Qualitative Research – TEXT 22/11/2001 3:26 PM Page i

Qualitative research in practice
Stories from the field

Yvonne Darlington and Dorothy Scott

Qualitative Research – TEXT 22/11/2001 3:26 PM Page ii

For fathers Bernard Darlington Arthur Scott

our Olsson

and Henry

Copyright © Yvonne Darlington and Dorothy Scott, 2002 All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without prior permission in writing from the publisher. The Australian Copyright Act 1968 (the Act) allows a maximum of one chapter or 10 per cent of this book, whichever is the greater, to be photocopied by any educational institution for its educational purposes provided that the educational institution (or body that administers it) has given a remuneration notice to Copyright Agency Limited (CAL) under the Act. Allen & Unwin 83 Alexander Street Crows Nest NSW 2065 Australi a hone: (61 2) 8425 0100 P Fax: (61 2) 9906 2218 Email: info@allenandunwin.com Web: www.allenandunwin.com National Library of Australia Cataloguing-in-Publication entry: Darlington, Yvonne. Qualitative research in practice: stories from the field. Bibliography . ncludes index. I ISBN 1 86508 522 7. 1. Human services—Research. 2. Human services—Research Methodology. I. Scott, Dorothy. II. Title. 362.072 Typeset by Midland Typesetters Printed by South Wind Productions, Singapore 10 9 8 7 6 5 4 3 2 1

Qualitative Research – TEXT 22/11/2001 3:26 PM Page iii

Content s

The researchers vi Table of research studies xi Preface xiii 1 From practice to research 1 Qualitative research in the human services 2 Generating research questions 4 The researcher–practitioner split 5 Quantitative and/or qualitative methods? 6 Stories from the field 8 Comments 17 2 Ethics and organisations 21 Gaining ethics approval for qualitative research 22 Informed consent 25 Intrusiveness 28 Confidentiality 29 Ethical and political complexities of research within organisations 31 Stories from the field 34 Comments 46 iii

Qualitative Research – TEXT 22/11/2001 3:26 PM Page iv Contents 3 In-depth interviewing 48 Choosing in-depth interviewing 48 interview process The 51 Focus groups 61 Stories from the field 63 Comments 72 4 Observation 74 Choosing observation 75 Combining observation with interviews 76 observation process The 77 Stories from the field 80 Comments 91 5 Tailoring data collection to suit the needs of research participants 92 Researching with children 94 Researching with people with an intellectual 102 disability Stories from the field 106 Comments 117 6 Mixing methods 119 Types of mixed method design 121 When findings conflict 126 Stories from the field 127 Comments 140 7 Analysing data 142Transcribing 143 Becoming familiar with the data 144 Coding 144 I need a computer program? Do 147 Stories from the field 147 Comments 156 i v .

programs and politics 177 Stories from the field 181 Comments 188 References 190 Index 203 v .Qualitative Research – TEXT 22/11/2001 3:26 PM Page v Contents 8 Presenting and writing up 158 Whose voices? 160 Structuring qualitative findings 162 Is there a book in it? 164 Alternative forms of presentation 165 Stories from the field 167 Comments 175 9 Epilogue: From research to practice.

the uses of fictional methods in social research. a spell joined the University of Sheffield in 1975.She has also completed an action research project culties. as a researcher and field an advocate. She has. v i . His current research interests are parenting by people with learning difficulties. Much of his work in this area has conducted jointly with Wendy been Booth. He graduated in Sociology from the University of Essex and. Wendy is involved in the of parents with learning difficulties. the monthly newspaper of Mencap.Qualitative Research – TEXT 22/11/2001 3:26 PM Page vi The researchers Tim Booth is Professor of Social Policy at the University of Sheffield. after as a research officer in a social services department. He ants and has published four books and numerous journal articles in the area of parents with learning difficulties. advocacy and supported parenting. narrative research with inarticulate inform. Wendy is a Research Fellow in the Department of Booth Sociological Studies at the University of Sheffield. with Tim Booth. She is the adviser to Huddersfield People First and writes a regular column in Viewpoint . which provided advocacy support for parents with learning difficulties and is involved in a further project which offers supported learning opportunities to mothers with learning difficulties. published four books and numerous journal articles on their research with parents with learning difficulties and the children of parents with learning diffi.

she the maintains homeless people in the inner city. and this has been a continuing theme in her research. she has completed about vii . young experiences of having been the subject of contested adults’ Familyproceedings as children. worker. In 1987 she became the first staff member of the and Woman Abuse Studies Unit at the University of Child North and has worked there ever since. She has been a research officer and lecturer in the School of Social Work and Social Policy at the University of Queensland and has recently taken up a position as Senior Policy Officer with Department of Housing. evaluation research and legal aspects of social work practice. She has completed qualitative research on adult women’s experiences of having been sexually abused in childhood. her professional involvement in this issue. Liz Kelly began researching in 1980 for her PhD on women’s experience of violence. and child welfare workers’ Court understandings of physical child abuse. that is assisting child protection workers to develop skills in evidence-based practice. She has previously practised as a social worker in mental health and family law settings. Anne completed her doctoral research on central the meaning of public space to homeless people in an inner city area of Brisbane. when these things are not separated in women’s and children’s lives. she has continued to work in an unpaid either directly with women or as an activist. As well as direct practice. With her colleagues at the Unit. In this role. Using Knowledge in Practice (UKIP). grounded and its links with policy processes and outcomes research became to her work. Within the School. Yvonne lectures and researches in the School of Darlington Social Work and Social Policy at the University of Queensland. prior to becoming a researcher. and in understanding interest alternative discourses and meanings. and continues to contact with be involved in policy and research relating to homelessness. Her research interest in this developed out of a dissatisfaction about the ways in area which forms of violence had become separated in research and different in service provision. she teaches in the areas of child welfare. client and family experiences of hope in mental illness. She is currently involved in a project. alongside capacity. A feminist activist London. Queensland. Her work there was the start of an abiding in the issue of homelessness.Qualitative Research – TEXT 22/11/2001 3:26 PM Page vii The researchers Anne began her professional social work career in 1990 Coleman worker/welfare coordinator at an inner city drop-in as the social centre for homeless people.

as well as issues around the creation of environments conducive to the development of child and family wellbeing. in Australia and Malaysia. Robyn is Head of the School of Sociology. Jackie has management the for-profit and health sectors and also in the experience in pro. extensive research experience in the disability and She has family fields. disability policy.development. generic munity social practice. Work and at the Her research interests are the non-profit human services sector. and qualitative work methods. Social Munford Work at Policy University. She is the author of thirteen journal articles and several book chapters.of welfare and early childhood services to families. and community from a research viii . She has lectured exten. feminism and social work. Catherine lectures and researches in the School of McDonaldSocial Policy Social University of Queensland. Jackie is the Research Director of the Barnardos Child Sanders Researchand Family Centre in New Zealand. She has worked as a social worker in child protection. disability services and policy.at both undergraduate and postgraduate levels on sively com.Qualitative Research – TEXT 22/11/2001 3:26 PM Page viii The researchers twenty-two different projects. She vision is currently involved in several research projects which have children and families as their focus. residential care and psychiatric social work. mainly on child sexual abuse. Palmerston North. and theorising the reconstruction of welfare. and Her professional background is in social work although she identifies more broadly as a person working in and around community human services. the mixed economy of welfare. She is currently engaged in several research projects. the mixed economy of welfare. She and Social Massey has qualifications in both social work and sociology from Massey University and the University of Calgary. a large research project on the reconstruction of including the community services industry. the reconstruction of the welfare stateshifts in the institutional arrangements supporting people. family policy and effective strategies for achieving family change. Her research interests include women and work management. The projects include research into factors associated with positive change in families seeking that are support home-based family support service. domestic violence and rape. Most of her writing is on the non-profit sector.

where she was involved in several to evaluate studies Children Act 1989 . maternal and child health. Her research are in practitioner-based research. Her academic back.is in history and social work. Cheryl Tilse is a lecturer and researcher in the School of Social Work and Social Policy at the University of Queensland. She graduated in Social Policy from the School. Her practice and ground policy are in child welfare. and child protection. in Australia and overseas. Within the health. policy Her current research includes evaluating housing options for older enhancing family and resident participation in aged people. care ix . welfare Her current research interests are the history of child protection. and has written seminal papers on epistemological and hermeneutic issues in social work. the organisational interests and context of practice and in ageing and aged care policy. Caroline is currently on secondment to the UK Department of Health. She has undertaken extensive qualitative research in the fields of child history. managing the children’s social care research program. These research are funded by the Public Good Science Fund—a programs government body which supports social science research in New Zealand. She has practised as a social worker in Australia and Canada in mental rehabilitation and corrective services. Dorothy Scott is Associate Professor in the School of Social Work the University ofat Melbourne and is currently on secondment as the Executive Director of the Ian Potter Foundation. she teaches in the areas of knowledge and practice. Later she moved to the School to take the lead role in a study of older children’s Cardiff Law views and experiences of the adoption process. mental health and maternal and interests child She has taught practice research at a postgraduate level health. and support in the transition to motherhood. School. In 1999 she social was awarded a medal of the Order of Australia for her contributions in field of child abuse. the organisational context of practice and practitioner research. post-partum psychiatric illness and the social welfare education and research. she joined Bristol University’s Socio-Legal Centre for Family Studies in 1991. After working in the health service sity of and voluntary sector. Cardiff Univer-Exeter in 1982.Qualitative Research – TEXT 22/11/2001 3:26 PM Page ix The researchers into factors associated with family well-being. Caroline is a Research Fellow at Cardiff Law Thomas University.

She is a member of the editorial board Australasian Journal on Ageing . asset management health and financial abuse of older people. MEd from the University of Manchester in 1983. She is now the President of the Hong Kong Kong Social Association and is a board member of numerous Workers social organisations in Hong Kong. of the Angelina Yuenis currently Associate Head and Tsang Associate Professor of the Department of Applied Social Sciences of the Hong Polytechnic University. She is conducting several g research on social support networks of women. She obtained her BSocSc in Kong Social from the University of Hong Kong in 1975. and her PhD from the University of Hong in 1995.Qualitative Research – TEXT 22/11/2001 3:26 PM Page x The researchers facilities. She is actively involved in welfare the development of social work education in mainland China and in training social work educators in China. the legal aspects of later life decision-making around care and accommodation options. the elderly and projects the unemployed in China. and the implementation of the user charges in aged care services. x . Her main research interest is on social support networks in Chinese communities using a rounded theory approach. MSW Work from the University of Toronto in 1978.

(1993) (‘The experience of childhood sexual hood abuse’) Liz Kelly Domestic Violence (1999) — Matters Catherine McDonald—Institutionalised organisations? A study of non-profit human service organisations (1996) (‘Institutionalised organisations?’ ) Robyn Munford and Jackie Sanders Working Successfully with — Familie (1996. 1998. places and homelessness Valley (2001) (‘Five star in Fortitude motels’) Yvonne Darlington—The experience of childhood sexual abuse: perspectives of adult women who were sexually abused in child.(1995) tive (‘Child protection assessment’) xi .Qualitative Research – TEXT 22/11/2001 3:26 PM Page xi Table studies of research Tim and Wendy Booth Parenting Under Pressure: Mothers and — Fathers with Learning Difficulties (1994) (‘Parenting under pressure’) Anne Coleman—Five star motels: spaces. 1999) s Caroline Thomas Adopted Children Speaking (1999) — Dorothy Scott—Child protection assessment: an ecological perspec.

1992) (‘Identification of posthealth partum depression’) Cheryl Tilse—The long goodbye: the experience of placing and visiting long-term partners in a nursing home (1996) (‘The long goodbye’ ) Angelina Yuen-Tsang Towards a Chinese Conception of Social — Support: A Study on the Social Support Networks of Chinese Working Mothers in Beijing (1997) (‘Social support networks of working mothers Chinese in Beijing’) xii . 1987b.Qualitative Research – TEXT 22/11/2001 3:26 PM Page xii Table studies of research Dorothy Scott—Identification of post-partum depression by child nurses (1987a.

Chapters 3. It is more a book about trate ‘doing’ and ‘being’ a researcher than about ‘how to’ do research research. observation. the shift from research back to practice. xiii . This is a common approach in the human services. Chapter 2 addresses ethical issues and the organisational context of research. In Chapter 1 we begin with an exploration of the journey from practice to research—how to generate a research question and how research questions that arise in the human services some mightbe explored through a qualitative research design. 4 and 5 focus particularly on data collection. Chapter 5 addresses challenges involved in interviewing children and people with an intellectual disability. The best ensuing focus on specific stages of the research process. chapters through data collection. We do this through the use of research practice narratives to illus. In Chapter 3 we consider in-depth interviewing and in Chapter 4. analysis and writing up to.Qualitative Research – TEXT 22/11/2001 3:26 PM Page xiii Preface In this book we aim to bridge the gap between theory and practice and between academic and practice contexts in qualitative research.stages of the research process. finally. In Chapter 6 we consider some approaches to mixing tive qualitaand quantitative methods where one method alone cannot adequately address a research question. different ways in which research can be modified to exploring better the needs of these groups of meet participants. where the issues researchers are faced with often complex and multiare faceted.

an aid to research troubleshooting.have written the book for anyone with an interest in We qualitative research. and how the day-tomade their day research context interacts to shape what is possible. Chapter 8ocuses on presentation and writing up. not an end in to an itself. this evolved into the idea of conducting in-depth inter. We decided first what issues we wanted in relevant to include. enabling them to draw on others’ experience in the design of their own projects. primarily through the medium of practice-based stories that illus. By bringing these examples together. We hope the book will encourage creativity others’ and. and then sought out researchers who seemed to us to have grappled with at least one of those issues in an interesting or innovative way. Each chapter trate includes two and often more ‘stories from the field’. but particularly for students and practitioners in human services. we hope to share them with a wider audience specific field to which each study than the belongs. We hope the researcher perspectives we the have presented will be of value to students and practitioner researchers alike. Finally. The book is thus also act as ahowcase of the variety of qualitative research that is being done s in different fields within the human services.showing how other researchers have dealt with particular by issues.Qualitative Research – TEXT 22/11/2001 3:26 PM Page xiv Preface In Chapter 7 we consider some data analysis processes that are common to many approaches to qualitative research. exploring different ways f of writing up research so that the findings can be communicated with different audiences. We envisaged not so much personal experiences about what it is like be a qualitative researcher as comments about how to researchers choices about what to do.were clear from the start that we wanted the book to We include comments from researchers about how they do qualitative research. Through our discussions. The book provides an introduction to qualitative research. in Chapter 9 we conclude with the feedback loop which takes us from research findings back to practice and policy implications—as practice research is a means end.particular stages of the research process. Thus the at least reader necessarily find ‘the answer’. This reflects our belief that there is great value in learning from experience. The work of all of the researchers included in the book was xi v . but will discover how will not others dealt with various aspects of qualitative research have practice.with researchers and including sections from those views interviews chapters.

We numbers had data we needed and it seemed important to use what we the could collect more that we would be unable to use.Qualitative Research – TEXT 22/11/2001 3:26 PM Page xv Preface known to us. We also and not liked continuity and coherence afforded by following the the researchers x v . We Some could have fitted equally well into more than one clearly chapter. We know many other researchers whose work could have had place ina this book. observation. We sent all of the excerpts to the researchers for comment before their final inclusion in the book. and we chose each researcher deliberately. the writing of the book has been a parallel with qualitative research. We struggled with this and tried more to strike a balance between remaining as true as possible to the actual word while maximising readability. We selected excerpts from what were very long interviews. Rather. and ‘smoothing out’ the text was important to this. in some cases. permission to do more. as three hours. In this way. We decided to include excerpts from a selection of the researchers in the first and final chapters and in the chapters on data analysis and writing up. This reflected the reason we chapter of included focused the major part of the interview. with a particular the book in mind. This reflects the messiness of research! The text is also more polished than the transcripts. but all the them and inter. also made the choices about where to place the excerpts. interviewing. Above all we spoken wanted to get the message across about what the researchers were saying. we sorted through the (the interview transcripts) to see who had had data interestingsay on these topics—and we had plenty to choose things to from! will be obvious to any reader that we have edited the It transcripts. We did not pre-empt the process content areas included in these chapters. views we pre-selected the researchers to be included in the chapters on ethics and organisations. and none less than an hour and a some as long half.were more broad-ranging than the one topic. In the end we decided to limit the overall and to use multiple excerpts from the interviews. we took this as validation of and support for the editing decisions we had made and. either through personal contact or through their writing. but without anyone in mind ahead of the interview. tailoring research to the needs of participants and mixed suit methods. We have deleted repetitions and digressions and generally edited the interview text to make it flow smoothly as written text. and we have made the corrections they requested. researchers asked us to do more to tidy up the Where the expression and grammar. To this extent.

xv i . In this way. thank each of the researchers we interviewed—for the We nessreadi. Our thanks to Ann Tierney and David Tregaskis for transcribing also go the interviews.Qualitative Research – TEXT 22/11/2001 3:26 PM Page xvi Preface and their studies through different stages of the research process this helps the reader to become familiar with their and hope workto follow the threads through the and book. As writing the book came to a close. and with for willingness to share their inside stories. we were struck by our how experiences and those we had drawn from our own fellow researchers were becoming synthesised into a collective ‘practice We hope that others. inspired this book.which they agreed to participate in the project. new issues and debates what will which may challenge what we have said and so extend arise the emerging body of knowledge on qualitative research in the human services. including those who may be wisdom’. will add their own experiences of research by reading to we have produced. given the their many unknowns associated with where they would end up. and to Lesley Chenoweth for her helpful comments on Chapter 5.

p. where operationalising where key constructs in behavioural terms is highly problematic (Is happiness the frequency of smiling behaviour?). there is a high. however great their technical interest. This book has more to do with the swampy than the high hard ground.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 1 1 From practice research to In the varied topography of professional practice. hard ground where practitioners can make effective use of researchbased and technique. there is a lot of lowland terri-in the human services field which connects these two parts tory of landscape and we believe that researchers in the human the servicesbe creating terraces which link the two parts of the should terrain. 1981. pp. are often relatively unimportant to clients or the larger society while in the swamp are the problems of greatest human (Schon. where the politics of the setting are often overwhelming and where values and ethical issues are critical and complex. Qualitative research has an important role of the to 1 . The difficulty is that the problems of the high ground. creating not territorial disputes. 42– concern 3). The swampy lowland of practice in the human services is a place there are rarely control groups. However. 36). and there is a swampy lowland theory where situations are confusing ‘messes’ incapable of technical solution. practice uses it’ The has been claimed to be one of the assumptions of positivism (Rein & hite. 1983. yet ‘scientific’ methods of investigation W have difficulty coping with the dynamic and complex social great world human services.belief that ‘science makes knowledge.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 2 Qualitative practice research in play in understanding this world and in complementing other forms of knowledge. such drawn as in the influential Chicago school of urban research in the 1920s. • Systematic observation of behaviour. In recent years specialisations such as medical and anthropology medical sociology have relied heavily on qualitative methods to explore issues relating to health and illness. Anthropology. Despite such diversity the core qualitative C methods can be described as follows: • In-depth interviewing of individuals and small groups. from the microcontext hospital ward or clinic through to the broader of the sociocultural context.recent decades. and through oral history methods have added in-depth interviewing to their reper. Qualitative research is not new. The techniques we will explore in relation to the analysis of transcripts of interviews or observational field notes are also applicable to pre-existing documentary data. used qualitative d methods field observation and informant interviewing to such as understand cultural patterns and social relationships. education and 2 . Historians have always analysed documentary evidence. Qualitative research methods have descended from several disciplines and belong to twenty or more diverse traditions (Miller &rabtree. Sociology has always upon both quantitative and qualitative methods. as their primary source material. Organisational theory has has been largely on case studies created from an amalgam of based observa-documentary tion. 1992).the social sciences in academia. Market research aries of was originally based on the social survey but now complements this focus groups to tap the processes and nuances of with consumer as does research on public opinion and voting opinion. much of it non-quantitative data such as correspondence. from its conception as toire in a iscipline in the mid-nineteenth century. and often utilised both approaches. material and interviews. Qualitative services research in the human For well over a decade there has also been a growing interest in qualitative research by academics within nursing. In this book we will focus on the first two methods. of documentary • Analysis data. Qualitative methods have extended well beyond the bound. intentions.

examine how research questions can emerge from different contexts and address very different questions. and a large service system consisting of a large number of organisations. a community of homeless people under siege in an inner city area. In many ways this is itself a aspects of parallel to qualitative research.qualitative approaches and sometimes by both at the tive and same We then draw upon several qualitative studies to time. Understanding the significance of past or current experiences lends itself to methods such as in-depth interviewing in which trustrapport are essential if an individual is to share thoughts and and feelings. draws upon a range of qualitative studies in This book the human services to illustrate how researchers develop their research work their way through a minefield of ethical and question. or the interaction of different social groups within a community. analyse it obstacles. maternal and child health with an nurses’ assessment of post-natal depression. nearly all of them bring 3 . The interviews we have conducted with qualitative for researchers have been taped and transcribed and excerpts this book from of these studies are used in each chapter in order to few a highlight that chapter’s theme. In some chapters. Some questions lend themselves to systematic observation identify the dynamics which may be operating in order to in a particular group or organisation. parents intellectual disability. political systematically collect appropriate data. including process this we also draw upon our own qualitative studies—when we one. Our examples cover diverse fields and units of attention and include women who were sexually abused in childhood. highlight authorial voice is always present in qualitative that the research. chapter we explore how the questions which arise In this from practice in the human services can be addressed by both quantita. do we speak to the reader in the first person in order to so. with and then disseminate the findings and implications of rigour the research. Research methodsin-depth interviewing and participant observation such as are particularly well suited to exploring questions in the human services which relate to the meaning of experiences and to deciphering the complexity of human behaviour. While most of the researchers were not working as practitioners at the time of the inception of their studies.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 3 From practice to research social work as they attempt to struggle with the issues which arise in their particular part of the swampy lowlands.

. Some of the ques. groups or whole communities. hard. or how to collaborate with others to group bring a change in an organisation. Given that research has traditionally been conceived within a ‘hypothetico-deductive’ model of science. tedious. prestigious.’. . families. education done and welfare practitioners. expert. it is little wonder that practitioners’ questions which from the swampy lowland do not come in the form of come ‘falsifiable’ propositions. ment. .Practitioners are often intimidated and alienated by the very notion of ‘research’. Beyond may the clinical level are similar questions which might relate to a whole of service users. all of services. the studies described have important implications for the way in which policies. scientific. If they had the time to write down their ideas or to share one of their questions with colleagues it might start with a phrase such as ‘I wonder if . When asked to offer words which sprang to mind 4 .which arise are clinical hypotheses—ideas about the tions possible background features or presenting features in a particular indi. generate multiple questions from their practice. Schon (1983) to observe that experienced professionals leading often more than they can know say. time-consuming. Most practitioners would not think of such questions as research. 1990). Generating questions research The world of research and the world of practice have remained fairly separate. cold. factual. difficult. Yet every day those working in the human services field. . In a word-association exercise one of us has many times with groups of experienced health. programs and services are developed and delivered. a community or an entire about service system. be it with individuals. measure-accurate.?’ or ‘I’ve got a hunch that . or working hypotheses about interventions vidual that result in a certain outcome in a particular case. This type of knowing is not always easy to state explicitly in a generalisable propositional form.or family. words that have been offered in relation to the word ‘research’ include: objective. tacit nature of clinical judgement leads the The often practitioner and others to dismiss their professional knowledge as unresearchable ‘intuition’ (Scott.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 4 Qualitative practice research in to their research many years of experience working in the human Regardless of the researchers’ backgrounds.

the following were among those typically subjective. The split researcher–practitioner The dichotomies between notions of objective and subjective and between art and science seem to parallel the dichotomy between of research and the world of practice. The ‘researcher–clinician split’ over the which in psychology in the 1950s has been attributed to occurred the attack mounted by Eysenck on psychotherapy as pseudoscience & Barlow. 1976). Yet in the same decade the growth of post-modernist traditions in the social sciences led to atrong resurgence of interest in qualitative research and saw s its expansion into fields such as discourse and narrative analysis in cultural studies. Yet in the same era the emergence of emancipatory and feminist research traditions pushed the bound. 677). Yet behaviourism strongly assisted in (Hersen the development of the clinician-researcher or practitionerscientist in psychology.the 1970s extreme behaviourist positions were In strongly challenged by the rise of humanistic psychology. A rapprochement of sorts is occurring between quantitative and qualitative research methods within the social sciences. as well as to education and welfare. warm. p. difficult. busy. ‘The issue of was still whether measure the subtleties of human nature and interaction one can will to be a problem once devised measurement rules can cease be shown to have a rational and empirical correspondence to reality’ (Bostwick & Kyte. with single subject research designs and other quasiexperimental methods being refined for use in clinical settings. soft. In recent 5 . The faith in empiricism to deliver knowledge for practice in the human services strong in some quarters in the 1980s. people.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 5 From practice to research in relation to ‘practice’. offered: pressured. This dichotomy the world is new and has been the focus of a debate which has not unfolded past half century.research methods even further to address the aries of power imbalance between the researcher and the researched and to allow the voice of the ‘subject’ to be heard through qualitative research. messy. gave empiricist approaches a new vigour in a social context of resource scarcity in which effectiveness and efficiency were dominant concerns for government. 1981. flexible. 1990s the extension of ‘evidence-based practice’ In the from medicine to the rest of the health field.

to some extent. feedback from those using services can be obtained easily through routine client satisfaction scales which rate different of a service. exploring define their needs and why and how they seek how people assistance in certain places. hypotheses such as ‘Clients who are rung the day prior to an appointment willmore likely to keep the appointment’ or ‘Clients who are be offered an appointment within three days of making the appointment will be more likely to keep the appointment’ are easy to test even in amall s human service organisation. For example. the outcomes of service. as quantitative the questions which arise in the human services require a broad repertoire of research approaches. occupational status or or people ethnic backgrounds can be identified. Quantitative methods? and/or qualitative Some questions are readily transformed into testable propositions and can be investigated using quantitative methods. To tap both dimensions of 6 .or under-representation of males or females of different age groups. and qualitative methods may have a place in however.reservoir of quantitative data which can be analysed ordinary very simply to answer many questions about service user characteristics. but hearing how the clients have aspects benefited a service in their own words will require more or not from than a standard quantitative approach.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 6 Qualitative practice research in years we have seen the terms ‘practice research’ and ‘practitionerresearcher’ transcend the narrow behaviourist model of the clinician-scientist and draw upon a broad range of both quantitative and qualitative modes of inquiry. the nature of service provision and. The reasons why this might be so cannot be so easily investigated using quantitative methods. Thus the over. Furthermore. hypotheses about patterns of presentation toservice in relation to different sections of the community can a be relatively easily investigated if the demographic profile of an area is known. We do not wish to dichotomise and qualitative methods of investigation here. Sometimes this involves a major endeavour to collect and analyse the necessary can also involve the analysis of data which can data but it be collected easily or which already exists. Similarly. The client information systems and management information systems used in human service agencies today provide an extra.

next question she may ask is of a very different order: The are What the multiple meanings of such a diagnosis for these women and significant others in their lives at this time? This is a hermeneutic is. swampthe high hard ground and analyse it up to there. question. with some risk to the diversity and nuances of defined. One of us has argued to their that ‘meaning construction’ is at the heart of much of the work in the human services field and that the core traditional professional tools 7 . questions from both the high ground and In the swampy lowland emerge from the same setting but cannot be trans. Thus a social worker in an formed oncology hospital who is interested in establishing a support unit of a group for women with gynaecological cancers may ask herself a range of different questions. it is about the construction of meaning. one ‘allow the data to speak for itself’ by reproducing all of the can indi. As she looks at a list of the patients in very a articular ward she may ask the following sorts of questions: p How women in the ward at this time have a similar many diagnosis? with this diagnosis are at a similar stage in the How many trajectory of their condition? What is their average length of admission? These are fairly straightforward numerical questions for which the data already exist. remain it is possible to take some qualitative data from the unheard. even done complexity and subtlety.into quantitative data. other situations. ‘What were the most useful/least useful aspectsservice?’ and ‘What suggestions could you make of the for improving the service?’ latter type of question does not presuppose a The particular classification of responses. This involves making qualitative judgements about their meaning before they can be allocated to a particular category. Of course. It is possible to turn qualitative the data of this nature into quantitative data if the categories are clearly Thus. the and recognising that those with literacy problems may data.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 7 From practice to research consumer feedback may require a combination of questionnaire items with predetermined response categories as well as a number of open-ended questions such as ‘How did you expect to benefit from this service?’. it is likely that much damage would be and. and in analysing such data the researcher has to inductively derive categories from the individual responses.responses but this merely leaves the task of making sense vidual of responses up to the reader. that The responses to such a question are unlikely to be easily classified into mutually exclusive categories that could be quantitatively analysed if they were.

On the other hand.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 8 Qualitative practice research in of client-centred interviewing and observation are not dissimilar to qualitative research methods of in-depth interviewing the and participant-observation (Scott. for example. that for m the professional practitioner such a question is immediately and directly with what she then does and leads to a series of connected equally questions. survival may allow some of the questions to be explored in qualitative methods a ifferent way. 1994a) is about the parenting experiences of UK parents with an intellectual disability. The study involved unstructured interviews with 33 parents.issues which can be explored in qualitative research and sity of the range of methods which can be used. The difference would be. same question which our social worker asks about The the meaning(s) of gynaecological cancers could equally be asked by a edical sociologist. by investigating the long-term outcomes of cancer support groups in relation to years of . the views of the women themselves d on experience in a group and what they may have found their helpful or unhelpful would be important. 1989). Would it be helpful or unhelpful to form complex aupport group for these women? What would be the best way s to facilitate such a group? How could one know whether the group was successful or not? For whom might it have been helpful or unhelpful and why? How might forming such a group affect and be affected by the current pressures on staff in the ward and the interprofessional tensions and dynamics? Researchers could explore (and some have done so) certain aspects of these questions using quantitative methods. followed by more intensive thirteen of those parents—six couples and one work with single 8 . however. Tim and pressure Wendy Booth—Parenting under This study (Booth & Booth. Stories field from the Here are examples of how some of the researchers we draw upon in book generated the questions which led to their decision this to adopt a qualitative approach. They illustrate something of the diver. They also reveal the centrality of the researcher as a person to the whole process. For example.

involving observation.as a researcher I was looking at how it affected families. informal interviews and document analysis. neither Tim nor I had ever at considered that people with learning difficulties had children. the readings of her work which for homeless people who had been involved in the she held study impact of the study. Wendy and Tim were interviewed this book. using observation as a data collection method. home . Anne talks about how and the she became interested in doing research with homeless people. .. We talk with them again later about together for their approach to interviewing people with learning difficulties and aboutthey wrote up the study. of policy. . Anne Coleman—Five motels star For her PhD research. being helped in a special care unit after She was the of her second child . We talk with Anne again later in the book about negotiating entry to the community. in-depth interviews. an spaces innersuburb of an Australian city. and whilst I was doing that research . and talking with them. and that’s what really birth gotinterested. She situated the study in Fortitude Valley. but in and amongst getting this change to know parents whose sons and daughters were in hospital. when we were looking at the closure of the long stay hospitals . analysing data. and that’s whenput in for a we grant. It was a multi-method in the study. I knew some of the professionals in us the who might be able to help us and give us an area idea of how many parents might be around. they knew seven mothers. Anne Coleman (2001) was interested in exploring the connection between homeless people and the public they use. Wendy is talking here about how how they to be interested in this came area. Wendy: I came into it about the mid-eighties. . 9 .came across a mother with learning difficulties.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 9 From practice to research parent—over the course of a year. As they came out.. they moved into their towns. I also got to know some of the sons and daughters. Here. The study was conducted city over a of some years at a time of urban redevelopment and period conflictarea about homeless people. I andthat point.

I to think became aware that many of the people that I knew in the Valley who had been homeless were still using those spaces even though there’d been this really public big urban renewal process in the area. I suppose it was partly because their lives aren’t full of a lot of significant other people.. So what happened was that people were really under pressure.. . I started work with homeless people in 1990 and I’d really had any contact with them before but never the they saw their lives and the things that way happenedmade perfect sense to me .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 10 Qualitative practice Anne: research in The central research question was ‘What do public in that area mean to people who have spaces been homeless there for quite some time?’ So I was really interested in exploring the connection between people and the public spaces they homeless use. I kept in enough contact with people. after three to them years’ with homeless people in a drop-in centre. but in spite of the obvious that they kept getting sent—‘We don’t message really you here any more’—they kept going back want and these spaces.. contact as as learning an immense amount I started to well form relationships with people. but in a different world. . After I finished that three work as a social work practitioner I was years’ lucky to be able to stay in the area.they were being moved on by the police and their gear was out of parks and dumped at night and all taken sorts of things had happened. of being able to learn things from people who lived so close to where I lived. that in fact it wasn’t just any to these public 10 . because of the still life journey that we’ve been on. but I became important to people The other side for me was that the privilege quickly. and I became more took and interested in trying to understand how all of more us human beings share certain things in common as but make unique meanings. So I started to think that using maybe quite understand what these spaces I didn’t meant people. reallyhold of my imagination.. And just as I was ways starting about doing some further study. and in doing that of course more learning and there was more exchange of skills took place and of looking at the world.

their This was the starting point of what turned out to be a circuitous route to finally settling on a qualitative methodology for my PhD research. considerable but 11 . aecognised area of public and professional interest r —number of articles had been printed in the press a and popular magazines.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 11 From practice to research space—that there was something about these specificand these specific people and it was the places connection between those two things that was important. Nevertheless. many women who had sought assistance in relation to depression. of the women I spoke with linked their present problems to their sexual abuse in childhood and there were also similarities in the waystalked about what had happened and what it they had in their lives. disclosed. I experienced pressure to do a quantitative study. Yvonne: In the early to mid-1980s I was working as a social in a community mental health centre. though not all. I became interested in the ways meant in which these women understood their experience of childhood sexual abuse and its long-term impact on lives. and others this was the first time they had talked for about abuse the with anyone. childhood sexual abuse in the course disclosed of counselling. At worker that child sexual abuse was only just emerging as time. and child protection responses were beginning to be developed. in my position as a female counsellor in a mental health team. was certainly little recognition of There child sexual abuse as a legitimate area for mental health intervention. the abuse had been previously either at the time of the abuse or later.of worthlessness or relationship feelings problems. For some. Yvonne Darlington—The experience of childhood abus sexual e Yvonne talks here about the background to her study. published in 1996. Many.

known as the community sector or nonsations profit So it’s a lot of organisations. . 1996). We talk with her again about her use of mixed and how she used a deductive approach to analyse methods her interview data. and my concern that of that women’s perspectives had been missing from the research in this won through in the end. I conducted two field.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 12 Qualitative practice research in however I constructed and reconstructed my research there was always one that I returned to questions. Catherine: The PhD came out of me wondering [why] everybodythe non-profit sector is different. and a shorter interview. this in the state and is 12 . heterogeneous . different from organisations organisations in the market . so I was looking at an organ. and I different accepted those assumptions for quite a long period of time until I started doing a PhD . . Catherine organisations? McDonald—Institutionalised Catherine McDonald used a mixed method approach involvingand semi-structured interviews for this study survey a (McDonald. complex. . indepth interviews each with ten women—one lengthy interview in which they talked about their experiences sexual abuse and the impact they felt of childhood it had on their lives. that was isational large. . There says that are assumptions embedded in all discourses these and conversations about the sector that it is somehow from the state or the market. .a whole field of organi. and I wanted sector. The they second interviews provided some useful insights about what it like to be a research participant and I have is included this material later in the some of book. to know why they were . — do women who have been sexually abused how as children make sense of that experience? The strength original idea. in which they talked about four how had experienced the first interview.. had aboutto six weeks later.space. Catherine talks here about how she became interested in the application of neo-institutional theory to the non-profit sector in Queensland. a social space.. .

What I did first was have a look whole bunch of writing about the sector at a and theorising around the sector and there’s a lot about of emerging in the USA in particular—theories of that the non-profit sector—and none of them answered the questions about why it was different. so I followed up a few answered refer. I was keen to understand the support ways in which our services might be contributing to family well-being and also to understand more about the 13 . New Zealand. 1996.which took me to a body of ences organisational . a prospective statistical analysis of client and intervention factors over a twelve-month caseload was undertaken relationships between a wide range of factors and to analyse client (Munford & Sanders. Jackie talks change here about how the study originated.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 13 From practice to research what people assumed. . Robyn Munford with successfully families and Jackie Sanders—Working Robyn Munford and Jackie Sanders used qualitative and quantitative strategies in a multi-phase study on what worked for families family support services from a non-government child receiving and family welfare agency in New Zealand. . Jackie: I was a Regional Director for Barnardos New Zealand were managing a large number of and we family programs. So then I applied that to non-profit human service organisations theory in Queensland. and followed up three months they after completion. The research was conducted with the support of Barnardos New Zealand and the Foundation for Research. provides the theoretical theory which justification for the perception of difference. Finally. The first stage involved an extensive file review of a twelve-month caseload for the whole organisation. and really it was to see whether or not theoretical explanation for the assumption this of difference held. The second stage primarily involved qualitative small sample of families was intensively tracked research—a as moved through the service. Science and Technology. 1999). They all invoked this assumption of difference but never actually the questions. 1998.

From that we developed an ongoing research partnership between Massey and Barnardos that has been successfully operating for six now. 1987b.1992). unlike those with acute psychotic disorders. We work put together a working party to try to secure funding— this successful and the rest. was We secured ongoing funding for four years to undertake aeally detailed analysis of the family support r program result we were able to produce a very and as a clear of the role of strengths-based work in picture family and also to begin to understand more change clearly the relationship between management styles and practice effectiveness. Dorothy: I was working as a social worker in a psychiatric ward Queen Victoria Hospital in Melbourne in of the the 1970s and early 1980s and I was very involved late in the treatment of women with post-partum psychiatric It gradually dawned on me that conditions. which was probably more personal than anything—it being the place where I had done my Master’s degree.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 14 Qualitative practice research in possible long-term effects of our services on clients and their families. funding and practice— that combine together to generate good interventions for clients. is history. as they say. I think we have an abiding interest in years the complex mix of factors—personal client. We began dialogue with people in sociology and social and Robyn was one of those people.better the link between management standing styles and practice approaches in terms of identifying waysthe two can best be aligned for the greatest that client We had a link with Massey University gain. structural. women with depressive disorders. seemed to be admitted to hospital very late—long after serious depressive symptoms had 14 . agency organisational. I also had an interest in under. Dorothy Scott—Identification depression of post-partum Dorothy talks here about the background to this study (1987a.

