s relational agency
Just as independence or individualism is part of an idealised cultural
script, so too is interdependence, also referred to as ‘collectivism’
(Greenfield 1994:7). Each script stems from alternative ‘philosophies
of life’ (ibid.), where Western cultures have commonly privileged
individualism, many Asian cultures value interdependence. In con-
trast to a Western script an idealised Asian cultural script presumes
that children begin developmentally as independent individuals and
must be taught and allowed to develop the skills of interdependence
(Greenfield 1994). Here selfhood is achieved not through autonomy
or individualism but through relational subjectivity with family and
community. Montgomery’s (2007) research with child prostitutes in
Thailand successfully articulates the expectation and responsibility of
children to foster interdependent relationships with family and com-
munity. Of course each us are both individuals and members of social
groups and so both are inextricably linked in the ways that we live
our lives. The point is that the particular script which is culturally
emphasised and idealised influences how the development of chil-
dren is organised and determines what is appropriate in their lives
and what is not. The balance between independence and interde-
pendence is determined by whichever concept is most highly valued.
The ideal adult citizen is then constructed as holding this charac-
teristic. Both concepts have strengths and limitations. Simplistically
one places the social above the individual and the other values
the individual above the collective. However, by placing such value
on independence, individualism and autonomy we risk overlooking
children’s capacity for relational agency. This oversight ensures that
the ways in which children exert themselves and achieve their goals
through and within the social context is often neglected.
A relational concept of autonomy is not overlooked entirely how-
ever. In contrast to previous clarifications of agency and autonomy
which idealise an independent selfhood, Emirbayer and Mische
(1998:970) define human agency as being, ‘the temporally con-
structed engagement by actors of different structural environments –
the temporal – relational contexts of action’. Such a perspective
The Illusion of Autonomy 87
is premised upon people being ‘socially embedded’ where our per-
sonalities and identities emerge within social relationships in our
social contexts (Mackenzie and Stoljar 2000). This ‘intrinsically social
and relational’ agency is achieved through the ‘engagement and
disengagement’ of actors in their contextual environments, which
Emirbayer and Mische (1998:973) refer to as ‘structured yet flexi-
ble social universes’. This relational model of autonomy refutes the
existence of the ‘in control’ model individual, as discussed earlier.
We place this relational agency, this interdependency, at the forefront
of the following empirical accounts.
Using a perspective of interdependency between participant and
researcher we potentially enhance the visibility of children as socially
connected with moral agency and power (Malaguzzi 1993). This rep-
resentation of childhood produces children as active in reciprocal
relationships rather than the abstract individual commonly found
in the language of autonomy, rights and ethics. Seeing these chil-
dren allows us to move beyond research where a dominant theme is
that of the right to protection, too often premised on the weak and
vulnerable child, whose right to self-protection as an autonomous
self is not yet fully developed, and instead, listen to socially situated
and connected children (Richards 2013). Recognising the relational
connections between individuals in the research process can be chal-
lenging. Detachment and neutrality may be the traditional language
of positivistic experimental methods (Smith 2005), but it is not nec-
essarily the language of children’s or indeed adult’s worlds (Richards
2013).
As we discuss in Chapter 2, research with children now commonly
takes place in environments familiar to the children: nursery, home,
school being typical examples (Harden et al. 2000; Lewis et al. 2004).
These places are assumed to be locations where children feel safe
and are more likely to contribute effectively (Harden et al. 2000).
Nevertheless, research with children in these social settings holds par-
ticular dilemmas, both social and ethical. What it can reveal however,
are the ways in which children negotiate relations and exert rela-
tional agency. In Sarah’s research, the young participants frequently
placed Sarah in different roles during the data collection activities.
For example, casting Sarah in a teacher role was quite common. She
found herself variously having to spell words, comment on neat-
ness of writing, and to determine or confirm whether an answer
88 Ethical Research with Children
was a correct one. Despite being uncomfortable in this role and its
contrast with her methodological values, Sarah complied in the inter-
ests of maintaining good relations and collecting data. Waller and
Bitou (2011) claim that current methods often used in participatory
research with children induce such pedagogical relations. A young
participant called Wim (aged eight years) confirmed her perception of
the similarity between Sarah and a teacher throughout the interview
by her continued compliance and deferral to her in decision mak-
ing, and also through requests for guidance in spelling and for what
she should write in order to address correctly the pages in the activ-
ity book she was given. The following questions are typical of ones
that she asked having verbally told a story and needing confirmation
that it would be appropriate to a specific page that had instigated the
telling of the story:
‘Can I put what I used to call my mummy here?’
...
‘Shall I put that here?’
Throughout the interview Wim sought approval for each contribu-
tion made and exerted her social agency by ensuring that Sarah
fulfil this role. Little wonder perhaps that when at the outset of the
activity with Wim, Sarah suggested, ‘you are in charge’, Wim simply
laughed. This interaction demonstrates that Wim’s understanding of
the power relations contrasted with Sarah’s, and her laughter related
more to her perception of Sarah’s foolishness in suggesting that it was
anything other than the usual adult/child power relation, rather than
evidence of a growing rapport that she took it to be at the time. In this
instance, Wim grasped the reality of the situational dynamics and
Sarah, as the researcher was innocent in relation to them. The dia-
logue demonstrates the tension between how researchers may assume
and wish agency to be and how it actually emerges in social contexts.
The implicit resistance on the part of Wim to assume the power-
ful role, but still to remain on good terms with Sarah as the adult
and yet comply with other aspects of this interaction, demonstrates
complex, agentic and successful negotiation of the adult spaces in
which she was placed and challenges simplistic notions of agency
and selfhood.
The Illusion of Autonomy 89
Walkerdine (1989:270) argues that the teacher (like the mother)
becomes the facilitator in the learning environment, not there to
instruct but to observe in a more passive role in contrast to the active
roles that are performed by the child. Adults here, like the adults in
child-centred research, take a ‘nurturant, facilitating, sensitive and
supportive role’ where intervention but not interference is required.
Walkerdine argues that these approaches are founded on an ‘illusion
of autonomy and control’ (ibid). However, we claim that the relation-
ship with participants in this facilitative and nurturing role does not
negate children’s autonomy but demonstrates an alternative social
understanding of the concept and therefore challenges contempo-
rary ideas of pursuing the traditional model of autonomy in research
into childhood.
The capacity to be agentic develops within dynamic, situated con-
texts and is always in a state of re-evaluation and reconstruction
on the part of the actors involved (Schutz 1962). Sarah variously
demonstrated throughout the fieldwork, ascribed characteristics of
‘mother’, ‘teacher’ and ‘researcher’. At times she performed the
role of teacher because her participants cast her as such as they
negotiated their way through the activity. At other times, she per-
formed ‘the good researcher with children’ through aspirations of
empowerment, capturing the voices of young participants, and on
occasion, ensuring that their right to withdraw, remain silent, or
leave stories untold, was protected in the face of frustration from
their parents. These roles were not selected at random but came
about as part of ongoing interaction with participants. The girls cast
these roles which Sarah duly performed, in part to connect with
them but also to illicit and encourage their participation in the
research. A simplistic focus on ‘selfhood’ in research with children
misses the significance of the network of relations, peer, family and
wider social networks that situate and inform children’s lives and
research relations.
Despite Sarah’s desire to ensure the voices of the young partici-
pants were at the forefront of the discussions on their intercountry
adoptions, the voices of others within the family also shaped the
data collected. While this complicated the ownership of the sto-
ries, it also revealed the social position of children within the
home and provided insight into research conducted in the home
(see Chapter 2). The following account taken from Sarah’s research
90 Ethical Research with Children
(Richards 2013) demonstrates the ‘messiness’ of these research rela-
tionships.
One family made the dining room and table available to me to
run the activity with the two daughters taking part. There were
two doors to the room, one led to the kitchen and one to the liv-
ing room. The room was therefore used like a connecting corridor
from kitchen to living room. The doors were pulled closed by the
mother as I began to talk to the two girls. As lunch time drew close
however, (and I had already been invited to join them and as a
guest felt obliged to say ‘yes’), both parents began to walk through
the room and then the mother (in particular) stopped to glance at
the books that the girls were making as part of the data collection,
and make the occasional suggestion about content. At the end of
the interview she came in to say that lunch was ready. As I cleared
away the pens and glue, Janet (mother) asked to read the journals
which were still on the table. This request was difficult to respond
to, the journals belonged to the girls, and this had been made
clear to them at the outset. To acquiesce would contradict their
ownership, and to refuse would be rude in their home, prior to sit-
ting down and eating lunch and before interviewing the parents,
already identified as potentially problematic in the establishment
of rapport and guest relations. I suggested that if the girls agreed
to show their journals then I was happy for them to do so. The
girls, perhaps unsurprisingly, agreed and their mother sat down
at the dining table and read through each page. She commented
on handwriting, spelling, and the content of some stories and the
absence of others which she felt that the girls could have or should
have included. On occasion she added written comments onto the
pages, which she felt were relevant or where her daughter(s) had
not written anything on. The Figures 4.1 & 4.2 illustrate this.
Whatever our individual expectations or aspirations of what
participatory research with children might be, we cannot remove
children from the social context that they occupy or separate their
voices from the intergenerational power relationships: the ‘con-
straints and choices’ (Twine 1994:11) through which they live their
lives. Nor should we attempt this if we really wish to capture their
life experiences.
91
Figure 4.1 The place where I live now
92
Figure 4.2 First days in my family
The Illusion of Autonomy 93
Consider the following examples from Sarah’s research which illus-
trate the ways in which family members co-construct their own
stories. Here are the experiences of a mother meeting her child for
the first time:
Sally: ‘We arrived and were told we were going to meet her the
following day and then there was a knock at the hotel door and
the guide was there and he said, ‘baby coming in half an hour’,
so it was a bit of a shock and we were jet lagged and hungry. And
half an hour later she was at the door screaming. They handed
her to me and she went very quiet and was looking around to
see where the orphanage director and carer was but she was not
crying. But after they had gone she just latched onto my husband
and decided to only go to him. The rest of the week that we were
in China she would not come to me.’
Ricoeur (1991) suggests that stories are performed for particular
social contexts and are therefore different each time with multiple
and changing key aspects as story lines. First meetings are the impor-
tant initial stories of belonging told by adopters. They are told to the
girls to help them begin to comprehend where and to whom they
belong as a result of adoption. The girls speak of the same events and
similar key figures, but with alternative explanations. In the account
below, Sally relies on a specific narrative within this adoption group
about subordinate approaches to care within the orphanages. She
implies that this care varies from what she and her husband begin to
provide, and suggests that this is why their daughter initially rejects
her. She gave this response after she had disclosed that her daughter
had refused to let her hold her until they had returned to the UK and
Sarah asked her how she felt about it.
Sally: ‘Um, it was okay, it must have all been really so strange and
I did wonder at the time whether it was because I had blue eyes
and must look very different. And we knew someone going out
at a similar time said that she felt the children were only picked
up by the orphanage director who was a man and that maybe the
women did not have enough time to pick the girls up and cuddle
them, so we considered that this was a possibility as well. I was
told that there were ten babies to a room with one carer so it
94 Ethical Research with Children
must have been quite busy. Whether it was that kind of thing
or whether my difference in colouring or both, I’m not sure.
But obviously we knew it was going to be really, really strange.
She eventually let me give her a chocolate biscuit and that was
fine.’
While mothers speak of strangeness, and rationalise rejection, the
girls construct similar but differing stories, positioning themselves
as babies to explain their rejection of their adoptive parents. Wim
(Sally’s daughter) tells her version of the same story:
Wim: ‘When I was in China, I was a baby so I don’t really know
what happened. I didn’t like Mummy so Daddy had to carry
me when we were in China. But when we got back home,
Mummy gave me a chocolate biscuit and I liked her then [she
laughs] Mummy thinks it was because Daddy had darker eyes
and Mummy has light blue eyes and I might have been used to
Daddy’s colour eyes.’
In this different example, highlighted below we can also see how such
important stories are co-constructed. Here, Lisa also recounts a story
of initial rejection by her towards her parents. Lisa’s younger sister
Jane interjects to explain her elder sister’s behaviour:
Lisa: ‘I cried. I remember when I kept stealing the biscuits. Well
Mummy and Daddy gave me a biscuit and I ate it and then they
gave [orphanage director] a biscuit and I took it because she was
right next to me. They were pink ones.’
Jane: ‘You didn’t understand. You were just a baby.’
Lisa: ‘My Mummy calls me Peaches because when she got me I had
a hole in my bottom [split in trousers traditionally used in China]
and when she held me for the first time she said it was as smooth
as a peach.’
Being ‘a baby’ explains the inexplicable, the rejection of the parents
they both now love and trust. Both girls laugh about this response to
their parents and use humour in telling how the eldest also got her
family nickname. Lisa claims these accounts as her memories. This
perhaps is not surprising as she has been told these stories over the
six years since her adoption, and has seen the video of it taking place
The Illusion of Autonomy 95
frequently. She has participated in the telling of the story with her
family members (as does her younger sibling despite being absent,
and not even born when the event took place) and thus here we
see how such family stories are woven together into narratives of
belonging (Richards 2013).
Acknowledgement that data is collected through these relation-
ships causes us as researchers to confront epistemological issues in
our research, to reflect on whose voices are being promoted and
how we construct our analysis. The ‘autonomous child’ discourse so
prevalent in participatory research allows for the decontextualised
abstracted (Komulainen 2007) voices of children to be elevated at
the expense of the socially embedded children. Consequently we
lose their expertise in social contexts and their negotiation of social
relations.
A further example in Sarah’s research of the agency exerted through
social interaction is provided by an account which again took place
in the home where a mother stayed throughout the data collec-
tion with her two daughters. On occasion this mother prompted her
daughters by reminding them that they had relevant information to
complete the pages (and that she knew this because she had pro-
vided them with the stories). As discussed in detail in Chapter 2, in
certain instances one of her daughters chose to remain silent. Her
mother began to reprimand her daughter and made suggestions as
to what could be put on the various blank pages. However, Jenny
(aged 12) continued to refuse to complete certain pages, focusing
instead on those in the book that could capture her contemporary
friendships, travel aspirations, and, on occasion, challenge parental
decisions about out of school activities, the lack of a family pet and
holiday destinations.
Sarah grew concerned that Jenny may have been coerced into par-
ticipating and so reiterated to both mother and daughter that leaving
pages blank were perfectly acceptable as a response, and that stopping
was also fine at any point. However, Jenny continued to complete
other pages and talked at length with her sibling about what her
sister was writing in her book. Interestingly, Jenny claimed author-
itative knowledge of her younger sibling’s unfolding adoption stories
and corrected her on occasion in the interview, yet claimed to have
no knowledge of her own. Sarah’s concern about Jenny’s informed
consent was confirmed when in the subsequent interview with her
mother, Linda revealed that her interest in participating had in part,
96 Ethical Research with Children
been to try to find a way to get her daughter to talk about a subject
that she was silent about within the family. At that point Sarah made
the decision not to include Jenny’s data. However, when she looked
in detail at her book and listened to the recording of the data collec-
tion, it became apparent that Jenny had talked about things that she
wanted to share and knew that her participation was on this basis.
This left Sarah with a dilemma, to exclude the stories that Jenny had
given on the basis that she was using the research to continue an
ongoing challenge towards her mother was not a particularly ethical
decision. Should Sarah exclude her voice? As a 12-year-old girl Jenny
used this event to continue a family dynamic with her mother and
had left pages blank clearly knowing that this was frustrating to her
parent. Jenny’s ability to both comply with her mother by taking part
but continue to remain silent about a subject that her mother wished
her to talk about demonstrates skills of autonomy within her social
relations.
As a guest in this home, Sarah struggled to understand the unfold-
ing social and political dynamics (Mayall 2000) and as a researcher
she questioned and anguished over the ethical ramifications of what
was taking place. Slote (2007) argues that parents, who allow for
the individualities of children, demonstrate respect for their opin-
ions and aspirations, ultimately, facilitate a nurturing environment
for the acquisition of autonomous thinking. The home environ-
ment is clearly a location where children socially engage and prac-
tise autonomous thinking within social interactions. For researchers
working in the homes of these children, autonomy and agency can
become problematic to determine given the social situatedness of the
children involved. The ethical implications for participatory research
can present real dilemmas during data collection activities. The dis-
connect between the dominant definitions and applications of these
terms and the ways that children practise them is not as yet well
acknowledged. Greater recognition of the relationships and interde-
pendencies of all those involved in the research process at the pro-
posal and ethics application stage would better prepare researchers
for the challenges that these can present in the field. The interdepen-
dencies of relationships, including that with the researcher, require
an ethic of care (Cockburn 2005), an essential and as yet arguably
neglected methodological aspect. A moral responsibility to care for
the participants and the relationships forged and developed in the
The Illusion of Autonomy 97
research is paramount rather than the adherence to a disconnected
set of individualised rights and responsibilities. In this perspective,
ethics are viewed as a principle of nurturance (Cockburn 2005), trust
and negotiation (Richards 2012b, 2013).
Conclusion
Within social interactions, the expected repertoire of adult/child rela-
tions allows the actors involved to predict behaviour associated with
these particular identities (see also Chapter 5). This allows for consis-
tent and coordinated social interaction. Yet improvisation on the part
of the actors within these social contexts is constantly occurring (Tilly
1994). Children are aware of their expected roles, yet they demon-
strate their capacity for evaluative judgement of their evolving social
contexts and can be seen to be agentic in shaping these social situ-
ations to their advantage. Consider Wim’s success in ensuring that
Sarah played the role of teacher despite Sarah’s wish that Wim make
the decisions, or the expertise with which Jenny used silence as a
strategy of resistance towards her mother.
Assuming a perspective of interdependence enables us to recog-
nise the ways that children participate and the contribution they can
make to social research. As we and others have previously argued,
participation is not straightforward or progressive, and it cannot sim-
ply be elicited through application of a particular method alone
(Waller and Bitou 2011), but it is always relational. Children are
socially embedded, and terms such as ‘autonomy’ therefore ill-define
their agency. An ethical stance which seeks to emphasise indepen-
dent agency, inaccurately positions children and seeks to elevate their
status inappropriately. An emphasis on interdependence in research
would more effectively reflect the social positioning of children as
social beings, not as autonomous agents, and provide a more effec-
tive way of promoting their participation ethics in research needs to
focus more on the embedded nature of children’s participation rather
than the pursuit of agency or a rational autonomy.
The relationships that are frequently a focus of research are
those within childhoods and amongst children, rather than the
multiple, intergenerational relationships where children success-
fully collaborate with adults and which demonstrate the interde-
pendencies of children and adults (Ennew 1994; Mayall 2000a).
98 Ethical Research with Children
The interdependency of adult and child, researcher and participant
remain overlooked and yet key to understanding the context in
which research with children is situated. Participant-led methodol-
ogy should seek to undermine the binary logic between self and
‘other’ that plays such a key role in identity construction and the
development of autonomy.
Being ethical in this perspective then is to achieve and sustain
an awareness of the social context, the social agents, the power
relationships and their influences on the research process. Work-
ing towards this requires both a personal reflexive approach towards
subjective values and positioning as well as an epistemological one
that explores the role of method in knowledge construction. Such
reflexivity remains underexplored in much research (Madden 2010).
Treading with care through these multiple relationships requires not
only a reflexive approach but also an ethic of care.
We contend that to engage in research with children, we need to
recognise more effectively the ways that children exert agency and
negotiate power relations within their social contexts. Rather than
autonomy of selfhood, children exhibit a relational social agency
that has yet to be recognised in ethical frameworks. Such procedures
continue to protect the vulnerable child while seeking to engage
with an autonomous one. Such contradiction cannot be ethical nor
reveal children’s actual characteristics. As this chapter reveals, chil-
dren exhibit agency in ways we may not expect or immediately
recognise. But abstracting their voices from the social context and
attributing autonomy denies the skilful ways that children manage
relations as interdependent beings.
5
Ramifications of Category
Entitlement: In What Ways Does
Who We Are Determine What
Others Will Say?
Introduction
The researcher-self as a significant presence in the field has become
more prominent in research discussions but, though established, the
issue remains contested in research debates. Our aim for this chapter
is to acknowledge the presence of ourselves in our research and begin
to explore the ways in which who we are shapes what we find and
why we find it. We use a specific focus of category entitlement to
explore the situated and connected category memberships of those
involved in our own research with the express purpose of extending
the debate beyond ideas that the researcher is influential, to explore
how this influence is established and articulated.
Extensive literature is available which explores the role of the
researcher in the field (see, for example, Wheatley 1994). Perhaps less
is evident in how the researcher’s subjectivity shapes the analysis of
the data, with the possible exception of notable feminist arguments
(Alvesson and Sköldberg 2009; Mauthner and Doucet 2003). In addi-
tion, limited literature is available which takes a reflexive perspective
concerning the identity of the researcher and how their identity traits
influence and shape the research process (Holstein and Gubrium
1995). Who takes part, what data they provide, how it is analysed and
finally written up and presented is often overlooked. Debate around
the extent to which the researcher shapes and influences the field
(Coffey 1999), how they use subjective knowledge to illuminate the
field in ways that objective observers cannot (Oakley 1981), or indeed
99
100 Ethical Research with Children
whether they contaminate the field (Atkinson and Coffey 2003), are
common but limited in their explorations.
To overemphasise our presence in research is a rather arrogant
(Atkinson and Coffey 2003) and an essentialistic (Howarth 2002)
position to assume. Yet to deny our position and its influence is
illusionary (Mauthner and Doucet 2003; Sin 2005) and somewhat
disingenuous. However, to actually recognise the extent of our influ-
ence in our own research is both challenging and time consuming
and, if acknowledged, opens our research to scrutiny through diverse
and contested values about validity and neutrality in the research
process. Dunbar et al. (2002:131) challenge researchers not to treat
interviewees as mere vessels from which stories can be abstracted
and presented as objective facts devoid of the social and historical
context in which they are told. They claim that this results in sim-
plistic portrayals of complex social settings. We contend that the
researcher producing this knowledge should not be abstracted from
their social context either. Both interviewer and interviewee need
to be acknowledged as situated participants, whose identities play a
pivotal role in the research produced. Song and Parker (1995) claim
that far more research is needed on how respondents and researchers
establish identities in social research. What emerges if we study inter-
views through identity attributes are displays of cultural practices by
its members (Silverman 1993:114) as they construct their identities.
It has been argued that identity and location cannot be erased
in an attempt to ‘clean up’ data (Howarth 2002:29). Yet the expec-
tation of objectivity is that such contaminants be removed or at
the very least minimised. Holstein and Gubrium (1995:114) con-
tend that active interviews should be built on the understanding
that ‘any attempt to strip interviews of their interactional ingredients
will be futile’. Despite these and similar claims, published research
frequently neglects to account for the attributes of those who col-
lect data, where this is done and how. Differences, demographic
and cultural between researchers and respondents are often ignored
with the possible exception of some ethnographic studies (see, for
example, Pole 2007) and yet these differences shape how accounts
from respondents are understood and analysed (Poindexter 2003).
We claim that similarities between respondents and researchers are
equally neglected and yet play a significant role in the interview
construction and subsequent analyses. This neglect not only reduces
the epistemological richness of research based discussions but it also
Ramifications of Category Entitlement 101
has ethical implications too. Fine (1994) argues that this separa-
tion of researcher and respondent contributes to the ‘othering’ of
respondents and their experiences and produces narratives where
researchers speak for others without acknowledging their own pres-
ence and investment in the final product (Poindexter 2003). This
is perhaps particularly so in research which aims to elevate the
voices of some groups such as children. We question the extent to
which abstracted voices devoid of context successfully promote val-
ues of empowerment and participation espoused so frequently in
participant led approaches with children.
This chapter explores the ‘processes of production and interpre-
tation’ (Aull Davies 2008:110) of data by applying three connected
topics that develop and use category entitlement to allow the influen-
tial identity traits of the researcher and respondent to be illustrated.
Category membership, self-disclosure and insider status are all related
to how individuals in social research establish relations and demon-
strate their authority to speak on the issues researched. We detail the
implicit and explicit connections established in the research con-
text and consider the ways in which these connections build an
entitlement on the part of the researcher to ask, understand and eval-
uate data as well as enable the respondent to establish a legitimacy
to provide responses on a given topic. We also consider the ethical
implications which become explicit when acknowledging the con-
nections between researchers and researched. What we offer in this
chapter is analysis of the relationships and identities which emerge
and influence the research but which are frequently left out of pub-
lished accounts. From this perspective all participants involved use
personal histories, identities and subject positions (Phoenix 2008) at
various stages of the research process to achieve diverse objectives
and perform various roles. We focus on interviewer and interviewee
identities and explore the influence of their category membership
and entitlement on the data collection and analysis. Initially we clar-
ify category membership and illustrate this outline with examples of
transcripts from our own research.
Category membership
Literature concerning the attribution of people to particular socially
constructed categories is well established. Hacking (1986) for
example, provides a comprehensive, philosophical discussion of the
102 Ethical Research with Children
emergence of particular categories, and the ‘making up’ of peo-
ple. He explores how categorisation ‘affects our very idea of what
it is to be an individual’ (1986:161) and argues that the particular
social and historical contexts in which categories such as ‘pervert’ or
‘Parisian waiter’, emerge, ‘create new ways for people to be’ (p.165).
What individuals are able to be and do within these categories
is inextricably linked to their description. Using Sartre’s argument,
Hacking compares these categories to imprisonment where our indi-
vidual potential and possibilities are constrained by the category
descriptions we are attributed to. Child, adult, researcher, adoptee,
or disabled, for example, are categories of particular relevance in our
discussion. Hacking contends that through the process of categorisa-
tion, experts or those with power, can create ‘realities’ or truths about
groups of people but, that as individuals, we also have the capacity to
resist such ‘realities’ (p.168). This perspective can easily be applied to
social research where the researcher, often categorised as an expert,
constructs realities about social groups and individuals through the
research process.
It is argued that social categories are instrumental in much social
research, often regarded as the ‘principle building blocks’ (Potter
and Wetherell 1987:116). People are taken to be members of rela-
tively enduring social categories. We each carry categorical labels,
such as mother, father, teacher, adult, child, and are attributed char-
acteristics accordingly. Potter and Wetherell (1987:137) argue that
we each construct versions of the groups or collectivities to which
we belong. These are then used to build and interpret our social
worlds; however, these structures are neither solid nor static, but are
shaped and adapted in each social interaction. Garfinkel (1967) and
Sacks (1992) claim that there exists a range of culturally available
knowledge, which we use to interpret and make sense of the actions
of others and which inform our own actions and behaviour. This
knowledge is embedded within social life, implicit and unnoticed
but such
categories are not nested in a clear-cut, natural way like Russian
Dolls, for example, one inside another, but are grouped into col-
lections by the use of potentially complex and contradictory sets
of interpretative procedures
(Potter and Wetherell 1987:128)
Ramifications of Category Entitlement 103
Sacks (1974) argues that people use knowledge of how categories are
organised as a resource when making sense of discourse in the social
world, which makes category membership an important research
concept.
Individuals categorised as members of a particular group are associ-
ated with specific activities and attributes which enable them to make
sense of their social environments. Category memberships enable
each of us to make sense of the discourse and behaviour of others
(Baker 2003; Sacks 1974; Silverman 1998). We use it to predict how
people in that category will behave and what activities they will do.
For example we can all predict some of the activities associated with
a category of mother or child (Sacks 1974).
People specifically indicate their category memberships as part
of their discourse and analysis of such conversational interaction
is most highly developed in the field known as conversation anal-
ysis (CA). This approach originated in the work of Harvey Sacks,
Emanuel Schegloff, and Gail Jefferson, is also closely related to
ethnomethodology and accounts for how we make sense of each
other and build knowledge through our category memberships. It is
defined by Heritage (1984:4) as
the study of a particular subject matter: the body of common-
sense knowledge and the range of procedures and considerations
by means of which the ordinary members of society make sense
of, find their way about in, and act on the circumstances in which
they find themselves.
Baker (2003:395) contends that such analyses consider how par-
ticipants make sense of each other, how knowledge is built, how
identities are negotiated, and how participants ‘characterize and con-
nect the worlds they talk about’. Howarth (2002:23) argues that a
‘theory that forges a link between what we say, who we are and
where we are located, provides valuable inroads into epistemological
debate’. Analysis of membership categorisation is described by Baker
(2003:401) as being a ‘powerful approach to uncovering the reflexive
relation of speaker, audience, and topic in research interviews’. From
this perspective interviews can be used to explore how participants
claim membership and accomplish category entitlement to knowl-
edge and experiences, and convey this through the interview process.
104 Ethical Research with Children
Yet examples of children using category membership in research is
curiously absent.
Using categories in everyday discourse involves language, social
expectations and attributes associated with membership of specific
categories, which makes analyses of these membership categories
highly relevant in research discussions. Category membership also
provides the participants and researchers with authority to offer
valid data collected in the interview process (Potter 1996). This is
done by members claiming to hold ‘expert knowledge or privileged
experience’ on research topics (Phoenix 2008:71).
Access to children can be challenging to achieve in a contemporary
climate of concern about adults relations with children (for example,
Monaghan et al. 2012). It is also one of the first ways that category
membership can be explicit in the research process. To gain such
access Sarah, in her research exploring intercountry adoption, used
various category memberships. In the ethics application Sarah identi-
fied specific characteristics and qualifications each indicating specific
knowledge and entitlements designed to enhance approval. Lecturer
in childhood, qualified teacher, adoptive parent and bereavement
counsellor were all specified, each category suggesting knowledge
assumed to be pertinent to the topic and respondents and each
making the ethics committee more comfortable for the research to
proceed.
Having eventually gained ethical approval the process of recruit-
ment began where category membership was also found to be
influential. Despite a long-standing significance of values such as
neutrality and objectivity (Poindexter 2003) in research discussions,
participants are commonly selected on the basis of being members
of a particular category and therefore holding specific and subjec-
tive knowledge relevant to the research topic (Howarth 2002). Baker
(2003:401) argues that researchers recruit representative respondents
using membership categories. Interviewees are then intended to
sound like members of that category, speak from within that particu-
lar category and are accountable for speaking as competent members
of that category. To encourage respondents to take part, researchers
can also identify their membership to specific categories. For exam-
ple in the information letter sent to parents, Sarah identified herself
as a lecturer in Early Childhood Studies explicitly placing herself
in a category where specific knowledge about children would be
Ramifications of Category Entitlement 105
assumed. It also implied a position of trust to encourage parents to
be more willing to agree to their child’s participation. In the subse-
quent phone conversations Sarah identified herself as an intercountry
adoptive mother where an element of trust might be implied through
being an insider of this small and rather private group. Sarah also
used her membership of this group to recruit participants at CACH
(Children Adopted from China) events where only adoptive fami-
lies were permitted to attend again explicitly situating herself as a
member of the group that she wished to recruit from. While acknowl-
edging the significance of objectivity in research debates Sarah, on
reflection, purposefully and strategically used her membership cate-
gories to achieve specific outcomes in recruitment. We question the
extent to which this is generally acknowledged in research and argue
that there is a need for a greater reflection on how memberships can
shape our recruitment strategies.
Notwithstanding the above arguments researchers also interpret
interview data from within their roles (Rojiani 1994) and particu-
lar categories which makes these category memberships an impor-
tant element of the research process. Poindexter (2003) argues that
where we position ourselves determines our position on the top-
ics we research. In accordance with this perspective we begin to
explore more fully the role of self-disclosure where explicit exam-
ples of social actors establishing category membership are illustrated
within the discourse. Challenges and opportunities which an insider
status provides are then explored before evaluating the use of cat-
egory membership in research relations upon which entitlement to
contribute is established.
Self-disclosure
The issue of self-disclosure is commonly absent in social research dis-
cussions and little attention has been paid to the interviewer’s use
of self-narratives of disclosure (Abell et al. 2006:223). Researchers can
be concerned that to identify when and how they self-disclose in
the field (if they have) may reduce the objectivity and impartiality
of the data they collect and cast doubt on its validity. Such disclo-
sure is often regarded as ‘contamination’ in conventional research
(Dunbar et al. 2002:138). Holstein and Gubrium (1995:13) note that
interviewers are generally expected to keep their ‘selves’ out of the
106 Ethical Research with Children
interview process’. Consequently such accounts are cleansed from
content where self-disclosure and connections between respondents
and researchers have occurred leaving the complexity and ambiguity
of these research relations underexplored.
To self-disclose is to offer a connection between the researcher and
the researched, to invest in them as social beings (Dunbar et al. 2002).
To actively self-disclose has been referred to as ‘creative interviewing’
where common but not necessarily comfortable ground can be estab-
lished (ibid.). Abell et al. (2006:224) argue that interviewers activate
category membership through self-disclosure in order to develop a
rapport, or to identify a difference with respondents. Oakley (1981)
argues that the researcher who is prepared to invest their personal
identity in the relationship gains greater insight into the lives of
those being researched in part because self-disclosure by the inter-
viewer can also effectively facilitate further interview talk from the
respondent (Abell et al. 2006). But to gather data as a result of such
connection opens up complex epistemological debates.
Our data illustrates that children, like adults, are using categories to
position themselves in the research context. Children can be seen to
account for themselves as competent members of the social category
assigned to them by the interviewer as well as claiming membership
to categories that the interviewer has not assigned them. We con-
sider examples where respondent’s membership to specific categories
is assumed by the interviewer (Phoenix 2008; Song and Parker 1995)
but challenged by the interviewee:
Kate (nine years): ‘Why do you want to talk to me just because I’m
adopted?’
Sarah: ‘I’d like to hear lots of your stories about friendships and
about activities and school and I would also like to hear your
adoption stories partly because we don’t know much about
these stories and partly because I have a daughter who was also
adopted from China. She is still very little and I am interested in
hearing what older girls think and feel about being adopted from
China.’
Kate: ‘What province is your daughter from?’
Sarah: [I name the province]
Kate: [with a look of surprise on her face] ‘That’s my province. I
was born there too.’
Ramifications of Category Entitlement 107
Kate: [sits down on the floor next to her older sister, opens the
activity journal and picks up a pen]
Sarah: ‘Kate, would you like to do the activities in the book and
talk to me?’
Kate: [nods as she reads the first page]
Lull (1985) contends that researchers need to develop and maintain
good relations with children and family members while also attempt-
ing to maintain an objective stance. Jordan (2006:177) claims that
to connect with families in social research and to obtain valid and
textured images of what their lives are like may require a level of self-
disclosure that many social science researchers such as Lull and others
would consider problematic. Jordan (2006:170) further argues that
in the field researchers are understood and labelled by family mem-
bers as they attempt to ‘construct a role for the researcher and define
the situation according to their needs, understandings and experi-
ences’. At first glance the above interaction might appear to be the
first stage in the development of good relations between Sarah and
Kate. However, what is also taking place is the establishment of com-
mon ground between two social actors through shared knowledge of
category membership.
Part of what takes place in the interaction above is the co-
construction of data (Jordan 2006) where Sarah is a component
of the cultural environment that she studies (Denzin 2000; Oakley
1981). Sarah’s interaction with Kate illustrates a clear example of a
researcher’s self-disclosure. It serves a specific purpose, an evident
attempt on Sarah’s part to connect with Kate and part of the ongo-
ing renegotiation of informed consent. Kate is active in identifying
what she needs from Sarah to make sense about why she is being
asked to participate and why Sarah is interested. Baker (2003:410)
contends that the social order and rationality that participants such
as Kate are seeking to ascertain is not imposed in an interview con-
text but ‘assembled’ by the participants using their ‘common sense’
member[ship] resources’. In consequence such interviews are displays
of ‘culture in action’ (Hester and Eglin 1997).
As discussed in Chapter 4, gaining informed consent is a dynamic,
ongoing and often complex process, that requires us to reflexively
consider (Sin 2005) how it is negotiated and re-negotiated in the
field. However, the complexity of gaining it and its relationship
108 Ethical Research with Children
to category membership and self-disclosure is overlooked in many
research texts.
In the account above Kate is seeking information that will help
her to either refuse or agree to participate in the interview. Ini-
tially Sarah gives some general information about why she wants
to talk to Kate but then discloses that she is an adoptive mother
with a daughter who was also born in China. It is this informa-
tion in particular that Kate employs in her next question by asking
for more personal information which she then goes on to use in
her decision to participate. Holstein and Gubrium (1997:123) argue
that during ‘active interviewing’ the role of the interviewer is to
‘incite respondents’ answers, virtually activating narrative produc-
tion . . . by indicating-even suggesting-narrative positions, resources,
orientations and precedents’. In response to Kate’s questions Sarah’s
self-disclosure effectively positions her in the wider canonical and
contested narratives about intercountry adoption (Richards 2012b).
This positioning as a member of the intercountry adoption com-
munity was perhaps taken as an indication to what narratives were
wanted and how they might be received. This assembling ensures
that both separate individuals, regardless of their adult or child sta-
tus, become recognised as members of the same group. This common
ground establishes the basis for the interview.
Sin (2005:290) argues that in seeking and gaining consent, objec-
tivity and impartiality ‘are mere illusions’. If such subtle articulation
of membership categorisation is indeed taking place in the early
interactions between researcher and respondent then connections are
being established during informed consent negotiations which shape
the data to come. How these connections are achieved and what
impacts they potentially have should be subject to greater interro-
gation in the wider research community. A further important point
to make here is that one participant of this exchange is a nine-
year-old girl where it is evident she is using category membership
to inform her decision making. This illuminates children’s capacity
and expertise to engage in research as competent social agents. Kate
demonstrates explicit social skills as she gathers the information she
needs from Sarah using category membership.
Intercountry adoption as a membership category revealed in the
interaction above is picked up later in this chapter but here we want
to identify the role and use of self-disclosure as a means to alleviate
Ramifications of Category Entitlement 109
anxiety about taking part in the research, establish a sense of trust
through such disclosure and to allow a rapport between Sarah and
Kate to develop. Jordan (2006:173) argues that researchers are stud-
ied by respondents for the impressions they ‘give’ and ‘give off’ in
order to anticipate how the researcher will respond to them and their
knowledge. If self-disclosure achieves these elements then it is poten-
tially a very useful (if epistemologically messy) tool and ‘strategy’
in the research process (Abell et al. 2006:223). Ethnomethodological
analyses of interview data are argued to be a far more reflexive
research practice than most other approaches. Baker (2003:411)
contends that through such analysis it
becomes impossible not to see the artful practices of interviewer
and interviewee in making the interview happen, and conse-
quently it becomes very difficult to unhitch ‘answers’ from their
(em)bedding in an actual, local situation of production
Some researchers claim that self-disclosure can be used to manage
and ‘counteract’ unequal power relations in research relationships
(Abell et al. 2006:221). Eder and Fingerson (2003) contend that
researchers can empower young respondents by self-disclosure which
enables them to share similar experiences. Sarah’s self-disclosure
clearly facilitated disclosures from Kate. It is argued that when those
being studied have a sense that they are understood by someone who
faces the same life challenges, a certain level of reciprocity and trust
can emerge (see also Oakley 1981). Sarah certainly was not claim-
ing similar life challenges but she did imply a connection through
adoption which seemed to influence Kate’s decision to take part.
Poindexter (2003), like Sin (2005), argues that self-disclosure plays
a role in positioning the researcher in the interaction. However, she
adds that this should not be ‘excessive, gratuitous, or self-serving’
(2003:401). Though it is hard to argue that self-disclosure is not self-
serving given the ways in which it is articulated in conversation.
No one self-discloses to deliberately disadvantage themselves in the
social interaction. Saying no to an unknown researcher in your liv-
ing room may be hard but saying no to a group member is perhaps
harder, particularly between adult and child. Sarah may have been
seeking to establish a rapport but she used ways to articulate connec-
tion and category membership to do this. Did Sarah’s self-disclosure
110 Ethical Research with Children
make it easier for Kate to consent or harder for her to refuse? These are
the dilemmas that come when we acknowledge that the research pro-
cess is more messy and more agonising than we often admit it to be.
Respondents may however, not always interpret the interviewer’s
behaviour in the intended way. Abell et al. (2006:241) argue that
through a sharing of experiences the interviewer paradoxically
exemplifies differences between themselves and the interviewee
in terms of age, gender, social class, race, religion and education.
Moreover, such strategies may display an interviewer’s greater cat-
egory entitlement to provide information about a particular topic
than the interviewee.
Though underexplored in research studies, self-disclosure can clearly
fulfil several functions in research strategies related to informed con-
sent, and building trusting relationships as well as have consequences
in interviews. Despite extensive debate and value placed upon objec-
tivity and neutrality, Reinharz (1979) argues that human research
should use the researcher as an instrument with tools like personal
experiences. These are increasingly becoming recognised as valid
sources of scholarly knowledge (Riessman 1994 in Shah 2006:217).
Sarah certainly used self-disclosure and category entitlement in gain-
ing respondents for her research and on occasion in the interviews as
a way of facilitating discussion. This approach was described by one
interviewee as being
[o]ne mother talking to another who has knowledge of what I am
talking about.
The quote above leads the discussion onto the challenges faced by
being categorised as an insider in the research context. Initially some
of the ethical implications of this status are identified.
Insider status
Coffey (1999:28) has claimed that the way we conduct our research,
the role we take, how we understand and interpret the knowledge we
are given is shaped by who we are and the categories we belong to.
Griffith (1998) extends this perspective by arguing that who we are is
Ramifications of Category Entitlement 111
continuously moving between aspects of the researchers biographical
profile, the relationship with the community under study and the
necessary ‘outsideness’ of being a researcher.
Sarah’s self-disclosure of her status as an intercountry adoptive
mother was instrumental in gaining respondents in what is a cel-
ebrated and contested field of family formation (Richards 2013).
Having gained respondents, her insider status as an intercountry
adopter arguably shaped the knowledge she was given and how she
understood the stories told. Those championing the role of insid-
ers in social research (see, for example, Leicester 1999; Oakley 1981;
Shah 2006; Taylor 2011) claim that ‘as a result of social distances,
interviewees may not trust researchers, understand questions, or
they may purposely mislead their responses’ (Miller and Glassner
1997:101) Though insider status may hold some significant advan-
tages (see Montgomery 2007), it also complicates the researcher’s role
and their membership in the group where they are considered to
be the insider (Taylor 2011). While some research approaches clearly
find this stance problematic, Coffey (1999:47) argues that these ten-
sions serve to remind us of important ethnographic dichotomies,
‘involvement versus detachment’, ‘distance versus intimacy’ and
‘stranger versus friend’.
The following account explores the articulation of Sarah’s status
as a member of the group being studied, and how this shapes the
interaction, is explicit:
Angela: ‘I explained about the question that people always ask,
you know the one . . . “Is she yours?” I explained to Lisa [her
daughter] that “they are not asking if you are my daughter but
actually asking if you are adopted”. “People do not always know
how to use the right words but that’s what they mean.” I wanted
to give her a framework for her about the questions and so I can
say to her “do you want Mummy to answer this question? And if
so what would you like Mummy to say?” And she said, “yes tell
them I am yours and that I am from China”, and so I said “ok”.’
‘You know the one’, is said based on Sarah being a member of the
category of intercountry adoptive mother. Angela as the intervie-
wee seeks Sarah’s acknowledgement to confirm the commonality
of intrusive questions about family members. The strategies that
112 Ethical Research with Children
the interviewee then goes on to outline in response to intrusive
questions are also provided in the context of being between one
mother and another where mutual understanding is assumed and
displayed. Krieger (1987) contends that such knowledge can produce
insight into the topic itself while also illustrating the complexities
of positions for the researcher. However, acknowledgement of insider
status and category membership on the part of the researcher is ini-
tially required along with a critical reflexivity of the ways in which
these memberships shape the research. Ristock and Pennell (1996:66)
claim that researchers should start their analyses from the personal
and indicate the ways in which personal locations and identities
inform and shape the research process. However, like self-disclosure,
insider status calls into question a researcher’s capacity to research the
field as others might. Familiar environments are argued to ‘produce
superficial accounts which fail to penetrate the taken-for-granted
relationships that are routine and mundane for the insider’ (Howarth
2002:22).
This is contested by researchers such as Leicester (1999), Oakley
(1981) and Shah (2006:210) who each claim that ‘interviewing indi-
viduals with similar experiences encourages the generation of richer
material’. Feminist social researchers have provided considerable lit-
erature to the debate about the strengths and, indeed, the limitations
of researching as a group member or an insider, defined by Thornton
(1995) as holding ‘subcultural capital’ and by Roseneil (1993) as being
‘empirically literate’. Many claim that prior knowledge of the group
and its members facilitates deeper understanding and a more trust-
ing rapport with participants (Taylor 2011). However, holding insider
status represents ethical challenges (Stacey 1996) in the field and
unforeseen dilemmas. The following account illustrates some of the
ethical complexities of being an insider in the research field.
‘I’ll do yours if you do mine.’
The above quotation came from a potential respondent in Sarah’s
research who expressed an interest in taking part with her children.
Participants were being recruited through the adoptive organisation
which Sarah was also a member of. Sarah asked if her children
had any questions about the research and whether they were also
interested in taking part. The answer suggested that the children
Ramifications of Category Entitlement 113
would participate if the mother wanted them to and that, in turn,
she would like Sarah to contribute to her own photographic study.
Sarah’s desire for respondents (she had very few at this point) was
strong and refusing such a request from a willing one risked alienat-
ing her, particularly problematic given their shared membership of
the organisation Children Adopted from China (CACH). However,
the children were not (at this point) offering their own consent.
Furthermore, Sarah was also being asked to disclose personal infor-
mation and artefacts that as a mother she perceived not to belong
to her but belonged to her daughter to either offer or refuse. Sarah
attempted to reduce tensions that come with saying no by suggest-
ing that she had to comply with an ethical framework laid out by the
university (which, of course, is true), and that this mother’s children
were perhaps not ready to offer their consent. In addition, Sarah’s
own daughter was not yet old enough to make an informed deci-
sion about offering the requested material. Sarah suggested that each
could get back in touch with the other if things changed. No further
contact was made.
Homans (1949 in Plummer 2001) has claimed that methodology is
about strategy not morality, but we argue that methodology moves
beyond simply a strategy which sets out a moral and ethical position
representative of little more than a pattern for ‘following the rules’
(Plummer 2001:227). It also involves the researcher in making com-
plex, ethical decisions in the moment (Wall 2010) through personal
moral perspectives. Such decisions are complicated by category mem-
bership relations. Multiple categories were articulated in the above
account: researcher, mother and member of CACH all with values
and conventions which need to be negotiated while maintaining
good relations with another member. Each of these discrete cate-
gory membership roles are connected, emerging and influencing each
other in the research field. Mauthner and Doucet (2003:421) claim
that ‘the interpersonal, political and institutional contexts in which
researchers are embedded’ play a key role in shaping their decisions in
the field (see also Bell and Newby 1977; Bell and Roberts 1984). How
these category memberships shape the role is still underexplored in
research publications.
All the examples in this chapter rely on the relevant knowledge
that comes with shared category membership and insider status
whereby the social actors involved are able to better understand
114 Ethical Research with Children
each other. Disagreement remains about whether or not researchers
should be members of the groups that they study, despite persua-
sive claims that subjective knowledge is necessary to appreciate the
life experiences of individuals. Maintaining a balance between ‘aca-
demic credibility’ and ‘community accountability’ can be difficult
and induce dilemmas about what to disclose as a researcher and
what to omit in the interests of being a member of the group (Taylor
2011:14). Burke (1989 in Taylor 2011) questions whether such use of
intimate knowledge can be considered ethical and has described this
stance as ‘privileged eavesdropping’. Others more simply claim that
researchers need to be skilled in maintaining the boundary of being
both researcher and insider (Taylor 2011).
What is evident is that insiderness and category member-
ship induce ethical dilemmas which all but the most reflexive
ethnographic accounts exclude from discussions (Madden 2010).
Situating ourselves both socially and emotionally in relation to
our respondents is an important element of the reflexive process
(Mauthner and Doucet 2003:419). However, it is too simplistic to
identify categories such as gender, sexuality, class or even family
formation and imply understanding, commonality or shared val-
ues as a result of these categories. Howarth (2002:23) claims that
an assumption that ‘common social identification produces shared
knowledge and experiences’ is ‘patronising and essentialistic’. The
account above, whereby Sarah was asked to reciprocate research
participation, revealed greater differences between researcher and
participant than shared values or experiences, but the interaction
was premised on an assumption of insider status between fellow
members of the same small group of individuals. In this instance
for the researcher, a category of insider status made consent nego-
tiations more problematic where discussions not only involved the
research activity but ongoing membership relations in a shared
group. Research has shown what Taylor (2011:13) describes as ‘rules
of engagement’. Belonging to specific groups and holding insider sta-
tus also has rules, and the two are not necessarily compatible. This
makes ‘role displacement and confusion’ (ibid.) important aspects
of these relationships, which should at the very least be identified
and explored for their ethical implications. Ultimately, researchers
‘can only strive to be intellectually poised between familiarity and
strangeness’ (Hammersley and Atkinson 1995:112). But without more
Ramifications of Category Entitlement 115
research-based interrogation of how familiarity and difference are
constructed and used in interviews and subsequent data analysis, lit-
tle can be gained beyond a token acknowledgement that familiarity
has to be managed.
Baker (2003:401) claims that ‘[i]nterviews are definitely not “time
out” from the social worlds that the participants are talking about;
rather, they are reflexive descriptions of those worlds’. This chal-
lenges the perception that interviews are ‘self-contained events’
(Phoenix 2008:71). Therefore, the identity attributes and category
memberships we embody in everyday life also shape the interview
and data creation process. Too often, identity categories are ini-
tially and sometimes tokenistically identified but then overlooked
as influences in the analyses or interpretative process. We suggest
that our respective multiple category memberships should be made
explicit, making it possible to discuss how these intersect with the
respondents when analysing data (Patai 1991:149).
Using category membership to identify potential respondents and
then constructing category membership during interviews are impor-
tant steps to successfully establish an entitlement to speak authorita-
tively about a topic (Phoenix 2008). It is this category entitlement
that we explore next.
Category entitlement
People construct identities through telling their stories and life expe-
riences. These descriptions construct and reveal the category mem-
berships of the tellers. Through category membership, entitlement to
speak authoritatively about a subject is established (Phoenix 2008;
Potter 1996). This entitlement is not static but can be constructed or
eroded within discourse (Potter 1996).
Sarah claimed her entitlement to discuss intercountry adoption
from China by stating her membership of the intercountry adop-
tion group from where her participants were drawn. For example, she
verbally reinforced this entitlement in the discussion with Kate by
stating that she was an adoptive mother. Similarly, Allison revealed
her own membership to the ‘parent’s of disabled children’s club’
when she swapped stories with Paul, a teacher whose daughter was
also an Augmentative and Alternative Communications Systems user.
Allison reflected on how this revelation forged an unspoken bond
116 Ethical Research with Children
between them, and they each gave a silent ‘nod’ to the others
entitlement to this exclusive, members-only club.
Mandelbaum (1978) contends that stories and accounts are col-
laboratively constructed. This perspective is evident in the previ-
ous account where Angela confirmed Sarah’s entitlement to discuss
intrusive family questions through identifying the typical ques-
tion that many parents of transracially adopted children are asked,
stating ‘you know the one’. Potter (1996) argues that category
entitlement removes the necessity to question how a person has
acquired topic knowledge. Group membership is deemed as suf-
ficient legitimacy for holding knowledge of a specific topic or
domain. Through affirming this mutual ‘insider’ experience between
researcher and respondent, entitlement is reaffirmed and the rap-
port between researcher and respondent can be strengthened (Abell
et al. 2006). This arguably increases the trust and potentially encour-
ages the respondent to speak more openly and may encourage the
researcher to reciprocate through self-disclosure. Both participants
in the discourse build this category membership and subsequent
entitlement, and both are consequently committed and invested in
the discussion.