I enrolled in a Master’s make degree my thesis I intended to develop a and for simple instrument for the nurses to identify postpartum depression and test it for its validity and reliability— a ery traditional research project really. While I was doing the 15 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 15 From practice to research developed and often after real damage had been done mother–infant relationship and the couple to the relationship. The nurses told me that they didn’t i have a to do mental-state examinations on mothers mandate — women came to see them to have their that babies and measured and that the mothers weighed were sensitive to nurses prying into private highly issues as their emotional state. and to interview mothers nurse about they saw as the role of the nurse and what what they find acceptable in regard to the nurse might exploring issues related to their emotional well-being. I suppose we would see it today as describing ‘best practice’. and semi-structured interviews with a mixture of closed and open-ended questions with a sample of 45 mothers in their homes. But it didn’t turn out as Intended. So I became interested in how primary healthworkers like maternal and child health care nurses be able to identify post-partum depression might and earlier referrals. which v otherssubsequently done with the Edinburgh Posthave Natal Depression Instrument. So instead I thought it might be useful to observe at length how nurses who were regarded peers as being particularly skilled in by their their interpersonal skills went about assessing women for depression. the nurses. So I used a number of methods—direct observation of three purposively selected maternal and child health nurses followed by in-depth interviews with at work. I also wanted to understand which governed the interaction between the norms the and the mother. In my literature such search read an article on a study on anxiety I also among of young babies and when I followed it mothers up spoke with the researcher she told me that and the nurses had discarded that instrument as soon as the research was finished! There seemed little point in developing an instrument which nurses were not prepared to use.

emotional and economic costs looking of dementia and part of that had been semistructured with 243 people who were caring for interviews aelative with dementia. I was interested in following their experiences and under.data. I felt that the semistructured interviews really hadn’t captured just how traumatic the experience was for them. Cheryl goodbye Tilse—The long This is a study of the experiences of older people who place apouse in an aged care facility and the conditions of the s facilityshape the nature of the experiences of visitors (Tilse. Cheryl: I was working as a research assistant on a large project at the social.. This research stimulated my curiosity about practice wisdom or clinical judge.and that has continued to be a great ment interest. this added an informal a participant. 1996). semi-structured interviews and documentWe talk with Cheryl later about her use of observation. So I became very interested in understanding what it was all about their perspective . my nurses were not in the study of although course.I was particularly from concerned happens to people who’ve been about what married 60 years when they have this involuntary for 50. I was particularly r interested in the interviews I did with spouses who had placed a partner in a nursing home.when one partner goes into a nursing ration. including indepth interviews.dimension to the study which observation ultimately be extremely valuable in understanding proved to the subtle nuances and ‘unconscious dynamics’ of the nurse–mother relationship. analysis. the usefulness of combining qualitative data collection approaches..kept notes on my own experiences of the But I service and although I had not intended these notes to be data source. writing up and dissemination of analysing findings. which It involved several methods of data collection. home.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 16 Qualitative practice research in thesis I had two children of my own and so I was a of the maternal and child health service client myself. observation.what it was like for them to try and standing continue 16 . sepa.

.My concern it was was their perspective wasn’t understood in practice that or policy... Tim grew out and Wendy Booth’s study can be seen in the context of the history of deinstitutionalisation and the emergence of reproductive rights for people with intellectual disabilities. Visitors were being constructed in policy in as resources to help the organisation and to provide for the resident. and as the researcher seeks greater depth of understanding there is an inevitable trade-off in the number of 17 .information about this context. it is important to note that each of these studies of a particular time and place. The researcher needs to acknowledge the limitation of their study’s findings in terms of context in which they were obtained and give the reader the suffi. it is probably obvious to the reader that the sample size in most of these studies is much smaller than one would expect in quantitative research and that the sample is often not selected systematically to ensure that it is representative of a particular Obtaining representative samples in these contexts population. Similarly. So I was in which they visited.. is often very difficult. the purpose of meeting and visiting and how nursing homes provided for family visiting on. For example. Comment s Several points about qualitative research emerge from these examples. Anne Coleman’s study is about a particular community of homeless people in a articular area that was undergoing urban redevelopment p and gentrification at a particular period of time.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 17 From practice to research or exit from the marriage within an organisational also concerned about the context setting. and I had a very firm resources view the purpose and meaning of visiting for this that group was particularly different.I was determined to try and catch and so what like for the older people. Cheryl Tilse’s study was a response to what she considered to be a too-narrow conception of role of spouse visitors in nursing the homes. Second. The reader needs to be cient aware so that they can make allowances when extrapolating of this the findings to other settings. First. of the issues for qualitative research is therefore the One degree of generalisability of findings across settings.

who can act as a sounding board with the on the question is best framed. arguing that all claims to knowledge are situated and partial (Marcus. the number of interactions or contacts investigated would have been infinitely larger than the number of individuals or families involved. then the sample is often much larger than first appears. as this will shape how it can how be investigated . This also places limits on generalisability.qualitative researchers are happy to acknowledge this Some and discuss their role in terms of ‘positioning’. Developing the embryonic question into a researchable form is more difficult and in qualitative research the question(s) may continue to be refined throughout the whole study. For some. the studies of Anne Coleman. the inquiry researcher cannot be and should not be written out of the text. we can see from these examples that in qualitative Third. if one considers the unit of attention as the phenomenon under investigation. The qualitative researcher is inextricably immersed in the research. Some important material may only in the form of unpublished reports. rather than the number of individuals.to explore. 1994). which have to exist be 18 . However. This is where it important to have others. it is a matter of recognising that tions they already have questions waiting to be crystallised and that the task of noticing the questions embedded in the conversations is one they about their work with themselves or have others. It is also important to discover what other research has already been done in relation to the question.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 18 Qualitative practice research in people able to be included in the study. relates to the development of the research question and. thus qualitative research requires a high level of ‘reflexivity’ or selfreflection about one’s part in the phenomenon under study. we would suggest that the conception of the question is usually not too difficult for those curious about the world around them and it may be more difficult choosing which of many ques. it permeates all parts of the qualitative research shall process. or the reader interested in generating their own research F questions. For qualitative researchers the questions they explore grow out some of strong ideological commitment and the pursuit of social a justice. as This we see. Cheryl Tilse or Thus in Robyn Munford and Jackie Sanders. preferably not too closely is involved site of the research. This may give research its driving force and provide an entrée into the particular social worlds—but there is also a risk that the researcher may avoid what they do not want or do not expect to finding find.

whereas in the United Kingdom ‘childis often synonymous with child welfare or child protection.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 19 From practice to research tracked down through a network of people. important to be aware that the same thing may be It is described differently in different places and sometimes the same terminology different things in different places. What you in order are interested in may be a concrete example of a broader phenomenon very useful to consult this broader literature. That way you also learn thus the skills of searching the literature. For example. means rather the ‘child care’ in Australia generally refers to the day care term of children of working parents. however. It theory led examining other studies which at first sight might seem to totally irrelevant (such as studies on the roles of policewomen and parish These. helped researcher out of a narrow mindset to fresh ways of lead the seeing the issue. but it is essential is that person fully understand your question and your setting this and it is preferable to do it together. proved to be very pertinent and priests). organisations and peak bodies or clearing houses. and it may be For example. This involves carefully selecting key words relating to the research topic and systematically searching data bases for books. highly desirable to undertake a thorough literature It is and search assistance of a specialist librarian will be helpful here the in accessing the numerous electronic databases which contain the published research literature. care’ If literature search is crossing service systems. role became a central conceptual framework for the study. or is going back the to time when different terminology prevailed. challenging again is to ask ‘What is my question a More of?’question to conceptualise your issue of interest. then it is a obviously to know the range of terms to use in any literature important search. 19 . journal articles and reports electronic which have been classified according to one or more of the key words. By conceptualising the question as one which was essentially about the core and marginal functions of a professional role (the role of the nurse was weighing and measuring babies and core the marginal role was assessing maternal emotional well-being). in the study described above on post-natal depression and role of the maternal and child health nurse. asking ‘What is the the question a question of?’ led to a body of literature which a literature search based on the more obvious key words would not have accessed. It possible to pay someone to do this for you.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 20

Qualitative practice

research

in

Research is essentially all about seeing the world in fresh ways, about searching again or re-searching the same territory and seeing a different light. In qualitative research we are it in particularlyin how others see and experience the world. This interested requires very aware of the lens which we bring to the task. us to be Perhaps it is a perceptual impossibility to look at one’s own lens at the same as one is looking through it, but this is one of the many time fasci- challenges of qualitative research. The excitement resides nating not much in reaching the destination, for we can never so completely enter the world of the other, but in the voyage and what might be found on the way. The swampy lowlands await you.

20

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 21

2 Ethics organisations and

The best one can do is to consider the ethical and political issues in particular research question, determine the areas of asking a concern the research, take into account professional standards prior to that been established and then consider the ethics of the have entire process as an individual case with its own social research and political ramifications (Minichiello et al., 1990, pp. 245–6).

It may seem strange to combine ethics and organisational issues, yetqualitative research in the human services, perhaps more than in in other area of research, doing the right thing by research any participants coexists with the pragmatic process of ‘getting in, getting on getting out’ of the research and setting. ethical principles which should guide research are The fairly clear. Both professional ethics and research ethics are based on similar core principles, such as beneficence and duty of care, and so might assume that there will be clear and congruent criteria one for determining ethical behaviour in research in the human services. Yet standard research and practice ethics statements let us down simply they are designed from different perspectives with because differentsight: the academic development of knowledge on ends in one the delivery of service on the other. They are not designed hand, for interface of these two the domains. chapter it is argued that there are complex issues In this which arise in research in the human services, particularly when the boundaries between practitioner and researcher roles, and those 21

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 22

Qualitative practice

research

in

between the roles of client and research subject, become blurred. The interconnectedness between the ethics and the politics of research is evident at every stage of the research process with the fundamental question being ‘Who owns the research?’. most Fromsome questions come to be framed as questions for inquiry how to ways in which research findings are disseminated and the utilised, the process is shaped by the interests and relative power of the various stakeholders. For example, the interests and power of agency management, service providers and clients may differ. These also exist within a social and political context. players The prevailing orthodoxy can allow certain questions to be asked and others. The political significance of a social problem will not influ-the priority given to it in research funding, the manner ence in which its findings are received and how its recommendations are implemented .

Gaining ethics research

approval

for

qualitative

In the past decade or so there has been an increasing awareness of complex ethical issues associated with research the involving and animals. Examples of grossly unethical humans practices human experimentation in medical research led to involving the development of mechanisms aimed at protecting the interests of participants. Research grants and the permission to undertake research under the auspices of one or another organisation are increasingly subject to processes under which the ethical issues associated with a particular study are screened by institutional ethics committees, sometimes known as ethics review boards. While the core principles governing the decision-making of these bodies are similar, they vary considerably in their operations, requirements and procedures. Prospective researchers should familiarise them- with the specific requirements of the relevant ethics selves body early on when considering a research very project. Human research ethics committees play an important gatekeeping role in all research involving human subjects and are likely extra vigilant in their consideration of proposals for to be research concerning any potentially vulnerable groups of people. Ethics committees have a duty to consider all possible sources of harm and satisfy themselves that the researcher has thought through all the 22

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 23

Ethics organisations

and

relevant issues prior to granting permission to proceed. The onus is firmly on the researcher to show that the proposed research will not harm and that adequate safeguards are put in place to cause ensure this. Even when practitioner researchers are experienced in working with a particular group which may be considered vulnerable in the research context, they will still have to demonstrate to an ethics committee that they have the appropriate skills to undertake the proposed research with this group. While ethical guidelines used in university and medical research settings were originally oriented to biomedical experimentation, as clinical drug trials, their mandate has now extended to in behav- and social science research. Ethics review boards are ioural often unfamiliar with qualitative research and this can create difficulties for researchers, particularly in the field of health. Attempts are now made to assist medical and health research ethics being review to develop a better understanding of qualitative research boards and associated ethical issues. For example, in Australia the its National Health and Medical Research Council has produced a special paper on ethical aspects of qualitative methods in health research Health and Medical Research Council, (National 1995). Some of the most common areas of misunderstanding in relation to qualitative research relate to the often small sample size and the of specific hypotheses, control groups and predetermined lack ques- which can lead to the false assumption that the tions, proposed not sufficiently rigorous. It is therefore important to study is addressconcerns directly in any such proposal. Institutional ethics committees may also be unfamiliar with the human service field, so those undertaking qualitative research in human services can encounter a double-layered lack of the under- Ethics review boards may not appreciate that the study standing. is part and parcel of professional practice, governed by professional ethics and under the auspices of an organisation with its own struc-of accountability. Thus, for example, workers who ture routinely collect feedback from service users may find that the ethics reviewis uneasy about such data being collected from people board whodependent upon the service and as a consequence may be are seen as constrained in their capacity to give freely their informed consent. The boundary between a clinical audit or quality assurance project in an agency and ‘research’ may also appear blurred. In our experience this is less likely to be a problem with medical ethics 23

emergent. 1989. 58). protecting privacy. It covers all be types of research. Other organisations. such as government departments. and (c) the more loosely defined. It takes the prospective children researcher research process and sets out clear guidelines in through the relation to designing valid research. as the distinction is usually better understood in health settings. implementing the research. may be concerned with the potential ramifications of the research and at times political research may be thwarted. There are three types of safeguard problems. Outside universities and hospitals the accountability mechanisms for research are still developing and research in some areas remains relatively unregulated. and on three confidentiality . p. 24 . In some human services agencies. not be an agency policy or set of guidelines on there may research wishing to undertake research may unwittingly and those find themselves in an ethical minefield. obtaining voluntary informed consent. In university-auspiced research in the human services it is usually necessary to have the permission of both the university ethics committee and the relevant human organisation(s). intrusiveness. human service agencies are accelerating their efforts Some to provide guidelines and structures to manage ethical research. Qualitative research in the human services poses particular challenges in relation to ethical considerations. the categories are loose and overlapping: (a) the although participant– relationship itself. within which are divulged investigator many confidences. design (Ramos. This child and agency has produced a Research Code of Ethics (Burnside. This chapter focuses main issues: informed consent. (b) the investigator’s subjective interpretation of the collected data. including quantitative and qualitative studies. 2000) which cancopied in its entirety for non-profit purposes. with the latter typically being one of services the prerequisites for the former.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 24 Qualitative practice research in review boards than with university-based ethics committees in the behavioural and social sciences. maximising benefits and disseminating the research results. and is particularly focused on safeguards for research with vulnerableand young people. The characteristics of qualitative investigation seem to generate decision-making problems for the investigator who particular seeks to the research participant. ostensibly on ethical proposals grounds. An excellent model of how this can be done is that of the Australian family welfare agency Uniting Care Burnside.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 25 Ethics organisations and Informed consent The capacity of an individual to give freely their informed consent to research is a core principle in research ethics.be dependent upon the agency for the provision tion may of much-needed field placements for their students. Even if the research is externally funded. any out-of-pocket expenses in the human services but the notion of voluntary consent is sometimes thought to be diminished if undue enticement exists in the form of payment. close For example. some organisations which routinely undertake social research with low-income families. Even where the researcher is not client’s worker. the researcher may be located in an organisation such as a university which has links and an interdependent relationship with the agency. or in association with. similar complexities might arise when the the research is being undertaken on behalf of. instances the agency commissions research and the In some relationship between the agency and the researcher is governed by a contract. the agency from which the client is receiving a service. university departments that provide professional educa. as the person apprehensive about the possible withdrawal of the may be service if they refuse to participate. it is a capacity that be diminished by a range of factors. Such factors can do influence the research process. researcher has a clear obligation to inform Thus. On the other hand. and whether that 25 . the potential research subjects that their access to services will not be affected they agree to participate in the research or not. have a policy of reimbursing research participants on ground that their time and knowledge is valued. such as the Brotherhood of St Laurence in Melbourne. One factor that can is commonly mentioned in research ethical guidelines is that of incen. A central issue in human service research is the complications which arise when the researcher is also the service provider. The the researcherneeds to think through the issue in the context of therefore their specific project and consider what might constitute undue enticement for particular participants. as the capacity of a client to withhold consent can be diminished by unequal power and the dependency typically entailed in the the worker–client relationship. and it is important and that arrangements of this kind do not adversely affect the interests of research participants.It is common to reimburse research participants for tives.

These are often the very capacity to people with whom human services are involved and whose input to the service providers and managers is sought through qualitative research. 1989. This can raise particularly complex issues for researchersa result of legal requirements. Prospective should be aware of any legislation which relates researchers to minors in their particular context. third parties such as children in child protection research. p. p. Under conditions in which it is likely that the capacity for freely given informed consent may be diminished. agency policy or who. For example.Research subjects. Most guidelines require that this be stated in writing and that ethics the research participant signs to the effect that they understand these conditions. The depth of this trust should gator to increase in proportion to the degree of shared intimacy and respondent vulnerability’ (Ramos. Generally speaking. minors under the age of eighteen cannot enterbinding contracts.Where children are to be involved in into such research and are too young for their consent to be legally meaningful. In relation to children. as professional be obliged in certain circumstances to divulge ethics.. A central issue in any research children is to what extent. might infor. they statutory clients or involuntary patients. researchers have an additional duty of care to potential participants. 1995.. 90). For people whose comprehension of English is limited.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 26 Qualitative practice research in they are free to withdraw from the research at any time. parents can exercise their powers as guardians to do either or both so (Commonwealth Department of Human Services and Health. research participants should be able to trust the investi. have a are diminished give informed consent. It has been argued that: ‘Regardless of the harm information divulged. 26 . have a right to be informed of any limitations on the confidentiality of what they may divulge to the researcher. they are capable involving of giving informed consent and under what circumstances and to what degree parents can give consent on their behalf.protect their welfare. like clients. the researcher may have a duty of care to mation. 59). there are additional issues and it is perhaps not surprising that the voices of children have remained largely inaudible as a result. if any. this agreement should be provided in their own language.should not be asked to give consent to things which subjects could them. cognitive capacity or the fact that people. Some by virtue of their age.For example.

than non-therapeutic research. in which there is levels no direct potential benefit. a distinction has been made In between ‘therapeutic’ and ‘non-therapeutic research’ in relation to children as subjects. so it may be appropriate for researchers to seek the views of both parents and children.easy to assess the level of risk. as there is a clear duty to inform as well potential in research of the possible adverse effects of subjects participation.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 27 Ethics organisations and Yet a minor’s capacity for informed consent is dependent upon age individual maturity. In a recent UK of children who had been adopted when they were five study years 27 . p. Communicating the purpose of research to children also requires skilful adaptation of the processes used for adults. 51). medical research. and this is related to their the potential harm or benefit which the research might involve for the child. In regard to research in which the proposed benefit might accrue to those other than the subjects. limits to which parents can give consent in relation There are to children’s participation in research. as much depends on Yet it is not the characteristics of the client and the psychological significance of the data being sought. Therapeutic research is intended to benefit the researchdirectly and has traditionally been regarded as having subject lower requirements in regard to obtaining consent. 1994. even if parents give their permission?’ (Koocher &eith-Speigel 1994. and allowing higher of risk. p. 1983. the question ‘Is it ethical to use children in research which is for remains: the “social good”.The US Department of Health and Human Services has that stated research may be acceptable on children under such conditions the research poses ‘no greater than minimal as long as risk’. considering probability and magnitude. Minimal Risk means that the risks of harm anticipated in the proposed research are not greater. Parental permission does not justify overriding a child’s or young person’s opposition to participation. Such a definition is ambiguous and is ultimately a matter of judge-by researchers and ethics bodies. and older children and adolescents and may be in a position to express their opinion about participating well in research. tests (HHS. The need to assess ment the possible risks involved in participating in research applies to adults as children. K 51). than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or quoted in Koocher & Keith-Speigel.

. 28 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 28 Qualitative practice research in or older (Thomas et al. Such risks can be reduced. Research on child abuse and neglect generally involves domains that consensually regarded as private. repeated interviewing about an experience of induces further trauma. and proved to sion of be particularly useful for children who had limited literacy or who did enjoy reading. data analysis (Chapter 7) and the shift research back to practice (Chapter from 9). However. In some human service fields this applies to almost the entire client population. 24). People are often interviewed personal matters. We interviewed Caroline Thomas.interviewers who are sensitive to the needs of ally trained subjects in the way they conduct interviews. but not eliminated.. there is an obvious victimisation conflict need to gather information for research . a the large typeface and colour graphics. if. it may also be neces. as analogue some clinicians believe. and access to appropriate services should be arranged in the planning stage of the study. The leaflet had photographs of researchers and used simple language.make available to interviewees opportunities to debrief sary to after the research interview. and perhaps even self-fulfilling prophecy that increased may warranted by the knowledge to be gained? (Melton & result Flood. short sentences. and parts not of interview are included in the chapters on tailoring research that to specific groups (Chapter 5). Many ethics review now regard this as a precondition for approving boards some projects. 23– 1994. children whose adoptive parents had their permission were provided with both a leaflet and given an audio cassette explaining the study. The audio cassette enabled the children to hear the researchers’ voices and to form some impres-the people who might interview them. For example. pp. . . are the with the anxiety. Intrusivenes s Qualitative research methods such as in-depth interviewing and observation can be highly intrusive. Research on child maltreatment may be susceptible to the research to iatrogenic effects in treatment. by using profession. . Such work is are commonly as more intrusive than researchers believe it to perceived be . 1999). scrutiny. sometimes relating to loss about highly and trauma.

While the intrusiveness of such qualitative research methods is very obvious. dataa disaggregated way. whether casethe a community.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 29 Ethics organisations and Observational methods can also be very intrusive. an organisation or a family. Thus. with in terms of developing data collection and storage systems in which not possible to identify the research subjects. the perusal of case files vidual or official records. it can be sometimes to disguise the data so that the setting or participants be difficult are completely unrecognisable.in qualitative research this is not always so tiality. or anxiety which the client’s anxiety might affect their ability to make use of what is being provided . One to different of dilemmas of reporting qualitative research is.a service provider may have a heightened vation performance may affect the quality of service offered. particularly to those familiar with the To reduce the risk of recognition. however. other researchers have access to the data in order Should to undertake secondary analysis? Being prepared to allow one’s 29 . The presence of an observer or the awareness that the interview is being observeda one-way screen or recorded on film can affect through the phenomenon under investigation. for instance. As a result of the concern—it is obser.may remain unaware. Who should have access to this material? permission necessary or will agency permission Is client suffice?document-based research be feasible if the permission of Would all relevant parties had to be obtained? Confidentialit y At first sight confidentiality seems a fairly straightforward ethicaland in the research literature is almost exclusively dealt issue. In research that is based on a case study method. can constitute a serious violation of privacy. even ‘unobtrusive’ research about which the indi. it is important to think through the ethical questions involved. such as presenting interviewees’ in responses questions or issues under theme-based headings. that the if purpose of the research is to show the phenomenon in a the holistic way. disaggregating the data can weaken its essence. whenever records are used for purposes other than that for which they were originally intended. But simple. This is not just a methodological also a serious ethical issue. research participants are routinely given assurances of confiden. it is possible to present field. it is Accordingly.

as they might perceive it. The challenge then is to disseminate the voices of those previously the public domain in ways in which privacy is unheard in protected. p.Even when the researcher ‘owns’ the data. once the research partis the public domain the researcher may have little control of over it is used and aspects may be selectively quoted. They above did taperecorded other children reading the transcripts of the adopted children’s interviews and played the interviews to prospective adoptive parents and social workers during their training sessions them to appreciate the subjective experiences of to enable adopted in their own words.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 30 Qualitative practice research in research data to be analysed by others is regarded as an important against fraudulent research. For these data in reasons researchers are usually required to keep their data for a number of years to enable others to have access to it. In contract research the typically belongs to those funding the research and thus data future is not controlled by those actually doing the research. 1999).disguise the verbatim quotes of interviewees by To paraphrasing them would defeat the purpose of much qualitative research. 14). in to see their the public domain can be a deeply violating experience even if their identity is not revealed.. The researchers doing the older adopted children study described this in a novel way (Thomas et al. For a research used participant words used or. p. Researchers must therefore be aware that what they write be may in ways other than they intended. access This create problems when applied to qualitative research as can ‘later different researchers may be inappropriate for projects use by which collected in-depth interviews on sensitive topics’ (Commonwealth Department of Human Services and Health. how ‘Researchers was frequently difficult to control the use of noted that it reports once they became a part of the public domain. This protected the privacy of children the adopted children while evocatively conveying their experiences powerfully than the written word allowed. misused. much more This example highlights the sensitive and individualised ways in which researchers can apply ethical principles. 13). 1995. 30 . 1995. and secondary data safeguard analysis useful as it can allow subsequent investigation of is very valuable relation to different research questions. They expressed about simplistic or sensationalist media concerns coverage’ (Commonwealth Department of Human Services and Health.

Staff at the institution under frequent attack from the media because of had been conditions in the institution and were defensive and resistant to the idea of my research. . formal permission did not really ‘get me in’. with little consideration given to their obligations to the staff or to the organisation (Rosenhan.I was less of a threat to 31 . My decision during this time to base myself in the locked unit assisted in the process of ‘getting in’ ..study on intellectually disabled women living in graphic an institution in Victoria in the early 1990s. . Human organisations are increasingly sensitive to the services political ramifications of research. yet issues in relation to informed consent.. There is a growing awareness these days of the ethical implications of such research. qualitative researchers have described in detail the Few and ethical processes of ‘getting in. Traditionally in the social sciences this has been seen as a political problem of ‘getting in. researchers intent on exposing the practices of one staff in psychiatric institutions posed as patients to gain entry. Research has the capacity to harm their the legitimate interests of the organisation and the professional and personal reputations of the individuals it employs. there is a risk that some organisations political issues as ethical issues in order to may reframe minimisepublic exposure as a result of legitimate adverse inquiry. issue—the In study. Kelley Johnson. getting on and getting out’. getting on and getting out’ political of their research settings. Doing research in institutions today much more careful negotiation of the respective rights entails and responsibilities of the organisation and the researcher. Research can consume valuable organisational resources such as staff also time. 1973). In this case.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 31 Ethics organisations and Ethical and political complexities of research organisations within In the past the ethical issues associated with conducting qualitative research within human services organisations tended to be overlooked. the approval of staff initially proved more difficult gaining than gaining the approval of senior management: . perhaps to a degree that will make such research more difficult to undertake in the future. who conducted an ethno. intrusiveness and confidentiality are equally as applicable to organisations and staff as to their clients. is an exception. Just as researchers once regarded what are now seen as legitimate ethical issues as merely political issues.

due to their fear may that management or the funding body has an ulterior motive and the future of the program is under threat. The reciprocal nature of my involvement 32 . From the perspective of staff. effectively locked away (Johnson & Scott. will feel a heightened performance anxiety. 1997. Issues relating to work and the degree of access which researchers have in the setting need to be clearly worked out and communicated to all of those concerned. p. the ethical question arises as to undercircumstances can the former give consent on behalf of what the latter? There is very little in the literature on research ethics to guide the researcher in this territory. but also This is perhaps particularly challenging when the research is initiated by organisation itself or by a funding body for the purpose of the eval. researchers can get in the way of people going about their normal are a potential source of interference. Staff were angry at the became closure and instituted industrial action bans which included the exclusion of all researchers from the site. needs to be placed in the context of its occurrence within the implementation program. I was 29). Even in an established program the phase of a new staff be apprehensive and hard to engage. ‘Getting in’ is therefore not just a matter of gaining official approval of engaging staff at various levels of the organisation. that the interests of management and service Given providers might be quite different. more difficult. particularly if often the program is experiencing the normal teething problems. staff uation. but sometimes the funding problems have source an evaluation to be built in from the outset. Sometimes that such fears are well founded. confidentiality and possible impact of the research on the participants. We would advise researchers in their communication with all parties and to to be honest develop transparent processes about informed consent. events can Even where unfold threaten the research.to delay evaluation until after the initial able implementation been solved. If this is so requires then an evaluation. my situation for a short time. be it a process or an outcome evaluation. While Kelley Johnson was which undertaking a government decision was made to close the her research institution.If it is a pilot program which is being evaluated. sufficient just to ‘get in’. this has been carefully negotiated. Once the decision was made to close the institution.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 32 Qualitative practice research in other staff since most of my time was spent confined in one unit. The researcher must also be It is not able to ‘get on’ with research participants. It is desir.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 33

Ethics organisations

and

with the staff in the locked unit and the length of time I had been part institution led to a decision to revoke the ban for my of the research & Scott, 1997, p. (Johnson 30).

Interestingly, in the light of this change, Johnson became focused on process of deinstitutionalisation, making her research a the uniqueof the impact of this policy shift on the intellectually study disabled their families and the staff inside the institution. women, This brought added challenges, not the least being that the families thus became research participants as well. The researcher also became the conflictual processes within the organisation as privy to staff the loss of jobs and the uncertainty of the process of faced closure.
. . . many of these families had put their own ambivalence and pain the walls of the locked unit with their relative. Now with behind the decision to close the institution, these feelings were released, and the existing processes of deinstitutionalisation did nothing to resolve them...the complexity of my study increased. I became involved in meetings and in work with the people closing management the institution. I found the gap between these encounters and life in the enormous. Because of the industrial disputes arising from unit the closure decision, I found myself privy to information from groups in conflict with each other (Johnson & Scott, 1997, p. 31).

It is fortunate that most qualitative studies in human service settings as challenging as Kelley Johnson’s turned out to be, but are not all research conducted in organisations will present some difficulties as organisations are complex and dynamic sociopolitical worlds. Conducting research within such settings inevitably adds to their complexity .
‘Getting out’ also creates its ethical and political challenges. Issues differences in interpretation also arise when feedback about occurs the report is completed. While the point of view of before participants process is important, it is also self-interested about a social and embedded in the power relations of the community. The final interpretation has to rest with the researcher, except in action research discussion and negotiation are a part of the research in which design (Commonwealth Department of Human Services and Health 1995, p. 12).

Organisational researcher Richard Scott has identified the later of a study as being particularly likely to be fraught by stages such problems. 33

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 34

Qualitative practice

research

in

...misunderstandings between the researcher and his subjects often to the surface on the occasion when the research findings come are published...even the researcher who does not centre his analysis on deviations (from rules or ideals) of one sort or another may still offend his subjects simply by applying his particular perspective, for he to take an objective and relative view of matters which attempts from the standpoint of his subjects are value-laden and unique. How much and what sorts of things to tell subjects about the research in progress much and what sorts of things to put into the and how published report—these are the kinds of ethical questions to which the open field researcher will find no easy solutions (Scott, 1969, pp. 571–2).

While such issues are relevant in all social research, they are of critical importance in qualitative research in the human services. loops from the research to policy and practice will Feedback often be central to the research, and the tensions which may arise need to anticipated and managed in an honest and open be manner.

Stories field

from

the

While it is tempting to ask for clearer guidelines to assist researchers in their decision-making about ethical issues, it is illusory to think there will be simple prescriptive solutions to the that complexand organisational dilemmas inherent in qualitative ethical research in human services settings. Ultimately many of the dilemmas require individual judgements based on the characteristics of specific situations. Unfortunately there is a lack of case studies or detailed descriptive accounts of how different researchers have grappled with ethical issues in the research process. lack of such accounts may be partly due to the The apprehension researchers feel about exposing their decisions ‘warts and all’ and leaving themselves open to criticism. We hope that the following accounts of how qualitative researchers have first-hand struggled of these issues may help others in their endeavours with some to conduct ethical qualitative research. After all, ‘Ethics is not just a ice thing to have; research is fundamentally weak without n it’ (Deetz, 1985, p. 254). In the following examples, the complexity and, at times, the interrelationship of ethical issues and organisational issues which involved in qualitative research in the human services can be are highlighted . 34

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 35

Ethics organisations

and

Dorothy assessment
Dorothy:

Scott—Child

protection

In looking back on my PhD thesis on child protection assessment what strikes me are the unanticipated ethical and political issues which arose as well as the that the issues I had anticipated proved to fact be complex and more difficult to handle than more I ad expected. This was partly because of my h close connections with the hospital setting in which the research was conducted yet without those close connections it is unlikely that I would have been able the research at all. At the time I undertook to do the research I was not working there but I had previously clinical consultant and group supervisor acted as a to unit in which the research was based, and so I the had relationships with many of the close staff. used in-depth interviews and observation to I intensively shadow a small number of alleged child abuse through the hospital unit, a statutory child cases protection agency and the police. I repeatedly interviewed involved with the same families professionals through- of the case, focusing on the factors out the life to which they were attending in their assessment. Where I also observed episodes of practice, possible ranging from observing interviews with children through a one- screen in the hospital unit, accompanying way child protection workers on home visits to attending staff meetings, case conferences and court hearings. I inter- the parents in their home three months after viewed the cessation of contact with the services. Obtaining the informed consent of parents proved to be a more complex ethical issue than I had antici- At the point at which each case was pated. selected, the parents in this study were in the immediate after- of discovering that their child might have math been physically or sexually abused. While no parent was a statutory client at the time they gave their consent to study, several parents became the subject the of child protection investigation and others a later expressed their fear of becoming so. Some parents were clearly in a state of crisis. In light of this, I chose 35

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 36

Qualitative practice

research

in

to delay approaching these parents in the immediate crisis in order to avoid seeking parental consent at aime when their capacity for informed t judgement most compromised. This entailed might be forgo- collection of data in the initial phase of ing the the thus illustrating the ‘trade-off’ which can case, occur between ethical and methodological priorities. were other ‘trade-offs’ of this nature. There For example, I decided not to tape-record interviews or to interview parents before their involvement with the various services had finished, both significant method- sacrifices, because of the risk that ological parents divulge information which could be might subpoe- legal proceedings. The decision not to naed in give parents the option of being interviewed throughout of service involvement was seen by some the period of colleagues as paternalistic and disempowering. my My reason was that it might be very hard for parents to trustconfidentiality of what they might say to me the whenknew that I was in close contact and on firstthey name with the professionals they were seeing. This terms may an be example of ‘justified paternalism’. children in my study were aged ten All of the years or less, and parental permission was sought to observe social workers’ interviews with the hospital children a one-way screen. This was a standard through practice in the hospital service and colleagues, trainees and clinical supervisors routinely observed interviews in way (with the permission of the parent and this the knowledge of the child). I did not seek parental permis- interview the children as I felt that sion to interviewing could not be justified due to the risk the children of further traumatisation. While the potential risks couldbe quantified, nor could the potential not benefits. However, as the study unfolded, the issue of children’s involvement in the research became less cut. For example, on several occasions in clear the follow-up home interviews with parents the children were unexpectedly present for some of the interviewI had arranged with the parents to visit although attime when the children would not be there. As a it 36

I accepted each social worker generally managed the way the issue of observation. however. For example. the situations were resolved by parents to put the children to bed or arranging deciding other activities. Just because others were observing the interviews for non-research purposes however. Although I was uncomfortable watching in which I was unsure whether the interviews childaware of being observed.there would be people behind the window times and not mention it again. In relation to the child. their social worker would tell them of my absence. necessarily justify does not. it is possible that was the comfortable I felt. a difference between [my] understanding of what the social workers had agreed to and their understanding 37 . While I had rejected the possibility of interviewing children because of the potential harm this might I saw observation through a one-way screen cause. it. In part this belief was shaped by it being routine in the hospital for colleagues to view practice inter. while others informed did the on each occasion and even showed child themobservation room and introduced them to the those observing. I was and that unable to attend.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 37 Ethics organisations and happened. for some unexpected reason.may have been and the greater its effect on vation the interview . Some social workers routinely the child at the initial interview that informed some. but this does illustrate some of the unantic. In relation to the informed consent of colleagues.complexities of naturalistic studies ipated involving children. In seeking consent from the parents for the observation of interviews with their child. I informed them I planned to be present during all the that interviewsif. It was hard views to assess the level of intrusiveness and risk for a particular child entailed in observation.through a one-way screen. as having a much lower level of intrusiveness and risk. the more intrusive the more obser. for children it may have a marked impact on some their capacity to express their feelings and make use of the therapeutic opportunity.

I become increasingly aware of the diffi. and would not using the data compromise the integrity of the research? The matter was resolved. For example. As the research progressed. Yet other when I presented my preliminary findings to the team some workers expressed concern that I had drawn social on that I had observed and heard during all intake while they had believed that I was only meetings at intake meeting to ‘pick up a new case’. probably notthe satisfaction of either party. this was objection an attempt to restrict academic the freedom. if she had not done so should I to secure have on this role? If I had. This placed me in dilemma. This type of data was clients very significant to the research questions. by removing to the verbatim quotations and substituting paraphrasing of comments. as the hospital social worker had intervenedthis. Should data be used which some a subjects was obtained under false pretences.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 38 Qualitative practice research in of what they had agreed to emerged late in the study. I had the very taken detailed notes throughout all of openly the intake meetings [that] included negative ‘off the cuff’ comments social workers and others had made about and other agencies. While he had already been returned home by the stage I became involved. even believed if was not done so deliberately? Alternatively.of presenting the findings of research based culty on 38 . In one case a 0-year-old boy was coercively removed from 1 his family by child protection authorities in a way which appeared both unethical and illegal. would I have been the study? colluding with injustice? Confidentiality also proved to be a more complex issue than I had originally anticipated. would I have taken endangered If I hadn’t. in relation to the hospital social workers [I] believed that they understood that I would be observing intake meetings to describe how a case and agencies involved were perceived. their Another ethical issue which unexpectedly was arose ‘researcher as whistle-blower’—whether it was appropriate for me to intervene in a situation in which malpractice appeared to have occurred.

The study specifically explored perceptions of the services. for future interactions between the participants? My research [thus] generated many questions of an ethical and political nature. some of which cut across both of these categories. lived and worked in the Valley and so was known to she had many homeless people in that community. In her interview with us. .there are a lot probably of documented things in methodologies about work homeless people.interactions.that in a sense the insider bit was my I knew entree into that community and I knew that it would overcome [problems]. real or imagined.. and interparents’ organi. the of being an insider and/or outsider was still central.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 39 Ethics organisations and an intensive analysis of cases without using illustrations which [might] be recognisable to the staff and/or the clients themselves. While this is very different community from negotiating with a formal organisation. which meant that much of sational the related to the often negative perceptions data research had in relation to one another. Anne: I lived in the area for about three years and this had overlapped with when I first started to work there . Anne reflected on the process of negotiating her entry into that particular of homeless people. A fellow who had tried to do some work in inner city about fifteen years before had the actually able to do it and he said that he felt not been the 39 . . issue While an outsider in the sense that she was not a homeless she was person. Can it be subjects said confidentiality has been preserved when a that service might recognise a case in which she or provider he involved and be able to identify the clients was or other service providers whose perceptions about themtheir agency are presented? What might be and the consequences of this. as in the example above.. but few answers! Anne Coleman—Five motels star We introduced Anne Coleman’s (2001) study of homelessness in Fortitude Valley in Chapter 1. about how tricky they are with with outsiders.

. . if you stepped on very sensitiveif you made a promise that you didn’t keep. at one stage a complete insider. At the same time. and work it would look like a very borderline unprofessional kind of relationship because of that closeness and involvement. . if you hurt somebody unintentionally.and he just couldn’t negotiate that at dictory all. For example .. And I about it but it seemed almost hypocritical thought to because even if I slept out for one night or if I me slept two months. Like they’d one thing on Monday and on Wednesday tell him they tell him something else that was totally would contra.. . If you value them [in return] it’s a very powerful thing when you work with them.a continuum conceptualisation where sometimes I was both of those things and I moved along the line constantly. When do that.I had to move a bit past the it was insider/outsider [dichotomy] and I had to come to. the reality was that as soon as out for it too horrible or I’d had enough I still had a got home to and I’d just say.‘Why don’t you go and sleep out?’ someone said. That not only seemed hypocritical.I was accepted to some extent in that community but I was never . . that seems a bit at odds with a lot And although of what’s written in methodology about insiders/ my feeling about it was really shored up outsiders.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 40 Qualitative practice research in people were deliberately playing with him.. if did what people perceived as ‘playing you games with them’. by I had done with people . I guess in the end the only way I could aliseconceptu.. ground. to go and I’d go. but off also something that they would really call me on.. you would immediately be moved— not move yourself—by people along that line from insider 40 . the relationship becomes you unbelievably probably in terms of white social close . I also knew that . there were a couple of worries for me in the insider position. ‘See you all later fellas’.I suppose work an example of that stuff is that if you work with indigenous people the relationship [between you] can become quite strong and I think an element of that is you’re seen as being a valued person in that the community. .