Phoenix (2008; see also Sacks 1974) argues that social actors
construct their entitlement through revealing related experiences
and how they position themselves in relation to the topic. How-
ever, she also argues that assumptions about entitlement can occur.
The following account is taken from Sarah’s research and reveals
shared knowledge, but not comfortable rapport between intervie-
wee and interviewer. Here Linda shares her experiences of receiving
a referral for adoption from China. This documentation included a
photograph of the baby she had been matched with.
Linda: ‘What arrived was the most unattractive photograph of a
funny little thing with hair sticking up obviously taken when
she was much younger than the paperwork suggested. Just for a
few minutes, I was disappointed. I think that I had this fantasy
picture in my mind, and it was not realised in the picture in front
of me and then I asked myself, “well what do you want, a fantasy,
pretty child or do you want to give a real child with nothing a
home?” And as soon as I thought that it was fine and I wouldn’t
have cared if she had turned out to be plain. It was the same with
Ramifications of Category Entitlement 117
my second daughter too. I saw her picture and thought Urgh! But
then you think, this is a child that nobody wants and the minute
I thought that it was fine. I guess the order of what I wanted them
to be like was firstly good natured, then clever and then to have
looks, but I would have accepted anything really.’
During interviews, interviewees can tell their stories and experi-
ences without knowing how the researcher may receive them but
mutual category membership provides clues about the type of sto-
ries expected and accepted (Phoenix 2008). In this narrative, Linda
admits to feelings and aspirations in motherhood that are not usual
for mothers to express, even if they may be very common. Her
account contrasts with what is culturally acceptable to say. How-
ever, as a member of an intercountry adoptive category Linda perhaps
feels entitled to reveal her feelings and can do so with an authority
of knowledge and behaviour within this group. Linda’s entitlement
to share these experiences with Sarah is therefore based on their
shared category membership. Sarah’s complicit understanding of
these non-normative statements on Linda’s part is both imagined
and assumed, while for Sarah, the desire to gather Linda’s stories,
and remain in good relations with this fellow member, mutes how
she might have otherwise responded. Shah (2006:211) reminds us
of the dangers of ‘over-rapport’ when we research a topic where we
hold what Stanley and Wise (1993:227–228) call ‘Epistemological
privilege’. Linda’s statements reinforced the need for Sarah to follow
advice given by (Bondi 2003 in Shah 2006:211) of remaining ‘emo-
tionally present and reactive to the interviewees’ responses’ as well
as reflecting on her own feelings. Category entitlement dictates that
people are treated as knowledgeable in certain contexts. Yet, assump-
tions about what one is entitled to say and how it will be received
on the basis of membership do occur, and being a member it is not
necessarily straightforward (Widdicombe and Woofitt 1995). Further-
more, people can fail to be treated as members, and some categories
are negotiable (Jayyusi 1984) where entitlement is either built up
through the interaction or undermined, as in the examples explored
below.
Particular life experiences of children are frequently the reason
for their participation in research. As Uprichard (2010) has previ-
ously argued, children do not participate in the way that adults
118 Ethical Research with Children
do in research generally but are included by virtue of their status
as school pupil, family member, young carer and so on. Categories
attributed to children, such as pupil, adoptee or disabled, categorise
children prior to and outside the normative category of childhood
and arguably serve to marginalise and exclude the very children
we seek to include in our research discussions. Their membership
of a particular category such as these is commonly the prerequisite
for taking part. In some cases, this membership not only predicates
specific knowledge and entitlements but carries a variety of traits,
characteristics and behavioural expectations (see Chapter 6 for fur-
ther discussion of adoptee attributes). Sacks (1974) refers to these as
‘category-bound’ activities associated with membership. For exam-
ple, in Sarah’s research young adoptees were expected to be engaged
in specific culture-related activities to ensure that their Chinese her-
itage was accessible to them. Some of the young participants spoke
at length about attending Chinese summer school, having Mandarin
classes, visiting China and being a member of CACH. However Finch
(1984) has argued that interviewers should not assume commonali-
ties. Category membership for children may not be something that
they have constructed themselves but can be a status attributed to
them (Drew 1978) by adults in general or researchers in particular.
Assumptions about which categories individuals wish to claim can
also occur. Sarah found this particularly so with the adoptees in her
research. The parents interviewed did not challenge the category of
adoptive parent as the primary reason for being interviewed and posi-
tioned themselves as being entitled to talk authoritatively within this
category. But as Hacking (1986) reminds us, such truths and realities
can be resisted.
A number of older girls (aged between 9 and 12 years) empha-
sised other categories that they claimed membership to during the
interviews and, on occasion, actively sought to distance themselves
and undermine their membership of an adoptive category. Some
girls constructed category membership of ‘best friend’ or ‘pupil’ as
indicated below:
Kate (nine years): ‘I like to hang out with my friends and go shop-
ping. I like to go swimming club and football club. I like to do
sporty things. I like to put on mini shows. I like going to my gang
show rehearsals.’
Ramifications of Category Entitlement 119
Louise (nine years): ‘My special friend is Sasha because she is my
best friend and very nice. We like hanging out and chasing and
annoying the boys.’
Jenny (aged 12 years) refused to acknowledge her membership of
adoptive category by remaining silent about her adoption and also,
on occasion, by challenging her mother about the obligatory nature
of the Chinese culture related activities that her membership in this
category and family demanded. In contrast, she was eloquent in her
aspirations to visit New York, to own a pet, about socialising with
friends and taking part in school activities.
Jenny: ‘School is fun because all of us – my friends all go round in
a big group talking, playing and laughing.’
In Mel’s (aged 12 years) statement below, she resists the category
through which she is being asked to speak:
Mel: ‘I am more than just adopted; I don’t feel like I am different
to everyone else just because I was adopted.’
These responses are part of a wider social construction of adoption,
and what is evident are the ways in which these young girls con-
struct their category memberships just as adults do. The girls here also
demonstrate the capacity to develop their identities by both estab-
lishing and resisting category entitlement. This occurs even within
adult/child power relations where parents and researchers assume
an emphasis on a category membership that the girls themselves do
not wish to give such prominence. As social actors, this negotiation
demonstrates a level of relational agency and autonomy that is rarely
attributed to children.
Conclusion
Our individual category memberships (both those we claim and
those attributed to us) shape who we are, who we can be and how
we perceive and are perceived in our social worlds. These mem-
berships influence our interactions and inform not only what we
think and say but what others anticipate we may think and say.
120 Ethical Research with Children
Our data cannot be extracted from this ‘complex web of intimate
and larger social relations’ (Mauthner and Doucet 2003:422). Our
‘scientific stories’ (Haraway 1991:106) therefore are not blank slates,
they are constructed through specific locations and interdependent
social relationships which in turn are informed by our category
memberships.
Such memberships extend far beyond fundamental and more obvi-
ous categories such as gender, sexuality or ethnicity. This can make
reflexive consideration of their influence a complex, rather messy
and, at times almost self-indulgent affair which may not reflect
the requirements of funders or institution (see Chapter 1). However
where we situate ourselves and our respondents, influences what we
ask, how we ask, and the responses we get. Categories establish our
entitlement to speak; they also shape what we can say and how our
voices will be interpreted. Voices thus become heard through the cat-
egories that we as individuals claim or that we have been attributed
to. Our voices are thus understood through the categories held or
claimed by those who listen. Acknowledging that establishing cate-
gory membership and constructing category entitlement in fieldwork
is taking place, it is a neglected and important ethical position for
each of us to take as we seek to analyse the ‘self-reflective accounts’
(Mauthner and Doucet 2003) we collect.
6
Privileging Voices
Introduction
It has already been established that the social studies of childhood
regard children’s voices as credible and worthy of being listened
to. Indeed, James and Prout (1997) go as far as to suggest that the
inclusion of their voices is a matter of need and right. As a conse-
quence, the drive to give children a voice, to make their voices heard
and understood is well documented in academic literature. How-
ever, Lewis (2008:14) describes the search for children’s voices as a
‘powerful moral crusade’, and she argues that since this quest has
gathered such a pace, it has persuaded researchers that innovative
means of gathering voices produce data that is authentic, realistic
and representative of all children. She goes on to suggest that its
assumed relationship with truth and experience has privileged the
concept of voice in qualitative enquiry. We agree with this sentiment
and also believe that this approach does little to engage critically
with methodological and epistemological limits of voice. Even as
self-confessed gatherers of children’s voices, we have come to ques-
tion the ethical positions of those who claim to represent others’
voices; who interpret personal accounts; and who sort and sieve data.
Through undertaking research of our own, we have come to under-
stand that it is within these very processes that conventional voices
and those that are the most easily heard and understood are privi-
leged, and those whose voices are harder to seek and to hear remain
marginalised.
The discussions in this chapter will illuminate the ways in which
some voices are heard and documented more than others. While we
121
122 Ethical Research with Children
draw predominantly on Allison’s research with disabled children and
young people, we consider that the key issues we raise relate to many
children whose voices remain unheard. We contend that children’s
influence in research, policies and service practice has been negli-
gible, and we argue that complex debates relating to competency,
age and maturity and the credibility of their statements are the main
causes for this. Some studies are adultist and thereby rely heavily on
data collected from parents and professionals. We therefore suggest
that marginalised children and young people’s voices tend to be over-
written by adults who talk at them, about them and on their behalf.
We suggest that as a consequence, the resulting silence is neither neu-
tral nor empty. It supports the ascribed collective identity of passivity,
vulnerability and dependency. This not only affects children’s social
status in general but continues to support the cultural schema and
collective understanding of childhoods.
Including children with unconventional voices often presents
researchers with significant challenges and facilitating their inclu-
sion requires a considerable investment in time, resources and the
development of new skills. In addition, it also calls for adaptations
to research designs and methodologies. However, the exclusions of
populations based on assumptions of vulnerability, incapacity or
incompetence is highly problematic. It is no longer acceptable to
exclude voices simply because they are considered incompetent or
pose challenges to traditional research methods. Therefore, recog-
nition of strength-based perceptions of children, who are regarded
as experts in their own lives (James and Prout 1997), as active citi-
zens with participation rights (UNCRC 1989) and within principles
of ethical symmetry (Christensen and Prout 2002) are central to the
discussions in this chapter.
Conventional meanings of ‘voice’ assume naturally produced
speech as a means of expression. Here a question arises: Are uncon-
ventional voices heard in the same way, or does having a different
voice implicate participation? The following discussions will explore
the concept of voice and examine the types of voices that we choose
to (or are able to) listen to. The ways in which some voices are
privileged will also be examined. We will use our collective nar-
ratives to illustrate how the taken-for-granted general disposition
and collective identity of children and childhood is influenced by
what Bourdieu terms the ‘doxa’ (1977:167) – the universe of the
undiscussed and undisputed.
Privileging Voices 123
This chapter sets out to examine the choices we make about the
voices we listen to, how we listen to them and why we consider some
to be more authentic than others. It builds resistance to what has
become the comfortable and accepted positioning and constitution
of ‘voice’.
Silences and inventions
Some research decisions affect whose stories get told and which
voices get used. Some voices inevitably get lost. While not disputing
that ‘listening to children’s voices’ should be embedded in and at the
forefront of ethical principles, we question whether liminal voices
are sufficiently captured for we believe that we naturally gravitate
towards those that are the most recognised and easy to hear.
Allison’s research with children who have little or no speech
enabled her to spend much time with the young participants, get-
ting to know them and them to know her as they worked and played
together in school. As the fieldwork progressed and they became
more familiar, the entries from her reflective journal illustrated the
pen pictures that she had begun to draw of each participant. The
memo’s she scribbled in the margins of the observations mostly
related to their cognitive ability and communication skills. On reflec-
tion, the entries also demonstrated that she had inadvertently started
to ‘grade individuals’ (Watson 2003:36), privileging some voices and
gravitating more towards the children and young people who were
easier to communicate with. This effectively attributed them with
higher status. For example, one of the young participants often
declined to communicate with her directly. He chose instead to look
away or hang his head. After several failed attempts, the conversa-
tions became less frequent and Allison interpreted his silence as an
inability to communicate. As a consequence, this submergence of
voice stigmatised him as ‘other’ and she ascribed his social status
as such. However, a research conversation with a member of staff
who had many years of experience of working with this young man
enlightened her:
If Jorge doesn’t want to communicate, he won’t communicate
with you either verbally, facially or using the machine. It’s just
a case of ‘I’m not talking today’. He has a lot of complex physical
needs, but if he could walk away sometimes, I’m sure he would.
124 Ethical Research with Children
Allison reflected on her assumptions. She concluded that not only
were her interpretations outrageously judgmental, they were quite
wrong. While Jorge could quite easily understand her and was per-
fectly capable of communicating with her, he had simply chosen
not to. Similarly, Paterson and Hughes (1999) concluded that peo-
ple with communication difficulties are often ignored and made
to feel inferior due to their inability to communicate according
to societally defined norms of accepted speech. With this now at
the forefront of her thinking, she realised that she should have
respected his silences as equally as his spoken narratives and paid
heed to the advice given by Agar (1980) who advocates that ethno-
graphers should realise that when a person withholds their opinion,
it does not necessarily mean that they do not have one. She felt
that she has also done what Crapanzano described as ‘marking
boundaries’ (1986:52) and used both space and silence to invent her
own construction of his identity, framing his dependency in what
Hockey and James refer to as ‘the metaphoric role of childhood’
(1993:10). This clearly reflects Bourdieu’s notion of habitus which
represents the social worlds of actors: ‘the relationship between
two capacities define the habitus, the capacity to produce classi-
fiable practices and worlds, and the capacity to differentiate and
appreciate those practices and products’ (1986a:17). Allison’s obser-
vations provided static definitions and based on this, she judged
that this young participant was unable to communicate, placing
him in a defining and deterministic category of ‘hard to reach’
and ‘unresponsive’. While Edwards and Imrie (2003) suggest that
reactions such as these may often be unthinking and spontaneous,
they warn that they can also be as a result of being unconsciously
learned.
Emerging from this scenario is what Bourdieu (1977:79) suggests
is unconscious behaviour, closely related to habit and habituation
whereby actions become taken for granted and undisputed. Jorge’s
practice of silence and his ability to ‘withhold access to his world’
(Mandell 1991:37) was clearly not appreciated. Because he had cho-
sen not to interact, he was a clean slate on which Allison was able to
write her own view of who he was and what he was able to do. The
misjudgement denied his competency and she perceived him inca-
pable of achieving specific outcomes. This, she reflected, could have
ultimately excluded him from the research project and his voice, like
Privileging Voices 125
many others before him, would have been lost. In addition, Jorge
may have experienced this abject identification on previous occa-
sions and, as a result, decided not to participate in any form of
further conversation. Indeed, as Gilligan (1993) warns, the relational
identities borne out of taken-for-granted attitudes can become a self-
fulfilling prophecy and responses often follow as a matter of course.
Clearly, this young man was prepared to allow the ventriloquised rep-
resentation of him to remain unchallenged and, as a consequence,
Allison’s (mis)interpretation assigned him to a pre-ordained habitus,
categorising him by impairment. Jessica’s experience of supervising
a student through a series of ethical approval processes highlights
that it is not just disabled children and young people who must deal
with being identified or labelled as incompetent or unable to par-
ticipate. The language of assent discussed in Chapter 1 assumes a
level of incompetency of all children whereby they are perceived as
being unable to understand the requirements of social interactions or
research relationships. These constructions serve to position them as
structurally vulnerable. However, in Allison’s example of Jorge, dis-
cussed above, there is also evidence of this young person’s ability to
negotiate competently adults’ expectations of him. Jorge uses silence
here as a demonstration of his power, it is a mechanism of resistance.
The staff in the setting state he would walk away if he could, in Jorge’s
case he cannot but he can and does remove his voice. This demon-
strates use of silence in a different way, as resistance and if we don’t
realise how children (and adults) use their voices in these ways then
we are not really listening.
The general dispositions of disabled children and young people are
often based on the overriding ascribed collective identity, ‘disabled’
or to put it another way, a taken-for-granted collective identity which
is often understood as a consequence of medical classification. As lan-
guage is a central aspect of the study of identity that affects and
dominates identity construction (Strauss 1977), Allison had based
this young person’s behaviour on stereotypical criteria associate with
‘disability’ and categorised him as passive and dependent. While
projecting negative assumptions such as these, she had denied dif-
ference, diversity and voice. She had privileged some participant’s
voices and subjugated others and done just as Fawcett (2000) sug-
gests, accepted social constructions and categories that tend to reflect
dominant and social groupings.
126 Ethical Research with Children
Further familiarisation visits however, helped Allison to gain a
greater understanding of Jorge’s personalised method of communi-
cation. Spending more time with him in the classroom enabled her
to get to know him better and as they built up mutual trust, he chose
to become more communicative. While critical self-reflections such
as these enabled her to form a nuanced understanding of the gen-
eral dispositions of disabled children and young people, they also
enabled her to move beyond the general to the specific and revise
her understanding of the diversities of individual childhoods.
The scenario has been used to illustrate the ways in which dis-
abled children and young people’s identities are based on accepted
perceptions and cultural schemas of disability, all of which are,
using Bourdieu’s terms of reference, embedded in ‘methodologies for
adapting generalised knowledge to a range of particular practices in
particular locations in time and space’ (1977:78–79). However, we
were curious to know how this process privileges some voices over
others. We concluded that if some children and young people are
not able to communicate in conventional ways, they may feel dis-
advantaged in their ability to contribute to research conversations.
They may also feel that their contributions are devalued. In addition,
and as the example suggests, the severely limited feedback from Jorge
allowed space for invention free from correction. Clearly, silences can
easily be filled by others who speak on behalf of those whose absence
of conventional modes of movement, speech or verbal communica-
tion renders them ‘speech-less’. Indeed, as Edwards and Imrie (2003)
suggest, some voices then simply disappear. Acts that constitute rela-
tions of hierarchy and power often lead to withdrawal of voice and
without recourse, silence ensues. Arguably, within this silence, iden-
tities can be misunderstood, misinterpreted or simply invented by
those whose voices are louder, more powerful. As a consequence,
some voices are privileged and some are silenced.
Influences that position children’s voices
While there has been an increase in policy priority to include chil-
dren and young people in decision making in general, Franklin and
Sloper (2009) found that very little is known about specific factors
that promote disabled children’s participation. Arguably, without
providing opportunities, platforms for voice or providing optimal
Privileging Voices 127
conditions that make the recursive activity of listening to children’s
voices a reality, limited examples of empirical research, especially
with those who have learning difficulties or severe communica-
tion impairments, will continue and external structures that mark
disabled children as vulnerable and incompetent will persist.
Images of dependency
Shakespeare (2000) contends that while disability continues to be
represented by images of dependency, the dominant and positive
imagery that surrounds helpers in our society portrays workers in the
caring profession as highly valued and respected. He goes on to sug-
gest that the establishment of expertise and power of helpers serves
not only to privilege their voices, but also to ensure their superior
status whereas the service users’ voices ‘remain silenced, objectified
and othered, in a discourse where they exist only to have things
done to them or for them’ (Shakespeare 2000:21). He went on to
pursue this point further by examining pre-existing ideas of the cul-
tural representation of disability and dependency. He argued that
‘normal’ bodies are legitimised and those with ‘in-valid’ or ‘deviant’
bodies are seen to be inferior or in need of help. While we believe
that the concept of representation can clearly be applied to chil-
dren with unconventional voices, once again we draw on Allison’s
empirical research to help justify this claim. She observed that some
of the young participants needed help to access their Augmentative
and Alternative Communication Systems (AACS). Her field notes
highlighted that the inaccessibility of some AACS limited the oppor-
tunities of use and this, she believed, excluded their voices. For
example, one young participant’s device was kept in a ‘rucksack on
the handlebars of her wheelchair’. She could not access the bag that
contained it without help. In addition, she could not open the bag
and take the AACS out without support and she also needed help
to set her AACS on the tray of her wheelchair. Her use of it was
totally governed by her ‘able-bodied’ adult helper. A conversation
with another young participant, Cherise revealed that she had to rely
on adults help to set the device up for her as she did not have fine
motor skills strong enough to undo the Velcro fastenings of the bag
that housed it. She also had to ask for help if the battery needed
recharging. She explained that as a wheelchair user, she could not
access the electric plug sockets that were positioned at floor level
128 Ethical Research with Children
in most classrooms. Allison concluded that both participants were
clearly disabled by their environment and there were many barriers
that served to exclude their voices. This example clearly strengthens
our argument: voices that are easily heard and within reach typically
become privileged. The stories told here support the argument for a
less polarised, more interactional approach towards including chil-
dren in research. It also suggests that without including marginalised
voices, we will continue to assume the collective identity of children
and therefore the homogenisation of childhood.
Borrowed voices
The majority of the children in Allison’s study relied on others to
programme their communication aids, so if any new vocabulary was
needed, the staff would take the device away for a few hours to input
specific words. One of the young participant’s readily demonstrated
the comprehensive dash-board bank of pre-programmed words and
phrases that she had on her AACS that she used to communicate.
She also told Allison that she ‘often typed her words herself’. Allison
observed that most of the pre-programmed vocabulary on the young
people’s AACS’s related to curriculum subjects and she felt that this
impacted on the study, for the research conversations relied heav-
ily on words lent to them by adults. On reflection, this made the
authenticity of voice questionable. Issues of access to and main-
tenance of the AACS’s, together with the limited choice of words
available to the participants clearly influenced the young partici-
pants’ voices. Just as in Sarah and Jess’s research, Allison concluded
that the voices of the young people were very much governed by the
adult practices and attitudes. Indeed, this demonstrates that those
with ‘normal’ or ‘conventional’ voices were legitimised and those
with ‘in-valid’ or ‘deviant’ voices were seen as inferior and in need
of help. As a consequence, the children and young people with lit-
tle or no speech became reliant on others with conventional speech
to provide them access to a voice or alternatively, to speak on their
behalf.
The symbolic representation of voice
Those with very little speech and others literally without voice are
often offered prosthetics in the form of high-tech communication
Privileging Voices 129
aids in order to secure the right to speak for themselves and make
decisions and choices about their lived experiences. While on the
one hand, such prosthetics can be seen as empowering and a means
of giving them a voice, on the other, they can be seen to act as sys-
tems of symbolic representation that lead to oppressive practices.
Indeed, in Allison’s study, the analysis of the data revealed that in
some instances, the children’s specific position practices in terms of
their levels of cognitive and physical ability were represented by the
specific type of AACS they used. During a research conversation, one
of the members of the teaching staff told Allison about a young AACS
user who had recently been given a new aid:
He’s recently got a new one, um, one with much higher specs
than he needs and we were interested to see if he would take it
on board intellectually. He had new pages he needed to remember
and recall. He has a lot more pages in this machine and he has sur-
prised us ‘cos he is gaining knowledge all the time which is really
good.
During another conversation, a member of staff compared one user
with another:
Stuart can be very able on his machine but he’s not as at home
with it as Jack, he needs much more guidance as to what page
he will find something on . . . if you can’t find it, Stuart can’t find
it . . . you need to push him.
On more than one occasion, the data illustrated that adults’ percep-
tions of the disabled children and young people’s position practices
were interpreted and based on their use of high-tech communica-
tion aids. The AACS represented the symbolic value of the children’s
cognitive foundations of status which, in turn, created hierarchies
of social location within the classroom. This indicated that some of
the staff believed that the type of aid allocated to each participant
reflected their identity. For example, the more functions the aid had,
the higher the user’s cognitive ability was assumed to be:
When someone is cognitively more able, they [the aids] are
fantastic and they can use them really, really well.
130 Ethical Research with Children
The smaller the aid, the less physically difficulties the user has:
They are small, that’s the problem for the variety and volume of
vocabulary the buttons get smaller and smaller and smaller, um,
so those with more profound physical difficulties or those with
visual impairments can’t use the more powerful machines
And, finally, the newer and more personalised the aid, the less likely
the user was dependent on external funding:
That new system is about £4,000 if it can connect onto the original
‘tell us’ which is what she’s got or £11,000 for the new ‘tell me’ –
Lizzie has a trust fund that might fund certain things for her and so
her mum has to go and ask the trust if they would like to resource
it – if not, its fairly unlikely she will get it.
The conversations with the staff about the AACS used by each partici-
pant not only categorised them but were based on what Cunningham
explains as ‘the basic stock of knowledge that people carry around in
their heads as a result of living in particular cultures or subcultures’
(1993:1). Identifying the participating disabled children and young
people, some AACS determined the users’ cognitive ability, others
gauged the users’ physical impairment and others calculated their
financial status. As Bourdieu (1986b) explained, social identity tends
to be defined and interpreted through difference. He also argued that
groups attribute cultural capital to particular behaviours and they
vie to impose their definition of which social phenomena consti-
tute legitimate behaviour on others. The above discussions suggest
that the normative expectations and social identities of the young
participants were interpreted from fixed unyielding notions based on
past relationships or from knowledge inherited from others and the
justification of the definitions provided a rationale for behaviour.
Such assumptions based on prior knowledge are not restricted to
Allison’s respondents but occurs throughout childhood. In Sarah’s
research it became apparent that ideas about what an orphan is, or
should be, informed literature, social work practice and the kinship
work done by families. At various points in the research Sarah found
some of the following ideas articulated by RECS, social work profes-
sionals, adoptive parents and in academic literature. Such knowledge,
Privileging Voices 131
accurate or otherwise is therefore productive. As Sarah writes in her
thesis:
Orphans are claimed to play a pivotal role not only in social wel-
fare provision but in mythology too (Ennew 2005). Orphaned girls
in particular, have been a recurring feature of fiction. As characters
they are commonly depicted as protagonists (Friedlander 2003) yet
vulnerable, loveable objects of charity, the ‘tropes of victim hood’
(Anzil 2013; Leinaweaver 2008:71). As victims, they are also avail-
able to be rescued and hold the key to transforming the lives of
those who do the rescuing (Walkerdine 1999:15). In classic and
contemporary literature, the orphan is often depicted as someone
with a blank canvas upon which identity traits can be inscribed.
Collier (2006) argues that through assimilation and appropriation
an orphan can be ascribed an identity in the image of her new
society and parents, suggesting that a form of colonisation takes
place. Walkerdine (1999:15) provides some fictional examples of
this perspective, including the story of Annie, who is depicted
as a working class, homeless orphan. Despite this destitute and
impoverished beginning, she is ‘a figure of immense transforma-
tive power’. Her fictional role is to use an endearing personality to
soften the heart of the adults around her and in so doing, secure
a place in a wealthy family. Saunders (2008:1553) argues that a
‘defining hallmark’ of such literature is the presence of sentimen-
tality, where sympathy for the plight of the character is evident
and required from the audience.
Despite Ennew’s (2005) argument that there is no global defini-
tion for ‘orphan’, the term does represent extensively held and
at times contradictory ideas of the becoming child, of powerless-
ness, of victim and also, conversely, of being lucky (Emond 2009).
The absence, or inclusion of a partial narrative of belonging which
each of these children’s orphaned status represents, lacks norma-
tive connections through family, biography and culture and is
considered to be socially problematic. In addition Moeller (1998)
argues that the image of orphan is a seductive one and, as with
other representations, is effective in undermining the realities of
the lives of these children. The role of orphan performs a nec-
essary identity role in the social construction of childhood. This
image is argued to evoke pity, sympathy and support (Holland
132 Ethical Research with Children
1992:148). Such a role depicts an ‘other’ childhood separate to
normative childhood which is identified as being embedded in a
biological family. Connolly (2000:174) argues that all children are
normatively socialised into ‘pre-existing racial and gender iden-
tities that have been set for them’. Collier (2006) suggests that
orphaned children are compelled to play a specific identity role
in their adoptive families, offering adoptive parents an oppor-
tunity to perform a nurturing protective and loving role. The
adopted child then allows the parent to perform and display a
version of the family. The orphan’s membership in this family is
argued to be contingent upon them being willing to comply with
the associated obligation and responsibilities and maintain the
social order (Bluebond-Langner 1978). However, I contend that
all children provide parents similar opportunities and their place
in the family carries similar responsibilities. Such opportunities
and responsibilities are common to both adopted and biological
children.
The orphan as a category upon which identity characteristics
may be ascribed has been explored in previous research (see
Dorow 2006). Gailey (2000:303) has previously claimed that racial
stereotypes of Asian women being ‘compliant and docile’ inform
adopters motives for adoption from China. Such is the linguis-
tic power of terms like orphan and abandoned, Chinese orphan
girls, prior to their adoption already have identity characteris-
tics inscribed on them. Some are unique and specific to girls
but other traits are assumed to be part of the adoptee’s identity
through other equally iconic linguistic imagery such as neglect,
deprivation and foreign orphanage. Such labels create images
which Fraser et al. (2004:39) argue reflect the power of adults
to place individual children into depersonalising categories ‘that
spotlight membership of abnormal categories ahead of member-
ship of childhood’. The language also conveys particular attitudes
of behaviour towards certain groups of children (ibid.). A variety
of attributes have been linked to an adoptee identity (Bartholet
1999; Jacobson 2008). These are often informed by psychological
research. An ‘adopted child syndrome’ (Kirschner 1990:3) identi-
fies adoptee traits which include attachment disorder, learning dis-
abilities, stealing, lying, violent behaviour, and even setting fires
(Jacobson 2008). These attributes are argued to be more prevalent
Privileging Voices 133
in transracial adoptees (Bartholet 1999). Whilst Kirschner’s claim
about a syndrome has been robustly challenged (see Friedlander
2003), the notion that adopted children may have particular dis-
orders by virtue of their orphan/adoptee status is taken as self
evident. Examples of parents and a professional social worker iden-
tifying such traits prior and post adoption are also found in my
research. ‘Myths and stereotypes’ about adoption are prevalent in
popular culture (Friedlander 2003:747) and shape how adoption is
framed.
Yet, just as Lizzie demonstrates, a perspective which suggests that
children are simply the products of such ascription is insufficient.
Children can and do play an active role in shaping their identi-
ties (Wall 2010), and once recognised, the social context where they
negotiate their subject positions must be accounted for (Connolly
2000). One of the most common ways that the girls in Sarah’s
research resisted such categories was to make statements such as the
following:
‘I don’t feel any different just because I was adopted.’
...
‘I am more than just adopted.’
Despite the power in these statements and evident attempts to move
beyond the category of adoption or adopted, the voices of actual chil-
dren can be muted by the identity traits placed on their shoulders
which they are to a certain extent compelled to perform within their
social specific contexts.
In relation to the systems in situ that categorise individuals,
Giddens suggests that bodily appearances relating to ‘all those fea-
tures of the surface of the body including modes of dress and
adornment, which are visible to the individual and to other agents,
and which are ordinarily used as cues to interpret actions’ (1991a:99)
have relevance to the trajectory of self, particularly in relation to
the ways in which bodies reflect social positioning. Clearly, the
extracts taken from Allison’s data demonstrate how the adults used
voice prosthetics as symbolic representations of disabled children
and young people’s identities. Indeed, Goffman refers to ‘props’ or
134 Ethical Research with Children
‘sign-equipment’ (1959:39) as representations of social positioning
and suggests that they can be used to reaffirm expressive bias which
eventually become accepted as reality. However, Allison concluded
that where AACS were used to categorise and identify disabled chil-
dren and young people, it not only constrained specific aspects of
the self but also governed responses of individuals, situating them in
social conventions that assume vulnerability and dependency.
The discussions demonstrate how social identities of both ‘dif-
ference’ and ‘similarity’ are used to categorise children and young
people, grouping them together homogenously according to the
central values of society. However, while this is useful in terms of
understanding their general dispositions, it is vital that their own
views are included, for we argue that the general habitus of children
and childhood does not reconcile other dimensions of their iden-
tity, nor does it account for the ways in which they have forged their
own cultures through acts of resistance. It does, however explain why
some voices are marginalised and others are privileged.
Powerful voices
While social practices and the recursiveness of social life tend to
settle into what Bourdieu (1986a) terms as habitus which repre-
sents the social worlds of actors, we would argue that membership
of social worlds and everyday institutional situations is not wholly
determined. Uncertainties of membership relate to intricacies of the
location and dislocation of multiple identities and disregards indi-
vidual agency. Arguably, language also regulates human interaction
and positions people socially. As ‘all human interaction involves the
communication of meaning, the operation of power, and modes of
normative sanctioning’ (Giddens 1991b:46–47), it follows then that
actors draw upon specific rules embedded in language to reproduce
corresponding structural properties of specific social systems.
The following conversations taken from Allison’s study illustrate
the ways in which memberships are bounded by institutional orders.
For example, a member of staff who had known some of the Sixth
Form pupils since they joined the school Nursery, described one of
the young participants as ‘backward and very passive’ and another as
‘all singing and dancing’. Another member of staff referred to one of
the young participants as ‘less able’ and another as a ‘very complex
Privileging Voices 135
case’. The point raised here highlights the ways in which voices and
competencies are constructed through language. It is also useful to
note Bourdieu’s (1977) expansion of the notion of habitus to speak-
ing. He refers to the use of symbolic power embedded in the language
used by dominated speakers in order to encompass the sense of value
to what one has to say in a particular situation. The examples above
demonstrated the ways in which the competencies of children were
clearly suspended and shaped by dominant adult’s voices. As a result,
it was adults who made decisions on behalf of the children. The dis-
abled children and young people became what Christensen (1999)
termed as muted actors and, by design, the ‘helpers’ became more
powerful and the ‘helpless’ more needy, less autonomous and less
capable.
At this point, it is useful to consider the factors that stabilise
expected everyday social functions that are embedded within the
examples given in this chapter. We need to critically examine how
these influence the participants’ voices. The examples authorise prop-
erties of the social practice of communication in school whereby
adult’s voices are privileged. Adults talk, children listen, joining in
only when invited, following conventions of social exchange. The
children are clearly reluctant to try out the suggested new practices
that encourage destabilisation of the ingrained social practices of
said membership. While Giddens (1979) believes in actors’ compe-
tent knowledgeability of their everyday world and recognises that
reflexive monitoring of action can reconstitute the (re)organisation
of society, an interesting lacuna in this analysis of reflexivity is high-
lighted. On this occasion, the young participants chose not to initiate
conversations or indeed use their discursive consciousness to alter
their general dispositions. While this was puzzling, it was not sur-
prising because as has already established, the disabled children and
young people’s lives were controlled daily by adults in that they were
told where, when and how to use their AACS’s. Even the words avail-
able to them were programmed for them by adults. Borrowing the
phrase used by Giddens ‘power is always exercised over someone’
(1979:341), this suggests that the membership of competency has
been ascribed for so long by adults that the set of normative rules
of conduct for disabled children are deeply engrained in their daily
behaviour. As a consequence, the disabled children’s voices have been
subordinated to a point where they have become passive and quiet.
136 Ethical Research with Children
Invitations to give voice are unexpected and the call for voice raises
anxiety. Indeed, if disabled children and young people are to be
empowered to gain competence and confidence, it must be worked
at discursively, socially and politically.
‘Seeing things differently’: Privileging children’s voices
The discussions so far have focused on the ways in which attitudes
and practices influence the general disposition of children and young
people. Placing children and young people in hierarchies accord-
ing to what they can and cannot do effectively privileges some
voices over others and the voices that are hard to hear often become
silenced. However, in order to gain a more holistic understanding
of children’s lives and lived childhoods, we believe that their stories
should be prioritised and their competence as social actors recog-
nised. The contours and shapes of children’s voices are therefore now
needed to contextualise the discussions further and so it is time to
move the discussion on and include the voices that are hard to hear.
An example of the ways in which a young person with little or no nat-
ural speech made himself understood becomes clear in the retelling
of one incident that occurred during one of Allison’s familiarisation
visits. She joined a French class where the teacher was telling the
class that their task was to translate a letter from Bruno, a French
boy, and compose a reply. As the teacher spoke, Leon yawned loudly
and used his AACS to type ‘boring’. He glanced at Allison, and sur-
reptitiously tilted the aid so that only she could read it. The aid was
switched to silent mode. She gestured that she had read the message
by raising her eyebrows. The teacher looked over and asked Leon
if everything was alright. He smiled and confirmed that it was and
quickly erased the message. When the teacher started to talk again,
he typed another message: ‘we did this before . . . ’ and yawned loudly
once more. Again, he erased the message quickly while the teacher
glanced at him, frowning. Once assigned the task, the young man
engaged fully on it until his classmate nudged him and then pur-
posely pushed his bag off the table and onto the floor. Leon was
not able to retrieve the bag without help. He was obviously agi-
tated by the incident and used his AACS to tell his neighbour to
‘Piss Off!’. This time he turned the volume up and so the whole
class heard what he said. The teacher ignored the fracas but walked
Privileging Voices 137
behind him and picked up the bag. She put it on the table told him
to erase the message and indicated that both should get on with their
work and stop squabbling. While the dynamics of interactions at
work here are undoubtedly complex and open to interpretation, the
above observation can be usefully employed to analyse the notion
that disability should not be based purely on others’ knowledge and
opinion. While Allison found that the majority of the staff were
concerned that the disabled children and young people were pas-
sive in terms of initiating conversations and expressing their needs
and desires, the above scenario clearly contradicts this. While we
consider that multi-voiced accounts are important to gain a bal-
anced view of childhoods, we argue that the stories of children and
young people tell themselves about their lives and lived experiences
should take centre-stage. Each voice should be offered equal sound-
scape. Clearly, given an appropriate platform, children and young
people can demonstrate active agency through voice and action for
quite simply, we believe that the recognition of social competency
and agency of those who have unconventional voices is ethically
vital.
Reflexive re-positioning
The increased emphasis on rights has challenged researchers to
recognise the importance of the inclusion of children’s own views.
Therefore, in order to gain a perceptual grasp of the complex influ-
ences of social contexts on the young people’s lives, understandings
of childhoods must be anchored in their own accounts. Without
including stories told by the children themselves, we believe that
narratives of passivity and incompetence will remain unchallenged.
The following extract from Allison’s field notes supports this shift in
recognition as it illustrates the ways in which a young participant
resisted the ascribed position of incompetent by creatively drawing
on specific resources available to her. A research conversation with
this young person helped Allison understand some specific aspects
of her expertise. As she competently demonstrated how she signed,
the ways she used her writing board, her high-tech communication
aid and her voice, she clearly illustrated that not only did she have
a vast range of communication techniques but also that she was
able to adopt them accordingly: depending on how she felt and who
138 Ethical Research with Children
she was communicating with. She clearly enumerated her likes, her
needs and her beliefs and she could express describe herself through
actions. Her confidence was impressive and Allison commented on
her expertise, to which she replied modestly: ‘I’m not an expert, but
my friend is. I’m only a little expert in comparison!’ Her sense of self
as a knowledgeable agent is illustrative of Giddens’ (1991a) concept
of reflexivity. She demonstrated her ability to read cultural life and its
textured flow of social interaction in a way that her actions and expla-
nations were constitutive of what she did and how she did it. Her
concept of self as a knowledgeable agent links to Giddens’ (1991a)
work on self-identity and also to Goffman’s (1959) suggestion that
knowledgeability is tied up with thousands of other interpersonal
settings and through the ways in which day-to-day life is organised.
As discussed earlier, disabled children and young people are often
positioned as incompetent and dependent but we believe that those
with little or no voice have plenty to tell us. If appropriate and equal
opportunities to speak are offered, a whole range of voices will be
heard.
Mead’s (1974) broad interpretation of the social nature of the con-
struction of self whereby each of us fashions our sense of self through
engagement with others can be usefully applied here. The narra-
tives within this chapter clearly demonstrate that children routinely
(re)construct themselves through their everyday experiences. Indeed,
as Elliott (2008) argues, to possess ‘self’ implies an ability to take
one’s own actions, emotions and beliefs as a unified structure as well
as acknowledging the perspective of the self from significant oth-
ers. While some of the previous discussions have signified systems
of identification as oppressive and limiting, the latter discussions
demonstrate the ways in which situated agents reflexively engage
with their contexts to build their sense of self. This demonstrates that
each of the young participants carry around stores of unique knowl-
edge of both external and internal structures in their heads. Neither is
prioritised but both are used reflexively within specific social contexts
to shape their position practice. Allison’s recollection of a conversa-
tion she had with one young participant, Lizzie might be useful to
further our understanding of this.
While negotiating schedules for carrying out semi-structured inter-
views with some of the young participants, a member of staff warned
Allison that one particular girl was uncommunicative and likely to
Privileging Voices 139
‘kick off at the slightest upset’. Indeed, during the familiarisation
visits, Lizzie chose neither to look at Allison nor communicate
directly with her. However, having established she was happy to take
part in the interview, the teacher indicated that we might find it eas-
ier to concentrate if they went out of the noisy classroom and into the
corridor. Allison was reminded of Lizzie’s response signals before they
left the classroom: a mixture of facial expressions, body movement
and pre-recorded responses on the AACS would be used to participate
in the conversations. Once they were settled, Allison started off the
conversation and confirmed that Lizzie was still willing to participate
in the research:
Allison: ‘Hi Lizzie, are you happy to talk to me today?’
Lizzie: [moves her head to the left which activates the head switch,
‘yes’]
Allison: ‘Thank you. Is it ok if I use my recorder?’
Lizzie: [moves her head to the left]
Allison: ‘If at any time you want to stop talking to me, you can tell
me to stop, or signal me to stop – is that ok?’
Lizzie: [pushes her head impatiently towards the head switch]
‘Questions.’ [pre-recorded message]
Allison: ‘OK, time for questions!’
Lizzie: ‘Where are we going?’ [pre-recorded message]
Allison: ‘Well, if its ok with you, we are staying here in the corridor
next to your classroom, is that ok with you?’
Lizzie: [inclines her head to the left; this means ‘yes’]
Allison: ‘Yeah, we’re just parked up here to stay in this quiet
corridor so we can talk, ok?’
Lizzie: [inclines her head to the left]
Allison: ‘Great . . . Lizzie, what do you . . . ’
Lizzie: [interrupts me with a pre-recorded message] ‘Where do you
live?’
Allison: ‘Where do I live? I live in . . . ’
Lizzie: [pause] ‘By the sea?’ [pre-recorded message]
Allison: ‘Yeah, by the sea. Have you ever been there?’
Lizzie: [inclines her head to the left]
Allison: ‘What did you think of it?’
Lizzie: [activates the head switch and scans the screen for her
choice of reply . . . I wait . . . ‘dump’]
140 Ethical Research with Children
Allison: ‘Oh! Right!’ [I laugh] ‘Well, its ok when the sun shines but
I agree it is a bit miserable when it’s raining! I wait a while and
then start the question again. Lizzie, what do you . . . ’
Lizzie: [she interrupts again with a pre-recorded response] ‘How
old are you?’
Allison: [I chuckle and answer] ‘’45 but don’t you dare tell anyone!’
Lizzie: [smiles and wriggles around a bit in her wheelchair]
Allison: ‘All these questions, Lizzie – I’m supposed to be the one
asking you questions!!!!’
Lizzie: [smiles and turns her head towards me]
The above extract demonstrates how the capability for autonomous
thought and reflexivity not only permits some sort of ‘regrooving’ of
the self in the broader context, but also emphasises the importance
of examining positional identities from a variety of perspectives.
Indeed, had this discussion relied merely on the interpretation of
the observations undertaken and on the adult-led conversations in
relation to this particular young person, Allison might not have inter-
viewed her and a great deal of rich description in relation to her
identity and agency would have remained undiscovered. As a result,
conventional voices, or those that were easier to listen to would
have been privileged and unconventional voices, or those that are
require different listening skills, would have been marginalised or
silenced.
Allison concluded that on the whole, while the observations and
the interviews with adults were worthwhile (see in addition discus-
sions of gatekeepers in Chapter 2) for gaining background informa-
tion about the disabled children and young people and allowed her
to build up a picture of the ways in which they used high-tech com-
munication aids on a daily basis, without the narratives of the young
participants themselves, opportunities to gain a more holistic under-
standing of their lives and an insight into their individual characters
would have been lost.
Routes to hearing some voices, as Allison’s research indicates, can
be challenging to navigate and require a willingness and capacity to
find paths through power relations embedded in the social contexts
through which voices emerge. The power to speak, or remain silent
are part of the socially embedded processes by which some voices are
listened to, afforded greater legitimacy than others or indeed have
Privileging Voices 141
the privilege to be heard at all. This does not just apply to the voices
of participants but also, crucially, to the voices of researchers, notably
those seeking to explore the construction of knowledge in contested
arenas as we explore below.
Hegemony of voice: The process of privileging
In arenas where Jessica has presented her work on sexualisation
and childhood, where what is advocated is a less sensationalist and
‘risk-based’ approach to children’s worlds, she has been approached
afterwards by fellow academics who have dubbed her brave for
embarking on discussions of such controversial topics. In addition,
Jessica’s workshops for practitioners on considering the positive ele-
ments of young people’s sexual lives were highlighted by attendees
as the only offering they had yet to experience as part of their profes-
sional development which did not focus on unplanned pregnancies,
STIs and sexual exploitation and wasn’t that a brave approach to take.
As Goode’s experiences have indicated (Channel 4 Press 2014), advo-
cating alternative ways of thinking that depart from the dominant
conversation in any given arena also have potentially very difficult
consequences (see Chapter 1). What can be seen from Goode’s expe-
riences researching with self-defined ‘minor attracted adults’ (2010)
is what we have dubbed a hegemony of voice. Her attempt, like that
of Plummer (Gilligan 2014), to critically evaluate public perceptions
of paedophiles as other than a dangerous, criminal perversion and
as a part of the wider human condition, generated significant pub-
lic outrage with an assumption that this somehow advocated sexual
attraction to children. This attempt to speak in ways that extended
the debate beyond this single hegemonic position led in Goode’s case
to disastrous consequences for her academic career. What this high-
lights is the potential ramifications for the construction of knowledge
of the questioning of this canonical narrative (Bruner 2004).
Sarah found that her voice was limited in various ways depending
on the context. In her data collection, remaining silent in the inter-
views was, on occasion, difficult as respondents sought to draw her
in to conversation about her own adoption experiences. A further
dynamic that Sarah experienced was that discussing her research in
other arenas was not always easy. In a review of adoption and race
literature, Kirton (1996, see also Högbacka 2013) remarks about the
142 Ethical Research with Children
lack of disclosure on the part of researchers to identify themselves as
intercountry or transracial adopters.
Rastas (2005 in Yngvesson 2010:33) claims that it remains prob-
lematic for intercountry adoptive parents to discuss their motivations
to adopt as such decisions are invariably perceived as racist. Gaber
and Aldridge (1994:1) provide further explanations of their silence
by claiming that any proponent of transracial adoption runs the risk
of being castigated as racist, which effectively makes ‘further argu-
ment unnecessary’. It is therefore unsurprising that many adopters
remain cautious about revealing or discussing their families.
Sarah’s personal position in her study emerged as being an impor-
tant epistemological aspect, and therefore it was necessary and
important to identify her proximity to the research subject. However,
she recognised why others may feel reticent to do so and experienced
trepidation in identifying this personal involvement. As part of her
research Sarah attended a number of conferences, which discussed
the issues surrounding transracial and intercountry adoption. Despite
numerous opportunities, she never felt able to disclose her status as
an intercountry adopter at these conferences or contribute in ways
that would disclose this status. In part, this was due to the negative
connotations attached to intercountry adoption and intercountry
adopters, but she also felt that any contribution she might make
would then only be perceived through her own mothering choices.
Her voice in this arena was silenced, to some extent a self-induced
silence, in part by the dominance of ideas which argue that par-
ents of transracially adopted child are insufficiently equipped to
assist children manage their experiences of racism in an ‘inherently
racist’ world (Barn 2000; Triseliotis et al., 1997:164). As intercountry
adopters of a transracially adopted children, parents like Sarah are
openly spoken about but far less often heard in this regard. In conse-
quence the activities that these families do engage in to support their
children in this regard remain for the most part unheard which is to
the deficit of a wider debate about how to support transracially and
intercountry adopted children.
Approaches to intercountry adoption, and indeed many other top-
ics, are shaped by dominant discourses whereby particular knowledge
and positions are privileged, and these positions become canoni-
cal narratives (Bruner 2004). These narratives shape the way a topic
can be discussed, written and thought about, determining what is
Privileging Voices 143
acceptable to say, it thus becomes the moral order. Part of the silenc-
ing of voices in some debates can also be linked to the category
entitlement or lack of entitlement of some to be seen to legitimately
contribute to the debate.
Phoenix (2008:66) uses category entitlement to analyse the ways
in which one of her respondents ‘a white mother of adult children of
mixed parentage’ establishes her entitlement to speak of her knowl-
edge and experiences of racism to Phoenix as a ‘black woman, in a
study of social identities’ (ibid.). Phoenix describes how this respon-
dent uses stories to position herself as knowledgeable of racism and
presents herself as being against racism. This perspective is further
emphasised by the stories of how she parents her children. The sto-
ries the respondent tells construct her as learning to stand up against
racism even in the face of injury or harm. Phoenix describes these sto-
ries as moral and part of how the respondent constructs herself as a
‘hero’, holding a world view of ‘racism as illegitimate’ (2008:69). The
respondent uses her stories to position herself as holding legitimate
authority to discuss racism.
Perhaps it is the absence of the opportunity to establish such an
entitlement to speak that mutes voices, either the absence of related
experience or the opportunity to share these experiences because
of how the canonical narratives are constructed. Hydén (2008:125)
argues that ‘without having experienced the cognitive process of
‘having been through’ an experience so that ‘you know about it’, it is
difficult to discuss the experience.’
In the case of research in the field of disability studies some schol-
ars argue that disability discourses are still dominated by people who
are not disabled (Kitchin 2000). Oliver (1992) finds this issue sig-
nificantly problematic and concurs with Hydén in arguing that this
traditional ‘expert’ model of research alienates, disempowers and dis-
enfranchises disabled research participants. The interpretation of the
knowledge and experience of disabled individuals by ‘non-disabled’
researchers is ‘compounding the oppression of disabled respondents’
(Kitchin 2000:26). A comparison could be made here to the voices
of children in the research process which come to subsequently
be collected, interpreted and thus dominated by adult researchers
(Coad and Evans 2008). Increasing interest in child-centred method-
ologies has led to the advocating of children as peer researchers
involved in all aspects of the research process from design to data
144 Ethical Research with Children
analysis. Such approaches attempt to address the imbalance of power
relations in research with children (Christensen and James 2000)
by attempting to give children greater degrees of control over the
research process (Thomas and O’Kane 1998). However, notions of
empowerment and of power itself are far more slippery (Foucault
1995) and nebulous than these approaches suggest. Power is nego-
tiated and co-constructed in social interactions rather than existing
as a static and structural phenomenon. Therefore, while we do not
hold to Hydén’s and Oliver’s arguments that such experiential knowl-
edge is necessary before empathy and understanding can occur, we
do concur with Phoenix (2008) in that our own particular attributes
and experiences can award category entitlement which legitimises
not only the authority to speak, but also the power to co-construct
the narratives told.
What can be seen from the examples above is an active demon-
stration of the hegemony of voice whereby particular positions and
arguments privilege or marginalise both what is being said and by
who. Just as researchers are often attributed the position of ‘expert’
on a topic (Danieli and Woodhams 2005) by the very virtue of their
status as questioner the positions they simultaneously occupy as
mother, woman, professional practitioner to name but a few can
render them vulnerable. This vulnerability results in the potential
of having our voices labelled as flawed because we reveal ourselves to
not just (or at all) being experts but rather subjects who are part of the
co-construction of knowledge and either don’t display the objectivity
still often desired within social science research, having views which
are shaped by more than just the label of researcher or by deeming
to speak in ways that are at odds with dominant knowledge in our
fields.