Angelina Yuen-Tsang —Social support networks Chinese working mothers in of Beijing Angelina is a Hong Kong-based researcher who is investigat.or I got quite a bit of stick when the tory way Counciloff a set of stairs where Murri homeless closed people and put up a barricade and then eventually had sat puta mural.support in mainland China—what constitutes help. So. within a split second you could go being ‘sister’ to being ‘that white bitch’—it from would quick and it would be that total and both be that of those positions would be 100 per cent heartfelt and sincerely held. ‘You’re close enough for us to trust we know that you won’t let us down and badly. ing social and 41 .. often not necessarily thing directly connected with me.. but we also We’ve know to those other people and we want you to you talk tell about this’ .I was seen for both those reasons of my as being the appropriate person in that case. . in that I think it was my insider/outsider status that it case was about . . I guess because I’d worked in that for aenvironmentthat idea that I would move along long time. And I think that a lot of what people told me was for that reason too. an indigenous mural that was painted up by people.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 41 Ethics organisations and to outsider..I think all the way through them they pretty clear that the research per se was for were me I think they probably trusted me that they but would get something out of it in the end. I copped quite a lot of stick around other that . me to some. people were very angry and there was no time one official to voice that to. . the continuum all the way through this research sat quite and in fact I think that’s probably what well alloweddeal with some tricky stuff sometimes . So.would happen . . So. . It could be somebody had heardthey’d been talked about at a meeting in a that deroga.they knew that I else was interested in all of this stuff because that was the topic thesis .. ... watched you over ten years.

a year to negotiate entry into one it took about particular community . .. she talks about how she needed to engage with those holding authority as well as make informal connections with formal people at the local level. I don’t want to be given all the cases are very special.. but I idn’t just talk to the high officials..I visited several communities and talked with the local officials. Her research from was published in 1997 Towards a Chinese Conception of Social Support: asStudy on the Social Support Networks of Chinese Working Mothers in A Beijing . This is a study of individuals and families within a particular community and issues relating to gaining the trust of the community were also central for Angelina.cases from the very which good families . 42 .so I looked around in Beijing to try to find And ..I didn’t want to rush things I felt that if I was to do a successful piece because of grounded theory research. So I had to find a where I could stay. finally decided on the Fuguo community I because I found that the local officials received me well and didn’t treat me as an external very researcherinto their community. We talk with Angelina about her data collection and analysis processes. the community that was the best fit. . .. .. They received intruding me naturally.I don’t deal want followed all the time and I don’t want to to be be treated as a VIP. Angelina: I believe in living in the community while doing participant observation. again later Here. and I feel that if I am not immersed in that particular community’s life.. I had to fit into that community and that community had to receive me . I found that they were treating me very as an ordinary academic who would like to just know about their community and there was no more big about my particular research . I cannot understand their way of thinking. where I could live and place have access to the people that I would like to interview . I talked to d the middle level and then the front-line level to see whether I could click with them.. .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 42 Qualitative practice research in under what circumstances will people receive help from outside and their peers and their family members.

. 1996. sort insider/outsider because they knew that I of an was Hong Kong but they treated me as a friend from —a friendly researcher. 2000a. shopping in the community—some the of them treated me as a teacher from Hong Kong. Jackie keeps the team on track [Jackie’s and pushes the deadlines—all teams need this. practitioners and community research members. other publications arising from this research two papers on ethical issues in qualitative research include with families (Munford & Sanders. and a 1999). the cultural the mix. 2000b)textbook on family support work (Munford & Sanders. not happen without the full-time efforts of a researcherrole]. We asked Robyn and Jackie to talk about the composition and func. academic and community partnerships and a involving diverse team of academics. Sanders et al.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 43 Ethics organisations Yvonne: and Angelina: To what extent were you treated as an insider or an outsider? Were you always an outsider or did you develop more of an insider status over time? to the local officials I always remained as I think an outsider . it evolved and the changes that occurred in how the over time? What it was like for each of you to team be part of this research team? I am very clear that this research would Robyn: Well.. Robyn Munford with successfully families and Jackie Sanders—Working We introduced Robyn and Jackie’s research on family support in New Zealand in Chapter 1.of the research team as well as well as some of the tioning ethical organisational and issues they encountered. . Dorothy: Could you say something about your research team — different roles which people had. living in community. In addition 1998. research 43 . Munford & Sanders. but to the people in the community — because I’m hanging around the community. This was a large services project.. to the research reports (Munford et al. Jackie and I re the longest-serving members of the team and a have maintained the consistency as others have come and I believe the secret of the success of the gone. So. 1999).

such as interview. gender. amazingly to this process and has had an dedicated ongoing commitment to always contribute to the researchthe fact that she has ‘a day job’ as well. By this I mean the research as an activity is a central part of their working week. running of 44 . despite Her into the university is really vital and we are link more lucky to have this connection and her than incredible commitment.particularly in managing the relationships lenges. We have really worked on the concept of the of teamwork and I feel are now at a point where we a clear understanding of what being a have team We have three people who are committed means. There are a lot of will chal. We have been really lucky to have ongoing fundingthink Robyn is right—without someone and I whose is to keep the research going. This core of three people do most hands-on work. is just money.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 44 Qualitative practice research in Jackie: is the team and the commitment to team meetings and clarity around tasks. We continually review our adds indi. say in teaching on to or practice. that really someone’s constant attention. who are We have a mix of cultures and experience and also this to the diversity. of us are employed part time with Two research and the other one. Robyn. research like job it this often fall by the wayside. assisting with recruitment ing.focus groups. We also have three team members who p bring differing perspectives in terms of culture. We are now fortunate to have on our team community members part of the agencies we are researching. community. keeping everything of the linked and maintaining a focus on both the together specific activity of the day as well as the research wider program of which each project must be research a art. end user. Goals are set and reviewed and members are clear about their roles. with service delivery staff and with funders. and who also from time to time undertake specific research tasks. then it is always going to fall off the bottomlist. to completing the research.contributions to the team and reflect vidual on progress to date. If this is require addedan already demanding job.

have a clear and significant contribution to make in terms of supporting us in the daily bringing a range of perspectives to work. I think it is impor. These other team members . for instance. There were some related issues around vice the that some of the qualitative research way strategies could ‘mimic’ early intervention work because of the emphasis upon building a relationship with participants and talking through the issues that brought them agency in the first place. the timing of interviews. In both of these situations. . for example. we used a fairly standard approach to problem resolution—being avail-to talk issues through until all parties were able happy that they had been resolved. Can you something about these issues and the say processes you developed for addressing them? We obtained ethics approval early on and have reviewed ethical procedures as necessary. and being open to modifications to research design so that new information about the impact of the research on practice could be heard and effectively responded to. We did not to the anticipate although once it happened. it seemed this issue. 45 . so obvious. involved some particularly complex Your study and interrelated ethical and organisational issues. were if a worker did not want to participate but a client did or versa.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 45 Ethics organisations and Dorothy: Robyn: Jackie: participants. . If our research caused problems for the agency and became we reviewed this and changed our intrusive practice. analysis and so on. This seems to work well and provide people with a lace to make a really valuable contribution to p the ongoing development of the research without having that they need to ‘go out there and do to feel some research’ or they have not contributed. analysis and other activities and to ensuring that we keep the focus on the bigger picture.be aware that in agency research the tant to primary responsibility is to make sure that the intervention takes place—that the client gets the support they needresearch does not have a right to undermine and that. issues related to the range of consents The ethical that required and what happened. planning data collection.

this means we get a less than ideal data times capture. Comment s It can readily be seen from the above examples of qualitative the researcher is indeed the instrument of their research that own research and that the interpersonal relationships and dynamics emerge are complex and deeply charged for both which can the researcher and the researched. Some. go in agency research we need to be able to be more responsive to the particularities of each situation we encounter and to modify our plans accordingly. Our background in qualitative research helped us to deal effectively with many of these challengesof the focus on developing and because maintaining good research relationships that will sustain intense So our work emphasised working research. I think the organisation was very of this issue and handled the boundaries aware really well. Such research calls for researchers of highest personal and professional integrity with a deep the capacity for reflexivity. Another issue was the potential for research data to be used in staff evaluations. is the way it is and we do not have a right but that to undermine interventions in the pursuit of good quality data . We maintained research records really carefully and in fact were never to furnish information to the organisation asked about performance issues.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 46 Qualitative practice research in So whereas in other research we may just be able to out there and capture the information we need. This was one of the reasons behind the decision to set up the that lay researchas functionally separate from the service centre delivery the organisation and also for building a arm of strong relationship with the university. with to find solutions to their concerns that people could be fitted into our methodology rather than then simply them ‘that was the way it had to telling be’. When we listen to these direct accounts we can see how unique each study is and how guidelines for conducting ethical research 46 .

can pose serious risks to their wellbeing. research is an essential tool in improving services and making them more accountable. It is a a voice powerful one to be used with tool and care. However. Qualitative particular has given groups of people previously research in denied the opportunity to be heard for the first time.research. argued that it may be unethical not to do research in It can be the human services in which the community invests scarce resources professional practice is itself often an untested and where social experiment with the potential to hurt as well as help individuals and their families. 47 . the risks must be balanced against the possible gains from conducting such research.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 47 Ethics organisations and can only ever be that—just guidelines. In that sense. particularly with people who have tative additional vulnerabilities. We can also see how quali. while no research is without risks.

from selection of participants. the perspectives of members of a range of groups. The in-depth interview takes seriously the notion that people are experts own experience and so best able to report how they on their expe. This is followed by an outline of the interview process. through the initial contact. Ultimately the choice to use 48 . This is hardly surprising. should be obtained. in-depth interviews have their relative strengths and weaknesses. such as clients and workers. The latter part groups of chapter includes two field stories of research that used inthe depth interviews . students and parents. Choosing interviewing in-depth Like any method of data collection.itself and ending. If we interview rienced different people about the same event or phenomenon. Where the research question requires it. we will inevitably get a range of perspectives. We have also included a section on view focus as a special type of interview situation. or teachers. approach given the common concern of qualitative researchers to understand the meaning people make of their lives from their own perspective. the inter.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 48 3 In-depth interviewing In-depth interviewing is the most commonly used data collection in qualitative research. chapter commences with discussion of the relative This and strengths weaknesses of in-depth interviews as an approach to data collection.a particular event or phenomenon.

1997.Sometimes interviews will be most appropriate. or tests are being performed Brown & Canter. Holstein and Gubrium talk of interviewing as an active. or explicit sharing and answers and/or negotiation of understanding in the interview situation which is not so central. 49 . Their immediacy and relational quality afford considerable flex. Respondents much repositories of knowledge—treasuries of are not so information awaiting excavation. meaning-making process. (Brenner. nor Meaning is simply transported through respondent replies. On this point. so to speak—as they are constructors of knowledge in collaboration with interviewers (Holstein & Gubrium. 1985. Any misunderstandings on the part of the interviewer or the interviewee can be checked immediately in a way that is just not possible when questionnaires are being completed. Very often the serve researcher to weigh up the pros and cons of a number of will have approaches the best choice available in the and make circumstances. have particular In-depth strengths. tions. p.to the data collection process. and often not present. is arguably more apparent the less structured and and more conversational the interview process. both in terms of areas ibility explored direction of the discussion. These are unlikely observation to all-or-nothing questions.interviews do. The advantage of being able to clarify what the other means. p. 3). There is an implicit. The sought best collection approach for any study is that which will yield data data best meet the research purpose and answer the research that ques. 114). therethen. though. Brenner. however. it is actively and communicatively assembled in the interview encounter. in other research procedures. they share the general advantages of face-to-face interviewing. sometimes or the analysis of existing records. or to provide an alternative data source that may to strengthen the overall findings. First. a combination be of approaches will be indicated—to answer different parts of the research question.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 49 In-depth interviewing them or not must be made in relation to the nature of the data and the practical constraints of the research context. active . Both parties to the interview are necessarily and ineluctably not merely elicited by apt questioning. In many cases. and the Brown Canter and say: Probably the central value of the interview as a research procedure is that it both parties to explore the meaning of the questions allows involved.

just as events have inevitably been reconstructed at the time of the interview. Interviews then. focus on how tends to displace the whats —the meanings— that serve assignificant the relevant grounds for asking and answering questions p. also happen at a cross-section in time and. Nevertheless. will the itself impact on participants’ subsequent organisation and understanding of their experience (Kleinman. 115).Qualitative Research – TEXT 22/11/2001 3:26 PM Page 50 Qualitative practice research in Viewing interviews in this way requires attention to the interview context as well as the content of what is said. A narrow turn. no matter that events. The process of telling their stories about the past & Frank. and over. While the emphasis on process has sharpened concern with. to process and the hows as well as the whats. what happened in the past. In-depth interviews are particularly useful when the phenomena under investigation cannot be observed directly (Taylor & Bogdan. even what they say now about how they felt but this does not give us access to the past. These retrospective and anticipatory are elements a world of experience that is not accessible via open up methods observation. in present. They also enable us to think or talk people about events that happened in the past and those with that yet to happen. further reconstructions are undoubtedly to come. in what is being conveyed interviews by respondents. Other than through diaries or other such as records the time. Even then. The only perspective that can be obtained is that of the present. we are as reliant on what the reporter chose to write down at the time as we on what interview respondents choose to tell are us. debate it is important not to lose track what is being asked about in of and. the epistemological status of interview data. 1998). and particularly in the interview context. Holstein and Gubrium caution against being so concerned with interview process that is actually said is what lost. 1988). As Bruner states: 50 . Thus they are an excellent means of finding out how people feel in relation to a given topic. thoughts and feelings being reported have the already We can find out how someone feels now about occurred. interviews in the present are the only way made at to access a person’s perceptions of past events. (1997. Participants’ perspectives can only be presented in the context of lives as they are being lived (Langness1981). It is important not to let this ability to talk with others about past experiences lead to a false sense of access to the past.

for with each new telling There is no the context varies. 1986. process participants control its process. and the absence of the questionnaire may obscure this the more (see Finch. on how research participants experience process and on the overall quality of the the research research. an impact. from involved the selection of participants through to the ending of the interview Each of these stages is important and will have encounter. 1988 on this point) all (Ribbens. p. . there is always so that power in the researcher role. interviews allow access to what people say but not to what do. The only way to find out what ‘actually happens’ in a they given situation is through observation (Chapter 4). all stories are partial. The process interview In-depth interviewing in qualitative research involves much more the actual interview interaction. (Bruner. as a set But narratives change. and Stacey. . fixed meaning in the past. The particular paradox that is worrying about depth interviews is that give the interviewee the power to control the interview itself. We now consider the than stages in setting up and conducting in-depth interviews. all meanings incomplete. Finally. 1989. There are other reasons to be cautious when using in-depth inter. the story is modified . 153). to be objectified in an depart interview In the end you are very powerful in this style transcript.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 51 In-depth interviewing Stories give meaning to the present and enable us to see that present of relationships involving a constituted past and a future. 1984. you and as a result they put themselves very much in your hands yet by exposing themselves in a one-sided relationship. Finding participants and selecting Experience of the topic under investigation and articulateness regarded as essential criteria for the inclusion are commonly of 51 . When you come toyou take their words away. Ribbens expresses concern inherent aboutissue. 587). particularly in relation to the vulnerability of this research participants vis-à-vis the researcher.No matter how much we try to construct the research views. for better or worse. p. the audience differs. of interviewing.

not always possible. 1998) is a useful approach to selecting participants. analysis and further data collection. a general rule. they rather add variation and depth of ’ (Strauss & understandingbold in original). the relative diversity or homogeneity As of 52 . 47). 1967. Later in this chapter. further participants are sought the specifically for their capacity to help fill gaps in the data that are thrown up in analysis . from whom. Such cases (Strauss don’t necessarily ‘ negate our questions or statements. 1989). Angelina Corbin. of course. be able to ‘develop some significant relationship with the phenomenon (1987. in order to obtain the broadest those possible the range of perspectives on the topic under reach of investigation. 109. Yuen. For Wertz and van Zuuren. With specific reference & van to phenomenological research. participants need to have or p. p. p. process Afterfirst few interviews. we would have the ‘dream team’ of research participants. It really does depend on study. or disprove them. Covering the range of experiences involves deliberately out seeking whose situations and experiences are different people from already obtained.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 52 Qualitative practice research in participants in qualitative research projects (Collaizzi.talks in some detail about how she used a similar process Tsang to obtain the participants for her study. 1994). while certainly having experience of disability. have limited capacity to report it articulately. and the likely variability of experience of the phenomenon under investigation. Generally speaking. Strauss & Theoretical Corbin. no easy answer to the question as to how many There is participants are required for a qualitative study. is sometimes referred to as negative case The process analysis & Corbin. 1990. In Chapter 5 we talk with Tim and Wendy Booth their approach to interviewing people with an about intellectualpeople who. Theoretically. 1987.If this condition were present in every participant. This is. whether or all the stages of grounded theory are being used in a not particular theoretical sampling. In cyclical of data collection. sampling (Glaser & Strauss. Morse. optimal numbers will be determined by the research topic and question—what you are wanting to find out. the research topic. there will be both theoretical the and practical considerations. 1990. Polkinghorne sees it as a requirement participants have ‘the capacity to provide full that research and sensitive descriptions of the experience under investigation’ (1989. Wertz Zuuren. 1990. under study’ 11). Polkinghorne. the researcher engages in a study. 1978.

from data collection through to analysis. within the constraints of time and other resources. it can be safe to stop. numbers. it wide of views and situations to build up a broad understanding range of topic. family contact. Where new broad patterns do not appear to stop be emerging. fathers’ and children’s experiences of postdivorce or Darlington and Bland’s (1999) study of client. Experienced researchers will know. it is important to a study to interview a players—as in Trinder. where interviewees’ perspectives are confirmatory rather than contradictory. however. literature beginning analysis) while building up the review.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 53 In-depth interviewing experience of the topic will impact on how many participants are needed. there is always something new to hear. document analysis and interviews. If. Even so. wise to obtain the maximum number of it would be participants order to document the extent of the views or possible in situations and to avoid the charge of choosing only the few identified cases fitted the researcher’s own that perspective. data collection has to somewhere. labour intensive and time consuming. Some of the ‘it depends’ issues here relate to the number of study groups. Above all. each its own pros and cons. often.will be important to talk to people representing a enon. say observation.each person’s experience and perspective will be While and different. and worker experiences of hope in relation to mental illness—it would to limit the number of each group. Similarly. however. Where there are many possible experiences of a phenom. Researchers often have to take as many personal participants as they can get. this will limit the total number of each participantsrelative to available research possible resources. that potential research participants are not always easy to find. Beek and Connolly’s (2001) number of study of mothers’. If 53 . In most cases it will not be possible to decide theoretically at the beginning just how many participants will be needed. Qualitative research is. in situations where there is relative the homogeneity. or if multiple interviews are to be conducted with participant. It is common to have to try a number of avenues. if a be wise combination of methods is to be used. leave plenty of time for with data collection and have something else to do (for example. for example. so there will be practical constraints on the number of people who can often be interviewed. the willingness to commit to reflection on deeply experiences. Participation in qualitative research requires a considerable commitment of time and energy and. the number of methods and the number of interviews.

of sufficiently enter into thorough exploration of the issues under relaxed to discus-or trusting enough to share their thoughts with the sion. this extends to the participant’s trust in the researcher’s behaviour and integrity at every stage of the research. interviewer. the ability to pick and choose respondents on theoretical grounds is a luxury we have rarely encountered. and once this is done the researcher can get on with the business of researching. we can’t use to state it data. For the researcher. will a great treat participants fairly. experiences project requires deal of trust—trust that the researcher will listen. Participation in a research about personal. if the researcher is not careful. In reality. It is relational. It develops. participants may take it for granted that standing the interviewer understands what they are talking about and skip over important aspects of their story. it requires be genuinely interested in the issue being researched that they andthe participant’s experience of it. will respect their limits about what they want to and will deal with the data fairly.develops. no matter how deep the sense of shared as understanding between interviewer and interviewee. the researcher–participant and relationship is subject to continuing negotiation and reworking. Without some sense of say. connec. the data collection process. is often included in research texts as an entity that Rapport is established at the beginning of the research.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 54 Qualitative practice research in obtaining numbers turns out not to be an issue. It could be reported as the interviewer’s experience but not as that of the interviewee. Making rapport) a connection (establishing The development of trust between researcher and participant is an essential part of the research process. But rapport is not a finite commodity that can be turned on and off by researcher.relationship. and perhaps traumatic. respondents are unlikely to be either tion. or does not. 54 . between the the researcher and the research participant. Like all relationships. As a sense of shared impede under.this interest and concern to the nicate participant. times when a strong connection between There are the researcher and participant can. We really do want to know exactly what participants think about a situation and at times we want them the obvious—if the participant didn’t say it. and that they be able to in commu. then the theoretical considerations can take precedence. It is not ‘established’ once for all.

this can meeting place on a separate day to the interview as this gives will take the participant time to follow-up any further questions that may arise for 55 . these 104). The contact initial Prior to meeting the researcher. but mistaken or incomplete. Where differ. to meet the person tions with they are being asked to talk about themselves. stage. ethnicity. it is well worth employing interviewers who share of the characteristics of the study group than the more researcher. enable people to tell their stories. may be constrained and potentially self-censor participants what say as a result of presumed. This can especially be issue in ‘insider’ research.they may have and. p. of the experiences. The initial contact (we prefer a face-to-face meeting) is their opportunity to find out more about the study. it is important that permission is obtained at are to be this Once discussion about the research process is exhausted.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 55 In-depth interviewing There is a related risk that the interviewer will think they know what the participant means and impose assumptions on the data without checking them out with participants. and prospective participant indicates they would like to proceed. Ideally. to ask any ques. youth workers and street-frequenting young street people and used them to recruit as well as interview young people. gender. Miller and Glassner (1997) suggest that too close an identification with one position in relation to the social phenomenon being investigated ‘which cultural stories interviewees may tell and may restrict how may be told’ (1997. age. they shared understandings between them and the researcher. Similarly. in their way. the they then be asked to sign a formal consent form. prospective participants have presumably been sufficiently interested to respond to information they have received about a study. Pe-Pua (1996) hired bilingual interviewers from among workers. If whom interviewstaped. dress or language—and where resources allow. attitudes and values of In her study streetfrequenting young people of non-English speaking background in Sydney. where researchers may assume an shared understandings between themselves and the participants. there needs To to be an openness to whatever perspective may emerge. most importantly. but have not yet consented to participate.between the researcher and those being interviewed are ences such rapport may not be easily established—differences of that race. disability.

in which they may still change their mind about participating in the study. as well as the vulnerability of the participant group and the wishes of individual participants. many of the advantages of in-depth interview will be diminished. under what circumstances might referral occur—would it self-referral only or would there be circumstances in which be the researcher might make a referral on behalf of the participant? The decisions made will depend very much on the research topic.relatively straightforward where participants have This is been recruited through a professional agency. On the other hand.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 56 Qualitative practice research in them and also allows a cooling off period. on whether appropriate resources are readily available and on the capacity of participants to access them. For example. the nature of the 56 . has a structure of some sort. Every interview. the idea of an ‘unstructured’ interview is the really a myth. pants. including the purpose of the study. An immediate issue is to be clear about the participant’s right to terminate the interview at any time. There is a fine the line between having enough structure to facilitate talk (why are we here?) and imposing a structure that becomes constraining for the participant. The interview itself is a structured social interaction— the act of setting it up brings its own structure and context. if and of the researcher imposes too much structure. This should always be discussed with partici-in order to work out a process for this to happen. The interview While interviews do vary in terms of how focused they are and the extent to which participants are encouraged to ‘direct’ the flow of conversation. no matter how free flowing in terms of topics and the order in which they are covered. Where the research topic has the potential to be distressing the for participant. it is important to work out a plan to deal with such distress should it arise. Often research ethics committees will require that provision human be made for referral to a trained professional should participants become distressed or disturbed during the course of the research. very There be little point in conducting an interview if the would interviewer some idea of why they wanted to talk with that did not have person what they would like to talk about. The extent to which interviews are focused depends on many things. At other times it may be necessary for the researcher to set up an arrangement with an agency for referral.

two together was a new experience . So I’d say it would be the fact never that it interview. however... Research interviewing can certainly into have therapeutic effects. and people with formal training and experience in communication and counselling skills start with something of an advantage. On this point. different types of questions will elicit different responses. Research interviews carry no expectation that this that interviewee will change their lives—we are. It is important that interviewers understand their role as researcher and do not slip counselling mode. this afford interviewees a greater capacity to explore their may experiences than is possible in some therapy contexts.I definitely didn’t feel under pressure here was an but sometimes in counselling or support group I feel under pressure so defensive and back off (Darlington. It does not necessarily commit the interviewer to covering those topicsany particular in order. While there are similarities in the communication skills used in research interviewing and clinical interviewing. but it may be precisely because it is not therapy is possible. 1993.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 57 In-depth interviewing data sought. In-depth interviews may also be longer and in them as less focused than interviews in therapy contexts. They are only less likely to be experienced as intrusive but may not yield explanations anyway. I was feeling that little girl againhas not happened to me. p. after all. seem ‘what’ and ‘how’ sometimes implicitly contain elements of ‘why’.. Becker says: 57 . ‘Why’ questions. This could be as simple as a list of topics to be covered during the interview.and feelings. In most cases an interview guide is recommended. Effective in-depth interviewing requires considerable skill.. may interrogatory and can lead to dead ends. I tend to get 111).. Descriptive questions about what and types of how happened are particularly useful in encouraging people things to describe their experiences. One participant in Yvonne Darlington’s sexual abuse study said: In the interview I could almost at times just see myself again and actually talking about things. the interested they are. Interestingly... on the other hand.I’ve been in counselling and I’ve let been in groups where a lot of this does get talked about and support I’ve had that same feeling. Paradoxically.I think maybe I To get the just go of something. and pragmatic concerns about time and other resources. the purposes of the two forms are. I’ve had some memories but no that .. In any interview. different and should not be confused.

Reflexive questions.studies require attention to subtle nuances of Some questioning in order to elicit the type of material sought. . were less when I constraining. where. reflexive). . as opposed to narrative ‘How?’ would likely evoke . influencing. .to the researcher. to tell a invited them story included whatever they thought the story ought to that include in order to make sense. The idea of iting using ‘influencing’ questions seems anathema in social research. ‘When you say . . If a word or concept is unfa. 58– anywhere 9). They are less useful in on how elic.a complex event or phenomenon. and we cannot think of a place for Tomm’s strategic questions. Clarificatory questions are also useful. Could you tell me a little more about that?’ sure what can provide the necessary clarification and open up further exploration. Lineal questions the what. to answer in any way that suited them. including their tions of thoughts others might see the situation.asked them. The types can be seen as reflecting two main types of interviewer intent: orienting (lineal. strategic and reflexive questions was developed for a therapeutic context but also has applicability to research interviewing. andsuch can be useful in opening up discussion of people’s as percep. 46). which are leading. I’m not miliar exactly you mean. ‘How?’ the simple questions. circular. the ‘what do you would happen if’ and ‘can you imagine’ questions that think may facilitate talk about future hopes and expectations.direct narratives (What happened next?). pp. 1998. as though you were asking about the deeper meaning of things. directive and have a corrective intent.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 58 Qualitative practice research in Somehow ‘Why?’ seems more profound. gave people more leeway. p. didn’t to be trying to place responsibility for bad actions or seem outcomes (Becker. Most researchers go into in-depth interviews with some kind of interview guide or list of topics to cover at some stage during the 58 . more intellectual. Where the focus is on descriptions of experience rather than descriptions of objects and actions. are facilitative in intent and can be useful while also in opening up discussion of hypothetical situations. circular) and influencing (strategic.Tomm’s (1988) typology of lineal. how and why questions that include all are commonly used in research contexts and are primarily investigative in intent. questions such as ‘What did you experience?’ and ‘What was it like for you?’ are more to elicit experiential data than questions such as likely ‘What happened?’ (Polkinghorne. as is often the case in interpretive studies. They didn’t demand a ‘right’ answer. a simple. 1989. Circular questions are more exploratory in their intent.

up everything that is said. less is fast generally best. for example when interviewing children. Always use new batteries and always take spares. using the other for backup when something can’t be heard properly. For than focus two tape recorders may ease anxiety about picking groups. Even if it were possible to take notes at the speed that the interview this would be very distracting for the interviewee progressed. While in that a practice many interviewees become used to the presence of a tape recorder quite readily. be taped.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 59 In-depth interviewing interview. Tape While good quality equipment will take you part of the way. in the that best facilitates that. As Burgess (1984) found in his way inter. it makes sense to use equipment that is least likely to cause distraction. Any attempt to rigidly follow a guide in a set order will inevitablythe conversational flow and cut across the aim of interrupt encouraging participants to talk about what is of concern to them. This can serve as a useful reminder of core aspects of the research question to be asked about. While there are no hard and rules as to what should be in an interview guide. a busy road outside. a flat microphone on a table works well. television in another room. being under path—these day-to-day noises do not stop just because flight a we 59 . Smaller is generally better. For interviews with more one person. preferably with a builtin microphone. Too much detail will be difficult to read and will distract the interviewer’s attention from what the interviewee is saying. different people’s stories unfold views in different ways and what may seem a logical order to one person be quite inhibiting for may another. Interviews are often conducted in people’s homes where there are distractions that may not be able controlled—children playing. quality is always a concern in qualitative research. and make it virtually impossible for the interviewer to attend to also the crucial relational aspects of the interview. Transcribe from one.with schoolchildren. there are considerations to naturalistic research than finding more the optimum taping conditions. Grbich (1999) suggests a roving microphone for interviews where you are likely to have to move around. Recordin g In-depth interviews should. It goes without saying good quality tape recorder is essential. to be the neighbour mowing the lawn. if at all possible.

It can be helpful to let the participant know where youin the process. Further ongoing on how to contact have been negotiated. we need to be tance carefulcross that bridge where we seem to be dictating to not to our research participants how they should be in their own homes. impor-of relative quiet and uninterrupted time. we need to think of the ending of each interaction. Being clear the time each has available to talk at the beginning of about the interview establishes the parameters of the interaction. for example. The important thing is that may well both researcher and participant are clear about what is expected and that researcher both follows through on mutually agreed the arrange. Ending Every social interaction has a beginning. Endings should not come as a surprise. Even so. in a manner for satis. ending of each interaction is no less important and needs The to be planned for. not just five minutes before the end of the interview. There will be times when the only option is to pause the While it is helpful to discuss at the initial contact the tape. The interview content itself can impact quality—people’s voices may fade if they become on tape distressed.to all parties. Such reminders should seems not to be given in plenty of time. particularly if the participant has become of the very involved in talking about their experiences and the conversation be coming to a natural end. if there are to be multiple interviews.does not create expectations of ongoing involvement ments and that are not able or prepared to they meet. will always be obligations to report back to participants There at some stage—being given a copy of a research report or information access it would be minimal.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 60 Qualitative practice research in want to interview someone. each part has to go well. 60 . Research factory interactions no are exception. a middle and an end. ‘I have just a couple more things are I’d to ask you about’. and the interaction to be completed successfully. it is generally helpful to include some reminders towards the latter part interview. In considering endings. they Also consider that you may need to leave time to negotiate another if an interview is unexpectedly long and time is running contact out. Always ask whether there is anything like else would like to say—and leave time for them to say it. as well as the ending of the participant’s involvement in the research.

and feelings. p. 61 . 434). problems. it is important to have on topics a ebriefing strategy worked out beforehand. the more intense the interaction. as can talking with a supervisor or trusted colleague. It shows you have an a enuine interest in them as a person and not just a research g topic. In the case of research that are potentially disturbing. but be careful not to jeopardise confidentiality . Focus groups share many of the advantages of in-depth interviews as a means of data collection. Sometimes the researcher will need some assistance to work through emotions aroused during interviews. It is unlikely sage to signal the opening of the floodgates of never-ending contact. Writing down d your reactions can be a valuable aid to debriefing. as some researchers may fear. for example. focus groups have been increasingly used in social science research since the 1980s.and the trend is towards expansion of the possibilities ing area for use rather than narrowly prescribing how and when their they be used (Morgan 1997). Focus groups Long used in market research.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 61 In-depth interviewing Generally. focus groups should varya continuum from highly structured through to on relatively unstructured and data obtained from focus groups can be analysed quantitatively or qualitatively. Remember that in many cases people achieve a level of sharing and disclosure about themselves in in-depth interviews in everyday life. the more time people will need to wind down emotionally. Like interviews. it is important that they feel they that is rare have dealt with sensitively and are not left feeling emotionally been raw or used. that ‘Focus group interviews are particularly well suited to collecting qualitative data about individuals’ definitions of in-depth. and meanings associated with opinions various phenomena’ (1987. This is still a develop. It is more likely to provide the time and for a real and meaningful interaction that enables space both researcher and participant to effectively disengage from the relationship. yes to a cup of tea that is offered can be an important Saying mes. Always leave time for day-to-day conversation at the end of the interview— some some will not want to engage at this level but for others it will people be important part of the closure process. Basch says.of acceptance of the interviewee as a person.

Focus groups also place limits on the amount of time ences. These relate to the extent to which participants may experience peer pressure to remain silent about views or to readily agree with more dominant views in some the group.to opt out if they do not feel able to participate tunity comfortably in group discussion. It is in the initial pre-group meeting that trust to be established. with any qualitative research.will be valued. each participant has to speak. and don’t that will be ‘protected’ from the group if needs be— they prospective are not likely to agree to take part. they participants may choose to participate and. so participants had developed considerable trust in the researchers by this stage. following two inused depth interviews. Osmond and Peile (2001) used focus groups in a study of child welfare workers’ use of theory in practice. 1987. may be reluctant to talk about personal and participants experi.activity involved presenting back to participants The first the thematic analysis of the material obtained from two rounds of 62 .is also in relation to the group interaction that potential It disadvantages of focus groups arise. feel comfortable about sharing their own experiences. In-depth interviews allow for concentrated and uninterrupted focus on the perceptions of one person and are preferred when this is sought.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 62 Qualitative practice research in Particular advantages of focus groups relate to the benefits of group interaction. such as the extent to which the cross-flow of communication sparks ideas that would not emerge as easily in a ne-to-one interview. Worse. their contri. Morgan. The group context also enables exploration of a range of subjective responses in relation or more topics in a relatively short space of time and to one relatively economically (Basch. Darlington. The focus groups involved two activities. Unless there is trust in the facilitator—that they will be heard when they speak. Mariampolski. be at emotional risk in the process. 1989. if their trust turns out to be unwarranted. may enable participants to riences. participants in focus As groups should be given as much information as possible about the purposeresearch and the topics to be discussed and given the of the oppor. With sensitive topics there is the potential for embarrassment.in a supportive environment. Groups also take the pressure off o participants to every question. They were as a third stage of data collection. Hearing others talk about their to respond expe. Issues of trust apply to focus groups begins as much as they do to in-depth interviews. they will not be pressured to speak when bution they want to or about things they don’t want to talk about. 1997).

attention needs to be paid to designing the types of questions that will be most effective in elic. Mariampolski (1989) suggests commencing with relatively ‘safe’ issues and encouraging everyone in the group to speak early on. For sensitive topics. As with in-depth interviews. Angelina Yuen-Tsang—Social support networks of working mothers in Chinese Beijing We introduced Angelina Yuen-Tsang in Chapter 2 where she talked the process of choosing a community in which to conduct about her research and her engagement with that community. at some stage the focus group will have feelings to move beyond description of events in order to find out about participants’ experiences of them. in order to reach some closure to the discussion of potentially stressful subjects. We were impressed by their willingness to discuss their practice in the group and so risk exposure. We followed this with a theory representation exercise that examined how participants utilised theory in practice.shared by participants—although the latter could have led setting to competition and defensiveness among participants had trust not been present. The study had 63 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 63 In-depth interviewing in-depth interviews in order to check the validity of interpretations made.the winding down period at the end. A return to ‘safe’ tant is topics is advisable. Stories field from the The first story from the field is from Angelina Yuen-Tsang’s study social support networks of working mothers in Beijing. of the the second from Yvonne’s study of women who had been sexually abused in childhood. If the focus of the research is iting about and experiences. Participants were requested to portray diagrammatically their perceptions of they integrated physical abuse theory in their practice and how then to explain their drawing to the group. We attributed this to the trust that had developed between participants and researchers and the commonality of work.the type of material sought. Equally there impor. only moving on to more sensitive topics when is evidence that participants are ready to do so.

The participant observation was also very important felt that. and I tried to observe the community at different times. For of social support. observation interviews with 27 working mothers. what the topics of their conversations are. I’d chat with the themsee the pattern of their daily life and how and they interact. I asked people about their pattern of. what their concerns are and things like that. even though I didn’t follow it the end. how they chat. including focus groups.I felt that I needed to know more about munity. 64 . Here she and in-depth talks the three stages of data collection and the process of about selectingwomen the to interview.. partly because I felt that I didn’t enough about the culture in the Beijing know com.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 64 Qualitative practice research in several stages of data collection. I didn’t learn a lot about how social support is being rendered in mainland China. So I lived in a local hotel right in the middle of the community. et cetera. and So go out at different times of the day and try to I’d see pattern of their daily routine . So I used focus groups to me to underlying issues that I may not be alert aware example. in order to understand the because I local community and its dynamics. Focus groups observation Angelina: and After gaining entry to the community. Even though it talked a lot about social support. I also asked them to comment on some concepts or ideas from the western literature. their perceptions. .. For example I’d get up early and wander around the community very because the community virtually woke up at around five or six a lot of people were going to work and so on. I have to know the community well. This helped me to design my question guideline. but at least the focus groups alerted in me a lot of issues which I was not aware of to earlier. I started to do some focus groups. . the issues they encounter and any tensions that they experience. because way most of the available literature was from overseas countries. their of life.

cultural issues around the process of Were there interviewing? I’m wondering was this an unfamiliar kind of encounter for the women? 65 Yvonne: Angelina: Yvonne: . after the first interview I decided to scrap the guideline and let the conversation just flow. Also. Angelina: The interviews Angelina: I remember that for the first interview. they encounter—do they really receive the help that they would like to receive and what are the consequences. And also helping me to understand the context of their because when I started to do the lives interviews..addition. I then decided not to use it at all tions because that if I followed the question guideline I found mechanically.from it.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 65 In-depth interviewing Yvonne: So a lot of that was orienting you to the community. I asked the interviewees’ permission and usuallybeginning they were a little bit cautious in the but gradually they began to enjoy doing the interview.. it ruined my relationship with the interviewee of conversation. I needed to know what they were talking about and to able to strike up a conversation. so that.. did you tape the And interviews? Yes. ques. So it helped a be lot. to get some sense of what life was like helping you in that community. events.so I still had a set and the flow of questions at the back of my mind but these questions free-flowing and open-ended. during these events critical what kind of support do they receive. from and how and what kind of difficulties do whom. I wanted to know their et cetera.forthe most important question was to find out me about the flow of their life course and the most critical they encounter. So these basically the major issues that I had.. Other were than all the little questions didn’t really matter. I tried were rather to follow [the interviewee’s] direction as it went . In values and their perceptions of the whole process. I had a very detailed question guideline set up after the focus but in the end I think I only asked a few groups.

that made them decide to be interviewed. some it was probably the local official’s but for . because in their encounters with friends and relatives they daily sel. but very close interaction in very. Did the women talk easily about their feelings? If this an unusual situation for them. It took some time to warm up but many revealed their feelings. .So they there a few who just did the interview as dutiful were interviewees.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 66 Qualitative practice Angelina: research in Yvonne: Angelina: I think it was something that most of them had not encountered.. even though a lot of them of academic friends were rather suspicious.. which could talk heart-to-heart with another they woman about their personal feelings or their problems. did it take a was while for some of them to warm up? Oh sure. yes. 66 . That’s why I think some of them enjoyed the interview because they felt that they were being with respect.talked about support and what kind of dom support they receive from their relatives and what kind of impact that support has on them. with And [the interview] was about something also they probably had not thought of before. intro. So I think most of them very interested in the process because were they had that sort of relationship in the past hadn’t —a not intense. but even they answered the questions and I managed to get the information that I needed. didn’t I . . duction so regarded it as a duty or whatever. and one much or of them said that I’ve helped them even two thoughintend to do so . don’t even their relatives or close friends or whatever. their There a few with whom I had a very good were rapport . They have not thought of things in this way. They felt that they had a treated chance about their underlying feelings which to talk they get the chance to share with other people. my Theythat it would not be easy to get them to said revealfeelings but I found it was not that difficult. A few of them have written to me afterwards saying how they have enjoyed the interview.