Conclusion
The discussions within this chapter have clearly demonstrated the
ways in which young participants’ voices are embedded in exter-
nal structures of power, difference, entitlement and social affiliation.
While acknowledging the importance in recognising the rights of
children and empowering them to express their views in matters that
concern them, we have argued that is it equally important that we
value and include all voices, not just those that are easy to hear and
Privileging Voices 145
interpret. The scenarios have illustrated that listening is a skill neces-
sary for researchers to acquire in order that they become part of the
in situ conversations, for we found that the ones that we were privi-
leged to be included in revealed that the children and young people
have much to say.
The rush to hear voices has masked the scrutiny needed to listen
and we therefore argue for a more critical recognition of the ways in
which unconventional voices are shaped and constructed by histor-
ical and cultural perceptions of disability. However, as this chapter
draws to a close, we are reminded of Jackson’s (2003) cautionary
advice. While clearly advocating working hard against the limits of
voice in order to hear those that have been silenced, she warns us to
resist the temptation to echo one’s own voice in silent data or filling
those silences with yet another voice of our own desires.
Conclusion
Specific stories have been presented within this book whose aim is
to capture a glimpse of a world beyond the normative construc-
tions of what ethical research with children looks like. We have
stepped beyond the boundaries of topic, interrogated discursive and
structural motifs of childhood itself, reflected upon our own roles
and identities in the research process and questioned the construc-
tion and application of ethics and ethical principles. The stories
we tell reflect upon the juxtapositions inherent between the eth-
ical principles that currently guide research with children and the
practical realities of such endeavours. Having to confront paradoxes
of position, power, voice and topic, to name but a few, means that
we inevitably challenge orthodox methodological discussions about
ethics. It is in the telling of such stories that we seek to move ethical
understandings further.
The involvement of children in research has become a canonical
narrative which carries certain key concepts such as empowerment,
agency, rights and participation, all of which build an ethical mirage
where these concepts, once articulated, go underexplored and unre-
flected. The paradigm shift which brought ‘child-centred’ research
into common practice challenged previous assumptions about child-
hood and children’s capacities and attributes, allowing for their
greater involvement and inclusion. However, much research with
children neglects to interrogate how these concepts or ‘cherished
conceits’ (Segal 1999) have been defined, how they are practised
by individual children in their research experiences and how they
are used by researchers. These ideas have thus become part of the
146
Conclusion 147
established habitus – the moral order which informs current research
practice to be followed and unquestioned. We are certainly not advo-
cating a return to previous practices with children in research nor
that concepts such as agency and participation not be integral to our
methodologies. But we challenge the lack of debate about how these
concepts are defined and performed by children in research. One
example of this can be found in Chapter 4 where we discuss how
children exert relational agency within the relationships and in the
social contexts where they are situated. Wim, for example, a partic-
ipant in Sarah’s research, consistently used her relational agency to
situate Sarah in the position of authority in the interview and indeed
laughed when Sarah claimed that Wim was in charge. Sarah was var-
iously positioned as researcher, teacher or mother in her interviews
with the young girls in her research. These relational positions were
co-constructed with her participants (Mishler 1986) and demonstrate
the capacity of these girls to negotiate with adults to assert themselves
to their advantage and comfort without challenging the normative
power relations between adults or child that an autonomy of selfhood
(Walkerdine 1989) would induce.
The neglect of reflecting on the ways that dominant attributes of
childhood are articulated, or resisted, inhibits our understandings
of the ways in which children exert power in their relationships
and the ways in which they participate in social interactions which
might challenge the conventional definitions of the very concepts
that inform our practice towards children. Such knowledge can be
lost to us by ignoring, in this case, the social context of research itself,
and what remains are disembodied children’s voices, abstracted from
the social context in which they emerged and were performed. ‘Voice’
has thus become the catchword used to imply all of the key con-
cepts above but with little or no reflection on the embedded power
relations, the positionalities and identities being performed in the
spaces and places where these voices were heard. Such nebulous cob-
webs of social interaction are all too often swept away in our desire
for sanitised research narratives when their inclusion could further
our understanding of how children demonstrated these skills in their
social relations. How identities commonly attributed to children in
research (indeed these attributes are often the only reason the chil-
dren are included in our research), such as pupil, daughter, sister,
adoptee, disabled, innocent, vulnerable, agentic, are complied with,
148 Ethical Research with Children
adapted or resisted by children in our research endeavours are stories
often left unheard.
The stories told here say little about our individual research topics
but instead highlight ways in which children are active in negotiating
their particular social contexts. They reveal ways in which children
exert a relational agency built less on our ‘cherished conceits’ and
‘persistent illusions’ (Segal 1999:110), concerning the attributes of
childhood and more about children’s capacity to transcend assumed
attributes and deficits, and instead use refined skills of managing
complex and dynamic social relationships. As such they highlight
little concerning children’s autonomy and independence but much
on their use of interdependency and social agency. As researchers, we
should include rather than evade such stories in our discussions as
pivotal in furthering our understanding of not just what children say
and do but how and why they say it – the social context which shapes
their voices.
When assumptions regarding agency, voice and indeed childhood
itself remain uninterrogated, when they are not considered as sit-
uated and embedded, the result is a homogenous treatment of all
research with children. What is often advocated as a result of the
paradigmatic shift towards research with children, as discussed above,
is ‘child-centred’ approaches to research and what are made are
claims that we are thus capturing the voices of the agentic child. Such
shifts are important in recognising children in research as subjects,
not objects; however, what we have witnessed in our experiences
is that claims of ‘child-centredness’ are often under-examined and
clear boundaries still exist regarding how, when, why and what chil-
dren are allowed to speak about. Just as Danieli and Woodhams
(2005) argue that the social model of disability has in some ways
come to dominate conversations about disability to the exclusion
of other perspectives, we argue here that child-centred research has
become a catch-all phrase for what is perceived as ‘good’ research
with few having the courage or indeed impetus to question some of
the assumptions it is founded upon.
Philosophical assumptions regarding children’s capabilities to con-
tribute to the construction of knowledge may have shifted within
particular domains, for example, regarding their views on education
or play but still remain limited in areas constructed as taboo or out of
bounds. Certain topics, as we have explored here, such as adoption
Conclusion 149
and non-normative family formation, gendered and sexual identities
or disabilities, are considered as sensitive to explore, particularly in
relation to children. This we argue is a direct result of the key tropes
that abound of childhood itself whereby adults are uncomfortable
hearing the views and experiences of beings whom are constructed
as intrinsically innocent (Faulkner 2011). The manifestation of this
innocence is context-bound and such imagery thus varies, we may
worry about and/or pity the orphaned child, the sexual child, the
street child, the asylum seeking child, the disabled child but what all
of these tropes have in common is the positioning of these children’s
experiences as outside of normative childhoods. Such classifications
are placed before the category of childhood and children are con-
sequently marginalised and othered by this categorisation. Asking
children to speak within such arenas thus generates anxiety despite
the fact that for these children what they are being asked to share is
their own experiences, elements of their own lives, which often they
do not view as taboo at all. We argue that more attention needs to
be paid to the voices of children within domains where they remain
ignored due to some perceived assault on adult expectations of sen-
sitivity or vulnerability. We do not advocate talking to children (or
indeed adults) in ways that may cause them harm or distress but what
we do argue is that the entire notion of ‘sensitive’ topics requires a
much greater degree of interrogation. Such definitions have profound
implications for research participants and researchers alike.
It is through articulations of power that particular positions are
privileged or indeed marginalised. If even the topic that we are
allowed to ask about and that children are allowed to respond
to, is governed by the social order of both research with children
and childhood itself then all aspects of the research process must
be conceptualised as influenced by socially embedded power rela-
tions. Here Foucauldian notions of power relations (Foucault 1972,
1995) are useful in understanding its fluidity in research encoun-
ters. Assumptions in research with children are often made regarding
dichotomies of power whereby adults are powerful and children are
powerless, yet what our stories here reveal is the complex ways in
which power is established, negotiated and articulated by all par-
ties through ideas such as category entitlement and interdependency.
Children’s social agency in these encounters is not easily contained
within a category of ‘powerless’ or ‘dependent’ and as such we should
150 Ethical Research with Children
not naively assume that our approaches, values and methods, how-
ever ideal in their intention, necessarily work to empower children
or to elevate their voices. Indeed in many situations we as adults
occupy a position of ‘powerlessness’, as explored for example in
Chapter 2, whereby the need to maintain guest relations in schools
and homes places researchers in difficult and often subordinate posi-
tions. Despite assumptions, we do not, cannot and are not always
occupying the position of ‘expert’ and our attempts to hand this posi-
tion to children themselves are also far from straightforward and not
necessarily always ethical. Problematising assumptions relating to the
positions that individuals occupy and strategies they may make use
of in research encounters is crucial to understanding research rela-
tionships and the construction of data itself. For example, there is an
assumption that the empowered child is the child who speaks, but
as our stories illustrate, children make effective use of silence as a
strategy of power in social relations (Lewis 2008).
The fluidity of power in research relations is however not actively
acknowledged in the governance of research itself. The tendency of
RECs is to adopt risk-averse approaches to research with children
and young people in the name of protection. Such positioning is in
direct contradiction to the elevation of children’s voices articulated
in rights documentation and academic discourse whereby their agen-
tic participation is considered paramount. Researchers’ thus seeking
this agentic child, the child as being, the child as expert, face a diffi-
cult challenge where they must in ethics documentation emphasise
the vulnerable child in need of protection but methodologically seek
to hear the voices of agentic children. RECs which emphasise assent
over consent or view particular topics as problematic, even illegiti-
mate, can thus come to be seen as barriers which researchers must
overcome rather than resources which can support the conduct of
ethical research. A tendency towards procedural ethics promotes a
desire to present only sanitised narratives of research whereby the
messiness, relational and situational elements of such encounters and
relationships is not accounted for (Guillemin and Gillam 2004).
In this book we have used the concepts of category membership
and entitlement to demonstrate the ways in which respondents in
our research established their authority to speak about topics. The
capacity to ask questions, to answer and to understand the social
interactions on a given topic are all developed through conversations
Conclusion 151
where we each legitimise our stories through the categories we claim
or are attributed to us. As Lawler (2014) argues, stories are used to
produce our identities. Category memberships are created through
discourse and allow us to anticipate the reception our stories will
receive and shape how we tell them to the audience present. Stories
told to Sarah by adoptive mothers were informed by the knowl-
edge that she was also an adoptive mother. Her membership of this
category allowed respondents to understand her interest as well as
assist them in determining whether or not to take part. As seen
in Chapter 5, the girls telling their adoptive stories also used cate-
gory membership, both accepting and resisting attributed identity
traits, as illustrated by Mel’s (and others) claim ‘I am more than
just adopted’. In addition in Chapters 3 and 6, Allison told stories
incited by her research which examined the physical and attitudi-
nal restrictions imposed on particular children’s voices by virtue of
assumed attributes and deficits. Here we also saw young people’s
active resistance of such categorisations.
Often, in order to establish an entitlement to speak as a member of
a particular category self-disclosure is used. Indeed, self-disclosure is
actually productive in Sarah’s research relations. Through openly dis-
cussing, when asked, her status as an intercountry adoptive mother
she set out to reassure young respondents through such discourse and
at least one girl took part as a result of such self-disclosure. As Sarah
identified, the dilemma in this instance about whether self-disclosure
made it easier for this respondent to say yes or harder for her to say
no, remains unresolved. As indicated in Chapters 2 and 6, not dis-
closing stories about her own adoption and mothering experiences
proved to be an issue in her interviews. This of course is in part
related to the method and methodology selected. Semi-structured
interviews where stories are co-constructed hold elements of cultur-
ally expected reciprocity particularly in relation to women (Holmes
1997), described by Oakley (1981) as being, when our questions are
asked back. The balance between a researcher and member of a par-
ticular group can be challenging to maintain when intimate stories
are being shared.
Such a balance is further complicated by the spaces where the
research takes place. As Jordan (2006) has argued, the relationships
between researcher and respondent are more ambiguous in the inti-
mate spaces such as the home. The social conventions surrounding
152 Ethical Research with Children
relations in the home ensure that researchers arrive as guests and are
often perceived as being first and foremost a guest in the home (Yee
and Andrew 2006). Moving beyond this category can be challenging
for the researcher and the interviewee (Bird 2003). Such intricate rela-
tions are common to each social context but performed in unique
ways which are not possible to anticipate prior to the interview.
Mayall (2000b) describes these relations as a triangle of conventions.
As we discuss in Chapter 2, such conventions shape the data in mul-
tiple ways. Sarah illustrates how negative characteristics commonly
attributed to intercountry adopters such as neo-liberal consumer
accessing a market (Dickens 2012), are reconstructed by an intervie-
wee into the very qualities required to progress through an intrusive
and overly bureaucratic system. Such careful identity construction is
inextricably linked to the location in which this interview took place
(Sarah’s home) and the category memberships of those involved; a
social worker and an intercountry adopter. To ignore the social con-
text in which these voices are heard and the category memberships
which inform the ways these stories are told is not only to sanitise
them, but undermine the very complexity of social research. As we
argue in Chapter 4, unless we include the ‘backdrop of [our] interpre-
tations’ (Elliott 2005:183), more commonly in research with children,
then the social embeddedness of children’s agency, and their capacity
to exert it, is lost.
As Sin (2005) argues, issues such as informed consent can seldom,
if ever, be disconnected from the social context in which permis-
sion is sought. This is further complicated by the social positioning
of children within the social context and subsequently in research.
This positioning is exemplified by the issue of informed consent.
In considering the positioning of children as unable to provide full
informed consent, where some use the alternative of ‘assent’ (as dis-
cussed in Chapter 1), we were presented with a motif of children as
irrational beings without the desired capacity to participate fully in
social life. We argue that this is an unacceptable classification of chil-
dren as incapable subjects which reinforces the perpetual ‘othering’
of children within the research process (Lahman 2008) and such prac-
tices contribute to the more general powerlessness of children under
an adult gaze. The risk here is that as a result of dominant develop-
mental frameworks which emphasise not an existing but a growing
capacity to engage in decision-making children become objectified
as passive social objects (Clark 2014). This includes the ability to
Conclusion 153
process information about and consent to participating in social
research. However, we argue that positions such as these fail to recog-
nise the extent to which even very young children engage in decision
making, whereby morality is part of the everyday life of all human
beings (see Mayall 2002).
Where informed consent is not required or deemed appropriate
it is still common to see the phrase assent used to define the con-
firmation of children’s participation in research. Assent is defined
as the child’s permission or affirmation of agreement to participate
in research (Broome and Richards 1998) and is regarded by some
researchers as a substitute for consent so that children do not con-
sent in their own right but assent to their parent’s consent (Powell
and Smith 2010). We have questioned whether the upholding of dig-
nity, highlighted in many ethical guidelines (BERA 2011; BSA 2002;
ESRC 2012) and in key human rights instruments (Bell 2008), can
be maintained while positioning children’s agreement as lesser than
that of adults within the research process; a position which ‘fails to
accord them the same rights as adults in terms of what their consent
means’ (Skelton 2008:23).
What we risk when we construct children and young people’s
‘consent’ as somehow lesser than adults is the privileging of par-
ticular voices over others. We maintain the status quo of children
as objects of research, which was supposed to be overturned by the
introduction and now dominance of participatory methods. This is
particularly the case for groups of children categorised as particularly
vulnerable by virtue of ascribed characteristics (such as the disabled
child, the adopted child, the sexualised child). As Allison’s research
demonstrates the children and young people who participated in
her study where the last to be consulted in a long and intricate
‘chain of consent’. This chain was directly related to the networks of
power belonging to particular adults surrounding the children, from
head teacher to class teachers to parents and finally, to the children.
Jessica’s experiences being guided away from online research with
children and young people (discussed in Chapter 2) mirror this con-
cern regarding the desire to but complexities of gaining authentic
informed consent from children and adults in particular spaces. The
result of such classifications is that particular stories are more readily
accessed and thus more readily heard.
As Frank (2010) argues, stories accompany us through life, from
birth to death, they entertain, educate and can inform us about right
154 Ethical Research with Children
and wrong and show us who we are. The stories we share within
this book illustrate how we applied, questioned and critiqued eth-
ical procedures and principles and made decisions in the moment
according to the situation and the social contexts we found ourselves
in. Such stories of course are not new in qualitative research; they
are even emerging in quantitative research (Elliot 2005). But we sug-
gest that they are as yet, underexploited in research with children.
The imperative to be ethical and be seen to be ethical when involv-
ing children, may act as a deterrent for researchers in telling stories
which reveal the doing of ethics to be more ambiguous and socially
situated than textbooks imply. Or that agency and empowerment
are less straightforward to evoke in our endeavours than the defi-
nitions and aspirations suggest. Or that capturing the ‘voice’ of the
child is enough when what subsequently evades us are the children
themselves.
We celebrate those who came before us and successfully chal-
lenged the assumed passivity and incapacities of children so explicit
in previous approaches to research on children. But we suggest that
the contemporary social order through which participatory research
with children is organised and delivered now needs greater critical
reflection. Doing ethics, being ethical, does not begin or end with a
signed form and permission from an ethics committee, this is merely
the beginning. Research is messy, it is fractured and it is imperfect.
These stories are also messy, they represent our stories of the ethical
decisions we made and as such they are not presented as ‘right’ or
‘wrong’. Rather as multiple, contingent, hidden and context-bound.
Although we, as authors, have in common similar values about
research with children and a shared vision of what it should look
like and we have articulated them here, we recognise that when we
seek to shine a light on these dark spaces we will thus create shad-
ows elsewhere. As such by sharing these narratives, our hope is not
to provide a guidebook for the conducting of research with children
but that other researchers are encouraged to tell their own stories, to
make their own unique decisions in their own moments and to move
beyond a simplistic dichotomy of ‘good’ or ‘bad’ research and ‘right’
or ‘wrong’ ethics.
References
Aapola, S., Gonick, A. and Harris, A. (2004) Young Femininity: Girlhood, Power
and Social Change, Basingstoke: Palgrave Macmillan
Aarsan, P. A. and Aronsson, K. (2009) ‘Gaming and Territorial Negotiations in
Family Life’ Childhood, vol.16(4), 497–517
Abebe, T. and Bessell, S. (2014) ‘Advancing Ethical Research with Children:
Critical Reflections on Ethical Guidelines’ Children’s Geographies, vol.12(1),
126–133
Abell, J., Locke, A., Condor, S., Gibson, S. and Stevenson, C. (2006) ‘Try-
ing Similarity, Doing Difference: The Role of Interviewer Self-Disclosure
in Interview Talk with Young People’ Qualitative Research, vol.6(2),
221–244
Agar, M. H (1980) The Professional Stranger, New York: Academic Press
Alderson, P. (1995) Listening to Children, London: Barnardos
Alderson, P. and Goodey C. (1996) ‘Research with Disabled Children: How
Useful Is Child-Centred Ethics?’ Children and Society, vol.10(2), 106–116
Alderson, P. and Montgomery, J. (1996) Healthcare Choices: Making Decisions
with Children, London: Institute for Public Policy Research
Alderson, P. and Morrow, V. (2006) ‘Multi-Disciplinary Research Ethics Review:
Is It Feasible?’ International Journal of Social Research Methodology: Theory and
Practice, vol.9, 405–417
Allan, G. and Crow, G. (2001) Families, Households and Society, Basingstoke:
Palgrave Macmillan
Alleyne, B. (2015) Narrative Networks: Storied Approaches in a Digital Age,
London: Sage
Alvesson, M., Ashcraft, K. L. and Thomas, R. (2008) ‘Identity Matters: Reflec-
tions on the Construction of Identity Scholarship in Organization Studies’
Organization, vol.15, 5–28
Alvesson, M. and Sköldberg, K. (2009) Reflexive Methodology (2nd ed.), Los
Angeles: Sage
Andrews, M. (2008) Never the Last Word: Revisiting Data, in Andrews, M.,
Squire, C. and Tamboukou, M. (eds.) Doing Narrative Research, London: Sage,
86–102
Andrews, M., Squires, C. and Tamboukou, M. (eds.) (2008) Doing Narrative
Research, London: Sage
Anzil, V. (2013) ‘Adopting “Imaginaries”: International Adoption in the
Spanish Press’ Adoption & Fostering, vol.37(1), 71–82
Archard, D. (2004) Children: Rights and Childhood (2nd ed.), London: Routledge
Atkinson, P. and Coffey, A. (2003) Revisiting the Relationship Between Par-
ticipant Observation and Interviewing, in Gubrium, J. F. and Holstein, J. A.
(eds.) Postmodern Interviewing, London: Sage, 109–122
155
156 References
Aull Davies, C. (2008) Reflexive Ethnography: A Guide to Researching Selves and
Others (2nd ed.), London: Routledge
Badham, B. (2004) ‘Participation – for a Change: Disabled Young People Lead
the Way’ Children and Society, vol.18, 143–154
Baker, C. D. (2003) Ethnomethodological Analyses of Interviews, in Holstein,
J. A. and Gubrium, J. F. (eds.) Inside Interviewing New Lenses, New Concerns,
London: Sage, 395–412
Barker, J. and Smith, F. (2001) ‘Power, Positionality and Practicality: Carry-
ing Out Fieldwork with Children’ Ethics, Place and Environment: A Journal of
Philosophy and Geography, vol.4(2), 142–147
Barker, J. and Weller, S. (2003) ‘Never Work with Children? The Geography
of Methodological Issues in Research with Children’ Qualitative Research,
vol.3(2), 207–227
Barn, R. (2000) Race, Ethnicity and Transracial Adoption, in Katz, I. and
Treacher, A. (eds.) The Dynamics of Adoption, London: Jessica Kingsley
Publishers, 101–125
Barn, R. and Kirton, D. (2012) Transracial Adoption in Britain Politics, Ide-
ology and Reality, in Phoenix, A. and Simmonds, J. (eds.) Multiculturalism,
Identity and Family Placement, Adoption & Fostering, vol.36(3–4),
25–37
Barthes, R. (1977) Music, Image, Text, Trans. Stephen Heath, London: Fontana
Press
Bartholet, E. (1999) Family Bonds: Adoption, Infertility and the New World of
Child Production, Boston: Beacon Press
Basset, E. and O’Riordan, K. (2002) ‘Ethics of Internet Research: Contesting the
Human Subjects Research Model’ Ethics and Information Technology, vol.4(3),
233–247
Beck, U. (1992) Risk Society: Towards a New Modernity, London: Sage
Benatar, S. (2002) ‘Reflections and Recommendations on Research Ethics in
Developing Countries’ Social Science and Medicine, vol.54, 1131–1141
BERA (British Educational Research Association) (2011) Ethical Guidelines
for Educational Research, Available at: www.bera.ac.uk/publications/ethical
-guidelines [Accessed: 29th January 2015]
Bell, N. (2008) ‘Ethics in Child Research: Rights, Reason and Responsibilities’
Children’s Geographies, vol.6(1), 7–20
Bell, C. and Newby, H. (1977) Doing Sociological Research, London: Allen and
Unwin
Bell, C. C. and Roberts, H. (eds.) (1984) Social Researching: Politics, Problems,
Practice, London: Routledge & Kegan Paul
Bhavnani, K. (1991) Talking Politics: A Psychological Framing of Views from Youth
in Britain, Cambridge: Cambridge University Press
Bird, S. E. (2003) The Audience in Everyday Life: Living in a Media World,
New York: Routledge
Bluebond-Langner, M. (1978) The Private World of Dying Children, Princeton,
NJ: Princeton University Press
References 157
Boden, R. and Epstein, D. (2006) ‘Managing the Research Imagination? Glob-
alisation and Research in Higher Education’ Globalisation, Societies and
Education, vol.4(2), 223–236
Boden, R., Epstein, D. and Latimer, J. (2009) ‘Accounting for Ethos or Pro-
grammes for Conduct? The Brave New World of Research Ethics Commit-
tees’ The Sociological Review, vol.57(4), 728–749
Boggis, A. (2011) ‘Deafening Silences: Researching with Inarticulate Children’
Disability Studies Quarterly, vol.31(4)
Borland, M. A. Laybourn, M. Hill and Brown, J. (1998) Middle Childhood: The
Perspectives of Children and Parents, London: Jessica Kingsley
Bourdieu, P. (1977) Outline of a Theory of Practice, Cambridge: Cambridge
University Press
Bourdieu, P. (1986a) Distinction, London: Routledge
Bourdieu, P. (1986b) ‘From Rules to Strategies’ Cultural Anthropology, vol.1,
110–120
Brooks, R., te Riele, K. and Maguire, M. (2014) Ethics and Education Research,
London: Sage
Broome, M. and Richards, D. (1998) ‘Involving Children in Research’ Journal
of Child and Family Nursing, vol.1(1), 3–7
BSA (British Sociological Association) (2002) Statement of Ethical Prac-
tice, Available at: www.britsoc.co.uk/about/equality/statement-of-ethical
-practice.aspx [Accessed: 29th January 2015]
Bruckman, A. (2002) ‘Studying the Amateur Artist: A Perspective on Disguis-
ing Data Collected in Human Subjects Research on the Internet’ Ethics and
Information Technology, vol.4(3), 217–231
Bruner, J. (2004) ‘Life as Narrative’ Social Research, vol.71(3), 691–710
Bryson, M. and de Castell, S. (1994) ‘Telling Tales Out of School: Modernist,
Critical, and Postmodern True Stories about Educational Computing’ Journal
of Educational Computing Research, vol.10, 199–221
Butler, I. and Williamson, H. (1994) Children Speak: Children, Trauma and Social
Work, London: Longman
Cameron, L. and Murphey, J. (2006) ‘Obtaining Consent to Participate in
Research: The Issues Involved in Including People with a Range of Learn-
ing and Communication Disabilities’ British Journal of Learning Disabilities,
vol.35, 113–120
Carlson, C. (2004) What Is Performance?, in Bial, H. (ed) The Performance
Studies Reader (2nd ed.), London: Routledge, 70–75
Cavet, J. and Sloper, P. (2004) ‘Participation of Disabled Children in Indi-
vidual Decisions About Their Lives and in Public Decisions about Service
Development’ Children and Society, vol.18, 278–290
Cerulo, K. (1997) ‘Identity Construction: New Issues, New Directions’ Annual
Review of Sociology, vol.23, 385–409
Channel 4 Press (2014) ‘The Paedophile Next Door’ Available at: http://www
.channel4.com/info/press/programme-information/the-paedophile-next
-door [Accessed: 21st January 2015]
158 References
Children Act 1989, London: HMSO
Children (Scotland) Act, 1995
Children (Northern Ireland) Order, 1995
Childwise (2013) Report: Trends in Media Use, Research Highlights for Children’s
Online Safety 50, Available at: http://www.saferinternet.org.uk/content
/childnet/safterinternetcentre/downloads/Research_Highlights/UKCCIS
_RH50_Childwise_Trends_Report.pdf [Accessed: 21st January 2015]
Christensen, P. (1999) Towards an Anthropology of Childhood Sickness: an
Ethnographic Study of Danish School Children. PhD Thesis, University
of Hull
Christensen, P. and James, A. (2000) Research with Children: Perspectives and
Practices, London: Falmer Press
Christensen, P. and James, A. (2001) What Are Schools for? The Temporal
Experience of Children’s Learning in Northern England, in Alanen, L. and
Mayall, B. (eds.) Conceptualising Child-Adult Relations, London: Routledge,
70–85
Christensen, P. and Prout, A. (2002) ‘Working with Ethical Symmetry in Social
Research with Children’ Childhood, vol.9(4), 477–497
Clark, J. (2013a) ‘Passive, Heterosexual and Female: Constructing Appropriate
Childhoods in the Sexualisation of Childhood Debate’ Sociological Research
Online, vol.18(2), Available at Accessed: http://www.socresonline.org.uk/18/
2/13.html
Clark, J. (2013b) Children’s Geographies, in Taylor, J., Bond, E. and Woods,
M. (eds.) Early Childhood Studies: A Multidisciplinary and Holistic Introduction,
London: Hodder, 272–292
Clark, J. (2014) Sexualisation and the Discursive Figure of the Child, in
Warehime, M. N. (ed) Soul of Society: A Focus on the Lives of Children and
Youth, Sociological Studies of Children and Youth, vol.18, Bingley: Emerald
Publishing, 173–197
Coad, J. and Evans, R. (2008) ‘Reflections on Practical Approaches to Involv-
ing Children and Young People in the Data Analysis Process’ Children and
Society, vol.22(1), 41–52
Cockburn, T. (1998) ‘Children and Citizenship in Britain: A Case for a Socially
Interdependent Model of Citizenship’ Childhood, vol.5(1), 99–117
Cockburn, T. (2005) ‘Children and the Feminist Ethic of Care’ Childhood,
vol.12(1), 71–89
Code, L. (1991) What Can She Know? Feminist Theory and the Construction of
Knowledge, New York: Cornell University Press
Coffey, A. (1999) Ethnographic Self, London: Sage
Collier Quy, R. (2006) Performing Childhood, in Trenka, J., Oparah J. C. and
Shin S. Y. (eds.) Outsiders Within, Cambridge: South End Press, 213–217
Collins, R. (2004) Interaction Ritual Chains, Princeton, NJ: Princeton University
Press
Connolly, P. (2000) Race, Gender and Critical Reflexivity in Research with
Young Children, in Christensen, P. and James, A. (eds.) Research with
Children Perspectives and Practices, London: Routledge, 173–188
References 159
Corsaro, W. A. (1997) The Sociology of Childhood, Thousand Oaks, CA: Pine
Forges Press
Crapanzano, V. (1986) Hermes Dilemma: The Masking of Subversion in
Ethnographic Description, in Marcus, G. and Clifford, J. (eds.) Writing
Culture, Berkeley: University of California Press, 51–76
Crompton, R. (2001) ‘Gender Comparative Research and Biographical Match-
ing’ European Societies, vol.3(2), 167–190
Conquergood, D. (1991) ‘Rethinking Ethnography: Towards a Critical Cultural
Politics’ Communications Monographs, vol.58, 179–194
Conquergood, D. (2002) ‘Performance Studies: Interventions and Radical
Research’ The Drama Review, vol.46(2), 145–156
Daniel, P. and Gustafsson, U. (2010) ‘School Lunches: Children’s Services or
Children’s Spaces?’ Children’s Geographies, vol.8(3), 265–274
Danieli, A. and Woodhams, C. (2005) ‘Emancipatory Research and Disabil-
ity: A Critique’ International Journal of Social Research Methodology, vol.8(4),
281–296
David, M., Edwards, R. and Alldred, P. (2001) ‘Children and School-Based
Research: Informed Consent or Educated Consent?’ British Educational
Research Journal, vol.27(3), 347–365
Denscombe, M. and Aubrook, L. (1992) ‘It’s Just Another Piece of Schoolwork:
The Ethics of Questionnaire Research on Pupils in Schools’ British Educa-
tional Research Journal, vol.18, 113–131
Denzin, N. (2000) The Practices and Politics of Interpretation, in Denzin, N. K.