These supports are holistic and more compre. . I lso wanted some who had a difficult a relationship in-laws or husband. but after the first ten or so. say for the first ten interviews. the occupation group.I felt that there must be families that were because not well supported. . I gradually realised that I needed more interviewees of certain types. Those who not worked in state-owned factories at that time were relatively in terms of their welfare. I said. the families that were harmonious. then what 67 Angelina: . reputed for their good relationships and so on.but there obviously are many families hensive without range of support. or three generations living or two generations together. not so harmonious. For example.. and some somefrom the state-owned factories. things like that. Can you talk a little bit just about that how process worked? I remember.. the family type and the type of residence . et cetera. that I sked the local officials to help me to identify a people with different characteristics. For example. I different age groups and different family compositions. If they were have strained relationships with their relatives. . say for example a nuclear family. So together it’s age group. I more and found that those living in three-generation families had more support. So I needed those who that full were isolated. and there must be families that that not that ideal.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 67 In-depth interviewing Samplin g Yvonne: With the interviews you used theoretical samplingsaid that in some cases it took a while to and you find the right person. So gradually I with their generated more questions . So I said I better off neededfrom the street-level factories and some some who self-employed because at that time there were were and more self-employed persons—working more the streets selling fruit. . for example. the the education background. So I worked out a table — needed people from different occupations. So I told them I needed some families that had problems. I need who are in a worse financial situation. initially they gave me the better families. selling clothes. without many relatives in more Beijing.

otherwise it would go on to forever. three meetings with each of them: the I had initial in which we negotiated their participation contact in study. I had a list of topics to cover but imposed no constraints on short the women in terms of the order in which topics were covered or how 68 . So in the beginning it was the demographic but gradually it was more the considerations theoretical considerations that guided me. In this chapter. but as far as my particular piece of research is concerned. as reported during the the followup interview. Ten women participated in the study. you feel confident that at the end of the day Do you obtained a thorough mix of types of situation? Well. then what happens? Things like that. Yvonne Darlington—The experience of childhood abus sexual e Yvonne: I talked earlier (Chapter 1) about how I became inter. So after the 27 interviews I felt that I had to make a doing stopit. after interviewing the 27 women I thought that the questions at the particular time were answered—but that I had of course there were many other questions that I would follow-up. which their was happening at that time. But you can never follow-up all like to the questions and you have to set a boundary. The in-depth and follow-up interviews were both taped and transcribed. the in-depth interview and the followthe up interview. I’ll q comment aspects of the interviews: my on two interviewing women’s reports of their experience style and the of in-depth interview. Interviewing style The in-depth interviews were relatively unstructured.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 68 Qualitative practice research in Yvonne: Angelina: happens to their support and if their support from work organisations breaks down. I don’t think it can be perfect.in this research topic and my choice of ested a ualitative methodology.

so the research was constructed in such a way that sexual abuse was brought into focus. Sometimes I asked questions that led women to state what in the context of our discussion. the woman came to the point I had jotted down than herself the interview. I took the approach of letting the women talk until they had exhausted what they had to on a particular topic. As farwas possible. to me. They were often clarificatory w in nature. I child encour. neither did I want to ‘lose’ the new avenue were for discussion. In the research context. Given the focus on the women’s stories. to be immediately relevant to her experience of sexual abuse. I needed to ably in the be sensitive to when a silence was a working silence and when it 69 . I managed this by jotting down brief notes.that I clarified any remaining views points. we both knew that the expectation was that they would talk about their experience of sexual abuse. The women differed consider. There were was. Except when clarification was needed. no more a word or two. I chose a deliberately reflective style of interviewing. as they wished to tell them. obvious.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 69 In-depth interviewing much they talked about each. I resisted the tempta-cut in and inevitably the connections the woman tion to was making between the various aspects of her life story became clear. curtailing the flow of what the women had to say Not also involved respecting their silences. This approach seems to have worked to the extent that the women about other aspects of their lives and some in fact did talk volun. More than often not. were other times when the women’s comments There related to things I wanted to explore further. While I did not directly want to break their train of thought and thus risk ‘losing’ what they about to say. There were times when what a say woman was saying did not seem. times an appropriate therapeutic response would have been when an empathic nod.time they took to collect their thoughts. but unspoken . my questions flowed from or built upon as a oman’s previous comments. I asked the naive questions so participants could actually state what was implicit.that other sorts of abuse. had had a more lasting impact on their of self than the sexual sense abuse. such as verbal abuse teered and putdowns in childhood. Even so. At these times.women to talk about those aspects of their lives that aged the they considered relevant to the study rather than just the sexual abuse. Nevertheless. about leads to follow-up later. It was only towards the end of the later in inter.

nor [did I] get. that participation in the study would not prove to be detrimental to their emotional health and that. I decided to err on the of not rushing in and always checked that there was no side more said in one area before moving to to be another. the women talked about their experience of being interviewed and how they felt afterwards. Limiting contact My choice to contain the women’s involvement in the researchrather than have it ongoing and open-ended was process an attempt to minimise any potential emotional distress that the women might experience through participation in the study and to provide a clear boundary around any expectation that they talk their experience of childhood sexual abuse in this about context. I now turn to two aspects of what the women said during the follow-up interviews that reflect methodological choices that I had These are: my decision to limit my contact with the made. Others listening to the tapes or reading the transcripts might wonder at directions taken and things said or unsaid.was some validation for overall choices I had up. I did not expect. and the decision to to the only include women who were referred through support groups and counsellors. were undoubtedly many things I could have done There differently. Both these decisions reflected my concern for the women’s well-being. a detailed critique of my style. they would already be linked to a trusted support network. should distressing issues arise. somewhat at odds with prevailing feminist This was research approaches.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 70 Qualitative practice research in indicated the exhaustion of a topic area. One reasons for conducting the follow-up interviews was of the my curiosity about how the women had experienced the interviews. Certainly some feminist researchers at that time 70 . but talking about how they experienced the interviews at in follow. as as anything else they had thought of that they would like well to add . The interviews follow-up During the follow-up interview. there made. womeninitial meeting and two interviews.

. 1993. Judith welcomed the follow-up interview as a chance to achieve closure: Even this morning when they asked me where I was going I said. it was good that I was going to do the follow-up—not I felt sort of in the air (Darlington. left up 109). Powerful . It also enabled me dered to concentrate fully on my role as researcher. Stanley & Wise. got the follow-up interview with the researcher’. p. I’ll probably go and the book. Harding. really 109). and I ‘I’ve even felt then that it was very necessary that I was going to follow it up. dolittle skip up the street. 1981. Compared to a feelingvulnerable after the interview (Darlington. taking responsibility for how I related to the women as researcher but without having 71 . Referral source My decision to include only women referred through counsellors groups also related to my concern that women and support should be readily able to obtain support in relation to any issues engen-by their participation in the research. 109). 1986). Cynthia (all names have been changed) said at follow-up that it was the absence of emotional tie between herself and me that enabled her to an share her experiences. . That’s the feeling. Comments by two of the women indicated that they the valued containment offered by this approach. 1983. p. I’ve read a few books on women’s stories and and I was thinking that’s how it must feel when they’ve that written An excited feeling. 1993. p. I feel that you are doing a job. 1993.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 71 In-depth interviewing encouraged multiple interviews and ongoing engagement with participants as research collaborators as ways of minimising the power differences between researcher and researched and highlighting their shared experiences as women (Oakley. Her comments at the end of the follow-up interview indicated that it had been a useful process for her: I feel much happier. I’m participating own free will so I don’t look at you as a threat to in this of my my innermost feelings (Darlington. She contrasted her relationship with me to that her family and of friends: I don’t feel that I’m vulnerable to you because we are not emotionally linked.

And what I was saying. she would have felt able too ashamed to do an interview like this. In relation to the timing of the interview. we have seen that in-depth interviewing involves much more than what happens in the interview itself. the process itself found helpful: I can’t believe it. in this context. the women who reported some degree of Of emotional distress following the interview.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 72 Qualitative practice research in to cross over to a counselling role for which. would have found it emotionally Cynthia said it had only been since she had come detrimental. I had mandate. Nevertheless. some had had regular coun. Some. but it’s like that I’ve I’ve been allowed to say.in the study would have been difficult for them had ipation theyalready had counselling in relation to their sexual abuse. all the women reported that of its own they pleased they had done the interview. maybe it was been worth on tape (Darlington. to go on and on for nearly three hours and it’s just okay to do that. It’s like never been able to do that before . some of the women identified a time when they would either not have made them. several of the women commented that partic. 1996. saying. . even in this group of self-selected women. crucial as thisWe have considered the selection of participants. First. putting 130). even six months earlier. Comment s In this chapter. but prior to that. I’ve actually talked nearly three hours . ‘It could have really me down. initial 72 . to believe that the sexual abuse was not her fault that she had been to talk about it. p. the is. she would have coped poorly with the stress of talking about such painful issues.’ Judith would not previously have been able taken to participate in research on this topic as it had only been her recent recall of her sexual abuse that made her experience acces. not and second. Others said their distress had subsided accord. .even sible to herself.appointments or support group meetings following selling the interview that had helped. like were Judith. if they had. over twothirds reported some degree of emotional distress following the interview . .available to be interviewed about their experience selves of sexual abuse or. . Irene thought that. Two themes in the women’s comments no supported this decision.

73 . on observation. reporting back to participants reaching closure. to allow us to stand aside from events and cannot. These include their high degree of interactivity and capacity for responsiveness to the research context. impact has implications for the well-being of the participants. as it did in larly so boththe stories included in this of chapter. past we consider a form of data collection that can do what interviews is. The researcher’s the interaction and the extent to which they handling of develop aworking relationship with their participants will trusting always on the nature of the data obtained and. In the following chapter. the relationship between researcher and participant. in many cases. finally. the ending of each contact. capacity to obtain information about things that and the cannot otherwise be observed.also considered the advantages of in-depth interviews as We dataa collection method. This is particu. Underlying all of these stages is and.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 73 In-depth interviewing contact. including thoughts and emotions as well as and future events. that watch as them they occur.when the research concerns sensitive issues.

the routines and interactional patterns in of their everyday lives. 1995). The many classic studies using ethnographic methods include Liebow’s studies of African American street corner men (1967) and homeless women (1993). and chapter includes two edited interviews with researchers who used observation as part of their research. 1975. observational research can provide an understanding of what is happening in methods the encounter between a service provider and user. and Becker et al. to combining observation with other data approaches collection observation roles. The second part of the sessions. Dalton’s (1959) study of formal and informal aspects of the world of managers. Hammersley & Atkinson. or within a family. We consider some of the strengths and limitations of observation. In the human services. a large organisation or a community . 74 . observation recording observations. a committee.’s (1961) study of the professional enculturation of medical students. a ward or residential unit. 1980) and sociology (Johnson. the timing and duration of methods.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 74 4 Observatio n Observation is a very effective way of finding out what people do particular contexts. Observation has a long history in ethnographic fieldwork in anthropology (Spradley. chapter commences with a brief introduction to some This the of practicalities of observation in human services research.

limited to observable social The phenomena. Different observers undoubtedly esting notice different things.but also effort in making arrangements to clear other view activities. reflects the material through observation is filtered through the observer. frowning or crying. And while observation can assist in understanding events as unfold. because the setting is not so controlled by the researcher. the researcher’s conceptual and whatever other biases and assumptions they framework bring research will all influence what is noticed and what to the sense of it. obtained The observer has first to see something and then to identify it as inter. Unlike interviews. the observer watches what happens between the others. Fortunately. who is intricately involved in interaction. observation affords access to events as they happen. observer is. even if non-verbal indicators of what these may be are evident. There is made is 75 . The research purpose. takes place at the same time as an activity that would observation be happening anyway. sometimes assumed that observation is more It is than‘objective’ interviewing. events that have already occurred or that have not they yet happened cannot be observed. In reality. Unlike interviews and document analysis.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 75 Observatio n Choosing observation Like all data collection methods. also controls what is recorded and thus brought The observer to analysis. Internal processes of cognition and emotion cannot be observed. Unlike the interviewer. observation has its strengths and limitations. Just as the information obtained from in-depth interviews interviewing style and skill of the interviewer. The presence of the observer will. Observation also generally requires little active effort on the of those being observed. both consciously and unconsciously. however. which can be part time consuming for participants. inevitably impact on setting to varying degrees. Observation alone cannot tell us why people do the things or what the particular activity means to them—even they do astute observation of non-verbal behaviour cannot provide access to a person’s own understanding of why they are smiling. People who know they are the being watched may alter their behaviour in all sorts of ways. These are the realities of research practice. however. we are relative rarely confined to just one way of collecting data.and worth reporting. every method involves trade-offs between strengths and relative limitations. taking not only the time for the inter.

Used in the early stages of a study. for example. This can be achieved either through prior familiarity with the setting or through a period of general observation at commencement of a study. Are all observers seeing similar things and making similar sense of them? Where it is possible that co-observers are operating with similar biases.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 76 Qualitative practice research in always the risk of imposing one’s own interpretations and assump. This type of observation could a more structured phase of observation or other data precede collection methods. to look for in the observations. the use the of co-observers may provide a check on observation. or taking successive stages of analysis stage of back to them for verification. It is important to build in safeguards to minimise such misinterpretation. either in formal interviews as a further data collection. Combining interviews observation with Observation can be used at different stages of a study and for different reasons. from the perspective of the study group. in order to work out what are the important interviews things. Where practicable. see Chapter with 5. there may be no better approach than checking out with the research participants themselves what their activity means to them. it can be useful way of understanding the context of the a phenomenon under investigation and working out what the important questions to be asked are. someone whose mindset is outside that of the researcher.what is observed and so failing to understand what tions on an activity means for those involved in it. Cheryl Tilse’s study is an example of this approach. Ultimately. and that agreement reflects their shared understandings rather than what is happening in the observation setting. Later in this chapter. Combining interviews and observation is a common approach in research with children and people with learning disabilities. 76 . Understanding of the context being observed is one approach. Anne Coleman talks about how used observation as a basis for getting to know the she research and helping to work out what issues to explore in context inter. Observation can be particularly useful where research participants have limited verbal skills. it may be helpful to include a naive observer. conducting first.An equally strong argument could be made for views. This is particularly valuable where the researcher is unfamiliar with the phenomenon.

and recording.. he is deceiving people about his feelings. he asks questions with purposes of which his respondents are likely to be unaware.This has two he s personal consequences: a pervasive feeling of guilt and.. Adler & Adler. he is not really doing so and. he observes even when he does not appear to when he does be doing so. passive participation. Spradley (1980) identifies five levels of s participation: non-participation. moderate participation. In aimilar way. place. wherever possible. ensuring that the people actually being observed just official gatekeepers) are aware of the (and not observer’s Even where observation is conducted from a covert presence. and in observing when they do not know it. studied 59). the timing and duration of observation sessions. however. Gans of the says: A final source of anxiety is the deception inherent in participant observation. partly in compensation. and like the formal interviewer. Observation roles Observation roles can be viewed along a continuum from complete through observer-as-participant to participant-asobserver observer to complete participant (Gold. the participant observer is acting dishonestly.. active participation and complete participation.) In a personal ipate account experience of participant observation. (Ethical issues and for in relation to the importance of obtaining informed consent to partic-in research were discussed in Chapter 2. thus.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 77 Observatio n The process observation We now consider some of the practicalities of observation. p. 1987). quite problematic. i spying on them. He pretends to participate emotionally not. In this chapter we are assuming fully negotiated observer roles thatnot involve deception. a tendency to overidentify with the people being (Gans. By do this mean being absolutely clear about one’s role as a researcher we and. covert In short.. 1982. deceives the people he studies. have accepted that most levels other than that ethnographers of complete observer or non-participant will involve a degree of decep-Using deception is. including observation roles. both tion. 1958. 77 . psychologically.even though [the fieldworker] seems to give of himself when he participates. ethicallyits potential impact on the researcher. whatever the level of participation. Traditionally.

firmly established prior to commencing the and observation. the goal of ‘just blending in’ is more likely to be achieved through a level of everyday participation. and to gauge whether boundaries need to be adjusted. If the researcher is clear about their purpose are and and is consistent in this. participation is unlikely to be appropriate when observing highly specialised activity. to be the primary issue. It helps be) the researcher monitor how things are going. As those being unknown observedmore familiar with the researcher’s presence. in turn. it will be easier for participants role. In this situation we would consider the degree of role clarity established. Non-participation to the extent of avoiding basic human interactions. In many general settings. being clear about where one is (or wants to at any point in time is invaluable in two ways. 78 . we assume that the observer’s been negotiated and that those being observed presence has are aware they are being watched. to accept the observer in that role and let them get about the business of observing. On the other hand. in reality some flexibility may develop in the role as the research progresses. The role of uninvolved observer may be more readily sustained early on in the research when the researcher is relatively in the observation setting.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 78 Qualitative practice research in such as behind a one-way mirror. Even where the observer to participate in such an activity. they are unlikely to is qualified be to do justice to both roles at able once. It can also be helpful for those whobeing observed. as The their presence will always have some impact on the setting. In a paradoxical way. including to whether one is primarily participant or observer clarity as at given time. participants who understand whyobserver is there and what they are doing may be less the botheredpresence and the observer. such as in an operating theatre or in a child protection or psychiatric case conference. even demands. there become may be invitations. would in likelihood draw more attention to the observer’s presence all and potentially heighten their impact on the setting. While in practice there will often be a continuum of involvement along these dimensions. such as responding to greetings. While the boundaries of the researcher’s role should be atednegoti. to participate. may be less likely by their to have a negative reactive impact on the setting. rather than the level any of participation per se. The level of participation that is possible or appropriate will vary from one setting to another.observer is always in some respects a participant.

single occurrence of an activity. or so much has been observed that long the observer forgets or lacks the energy to record or reflect on what has happened . This can be useful for the be minute 79 . On the other hand. for example. or during inclement weather?).Realistically.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 79 Observatio n When to observe and for how long? No social setting is static. it is important observation plan be broad enough to include any that the significant in activity that may potentially alter the variations conclusionsfrom drawn the research. There will always be a range of activities whether during a day. It makes sense to expend observation time at the times when what is being valuable observed likely to happen. 1993. or in parents’ availability to pick children up). duration of any period of observation needs to be The carefully considered. observed and to take a cross-section of occurrences. or throughout the year (Do parents pick their up children more commonly at the start of a new term. observations of is most parents up and dropping off children to and from school picking would occur at those two times of the day.quantitative analysis.. highly structured recording For mayframes (Trickett. 1997) that enable data be used to reduced as they are recorded. Whatever the research question. The best guide to deciding when to observe will be the research It is important to be clear about what is being purpose. there would be little point in leaving before a session ended. Recording observations As with so many other aspects of research practice. within each activity there are likely to be peaks and slow downs of occur. Patterns may vary need to on different days of the week. however (there may be daily changes in patterns of children’s activities. A sampling process thus needs to occur. observation requires considerable concentration and sessions should not be so that alertness fades. what and how to record depends on the purpose of the observation much and the data are to be how analysed. and over time. Observation sessions certainly have to be long enough to observe the social processes that are the subject of the study— in study of changes in client–worker interactions at various a stages of counselling sessions. one cannot hope to observe every rence. a week or a month. For example. Singh et al.

80 .between a mother and her infant. Their suggestion is to jot brief notes during the observation and to write these up down as field notes after leaving the field.of visitors to the nursing homes. Lofland and Lofland (1995) stress the necessity of recording as soon as possible after observing and suggest a practical process for dealing with the often impractical task of writing copious notes in the field. Cheryl was able to observe long for longer periods of time than Anne. while Anne Coleman actions was interested in a much broader sweep of activity in Fortitude Valley. ‘overloaded’ with things she wanted to became record. In each but intensity of observation and recording had implications for the how each session lasted. These raw field should be identified separately from the researcher’s notes own reflections and conceptual material. Stories field from the In each of the following inside stories. In the stories reported below. whether through the risk of missing while somethingis important or through concern for how those else that being observed will respond to one’s writing. Recording for nication qualitative less structured but decisions still need to be made analysis is aboutto focus the observation—this could be at a very specific how or more general level.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 80 Qualitative practice research in analysis of interactional processes such as the non-verbal commu. In general. We asked both Anne Coleman and Cheryl Tilse to begin by talking about why they chose observation as a major of data collection. These two very different examples how provide some useful insights into the versatility of observation as a method collection and some of the issues to consider when of data thinking about using observation. case their purpose flowed from the research question. paraphrasing and more general recall. after a couple of hours. These notes should be as faithful a recollection of what happened as possible. who. which themselves may range brief impressions to more formal analytic from notes. although the interview excerpts included here focus specifically on the use of observation. and clearly distinguish between exact quotations. then to move on to the practicalities method of they went about it. but always no later than full the following morning. observation was used in conjunction with several other methods of data collection. Cheryl Tilse focused her recording specifically on the movements and inter.

. Choosing observation Yvonne: Anne: Why did you choose observation as an approach to collection? data Simply because I’d known people in this group long enough to know that in fact they could be quite devilish . . what am I going to believe? dilemma It’s a very divided community so if you talk to also one person they will tell you this is a fact and there’s no question about that and if you talk to somebody else tell you something else is a fact. a search of documents that relatedlocal urban renewal process and. just for the fun of it. then ininterviews depth interviews with homeless people. They could. Anne has published on ethical issues encountered during observation (McAuliffe & Coleman. tell you biggest story and then tell you something else the the day and then you were caught in that next terrible about. 1997). . . The phases were observation. . followed by the informal with a range of people in public spaces. each phase building on one before. . five-phase study. the observation first up was that because of [my] famil.I wanted to go back and just have a look whole place and the range of things at the that happened in those spaces before I actually started to focus myself in again .[with the area] I had a fair degree of iarity knowledge that some of that knowledge would but I knew be outdated .other really important reason for doing The . She is talking here about her use of observation as a data collection method. And you they’ll find that none of those things are actually out facts.. finally.. 1999) and social policy implications of the study (Coleman. What were the benefits of that first stage of observation? 81 Yvonne: .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 81 Observatio n Anne Coleman—Five motels star Anne’s was a multi-method. well..So I knew that observation was going to be a really useful check what was said to me against what way to I’d actually seen myself . a second phase to the of observation.

as soon saw me. What writing down there?’ A lot of them are you were homeless people I knew but a great number of them just local people who felt that this was were their too and they wanted to know who I was. . t So it 82 . that people I anted to hear . So in between encounters I’d w be sitting down taking some notes and if I was just sitting looking for some interesting people to have around aalk to I’d be taking a note of what I was seeing. were some spaces in the Valley that that there alreadyshared public spaces.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 82 Qualitative practice Anne: research in Well. It also gave me a chance to see clearly what I suspected. It gave all sorts of people ahance to get used to me and that was really c impor..I was aware that in a even I’d talked to people about the fact that I though was coming back as a researcher. the first one was that I got re-orientated and that because... and what I was writing in this about. One of the things that happened all the tant . in my field journal you can was see there was a level of tension and expectation that about me going back in there. space writing book. Informal spaces Anne: interviews in public Phase two was like the active engagement—I was in those public spaces I’d identified in the first phase and purpose was to talk to anybody that used my those ‘What do you think about this place? What spaces. . do like about the Valley? Why do you come here? you Is interesting? Does it make you feel scared?’ it Anything wanted to tell me about the Valley. but that those spaces had changed in the four years that I hadn’t been working in the Valley. So it gave me a chance to Anne reorientate myself and other people to that new role. actually.. time that constantly people would come up to me was and ‘What are you doing? What’s in the book? say. for most people. I was Anne the social worker as they not the researcher. and that was about my beingvery different role . .. where in fact were homeless people and mainstream community people did have a reasonable level of interaction.

Knowing stop Yvonne: Anne: when to When did you start to feel that you had enough data? before I got to the end of the in-depth Even interviews. were the benefits of that second Yvonne: And what stage? Anne: I became much more confident. my attention was the space itself. . Anne went back for a second stage of observation towards the end study. It was also just another look at things I started in-depth interviews with before homeless the more I saw before I went into people and those interviews obviously the better the interviews were going to be. but there’d been some very interesting and quite significant things that had happened in the Valley while I was doing the in-depth interviews and I wanted to go back and capture what these were about.I was starting to hear the same sorts of things from. Anne: After I did the in-depth interviews I thought I was finished until I went back and did the feedback. even though she had not initially planned to do of the so. In the first what was motivating me and the focus of phase. . the focus was people . you know. . police [who] were saying about people who have been here in things the community—they identified ‘homeless people’ as being local community as opposed to ‘itinerants’ whooutside people—but I started to hear the same are sorts 83 . In the second phase.. how did that phase differ in terms of its Yvonne: So purpose the from first phase? Anne: The major difference was its focus . Probably because all that initial stuff about being the observer of had largely been resolved so I wasn’t getting interrupted. . .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 83 Observatio n wasn’t strictly observational though observation happened .. come up and ask me what I was People didn’t doing any more . . real commonality that was starting there was a to surface .

phase to come and if I’d missed anything back major would say at that stage. But some much days. I’d then leave was in and was going straight home. you vary the time of day that you Did observed? 84 . ‘Hey wait a minute’. I’d sit down at if I the computer and start to write up a set of notes based on ones I’d taken in the field. So even though I was getting it from a different point of view. methodological made as I went along. So for how would that take you to write up your long notes? Anywhere from—if it had been a quiet day. if there was more complicated stuff going if anything I’d observed had had a big on. Sometimes. the same sort of stuff being replicated across different groups was and across different homeless people. But part of decisions that I the reason in the end that I kept observation periods to hours was because I couldn’t keep up with two writing them up. there hadn’t hadn’t been happening—maybe two hours. So I quite satisfied and somewhat felt relieved. people so there was nothing lost that couldn’t be got back. Also. two hours observing. I knew that there was still a feed. or impact. Timing. be another couple of hours of journal there’d time because I also kept a separate journal to record my feelings and also to process. up the field notes would take four writing hours.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 84 Qualitative practice research in of things from homeless people about how they perceived themselves. recording Anne: duration and Yvonne: Anne: Yvonne: I couldn’t do any more than about two hours at a because I couldn’t absorb it and I couldn’t time hold my memory if I went much over that. I guess. generally speaking. So I’d it in jot down what were basically memory prompts while I the field for that two hours. and then write them up as soon initial as I got home. If there was going the to a delay then I’d go somewhere private and fill be the notes out. there been a lot of people round. So it was all starting to converge.

they’d go. to let people know she wanted to conduct some indepth interviews.. 85 . generally. to let anyone who was interested know that she was still around and involved in the research. Keeping homeless people informed about the study Anne used an innovative approach to keeping in contact with this population. thinkwas a solid enough picture to work that from. Five minutes later—totally is different people. So. when those across changes what groups came in and out..’ So even people who couldn’t read knew that this was a bona fide communication about this particular piece of research.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 85 Observatio n Anne: Another useful thing I picked up in the first observational phase was that public spaces in Fortitude Valley and they can literally change in the change movement If you were looking in the opposite of a hand. I in the end the night stuff was underthink represented. to advertise her feedback sessions and. Here. groups of Sowas clear from pretty early on that I had to I consider the 24-hour clock and that I had to be aware of what on all through that clock. So that’s what I went did. The graphic was just one those standard ones you get in computer of packages. So at one minute apace can be where this particular group is and s this happening.I kept the words with people to a minimum and I put a graphic on it that after a while time somebody saw something with that every graphic on. Anne: I put a flier out saying that I wanted to do interviews and why . So it was definitely were weighted on the day-time side but I still spent enough time at night observing to have a clear idea about what went on and I could identify when the transition times the 24 hours were. totally different things happening. she talks about how she let people know that she wanted to do some interviews.a total of somewhere between 120 and There was 150 and probably only about a third of those hours hours night-time hours. I happened. ‘Oh this is a thing about Anne’s research. direction you’d completely miss it.

This is your local area. would like what’s happening. also conducted brief.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 86 Qualitative practice research in but it was a suitcase being opened up and out of the suitcase were springing all these high-rise buildings. that went nication on top of it.is your home. the graphic became a sort springing of signal all the way through. 1997a. 1994). So. Here Cheryl talks about her use of observation as a data tioncollecmethod. of captured my sense of what So. semi-structured interviews with She staff how they viewed visitors. She has published two papers on her use of participant in this study (Tilse. much She used the six units in which they had placed their partners then as a for observation of how visitors were treated and provided focus for. Choosing observation Yvonne: Cheryl: How did you come to choose observation as one of your data collection methods? It was partly a commitment to try to understand the complexity of the experience and my view that you 86 .I wanted to make with people. you open it up and now look this what’s up out of it. And people who weren’t literate the then say to people. it kind was happening for these people. what’s going on?’. 1997b) and has also observation reportedthemes from her in-depth interviews with spouses on the (Tilse. Cheryl goodbye Tilse—The long Cheryl’s study of the experiences of older people who had placed a artner in a nursing home used several methods. If there was any commu. She p first conducted in-depth interviews with nine men and nine women recently placed a partner in a nursing home or a who had dementia began with in-depth interviews as her concern was hostel. She very with trying to understand the perspectives of the spouses. and did a content analysis of about any documents that the nursing home had produced for or about families. so people who couldn’t read were able to be involved as well. ‘Here’s one of Anne’s fliers.

not a I was always partic-in the sense that I didn’t have a relative in ipant. seen And felt it provided lots of opportunity for people I also to 87 Cheryl: .I to watch interactions between staff also wanted and families. a theoretical interest in the use of space I also . So Ieally wasn’t part of the place or pretending to be r part place. I pick thought there was the risk of [obtaining] socially desirable responses.. In do residential at that time there was a big interest care policy in families.. If I just interviewed staff. where observer werealong that line? And did that change at all you during course the of the observations? clear that I was an observer. . And part of the outcome standards was about open to visitors and welcoming visitors being and home-like environments.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 87 Observatio n couldn’t understand how family visitors were provided treated without actually being in the for and settinghad. I anted to understand whether visiting was w primarily and private or whether it was social individual and collective. to include and exclude people in health settings. . . I wanted to n be as a researcher. as ethically I felt I had to be. unit I’m not a resident of the unit so I really am and an observer and what I’ve come here to do is observe one feature of life—the treatment of family visitors’. So there was a whole lot of rhetoric about families and I guess that was the other I wanted to observe because I didn’t want reason to up the rhetoric. So I guess it and how was on an understanding that what people say based theyis often different to what happens . the I was saying. Observation role Yvonne: If you can imagine a continuum between the complete and the complete participant. I set myself up as a researcher. carried of the a otebook and made notes very overtly. ‘I’m not a staff member of the unit. You could only understand [that] by watching what visitors did and how they interactedstaff interacted with them. and between families and other families.

So participant at that level. and I’d tell them and say. But people came in and were talking to staff members issue and then they were gone and about an I’m that they weren’t aware that I was there. So you help would find yourself engaged in that sort of thing.. I wasn’t doing anything else.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 88 Qualitative practice research in Yvonne: Cheryl: say. So I kept myself at quite a distance from interactions. So you did get engaged in that way but always very clear to me and I tried to make it was it very clear to staff that I was just observing visitors. I asked staff to tell any visitors that I Probably wasthe unit and what I was there for and I left on material.and researchwise... what’s it like to be a visitor?’ So I had a way ‘Well. level I was very clearly an observer— So at that ethi.’ And I would engage through and being part of their visiting and talking meeting their family. I guess occasionally I felt I was most a articipant in that I was there and somebody p with dementia would come up and start talking to me and I would have to respond. I would then try and the resident join in the activity.if I was pretending to be a staff able with member. are you doing? This is really interesting. I was keen to observe from a distance as I wasn’t actually interested in what people said to each other—more [in] how the was used and how people got included space or excluded. say. ‘What Come and have a talk to us. Did people always know from they were being watched? not. So it’s always was a I more in practice than it is in messy theory. With visitors Some of the visitors would come over and as well. ‘What are you doing?’. that you were very overt about You said your observing—you had your notebook and you were in a on view—but also that you tended to sit sense away direct interactions. It sure was 88 . especially if the staff were running a particular activity. in talk to people that I wouldn’t have been to comfort. But I also had to say I was a participant in that I’m visiting and experiencing all this—I can’t find a place to sit and I’ve been of hereday and the tea trolley just passed me all by.For most of it I sat cally and observed and listened and watched..

late afternoon. The time I slept over in seven the nursing home there were no visitors. think that really interested in observing visitors and the I was unit relation in to visitors. I did come to slip the furniture or the shrubbery at into times. . ‘Oh you’re still here?’ . this is visitor one. Ithat was important in terms of their trust. I had a whole lot of things that I to lose was looking at so I’d draw the setting and then when came I’d often draw. so I think they did lose track of the fact I was there. and it was quite a feminine environment w in terms of residents.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 89 Observatio n interesting because I spent the whole—as much ascould—almost the whole shift there. Timing of observations Yvonne: Cheryl: the Yvonne: Cheryl: I’m interested in the timing and duration of your obser..So I hours. start material. what would you say would be In hindsight an optimum period of observation? what you’re observing and the depth It depends on of you’re trying to understand so it’s really hard what to I think after more than three hours you must say. shifts? I did was try to sample the shifts when What visitors were most likely to come. I was there I for a long time and I did become part of the furniture and I noticed staff would suddenly say to me. particularly in some of the that units they were busy with bathing and where showering just sitting in the dayroom and there and I was were visitors in the dayroom..You’ve said you observed on all three vations. visitors and staff. didn’t six in the morning when they were go at showering people when they told me that no and feeding visitorsI came in the afternoons. sort of mid. and visitors I’d 89 . So it wasn’t the whole shift but it was what they told me was the likely time there’d be visitors on the unit most because it very clear to them I wasn’t observing I made care..to came.. So I’d stay three or four I think they were six-hour shifts. and they use d to say no visitors come after and that was true. . and the fact that I’m other a oman.

. They ended up there and then long they stayed and who talked to how them. . like did they speak to the simple registered a whole list of things I was interested nurse. interpretation 90 Cheryl: . They sat there on the verandah for two hours and it wasn’t something that you would miss. must have been so much going on. Yes. . But How most time it wasn’t high intensity. I had different diagrams for different Often visitors it got very complicated but it was that sort because of level of observation that kept me engaged. was the use of space. Were There there when you saw things that were interesting times but to say to yourself.. ‘Well. who did they talk to . You used diagrams and had some broad categories—how struc.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 90 Qualitative practice research in Yvonne: Cheryl: watch where they went. could see. I used those little shorthand note. I wasn’tto look at how residents were being treated.was that and were there other ways tured you recorded? I wanted something that was obvious note-taking but too obtrusive and clearly not tape recorders not or anything like that. One was entry and One exits. ut it was the fact that I was only trying to b observe things. did they enter. I guess. Recording observations Yvonne: the I’d like to ask you now about recording. that’s really had interesting what I’m here to look at’? Were you but it’s not ablefocus yourself in that to way?It wasn’t that intense. I had in. in most places. . So occasionally a whole lot of visitors came at once . It was over a of the fair spread of time. The work for the workers was very but in fact for the visitors it was quite a busy slowerPeople came and stayed a few hours so I pace. trying It was quite focused on one particular thing.because on one side I’d put clear books descriptions side of the page I kept for analytical and the other or interpretative notes or questions I had to follow-up — appears to be happening. . I should check ‘This this So I kept my analysis and ongoing out’.

as both a filter of what is recorded and a part of the research context. 91 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 91 Observatio n of questions separate from pure description. of course. it also possible to take steps to minimise bias and inaccuracy is in observation. that observation cannot tell us is about. can tell us Obser. the the shiftthen I would write down what I saw. the common practice in qualitative research of hence combining observation and interviews. The examples also highlight the central roles of the observer. because I had this thing about to draw diagrams of who was where space. the time I arrived.groups ular of research participants. so I remember it when I was analysing what could was happening—to remember how long people spent outthe verandah without a staff member speaking on to them. and the observer in turn has an impact on the observational environment. Comment s An important message from both these examples is that observation things that other methods of data collection can’t. Assumptions and biases need to be stated. the unit I was on. while what is observed will always be filtered through the observer’s mindset. So I’d just record at the n startdate. as can taking can assist seri. In the following chapter. I had a otebook for each unit. Being conceptually clear about what is being observed rigour and consistency in observation. vation not filtered through someone else’s perception of what is happening. Only what the observer notices and decides is relevant is recorded. It those perceptions.enables us to see events and interactions as they unfold. we consider some ways in which data collection approaches may be modified to suit the needs of partic. But it and was always important.the physical limits of one’s capacity to observe and ously later record.