and Lincoln Y. S. (eds.) Handbook of Qualitative Research (2nd ed.), Thousand
Oaks, CA: Sage, 897–922
Denzin, N. K. (2003) Performance Ethnography Critical Pedagogy and the Politics
of Culture, London: Sage
Department of Health (1991) The Children Act 1989 Guidance and Regulations:
Volume 6 Children with Disabilities, London: Department of Health
DePalma, R. (2010) ‘Socially Just Research for Social Justice: Negotiating Con-
sent and Safety in a Participatory Action Research Project’ International
Journal of Research and Method in Education, vol.33(3), 215–227
DeSchauwer, E., Van Hove, G. and Loots, G. (2009) ‘ “I need help on Mondays,
it’s not my day. The other days, I’m ok”. Perspectives of Disabled Children
on Inclusive Education’ Children and Society, vol.23, 99–111
Devine, F. and Heath, S. (1999) Sociological Research Methods in Context,
Basingstoke: Palgrave Macmillan
Dickens, J. (2012) ‘Social Policy Approaches and Social Work Dilemmas in
Intercountry Adoption’ in Gibbons, J. L. and Rotabi, K. S. (eds.) Intercountry
Adoption Policies, Practices, and Outcomes, Farnham: Ashgate pp.29–42
Dingwall, R. (2008) ‘The Ethical Case Against Ethical Regulation in
Humanities and Social Science Research’ Twenty-First Century Society,
vol.3(1), 1–12
Dingwall, R. (2011) Absurd Decisions by Ethics Committees, Avail-
able at: www.socialsciencespace.com/2011/02/absurd-decisions-by-ethics
-committees [Accessed: 29th January 2015]
160 References
Dockett, S., Perry, B., Kearney, E., Hampshire, A., Mason, J. and Schmied,
V. (2009) ‘Researching with Families: Ethical Issues and Situations’ Contem-
porary Issues in Early Childhood, vol.10(4), 353–365
Dorow, S. (2006) Transnational Adoption. A Cultural Economy of Race, Gender
and Kinship, New York: New York University Press
Douglas, M. (1979) Man, Myth and Magic, in Cavendish, R. (ed) Taboo,
London: BPCC/Phoebus Publishing, 2767–2761
Douglas, M. (1991) Purity and Danger, London: Routledge
Drew, P. (1978) ‘Accusations: The Occasioned Use of Members’ Knowledge of
“Religious Geography” in Describing Events’ Sociology, vol.12(1), 1–22
Duchinsky, R. and Clark, J. The Sexualisation of Boys? Purity, Danger and
Masculinity, unpublished paper
Dunbar, C., Rodriquez, D. and Parker, L. (2002) Race, Subjectivity and the
Interview Process, in Gubrium, J. F. and Holstein, J. A. (eds.) Handbook of
Interview Research, London: Sage, 279–298
Dworkin, G. (1988) The Theory and Practice of Autonomy, Cambridge:
Cambridge University Press
Eder, D. and Fingerson, L. (2003) Interviewing Children and Adolescents, in
Holstein, J. A. and Gubrium, J. F. (eds.) Inside Interviewing: New Lenses, New
Concerns, London: Sage, 33–53
Edwards, C. and Imrie, R. (2003) ‘Disability and Bodies as Bearers of Value’
Sociology, vol.37(2), 239–256
Edwards, R. and Mauthner, M. (2012) Ethics and Feminist Research: Theory
and Practice, in Miller, T., Birch, M., Mauthner, M. and Jessop, J. (eds.) Ethics
and Qualitative Research (2nd ed.), London: Sage, 14–28
Elbers, E. (2004) ‘Conversational Asymmetry and the Child’s Perspective in
Developmental and Educational Research’ International Journal of Disability,
Development and Education, vol.51(2), 201–205
Elliott, J. (2005) Using Narrative in Social Research, London: Sage
Elliott, A. (2008) Concepts of the Self (2nd ed.), London: Polity Press
Emirbayer, M. and Mische, A. (1998) ‘What Is Agency?’ American Journal of
Sociology, vol.103(4), 962–1023
Emond, R. (2009) ‘I Am All about the Future World: Cambodian Children’s
Views on Their Status as Orphans’ Children & Society, vol.23(6), 407–417
Ennew, J. (1994) Time for Children or Time for Adults, in Qvortrup, J., Bardy,
M., Sgritta, G. and Wintersberger, H. (eds.) Childhood Matters Social Theory,
Practice and Politics, Aldershot: Avebury, 1–23
Ennew, J. (2005) Prisoners of Childhood: Orphans and Economic Depen-
dency, in Qvortrup, J. (ed) Studies in Modern Childhood: Society, Agency,
Culture, Basingstoke: Palgrave Macmillan, 128–146
ESRC (Economic and Social Research Council) (2012) Framework for Research
Ethics, Available at: www.esrc.ac.uk/about-esrc/information/research-ethics
.aspx [Accessed: 29th January 2015]
Faulkner, J. (2011) The Importance of Being Innocent, Cambridge: Cambridge
University Press
Fawcett, B. (2000) Feminist Perspectives on Disability, Harlow: Pearson Educa-
tion Ltd
References 161
Feinberg, J. (1980) The Child’s Right to an Open Future, in Aiken, W. and La
Follette, H. (eds.) Whose Child? Children’s Rights Parental Authority and State,
Totowa: Rowman and Littleford, 124–153
Finch, J. (1984) ‘It’s Great to Have Someone to Talk to’: Ethics and Politics of
Interviewing Women, in Bell, C. and Roberts, H. (eds.) Social Researching:
Politics, Problems, Practice, London: Routledge and Kegan Paul, 70–87
Fine, M. (1994) Working with Hyphens: Reinventing Self and Other in Qual-
itative Research, in Denzin, N. K. and Lincoln, Y. S. (eds.) Handbook of
Qualitative Research, Thousand Oaks: Sage, 70–82
Fisher, A. P. (2003) ‘Still Not Quite as Good as Having Your Own: Toward a
Sociology of Adoption’ Annual Review of Sociology, vol.29, 255–276
Flewitt, R. (2005) ‘Conducting Research with Young Children: Some
Ethical Considerations’ Early Child Development and Care, vol.175(6),
553–565
Foley, P. (2011) Democratic Spaces, in Foley, P. and Leverett, D. (eds.) Children
and Young People’s Spaces, Basingstoke: Palgrave Macmillan, 89–101
Foucault, M. (1972) The Archaeology of Knowledge (Sheridan Smith, A. M.
Trans.), New York: Pantheon Books
Foucault, M. (1995) Discipline and Punish: The Birth of the Prison, New York:
Vintage Books
Foucault, M. (1998) The History of Sexuality: The Will to Knowledge, London:
Penguin
Frank, A. (2010) Letting Stories Breathe: A Socio-Narratology, London: University
of Chicago Press
Frankel, S. (2012) Children, Morality and Society, Basingstoke: Palgrave
Macmillan
Franklin, A. and Sloper, P. (2009) ‘Supporting the Participation of Disabled
Children and Young People in Decision-Making’ Children and Society, vol.23,
3–15
Fraser, S., Lewis, V., Ding, S., Kellett, M. and Robinson, C. (eds.) (2004) Doing
Research with Children and Young People, London: Sage
Fraser, S., Maher, J. and Wright, J. (2010) ‘Between Bodies and Collectivities:
Articulating the Action of Emotion in Obesity Epidemic Discourse’ Social
Theory & Health, vol.8(2), 192–209
Freund, P. E. (1990) ‘The Expressive Body: A Common Ground for the Soci-
ology of Emotions and Health and Illness’ Sociology of Health and Illness,
vol.12(4), 453–477
Friedlander, M. L. (2003) ‘Adoption: Misunderstood, Mythologized,
Marginalised’ The Counseling Psychologist, vol.31(6), 745–752
Gaber, I. and Aldridge, J. (1994) Introduction, in Gaber, I. and Aldridge,
J. (eds.) In The Best Interests of the Child Culture, Identity and Transracial
Adoption, London: Free Association Books, 1–5
Gailey, C. W. (2000) Race, Class and Gender in Intercountry Adoption in
the USA, in Selman, P. (ed) Intercountry Adoption Developments, Trends and
Perspectives, London: BAAF Publications, 295–313
Gallacher, L. (2005) ‘The Terrible Twos: Gaining Control in the Nursery?’
Children’s Geographies, vol.3(2), 243–246
162 References
Gallacher, L. and Gallagher, M. (2008) ‘Methodological Immaturity in
Childhood Research? Thinking through Participatory Methods’ Childhood,
vol.15(4), 499–516
Garfinkel, H. (1967) Studies in Ethnomethodology, New Jersey: Prentice-Hall
Garrett, P. M. (2009) Transforming Children’s Services? Social Work, Neoliberalism
and the Modern World, Maidenhead: Open University Press
Giddens, A. (1979) Action, Structure and Contradiction in Social Analysis,
Berkeley, CA: University of California Press
Giddens, A. (1984) The Constitution of Society, Cambridge: Polity Press
Giddens, A. (1991a) Modernity and Self-Identity: Self and Society in the Late
Modern Age, Cambridge: Polity Press
Giddens, A. (1991b) The Consequences of Modernity, Cambridge: Polity Press
Giddens, A. (1993) New Rules of Sociological Method: A Positive Critique of
Interpretive Sociologies (2nd ed.), Cambridge: Polity Press
Gilligan, C. (1993) In a Different Voice (2nd ed.), London: Harvard University
Press
Gilligan, T. (2014) ‘Paedophilia Is Natural and Normal for Males: How
Some University Academics Make the Case for Paedophiles at Summer
Conferences’ The Telegraph, 5th July
Goffman, E. (1959) The Presentation of Self in Everyday Life, London: Penguin
Goode, S. (2010) Understanding and Addressing Adult Sexual Attraction to
Children, Abingdon: Routledge
Graue, E. M. and Walsh, D. J. (1998) Studying Children in Social Context:
Theories, Methods and Ethics, London: Sage
Greenfield, P. M. (1994) Independence and Interdependence as Developmen-
tal Scripts; Implications for Theory, Research and Practice, in Greenfield,
P. M. and Cocking, R. R. (eds.) Cross-Cultural Roots of Minority Child
Development, New Jersey: Erloaum, 1–37
Greig, A., Taylor, J. and Mackay, T. (2013) Doing Research with Children a
Practical Guide (3rd ed.), London: Sage
Griffith, A. (1998) ‘Insider/Outsider: Epistemological Privilege and Mothering
Work’ Human Studies, vol.21, 361–376
Guba, E. G. and Lincoln, Y. S. (1994) Competing Paradigms in Qualitative
Research, in Denzin, N. K. and Lincoln, Y. S. (eds.) Handbook of Qualitative
Research, Thousand Oaks, CA: Sage, 105–117
Guillemin, M. and Gillam, L. (2004) ‘Ethics, Reflexivity, and “Ethically
Important Moments” in Research’ Qualitative Inquiry, vol.10(2),
261–280
Guta, A., Nixon, S. A. and Wilson, M. G. (2013) ‘Resisting the Seduction of
“Ethics Creep”: Using Foucault to Surface Complexity and Contradiction in
Research Ethics Review’ Social Science and Medicine, vol.98, 301–310
Haavind, H. (2005) ‘Review Essay Towards a Multifaceted Understanding of
Children as Social Participants’ Childhood, vol.12(1), 139–152
Hacking, I. (1986) Making Up People, in Heller, P., Sosna, M. and Wellberry,
D. (eds.) Reconstructing Individualism, Stanford: Stanford University Press,
222–236
References 163
Hallowell, N., Lawton, J. and Gregory, S. (2005) Reflections on Research:
The Realities of Doing Research in the Social Sciences, Buckingham: Open
University Press
Hammersley, M. (2009) ‘Against the Ethicists: On the Evils of Ethical Regula-
tion’ International Journal of Social Research Methodology, vol.12(3), 211–225
Hammersley, M. and Atkinson, P. (1995) Ethnography Principles in Practice (2nd
ed.), London: Routledge
Hammersley, M. and Traianou, A. (2011) ‘Moralism and Research Ethics:
A Machiavellian Perspective’ International Journal of Social Research Method-
ology, vol.14(5), 379–390
Hammersley, M. and Traianou, A. (2012) Ethics in Qualitative Research: Contro-
versies and Contexts, London: Sage
Haraway, D. (1991) Simians, Cyborgs and Women New York: Routledge and
Kegan Paul
Harden, J., Scott, S., Backett-Milburn, K. and Jackson, S. (2000) ‘Can’t
talk, Won’t Talk? Methodological Issues in Researching Children’ Sociolog-
ical Research Online, 5(2), Available at: http://www.socresonline.org.uk/5/2
/harden.html [Accessed: 20th June 2013]
Harden, J. (2012) ‘Good Sitting, Looking and Listening: The Regulation
of Young Children’s Emotions in the Classroom’ Children’s Geographies,
vol.10(1), 83–93
Harris, A. (2003) ‘gURL Scenes and grrrl zines: The Regulation and Resistance
of Girls in Late Modernity’ Feminist Review, vol.75, 38–56
Hartsock, N. (1996) ‘Postmodernism and Political Change: Issues for Femi-
nist Theory’, in Hekman, S. J. (ed) Feminist Interpretations of Michel Foucault,
Pennsylvania: Pennsylvania State University Press, 39–58
Haugen, G. M. D. (2008) ‘Children’s Perspectives on Everyday Experiences
of Shared Residence: Time, Emotions and Agency Dilemmas’ Children in
Society, vol.11, 71–94
Heath, S., Charles, V., Crow, G. and Wiles, R., (2007) ‘Informed Con-
sent, Gatekeepers and Go-Betweens: Negotiating Consent in Child and
Youth-Orientated Institutions’ British Educational Research Journal, vol.33(3),
403–417
Hendrick, H. (2000) The Child as Social Actor in Historical Sources. Problems
of Identification and Interpretation, in Christensen, P. and James, A. (eds.)
Research with Children: Perspectives and Practices, London: Falmer Press,
36–61
Hendrick, H. (ed) (2005) Child Welfare and Social Policy, Bristol: Policy Press
Heritage, John (1984) Garfinkel and Ethnomethodology, Cambridge: Polity Press
Herz, R. (1996) ‘Introduction: Ethics, Reflexivity and Voice’ Qualitative Sociol-
ogy, vol.19(1), 3–9
Hester, S. and Eglin, P. (1997) Culture in Action: Studies in Membership Categori-
sation Analysis, Washington: University Press of America
Hewson, C., Yule, P., Laurent, D. and Vogel, C. (2003). Internet Research
Methods: A Practical Guide for the Behavioural and Social Sciences, London:
Sage
164 References
Hill, E. J. (2005) ‘Work-Family Facilitation and Conflict, Working Fathers and
Mothers, Work-Family Stressors and Support’ Journal of Family Issues, 26(6),
793–819
Hine C. (2005) ‘Internet Research and the Sociology of Cyber-Social-Scientific
Knowledge’ Information Society, 21(4), 239–248
Hobbs, D. and May, T. (1993) Interpreting the Field: Accounts of Ethnography,
Oxford: Clarendon Press
Hochschild, A. (1983) The Managed Heart: Commercialization of Human Feeling,
Berkeley: University of California Press
Hockey, J. and James, A. (1993) Growing Up and Growing Old: Ageing and
Dependency in the Life Course, London: Sage
Högbacka, R. (2012) Maternal Thinking in the Context of Stratified Repro-
duction: Perspectives of Birth Mothers from South Africa, in Gibbons, J. L.
and Rotabi, K. S. (eds.) Intercountry Adoption, Policies, Practices and Outcomes,
Surrey: Ashgate, 143–159
Holland, P. (1992) What Is a Child?: Popular Images of Childhood, London:
Virago
Hollway, W. and Jefferson, T. (2000) Doing Qualitative Research Differently: Free
Association, Narrative and the Interview Method, London: Sage
Holloway, S. and Valentine, G. (eds.) (2000) Children’s Geographies: Playing,
Living, Learning, London: Routledge
Holloway, S. and Valentine, G. (2001) ‘ “It’s Only as Stupid as you are”: Chil-
dren’s and Adult’s Negotiation of ICT Competence at Home and at School’
Social and Cultural Geography, vol.2(1), 25–42
Holmes, J. (1997) ‘Women, Language, Identity’ Journal of Sociolinguistics,
vol.1(2), 195–223
Holstein, J. A. and Gubrium, J. F. (1995) The Active Interview Thousand Oaks,
CA: Sage
Howarth, C. (2002) ‘Using the Theory of Social Representation to Explore
Difference in the Research Relationship’ Qualitative Research, vol.2(1),
21–34
Hudson, J. M. and Bruckman, A. (2004) ‘ “Go away”: Participant Objections
to Being Studied and the Ethics of Chat Room Research’ The Information
Society, vol.20, 127–139
Huxley, P., Evans, S., Davidson, B. and King, S. (2005) ‘The Costs of Scrutiny
in Applied Health and Social Care Research: A Case Study’ Research Policy
and Planning, vol.23(1), 55–60
Hydén, M. (2008) Narrating Sensitive Topics, in Squire, C., Andrews, M. and
Tamboukou, M. (eds.) Doing Narrative Research, London: Sage, 121–127
Jackson, A. Y. (2003) ‘Rhizovocality’ Qualitative Studies in Education, vol.16(5),
693–710
Jacobson, H. (2008) Culture Keeping, Nashville: Vanderbilt University Press
James, A. and James, A. (2004) Constructing Childhood, Basingstoke: Palgrave
Macmillan
James, A., Jenks, C. and Prout, A. (1998) Theorising Childhood, Cambridge:
Polity Press
References 165
James, A. and Prout, A. (1997) Re-Presenting Childhood: Time and Transition
in the Study of Childhood, in James, A. and Prout, A. (eds.) Constructing and
Reconstructing Childhood, London: Falmer Press, 230–250
Jayyusi, L. (1984) Categorization and the Moral Order, London: Routledge and
Kegan Paul
Jenks, C. (1996) Childhood, London: Routledge
Jenks, C. (2005) ‘Journeys into Space’ Childhood, vol.12(4), 419–424
Johnson, K. (2012) Challenging the Discourse of Intercountry Adoption:
Perspectives from Rural China, in Gibbons, J. L. and Rotabi, K. S. (eds.)
Intercountry Adoption Policies, Practices, and Outcomes, Farnham: Ashgate,
103–117
Jordan, A. B. (2006) ‘Make Yourself at Home: The Social Construction of
Research Roles in Family Studies’ Qualitative Research, vol.6(2), 169–185
Kanuka, H. and Anderson, T. (2007). ‘Ethical Issues in Qualitative E-learning
Research’ International Journal of Qualitative Methods, 6(2), Available at: http:
//www.ualberta.ca/∼iiqm/backissues/6_2/kanuka.htm [Accessed: 20th May
2015]
Kehily, M. J. (2012) ‘Contextualising the Sexualisation of Girls Debate:
Innocence, Experience and Young Female Sexuality’ Gender and Education,
vol.24(3), 255–268
Kellet, M. (2005) How to Develop Children as Researchers, London: Paul
Chapman Publishing
Kellett, M., Robinson, C. and Burr, R. (2004) Images of Childhood, in Fraser,
S., Lewis, V., Ding, S., Kellet, M. and Robinson, C. (eds.) Doing Research with
Children and Young People, London: Sage, 161–174
Killen, M. and Wainryb, C. (2000) Independence and Interdependence in
Diverse Cultural Contexts, in Harkness, S. and Raeff, C. (eds.) Individu-
alism and Collectivism as Cultural Contexts for Development, San Francisco:
Jossey-Bass, 5–22
Kirschner, D. (1990) ‘The Adopted Child Syndrome: Considerations for
Psychotherapy’ Psychotherapy in Private Practice, vol.8(3), 93–100
Kirton, D. (1996) ‘Race and Adoption’ Critical Social Policy, vol.46(16), 123–136
Kitchin, R. (2000) ‘The Researched Opinions on Research: Disabled People and
Disability Research’ Disability and Society, vol.15(1), 25–47
Krieger, S. (1987) ‘Beyond Subjectivity: The Use of the Self in Social Science’
Qualitative Sociology, vol.8(4), 309–324
Komulainen, S. (2007) ‘The Ambiguity of the Child’s “Voice” in Social
Research’ Childhood, vol.14(1), 11–28
Lahman, M. K. E. (2008) ‘Always Othered: Ethical Research with Children’
Journal of Early Childhood Research, vol.6(3), 281–300
Laidlaw, J. (2014) The Undefined Work of Freedom: Foucault’s Genealogy and
the Anthropology of Ethics, in Faubion, J. D. (ed) Foucault Now, Cambridge:
Polity, 23–37
Lange, A. and Mierendorff, J. (2009) Method and Methodology in Childhood
Research, in Qvortrup, J., Corsaro, W. and Honig, M. (eds.) The Palgrave
Handbook of Childhood Studies, Basingstoke: Palgrave Macmillan, 78–95
166 References
Lansdown, G. (1994) Children’s Rights, in Mayall, B. (ed) Children’s Childhoods
Experienced and Observed, London: Falmer Press, 33–42
Lawler, S. (2008) Stories in the Social World, in Pickering, M. (ed) Research in
Cultural Studies, Edinburgh: Edinburgh University Press, 32–49
Lawler, S. (2014) Identity Sociological Perspectives (2nd ed.), Cambridge:
Polity
Lee, N. (1999) ‘The Challenge of Childhood; Distributions of Childhood’s
Ambiguity in Adult Institutions’ Childhood, vol.6(4), 455–474
Leicester, M. (1999) Disability Voice-Towards an Enabling Education, London:
Jessica Kingsley Publishers
Leinaweaver, J. B. (2008) The Circulation of Children Kinship, Adoption, and
Morality in Andean Peru, Durham: Duke University Press
Lewis, M. (2001) Learning to Listen: Consulting Children and Young People with
Disabilities, London: Save the Children
Lewis, A. (2008) ‘Silence in the Context of “Child Voice” ’ Children and Society,
vol.24, 14–23
Lewis, A. (2010) ‘Silence in the Context of “Child Voice” ’ Children & Society,
vol.24(1), 14–23
Lewis, V., Kellett, M., Robinson, C., Fraser, S. and Ding, S. (2004) The Reality of
Research with Children and Young People, London: Sage
Lewis, A. and Porter, J. (2004) ‘Interviewing Children and Young People with
Learning Disabilities: Guidelines for Researchers and Multi-Professional
Practice’ British Journal of Learning Disabilities, vol.32, 191–197
Livingstone, S. M. (2002) Young People and New Media: Childhood and the
Changing Media Environment, London: Sage
Livingstone, S. (2006) Children’s Privacy Online: Experimenting with Bound-
aries Within and Beyond the Family, in Kraut, R., Brynin, M. and Kiesler,
S. (eds.) Computers, Phones, and the Internet: Domesticating Information
Technology, Oxford: Oxford University Press, 145–167
Livingstone, S. (2008) ‘Taking Risky Opportunities in Youthful Content Cre-
ation: Teenagers’ Use of Social Networking Sites for Intimacy, Privacy and
Self-expression’ New Media & Society, vol.10(3), 393–411
Livingstone, S. and Haddon, L. (2009) EU Kids Online: Final Report.
Deliverable D6.5 for the EC Safer Internet plus programme, Available at:
http://eprints.lse.ac.uk/24372/ [Accessed: 9th Jan 2015]
Lloyd Smith, M. and Tarr, J. (2000) Researching Children’s Perspectives: A Soci-
ological Dimension, in Lewis, A. and Lindsay, G. (eds.) Researching Children’s
Perspectives, Buckingham: Open University Press, 59–70
Luff, D. (1999) ‘Doing Social Research: Issues and Dilemmas’ Sociology,
vol.33(4), 687–703
Lukes, S. (2005) Power: A Radical View, Basingstoke: Palgrave Macmillan
Lull, J. (1985) Ethnographic Studies of Broadcast Media Audiences, in
Dominick, J. and Fletcher, J. (eds.) Broadcasting Research Methods, Boston,
MA: Allyn & Bacon, 80–87
Mackenzie, C. and Stoljar, N. (eds.) (2000) Relational Autonomy: Feminist Per-
spectives on Autonomy, Agency and the Social Self, Oxford: Oxford University
Press
References 167
MacNaughton, G., Hughes, P. and Smith, K. (2007) ‘Young Children’s Rights
and Public Policy: Practices and Possibilities for Citizenship in the Early
Years’ Children and Society, vol.21, 458–469
Madden, R. (2010) Being Ethnographic A Guide to the Theory and Practice of
Ethnography, London: Sage
Madge, C. (2007) ‘Developing a Geographers Agenda for Online Research
Ethics’ Progress in Human Geography, vol.31, 654–674
Madison, D. S. (1988) Performance, Personal Narratives, and the Politics of
Possibility’, in Dailey, S. J. (ed) The Future of Performance Studies: Visions and
Revisions, Washington: National Communication Association, 276–286
Madison, D. S. (2005) Critical Ethnography Method, Ethics and Performance,
London: Sage
Mahon, A., Glendinning, C., Clarke, K. and Craig, G. (1996) ‘Researching
Children: Methods and Ethics’ Children and Society, vol.10, 145–154
Malaguzzi, L. (1993) History, Ideas and Basic Philosophy: An Interview with
Lella Gandini, in Edwards, C., Gandini, L. and Foreman, G. (eds.) The Hun-
dred Languages of Children: Reggio Emilia Approach – Advanced Reflections,
Greenwich: Alex Publishing Corporation, 49–98
Mandelbaum, J. (1978) ‘Assigning Responsibility in Conversational
Storytelling: The Interactional Construction of Reality’ Text, vol.13(2),
247–266
Mandell, N. (1991) The Least Adult Role in Studying Children, in Waksler,
F. C. (ed) Studying the Social Worlds of Children: Sociological Readings, London:
Falmer Press, 38–59
Mann, C. and Stewart, F. (2000) Internet Communication and Qualiltative
Research: A Handbook for Researching Online, London: Sage
Marchant, R. and Crisp, A. (2001) What We Think: Views of Children and Young
People in Southwark, Brighton: Triangle
Masson, J. (2000) Researching Children’s Perspectives: Legal Issues, in Lewis,
A. and Lindsay, G. (eds.) Researching Children’s Perspectives, Buckingham:
Open University Press, 34–45
Mauthner, M., Birch. M., Jessop, J. and Miller, T. (2002) Ethics in Qualitative
Research, London: Sage
Mauthner, N. S. and Doucet, A. (2003) ‘Reflexive Accounts and Accounts of
Reflexivity in Qualitative Data Analysis’ Sociology, vol.37(3), 413–431
Mayall, B. (2000a) Conversations with Children: Working with Generational
Issues, in Christensen, P. and James, A. (eds.) Advocating for Children:
International Perspectives on Children’s Rights, London: Falmer Press, 120–135
Mayall, B. (2000b) Conversations with Children: Working with Genera-
tional Issues, in Christensen, P. and James, A. (eds.) Research with Children:
Perspectives and Practices, London: Routledge, 109–124
Mayall, B. (2002) Towards a Sociology of Childhood. Thinking from Children’s
Lives, Maidenhead: Open University Press
Mayall, B. (2008) Conversations with Children: Working with Genera-
tional Issues, in Christensen, P. and James, A. (eds.) Research with Chil-
dren: Perspectives and Practices, (2nd ed.), London, Routledge Falmer,
109–122
168 References
McDowell, L., Ray, K., Perrons, D., Pagan, C. and Ward, K. (2005) ‘Women’s
Paid Work and Moral Economies of Care’ Social and Cultural Geography,
vol.6, 219–235
McKie, L. and Callan, S. (2012) Understanding Families, London: Sage
Mead, G. H. (1974) Mind, Self and Society (2nd ed.), Chicago: University of
Chicago Press
Miller, S. (2000) ‘Researching Children: Issues Arising from a Phenomenolog-
ical Study with Children Who Have Diabetes Mellitus’ Journal of Advanced
Nursing, vol.31, 1228–1234
Miller, J. and Glassner, B. (1997) The “Inside” and the “Outside”: Finding Real-
ities in Interviews, in Silverman, D. (ed) Qualitative Research; Theory, Method
and Practice, London: Sage, 99–112
Mishler, E. G. (1986) Research Interviewing: Context and Narrative, Cambridge,
MA: Harvard University Press
Moeller, S. (1998) Compassion Fatigue: How the Media Sell Disease, Famine, War
and Death, New York: Routledge
Monaghan, L., O’Dwyer, M. and Gabe, J. (2012) ‘Seeking University Research
Ethics Committee approval: The Emotional Vicissitudes of a “Rationalised”
Process’ International Journal of Social Research Methodology, vol.16(1), 65–80
Montgomery, H. (2007) ‘Working with Child Prostitutes in Thailand: Prob-
lems of Practice and Interpretation’ Childhood, vol.14(4), 415–430
Morris, J. (1998) The Experiences of Disabled Children and Young People Living
Away from Their Families, London: The Who Cares? Trust
Morris, J. (2003) ‘Including all Children. Finding out the Experiences of Chil-
dren with Communication and/or Cognitive Impairments’ Children and
Society, vol.17(5), 337–348
Morris, L. (2006) A Foundation for Rights or Theories of Practice?, in Morris,
L. (ed) Rights: Sociological Perspectives, London: Routledge, 240–252
Mullin, A. (2014) ‘Children, Paternalism and the Development of Autonomy’
Ethic Theory Moral Practice, vol.17, 413–426
Murphy, E. (n.d.) Images of Childhood in Mothers Accounts
of Childrearing, Nottingham University, Available at: http://
eprints.nottingham.ac.uk/775/1/Images_of_Childhood_in_Mothers’
_Accounts_of_Contemporary_Childrearing.pdf [Accessed: 30th July 2012]
Noelle-Neuman, E. (1984) The Spiral of Silence: Public Opinion, Our Social Skin,
Chicago: University of Chicago Press
Oakley, A. (1981) Interviewing Women: A Contradiction in Terms, in Roberts,
H. (ed) Doing Feminist Research, London: Routledge, 30–61
Oakley, A. (2007) Fracture: Adventures of a Broken Body, Bristol: The Policy Press
Oliver, M. (1992) ‘Changing the Social Relations of Research Production’
Disability, Handicap and Society, vol.7, 101–114
Oliver, M. (1996) Understanding Disability: From Theory to Practice, Basingstoke:
Macmillan
Papacharissi, Z. (2009) ‘The Virtual Geographies of Social Networks: A Com-
parative Analysis of Facebook, LinkedIn and ASmallWorld’ New Media and
Society, vol.11(1–2), 199–220
References 169
Parnell, R. and Patsarika, M. (2011) ‘Young People’s Participation in School
Design: Exploring Diversity and Power in a UK Governmental Policy-
Case Study’ Children’s Geographies, vol.9(3–4), 457–475
Parry, G. (1991) Conclusion: Paths to Citizenship, in Vogel, U. and Moran,
M. (eds.) Frontiers of Citizenship, London: Macmillan, 166–201
Parton, N. (2006) ‘ “Every Child Matters” the Shift to Prevention Whilst
Strengthening Protection in Children’s Services in England’ Children & Youth
Services Review, vol.28(2), 976–992
Patai, D. (1991) U.S. Academics and Third World Women: Is Ethical Research
Possible?, in Gluck, S. B. and Patai, D. (eds.) Women’s Words: The Feminist
Practice of Oral History, New York: Routledge and Kegan Paul
Paterson, K. and Hughes, B. (1999) ‘Disability Studies and Phenomenol-
ogy: The Carnal Politics of Everyday Life’ Disability and Society, vol.14(5),
597–611
Phoenix, A. (2008) Analysing Narrative Contexts, in Andrews, M., Squire, C.
and Tamboukou, M. (eds.) Doing Narrative Research, London: Sage, 64–78
Pike, J. (2008) ‘Foucault, Space and the Primary School Dining Room’ Chil-
dren’s Geographies, vol.6(4), 413–422
Pink, S. (2012) Situating Everyday Life, London: Sage
Pittenger, D. (2003) ‘Internet Research: An Opportunity to Revisit Classic
Ethical Problems in Behavioural Research’ Ethics and Behaviour, vol.13(1),
45–60
Plummer, K. (2001) Documents of Life 2: An Invitation to Critical Humanism (2nd
ed.), London: Sage
Poindexter, C. C. (2003) ‘The Ubiquity of Ambiguity in Research Interviewing:
An Exemplar’ Qualitative Social Work, vol.2(4), 383–409
Pole, C. (2007) ‘Researching Children and Fashion: An Embodied
Ethnography’ Childhood, vol.14(1), 67–84
Potter, J. (1996) Representing Reality: Discourse, Rhetoric and Social Construction,
London: Sage
Potter, J. and Wetherell, M. (1987) Discourse and Social Psychology: Beyond
Attitudes and Behaviour, London: Sage
Potter, C. and Whittaker, C. (2001) Enabling Communication in Children with
Autism, London: Jessica Kingsley
Powell, M. A. and Smith, A. B. (2010) ‘Ethical Guidelines for Research
with Children: A Review of Current Research Documentation in New
Zealand’ Kotuitui: New Zealand Journal of Social Sciences Online, vol.1(2),
125–138
Prentsky, M. (2001) ‘Digital Natives, Digital Immigrants Part 1’ On the Horizon,
vol.9(5), 1–6
Priestly, M. (1998a) ‘Childhood, Disability and Disabled Childhoods – Agen-
das for Research’ Childhood – A Global Journal of Child Research, vol.5(2),
207–223
Priestley, M. (1998) ‘Constructions and Creations: Idealism, Materialism and
Disability Theory’ Disability and Society, vol.13(1), 75–94
Prout, A. (2005) The Future of Childhood, London: Falmer Press
170 References
Prout, A. and Hallett, C. (2003) Introduction, in Prout, A. and Hallett, C. (eds.)
Hearing the Voices of Children: Social Policy for a New Century, London:
Routledge Falmer, 1–8
Punch, S. (2002) ‘Research with Children: The Same or Different from
Research with Adults?’ Childhood, vol.9(3), 321–341
Qvortrup, J., Bardy, M., Sgritta, G. and Wintersberger, H. (eds.) (1994)
Childhood Matters, Aldershot: Avebury
Rabiee, P., Sloper, P. and Beresford, B. (2005) ‘Desired Outcomes for Chil-
dren and Young People with Complex Health Care Needs, and Children
Who Do Not Use Speech for Communication’ Health and Social Care in the
Community, vol.13(5), 478–487
Reinharz. S. (1979) On Becoming a Social Scientist, San Francisco: Jossey-Bass
Riessman, C. K. (1994) ‘Subjectivity Matters: The Positioned Investigator’, in
Riessman, C. K. (ed) Qualitative Studies in Social Work Research, Thousand
Oaks: Sage, 139–52
Richards, S. (2012a) ‘ “Why are we not Allowed to Comfort Children when
they’ve been Told off?”: The Social Positioning of Children and Their Capac-
ity for Ethical Agency’ Childhood Remixed, vol.1, 48–54, University Campus
Suffolk
Richards, S. (2012b) What the Map Cuts up the Story Cuts Across: Narra-
tives of Belonging in Intercountry Adoption, in Simmonds, J. and Phoenix,
A. (eds.) Multiculturalism, Identity and Family Placement, in Adoption &
Fostering, vol.36(3–4), 104–111
Richards, S. (2013) ‘Stories of Paper and Blood: Narratives of Belonging in
Families with Daughters Adopted from China’. PhD dissertation, Institute
of Education, University of London
Richards, S. (2014) ‘HCIA Implementation and the Best Interests of the Child’
ISS Working Paper Series/General Series, vol.597, 1–18. International Insti-
tute of Social Studies of Erasmus University (ISS). Retrieved from http://hdl
.handle.net/1765/77407
Ricoeur, P. (1991) From Text to Action, Illinois: Northwestern University Press
Ringrose, J. (2011) Beyond Discourse? Using Deleuze and Guattari’s
Schizoanalysis to Explore Affective Assemblages, Heterosexually Striated
Space and Lines of Flight Online and at School, Educational Philosophy and
Theory, vol.43(6), 598–618
Ringrose, J. and Eriksson Barajas, K. (2011) Gendered Risks and Oppor-
tunities? Exploring Teen Girls’ Digital Sexual Identity in Post Feminist
Media Contexts, International Journal of Media and Cultural Politics, vol.7(2),
121–138
Richman, A. (2007) The Outsider Lurking Online; Adults Researching Youth
Cybercultures, in Best, A. L. (ed) Representing Youth: Methodological Issues in
Critical Youth Studies, New York: New York University Press, 182–202
Ristock, J. L. and Pennell, J. (1996) Community Research as Empowerment:
Feminist Links, Postmodern Interpretations, Toronto: Oxford University
Press
References 171
Robinson, K. (2013) Innocence, Knowledge and the Construction of Childhood: The
Contradictory Nature of Sexuality and Censorship in Children’s Contemporary
Lives, London: Routledge
Rodgers, J. (1999) ‘Trying to Get It Right: Undertaking Research Involv-
ing People with Learning Difficulties’ Disability and Society, vol.14(4),
421–433
Rojiani, R. H. (1994) Disparities in the Social Construction of Long-Term Care,
in Riessman, C. K. (ed) Qualitative Studies in Social Work, California: Sage,
139–152
Rose, N. (1999) Powers of Freedom: Reframing Political Thought, Cambridge:
Cambridge University Press
Roseneil, S. (1993) Greenham Revisited: Researching Myself and My Sisters, in
Hobbs, D. and May, T. (eds.) Interpreting the Field: Accounts of Ethnography,
Oxford: Clarendon Press, 177–208
Sacks, H. (1974) An Analysis of the Course of a Joke’s Telling in Conversa-
tion, in Bauman, R. and Sherzer, J. F. (eds.) Explorations in the Ethnography of
Speaking, Cambridge: Cambridge University Press, 337–353
Sacks, H. (1992) Lectures on Conversation, Volumes I and II, in by
Jefferson, G., with Introduction by Schegloff, E. A., Oxford: Blackwell
Sandbaek, M. (1999) ‘Adult Images of Childhood and Research on
Client Children’ International Journal of Social Research Methodology, vol.2,
191–202
Saunders, J. S. (2008) ‘Spinning Sympathy: Orphan Girl Novels and the
Sentimental Tradition’ Children’s Literature Association Quarterly, vol.33(1),
41–62
Scott, J. (2000) Children as Respondents: The Challenge for Quantitative
Methods, in Christensen, P. and James, A. (eds.) Research with Children:
Perspectives and Practices, London: Falmer Press, 36–61
Schutz, A. (1962) ‘Choosing Among Projects of Action’, in Natanson, M. (ed)
Collected Papers, vol.1 The Problem of Social Reality, The Hague: Martinus
Nijhoff, 67–94
Segal, L. (1999) Why Feminism? Gender, Psychology, Politics, Cambridge: Polity
Shah, S. (2006) ‘Sharing the World: The Researcher and the Researched’
Qualitative Research, vol.6(2), 207–220
Shakespeare, T. (1996) Disability, Identity, Difference, in Barnes, C. and
Mercer, G. (eds.) Exploring the Divide; Illness and Disability, Leeds: Disability
Press, 94–113
Shakespeare, T. (2000) Help, Birmingham: Venture Press
Shakespeare, T. (2006) Disability Rights and Wrongs, Oxon: Routledge
Short, G. (1999) Children’s Grasp of Controversial Issues, in Woodhead, M.,
Faulkner, D. and Littleton, K. (eds.) Making Sense of Social Development,
London: Routledge, 153–169
Sieber, J. and Stanley, B. (1988) ‘Ethical and Professional Dimensions of
Socially Sensitive Research’ American Psychologist, vol.42, 49–55
Sikes, P. and Piper, H. (2010) ‘Ethical Research, Academic Freedom and the
Role of Ethics Committees and Review Procedures in Educational Research’
172 References
International Journal of Research and Method in Education, Special Issue: Ethics
and Academic Freedom in Educational Research, vol.33(3), 205–213
Silverman, D. and Gubrium, J. (1989) Introduction, in Silverman, D. and
Gubrium, J. (eds.) The Politics of Field Research: Sociology Beyond Enlighten-
ment, London: Sage, 1–13
Simons, K., Booth, T. and Booth, W. (1989) ‘Speaking Out: User Studies and
People with Learning Disabilities’ Advances in Psychiatric Treatment, vol.1,
207–213
Simons, H. and Usher, R. (2002) Situated Ethics in Educational Research,
New York: Routledge
Sin, C. H. (2005) ‘Seeking Informed Consent: Reflections on Research Practice’
Sociology, vol.39(2), 277–294
Skelton, T. (2008) ‘Research with Children and Young Children: Exploring
the Tensions Between Ethics, Competence and Participation’ Children’s
Geographies, vol.6(1), 21–36
Shah, S. (2006) ‘Sharing the World: The Researcher and the Researched’
Qualitative Research, vol.6(2), 207–220
Silverman, D. (1993) ‘Beginning Research’. Interpreting Qualitative Data. Methods
for Analysing Talk, Text and Interaction, London: Sage
Silverman, D. (1998) Harvey Sacks: Social Science & Conversation Analysis,
Cambridge: polity Press
Slote, M. (2007) The Ethics of Care and Empathy, Abingdon: Routledge
Smart, C., Neale, B. and Wade, A. (2001) The Changing Experience of Childhood:
Families and Divorce, Cambridge: Polity Press
Smith, F. and Barker, J. (2000) ‘Contested Spaces: Children’s Experiences of
out of School Care in England and Wales’ Childhood, vol.7(3), 315–333
Smith, Mark J. (2005) Reinventing the Social Sciences: Towards a Post Disciplinary
Future, London, UK: Sage Publications
Smolin, D. M. (2006) ‘Child Laundering: How the Intercountry Adoption
System Legitimizes and Incentivizes the Practices of Buying, Trafficking,
Kidnapping, and Stealing Children’ Wayne Law Review, vol.52(1), 113–200,
Available at: http://wors.bepress.com/david_smolin/1 [Accessed: 9th Jan-
uary 2015]
Smyres, K. (1999) ‘Virtual Corporeality: Adolescent Girls and their Bodies
in Cyberspace’ Research Methodology Online, Available at: http://www.socio
.demon.co.uk/magazine/6/smyres.html [Accessed: 21st May 2015]
Song, M. and Parker, D. (1995) ‘Commonality, Difference and the Dynamics
of Disclosure in In-Depth Interviewing’ Sociology, vol.29(2), 241–256
Spriggs, M. (2010) Understanding Consent in Research Involving Children: The
Ethical Issues, The Royal Children’s Hospital, Melbourne, Available at: www
.mori.edu.au/media/62539/handbook.pdf [Accessed: 9th January 2015]
Spyrou, S. (2011) ‘The Limits to Children’s Voices: From Authenticity to
Critical Reflexive Representation’ Childhood, vol.18(2), 151–165
Stacey, J. (1996) Can There Be a Feminist Ethnography?, in Gottfried, H. (ed)
Feminism and Social Change: Bridging Theory and Practice, Illinois: University
of Illinois Press, 88–101
References 173
Stainton Rogers, W. (2001) Constructing Childhood: Constructing Child Con-
cern, in Foley, P., Roche, J. and Tucker, S. (eds.), Childhood in Society:
Contemporary Theory, Policy and Practice, London: Basingstoke, 26–33
Stalker, K. (2012) ‘Researching the Lives of Disabled Children’ Children and
Society, vol.26, 173–180
Stalker, K. and Connors, C. (2003) ‘Communicating with Disabled Children’
Adoption and Fostering, vol.27(1), 26–35
Stanley, L. and Wise, S. (1993) Breaking Out Again: Feminist Ontology and
Epistemology, London: Routledge
Stanley, L. and Wise, S. (2010) ‘The ESRC’s 2010 Framework for Ethics: Fit for
Research Purpose?’ Sociological Research Online, 15(4), 12, Available at http:
//www.socresonline.org.uk/15/4/12.html
Stern, S. (2004) Studying Adolescents Online: A Consideration of Ethical
Issues, in Buchanan, E. (ed) Readings in Virtual Research Ethics: Issues and
Controversies, Hershey, PA: Ideal Group, 274–287
Stones, R. (2005) Structuration Theory, Basingstoke: Palgrave MacMillan
Strauss, A. L. (1977) Mirrors and Masks: The Search for Identity, New Jersey:
Transaction Publishers
Tapscott, D. (1998) Growing Up Digital. The Rise of the Net Generation, New York:
McGraw Hill
Taylor, L. (1998) ‘Confidentiality: Competing Principles, Inevitable
Dilemmas’ Journal of Education and Psychological Consultation, vol.9(3),
267–275
Taylor, J. (2011) ‘The Intimate Insider: Negotiating the Ethics of Friendship
when Doing Insider Research’ Qualitative Research, vol.11(3), 3–22
The National Children’s Bureau (2003) Guidelines for Research, Available
at: www.ncb.org.uk/research/research guidelines.pdf [Accessed: 21st August
2013]
Thomas, N. and O’Kane, C. (1998) ‘The Ethics of Participatory Research with
Children’ Children and Society, vol.12, 336–348
Thomas, N. and O’Kane, C. (2000) ‘Discovering What Children Think: Con-
nections Between Research and Practice’ British Journal of Social Work, vol.30,
819–835
Thorne, B. (1980) ‘ “You Still Takin’ Notes?” Fieldwork and Problems of
Informed Consent’ Social Problems, vol.27(3), 284–297
Thornton, S. (1995) Club Cultures: Music, Media and Subcultural Capital Oxford:
Polity
Tilly, C. (1994) ‘History and Sociological Imagining’ Tocqueville Review, vol.15,
57–74
Triseliotis, J. P., Shireman, J. F. and Hundleby, M. (1997) Adoption: Theory, Policy
and Practice, London: Cassel
Tronto, J. C. (2009) Moral Boundaries a Political Argument for an Ethic of Care,
New York: Routledge
Twine, F. (1994) Citizenship and Social Rights, London: Sage
Twycross, A. (2009) ‘An Interprofessional Approach to the Ethics of Undertak-
ing Research with Children’ Nurse Researcher, vol.16(3), 7–20
174 References
United Nations (1989) Conventions on the Rights of the Child, United Nations:
Geneva
Uprichard, E. (2008) ‘Children as Being and Becoming: Children, Childhood
and Temporality’ Children and Society, vol.22(4), 303–313
Uprichard, E. (2010) ‘Questioning Research with Children: Discrepancy
Between Theory and Practice?’ Children & Society, vol.24(1), 3–13
Veale, A. (2005) Creative Methodologies in Participatory Research with Chil-
dren, in Greene, S. and Hogan, D. (eds.) Researching Children’s Experience,
London: Sage, 253–269
Walkerdine, V. (1989) ‘Femininity as Performance’ Oxford Review of Education,
vol.15 (3), 267–279
Walkerdine, V. (1999) ‘Violent Boys and Precocious Girls: Regulating Child-
hood at the End of the Millennium’ Contemporary Issues in Early Childhood,
vol.1(1), 3–23
Wall, L. (2010) Ethics in Light of Childhood, Washington: Georgetown Univer-
sity Press
Waller, T. (2010) ‘ “Let’s Throw that Big Stick in the River”: An Exploration of
Gender in the Construction of Shared Narratives around Outdoor Spaces’
European Early Childhood Education Research, vol.18(4), 527–542
Waller, T. and Bitou, A. (2011) ‘Research with Children: Three Challenges
for Participatory Research in Early Childhood’ European Early Childhood
Education Research, vol.19(1), 5–20
Walter J. K. and Ross L. F. (2014) ‘Relational Autonomy: Moving Beyond the
Limits of Isolated Individualism’ Pediatrics, vol.133(1), 16–23
Wang, X. (2012) ‘The Construction of Researcher-Researched Relationships
in School Ethnography: Doing Research, Participating in the Field and
Reflecting on Ethical Dilemmas’ International Journal of Qualitative Studies
in Education, vol.26(7), 763–779
Ward, L. (1997) Seen and Heard: Involving Disabled Children and Young People in
Research and Development Projects, York: York Publishing
Watson, N. (2003) Daily Denials: The Routinisation of Oppression and Resis-
tance, in Riddell, S. and Watson, N. (eds.) Disability, Culture and Identity:
Introduction, Harlow: Pearson Education Ltd, 34–51
Watson, N., Shakespeare, T., Cunningham-Burley, S. and Barnes, C. (1999)
Life as a Disabled Child: A Qualitative Study of Young People’s Experiences
and Perspectives, Available at: www.esrcsocietytoday.ac.uk [Accessed: 10th
November 2013]
Westcott, H. and Littleton, K. S. (2005) Exploring Meaning in Interviews with
Children, in Green, S. and Hogan, D. (eds.) Researching Children’s Experience:
Approaches and Methods, London: Sage, 141–157
Wheatley, E. (1994) ‘Dances with Feminist: Truth, Dares and Ethnographic
Stares’ Women’s Studies International Forum, vol.17(4), 421–423
White, H. (1980) ‘The Value of Narrativity in the Representation of Reality’
Critical Inquiry, vol.7(1), 5–27
Widdicombe, S. and Wooffitt, R. (1995) The Language of Youth Subcultures:
Social Identity in Action, London: Harvester Wheatsheaf
References 175
Wright, K. (2008) ‘Researching the Views of Pupils with Multiple and Complex
Needs. Is It Worth Doing and Whose Interests Are Served by It?’ Support for
Learning, vol.23(1), 32–40
Young, L. and Barrett, H. (2001) ‘Ethics and Participation: Reflections on
Research with Street Children’ Ethics, Place and Environment, vol.4(2),
130–134
Yee, W. C. and Andrews, J. (2006) ‘Professional Researcher or a Good Guest?