* We not saying that are children and all people with an intellectual disability requireall a ifferent approach to data collection. cable we consider how data collection processes may need to be modified to the needs of particular groups. dementia or acquired a brain Knowledge of particular conditions and their injury. we mainly use the terms ‘intellectual disability’ and ‘people with an intellectual disability’ which are the preferred current Australian nomenclature. 92 . We acknowledge that other terms. are used in other contexts. Rather. using children and people suit with an intellectual disability as examples. These groups are far d too diverse to suggest that one approach would fit all for a start.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 92 5 Tailoring data collection to suit the needs of research participants The data collection processes we have talked about so far are appli. our position is that some children with (the so the older they are) and some people with an less intellectual(the less so the milder the difficulty) will require disability particular be made by the researcher to promote their full and efforts to active participation. Some of the material in this chapter will also be relevant to research with people suffering cognitive impairment as result of conditions such as stroke. ‘developmental disability’ or ‘learning difficulty’. * In this book. such as ‘learning disability’.to qualitative research with any group. cognitive the individual is obviously important if the researcher effects on is make the adaptations necessary to achieve the best to possible communication with the research participant. and in many cases research with these groups will be no different from that any other group. In this chapter.

the absence of voices of children and people with an intel. only a first step. and held concerns that participation could be detrimental. In many cases. considering them too hard to get information from or unable to express a point of view. Both children and people with an intellectual disability have. and that their perspectives are a valuable source of input into decisions regarding themselves as individuals and the development of services more generally. played an important role in highlighting. in recent times.They have been overlooked as potential research ture. they may just not have known how to work directly with children or with people with an intellectual Researchers’ lack of skill in working with these disability. but on own behalf—parents. and attempting to redress. Ethical issues. particularly in relation to obtaining informed consent.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 93 Tailoring collection data We focus here particularly on practical aspects of engagement with research participants and data collection. remainder of this chapter focuses separately on the The practicalities of working with children and with people with an intellectual disability in the research context. Researchers may have feared intruding to speak or doing the wrong thing. 93 . until relatively recently. Another hurdle to the inclusion This is of children and people with an intellectual disability in research the skills of the researchers working with these concerns groups. Researchers have too often avoided including these groups as research participants. and instead asking others on their behalf. either through the researcher acquiring the can be necessary skills or through working in a team with someone who already these has skills. have been considered in Chapter 2. It is not that children be and people with an intellectual disability have been underresearched that they have had limited opportunities to speak per se. carers and case records have their all been prefered as sources of information about them. or to unreliable sources of information. teachers. considered either too vulnerable to be troubled by researchers. participants. remained ‘voiceless’ in the research litera. groups addressed. An essential step is accepting lectual that children and people with an intellectual disability do have important things to say on their own behalf.disability in research. Children’s rights and disability rights movements have.

1985. In this section. we in no way view children. in the best interests of children to (Wilkinson. 1993. 1989). Carey. Just how and to what extent this in should is a subject of continuing debate. 121). for example. and is areas also relevant to the research context: States Parties shall assure to the child who is capable of forming his or her own views the right to express those views freely in all matters the child. and thus argue for expanded rights for children. even of similar ages. p. Donaldson. emphasises the situational context of children’s cognitive and language capacity. dependent child requiring special protection and also the child who is a potential is adult-like.to adults. 94 . 1991)’ (Wilkinson. we draw together some of the considerable wisdom practice that has developed in research with children. 148). Brannen & O’Brien. Fixed age-related stages of children’s isations cognitive development and their use and understanding of language are being increasingly challenged (Donaldson. 1995). 1989) steers a course middle in that it ‘embraces both the vulnerable. and thus the rightful recipient of a range adult. both the behaviour of others and what is expected of them to make sense to need them. or civil. the views of the child being given due affecting weight in with the age and maturity of the accordance child.can be made. as homogenous groups about whom rigid those general. suggesting that children are generally more likely to display competence when they ‘dealing with “real-life” meaningful situations in which they are have purposes and intentions and in which they can recognise and respond to similar purposes and intentions in others’ (1978.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 94 Qualitative practice research in Researching children with There is increasing recognition of children’s rights to be involved decisions affecting them. In this discussion. Keil. to protectionist stance that takes account of children’s a differences as well as what they have in common. political.1 of the Convention is germane to practice in such as family law. Thus. 1993. and seeks from adults preserve the differences. United Nations Convention on the Rights of the The Child (United Nations Centre for Human Rights. 1978. Article 12. Perspectives on happen children’s from liberationist views which focus on children’s rights range larities simi. child protection and education. in particular on interviewing and conducting participant observation with children. p. legal and social rights similar to those attaching to of the status of adult (Carney.

and to be able to communicate this to them. Very often. In all cases. and these are decisions that need to be made in and the context of the research purpose and the setting in which it is being conducted .involve.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 95 Tailoring collection data The experiences of researchers who have worked with children are presented here as a guide to some approaches that may be useful. and any techniques that will be It is also vital that they understand the importance of used. We then consider family some practical issues relating to the interview context itself. we consider first the issue of deciding who to inter. to be really interested in what they think and feel. or whether to interview children alone others. In general. Interviewing children In this section.and trust. is crucial to eration the child’s participation in the research. not influencing what the children say. as well as formal consent. and the necessity to like children. a couple of overriding principles child. where parents and children are being interviewed. These are the importance of neither over-estimating nor under-estimating a child’s capabilities. in working with children. the ‘best’ thing to do has to be worked out in the context individual relationships between the researcher and of the each There are. There is no right or wrong approach. to be comfortable with them. Who to interview and in what order? In research with children there will always be adults whose coop. 95 . In the final analysis. interviewing parents first gives them an opportunity to get to know the interviewer and also to develop a clear idea of what the inter. however.and in what order. All have or with pros cons. Researchers a number of approaches to deciding the order in have taken which interviews are conducted. not as recipes for how to work with children. parents should be informed as to views the nature of the children’s interviews. This is important in studies where view other members are also being interviewed. both prior to the initial interview and before any subsequent interviews. in any context. others who are significant in the child’s life will also be interviewed.

children struggling to be heard among the adult voices. interviewers to their own research agenda rather than taking keeping firmly the to follow the child’s view. This can 96 . All the children’s interviews were conducted by a social worker experi. and children presenting a children united with parents. but may other difficulties. 1999) found the presence of adults at times altered the behaviour of all concerned—the interviewer. the context of an interview with a child will shape the interaction.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 96 Qualitative practice research in Interviewing the parents first may also alert the interviewer any to indicators that children may later be exposed to pressure by a arent to disclose what they said in an interview. Analysis of transcripts revealed interviewers deferring to and siding with parents’ views. and to make the interview context as child-focused as possible. even though the child’s perspective was between the the major focus of the research. Hood. It is important to take the time to develop a friendly and informal relationship with the child. see pose also Mayall. The interview As with any interview situation.in working with children and others on the team enced alternated interviewing mothers and fathers. deferring to adults’ views. the child and the parent.so in research on sensitive topics such as separation cially or divorce. in relation to any questioning about front. Very often the interviews ended up largely as exchanges adults. Beek and Connolly (2001) used separate interviewers for each member of the family. intergenerational issues within the family. Children need an emotionally environment if they are to feel comfortable supportive enough to participate in an interview. at times a decision may have to be made as to whether or to proceed with a child’s not interview. In a study on contact after divorce in which both parents and at least one child were interviewed. This may be p espe. parents interrupting and time speaking for children. While this between approach prevent a parent intent on asking children what they would not had from doing so. separate from what others might have said. Kelley and Mayall (1996. Trinder. it at least established a sense of the said importance of each having their own say. Interviewing parents and children together may help deal with parents’ concerns about what is being asked of children.

They of drawings had four. p. children within and outside the protective care system. in through pairs small groups. the interviews in this way enabled the children to Beginning talk something they were already familiar with. asking them to ‘Draw all the things you yourself do that you healthy and keep you healthy’. 1986. a study of children’s perceptions of harm and risk. 231). They also suggest right letting direct the interview process as far as possible. without influencing their on how drawing. At times children this involve the adult letting go of their adult control and will authority and ‘playing dumb’ in order to let the child set the adult right and try to get the adult to understand what they mean: ‘Once really the ethnographer is established as a “dummy” in need of guidance (a lasting “one-down” status). provided a focus for the interview and content. choosing the the venue and activities. set them at about ease. Tammivaara and Enright (1986) suggest using classroom settings such as ‘show and tell’.to twelve-year-old children a drawing pad prior to sent the interview.giving them the choice of being interviewed singly. and allowing the child as much control familiar overprocess as possible. the child informant will often provide explanations and information voluntarily and at almost any time in out of the formal interview encounter’ (Tammivaara & or Enright. Parents were given make advice to encourage the child. In a study of children’s ideas about health. Tammivaara and Enright suggest getting away from a straight question-and-answer format as children will perceive this as an examination-like context and provide what they perceive to be the kind of answers for this context.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 97 Tailoring collection data be achieved through using a venue. Another way to maximise children’s comfort in the research setting is to combine data collection with another activity that is already familiar to the child. and putting them in control of the tape or in recorder and the duration of the interview. Backett and Alexander (1991) commenced their child interviews with discussion the children had done prior to the interview. and stating when they want to stop. or to use familiar game discussion routines 97 . asking teachers to include questions as part of everyday activities (such as classroom on a particular topic). They aimed to allow children as much control as possible in the interview situation. such as choosing to participate. their In own anxieties and their experiences of adult support and professional Butler and Williamson (1994) interviewed 190 intervention. context and reporting media to the child.

for example whether that with Siegert suggests that a combination of participant peers. Goodman and Saywitz suggest that in order to maximise children’s communication : The interviewer should listen to the child’s narrative report. but it is questionable can be generalised to other contexts. assume that the younger the child. what choices the children had. in contexts defined by adults.tell us a lot about children’s competences in adult– mately child interaction. 98 . it is important to take leadthe from the child as to what they find interesting enough to talk (Beresford.. Beresford also suggests taking the time to to be establish and phrases used by the child at the beginning of the words the interview and using the child’s language. Similarly. not onlythe content. context of child abuse investigations. the shorter Interviewers can the sentences the child will comprehend. such as the timing of the inter. flexible and struc. observation discussions among children would be more and group appropriate gather data about peer ways to interactions. 32– 3). Steward. p. In the Bussey. views. the fewer verb-noun units per utterance. 1993.Whatever choices are made about how to interview children.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 98 Qualitative practice research in within interviews (such as asking children to make up a story or draw a scene. (1986) raises some interesting issues in relation to the Siegert process of interviewing children and the nature and quality of data obtained. This includes anything ently in about the research was conducted that could have affected the way the children’s responses or performance. Backett and Alexander suggest.reports on the research. or to play ‘let’s pretend’). The interviewer should then match his/her own sentence length and complexity to the child’s in order to maximize the communication. This avoids making assumptions about aboutthe researcher thinks might be most interesting and relevant what to them or holding too firmly on to one’s own topic list of what seems important.the venue. pp. he argues that an interview between an adult and a child can legiti. tured 35). or who else was present. In discussing research on children’s social competence. contextual issues should be discussed thoroughly and transpar. 1997). ‘Keep the interviews open. the fewer syllables per word and the more they will depend on familiar contextual cues to glean meaning (Steward et al. In research on children’s perspectives. the activities. but also for the child’s spontaneous use of for language.around the daily experiences of the child’ (1991.

assisted and language they used to discuss these and related topics. and Enright say: ‘Young children generally find Tammivaara doing something with something and talking about that something to be easier. Participant children observation with Fine and Sandstrom (1988) provide an excellent introduction to participant observation with children of various ages.e. and spending some time ‘warming up’ getting to know the and children.. we draw on their work in the following discussion. and with six or more it became going with difficult the attention of the group and to draw out quieter to hold children. The focus knowledge groups them to understand children’s level of knowledge. however. 1995. 1997). They found it useful to move from less sensitive hour to more sensitive topics and to avoid abstract questions. Hoppe et al. and more interesting than only talking something that isn’t physically present (i. be reserved for individual interviews the or pencil and paper exercises. instead asking questions that led children to talk about their own concrete experience. and can and less be a way of encouraging more reticent children to offer useful informa. They also suggest making the setting as informal and unschool-like as possible. 99 . a an idea)’ (1986. as it was difficult to get lively discussion smaller groups. Topics with A the potential to cause discomfort or embarrassment for any child in group should. Beresford. 1992.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 99 Tailoring collection data Making use of pictures. 232). routine. books and other props also gives children something to do. and sessions of rators one or less. use of They also mode. more comfortable. taking the pressure off the verbal interaction. The group context may well be more familiar intimidating than the individual interview. et al. the They recommend groups of five. Group discussions groups and focus Group discussions or focus groups can be used as a variation on individual interviews. an about event. p. (1995) used focus groups with primary Hoppe age school children to develop measures for a study about children’s and understanding of HIV and AIDS.experienced in working with children.might not provide in an individual interview (Mates tion they &llison. recommend homogeneity of age and gender.

While clearly not ‘one of them’. any adult will inevitably be seen to some extent as an authority figure. when ‘the child begins to belong to a group that is meaningful to him or her. they initial graduallyto accept his presence and involvement in their seemed activities. He was present and available in the children’s the activity waited for them to make the first moves. adult It important that the adult participant–observer has some reason is for there that is understood and accepted by the children. in the role of an adult friend: The final major type of participant observation role. Corsaro (1985) chose a ‘reactive’ approach to engaging with children. and issues of children’s reactivity to an presence need to be taken into account (Fine & Sandstrom. especially where the researcher holds no formal authority in relation to the children. Fine and Sandstrom recommend adult the researcher take the middle ground. While consent group relations for 100 .We believe there is a methodological value in maintaining the differences between sociologists and children —a feature of interaction that permits the researcher to behave in certain ‘nonkid’ ways—such as asking ‘ignorant’ questions (Fine & Sandstrom. as being airst step to developing a relationship with them. is to become a friend to one’s subjects and interact with them in the most trusted way possible—without having any explicit authority role. While some suggest that adult observer can divest themselves of their ‘adultness’ the so interact with children as an equal (Goode.Even so. responding to his occasional attempts their behaviour with ‘You’re not a teacher’ or ‘You to control can’tus what to do’ (1985. tell 31). 17). 1988). in our view. can be studied’ (1988.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 100 Qualitative practice research in The role the adult observer takes vis-à-vis the child will be a powerful shaper of the research.. also essential. the children also did not regard as hima formal authority figure. 1988). or in a and least. p. and Sandstrom suggest that observation is Fine possible with children from age three. this will always be an ideal type because of the demographic and power differences involved. and of what children will allow the observer to see and/or to participate in. 1988. and. p. 1986). In a participant observation study of pre-schoolers’ peer behaviour. p.. 36). After areas. As indicated above.way (Mandell. but some tentative advances from a couple of children. and the one emphasised in this book. as a consequence. The f researcher’s where they fit in needs to make sense to the presence and Establishing rapport with adult authorities or caregivers is children.

what to clarify the meaning of what has been observed. Mayall (1999) suggests ongoing consultation with children as the research progresses and using their input on interim findings in development of later stages of the research. in ways in which the preschoolers would never conceive.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 101 Tailoring collection data children to participate in research must be obtained from parents. and the researcher who wishes to gain rapport with informants must recognize this (Fine & Sandstrom. by its very nature. if necessary. there is increasing recognition of team are the 101 . as activity. is structured ipate in. 46). what the issues to focus on.participant observation. 1988. to control the researcher’s access to their and. p.as possible about the research and the role of the mation researcher: children should be told that there will be an adult ‘Perhaps the who watch and play with them to learn what they like and will what do. what to look for. even where adults have already given consent to the researcher’s presence. p. and so on resting with the researchers. by researcher—with decisions about where to observe. While older children may find ways to control what the researcher doesn’t observe.Older children are in a much better position to vote with their feet—both in terms of whether they agree to be part of the research it proceeds. Provided the researcher does not also hold a formal authority role in relation to them. older children can decide whether trust this person and have a measure of control over or not to what aspects of their worlds they will allow the researcher to participate in and/or to observe. The preadolescents had the authority to decide when I could be present. This at least the gives the children being researched some opportunity to comment on the researcher has made of their activity and. they recommend that even young children be given as much infor. Involving children in the research process While examples of children forming an active part of the research relatively rare. means that preadolescents will have a fair degree This power of authority to shape the role of the researcher. 50). certainly what they do and do not does and partic. This simple explanation might be sufficient to they provide a of informed consent consistent with the informants’ measure understanding’ (1988.

et (Health al. and Alderson was funded to work with the staff and children on writing a book about their school. One example of children being actively involved as researchers is an edited book of the teaching and learning approach of Cleves in London (Alderson. p. 1997). data 102 . and to be subse. maximising children’s input into the research. wherever possible. The school had a reputation School for being a model for inclusive education and independent and group learning. s The committee decided what topics would be included in the book and conducted interviews throughout the school of people’s experiences—as students. Even where children are not actually engaged in the research process. both verbal and nonwith verbal. A team of Year 6tudents participated actively in the writing committee. Laybourn and Borland’s (1996) study of children’s Hill.. to provide feedback to children after data analysis as a check against possible the misinterpretation of the information they provided (Beresford. is talking research. 133). 1996. A similar approach was used for both. perceptions of their emotional needs and well-being was conducted in order to develop health education materials aimed at helping adults more responsive and supportive towards children. but including various points at which ‘children were asked to specify or make choices about the issues to be discussed’ (Hill et al. but having to work within time and resource constraints. The researchers conducted focus groups and interviews with children aged five to twelve. a variety of structured activities. 1998). 1999). more than any other. Qualitative researchers like to really a understand life experiences and events mean to people and what various thus tend to prize highly articulate and reflective research we partici-Even when observation represents a significant part of pants. The researchers clearly saw this as a compromise position. it is important. Researching with people with an intellectual disability Working with people with an intellectual disability presents a real challenge for qualitative researchers as this.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 102 Qualitative practice research in importance of understanding more about children’s experiences if are to develop educational and social programs and services we that adequately meet their needs. teachers and parents.. cited in Borland.formed the basis for an information leaflet for quently parents Education Board for Scotland 1997.

observation should take place in a variety of settings. for their voices to be heard. in the research context. Biklen & Moseley. preferably of the participant’s choosing and at a pace suitable to them.thus requires at least three things. Interviewin g Interviews are best conducted in a familiar environment. 1994b. Official records and accounts from significant others person’s life history and preferences can also be used of the as supplementary sources of information (Biklen & Moseley. to check out we have observed and to find out what the activity we what have observed means to those involved. Understanding context participants and their In any research with people with an intellectual disability. both formal and informal. We an acknowledge is still evolving and that there is developing that this field knowl. This section offers some suggestions for ways of communicating verbally with people with intellectual disability. A series meetings may be more effective than relying on one of brief formal (Booth & Booth. This will assist the researcher’s prior to understanding of the research participants and their situation. Undertaking qualitative research with people with an tual intellec.is or is not possible and may also assist participants cation that to become familiar with the researcher’s presence. that we accept their experience of themselves and are. inclusion of a person well known to the informanta useful way to handle language difficulties and may be communication problems. provide a safeguard against making incorrect assumptions about the level of communi. 1994a. we most often also want to talk with people. that we disability value the experiences of those who are not as articulate or verbal as we second. that we find ways to elicit their experience. First. Where possible. their as valid—and not as either inferior or a threat to our own world way of being in the world. though they should always be well briefed about the purpose of the research and the importance of obtaining the 103 . Biklen and Moseley (1988) recommend an initial period of observation interviewing. 1988).experience with research participants who have edge and nonverbal means of communication.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 103 Tailoring collection data collection. interview In some situations. 1988). and finally.

In summary. In for summary. their presence may also constrain what the respondent reports (Biklen & Moseley. and Booth (1994b) found joint interviews with Booth significant sometimes useful in opening up discussion but others were also carried the risk that the respondent’s voice was excluded if a more articulate significant other took over. We include here some specific approaches that researchers have found useful in interviewing people with an intellectual These are disability. How are supervisors from counsellors?). B • The use of visual cues such as pictures and photographs can assist in eliciting information from children and adults with an intellectual disability (Minkes. wordnswers. compromise approach is to have participants interviewed A by someone they know who is. • Open-ended questions may elicit only one. 1988. The teachers had rapport with the children. independent of the research topic. 1994a. 1988). getting to know the people you are working and with taking the lead from others who know them well and communicate effectively with them are invaluable principles in research with people with an intellectual disability. Booth and Booth (1994a) also provide helpful suggestions finding out about the time and frequency of events. 1994b).Qualitative Research – TEXT 22/11/2001 3:26 PM Page 104 Qualitative practice research in respondent’s own perspective. by no means prescriptive. Some people will be better at this others. 1994a. instead. and were also able to interpret in the nonverbal responses. different things and activities separately (Biklen & Moseley. for used teachers to interview children with an intellectual example. disability about their experiences of residential respite care.or two1988). at the same time. and so build up an incremental picture questions of the informant thinks (Biklen & Moseley. than others embellish it with their own perspectives. • Avoid comparison questions (for example. Some will filter out important parts of the story. Booth what &ooth. thus assisting them to already relax interview situation. and in no way substitute for taking the time to get to know the people involved and their capabilities. Minkes. It is generally more effective to ask a series of a specific separately. Robinson & Weston. 1994) and particularly useful in facilitating responses to openare ended questions (Booth & Booth. they suggest: 104 . Robinson and Weston (1994). 1994b). ask about people.

study of young people’s experiences of special In a education. We were aware that. p. might be The alternative was to exclude a substantial number of children whose disabilities precluded their making written responses but who were capable of expressing their views in other ways (orally or just as by signing.to the context are very much part of good siveness practice. and students’ p responses modified to take account of the scriber as audience. Robinson and Weston (1994) used an interview/discussionbe used with all children. the guaranteed anonymity of a rivately written response was removed. • Use their children as a chronometer—gauging other events in terms of how old the children were at the time. but for those who did not communicate easily of all by speech. for example. The researchers traded off potential ‘contamination’ of the data against as many students as possible to express their enabling views. find another marker. Being flexible collection in data If data collection is to be tailored to the needs and capacity of each participant. staff members and other children. Wade and Moore used a structured questionnaire and openended sentence completions. Here. 1993. Some students were able to complete the exercises on their own while others required assistance from teachers to read out statements and scribe responses. then flexibility of approach is inevitable. for example) (Wade & Moore. whether something had happened before or after another whose date was known . These included photographs of the residential homes. but flexibility of approach and respon. several visual cues were used. Minkes. Not only is there less emphasis on standard questions and administration procedures than in quantitative research. 21). pictures of everyday objects and small books that could be used by the 105 . for example. whether they format that could commu-by speech or other means. The same questions were nicated asked children. In their study of children’s experiences of residential respite care. in these cases. qualitative methods are particularly appropriate. whether the duration of a relationship has been weeks or years.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 105 Tailoring collection data • Settle for approximate times where these will suffice. • Wherever possible.

a on facing sheets. Caroline Thomas talks about her research with adopted children. according to their own levels of skill and interest in the report. Elsewhere. 1995) have developedtexts’ of research reports as a means of making ‘parallel research to people with an intellectual disability. Again. Involving people with an intellectual disability in research Linda Ward (1997) provides practical guidelines for involvingchildren and young people at every stage of research. Stories field from the In the following inside stories. project Ward and Simons (1998) describe several ways in which people with an intellectual disability have been able to participate in research including: helping to shape the research agenda.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 106 Qualitative practice research in children and interviewers to develop a pictorial record of their experiences at the home (the children who did this were able to keep book). through for research to be conducted on issues of importance lobbying to people with an intellectual disability. and Tim and Wendy Booth talk their work with parents who have learning difficulties. Researchers at the University research of Bristol (Bashford. not both the asseparate document. about These stories illustrate the ways in which some of the principles inside and techniques we have discussed in this chapter can be put into practice. doing research themselves. but in parallel form within the report. processes. and being involved research in research dissemination. Townsley & Williams. They accessible include formal and complete report and a simplified version. teachers used their existing knowledge of their the children to devise imaginative ways of using the available cues. They argue that providing access to both the original simplified versions helps to break down pervasive and the ‘them’ attitudes and also avoids making judgements about the and ‘us’ skill of those accessing the report. advising and assisting projects. disabled from development through to dissemination. 106 . Any user of the research levels may the simple or complete version or move between the read two. Dissemination of research to people with an intellectual disability requires serious consideration of the form in which results are presented.

authored by Caroline Thomas and Verna study Beckford. Most of the children were inter. is published with Nigel Adopted as Children Speaking by the British Agencies for Adoption and Fostering (1999). friends and then profes. due to concern about two years from the advisory group that children’s perspectives were not being included . who The report. and the views the children children themselves 107 . The study was one of a set of studies funded by the UK Department of Health specifically to contribute to an adoption law review early 1990s. Caroline and her colleagues have reported the following insights from their research: From our direct experience of communicating with older adopted during the interviews. with cent aged between eight and twelve. Forty-one children were interviewed for the study. The third stage of the sionals interview the children’s experiences of the adoptive process and explored the they had received during that process. encouraged the children to nominate and rehearse a process by which they would indicate they wantedanswer a particular question or to stop the interview not to and the children to sign a simple consent form. The interviews commenced with a present warmup period during which the interviewer obtained permission to tape. All the interviews 83 per were conducted in the children’s homes. research was commissioned in in the relation to adoptive parents. In the end help phase. but there were also three interviews viewed with of siblings. were offered certificates for taking part in the study and children a ersonalised thankyou was sent p later. Initially. This particular study was commissioned into the larger series of studies.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 107 Tailoring collection data Caroline speaking Thomas—Adopted children Caroline worked with a research team in Wales to find out about children’s experiences of the adoption process. explained confidentiality.involved in their lives. and six interviews where a parent was pairs either or within earshot. adoption agencies and post-adoptive services for older adopted children. in private. Lowe and Mervyn Murch. The older study on children who were adopted at the age of five or later focused and had previously been looked after by the local authority. They were aged between eight and fifteen at the time of the interview. The children asked were asked to create an ‘eco map’ of the families.alone.

expand.approach given that as far as we were tative aware children’s views about support hadn’t adopted been explored before. • Allow children plenty of opportunities for asking questions.but and I felt uncomfortable about that. There wasn’t anything to build on..whose well-being wasn’t too high up on ties and the 108 . books. and the interviews. • Use communication tools such as games. simplify. videos and so on (Thomas et al. we talk with Caroline about choosing to conduct indepth interviews with the children. pp. Choosing approach Caroline: a qualitative research We were sure that we needed a predominantly quali. and didn’t have the training to interpret them. life story work and contact issues. we didn’t have the metric training to administer them. It exploratory. or build on explanations if appropriate. • Elicit children’s fears and offer reassurances. While we would have been very happy to involve and engage other people in helping us. • Be aware of the possible impact of emotional distress on children’s understandin g. So we felt that that was the was best approach given that situation. We would have learned that these were children who had emotional and behavioural difficul. 131– 2). Here. 1999. • Match their explanations of new ideas to the children’s age and levels of understanding. At that time. Verna partly we didn’t have the training to use because psycho-tests. • Ask children for feedback to see if information and explanations been have remembered and understood. prompt cards. we thought we would be losing some control of the study by doing that and also we wondered what we would be learning. we learnt that it is particularly important for adults to: • Express themselves simply and clearly and use concepts which are familiar to the children... • Repeat.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 108 Qualitative practice research in expressed about their understanding of adoption. were encouraged to consider including We sort some of assessment of the children’s well-being . the process of recruiting and engaging with the children.

We had to convince them of the worth of all that we were doing.time [involved] in using a battery of tests. the interviews would have we been a different experience for the children and wholly Ihink that that has definitely been borne out by t other I’ve done subsequently that has work involved psychometric testing. with the children speaking about their we were expe.. we had a Yes. It was wanted absolutely 109 .On stay average. how found yourself adapting this because the you research participants were children—can you say a little bit about your own thinking and how that unfolded? In the recruitment process what was absolutely key to Verna and to me was that it was the children we needed. just about the level of intrusion—it’s about It’s not the length of time that interviews can take and the need to relate that to children’s capacity to concentrate and with what they’re talking about. . We already knew that.. also wanted to be sure that the But we children to speak to us at the outset. I think we have to be extremely careful whatabout put children through in the interview we process.of adoption for about an hour and a half. That what we really wanted to get at and focus was upon. but to get to them we had to get past their gatekeepers—their adoptive parents. What we didn’t know was what they thought about what they had been through and what they remembered about it. Recruiting children Dorothy: and engaging the Caroline: Can you say a little bit about the process of the research? Thinking of moving through the recruitment and engagement and finally the interviewing. .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 109 Tailoring collection data Dorothy: Caroline: scale. really had to justify to others the choice of So you indepth interviewing as the method? also concerned us was . view Had included tests. What certain which only allowed us to visit the budget children we were anxious about the amount of once and inter. riences and seemed quite long that enough.

And.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 110 Qualitative practice research in crucial to us that we didn’t pressure them and that we were giving them as much information as possible and the study. Is there anything doing that would add to you that? Possibly the book doesn’t convey the differences in those individual interviews. The about chal. With ‘all over himone point I ended up at Bagatelle in the playing middle of the interview. once we tested it out with we other children who were about the same age and who were imagining themselves into the situation of the children who would be receiving the materials. whether we needed different materialsand girls.That was the big range chal. But we also had to produce one material going to appeal and make sense to a that was huge of different children.there. physically and mentally..there was to try to communicate that to them lenge in opportunity. They gave us reassurance we needed that the materials the we’d produced would be acceptable. It was sometimes important more to 110 Caroline: .. rather be than it in a standardised way. Then we went out into the garden and walked around the garden and played football for a while before coming back and doing a little little bit of the interview. but finally came up with something that were confident in. we thought lenge about we needed to produce different materials whether for different ages. individualising the interviewing really each to able to engage and reach this child. We wanted them to be about us clear what we were expecting from them. We did go through all those options. for instance. The interviews Dorothy: You’ve written about adapting the interviewing [for child].just talking to me and very relaxed fortable and articulate—we could have gone on talking for much than two hours—to a little boy who was longer literally the place’. I mean. and whether we needed to try for boys to make them culturally sensitive. from speaking or 14-year-old girl who was extremely to a 13com.

I wondered how the child felt after the inter. That was quite really challenging . the most challenging area of that the research? I think so. On one or two occasions I had to respond children’s distress about what they were to the talkingand wondered whether I had done that about well enough. to follow the child. .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 111 Tailoring collection data Dorothy: Caroline: Dorothy: Caroline: let go of the ambition of covering all that I wanted to cover. was most challenging about the What interviewing itself? Possibly sometimes feeling as though I was moving into territory that I didn’t want to explore and havingthe stoppers on—putting boundaries up to put to discourage disclosure about certain experiences that we weren’t interested in. . extremely Sometimes it was hard just trying challenging. about the research as a researcher but your concerns were fundamentally about the children as children— was the hardest part. but it helped that after just a few interviews I was confident that the children’s material was going rich enough for us to say some to be interesting .of people with learning difficulties was funded for riences two by the Nuffield Foundation and is published years Parenting as 111 . . . so I was able to put aside people’s things . concernswouldn’t be able to get something useful that we from the children.was over and whether the adoptive view parentsbe able to pick up on any issues that were would left So that was challenging. And I think that over. though that the need to be adaptive made the I say interviews it made some of the interviews interesting. trying to get something from the interview for the study but at the same time not exploiting the childinteresting that your concerns are not really It’s . Tim and pressure Wendy Booth—Parenting under Tim and Wendy Booth’s biographical study of the parenting expe. . and to try to find just a little bit of the adoption that the child could talk process about.

They have subsequently conducted a study of children of people with learning difficulties. Realising the approach Wendy: need for a different I always remember the very first interview where I with the idea of how I had been trained to went in do interviews. on writing up narrative reports of inarticulate 1994b). And I drew up outside the house and the door opened and the mother came out and greeted me and I went in and was having a party. And it was a really good introduction of how it was going to be 112 . interview for this book. of their The number of interviews in the first stage ranged from one to six. Six the second stage. we have also included some references to their published work. Tim and Wendy have also reported on their use of in-depth interviewing with parents with learning difficulties (Booth & Booth.couples and one single parent then participated in hood.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 112 Qualitative practice research in Under Pressure: Mothers and Fathers with Learning Difficulties (1994a). of whom 25 had learning difficulties. Both stages also in involved informal contacts such as brief social calls. focusing specifically on experiences of parent. 1998). phone numerous callstrips and outings with parents. We didn’t have any formal questions. 1998). including five couples where both parents had learning difficulties. research participants (Booth & Booth. Tim and Wendy expanded In their on some of the points they have written about elsewhere. In each family. We just had an aide-mémoire. I would go in with an idea of the sort of questions I wanted to ask. andthe second stage from nine to twenty. 1996) and on maintaining ongoing relationships with research participants (Booth. It was her birthday and she the was full of people and it was a case of house just getting to know her in that situation really. The study of parents was conducted in two stages. The first involved unstructured interviews with 33 parents. published Growing Up with Parents who have Learning as Difficulties (Booth & Booth. which involved the compilation of in-depth accountsongoing situation as parents over the course of a year. at least and one had at some stage received services especially intended parent for people with learning difficulties. While we mostly rely on the interview material.

. It was I guess a way of realising her that Ias well forget all those ways I’d learnt. that may this going to be very different and I’m going to was have to have new ways of talking to people. . recognition that there are reciprocal duties and responsibilities 113 . so I just had to remember what she was telling me. It’s a real challenge of methods here.or they are referred somewhere they don’t where want . Textbook what methods won’t work with this group of people so you have to innovative and devise methods that will unlock be the So that means a lot of experimentation.and measures that [very often] in the end ments have put them down in some way—they get referred to another school or they get debarred from going some. write notes or anything. until you hit the right button—and a lot of it’s likely to be a different button with quite different people as well.you get nowhere. Tim These people have been subjected to endless : assess. Her husband was there and her two children and all friends . means failure. it door. They’ll parrot opinions. They book clam up. Taking the relationships time to develop Tim and Wendy were very aware of the potential for exploiting an intellectual disability by offering a muchpeople with needed ‘friendship’ in order to get them to participate in the research. .So the obligation from this method is pose. merely provide you with they think you want to hear. So you have to overcome that to go reticence can get people to talk and in doing that before you as well obligations come and it’s important to recognise doesn’t want to end up in a situation that. They won’t talk.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 113 Tailoring collection data Tim : really because I wasn’t able to record. One where tricking people into talking by you’re pretending you’re a friend when all you’re doing is using the guise of friendship to get information that serves your pur. I learnt so much that day about that family. How do you access people’s stories? If you work from the text. .

but in truth.. Partly of course this description wasn’tresearchers deciding to do it this way. you have to free entera kind of relationship that goes beyond into the straight researcher relationship. the kind of contract that that.So that means that at the of the interview you can’t just close the door end and goodbye and leave.. the use that drive club that they were going to. I’m here’. and we have never particularly that notion.I think in most other ring studies researchers would protect themselves against that sort commitment by drawing clear boundaries of around and saying this is what I am.. go around to the the 114 .[To Wendy] In all these studies. it’s simply doing. in return and honour those. you’ve come out with a lot of friends the who get in touch. ‘Yes sure no problem’. you’ve come out of them with of some who have wanted to stay in contact.So you with would opportunity.. the walk to the shops. but it is not an embraced inaccurate of what we did. share problems.. this is what their role I’m this is where it ends.. and so the only thing you could do to the is quickly adapt and say. but if people are expecting more of you than then the obligations. The fact of the matter is that you would have rung up beforehand. who sometimes still just for a chat . not possible to do that with this particular group of people in order to grow that trust in which they because feel to talk honestly from the heart. you established implicitly with them—you’ve got have to honour those other ambitions.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 114 Qualitative practice research in here and that if as a friend people are giving you infor. ‘Is it all right if I to come you?’.. so people overyears. that nowadays people describe things I suppose as participatory methods. ‘Oh were hi Jacqueline or John. and you would say. I am just ‘Oh going shops’. it us as was dictated by the people that we were working with andthe way in which they conducted their own by lives. and they would say. you would have arranged an appoint. I had forgotten you were coming. say fair enough.. If that’s what people want...you have to take on the duties of friendship mation. you would get there just as ment for they leaving the house.an interview.

you intended to do in something as researchers we call an interview. a concrete rather than an abstract frame of reference. 1996]. But he couldn’t lie. about approaches they had used to elicit information from specific the parents . His longest sentence in word. a present orientation. it seemed that developing trust was only part of the We were also interested to talk to Tim and Wendy story. You would do it on the hoof. He wasn’t able to He dissemble. Let me give an example. Given this.. I mean. Approaches interviewing to Tim and Wendy found that the parents in their study rarely took direction of the interviews themselves. four a half hours of interviewing was. He replied to every question with B one two words at most. a literal rather than a figurative mode of expression. methods the textbooks will tell you not to ask leading questions. questions . don’t ask leading questions with this but if you group. What you need to know is enough about the person you’re talking with to be able to judge the veracity of their replies. So if you asked a leading question and he said yes or 115 . 421). 1994b. our informants’ tended to display some or all of the conversation following characteristics: an instrumental rather than an expressive vocabulary. and a responsive rather than a proactive style (Booth & Booth. our informants were more inclined to answer with a single word. and in the course of doingyou are chatting and you would do what that.’ Otherwise he replied in monosyllables to whatever questions he was asked. always told the truth.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 115 Tailoring collection data supermarket with them. a focus on people and things rather than feelings and emotions. Although varying in their language skills. Tim We’re back to the difference between : textbook and methods in the real world. ‘Two rabbits and and a ferret. a short phrase or the odd sentence . Danny Avebury had no words [see Booth &ooth. They have said elsewhere: Generally speaking. p. you won’t get any answers.

particularly when you are in that first session. by getting him of his to no to leading questions or by getting him to say assent to one . you can still be misunderstood—even sometimes if are asking in a very simple fashion . but they can they make feelings known by those few words. I think I have always gone in and to people as I would anybody else.Is there anything else on communication—on the practicalities of interviewing? Use of photographs as props for interviews is very useful. . and ground asking to show you their albums. more about communication.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 116 Qualitative practice research in Yvonne: Wendy: Yvonne: Tim : no. but cult it’s able to pick up on key words that people being use their sentences. because of the way they can are responding . If that is the conversation I am really gettingI could still talk to people in the way I back. Somebody might only use in senten. It is you diffi-to actually explain how you do that. their back.. of where people have come from.and history and that sort of thing.four or five words in it. so it their doesn’tmatter.. so I am like to begin with. do you work that out in any given How interview situation ? suppose it is something that I do I automatically trying to remember what it was now. .of not patronising people with tance learning difficulties by pitching a discussion at too low a level. one of Thinking the things that you talk about in your book is the impor. . or I feel they understand me. . and talked then gradually adjust the way I am talking. where you are just trying to get some idea of the family. so picturelife by either ruling things out. and they could understand me. to keep putting things in are not a ifferent way and hoping that—not with d everybody. would talk. you could be sure that his reply was an honest you could build up a story. but you know ces with that are all very key words that they have picked they — don’t have the grammar. So the response is congruent with the question. You can people work 116 . build up a one. if I think they understanding me.

specific examples—that is so important with this group of people. what is.sequence. 117 . As I have said before. behaviour.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 117 Tailoring collection data Wendy: through family photographs. c Both research teams exercised considerable flexibility in their approach in order to elicit information about participants’ experiences— whether abandoning an interview to play a game with a restless child or going shopping with a mother with a learning difficulty—and so able to obtain a depth of understanding that would have were been unlikely had they relied on a more conventional format. both illustrate the importance of an individualised approach to working with research participants—taking the time to get to know and engage with research participants on their own terms. There is little in relying on conventional notions of what happens in point an interview if participants have neither previous experience of such aontext nor a view of what an interview ‘should’ be like. like she cating and would name of her newest grand-daughter. and it will only be short sentences. just one or two key words. but maintaining that focus. because I know one mother get who I say does have very limited skills in would communi. and she will come back. and because you know them. that is where it is important that you do I think know something about them. Comment s While these examples reflect very different research studies. logical that concrete focus upon events that happened. that you have taken the time to to know them.if she says to me one word. where they have them. really you what it is about. and people’s ability to be able to place events in chrono. as very useful prop and I guess the other a important too is maintaining a concrete frame thing here of reference in terms of the questions and how you progress the interview—not dealing with abstract Being wary about things like time notions. but this is how we can very communicate about it. and then you can start know having a conversation. I say the know wants to talk about her and I can ask that she her a question.

and Tim from and Wendy Booth’s ongoing involvement with research participants their commitment to honouring reciprocal obligations arose from to participants . at times containing them disclosure beyond her specific research brief. own right or in partnership with research teams. Along with increased expertise and willingness on the part of researchers to engage with children and people with an intellectualas research participants. either in their We expect in the coming years to see many more instances of children and people with an intellectual disability taking a role in every stageresearch that concerns of them. 118 . there is also increasing disability recogni.the contribution these groups have to make as tion of researchers. Caroline Thomas was careful not to take advantage of the children’s ready trust.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 118 Qualitative practice research in Both studies also illustrate the importance of sensitivity to and ethics boundary issues.

as outcome of everyday pragmatic research decisions. those who regard qualitative and quantitative methods as representing incompatible paradigms. Kidder and Fine say. 1986. 1977. and for whom mixing competing and is tenable (Rist. 1989). there is no compromise’ (1985. Guba (1985) says. or as the appro. we Instead would see researchers become bicultural’ (1987. 80).The situationalists adhere to the notion of separate but paradigms increased understanding that can be obtained value the from examining aspects of social life from different perspectives (Filstead. are dealing with an ‘We eitheror proposition. but at the same time. p. At the one extreme are the purists.of homogenizing research methods and cultures.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 119 6 Mixing methods ‘Mixed methods’ is. p. 119 . Mixing methods has been the subject of considerable debate in the social sciences and has variously been regarded as anathema. ‘We the call for “synthesis”. Oakley. not Guba & Lincoln. we want share to preserve the significant differences between the two cultures. Kidder & Fine. in social science research. the pragmatists and those taking a middle-ground. situationalist approach. like to Similarly.in some situations but needing to be carefully priate justified. Guba 1985. In this chapter we discuss studies methods of that combine these methods in four common ways. 1979. a shorthand way of saying that a study has combined qualitative and quantitative data analysis. Smith & Heshusius. in which one must pledge allegiance to one paradigm or the other. 57). 1999). the purists.Caracelli and Graham (1989) identify three positions: Greene. 1987.

We take Questions the that any combination or sequence of methods or view approaches provided it can be justified by the research purpose can be used. are going to require quantitative thing analysis. accepting that each may contribute to the strengthening of an evaluation design. pragmatists (e. Mark & (Cook & Shotland. p. that qualitative and quantitative methods ‘represent fundamentally different epistemological conceptualizing the nature of knowing. 1979.Greene. and procedures for comprehending these phenomena’ but he also argues for the inclusion of quantitative (1979. 1988. Bryman. 1985. ‘Certain questions simply cannot be answered by quantitative methods. From a pragmatic stance. p.g. recourse to these broader intel. 16). on the one hand. or include both qualitative and quantitative elements. about meanings require qualitative analysis. The ‘to mix or not to mix’ debate has lectual in abated somewhat in recent years. about how much or how many. 1989)—those 1987. or measuring one against another. (1985. 1994) tend to link choice of method directly to the research purpose and the questions being asked. 45). Walker. he argues that in practice helped to ‘a great many decisions about whether and when to use qualitative methods seem to have little.issues’ (1988. Questions that are either openly or implicitly about quantity. 108). research projects have a range of purposes Many incorporating several research questions. As Walker says. 1987. Bryman suggests that the epistemological differences between qualitative and quantitative research have become exaggerated. Caracelli and Graham. and 120 . Kidder & Fine. While he concedes that ‘much of the exposition of the epistemological debts of qualitative research afford it some credibility’. with the emergence of a fact more pluralistic approach in which methods are chosen on the basis of their suitability to address specific research needs. and qualitative methods to address different aspects of an evaluation. commissioning more likely to be concerned with getting answers evaluations are to the questions that interest them than with the ideological all purity of how those answers are obtained. These questions may be primarily quantitative. primarily qualitative. 1992. Miles The & Huberman. frameworks for social reality. if any. is perhaps not surprising that evaluation researchers It have been prominent in debates and developments in mixed methods Reichardt. while others cannot be answered by qualitative ones’ p.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 120 Qualitative practice research in Filstead is adamant.

as well as measurement decisions. 1989. Caracelli and Graham (1989) reviewed 57 evaluation used mixed methods and identified five main studies that purposes for combining methods: triangulation. develop. illustration and clarification of the results from one method with the results the from other method. newperspectives of frameworks. where development is broadly construed to include sampling and implementation. • Triangulation seeks convergence. enhancement. 80 per cent of the primary purposes of the 70 total purposes were either complementarity and half or expansion . and expansion. the recasting of questions or resultsone method with questions or results from the from other method . This includes being clear about what each approach brings to the study. Complementarity seeks elaboration.of results from the different dence • methods. • Initiation seeks the discovery of paradox and contradiction. • Development seeks to use the results from one method to helpdevelop or inform the other method. Types of design mixed method In this section. • Expansion seeks to extend the breadth and range of inquiry by using different methods for different inquiry components al. corroboration and correspon.are many reasons why mixed methods may be There appropriate. Of the studies they reviewed.initiation. p. Greene. complementarity.. we consider four common approaches to mixing methods. Qualitative quantitative then This design occurs when the findings of the qualitative research are to develop the quantitative phase of the research. ment. and their limitations.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 121 Mixing methods the different components are adequately articulated and integrated. (Greene et 259). but one that is sometimes 121 . This used is a common approach to mixing methods.

the use ofproperly thought out mixed-method design implies that all a the components have a specific purpose and are equally necessary to the overall research.one is inherently of more value than the other. to develop and define concepts that represented the needs and concerns experi. consistency validity by panels of doctoral students and and content expectant parents before final inclusion in the scale. Issues and questions emerged from the data collected in the first stage project which we felt obliged to pursue. This has led to the perception among (1970.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 122 Qualitative practice research in misunderstood. development of a scale to assess and evaluate the In the learning needs and concerns of prospective parents. considers that ‘techniques of participant observation are extremely useful in providing initial insights and hunches that can to more careful formulations of the problem and lead explicit hypotheses’. to be included in a largescale survey of British households. Without the use of qualitative 122 . not only of the for clarification of the qualitative data but to meet the design imperatives of a quantitative. Prospective items were rated for clarity. longitudinal study. some researchers that the qualitative phase of such studies is somehow part of the real research so much as paving the way lesser. This is an example of a rocess designed to meet quantitative psychometric estimates p for testing while also preserving the meaning of the items from scale the perspective of expectant parents. and to study as a whole. not for We would see the qualitative phase of such a study as it. 45–6). Imle and Atwood (1988) attempted to use qualitative and quantitative methods in equal but different ways. as often may be the case. Blalock (1970). Thinking of the different the methodological terms of their purpose gets away from any components in sugges. The inductively generated enced concepts were then used to develop the Transition to Parenthood Concerns (TPC) Scale. Some primarily quantitative researchers see aimited role for using qualitative research in an exploratory l capacity. They used qualitative methods first. but are rarely sufficient in themselves to stand alone pp. but always for example. followed by quantitative research. Laurie also used qualitative approaches—interviews and group discussions—in the development of a questionnaire module on the allocation of resources within households.by expectant parents. having a developmental purpose in relation to the quantitative phase. Even tion that where one method is more prominent.

1996. of self-reported delinquency and social attitudes.interviews with 150 of the original sample about their to-face views on delinquency. they iden. 165). In both these studies. the quantitative phase was used to identify groups of particular interest that were followed up for more in.the use of multiple methods enabled the scope of questions designed for the BHPS [British Household Panel the Study] to be far broader than initially envisaged (Laurie.. Quantitative qualitative then This approach occurs when the findings of the quantitative research are needed to develop the qualitative phase. cited in Bryman. it is doubtful that many of these issues would have emerged. 53). organisational analysis that incorporates micro and macro levels of analysis..groups which differed sharply in their degree of tified involvement in delinquency.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 123 Mixing methods methods which allowed the detailed exploration of intrahousehold resource allocation. actions. From this. The data obtained at this stage formed the basis for intensive interview study of 60 young people with an contrasting involvement in delinquent levels of activities. schedule 1992. p. with macro sociological processes such as habituated the production and maintenance of dominant organisational forms. the results frommethod directly inform the other. 123 . In both this and depth the qualitative to quantitative design first discussed.analysis using qualitative methods. 1988) initially asked 600 young 12 to 15 years old.) We talk to sectors Catherine use of mixed methods later in this about her chapter. Reicher In and Emler (1986. Catherine McDonald’s (1996) study of the application of neoinstitutional theory in the non-profit sector is another example of the of quantitative research to inform a later stage of use qualitative (Neo-institutional theory is an approach to research. to complete questionnaire measures people. p. what happens in later one phases is dependent on the findings from earlier phases. It ‘links micro sociological processes in the form of taken for granted. In the second stage they conducted structured face. or fields’ (McDonald. a three-stage study on juvenile delinquency.

Even so. Hohn and Kime (1995) in their study of parental experiences of being informed about their disability. This approach was used by Krahn. In this case.and a follow-up questionnaire to social services ology. and research progressed. Qualitative and of this quantitative methods were used. or some combination of these. The purposes of a mixed-method study of this type would generally be triangulation.rated the professionals highly. Parents also completed quantitative ratings about the professional who informed disability. sequentially or in any particular order. After considering several alternative explanations through means. the researchers concluded that the openstatistical ended interviews both enabled greater nuances of experience to be expressed and encompassed broader aspects of the informing the ratings. such as rewording some survey questions to reflect current practice termin.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 124 Qualitative practice research in Qualitative concurrently and quantitative Mixed qualitative and quantitative designs do not always have to be interdependent to the extent described above. for different components of the evaluation at different stages of the research. there were various modifications in design. the interviews tently revealed considerable dissatisfaction with how the professionals handled the situation. complementarity or expansion. As the brief.and qualitative components of a study are carried itative out independently and the conduct of one does not depend on the other. Sometimes quant. Interviews about the diagnostic experience child’s were conducted with parents and analysed thematically. number the methods themselves were not interdependent. agencies was abandoned on the basis of poor response in favour of a smaller of more targeted telephone interviews. Liz Kelly’s (1999) evaluation of a crisis intervention service to follow-up police responses to domestic violence is another example approach to mixed methods. or relate to different aspects of the a research. thus revealing a greater depth process than and complexity to the informing process than had previously been reported in the literature. In this case it does not matter whether they are conducted concurrently. Differentmay be used to address the same research question. as methods in classic triangulation. in that the findings 124 . While parents consis. different methods were used them of the to obtain different sets of data that together may shed light on how parents experienced the diagnostic process.

Following the first interview. Osmond and Peile (2001) conducted two rounds of in-depth interviews. the different As with approaches by side or be iterative. with the analysis from one may sit side phase forming the basis for the next phase of data collection. at the case-specific how level. the first interview participants were asked to select and In briefly describe a case in which intrafamilial physical abuse was a major The interview then focused on eliciting explanations issue. the second interview encompassed concerned the range of types of cases with which the worker was involved. followed by focus groups.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 125 Mixing methods obtained through one method were not used in the development of another. there had/had not been a previous history of abuse. the child how) was younger/older. In a study of child welfare workers’ understandings of child abuse. was designed in such a way that each stage The study was dependent on the previous stage. with twelve child protection workers from the Queensland Department of Families. They were asked to identify any connections or associations between also the explanations they had provided. quantitative/qualitative designs. the research team compiled lists of all explanations that each worker had identified in the first the inter. etc. in the Child and methods Woman Studies Unit at the University of North Abuse London. explanations to them during the second interview and used were presented as a for broader discussion of reasons why physical child basis abuse The workers were asked to consider a range of occurs. more generally. aboutand why the abuse happened. full In focus groups the discussion moved to an even more the general 125 . Mixing qualitative approaches data collection Just as it is possible to combine quantitative and qualitative methods to more thoroughly investigate a research problem. would the explanation change (and if: the mother/father was the identified abuser. Darlington. alternative contingencies—for example. Thus while the first interview a specific case. a in order more thorough understanding of an issue or problem can often be gained combining a number of qualitative data collection from approaches. We talk to Liz later in this chapter about her use of mixed in this study and. there had/had not been other forms of abuse identified. only one/more children in the family had been abused.both explicitly and implicitly. Each worker’s view.

although the nurses reported that they did not this was the case. Dorothy Scott (1992) used quantitative and qualitative methods in her study of the role of child health nurses in relation to the social and emotional well-being of mothers of young children (we intro. In this study. with maternal discomfort. Forty-five mothers were duced interviewed using semi-structured questionnaires and three nurses were observed at work for a week each. on and the hand. The discrepancy in the data was in this case not a problem 126 . however. there was greater congruence with the nurses’ responses. a clear in indication work may be required to understand better what that further is happening . The mothers’ interviews included of structured scales concerned with their attitudes to a number the nurses’ role. It is possible that this could have been due to sive’ the observer’s presence. well-being traditional child-focused issues if the mother seemed to perceive this as intru. p. Mothers were observed to respond tentatively when nurses initiated discussion of maternal emotional and nurses ‘readily retreated to the safer. The observational data indicated the issue was much interactional than was evident from the mothers’ more interviews the nurses seeming to retreat at signs of alone.(1992. The mothers had reported very high of acceptance of nurses inquiring about their emotional levels statehow they were coping with their baby.this study in Chapter 1). the mothers. quite different results were obtainedthe interviews with mothers and the observations through of mother–nurse interaction. example. It is. When conflict findings Using different methods in the one study always carries the possibility of obtaining contradictory findings. The nurses. however. believing that their interactions with think the mothers were similar in the absence of the observer. In the observations. This should not itself be considered a problem. based around a composite list of explanations from all twelve participants . 352).Qualitative Research – TEXT 22/11/2001 3:26 PM Page 126 Qualitative practice research in level. the observation of what mothers and In this did nurses additional data that challenged the findings from provided the interviews. seemed cautious about extending their role to other these giving lower acceptability ratings to these items than areas.

finally. what matters is how one responds to them: Discrepancies between the findings deriving from research in which quantitative and qualitative research are combined are not in the least unusual. Liz Kelly talks about her evaluation of a program methods. It is a complex program with three discrete aims. as such. 1988. Rather. right (Bryman. Cheryl Tilse talks about her use service of combination of qualitative approaches in her study of older a people who had placed their spouse in a nursing home. are best regarded as an opportunity rather than a constraint. or whether one Whether. explanations can be found. it is not in its spirit that one should opt for one set of findings rather than another. Catherine McDonald then to domestic talks her use of mixed methods in her study of non-profit about human organisations and. each of which had to be evaluated in terms of both outcome and process. did on Domestic Violence Matters—a program to implement a crisis intervention service to follow-up police responses to domestic violence.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 127 Mixing methods in itself. University of North London. here. to implement a crisis intervention service to follow-up police responses violence in the UK. Liz Kelly—Domestic matters violence We talked with Liz Kelly about the evaluation she and her colleagues at the Child and Woman Abuse Studies Unit. as needs back to the field and look again. Further. simply Discrepancies may also prompt the researcher to probe certain issues in greaterwhich may lead to fruitful areas of inquiry in their own depth. discrepant findings can to go be a catalyst to carrying an analysis and understanding forward and. view is that discrepancies in the data should be Bryman’s expected. it is in the spirit of the idea of triangulation that inconsistent results may emerge. The semi-structured interviews measured the mothers’ attitudinal norms while the observational data measured the behavioural norms which appeared to govern nurse–mother interaction. Stories field from the We include here three inside stories about the use of mixed First. 134). the methods ‘measured’ different things. p. The study is published Domestic Violence as Matters 127 .

Then towards the end I did two groups with police officers to explore with focus them of the questions that had come up about some ways the ambitions of the project hadn’t been that realised 128 . that both are equally important to approach the research. The second was calling to increase a law enforcement response and the third to develop inter-agency work in the local was areas. the had discrete aims. One was to deliver crisis three intervention by civilians after police intervention to anyone about domestic violence. We also tried to get actual operational data the police about the number of arrests and from about the outcomes of the cases. and then a detailed questionnaire to the women who used the service. . and again I hung around and observed the police officers. well. Liz and her colleagues used both quali. particularly those in specialist unit who were supposed to deal the with domestic violence . Liz is absolutely clear that three neither is privileged.and quantitative methods to address different aspects of tative the strands of the program. I would also pop in and hang around or I would go to just collect hang around for longer and I would go data and to canteen and chat with the workers and so I their got aof [how they were] doing the work. The police strand involved three much shorter questionnaires at timed intervals at the beginning of project. and towards the end. So each of those strands was addressed . in the middle. to the try see whether there was a shift in and understanding around domestic violence being a crime and police responses. She talks here about her use of mixed methods. Liz: Domestic Violence was the most Matters piece of research I have ever complicated project done . was no simple way of addressing all There those and [they wanted] a process evaluation as things. and all interviews with the workers delivering the service. and I did feel that at different times in the day and sometimes in the evening . . through multi-methodological strategies. The crisis intervention through having a database that was addressed the workers filled in. .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 128 Qualitative practice research in (1999) by the Home Office. . .

one at the beginning and t one towards the end again to look at change and contact project and we decided that instead of with the doingwe would do telephone interviews with 20 or that. what they thought and what ideas they had about how it about it could have been done differently. and they are going to know their interaction was. did [the response] us happen 24 hours. and how many subsequently they were in contact with times the project. to show how the methods interrelated and what each brought to the evaluation. What the database would just rely tell was. Using methods mixed We asked Liz to elaborate on one of the strands of the project in more detail. And also I think they didn’t acting necessarily 129 .going to want to believe that their work has ously an impact.with. but they are not going what to know what that meant to the person they were inter. we initially sent a questionnaire to 200 voluntary agencies but we didn’t get a very good return—that was supposed to be awo-stage questionnaire. but it doesn’t tell you about the process. Then with the inter-agency. how or in what way. which within other agencies they were referred on to. let’s talk about the crisis intervention. It : couldn’t on the database. It couldn’t just rely on talking to the workers. It doesn’t tell you whether much it made a difference to women. Liz Okay.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 129 Mixing methods and how they would explain it. because they are obvi. are the kind of benchmarking They performance things that people measure. 30 agencies that the project had had the most contact with . but those measures tell they performed at that level—they don’t tell you you anything about the quality of the work and what impact it had on anybody. as they should. as it was supposed to. how many women. evenorobserving the workers. That’s basic information.

we had to have feedback from the users. Because it was crisis intervention it had to whoever was there. . they were more able to do that. The idea users of doing in-depth interviews with that number is just impossible. etc. was it wasn’t all the things that people want to measure.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 130 Qualitative practice research in have an overview.of the process for any individual woman. sarily and I think sometimes they were so busy. They were just responding to what was happening in the immediacy of the situation. and we data. so they wouldn’t have the sense comes neces. Within that questionnaire there was a lot of space for qualitative but not collected in the traditional information. and also you can’t analyse that amount of So there are practicalities involved. whether they were useful. if it. they didn’t use database as a way of having a sense of what the is happening to this woman over time. how long did it take them to come. It was fundamental things: that the workers really named 130 . They didn’t work with cases as a caseworker. There were few forced-choice questions. The workers wouldn’t know whether any of the referrals they made were taken up. had a method that didn’t require so much of to use our and didn’t require enormous amounts of time time for analysis. then there was a question. would probably have most information and they on cases where the model worked best. if so what was they do something that wasn’t helpful. So we but had to have it in a way that was also manageable doable. those would be the cases that they would know. and what made a difference to them. Other times when they had less workload. given that there were over for us and 1000 of this project over three years. how would And you describe the responses of the police to you—did they do anything that was particularly useful. And then the same kinds of questions what for support workers . so we used a questionnaire. [other than] did very the police come. so the the women for whom this model worked. way qualitative data tends to be thought of. When we asked them the whathelpful. who used the project. did so was it. You would respond to be what in. .

Liz I began as a qualitative researcher and then the : second grant we got here was to do an research exploratory study of sexual abuse. Ihink it wouldn’t have been as powerful if we t had 20 or 30 in-depth interviews. what resonated with was us. but also.policy-makers. That obviously prevalence had a quantitative study on some level and so to be we needed a staff member who had that expertiseI didn’t. which wasn’t the men explanation women were currently using that the themselves. Linda Regan. We there thought was much more abuse by brothers and that there we didn’t think there was as much abuse by uncles —it about what we noticed. because has here ever since. the importance of encing beingto ascertain whether things really are in the able data because while I absolutely accept that or not. that they were really clear that it wasn’t the women’s fault. Liz also commented generally on the use of mixed methods at the Child and Woman Abuse Studies Unit.. actually that isn’t always [what the data and are 131 .respondents were saying the same tion of thing. and me understanding the importance of numbers if you are talking about influ. What we had done was questionnaires where a really significant 230 propor. and so in a way it’s the been linking of both our skills. we our subjectivity and subjective meanings bring to work that we are doing. But equally. being confronted with the fact that chose to be violent. and that person..Qualitative Research – TEXT 22/11/2001 3:26 PM Page 131 Mixing methods violence. at the same time I the think we need to be rigorous about whether things are there We had this really interesting experience in or not. What I think is interesting about that was that we wouldn’t have found that out it if we hadn’t had the open-ended questions in the questionnaire. But also. the prevalence study where three of us coded 1244 questionnaires and we ended up with perceptions about we thought was in the data and actually it what wasn’tto the extent that we thought it was. I think it’s about her understanding the importance of meaning.

So that led me to think. the research material and not an ideological position. I have a suspicion that for for your some not everybody. Liz also talked about the advantage of mixed methods from research participants’ points of view.The other have side of that I think is a more realistic position. actually. . but I never think that does mean that you are held to account and to accusations of bias in a way that people who pretend position are not. and the kind of discipline of paying to what people are actually telling attention you .. . . given that we work in an area where there possibility that what we say. include this or exclude that. how do And .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 132 Qualitative practice research in saying] . . researchers have to through with participants how they will talk disguise their identity. Whereasquestionnaire it’s very clear—we are not with a asking name. so one of the ways we a neutral have responded to that is to never publish anything where we weren’t really certain that it was actually in the so that we can defend the position from data. if it does. then the qualitative for me is.. I feel impact we a responsibility to be careful. that actually counting how many think or say a particular thing matters. With interviews. we can things be 132 . is always more messy than tick boxes. that has a direct on the lives of women and children. gives a kind of freedom to say things. . might influence policy and. We have hidden the fact that we are feminists. okay. what do make they what is the complexity of them. I think it gives you a kind of discipline. methods have been an effective means Yvonne: And mixed of doing that?Yes. but for some people that people. so they about the reveal . because life mean. you sense of some of these things. what we is a conclude. Liz I think the thing about questionnaires that is : very interesting is that they automatically give confiden-a way that participant observation tiality in or interviews don’t. a kind of them security fact that it is confidential. suggesting that different people may to provide information about themselves in different prefer ways.

1999). Why methods? mixed Catherine: The reason that it went to a mixed-method type approach was [actually] two reasons. so I think about methods as enabling different kinds of mixed telling. whatever things it might be. She has subsequently reported on methodological issues in relation to this study (McDonald. and that maybe we need to think about as complicated. And then if we spoken start talking about women and children with disabilities. Stage one The involved a survey of 500 non-profit services. Catherine organisations? McDonald—Institutionalised Catherine McDonald used a combination of quantitative and qualitative methods for her PhD study on the application of neoinstitutional organisational theory to non-profit organisations in Queensland. ten minutes. from which twelve services with either very high or very low organisational commitment were identified. so the 133 . One was because I was looking at a whole field of organisa. Catherine then sought to test each of the propositions of neo-institutional theory in each of these services.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 133 Mixing methods incredibly arrogant and presumptive about our ability a context where it is safe to talk about to make these in like five minutes. 1997. we enter another whole realm again. and through analysis of public documents on the goals and financial status of each of the organisations . is a much easier form a of communication in which to tell complicated difficult and there will be other people for things. research was conducted in two stages. She did this through semi-structured interviews with the CEO and with one worker directly involved with clients. whom communication is easier. will be some for whom the questionnaire so there is much easier thing to do.one approach could never deal with tions. that there is no perfect people method. We were particularly interested in Catherine’s rationale for using mixed methods and the way she combined them in this study.

. Organisational commitment itself is a fairly complex variable and there are a lot of dimensions to it... because nearly all of the literature around non-profits argues that these people are highly committed . . some The survey Catherine: I had to spend a lot of time in the first place constructing a sample frame .. right.It would be not to use that key variable to try and silly denote where some qualitative work would the sites make sense..work most people go qualitative. So where they varied on the look one variable a lot.. saying There are excellent instruments developed. .I started off with a key variable in all of this. and then developing that pick up all the component instruments parts. So I constructed a sample frame from three or four different sources. . they were obviously key organisations for saying where these sorts of processes that I was talking about were either being enacted or not being This means you could get a real fix on enacted. What I wanted to do was get a feel the whole series of organisations on a key for variable pick out organisations where I could go and then and at stuff in-depth.. methods then quantitative and back to qualitative. there is a whole body of theory what that variable actually consists of.. organisational commitment. about the component parts of the concept.. the theory . So I thought. I went quantitative to qualitative. . In me mixed. but there has been a lot of work done developing what the concept actually is.. So that led to mixed methods almost inevitably. you have to come at it from lots of different angles to get a fix on it.You can never find all organisations and there’s no one list of the them anywhere . The other reason was the body of theory itself invoked different levels that of analysis so you had to develop a methodology that worked at those different levels of analysis.[there about were] 134 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 134 Qualitative practice research in complexity of the social phenomenon that I was examining . . .

In highly institutionalised organisations there was virtually no internal inspection. They were. And that I in low ones. And became the sites for me to go and do they further drawing on a model from neowork. All the propositions. . the The theory fitted like a glove. They were semi-structured interviews and pretty in-depth because what I was actually doing was getting them to talk about their organisation and really them to reveal how they constructed getting that 135 . all the opposite things were there. . . Then I did a random out of that. .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 135 Mixing methods about 1200 organisations in it . sent off about 500 sample questionnaires with a return rate of about 50 per and ended up cent. theory. The interviews Catherine: It was 24 interviews all up. I observed. But I then to go and actually talk to people in those had organ.about the sorts of things that the theory isations said would be happening.. congruence. all the thingssaid I would observe there.. And It was eight propositions of why these organisations we would call highly were what institutionalised and therefore considered to be different . If it was a highly institutionalised there would be values organisation. lowest And just what the theory said would happen. So I took the very highest and the upper very and went off and had a look at them.What I wanted that for was to find outliers find and people with a high organisational commitment and those with low organisational commitment. . It was just all taken for granted that everything would be okay. There were about eight organisations which had scored very low on the organisational commitment and the rest were in the ranks . So was it in high-commitment organisations the theory the was absolutely validated. .a whole series of things. One of There was them evaluation—normal processes of inspection was and evaluation would be suspended. institutional that was a complex model in itself.

has got many different characteristics enon that and need multiple ways of getting at it so you you’re a credible composite picture. The analysis document Catherine: I also used organisational records to look at some of propositions . . So I to the needed at their ratios of financial dependence to look which looking at accounts. You’ve got building to enough of the different types of data to build get the picture and each of them have to be conceptually there’s got to be an overarching congruent... so theory or framework that hangs it all together. then the interviews and Yvonne: So you then the content analysis.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 136 Qualitative practice research in organisation.. I also wanted to have a look at their financial records one of the other things that the theory because said certain patterns of financial dependence is that on external constituents means that they have to conformexpectations of those constituents. That was it. did the survey... would be so broadly stated that tives. .I wanted to have a look at all the their and objectives because one of the things aims the theory said is that the organisation’s aims and objec. What I was doing was building a picture. They all needed different ones . things were like organisation has been formed to save ‘this children’. And they raved—each interview was an hour-and-a-half to two hours long. You know. looking at the sorts of goals and things they said about themselves and the financial stuff was just really simple sums. they completely untestable. If you don’t that’s have 136 ..its goals. which sounds as though it included a combination qualitative and quantitative analysis Catherine: . I had eight propositions of composite what organisations should be doing and not all these of them were amenable to a single analytical technique or methodological data-gathering technique. and I think the trick with mixed methods. You’ve got this phenom.With the goals meant stuff. that was just a content analysis.

Catherine: Yes. I just don’t know how I would have done it otherwise.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 137 Mixing methods that overarching analytical framework that puts it all together. I was using qualitative work a positivist within a positivist framework. find the the theory—use the qualitative. like they do qualitative work other so can go and determine what the key concepts they are can then do quantitative work by using and instru. Yvonne: Without the quantitative stage. People often do it l the way around. which was not possible . So it depends which way going. One of the things that became and clearer to me was that this was clearer essentially model. find what the important are and the central relationships constructs between them. It’s a very different way you’re of using qualitative work. .that measure what they’re looking at. you would have had to considerable prior qualitative work to try to do identify the services inductively. I would have had to do an awful lot more than 24 interviews. . which then hopefully would lead them to a ody of theory from which they can then b construct their quantitative work. that’s right—to really test out whether it was aiving and breathing theory. A lot of people start in the field. before you could start to test theory. Yvonne: There’s also something about your purpose. it’s a real shortcut. how you do you’re it. you’re very clear that the purpose of because the survey was to identify organisations for the next stage could argue in a sense that this was to get so you to point where you could use the the theory? Catherine: Yes. so you could then commence the qualitative research at a theory testing stage. And I’d never really know that I was sampling right and that’s the whole bedevilment when doing theory testing. The survey enabled you to use the the construct of organisational commitment to find the services quite quickly. 137 . then you’ve just got a pastiche. I had ments the theory first and I was going from the theory down into field. isn’t there.

but most people don’t have the direct oppor.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 138 Qualitative practice research in Potential methods pitfalls in mixing Yvonne: What are the pitfalls with mixed Catherine: methods? bit won’t speak to the other. Yvonne: What kinds of things did you pick up from having multiple data collection methods that you wouldn’t have picked up from one alone? Cheryl: I interviewed the spouses first so I came in with a erspective of the spouses and an understanding p of what the purpose and meaning of visiting for them That was important in how it situated me. . I think That one that’s it—that they have no linkage.for that. I got was. No. She also conducted in-depth interviews and analysed documents. Cheryl goodbye Tilse—The long Cheryl’s use of observation (Chapter 4) was just one of the datacollection approaches she used in her study of visiting practices in nursing homes. here about what each of the methods brought to She is talking the study . lots the two methods that I wouldn’t have got from fromwatching. But if 138 . the really other from a research point of view is that you’re pitfall jack trades and master of none and so that’s of all pretty . . [You] can never be expert in all of scary them. of why families visited. keep the I guess analytical methods that you’re using as simple as possible you’re going to get out of your depth otherwise pretty quickly in the analysis. She is very clear that no one method would have enabled as full an understanding of the visiting experience as achieved with the combination of data-collection she approaches. Theoretically I suppose what you should do is gather you experts in those various methods around and them. From just watching I couldn’t get just who staff] saw as difficult and why they were seen [the as difficult and how the staff understood the purpose and meaning of visiting. So if you’re going to use mixed tunity methods keep it as simple as possible.

The observation was also important much to because I actually situated myself as a me visitor. You could just observe they don’t to people. showed it wasn’t the ‘terrible staff’—I’d listened me that to families but staff understood the situation differently. How did they feed into each other at final the end ? suppose the most important part of the final I product perspective of the older spouses. done the observations there’s a risk If I had just that I have had no staff perspective. whether I was introduced to w people. reflected on my own experience of whether so I’ve I as welcomed. All their signals are ‘We don’t listen have for you’ and yet when you talked to them it time came as something quite different. With those different methods and the analysis. talking to them.I could actually see how lonely visiting in a congregate be. I’m interested in the relative importance of each to the product. the way people able were approached. I was given a cup of tea when the tea whether trolley by or whether I was went ignored. So I think across [mixing methods] enhanced the complexity of understanding. which made me understand setting could that better. you have done it just with the in-depth Could interviews and the observations? would have been a weakened I think it methodology. I was the really them to be central because in the final wanted analysis it was really about them and what it meant to them then the context in which they and visited..The that came out were in the interviews main things with spouses and the observations.. 139 Yvonne: Cheryl: Yvonne: Cheryl: . you couldn’t and without actually understand that.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 139 Mixing methods I hadn’t had the observations I wouldn’t have been to pick up the use of space. I’m just would observing and I’m identifying strongly with these older but the fact that I also talked to staff people. The the interviews confirmed and supported things but with staff they weren’t strong enough to stand alone. how impossible it was to talk to other when they were sitting down the end of people the verandah with a spouse who couldn’t speak . . .

when clearly Just thought through and purposeful. Nobody talked staff to them much about their work or their perspective on things. to come in as a researcher and say their I’m interested in visitors—I’ve got no interest in your life than that this is where you work. epistemological concerns researchers may have. tend to be complex and very often require a their range of research approaches in order to answer them adequately. It’s people not understanding what the issues are for each I think also that the observations would other. An important approach key though is the need to be clear about which approaches are being what the purpose and intended outcome of each is.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 140 Qualitative practice research in of saying it’s not ‘good guys’ and ‘bad guys’ here. Real-world research questions. the risk in letting go of specialisation is that research skills will become too thinly spread. a mixedmethods to research also has this potential. but in fact have the really liked being interviewed. their So think [without the staff interviews] the I observations been harder to do and I would would have have understood less. tical by very nature. In this chapter’s stories. how interrelate. convinced that no of qualitative one approach would have yielded the depth and range of understanding possible with a mixed approach. And some of them had very practical and good of what they’d like to do but felt constrained ideas by workload or by the attitude of [senior staff]. Liz Kelly and Catherine McDonald have explained their rationale for using both quantitative and qualitative methods. potentially 140 . if you didn’t value views. and used.as ‘eclectic’ counselling practice. be weaker standing alone because it would be harder to observe if you didn’t talk to staff. I thought other I’d a lot of resistance to interviews.Tilse provided an equally clear rationale for the Cheryl combination approaches that she used. can respond to complex needs more adequately than any one approach on its own. in Whatever prac-terms it makes good sense. While complex problems can be responded to they and investigated more adequately. Comment s Mixing methods is an increasingly common approach to research.

In Liz Kelly’s team. we leave data collection to focus on In the nextthe stage of research. We need to aim for increased response options without diminishing the depth of breadth of skills.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 141 Mixing methods reducing the benefit of the approach. 141 . as well as a of common in advancing understanding and the alleviation of purpose violence women against and children. be addressed by working in teams with qualitative This can and quantitative specialists. following chapter. data analysis. this works well because of an underlying appreciation of the worth of both approaches andthe strengths of each of the team members.

observation field notes. 1998). Whatever the research purpose and question. Strauss. Lofland & Lofland. 1995.over twenty separate approaches that between tified them encompass a broad range of methodologies and epistemological perspectives. notes on tapes or interview context and process. Tesch (1990) iden. we are absolutely clear of the need to be as explicit as 142 . analytic notes and memos. we do not hold firmly to any one way of doing While data analysis. from whom. Strauss & Corbin.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 142 7 Analysing data There are many ways to do qualitative analysis. for analysis may come from many sources and be in Data many forms. or content analysis of existing materials. Miles & Glaser & Huberman. There are also many texts providing detailed guides to practicalities of doing qualitative data analysis (see. there are many different approaches to qualitative While data analysis. how choices to focus the analysis and how to structure the research report. for the example. Dey. In this chapter we have opted to avoid ‘potted of research methodologies and restating analysis versions’ techniques that are readily available elsewhere. 1967. 1993. observation. and extent of analysis of data from the various sources The level will depend on the purpose for which the data was collected and involves choices that need to be made for each project. 1998. Taylor & Bogdan. Strauss. 1987. and may include interview transcripts. 1994. in practice most approaches involve similar stages. certain analytic have to be made—what data to collect. or journal entries. They may be obtained through interviews.

narrative analysis (Riessman. approachesto identify themes or patterns in the data. This is followed by the presentation m of some perspectives on data analysis from four of our researchers: Catherine McDonald uses a deductive approach to analysis. 1995). crying or sighing. the The researcher’s field notes can provide at least one perspective on these and thus should be sufficiently comprehensive to include issues contextual factors that may have a bearing on the any research. are scribing Tran.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 143 Analysing data possible about what is being done and why. Estimates of the time lack it takes to transcribe an hour of tape vary according to the level of detail required for the transcripts. the quality of the tape. while the approaches used by Cheryl Tilse. A reasonably good quality recording. These that seek studies include that are conducted inductively. with an interviewer and one respondent. most qualitative research.is a time-consuming process. especially for researchers who the resources to employ a transcriber. transcribed at a level generalof detail by a proficient transcriber with a transcribing take around three hours per hour of machine. transcription at a general level For of detail would include.she studied. Coleman and Caroline Thomas are essentially Anne inductive. 1993). identification of long pauses the number of seconds or minutes typed in) and (with indications bracketed of obvious emotional content such as laughing. and the number of voices on the tape. including non-verbals and aspects that does of interview setting that may impact on what is said and how. There will inevitably be a great deal of contextual material not get onto the tape. 168). or discourse 143 . The following section focuses on some key stages of dataqualitative Our comments are particularly relevant to analysis. such as conversation Some analysis (Psathas. a ore deductive approach. at a minimum. with a view to generating as well as those to test an existing theory which use new theory. Transcribin g Most researchers would agree that any audio-recorded data that to be systematically analysed will need to be transcribed. p. testing the applicability of neo-institutional theory to the non-profit isations organ. will tape. specialised forms of analysis. Taylor and Bogdan suggest that all qualitative research reports should (1998) ‘provideinformation about how your research was conducted enough to enable readers to discount your account or to understand it in the context of how it was produced’ (1988.