Ethical Dilemmas Involved in Researching Children and Families in the
Home Setting’ Education Review, vol.58(4), 397–413
Yngvesson, B. (2010) Belonging in an Adopted World, Chicago: University of
Chicago Press
Ytterhus, B. (2012) ‘Everyday Segregation Amongst Disabled Children and
Their Peers: A Qualitative Longitudinal Study in Norway’ Children and
Society, vol.26, 203–213
Zelizer, V. A. (1985) Pricing the Priceless Child: The Changing Social Value of
Children, New York: Basic Books
Zignon, J. (2008) Morality: An Anthropological Perspective, Oxford: Berg
Zignon, J. (2009) ‘Within a Range of Possibilities: Morality and Ethics in Social
Life’ Ethnos, vol.74, 251–276
Index
AACS, see Augmentative and autonomy, illusion of, 78–9
Alternative Communication children’s relational agency,
Systems (AACS) 86–97
Aapola, S., 54 constructing child in social
Aarsan, P. A., 34 research, 79–81
Abebe, T., 32 and place in ethics, 81–6
Abell, J., 105, 106, 109, 110, 116 relational concept of, 86
academic credibility, 114 relational model of, 87
active interviewing, 108 teacher’s roles, 89
adoptive category, 119
intercountry, 117 Badham, B., 70
membership of, 118–19 Baker, C. D., 103, 104, 107, 109,
Agar, M. H., 124 115
agency, divergent understandings Barker, J., 21, 32, 33, 45, 49, 51, 60
of, 63 Barn, R., 39, 142
Alderson, P., 18, 30, 47, 62, 85 Barrett, H., 33, 60
Aldridge, J., 142 Barthes, R., 1
Allan, G., 34 Bartholet, E., 132, 133
Alleyne, B., 56 Basset, E., 56
Bebo page, 56
Alvesson, M., 5, 38, 99
Beck, U., 30
Anderson, T., 54
Bell, C. C., 1, 17, 113
Andrews, J., 33, 35, 37
Bell, N., 153
Andrews, M., 4, 5
Benatar, S., 15
Anzil, V., 131
Bessell, S., 32
Archard, D., 16, 18
Bhavnani, K., 54
Aronsson, K., 34 BHPS, see British Household Panel
Asian cultural script, 86 Survey (BHPS)
assent Bird, S. E., 38, 152
alternative of, 152 Bitou, A., 43, 88, 97
defined, 16–17 Bluebond-Langner, M., 62, 132
vs. informed consent, 16–20 Boden, R., 19, 20, 24, 29, 30
Atkinson, P., 100, 114 Borland, M. A., 42
Aubrook, L., 46 borrowed voices, 128
Augmentative and Alternative Bourdieu, P., 122, 124, 126, 130, 134,
Communication Systems 135
(AACS), 9, 47, 64, 115, 127, 135, British Household Panel Survey
139 (BHPS), 17
Aull Davies, C., 101 Brooks, R., 13, 16
autonomous self, 85 Broome, M., 17, 153
176
Index 177
Bruckman, A., 55, 58 classification of, 18
Bruner, J., 34, 141, 142 cognitive foundations of status,
Bryson, M., 53 129
Butler, I., 70 colonization of, 45
complexities of working with, 5
CA, see conversation analysis (CA) constructed innocence of, 28
CACH, see Children Adopted from in contemporary research, 83–4
China (CACH) critical reflexivity and, 74–6
Callan, S., 34 dangers of online activity, 59
Cameron, L., 71 developmental stages of, 80
Carlson, C., 37 dilemmas of researching with, 9
category entitlement, 115–19 disabled, see disabled children
description of, 101–5 discursive constructions of, 16, 54
insider status, 110–15 discursive motifs of, 9
self-disclosure, 105–10 domains for research, 6
specific focus of, 99 dominant attributes of, 147
category membership, 104, 150–1 empowerment of role of, 27
to identify potential respondents, homogenisation of, 128
115 innocence, 6, 7
insider status and, 112, 114 institutionalisation of, 51
mutual, 117 in institutional settings, 46
shared, 113–14 institutional spaces of, 45
Cavet, J., 70 life experiences of, 117
Cerulo, K., 38 micro-level analysis on, 3
chain of consent, 153 moral meanings, development
cherished conceits, 146, 148 of, 18
childcare, forms of, 46 nature of, 26
child-centred perspectives, 19 normative assumptions about, 8
child-centred research, 146 objectification and
adults in, 89 homogenisation of, 81
discussions of, 85 online worlds, risks of, 53
to empower participants, 48–9 over-bureaucratisation of
guidelines for, 19 processes, 4
notions of, 76 in participatory research, 2, 5
philosophies of, 52 in primary school classroom, 48
Children Act 1989, 3, 69 researching with, 2–3
Children Act 2004, 3 scientific and political interests
Children Adopted from China in, 80
(CACH), 105, 113 sexual attraction to, 141
children/childhood in social research, 79–81
active participants in research, 3–4 social studies of, 16, 19
agency and participation of, 147 sociological interest in, 2
in Allison’s study, 128 stories, 1
Article 12 of UNCRC, 3 strength-based perceptions of, 122
assent vs. informed consent, 16–20 structural vulnerability of, 6
attributes of, 146, 148 understandings of, 137
178 Index
children/childhood – continued data collection process, 69
use of technology, 53 David, M., 32, 46, 47, 49
virtual worlds of, 53–60 de Castell, S., 53
voices of, see voices of children Declaration of Helsinki (1964), 14
Children (Northern Ireland) Order Denscombe, M., 46
1995, 3, 69 Denzin, N. K., 37, 40, 107
Christensen, P., 41, 45, 81, 122, 135, DePalma, R., 52
144 Department of Health, 65
Clark, J., 18, 26, 32, 45, 60, 152 dependency, images of, 127–8
Coad, J., 143 DeSchauwer, E., 70
Cockburn, T., 1, 81, 85, 96, 97 Devine, F., 1
Code, L., 82 digital immigrants, 59
Coffey, A., 37–9, 99, 100, 110, 111 digital natives, 59
cognitive ability, 123 Dingwall, R., 9, 15
collectivism, 85, 86 disabled children, 48
Collins, R., 24 Allison’s research conversations
communication with, 63–4
aids, 66, 128–30 awareness of, 71
difficulties, 124 cultural schemas of, 126
forms, 47 general dispositions of, 125–6
Jorge’s personalised method of, nature and context of, 67
126 normative rules of conduct for,
prompted vs. spontaneous, 74 135
skills, 123 participation, 126
social practice of, 135 positive inclusion of, 71
techniques, 137 realities of inclusion, 70–2
community accountability, 114 regularisation of, 48
competencies of children, 15, 24, 46, rights of participation in research,
47, 50, 62, 64, 67, 72, 85, 122, 64–9, 76
124, 135, 136 symbolic representations of, 133–4
Connolly, P., 132, 133 voices of, 65, 68
Connors, C., 62, 71 Dockett, S., 51
Conquergood, D., 37 Dorow, S., 132
contemporary research contexts, 16 Doucet, A., 99, 100, 113, 114, 120
conversation analysis (CA), 103 Douglas, M., 6, 7, 27, 28
Corsaro, W. A., 33, 62 Drew, P., 118
Crapanzano, V., 124 Duchinsky, R., 60
Crisp, A., 66 Dunbar, C., 100, 105, 106
critical ethnography, 37 Dworkin, G., 82
Crompton, R., 34
Crow, G., 34 Economic and Social Research
curriculum based language, 47 Council (ESRC), 14
cyberbullying, 53 Eder, D., 109
educational institutions, 45–53
Danieli, A., 7, 144, 148 distribution of power in, 46
Daniel, P., 46 traditional, 46
Index 179
Edwards, C., 124, 126 Foucault, M., 5, 18, 24, 29–31, 45,
Edwards, R., 16 49, 50, 144, 149
egarement, Foucault’s concept of, 29 concept of egarement, 29
Eglin, P., 107 concept of examination, 18
Elbers, E., 80 concept of governmentality, 24
Elliott, A., 138 notions of power relations, 149
Elliott, J., 152 Frank, A., 1, 153
Emond, R., 131 Frankel, S., 18, 85
Ennew, J., 97, 131 Franklin, A., 70, 126
Epstein, D., 24 Fraser, S., 23, 132
ESRC, see Economic and Social Freund, P. E., 24
Research Council (ESRC) Friedlander, M. L., 131, 133
ethical enthusiasm in childhood
studies, 22–9 Gaber, I., 142
ethical research process, 62 Gailey, C. W., 132
ethical symmetry, principles of, Gallacher, L., 1, 43, 45, 50, 79
122 Gallagher, M., 2, 43, 45, 50, 79
ethics Garfinkel, H., 102
autonomy and place in, 81–6 Garrett, P. M., 81
dominant principle of, 82 Giddens, A., 30, 67, 68, 133–5, 138
forms of, 18 Gillam, L., 33, 150
hybrid relational model of, 56 Gilligan, C., 38, 125
Jessica’s explorations of, 52 Gilligan, T., 29, 141
procedural vs. practice, 24 Glassner, B., 111
vs. reflexivity, 5 Goffman, E., 133, 138
in research process, 1, 19 Goode, S., 23, 24, 29, 85, 141
ethnography, 5, 45 Goodey, C., 85
reflexive, 114 governmentality, Foucault’s concept
virtual, 55 of, 24
Evans, R., 143 grade individuals, 123
examination, Foucault’s concept of, Graue, E. M., 22, 23, 47
18, 30–1 Greenfield, P. M., 82–6
Greig, A., 5
Facebook, 56 Griffith, A., 110
Faulkner, J., 149 group membership, 116
Fawcett, B., 125 Guba, E. G., 34
Feinberg, J., 83 Gubrium, J. F., 19, 99, 100, 105, 108
femininity Guillemin, M., 33, 150
forms of, 43 Gustafsson, U., 46
social constructions of, 38 Guta, A., 18, 24
feminist social researchers, 112
Finch, J., 118 Haavind, H., 41
Fine, M., 101 habitus, 134, 135
Fingerson, L., 109 Bourdieu’s notion of, 124, 135
Fisher, A. P., 27 of children/childhood, 134
Flewitt, R., 19, 52 pre-ordained, 125
180 Index
Hacking, I., 101, 102, 118 informed consent, 3, 57, 152–3
Haddon, L., 59 approach to gaining of, 59
Hallett, C., 2 children’s assent vs., 16–20
Hallowell, N., 18 gaining, 72–4
Hammersley, M., 9, 15, 16, 22, 23, importance of, 46
46, 114 issues of, 46, 58
Haraway, D., 120 negotiations, 47
Harden, J., 34, 43, 48, 87 principles, 16
Harris, A., 52, 54 provision of, 46
Haugen, G. M. D., 62 role in gaining, 52
Heath, S., 1, 33, 46, 49, 51 seeking, 69–70
hegemony of voices, 141–4 innocence, 28, 29, 54
HEIs, see Higher Education of children/childhood, 6, 7
Institutions (HEIs) concept of, 26
Hendrick, H., 19, 81 Instagram, 56
Heritage, John, 103 Institutional Review Boards
Herz, R., 33, 60 (IRBs), 14
Hester, S., 107 intercountry adoption, 8, 108–9, 115
Hewson, C., 58 categories of, 117
Higher Education Institutions IRBs, see Institutional Review Boards
(HEIs), 14 (IRBs)
Hill, E. J., 34, 43
Hine, C., 56 Jackson, A. Y., 145
Hobbs, D., 1 Jacobson, H., 132
Hochschild, A., 23 James, A., 1, 16, 19, 45, 64, 79–81,
Hockey, J., 124 121, 122, 124, 144
Högbacka, R., 141 Jayyusi, L., 117
Holland, P., 31 Jefferson, T., 19
Holloway, S., 32, 53 Jenks, C., 34, 53, 79
Hollway, W., 19 Johnson, K., 27
Holmes, J., 38, 151 Jordan, A. B., 33, 35–7, 40, 107, 109,
Holstein, J. A., 99, 100, 105, 108 151
Howarth, C., 100, 103, 104, 112,
114 Kanuka, H., 54
Hudson, J. M., 55 Kehily, M. J., 6, 26
Hughes, P., 124 Kellett, M., 41, 81
human agency, 86 Kirschner, D., 132, 133
Huxley, P., 16 Kirton, D., 39, 141
Hydén, M., 27, 143, 144 Kitchin, R., 53, 143
Komulainen, S., 95
images of dependency, 127–8 Krieger, S., 112
Imrie, R., 124, 126
in-control model, 82, 83 Lahman, M. K. E., 18, 31, 152
individualised self, 85 Laidlaw, J., 5
individualism, 86 Lange, A., 81
rights of, 83 Lansdown, G., 23
Index 181
Lawler, S., 4, 37, 151 Mierendorff, J., 81
Lee, N., 26, 27, 78 Miller, J., 111
Leicester, M., 111, 112 Miller, S., 23
Leinaweaver, J. B., 131 minor attracted adults, 141
lesbian, gay, bisexual, transgender Mishler, E. G., 37, 147
(LGBT), 52 Moeller, S., 131
Lewis, A., 5, 34, 42, 43, 71, 121, 150 Monaghan, L., 15, 21, 23, 24, 30,
Lewis, M., 66 104
Lewis, V., 87 Montgomery, H., 2, 13, 86
LGBT, see lesbian, gay, bisexual, Montgomery, J., 62
transgender (LGBT) moral codes, 5–6
Lincoln, Y. S., 34 moral reasoning, 18
Littleton, K. S., 73 Morris, J., 70, 71
living room as research field, 35–7 Morris, L., 77
Livingstone, S. M., 34, 57, 59 Morrow, V., 18, 30
Lloyd Smith, M., 70 Mullin, A., 82
Locke, A., 18 Murphey, J., 71
Luff, D., 39 Murphy, E., 53
Lukes, S., 18
Lull, J., 36, 107 narrative research, defined, 4
‘lurking’ in online spaces, 25, National Children’s Bureau, 70
55, 57 The National Children’s Bureau, 70
Newby, H., 1, 113
Mackenzie, C., 87 Noelle-Neuman, E., 13
MacNaughton, G., 62 normative sanctioning, modes of,
Madden, R., 98, 114 134
Madge, C., 55, 56, 58 Nuremberg Code (1947), 14
Madison, D. S., 5
Mahon, A., 72 Oakley, A., 1, 38, 99, 106, 107, 109,
Malaguzzi, L., 87 111, 112, 151
Mandelbaum, J., 116 O’Dwyer, M., 22, 23
Mandell, N., 124 O’Kane, C., 144
Mann, C., 58 Oliver, M., 71, 143, 144
Marchant, R., 66 ‘one size fits all’ approach, 72
Masson, J., 21 online activity, dangers of, 59
Mauthner, M., 16, 19 online content, defined,
Mauthner, N. S., 99, 100, 113, 114, 55–6
120 online spaces, 53, 57
Mayall, B., 19, 34, 35, 63, 76, 96, 97, for identity construction, 56
152, 153 ‘lurking’ in, 55
May, T., 1 public and private in, 56–7
McDowell, L., 46 O’Riordan, K., 56
McKie, L., 34 orphans
Mead, G. H., 138 defined, 131
membership category, 108–9 role of, 131–2
analysis of, 103 othering of children, 18, 45, 101
182 Index
parental consent, 17, 57, 58 Rabiee, P., 71
Parker, D., 100, 106 RECs, see Research Ethics
Parnell, R., 51 Committees (RECs)
Parry, G., 78, 85 reflexive re-positioning, 137–41
participatory methods reflexivity, 6, 10, 33, 60, 67
canonical narratives of, 2 analysis of, 135
dominance of, 153 concept, 138
importance of, 48 critical, 74–6, 112
participatory research, children and ethics, 5
in, 2, 5, 43, 88, 90, 95, 96, Reinharz. S., 110
154 relational agency, children, 86–97
Parton, N., 80 research
Patai, D., 115
Allison’s, 9, 20, 47, 48, 51, 52, 63,
Paterson, K., 124
66–8, 70, 72–6, 115, 122–30,
Patsarika, M., 51
133–4, 136–41, 153
Pennell, J., 112
‘child-centred’ approaches to, 148
persistent illusions, 2, 35, 148
with children/young people, see
Phoenix, A., 101, 104, 106, 115–17,
research with children/young
143, 144
people
physical spaces, 56, 62
critical reflexivity and children in,
Pike, J., 45
74–6
Pink, S., 32
with disabled children, see
Piper, H., 15
disabled children
Pittenger, D., 54, 55
epistemological richness of, 100–1
places for research, spaces and, see
spaces and places for research fluidity of power in, 150
Plummer, K., 1, 29, 33, 113, 141 involvement of children in, 146
Poindexter, C. C., 100, 101, 104, Jessica’s, 9, 16–17, 20, 21, 25, 52,
105, 109 54–8, 68, 125, 141
Pole, C., 13, 32, 33, 35, 100 rights of participation in, see rights
Porter, 71 of participation in research
Potter, C., 74 role of insiders in, 111
Potter, J., 102, 104, 115, 116 Sarah’s, 8, 24, 25, 27, 35, 36–41,
Powell, M. A., 17, 23, 153 43, 49, 52, 69, 84, 87–9, 93,
powerful voices, 134–6 95–7, 104–18, 128, 130, 131,
Prentsky, M., 59 133, 141, 142, 147, 151, 152
privileging, voices of ‘selfhood’ in, 89
children/childhood, 136–7 validity and neutrality in, 100
Prout, A., 1, 2, 41, 80, 81, 121, researcher, 5, 7, 11–30, see also
122 research
public space, 53, 55 identity, performance of, 37
Punch, S., 13, 73, 80 issues faced by, 8
Purity and Danger (1991), 28 mistrust of, 23
motives of, 10
quantitative research, 154 self-disclosure, 107
Qvortrup, J., 51, 79, 80 shapes and influences, 99
Index 183
Research Ethics Committees (RECs), Schutz, A., 89
9, 14–16 Scott, J., 17
alternative and competing Segal, L., 2, 13, 35, 83, 146, 148
agendas of, 24 self
application, 24–5 autonomous, 85
contradictory demands of, 20 concept of, 138
decision making, 19 construction of, 138
university, 21 individualised, 85
research with children/young people ‘regrooving’ of, 139
assent vs. informed consent, self-disclosure, 101, 105–10, 116,
16–20 151
in educational spaces, 51 and category entitlement,
in home, 33–5 110
in living room, 35–7 role and use of, 105, 108–9
Sarah’s, 35–41 selfhood, rights of, 83
sensitive topics on, 15 self-improvement, continuous
spaces and places for, see spaces process of, 82
and places for research self-protection, 87
taboos and boundaries, 22–9
semi-structured interviews, 138
third-party involvement in, 20–2
sensitive topics, 26–7
Richards, D., 153
sexual cultures, 26
Richards, S., 17, 40, 41, 43, 84, 87,
sexualisation, 26
90, 95, 97, 108, 111
Shah, S., 110–12, 117
Richman, A., 55
Shakespeare, T., 66, 127
Ricoeur, P., 93
Short, G., 19
Riessman, C. K., 110
rights of participation in research, Sieber, J., 26
64–9 Sikes, P., 15
disabled children, 64–9 silences as relational strategies,
informed consent, 69–70, 72–4 41–4
realities of inclusion, 70–2 Silverman, D., 19, 100, 103
Ringrose, J., 54, 56–9 Simons, H., 33
‘risk-based’ approach to children, Simons, K., 73
141 Sin, C. H., 100, 107–9, 152
Ristock, J. L., 112 Skelton, T., 15, 17, 19, 30, 153
Roberts, H., 113 Sköldberg, K., 5, 99
Robinson, K., 6, 26 Sloper, P., 70, 126
Rodgers, J., 72 Slote, M., 96
Rojiani, R. H., 105 Smart, C., 62
Rose, N., 82, 85 Smith, A. B., 17, 23, 153
Roseneil, S., 112 Smith, F., 45, 49
Ross, L. F., 82 Smith, M. J., 87
Smolin, D. M., 39
Sacks, H., 102, 103, 116, 118 Smyres, K., 55, 57
Sandbaek, M., 23 social identity, 130
Saunders, J. S., 131 Song, M., 100, 106
184 Index
spaces and places for research, UK Human Rights Act, 69
32–3 unconventional voices, children
educational institutions, with, 63, 122, 127, 137, 145
45–53 UNCRC, see United Conventions of
in home, 33–5 the Rights of the Child (UNCRC)
living room as research field, UNICEF, core principles of, 35
35–7 United Conventions of the Rights of
management of our identities, the Child (UNCRC), 3, 16, 20,
37–41 65, 69–70
silences as relational strategies, Uprichard, E., 27, 79, 117
41–4 Usher, R., 33
virtual worlds, 53–60
spatial-temporal regulation, 47 Valentine, G., 32, 53
Spriggs, M., 55 Veale, A., 49
Spyrou, S., 74 virtual ethnography, 54, 55
Stacey, J., 112 virtual spaces, 25, 33, 56, 60
Stainton Rogers, W., 53 virtual worlds, 53–60
Stalker, K., 62, 71, 72 voices of children, 2, 149
Stanley, B., 26–7 Allison’s research on, 140–1
Stanley, L., 24 ambiguity about, 18
Stern, S., 55, 58 borrowed, 128
Stewart, F., 58 common notion of, 75
Stoljar, N., 87 concept and practice of, 3
Stones, R., 68, 77 concept of, 122
storytelling, 4 conventional meanings of, 122
Strauss, A. L., 125 hegemony of, 8, 141–4
symbolic representation of voices, images of dependency, 127–8
128–34 perceived importance of, 47–8
powerful, 134–6
taboos, 6–7, 22–9 privileging, 136–7
Tapscott, D., 59 reflexive re-positioning, 137–41
Tarr, J., 70 research on, 9
Taylor, J., 111, 112, 114 self-confessed gatherers of, 121
Taylor, L., 73 silences and inventions, 123–6
Thomas, N., 144 submergence of, 123
Thorne, B., 16 symbolic representation of,
Thornton, S., 112 128–34
Tilly, C., 97 types of, 122
traditional research methods, vulnerability, assumptions of, 122
challenges to, 122
Traianou, A., 15, 16, 23, 46 Wainryb, C., 83
Triseliotis, J. P., 142 Walkerdine, V., 85, 89, 131, 147
Tronto, J. C., 82 Waller, T., 43, 88, 97
Tuskegee syphilis study, 14 Wall, L., 1, 33, 113, 133
Twine, F., 85, 90 Walsh, D. J., 22, 23, 47
Twycross, A., 17 Walter, J. K., 82
Index 185
Wang, X., 32 Wise, S., 24, 117
Ward, L., 71 Woodhams, C., 7, 144, 148
Watson, N., 48, 70, 123 Wright, K., 70
Weller, S., 21, 32, 33, 51, 60
Westcott, H., 73 Yee, W. C., 33, 35, 37, 152
Wetherell, M., 102 Yngvesson, B., 142
Wheatley, E., 99 Young, L., 60
White, H., 1, 13 Ytterhus, B., 66
Whittaker, C., 74
Widdicombe, S., 117 Zelizer, V. A., 34
Williamson, H., 70 Zignon, J., 5, 6, 33