57). but essential to address this issue specifically with each it is job. the transcriber. It is also important to decide a process for dealing to with or phrases the transcriber cannot hear or is unfamiliar words with. with the transcript in hand. it can also be hard work emotionally. may require more detailed transcription. 1999). For be particularly intense. It is difficult obtain a good sense of familiarity with the data during into depth interviews or while observing. ‘A focus analysis often emerges. or becomes clearer. about their emotional response to the transcripts (see talk Matocha. In qualitative research this process is 144 . as there are so many other things to attend to during data collection. undistilled by other aspects of the interview and the research context. transcribers Experienced will understand the importance of confidentiality. and having symbol an understanding that the transcriber will not try to guess. who should be well briefed about the level of detail required. Becoming familiar with the data Once the transcript has been checked for accuracy.Wherever possible. listening to the again. Gee on the other hand suggests ‘transcribing than may in the end be relevant’ (1999. This will vary from study to study and from researcher researcher. there is no substitute for this next forms overall stage of immersing oneself in the data. p. p. be prepared to pay for an experienced transcriber. Codin g Coding is the process of creating categories and assigning them to selected data (Dey. for more detail 88). as I see what for respondents say’ at this stage (1993.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 144 Qualitative practice research in analysis (Gee. Where sensitive topics are raised in inter. For Riessman (1993). 1993).suggest identifying such places in the transcript with an We agreed such as a series of crosses or question marks.Transcribing is not only a specialised technical skill. the powerful emotions experienced can 1992).be prepared to spend time to allow the transcriber to views. Riessman recommends a more general transcription of the entire interview and then retranscription for detailed analysis of sections of particular interest. While the researcher undoubtedly impressions. can be an invaluable tape way of getting a fuller sense of what the text is about.

encounters. the of the study. coding is part of data management and involves numerically transforming preparation for analysis. Bogdan and Biklen (1992) also present a seful list of possible kinds of codes to develop and likewise u stress that they are meant to assist in categorising data and not to be rigidly adhered to. and so be made early on—data purpose collection. strategy codes. and structure codes. p. t very different processes are involved. 122). groups. perspectives held by participants.every stage of the research from here ence on.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 145 Analysing data sometimes referred to as indexing (Mason. They stress that this approach is meant to provide a mindset for coding that ‘should you a general orientation to provide kinds of things for which the to look in coding not a preformed schemata of things for data. Their list includes setting/context codes. episodes. In practice. and any given study may be focused at feelings or one or more levels and one or more aspects. roles. data. In qualitative research. In quantitative research. settlements. social worlds. a belief that a focus may for analysis will emerge at some stage. ought kind 145 . focus have identified a number of levels at which analysis can be focused. They open up the possibilities but also make the need to focus abundantly clear— no study can hope to look at everything. for example. meanings. participants’ processactivity codes. 1996). Again. definition of the situation codes. such as these are helpful in so far as they draw attention Lists to the vast array of possible ways to focus analysis. inequalities. These may range from the microscopic to the macroscopic and be social practices. decisions about the levels of analysis and types of codes should flow from the research and question. relationship codes. Within the specific aspect of focus may be cognitive each. While coding is aerm used in both quantitative and qualitative research. which to code’ (1995. a study social may on just one or a number of these types of focus codes. Qualitative analysis is generally concerned with identifying patterns in the data—different ways in which the data relate to each The kinds of patterns identified depend very much on other. the data in coding is an integral part of the analysis. The danger is that one they provide a false sense of security. and methods codes. involving sifting through the making sense of it and categorising it in various ways. lifestyles or subcultures. ways of thinking about people and objects. Lofland and Lofland (1995). may relationships. to have been focused in such a way as to obtain the too. event codes. The analytic choices made here about what to code and how will influ.

family members and workers). 1999). 146 . data (consumers. In keeping with this phenom. While team coding can be a powerful motivator for rigour. the processes have to be very explicit and consistently applied. if done poorly the potential for comprising both reliability and validity is multiplied.focus. Once we were confident of our own and other’s coding. Fielding and Lee found numerous instances of differences among team members getting in the way of completing qualitative also note the potential of teamwork to enhance analysis. but qualitative research: [Team research] obliges researchers to be more explicit about their assumptions and particular understandings of qualitative research. Coding team in a While coding in a team is essentially no different to coding alone. too tight a focus on particular types of data at an Even early stage of analysis carries the risk that unexpected and unanticipatedbetween the data will be missed. ask questions like: Does this bit fit in a category stopping to we already have or is it really a new category. the interview transcripts were coded with enological a articular view to relationships and experienced p emotions. including n looking for differences as well as similarities. something we haven’t come across before? Does it require a new code? Having to argue for new code to a fellow researcher proved to be a good test any of whether it was really needed. where both authors We both coded the first few interviews in each of three coded. Team coding worked well on a study on understanding hope in mental illness (Darlington & Bland.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 146 Qualitative practice research in of data that will enable the research question to be answered. and through sets this developed our code book. We suggest trying relationships a umber of different ways of looking at the data. This approach does not predetermine what themes will emerge but certainly shapes the kinds of themes. so. Yvonne Darlington’s (1996) study of sexually abused women was primarily with women’s subjective experience of the impact concerned of childhood sexual abuse in their lives. For example. we divided the remaining interviews each between still cross-coded a selection of each other’s work. us but Discussions about difficult or unclear coding decisions were invaluable.

p. particularly for studies where there are large amounts of data. There is no doubt that computer programs to done assist qualitative data analysis can be of enormous benefit. a potential downside to this. the risks are that this that canmistaken for the analytical work integral to coding and that. be after of work. coding which 119). It is important programs to choose a program that will support rather than constrain your analysis. one is no closer to understanding what the a lot data mean . Do I need program? a computer The answer to the question of whether a particular piece of research a computer program will always be in the first requires instance. selected parts of the data for detailed analysis.analysis are code and retrieve programs and theory tative building (Fielding & Lee.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 147 Analysing data Team research makes the research process more transparent. While this can be an advantage in closure is not reached too early on. These intended to provide comprehensive coverage of the are not possible 147 . The availability of training and/or ongoing technical assistance is also an important consideration. It will depend on the amount of data and what is to ‘maybe’. such as in teams the unchecked proliferation of codes and the contrasting problem of is too ‘thin’ or superficial (Fielding & Lee. The two types currently most commonly used in quali. our stories are relatively brief excerpts from the interviews with some of the researchers introduced earlier. in that the capability of There is the program may encourage sloppy data management early on — coding anything and everything ‘just in case’. thus knowing what you want a program to do is very important. 1999). or justifying very codes on the basis that one can come back later and broad refine the coding system if need be. be with it. 1998. There are an ever-increasing number and variety of programs available. or bring to the fore. Working might help to counteract other pathologies. at the same time. retrieve. Grbich. A computer program can manage amounts of data that it would be impossible for any researcher to keep in their head and. 1998. Stories field from the In this chapter.

The array of approaches presented here is indicative of the complexity of the and of the need for researchers to make choices about field data analysis (as with every stage of the research process) in the context particular project. brackets. they’re all about grief and loss and people 148 Yvonne: Cheryl: . We have included excerpts from our of the inter. How important is it to have a good transcriber and a ood g relationship with your transcriber? really important. So the prime tool to help the analysis have the interviews transcribed in was to absoluteI asked the transcriber to put in detail. I thought it was also I think it’s really important that the same person did them all because what I was talking about and she knew she knew the and she knew what I was trying to do. that sort of thing but it wasn’t the very detailed transcribing of conversational fine analysis. It study was important to say things like. pauses. ‘Don’t guess. It’s about you typing them up for me about and tell me what’s not clear and then it’s my job just to make the thing perfect. It’s really not that. Transcribin g Cheryl: I had one person who is an experienced transcriber the interviews transcribe The Ethnograph [computer in program] format.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 148 Qualitative practice research in approaches to qualitative data analysis or of all the stages in the analysis process—that would require a book in itself. I will go back and check it. Cheryl goodbye Tilse—The long Here Cheryl talks about transcribing in-depth interviews and analysing the data from the interviews. the observations and the nursing home documentation. crying. Catherine McDonald views and Caroline Thomas. Anne Coleman..with Cheryl Tilse.’ It also good that I had a good relationship with was her that she got very distressed—I mean they’re in really sad tapes. It was very much ‘type down what they say’..

I think it was important that somebody was really to what they were typing and really trying listening to it down. .’ So. .and coded them and kept working on scribed the general themes. was always very clear suppose I I The Ethnograph that just a cut-and-paste tool and that I was doing was the analysis. I would go back view to Interview 1 and say. It wasn’t a simple typing get job. it was very moving and I thought. ‘Here’s all the interviews. When you heard just was I how profoundly unhappy some people were. ‘Well.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 149 Analysing data crying—and I could say to her. how do you go c when sad? I find them really sad. I kept a code book and so if in Inter-3 I introduced a new code. ‘I get up and walk by around the kitchen. ... ahance to debrief or say.I as it was going to be a chapter on four interviews if — are the common themes. here are quotes here that 149 . those The Ethnograph was handy because like you could add things together and do things thattook four interviews and then I wrote them up . So it was really important to give her this’ .’ We did a lot of it’s so thistelephone and she’d say. have a bit of a cry and go back to it. ‘She sits at home typing. Analysing interviews Cheryl: the in-depth I had mounds of transcripts—very long interviews. now I’ll sit down and code I did a couple of interviews and got them them.’ tran. rather than so it say. That was really important—because views you suddenly got a sense of ‘I’m calling it something else early stages but the more I understand what in the it I can code it slightly differently.for. ‘Is it there called something I didn’t actually introduce a code else?’ so that in Interview 10 that I hadn’t checked all the other inter. I just did the usual process of coding line by into categories . trying to code as I went line along was a developing process. so I need to is find codes’. ‘How are you doingthis stuff?’ because I found when I re-listened with to tapes it was actually more upsetting than the whendoing the interview.

And when I’d done eight. It was a really good way for four mestay on top of such a mound of material and to feel I was writing about it all the time.. Or I’d identify which find inter. . I addressed that in the wasn’t next chapter . I was trying to see what was staying in as common themes or what was dropping really kept on top of the analysis . do also looked they differ?’ If I re-analyse it and say. The other that thing was use the face sheet option I did The Ethno[in ] for separating out the themes. And when I knew because who it common with.. dementia and nondementia. I at age groups .I out. to say. I’m looking at this way. how do these themes come it Th out?’. which was almost like case study reporting and saying one man is hardly quoted. ‘Well. . ‘If it’s a common theme then I have to it in every interview’. . When I saw a common theme I’d take that theme and then say.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 150 Qualitative practice research in support these themes. But I men also talked about three people who really didn’t then fit. . I wrote the chapter then again on eight interviews. .. .did not reflect the theme. But the process of analysis was making sure that I said it was common it was common . . . ‘Well. So I ended up with another chapter on difference and divergence and that’s where I look at and women and [other differences]. but what I did in then the draft was actually select out the examples final that encapsulated the theme best. e Ethnograph was very useful for doing that— for grouping and seeing the differences .than a spouse who has a physical ation disability. 150 .. . that there might be trying to some who didn’t fit and I’m not just selecting people the four or five interviews that are very rich same [data]. because the literature says that the spouse with dementia is a much more traumatic care-giving situ.I was ‘this determined him out but it was very clear he not to leave didn’t fit in’. to have an example from each interview I wanted and I could write it down . . here are the differences. So I was views constantlykeep in this view. . so I did big versions . . The things graph lookedI for were gender.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 151

Analysing data

Analysing data
Cheryl:

the

observational

What I did here was try and sort through the observa- was trying to link them to themes of tions. I inclusion and exclusion and so I really looked at them in those rather than just describing this happened to terms this visitor, this happened to that visitor—I looked at themand said, ‘Well, what practices and all provisions people and what excluded family included visitors?’ to structure it that way so it allowed me So I tried to contain heaps of information—notebooks of descrip- what happened to people. I had some tions of basic research questions, for example, ‘How were they provided for?’ So I did a description of what the pro- was—were there family visiting rooms? vision Were kettles? Was there equipment for families, there note- for families to write things in? So I looked books at provision and then I looked at the treatment of visitors I talked about treatment it was just and when this theme of inclusion/exclusion.

The analysis
Cheryl:

document
The content analysis basically just picked up what [the nursing homes] had—they’d have guides some to family visiting or notes for families. Then guides I id a content analysis and identified what roles d they constructed for families—are they resources? Like, appreciate your help with feeding and ‘We bringing or are they seen as, ‘Come join us. clothes’, You’re the family. We’d like to get to know you.’ part of You the underlying themes. So it was really know, inter- seeing how particular facilities constructed esting, very roles [for families]. clear

Anne Coleman—Five motels

star

Anne talks here about her thematic analysis of the data from her observations and informal interviews for her research with homeless 151

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 152

Qualitative practice

research

in The Ethnograph

people (Coleman, 2001), and about her use of program for managing text-based data (Seidel 1998).

Analysing data
Anne:

the

observational

In Phase One [observation] I colour-coded everything. only interested in three things— I was really people, and what they were doing in the spaces. spaces So I took the computer notes that I’d typed from the original field notes and I went through every page of those notes with three coloured highlighters— green for spaces, orange for people and purple for acts— and space named, every person, group, member every ofgroup, whatever, I colour-highlighted the whole a lot. fairly visual so once I got it to this stage it I’m was It made sense to me because I’d never keep easy. all together in my head. And the journal—I didn’t that do much analysis on the journal because in a sense it wasanalysis of what I was feeling that day and an whatthinking and it was also a record of why I’d was I goneway methodologically and not the other. It one was all the way through in the final write-up as a used way of illustrating, or making a point, or putting a date on something, but it wasn’t analysed and used in that . . sense . In Phase Two [informal interviews] I repeated way the recorded the observations in Phase One I’d so any time I wasn’t directly engaging with people that or following that up I’d be having a bit of a sit and writing down what I’d seen. So that was an identical I didn’t want to record those process. conversations they were brief and I was going because I knew to have most of them on the run, for example, ‘I’m from Queensland University, I’m interested in this, do you to the Valley often? Have you got five come minutes And that’s all it would be. I’d have to chat?’ the conversation and then I’d run away and write some memory-dot points, this is this, this and this. The only exception was if somebody had said something in a ay that was just so brilliant or so clever or just ‘I w wish 152

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 153

Analysing data I’d said that’, then I would write that down. I repeated the process of typing the [handwritten] field notesthe field notes computer format. I’d go home into and those dot point notes write up the from conversation as a separate conversation...and again the analysis of that was basically thematic. What I was looking for the breadth of the range of opinions, so I was wouldbeen really disappointed if everybody had have said exactly the same thing—they didn’t of course.

Usin The interview g Ethnograph s
Anne:

to order the data from the indepth

I like to hold something in my hand while I’m readingI was extremely concerned about whether it and using The Ethnograph would somehow take the life out of the conversation or transform the information in some but I used it partly because I wanted to learn way, it this was the perfect opportunity, and the and other that it was really useful for was as a thing management it was really useful, I’ve got to say I tool. Although kept back to the transcripts, to the hand-held going ones. I did a lot of stuff by hand...I did the And initial of rounds of coding by hand. When I couple thought broad enough idea about the whole length I had a and breadth of the content and I wasn’t coding myself in tightly, I then entered them too The Ethnograph in all the numbered lines and started to pull got the themes out there. But I never totally relied on Ethnograph . To me, it was a management tool.

an d Th e

Catherine organisations?

McDonald—Institutionalised

Here Catherine talks about how she analysed the in-depth inter- for her study on the application of neo-institutional theory views to non-profit sector in Queensland. Catherine’s work is an the example structured approach to qualitative data analysis—a of a very deductive approach, essentially testing an existing theory. 153

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 154

Qualitative practice

research

in

Analysing in-depth interviews using a deductive approach
Catherine: I transcribed all my interviews and I had preset categories drawn out of the theory. From the theory I rew out what you would call third order d concepts, which were theories. First order concepts were just descriptors, which I’d cut up into second order ones — another level of analysis, then into the third order. So I went through all the interviews once, did the first analysis, then went through them again order and looked at the [second order concepts]. I Th used e Ethnograph . Yvonne: What happened in the first order Catherine: analysis? interviews through without my schema I read the [of concepts in neo-institutional theory]. Everything that Ihought was interesting I coded and gave a code t word. a purely descriptive process and I ended It was up a coding scheme of about 50 or 60 descriptors. with Ihen pulled out each of those descriptors, had a t look at them and did searches on them to see what they saying. They would either split into higher were order concepts or sometimes two descriptors would comeone to be a higher order concept. I then into went and re-coded all the interviews and back found order concepts. Sat back, felt sick. Went second back schema that I’d developed from the to the theory, all those concepts and printed them extracted out. Yvonne: These are your second order Catherine: concepts? order ones. And then I looked at The second them according to the theoretical schema and I recoded the whole thing again. So after three lots of coding I nded up with conceptual indicators and there e were beautiful examples of all of this stuff coming out. It worked, but it was very tiresome... have gone straight to the higher Yvonne: Could you order concepts—used your schema right from the start— or you need to go through those did Catherine: stages? In hindsight, I reckon I could have. I mean I didn’t I was learning as I went along, but because I’ve noticed since then when I’ve been doing this sort of thing that I could just pull [the schema] out and go 154

how researchers get quicker and quicker . more familiar with m the theory. I wanted unusual to to see the patterns. . I still wanted to be able was to convey whether the views expressed [by a child] were or whether they were common.’ So. and I think it is two levels of confidence. I what knowthat’s an indicator’.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 155 Analysing data straight to the interviews. Caroline speaking Thomas—Adopted children Here Caroline talks about data analysis and. I can see I go. could do the descriptive stuff but she couldn’t She do I was doing. But I wanted T be sure that I duces to had actually surveyed the breadth of the material and I to try to guard against being too selective wanted on basis of what had stayed with me . just could not do the it. She couldn’t do it. which was going. I felt very unsure of myself about the qualitative analysis in that field. but it’s just confidence that. I still a little bit uncomfortable with it but I’m feel getting surer now. I’m now a ore experienced researcher. . Dorothy: Can you say a little about the data collection and analysis process? For instance. that’s so and so. . and because I’ve applied it once I now what it is when I see it. But the interesting thing was I got a research assistant and tried to get her to that do same thing. you did a content and chose to analysis NUD*IS . in particular her use of NUD*IS software program. Although the this an exploratory study. code and then to use time the volumes of material NUD*IS sorts and reprothat for you. I knew 155 . One is I know the theory so well. ‘Oh yes. . and I trust my own instincts whereas I didn’t the first time around—I thought I was do that stupid and couldn’t possibly be right. to try to get some sense of try the proportions of children who said similar things. I can see it know and ‘Oh yes. is that T right? Caroline: use were discouraged from doing it because of We the needed to transcribe. for her interviews with the T their children about experiences of adoption.

whether you were getting Did you look certainof responses in relation to the gender of the types child age at which the child was placed for or the adoption? We did think about doing that but given that it was a small sample and that not all the children such had asked the same questions.we decided not been to in the end. And were you the happy with NUD*IS did for what I loved it. we provided more sources general comments about some of the processes common to a range of qualitative data analysis approaches. But given it was took longer the time I’d used that sort of tool for looking at first inter. I think. They don’t provide These four a definitive recipe for qualitative data analysis. fewer in future. at. this exploratory study you were interested in So in the overall themes that are coming through about the adoption process—the chronology of those points of process you’ve identified. 156 . They do provide some glimpses of what happens in in practice. how some researchers have gone about their analysis. either individually or composite. stories continue that theme. say. there like it and who find it cumbersome and using time consuming but if I did the exercise again I wouldto use a similar tool.material. I had too elaborate a want coding and consequently the coding of the system transcripts than it should have. it wasn’t a surprise that I’d overview cateredthan under-catered with the coding. Instead.. I would rather use sub-codes and more global codes..Qualitative Research – TEXT 22/11/2001 3:26 PM Page 156 Qualitative practice research in Dorothy: Caroline: Dorothy: Caroline: that the findings were not statistically significant but Ihought it was important to know whether what t was quoted was a one-off example of an being extremeor whether there was some case commonality. Comment s At the beginning of this chapter we said we did not intend to provide a comprehensive coverage of qualitative data analysis— the diversity of approaches is far too great and there are many excellent for this information. I knowyou? are people who don’t T Yes.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 157 Analysing data They also highlight the crucial role of the researcher who. to be clear about the steps taken in data analysis and to able to defend those steps. Qualitative data analysis is a dynamic process and no can method isolation from the world of research practice. has to make many decisions about approach what actually to do in practice.research where there are many possible approaches tative and the researcher is so integral an instrument at every stage where of process. have a part in the experience continuing of the complex world of qualitative data evolution analysis. part of the the accumulatedof qualitative researchers. The stories presented here. stand in Any approach is mediated by the researcher and is only as good as its capacity to assist the researcher to make sense of the data collected. This is arguably even more so in be quali. It is always incumbent on the researcher to be rigorous and purposeful. whatever is being used. 157 . meant to imply an ‘anything goes’ approach to This is not data analysis.

1990. There are c clearlyto the researcher’s control over if and how their research limits will 158 New . there is an added responsibility to report. Weiss. Where research participants have shared their experiences in good faith for the research to be used to create aware. (1985).Qualitative Research – TEXT 22/11/2001 3:26 PM Page 158 8 Presenting and writing up Presentation and writing up are integral parts of the research process—no research is completed until it has been reported on. research has been funded. 1997. It is in rare for research findings to be presented in just one fact form. & Rogers. as a society. Patton. 1999) commonly draw a distinction between Owen the dissemination of evaluation findings and their use—a distinction that is equally relevant for all social research. respond to issues or them. Presentation and writing up are. an academic journal article (1988) and anOther Woma article n for a mass-circulation magazine (1986). there will be formal Where reporting requirements. Richardson (1990) provides a practical account of the publication of research on single women involved in long-term her relationships men in three forms: as a popular with married The book. just part of the process of ensuring that research findings are directed in such a way they make a difference to our understanding of that particular problems and to how we. 1990. of course. each Any with a different purpose and directed at a different audience. Dissemination is arucial first step but it in no way guarantees use.some issue or problem or to highlight implications ness of for practice or policy. one study may be reported in a variety of forms. Program evaluation writers (Alkin.

for a thesis (Lewins.of the researcher. that could influence thinking about a pro. and others. particularly whenup against the antagonism of interests embodied within runs it the current political balance. 1998). 1985.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 159 Presenting and writing up be used. Pirkis & Gardner. on the process and experience of writing (Woods. p. Rather. It enlighten them. They can necessarily apply not in the short term . Rountree & Laing. 1986. 1990. are many publications whose purpose is to help There researchers. B 98). These focus variously on writing style (Murphy. Cousins and tual use Leithwood (1986) suggest six factors that may influence the utilisation of a articular piece of research: the quality of the research. Anderson & Poole. we focus on We ways of presenting qualitative findings. Becker. We talk more about the impact of research—the shift from research back to practice— in the following chapter. do not go over this ground here. Strunk & White. But decision-makers’ ratings of research that such research can contribute to their work and the indicate work of other appropriate decision-makers. 1985.always precedes instrumental use. 159 . 1993. such as 1996. 1980. leading to some tangible action. will influ-whether research gets taken up by practitioners ence or policy-makers. arrangements. Political considerations. is not an easy task. programs and institutions cannot readily be put to work in a direct and immediate way.without leading to immediate gram action. in fact. or enlightenment. including decisions about voice voice/s to present) and decisions about structuring (whose findings. the clarity of communication with potential users and after the research. and the timeliness of reporting. Owen and Rogers (1999) suggest. write better. Wolcott. Research that challenges accepted ideas. some have argued for a wider definition. Weiss and Bucuvalas introduced the notion of conceptual use. even fashions in ideas. It cannot be plugged in to solve problems. 1998) or a journal article (Williams & Hopps. 2000). the p credi. that some form of concep. While we tend to think of use as instrumental use. 1988. Lofland & Lofland. as data are often voluminous and The latter of their very nature reflect a diversity of experiences and perspectives. the nature and implications of during the findings. but it affects do way the it they they do (Weiss think about issues and gradually it can affect what &ucuvalas. 1995) and on writing for particular purposes. the relevance of the research to the needs bility of decision-makers.

talking about how she We also structures qualitative findings. (1998) suggests two dimensions that relate to Padgett the question of voice—the ‘etic-emic’ and the ‘reflexivenonreflexive’. etic-nonreflexive report approximates the classic The ‘realist tale’ (van Maanen. different to writing a approach report. just as there will be multiple voices.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 160 Qualitative practice research in defying easy or neat categorisation. talking about their approach to writing up the stories of inarticulate research partici. 1990)? This may well change for different parts of them the report. It some is inevitable that these voices will at times be competing and contradictory—one of the challenges of writing up qualitative research is find a place for all the perspectives to involved. the research process and the reasons for conducting the research will. This will be so where research has participant been conducted in a team. By intersecting the dimen. for participatory and action research. 1988). and to Anne Coleman. the methodology. written in the third person. of course. research 160 . audiences talking the public research readings she held for homeless about people. be multiple researcher voices. talking about presenting findings to pants. The nonreflexive-reflexive an dimension extent to which the researcher’s experiences reflects the and perspectives are included in the report. For example. The emicnonreflexive report. different and for different purposes. This is followed by discussion possibilities for alternative modes of of some presentation. characterised by ethnomethodological and phenomenolog. each representing a sions. There may. at least of the voices will be both researcher and participant. in most cases. be told fromresearcher’s point of view. emphasising the researcher’s authority to report the lives of others. to Tim and Wendy Booth. return to Liz Kelly. Whose voices? One of the most important decisions to make in writing up research voices to use—whose stories are they and who is is what telling(Wolcott. but containing little of the researcher themselves. whereas other sections may the include a combination of researcher and participant voices.we can imagine four quadrants.emphasises the lived experience of ical approaches. dimension relates to the extent to which a report The etic-emic is written from the perspective of the research participants (emic) or of ‘objective’ outsider (etic).to Cheryl Tilse.

their professional background. thesis.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 161 Presenting and writing up participants. how ology and the analysis unfolds. 1994). for them. At the least this should include material on how the research came to be conducted and any assumptions the researcherthe work—their positioning in relation to participants. 1988)focus largely on the researcher’s perspectives. This approach serves two functions. 1996) but they do not substitute for a research report.sense they make of it and the approach to writing up are retical all researcher-dependent choices and should be acknowledged as such. confidentiality requires that what 161 . 67). 68). the We forget words lifeless on the page. and the repetition of materials that remain are virtually identical becomes tedious’ (1990. however. what theo. p. either to themselves or to others. with limited interpretation by the researcher (Holstein & Gubrium.findings. what patterns the researcher identifies. The researcher is. They that contribute to our understanding of how research really happens (Hyde. p. Moran-Ellis. recommend an approach that keeps the researcher in We but does not so privilege the researcher’s experiences that participants’lost or overshadowed. 1994. He also cautionsbeing overly wedded to reporting examples of subtle against differ. It reports the substantive content of research but also takes seriously issues of reactivity and the reflex-It is commonly used for formal presentations such as a ivity. Emic-reflexive reports are ‘confessional tales’ (van Maanen. brings to to broader topic. They also show the reader the evidence upon which the researcher’s interpretations overuse of quotes can become tedious and the are based.in the data that are unlikely to be recognised by readers ences whonot as immersed in the study as the researcher is. Etic-reflexive reports reflect a combination that they present a largely etic report of approach in research but include additional sections detailing the findings researcher’s of the research—how their presence may have experiences impacted on the research and how the research impacted on them. Direct quotations from participants are integral to qualitative research reports—they bring the research to life. voices are central qualitative research and this should be reflected in to any the written report. and so on. But point made can get lost in the words. It the also makes sense to include the researcher’s active voice in the method. How the research is conducted. ‘Most of us are see hear our informants as we enter their words onto a and manuscript.In studies where participants may be readily identifiable.that our readers cannot do that. Wolcott’s advice is to being the point: ‘Save the best and drop the rest’ (1990.

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 162

Qualitative practice

research

in

individuals actually said be disguised. In such cases it may be wiser on aggregate comments at a higher level of generality to rely thanmake liberal use of direct to quotes.

Structuring findings

qualitative

Inevitably, decisions have to be made about how to order the data written report. Where the research has focused on a in the social say a child protection office, it may lend itself to setting, being reported according to particular aspects of the setting (worker– client interviews, court work, staff meetings or informal staff interactions), different stages of assessment and intervention, or the perspectives participants (parents, children, child protection of different workers, managers, or professionals from other agencies). there are several equally plausible and effective ways Where of structuring the report, the purpose and overall framing of the study will help in making the choice. If a study was, for example, about emotional experiences in marriage, the report may lend itself to being structured according to a range of emotions. If the study was a comparison of emotional experiences across a range of relation- the report may be equally well structured according ships, then to emotions or to different types of relationships. the decision rests with the researcher as to Ultimately, which structure is going to provide the most effective vehicle for the findings. What is the message and how can it best be got across? Remember, though, the simpler the better—the structure shouldbecome so complex or cluttered that it gets in the way of never the content. Yvonne Darlington structured her study of women who been sexually abused as children (1996) into three broad had topic impact on self, impact on relationships with others and areas: expe- of professional intervention, healing and recovery. riences This provided a manageable structure for reporting a huge array of data. Becker’s position is that there is no ‘one right way’ to organise written research reports. He understood that his thesis on school- evaluations of their relationships with students, teachers’ parents, the principal and other teachers could be organised around kinds of schools or kinds of work relations.
Whichever way I chose, I would have to describe teachers and working-class kids, teachers and slum school colleagues, teachers and

162

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 163

Presenting and writing up
the principals of middle-class schools, and all the other combinations school types created by cross-classifying relation of relations and and class...Either way, I would report the same results (although in a different order) and arrive at essentially the same conclusions (though they were put in and their emphases would differ). the terms Whatabout the implications for social science theory and social said I policy differ, naturally. If I used my results to answer would different the answers would look different. But none of that questions, would the work that lay immediately ahead of me as I began affect writing my thesis. Why worry about it? (Becker, 1986, p. 58).

While quantitative studies lend themselves more naturally to the succinct presentation of results and discussion of relevant findings, the distinction can be less clear-cut in qualitative research, where description and interpretation are more closely interwoven. We have it helpful to divide a report into major topic areas and found to include a separate discussion section within each, after the thematic description. In this way, the author’s interpretation does not become too intermingled with the voices of the research participants, but occurs close enough to the thematic analysis to allow the reader to how it has been interpreted. Whatever choice is made, it see is important to be clear about who is talking at any point. have been talking particularly about qualitative studies We that focus closely on participants’ accounts and draw their interpreta- the data, developing theory using an inductive tions from logic. comments are less applicable to theory testing studies. These The tend to be more conceptually driven than data driven and latter the of writing-up issues we have been addressing often do kinds not arise—the theory being tested, or the conceptual framework driving the study, also provides the structure for the report. writing up qualitative research it is not always In immediately obvious what tense should be used for the various parts of the report. This is especially so in studies which include direct quotes from participants, where the primary inclination will often be to present them in as immediate a way as possible. Wolcott providesadvice, however, in the suggestion to write useful descriptive the past tense right from the start, even if this passages in seems even disrespectful to participants you may still be odd or workingwhose reported experiences are still vividly in your mind with or — certainly one works with their words for a long time through the analysis and writing up processes. He says: 163

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 164

Qualitative practice

research

in

By the time your manuscript has gone through many iterations, editorial review, and publication you will discover that the past tense no longer seems so strange, and you will not have left informants and saying whatever they happened to be doing forever doing and saying during your brief tenure (Wolcott, 1990, p. 47).

Is there a book in it?
A question asked by many researchers, especially postgraduate students, is whether they should attempt to publish their research asbook. There are no clear-cut answers. It depends on the nature a of work and whether there is a publisher which considers there the is market for it. Some research reports (theses) form an a interlinked it ‘makes sense’ to keep the bits together—equally, it whole and maymake sense to separate the bits, as the integrity of the work not or potential impact would be diminished. Publishers will its primarily be concerned with whether the book will sell, so it needs to be on aopic with reasonably broad and current appeal, and with an t identifiable audience. important to get advice. Supervisors and thesis It is are examiners useful people to start with—they will have read the thesis and be able to comment on whether they can picture it making may the transition from thesis to book. Thesis examiners are often asked to comment on how publishable a work is, and may include specific comments about possible forms of publication in their report.you do decide to try to publish a book from the thesis, the If stepfirstto find a publisher. Choosing carefully can save a lot of is time. as prospective publishers those who are known to Select includeof the type you want to write in their list and then works prepare a book proposal. Any proposal should include the rationale behind the book, a brief outline of each chapter, the proposed word length target readership. The book proposal is more than and the just sharing information about what you would like to do. Its aim is also to convince the publisher that this is a project worth taking on. If are successful first time, congratulations. If not, consider you trying There are many reasons why a publisher won’t again. accept aat a particular time; some may suggest other publishers project you could approach. a book from a thesis is always a major task that Writing requires hard decisions about what is left in and what goes out. That can be 164

Qualitative Research – TEXT 22/11/2001 3:26 PM Page 165

Presenting and writing up difficult at first, after all the time spent working on the great labour that is a thesis. It is probably a good idea to leave it a of love while anyway—to get some distance from it. Consider writing some articles in the meantime—this is often easier, as they tend journal to chunks that can be pulled out without disturbing ‘the be thesis’ itself. In writing a book based on a thesis, you may find that some prominent sections of the thesis take up only a small part of the booknew sections may have to be written. The literature will and always be updated. While it is difficult to generalise, the book need to will to be shorter overall, the literature review will be less tend exhaus- the methodology will occupy a less prominent tive, and position. Depending on the audience, the methodology may best be included as an appendix. As a general rule, the more scholarly the intended the more extensive the methodology will need to audience be. Yvonne Darlington rewrote her PhD thesis on women’s experiences of childhood sexual abuse (1993) as a book for a practitioner audience (1996). This required considerable restructuring of the original material: the book is less than half the size of the thesis; the methodology is placed as an appendix, and greatly summarised; review is shorter and more targeted to the the literature central of the book; and one major section of findings— content the women’s remembered experiences of abuse and its impact in childhood—does not appear, as the book is focused on longerterm impact, experiences of counselling and processes of healing and recovery.

Alternative presentation

forms

of

Research need not be disseminated only in written form. Paget had excerpts of her research on mis-communications between doctor and patient adapted and presented as a piece of formal theatre,and performed by theatre students. The performance staged text drawn from a written report on the research, including was the doctor–patient exchanges as well as Paget’s analysis. Paget says of this work:
Performance promises a far richer and more subtle science of culture analytic text can establish. But it makes different than the demands. a narrative, drama, action, and a point of view. This It requires work succeeded as theater because the original text had a narrator who

165

1990. p. 1989. 1990) have conducted staged readings of their research on the social organisation of professional theatre. First. p. Becker. p. 1999). not in short quotes used as evidence for the generalizations we want to make. 95). You can’t disguise the speaker’s own meaning when you use long quotes. Scripts several are ‘multivocal’—this one has twenty-five voices speaking—and the multivocality further deprivileges the analyst (1989. McCall and Morris (1989. p. also. talking as themselves but also taking roles of research participants. because it had a good reason to be told (Paget. there is room for three researchers’ voices in the presentation of the all ‘To reduce our three voices to the single authorial voice in a research: conventional social science text (and thus hide our negotiations. Theirs presents an on the innovative approach/response to ‘telling it like it is’. because it contained a real dialogue. You make them say just what you want the audience to hear and can’t no more (1989. the researchers are the performers. 152). and the ideas we share) would misrepresent the disagreements. the researchers used an children interesting way to protect the confidentiality of the children while allowing had to say to be heard verbatim in the public what they domain.. more material that isn’t exactly about the point being made. wayproduce performance science scripts’ (Becker et al. This protected the privacy of children the 166 . playing the recordings to prospective adoptive parents and social workers during their training sessions them to appreciate the subjective experiences of to enable adopted in their own words. in long speeches than or in conversations. Becker & Meshejian. Long quotes contain more ‘noise’. see also McCall.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 166 Qualitative practice research in In a different but equally innovative approach. 95).. we 94). They taperecorded other children reading the transcripts of the adopted children’s interviews. reported dramatic events. Turning the sociologists into characters in the script makes them less authoritative and easier to argue with. with less intervention by us. Both have the effect of deprivileging the researcher: In Caroline Thomas and her colleagues’ study on older adopted (Thomas et al. In these theatrical presentations. especially if you give them voices and let them disagree with each other. also argue that the research participants’ voices come They through strongly in the performance more medium: We let the voices of the people who talked to us be heard more fully usual. and.

the then 167 . 1999).Qualitative Research – TEXT 22/11/2001 3:26 PM Page 167 Presenting and writing up adopted children while evocatively conveying their experiences powerfully than the written word would have much more allowed. I don’t think I entirely succeeded. The information that we data ought had from the police wasn’t available. without having core . It draws on different parts of like. so it to have was having to kind of wander around. the chapter on the police is clumsy and I think dense and tries to use too much of the data without distilling it. Stories field from the This chapter’s stories are again drawn from our interviews with researchers. That was the hardest one to do partly because the was incomplete.to any study—is reflected in her resolve to integrate bution the findings from both approaches in written reports on her research (Kelly. the and there is a thread weaving through it data that different parts of the data illuminate and help explore. Liz Kelly—Domestic matters violence Here Liz talks about her preferred approach to writing up mixed methods research. . then it is okay. central a I think the thing for me is that if you have a clear thread or couple of threads that you can follow and data weaves into this thread. Looking I think Domesti at c Violence Matters now. Writing up a mixed methods study Yvonne: Liz : How do you write up a mixed methods study? with enormous difficulty. . The chapter on crisis intervention I reallyI think it works. Her position that qualitative and quantitative equal but different—each making a unique methods are contri. Readers who are interested in the published work of these researchers and the other researchers interviewed for this book can follow-up the references in the bibliography.

which are the more complex data in the middle and then which are the that give you some kind of clear insights. One of the ways also we it is to show contradiction. For example. or something else? partly that this was the first piece of It is research where the police were subjects. there are some can police who are offensive and whatever. Quantitative data can’t give you a mas as sense actual messiness of real life. understanddifferent way. is that you use the more quantitativescene-setting—this is how many. in relation have to domestic violence. I think I have come that actually. but most of them are strugglingwhat to do. a sort of more accessible description. I think a lot of times qualitative data are used as a kind of illustration. as I am now. Qualitative data helps us arrest in a to communicate some of the complexity and the dilem.well. because on the one q hand the demand has been for more victim-centred and empathetic understandings in response. I don’t think I was as clear about that when I wrote it. how much of that—and then how the qualitative data enables you to explore. or with that. the demands to see of women’s movement have placed the police in the a uite complicated position. there use is lot of police officers who understand arrest in one this and another lot of police officers who way. and it is partly through with researchers from other countries talking who also done research on the police. and I don’t thinkit is stated there as strongly as I would that now. If data you have that. and qualitative of the data show you that—okay. I suppose one of data the things you do. but on the 168 . and to make sense of to know often conflicting messages. and there officers are some who are great.whether this bit hard of goes with this. yes..Qualitative Research – TEXT 22/11/2001 3:26 PM Page 168 Qualitative practice research in Yvonne: Liz : you can work out which data are ‘setting-thescene’ and come at the beginning.. unintentionally. In dialogue. that about having time and distance from Was the study. then it is actually quite difficult. because it’s to know what to select. this data as the is much of this. but there are other ways that you use it. or there are too many possible don’t threads to follow.

and not enough signposts as to why is this information here. what is it doing. or in a way that it is more complicated than that.. to find the thread. but when I was writing it up I hadn’t quite grasped it clarity I have now. chapter and have enabled me to move in and out of would the qualitative and quantitative in a way that had a sense of direction. conceptual. So to me they need to be together to illuminate. difficult and dilemma-filled reality. The numbers don’t enable you to know that.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 169 Presenting and writing up other hand to respond to domestic violence as arime . but actually sitting underneath these these numbers is a much more complex. I would have been the thread in the think that police that would have given it a different feel. but these things you do a subconscious level without [realising]—you on are making me bring to the surface things that are just implicit in my work—that I do see the two kinds as being in a dialogue with each of data other. I feel that chapter provides lots and lots of information. Had I [had that with the clarity]. or if you are going to going even understand it properly. but you need to know it if you are to do anything useful. all And .. either in a sort of confirmational way. or at end of the day does one become more the influential than the other? Or does that depend on the context of where you are reporting? 169 Yvonne: Liz : Yvonne: . you do in practice is to interweave the So what qual.I don’t to this think I know how to write it in any other way would now. is interesting and it hasn’t been a which principledto do this with some kind of deeply decision sophisticated rationale at all. and some.the interpretative. and I c Domestic Violence wa think of a journey where I arrived at this s Matters part understanding... itative in the same section even . I think it is to do with how I understand multi-methods. in the same chapter. . Do you see the methods as more or less equal. . . those that are what the researcher’s job is in things relation material.it is about saying it is all very well to times have numbers. Sometimes they just illuminate each other. analytic.and quantitative data.

They are part of the process of discovery and sense-making of the material. in verbal and written form. andof that. there is my perception and [there are] external perceptions.by. the stories are still there and the quotes are still there so Ihink I let a lot of it speak for itself. information. For the practitioners [nursing home staff] all I as really keen not to have a critique of what they w did to say ‘here’s a way of understanding what but some 170 . Presenting audiences Cheryl: research to different Some of the things were the same. or be that influ. and there are in lotsaudiences that are in a position to make a of differ-different kinds of differences. I see them as equivalent and as necessary to one another. content and process of presentation for different audiences. is because trying to speak to the audiences in the you are way they find easiest to absorb or accept. So I wrote lots of quotes and see then ‘But other people said similar things to you’. With the participants [relatives of nursing home residents] I was really keen for them to themselves in it. you know if you are trying to do this difficult job of writing for multiple audiences then I think thata reason for trying to use them together. Cheryl goodbye Tilse—The long Cheryl has presented her work. but me. is always to try to speak to audiences simultaneously. that while for practitioners it is often the experiential that counts.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 170 Qualitative practice Liz : research in Well. 1997b). Here she talks about how she has tailored the 1997a. said. I always thought strength was in the stories and the quotes that the — I think they’re very compelling and they’re still with Some of these I wrote back in 1994. 1996. So our ambition ence. but I knowfor policy-makers it is numbers that count. One of our guiding enced principles knowledge that is of use to those who is to create arethe position to make a difference. But I do put t an interpretation on it. to several different audiences and in a range of formats (Tilse. So again.

I suppose. and it is very rich powerful. always grounded in trying to convince them I knew I was talking about because I’d talked to what older and this is what I had seen happen. I’ve always seen it as a policy thesis. and always through at examples. about understanding issue of residential care of older a key policy people and trying to make policy more receptive to a broader to understanding what some things approach are about. For practice audiences I was that these are the sorts of things people told saying me about. Youjust frame it. . this is what happened for them and this is how some people respond to it and it seemed to I saw work So I think that’s the strength of the well. that it really speaks for itself a lot. It’s about how you incorporate meaning to activities such as family participation rather than simply rational decision-making models. same So examples are the same. . All your writing is targeted at can an audience—you know who you’re talking to and the points you want to make. people Although it’s all about human relationships and meaning. it’s just what you the select and what you leave out and the way you write that are different. But it’s more similar than you would imagine because it’s the same stories. qualitative material. [just] told at times with a different level in mind. So I write differently for the policy audience. So it was very much—I and didn’t talk about roles. but the qualitative data is so it speaks for itself . So I was trying to get some practice do ideassome new thinking. Tim and pressure Wendy Booth—Parenting under Here Tim talks about the challenges involved in writing up— and doing justice to—the accounts of people with an intellectual 171 . resources. At a very policy conference you talk in terms of policy and use the examples to question the policy. rights and relationships — a human level.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 171 Presenting and writing up of the things that bother you are about—why families are so demanding and why older men don’t seem to know what to do when they come and what can youabout it’.

and we have that the words. deal with some of those issues? You How do you are saying if you don’t in a sense take over and write the 172 Yvonne: . which gladly expand on at length. because it emphasises their [inarticulate.and actually is also doing a disservice to sible task the informant. the inherent incoherence of most transcripts may be exaggerated because they lack the same command of grammar. make of a transcript is presenting them with an sense impos. is while people have the stories.In tell the of the group of people that we are talking case about. with a 1996). when.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 172 Qualitative practice research in disability. An approach to writing up the stories of inarticulate informants Tim : The challenge is how do you tell the story? How do you the stories of people who lack words?.Lacking all that and asking a reader to selves. You wouldn’t be able to use it to convey information to reader.So there is a been real challenge there for researchers. in the sense of how do you cope with this quite fundamental dilemma. are though it is a real challenge. lacking the kind of struts and trate supports interviewer or researcher had—prior that the knowl. based upon ten years of research. have able to express themselves. an understanding of their edge of lives. Now... because the reader wouldn’t be able to a pene. most only of time you would end up with an incoherent the story.. they might very well ness]. frankly.My considered view. of being and presenting the material verbatim. we can lend them the words to enable them to their stories.the prose... That in itself brings all kinds of tell ethical issues of legitimation and representation. 1994a) as well as that of children of culty parents learning difficulty (Booth & Booth.the person. if as a researcher you believe in the notion absolutely true to the data that you collect. issues.(Booth & Booth.in context.. but I am sure I would you not here to listen to me go on about it now. the same vocabulary and the same oratorical skills. He draws on their study of parents with a learning diffi. experience of the way in which they express them.

that they were presented with the text. but what kind of safe. which takes culties the of a conversation between two sisters. but by the the of the words isn’t necessarily the same. Anne’s oral presentation of work to the homeless people in her study has some her parallels with performance science. Tim : Anne Coleman—Five motels star As part of the process of feeding back findings to the people who participated in her study (Coleman. ants.do you build guards in? I suppose in thinking about how you do it. in and okayed they it. but it was the best I ould do—I wanted to take the whole of the c thesis to people and say ‘Is it all put together back okay?’ Now obviously I knew that a large number of people give a damn about the literature review wouldn’t or what I said in the methodology. I would hold in my mind a chapter that appeared probably in growing up with parents who have learning the diffi. so I did some readings at a drop-in centre [that] I knew a lot of these people accessed. 1996]. albeit in a less formal way. Everything that is in the chapter was spoken informants—nothing has been added. it is not going to get told. Anne had conducted a series of research readings of her work. order The ordering of the sentences isn’t necessarily the same. Again I flagged them with the graphic image 173 . 2001). Research people Anne: readings for homeless This was not entirely satisfactory. have been cuts made. Many form of techniques that we used in constructing the that chapter are techniques that would apply to our work generally. and the There material very heavily edited into a coherent text has been that is true to their story and was validated by the inform.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 173 Presenting and writing up story.book [Booth & Booth. While she did not talk of her presentation of findings as performance. elisions.

And then with another one was when quite a few Murris turned typical up day and there were quite a few other people. as if to say ‘Oh said yeah. Read they’d us another bit about somebody saying something. I’d always get got that confirmation. . smiling or nodding. that one. There was just This one and we spent about three hours together. Nobody would say anything. Oh yeah. . So I actually got to read some of methodology and talk about some of those the issueshim. read that other bit. . And once they’d warmed up they’d start to ask questions or they’d start to say things. Tell a bit more about us that. Oh I know yes. one It payday. and the atmosphere was.’ So I guess that’s not we the that people usually get their [research] verified way but that group of people that was a strong with verification. That’s fascinating know but we’ve heard it all before. we what you think Anne. That particular session was about acknowledgement for them—I’d start reading and say. ‘No. You’d see people. ‘Where are we? Where are we in there? Yeah. ‘We’ve all was had drinks. no. we’re just going to go up to the some club Anne’s going to tell us a story.’ And that and was different. typicalhave been one like? one is typical at one extreme. It was shorter. but when you it to the ‘read us our bits’ stage. That was just delightful. 174 . even the people who nothing.’ So they wanted the bits where they were quoted? Yes. the consistent that for the first half an hour it was like I thing was wasexpert. they’d just the sit there.know what that feels like. she got that right. no. .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 174 Qualitative practice research in Yvonne: Anne: Yvonne: Anne: and I went down there and read big sections of it to some people happened at the readings? What would a What . That guy was productive because he was interested not just really in findings and hearing the bits about him but he the was interested in all of it.’ So that was really like just selected bits but it was a good process of confirmation for me at the end . It was probably totally only an hour and a half because of course maybe people had other things to do and they wanted to go and have another drink. at every session.

illumi. They all retain the qualitative richness and uniqueness of the data. The purpose of research in the human services is always to have some impact. for change to occur as a result of the work—whether a some direct instrumental effect in the form of changes to practice or policy or a ore general contribution to the development of thinking in m an We focus on the impact of research in the final chapter. whether in Liz Kelly’s attempt to inte. He opted instead to edit the text to their makestory more easily accessible to the the audience. There necessarily a marked delineation between these stages. The researchers are aware of the power of qualitative research to move audiences. People’s stories are treated respectfully and not used for gratuitous effect.’ Others. Comment s These four stories illustrate a variety of approaches to presenting research to diverse audiences. is not While 175 . They’ve area got good communication networks so as long really as keep tapped into the network you can get you the information out about what you’re going to do. It’s all right with know us. but do this always with a particular purpose and audience they in mind. And I now think that it’s a good way. Some people said. wanted to hear They it.and communicate both the numbers and the grate underlying Cheryl Tilse’s use of quotations from spouses to messiness. trust you’.policy point. but other people were really We keen. we know you. or Anne Coleman’s reading back to nate a homeless people what they had said during the research. we what you’re going to say. excerpts also illustrate that writing up and These dissemination are processes that need not end there. Tim Booth is also aware of the communicative potential of the stories of his participants with learning difficulties dilemma that this would largely be lost if he used but faced the just verbatim accounts. ‘We couldn’t care less.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 175 Presenting and writing up Anne felt pleased with the readings and believes this approach to presenting findings has implications beyond her study: Anne: I think that this has overcome one of those big methodological gaps in research with people in this about how do you do feed back. area. ‘Look.

it also happens along the way—research can have an p impact any at stage. 176 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 176 Qualitative practice research in research dissemination starts in earnest at the completion of a roject.

It is also hard to differentiate particularly in the impact of one study from that of others as one piece of research can to another. policy-makers and just practi. Surprisingly.research in the human services can be used to enhance itative the response of the service system and the broader community to complex human problems. very little attention has is a key been 177 . Some studies have an obvious and imme. creating multiple ripples in a reservoir of lead research and practice in which it is impossible to determine the ripple from particular a stone.effect while the effects of others may be almost diate imperceptible. services. possible part of this. researchers have an important role to play in determining the impact of their Making recommendations in the most effective ways study.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 177 9 Epilogue: From research to practice. will mostly be people other than researchers who While it put will into practice the implications of their findings. In this chapter we look briefly at how them in their qual. the short term.study as human services managers. thus completing the loop back to practice. We conclude by giving a few examplesour from interviews with researchers. It is not always easy to identify the impact of research. programs and politics We began this journey looking at how qualitative research can be generated from the swampy lowland of practice in the human We end it by examining the impact of such research. researcher may not be aware of the impact which a The has.may not be fully aware of the research which is tioners influencing decisions.

• Work closely with agency personnel throughout the process to minimise the threat which unexpected recommendations can and engage stakeholders who have the power to pose. The pay off from these advances comes in the recommendations we make. commented that: Recommendations have long troubled me because they have seemed the weakest part of evaluation. thinking carefully before recommending fundamental changes. • Consider the contexts into which the recommendations must fit and make realistic recommendations. • Don’t wait until the end of your evaluation to begin thinking about recommendations—record possible recommendations from the commencement of data collection. We have made enormous progress in ways of studying programs. 1988. • Decide how specific you want your recommendations to be and consider the possibility of providing options for decisionmakers.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 178 Qualitative practice research in given in the research literature on how to do this. • Make your recommendations easy for decision-makers to understand. 90). But we have made little progress in how to construct useful very recommendations p. applicable also to other types of They are research. Patton. • Consider all issues in your evaluation to be ‘fair game’ for recommendations. categorising them in meaningful ways (for example. pp. consider becoming involved in tion strategy the implementation itself. if invited. not just those the research was designed to investigate . • Draw possible recommendations from a wide variety of sources. including earlier studies of similar programs and program staff different of levels in the organisation. • Show the future implications of your recommendations in as much detail as possible and consider planning an implementa. methodological diversity. implement them.and. (Patton. and a variety of data-collection techniques and designs. Perhaps there needs to be a research project on how best to write research recommendations! Below is a summary of the suggestions which Hendricks and Papagiannis (1990. 122–5) have proposed for making recommendations in relation to program evaluation. an expert in qualitative program evaluation. short-term and long-term) and adapt the way recommendations 178 .

across The challenge is even greater when crossing national and cultural boundaries . Extra. However. given context the dynamic and turbulent field of the human services. The external environment will also greatly influ-whether there is a receptive climate for the research. 179 . as do demographic differences. Most research will have implications which go well beyond a specific research site. This may mean that the findings are seen as it was datedso are easily passed and over. Conversely. as they tend to be taken for granted. and may need to be disseminated in different ways to different audiences. as very different legislation and service systems services exist regions and states. This helped to give the study a higher profile than it might otherwise have received. The suggestions above are focused on making recommendations in relation to the research site of a particular program which has been evaluated. Often ence the origin of the research occurs within a particular policy and social which also affects how it is received. making recommendations researchers need to be When mindful of the context of their study in terms of its time and place.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 179 Epilogu e are presented to the way in which the decision-makers normally receive information. the research finally delivered in a very different context to that in may be which conceived. An example of this is the study on adoption experiences of older children undertaken by Caroline and her colleagues. the findings can sometimes take on greater salience in an external environment which has changed significantly since the research commenced. This is a very particular when crossing different service systems in the challenge human field. Researchers and those who utilise issue research must therefore be very aware of the core components of the context the research was done and the core components of in which the context in which the recommendations arising from it may be implemented . as was discussed in the previous chapter. Sometimes the most important contextual components of the both research site and where it is hoped to apply the findings are not visible.the findings of a study to other contexts is always a polating centralin human services. The publication of the book Thomas coincided high-level political attention on the issue of with very children languishing in institutions when they were eligible for adoption.

The question then arose child as how applicable the findings and recommendations were to in relation to the new context. salience We prefer to describe it as a process of transplanting innovation across landscapes. The researcher will need to think carefully about what to be changed to make the program or policy fit the needs context into which it is being introduced. while being mindful of the risk of damaging what might be the core ingredients which make a program work. This is true for both quantitative and qualitative research. Some human services researchers are actively in a strategic process of developmental research involved which and Leavitt (1990) have described as ‘missionFraser oriented and entrepreneurship’. There is obviously a trade-off between maintaining the integrity of a model and adapting it to fit new contexts. In developmental research of research this the evaluation is just one part of the process described nature by Thomas (1978) as consisting of the following phases: • Analysis (identification of need) • Development (designing the social technology) • Evaluation (assessing the program) 180 . researcher’s role may go beyond dissemination and The making recommendations.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 180 Qualitative practice research in In industry this process is called ‘technology transfer’ but in relation to the human services the term does not give sufficientto the contextual features which need to be considered. precisely why the study investigated interThis was organisational and inter-professional conflict—but by the time the research was completed. the highly politicised controversies surrounding the protection system had diminished. the conflict between various organisations in the child protection system was particularly intense during the period of data collection. while not minimising the significance of the study’s findings to other settings. The metaphor of taking a plant from different the climatic and soil conditions in which it originally grew and trying to grow it elsewhere draws awareness to the different contextual characteristics which must be considered if the transplant is to be successful. In making recommendations researchers need to acknowledge the possible limitations on the transferability of their studies. the generalisability and transferability of research are Thus in direct proportion to the degree to which the findings are contextbound. In Dorothy Scott’s child protection assessment study (outlined in Chapter 2).

Liz Kelly discusses how was her study’s different conceptualisation of domestic violence intervention struck a chord with workers in the field of family violence (Kelly. We conclude with Dorothy Scott reflecting on the longterm outcomes of her research undertaken with maternal and child health in the early 1980s (Scott.(Tilse. as some of the following of stories well. illustrate Stories field from the We have selected quite a few of our stories to highlight the impact that qualitative research in the human services can have.I’vea lot of people in the last ten years. Caroline Thomas then speaks briefly about views the impact her adoption study has had in the year since its publication (Thomas. 1987b. 1996). 1999). 1987a. Jackie Sanders also talks about how the research she 1999). The first one we present is Anne Coleman’s study which was still in progress at the time we interviewed (2001). Fraser and Leavitt (1990) use the US model of intensive familybased services. ‘Homebuilders’. nurses Anne Coleman—Five motels Anne: star Yes. an excerpt from the interview with Cheryl Tilse... as a case study of this type of research. However. both in the term. most research in the human services will not be partsuch a linear strategic process. I’ve quite a strong sense of [impact] now. in talked to the 181 . short through to the long term. which typically involves a large number of people and organisations. This is followed with may have. done from students and practitioners to policy-makers and funding sources. has with Robyn Munford has been used by a range of people. her. sometimes commencing during the study itself.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 181 Epilogu e • Diffusion (disseminating information) • Adoption (implementing the program). and how the political environment in which it released gave it a special salience. 1992). were interested in finding out about the impact that We research even at such an early stage. It picks up from the previous chapter and relates to reaching audiences beyond the research sites through conference presentations and media inter.

but I can see a probably shift local level. and I know that the longmuch term homeless people whom I’ve looked at in this research won’t be much better off.. I now hear tive? the Councillor saying things like.I targeted workers and . If you’re now going to tell to leave these spaces. and I certainly spent of the after I’d written the thesis talking about the need year to understand the perspectives of older spouses. in terms of making life different at the for people that I’m concerned about.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 182 Qualitative practice research in bureaucracy. . I did a lot that—I got invited to the annual general meeting of of Gerontology Association.I feel make like been a very useful piece of work it’s already. in homelessness services and in politics the sorts of things I’ve said are simply that .’ So I know that not in terms of the federal government response it’s pretty unchanged. actually right. So saying these older spouses need a voice.’ move Some in Brisbane City Council—not necessarily people the levels—are starting to take on this idea of top ‘Yes. .. where it the will a difference is at the local level. where will they go? them And do you expect that sort of response to be how effec-But in the last twelve months. . These are the spaces that capitalism has left for them.. really I was ties.. I think I had an impact on all the facilities that I worked in. So. I did some in-service training for some church-run nursing homes sessions where they brought in all their workers . These are all the same that’s people been moving on from these spaces for ten we’ve years and they’re still the same people so obviously this is a very good way to go about it. ‘We will no local longer homeless people on from this area. . and I did a talk for nurses in aged care h facili. I was to get to front-line workers rather than trying social workers or policy workers. I was trying to increase the understanding of the issues for them. and it’s pointless to move people on.I wrote a report in 182 . Cheryl goodbye Cheryl: Tilse—The long I didn’t change the world. I did a talk with staff at the a ospital.

. here What do to help?’ .. I that that’s probably the least likely way of know getting information to front-line staff but it’s a good way to get it to management. I we I work in a hospital. they were working in what nursing and might be able to set up a program homes . particularly on social 183 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 183 Epilogu e practical and simple language that I sent back to the facilities—how to understand the issue. which meant that it went Practicea out to wider audience. including a GP ringing from another state saying. from a social worker in a hospital. I had phone calls from that.so I about a lot of things that you could do talked in practice and how I’d seen a support group work and were the elements of one that worked what for families. ‘I’ve got an old man who’s putting his wife in a nursing home. I know when I talk about [the study] that I can talk in ways that move people. ‘What can stuff... I know some people read it.I got a call last year..I feel that I should have can I done with it but it was always a matter of more time. And whenever I present always get practitioners saying... Geraldine Doogue talked about the study on ABC National Radio and there were articles in Sydney the Morning Herald and in General the magazine. what was it about and what not to do.saying.. social workers—saying you really to take care of the people who are have placing with dementia because it’s a very someone traumatic here are the ways you can help and thing and this is what the issues sometimes are.’ Social rang from interstate and talked to me workers about they could do. the conference and we want ‘We heard you at to develop a support program for families. How can we address do? this thing?’ I’ve targeted particular groups in the dementia area—hospitals. Caroline speaking Caroline: Thomas—Adopted children The feedback that I’ve been given has been very positive in terms of its impact.

’ Liz Kelly—Domestic matters Liz : violence Interestingly. I wonder many if is because of its simplicity. who do direct work with children.Our when whole approach was really about bringing together expertise and views from lots of different sources... you taught me this.some confidence. finished when people were asking questions about it— and was an interest in the outcomes—I think there it probably would have had more impact. and didn’t to convey in the book was how I felt manage we learned a lot from very good practitioners. I hope it will give them the inspira. At conferences people tell me it has changed the way they do things. ‘Really. Perhaps what I tion.. When I make presentations to practitioners I always feel a little bit uneasy handing them what they have given to me. about how to work with children in the process of adoption? were fortunate enough to include some Because we of tools in the book.. do you think practitioners will do with it? What What you hope that many practitioners would would have learned.I think that is partly to do with [the fact] it stayed in the Home Office for three years that after finished. who has been well. which meant that there was a it was timing Had it come out a year after it was issue. and [that] it’s quoted the place.. fundamentally. that we have been paid things less money for . depending on the audience. It has. in a this way.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 184 Qualitative practice research in Dorothy: Caroline: work practice. in our consultation processwe were developing the tools. doing policy-related research for many years—has told me pretty sure that it’s having more impact he’s than other studies he’s been involved in. Certainly my project director all over — one of them in particular. I think that the study—in terms of the document itself—has been less influential than other we have done. back to Sometimes I say. including social work practice itself. though.. how we present material about affected domestic 184 .

then their problem. The materials are used in teaching both practice and research in universities and other tertiary institutions. So there’s a part at the end positions that about institutional power. and too little of the interventionthat.that the bit that they found the most useful tioners is part of the process analysis where.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 185 Epilogu e violence. . formal power. Robyn Munford with successfully families Jackie: and Jackie Sanders—Working I think that [our] research has been one of a number of contributions to the growing interest at a policyin strengths-based work with families. I level knowthe research findings have been used in that negotiations with funders of family services to help explain why certain funding approaches won’t work in practice argue for increased recognition in and to funding for a ‘whole person’ or ‘whole contracts family’ to funding. about advocacy and outreach approaches and this study has informed that. the things we emphasise in talks. who have traditionally provided refuge. of power. .we conceptualise domestic violence in terms of forms of power. 185 . We have had lots of approach positive from practitioners relating to the feedback findings endorsement it gives to models of and the good practice. so it has very much that’s shifted our questions and challenges to those who work in the field—there is a whole process before women are ready or able to leave. talks personal gender power and people have said power and that find that really helpful in making sense they of processes in projects and interagency groups. .. And I have had feedback from practi. too much of the intervention focuses on leaving and the assumption that this is the option you offer women and if they don’t want it. which is again about leaving. both [in New Zealand] and in Australia. . and we have just written a review for the Home Office . the basic messages .. It has also been a challenge to focuses on the women’s services.

We have been aware that it is not enough just to do always the research and hope that people will find it. it never felt like I was taking the research down a predetermined pathway— rather 186 . From the tioners.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 186 Qualitative practice research in and we understand that there is interest in them in the United Kingdom as well.against government moves to restrict access strating to maternal and child health service! In all of the the intervening fifteen years. Dorothy Scott—Identification depression Dorothy: of post-partum In Chapter 1 I described how my research in the 1980s with Victorian maternal and child early health grew out of my clinical work with nurses women with post-partum psychiatric disorders. receptive We to establish and maintain a dialogue with a try wide of users of our work to keep on hearing what range it they need in terms of information and how is they it delivered to them and then to incorporate need that the work we into do. years I would find myself on the steps of the later Victorian with a megaphone in my hand. we really want to make a A contribution to family change—we can’t do that directly but can do it by providing good information to we practi. We have found people as we incrediblyto our work and that is really pleasing. demon. My initialof increasing the capacity of the maternal goal and health service to identify and refer women child to psychiatric services was broadened to explore the conditions under which the nurses might encompass maternal emotional and social well-being in their I would never have guessed that fifteen role. mostly women pushing prams.managers and policy-makers. outset we wanted to produce research that made sense to these people and that they could use. Parliament addressing a large crowd. so we are active in circulating our materials and trying very hard to make them accessible in as wide a range of settings possibly can. and this has driven our work and continues to drive it. s researchers.

This because the time was ripe for the sort of was study I ended up doing. 187 . and mental health journals to reach other professional educate them about the central audiences and role maternal and child health nurse. low level lems. These presentations were recorded on audio cassettes and videos so they could be used to reach nurses in rural I also published papers in child welfare areas. Leading maternal and child health were becoming concerned about postnurses natal depression and keen to broaden their role to encompass maternal emotional and social well-being. The were study described and conceptualised the clinical judgement maternal and child health nurses at of exemplary aime when the nurses themselves had a strong t desire to broaden their role and enhance the status of their profession. During of the the decade following the study there was a marked growth in professional and public awareness and concern about post-natal depression and this also increased the profile of the study. Interestingly. strongly supported the causal these studies signifi-of the factors which I had identified the cance nurses in assessing whether a mother was using depressed (infant temperament and sleeping and eating prob.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 187 Epilogu e [it] was dragging me along unfamiliar pathways. The in nursing education from a hospitalshift based apprenticeship model to a university-based profes-education some years earlier had helped sional to create a new generational cohort of nurse practitioners who were more psychologically oriented and who eager to work in different ways. A paper I published in Australian the Journal of Advanced Nursing (1987b) was positively received and I was invited to give addresses at maternal and child health conferences and inservice courses where I was introduced as education an ‘honorary maternal and child health nurse’. [Mine] was one of the first studies on maternal depression and most [of those] that followed were large quantitative studies. and a umber of situational stressors such as moving n house.poor maternal family background. of support from partner and/or her own mother.

our inside stories demonstrate that qualitative research in the human services is. not an enditself but a means to an end. and this book has drawn upon the experience of some particularly committed and enthusiastic missionaries! into the swampy lowlands of research. and supports implications a roadened role for maternal and child health b nurses. Instead. Hence the campaign to protect the service and the mobilisation of mothers and their prams on the steps of Parliament.of the prevailing tivity climate. and in joining a struggle have for justice. the effects of not the research may not register until some time after it has been completed and the impact will be largely determined by the recep. financial worries). to use The journey the phrase of Donald Schon with which we began the book. The cumulative weight of now large body of research carries the consistent for policy and practice. we hope will ensure that 188 . qualitative research is rarely a linear process and it often takes the researcher down unexpectedThe impact of the research may start very early on. Because of this it is not possible to from provide the aspiring qualitative researcher in the human services with a map and a set of instructions which. pathways. andpart and parcel of the research itself. as we have said before. I think that the moral of this story for methat researchers in the human services may well is find themselves involved in giving a voice to those who not been heard. The transformation which had occurred in the maternal and child health service in Victoria. Alternatively. the election of the Kennett Government. social Comment s As some of our inside stories illustrate. which introduced policies which limited access to the service and narrowed the focus to one of ‘paediatric surveillance’. that they get to their destination. In that sense it is indeed in ‘missionoriented research’. if followed detailed faithfully.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 188 Qualitative practice research in illness. is more to a mystery tour than a trip which progresses along a set akin route the start to the end. Above all. even in studies which be are thought of as ‘action research’. was threatened in the early 1990s by however.

189 . we have been able to have provide a and some idea of the demands and delights of the compass diverse which might be encountered on such terrain journeys.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 189 Epilogu e through reporting the experiences of qualitative researchers who undertaken prior expeditions.

vol. M. Brisbane. University of 1986 Chicago Press. 3rd 1998 edn. S. David Fulton. 1991 ‘Talking to young children about health: methods and Health Education Journal . L. C. S. ——1998 Tricks of the Trade . (ed. K. London. Becker.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 190 Reference s Adler...) Learning and Inclusion: The Cleves School Expe1999 rience. H. C. & Poole. Assignment and Thesis Writing . University of Chicago Press. 190 . Newbury Park. Chicago. vol. Becker. M. H. Williams. Becker. pp. 1990 CA. P. B. S. & Strauss. M. John Wiley & Sons. 42. Backett. 1. Boys in 1961 White: Student Culture in Medical School. & Adler. 8.. & Alexander. 1987 ‘Focus group interview: an underutilized research technique for improving theory and practice in health education’ Health Education Quarterly . McCall. 36. pp. Geer. Anderson. 1989 ‘Theatres and commu. no. vol. L. 411– Bashford. C. pp. CA. Chicago. Newbury Park. Townsley. 93– nities: 116. Chicago.three scenes’ Social Problems. P. Writing for Social Scientists . E. H. Membership Roles in Field Research. 1995 ‘Parallel text: making research accessible to people with intellectual Interdisabilities’ national Journal of Disability. vol.. E. L. 3. findings’ 34– 50. Debates on Evaluation . Development and Education . Sage. Alkin. C. A. Basch. & 48. H. 4. A. J. 211– 20. & Morris. P. no. 14. Hughes. L. University of Chicago Press. 1987 Sage. Alderson. pp. no.

The Anthropology of Experience . W. Hill. 1995 ‘Childhood and the sociological paradigms gaze: and Sociology . 415– 24. B. pp. Brown. London. E. University (eds) Press. H. M. N. W. Laybourn. Quality and Quantity in Social Research . 39. M. S. London. D. J. paradoxes’ 4. & Moseley. Have Learning Difficulties Who .. K. 729–37. Boston. Bryman. Booth.. & O’Brien. R. K. no. 13. pp. 3. J.) ocial Work Research and Evaluation . pp. C. Middle Child1998 hood: The Perspectives of Children and Parents . 26. Booth. 29. C. Chicago. 11. ——1994b ‘The use of depth interviewing with vulnerable subjects: lessons from a research study of parents with learning difficulties’ Social Science and Medicine . Social Policy Research Unit. pp. 155– Blalock. & Biklen. S. & Booth. Bruner. Jessica Kingsley. G. Allyn & An Bacon. Bruner E. A. An Introduction to Social Research . 1998 ‘Doing research with lonely British Journal of people’ Learning Difficulties .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 191 References Beresford. 1986 ‘Ethnography as narrative’ in V. 1. London . 1981 ‘Measurement’ in R. pp. J. London. Bostwick. Biklen. 2nd edn. 55– subjects’ ——1998 Growing Up With Parents69. Brenner. 3. J. no. vol. M. Borland. I’m Catholic”: qualitative methods in the study of people with severe handicaps Journal of the Association for Persons with Severe Hand’ icaps. ——1996 ‘Sounds of silence: narrative research with inarticulate Disability and Society . & D. vol. E. pp. Bogdan. vol. M. R. Academic Press. Brenner. Qualitative Research for Education: 1992 Introduction to Theory and Methods . J. Turner &M. A. 1997 Personal Accounts: Involving Disabled Children in Research. vol. Open University Press. 132– 4. Grinnell (ed. 191 . pp. Peacock & Sons. 139– of Illinois 55. M. no. Buckingham . Il. 1970 Englewood Cliffs. M. & Kyte. Brown. W. S Itasca. Prentice-Hall. & Brown.. no. NJ. T. Canter The Research (eds) and Approaches. 1988 Unwin Hyman. 1– Interview: Uses 8. Parenting Under Pressure: Mothers and 1994a Fathers with Learning Difficulties . Brannen. no. vol. & Canter. 1985 ‘Introduction’ in M. Routledge. 1988 ‘ “Are you retarded?” “No.Norwich. 62. 4. F.

S. Cambridge. London. 1986 ‘Current empirical research on evaluation Review of Educational Research . 57– tive 78. A. Darlington. & Leithwood. 48– 71. Canberra. R. D. 3. Cook. 3. Butler. 192 . Collaizzi. G. Avebury. & Bland. 17– illness’ 23. Dalton. Friendship and Peer Culture in the Early Years . Burnside. Y. Massachusetts 1985 Institute of Technology Press. Trauma and 1994 Social Work . Longman. NJ. The University tude of Queensland . Sydney. CA. Oxford University Press. ——1996 Moving On: Women’s Experiences of Childhood Sexual Abuse and Beyond . S. S. vol. P. North Ethics Parramatta. B. Wiley. Qualitative and Quantitative 1979 Methods in Evaluation Research . & Reichardt. Y. W. 1997 ‘Empty streets: current policy relating to longterm homelessness’ in P. Burgess.)and Quantitative Research. (eds) Markets and Communities: Remapping the Boundaries Volume 1 . MA. 1999 ‘Strategies for encouraging and maintaining hope among people living with serious mental Australian Social Work . I. pp. Men Who Manage . Conceptual Change in Childhood . 1985 Ablex. R. C. Darlington. 1978 ‘Psychological research as the phenomenologistR. Beverley Hills. utilization’ 331– 56. Valle & M. no. A. places and homelessness in Forti. T. F. pp. Aldershot. 1993 ‘The experience of childhood sexual abuse: perspectives of adult women who were sexually abused in childhood’ unpublished PhD Dissertation. Cousins. Children Speak: Children. Coleman. Sydney. Burnside 2000 Research Code of . AGPS. 64. Norwood. vol.Valley’ unpublished PhD Dissertation. Saunders & T. Brannen on their Mixing Methods: Qualita(ed. The Federation Press. In the Field: An Introduction to Field Research . H. pp. New 1959 York. M. no. K. Harlow. Social Policy Research Centre. The University of Queensland .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 192 Qualitative practice research in ——1992 ‘Quantitative and qualitative research: further reflections integration’ in J. ——2001 ‘Five star motels: spaces. 59– 71. & Williamson. A. J. 52. 1984 George Allen & Unwin. pp. Sage. pp. Eardley States. Commonwealth Department of Human Services and Health 1995 thical Research and Ethics Committee Review of Social and BehavE ioral Research Proposals. New York. Corsaro. King views it’ in Existential-phenomenolog(eds) ical Alternatives for Psychology . Carey.

70– 87. Routledge. W. culture and Human Studies . Qualitative Data Analysis: A User-Friendly Guide for Social 1993 Scientists . 1999 An Introduction to Discourse Analysis: Theory and Method . London. innocents’ pp. S. Martin Organizational Culture. in supervision’ press. CA. G. 33– 48. 53– 61. Y. Sage. London. Gold. 1990 ‘Creating social change: “mission”oriented research and entrepreneurship’ in J. Burgess Research: A Sourcebook and Field Manual . D. Booth J. M. Sage. CA. J. 1979 in evaluation research’ in T. 165– Families Aldine. 193 . S. Frost. 1986 ‘Kids. Fraser. G. I. & Leavitt. (eds) New York. 1985 ‘Ethical considerations in cultural research in organizations’ P. Donaldson. 106.) Field George Allen & Unwin. Computer Analysis and Qualitative 1998 Research. & Strauss. L. N. pp. London. Sydney. Tracy & C. Reaching High Risk . Moore. C. Goode. L. Routledge Social & Kegan Paul. A. J. Problems and Practice . Beverly Hills. M. 83– vol. 2002 ‘Child welfare workers’ of theory in working with physical child abuse: use implications for professional Families in Society . pp. M. 1982 ‘The participant observer as a human being: observations on the personal aspects of fieldwork’ in R. Finch. L. Louis. Knowing Children: Participant 1988 Observation with Minors .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 193 References Darlington. Grbich. Fielding. Lundberg & J. Sage. & Sandstrom. pp. 36. B. Routledge. 253–69. K. & Lee. Deetz. Sage. pp. 217– Forces vol. Bell & H. P. New York. Roberts (eds) Researching: Politics. C. D. Fontana/Collins. G. J. & Peile. J. pp. (ed.. Kinney. A. Children’s Minds . M. 1984 ‘ ‘‘It’s great to have someone to talk to”: the ethics and politics of interviewing women’ in C. G. R. 23. Reichardt Qual(eds) and Quantitative Methods in Evaluation Research itative . Whittaker. Beverly (eds) Hills. Dey. C. 1967 Aldine. 1999 Allen & Unwin. A. Glaser. Gans. L. The Discovery of Grounded Theory . observation’ pp. Newbury Park. J. London. H. Cook & C. Fine. M.‘Qualitative methods: a needed perspective Filstead. 1978 Glasgow. Osmond. R. 78. London. 1958 ‘Roles in sociological field Social . Gee. S. 9. Qualitative Research in Health: An Introduction .

Sage. C. Routledge. ‘Children as research subjects: a risky Children and Society . Lincoln Y Handbook of Qualitative Research . Riessman Qualitative Studies in Social Work Research . pp. 1995 ‘Using focus groups to discuss sensitive topics with Evaluation Review . F. (eds) Sage. G. NJ. no. 10. E.. & Wilsdon. e Harding. Y. London. Kelley. CA. J. Holstein. M. D. Hoppe. CA. 10. M. 1996 ‘Engaging with primary.. Thousand Oaks. Denzin &. vol. Guba. & Atkinson. 262– 72. 2. ——1995 The Active .. 1997 Health Education Board for Scotland. 1989 Sage. 129– considerations’ 44. P. vol. Hammersley. Hendricks. Caracelli. 1990 ‘Do’s and don’ts for effective recommendations Evaluation Practice.children about their emotions and well-being: aged methodological Children and Society .. Beverly Hills. 2nd edn. R. pp. 1994 ‘Reflections on a journey: a research story’ in C.. Edinburgh. pp. & Mayall. 194 . R. 61– Nursing 75. & Barlow. Method and Practice . M. & Papagiannis. no. J. pp. ——1997 Interviewinterviewing’CA. G. S. Thousand Oaks. D. pp. 1988 ‘Retaining qualitative validity while gaining quantitative reliability and validity: development of the Transition to Parenthood Concerns Advances in scale’ Science . Laybourn. A. pp. 1976 Pergamon Press. W. vol. & Lincoln. NY. Health Education Board for Scotland Messages from Children. & Atwood. & Graham. Elmsford. 79– 104. F. A.) Research: Theory. pp. K. ethnomethodology and interpretive practice’ in N. London.) Revolution . 1. Guba. J. 1. Ithaca. S. 3. 11. J. 1985 ‘The context of emergent paradigm research’ in Y. M. 121– ’ 5. D. A. Hill. Single Case Experimental Designs . A.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 194 Qualitative practice research in Greene. P. vol. 113– Hood. & Gubrium. pp. Newbury Park. C. vol. Morrison. 11. 2. M. A. no.) Sage. M. no. S.. 11. Lincoln Organizational Theory and Inquiry: The Paradigm (ed. Gillmore. no. (ed. Cornell 1986 Feminism University Press. M. vol. CA. 19. Ethnography: Principles in 1995 Practic . Silverman ‘Active in Qualitative (ed. K. & Borland. Imle. Sage. 2. B. E. pp. CA. M. S. Fourth Generation Evaluation . M. 169– 89. S. Hersen. J. 1994 ‘Phenomenology. 102– children’ 14. Sage. J. V. Hyde. pp. 117– enterprise’ 28. M. The Science Question in . Thousand Oaks. Wells. 255–74. 1989 ‘Toward a conceptual framework for mixed-method evaluation designs’ Educational Evaluation and Policy Analysis . no. 199629. E.

and educational research with children’ in M. Kinds and Cognitive Development . Cambridge. 4. Thousand Oaks. F. Denzin & Y. no. Little. Shotland (eds) Multiple Methods in Program Evaluation: New Directions for Program Evaluation 35 . L. Langness. CA. Analyzing Social Settings: A Guide 1995 Qualitative Observation and Analysis to . H. New 1975 F. 24. N. Kleinman. Mark & R. 145– tive Lewins. Ethics and &. get subjects to open up and feel good about telling the Applied truth’ Marketing Research . pp. Kidder. Hohn. Chandler & Sharp. Brannen Mixing Methods: Qualita(ed. pp. & Kime. vol. Belmont. F. Lives: An Anthropological Approach 1981 Biography . M. Healing and the Human Condition . Home Office. L. no. vol. 195 . when methods: stories converge’ in M. E. 1. 1. Keil. L. pp. Sage. Novato. C. pp. Krahn. vol. 1994 ‘Scientific issues in psycho. 204– Clinical 13. & Frank. 6– 11. 4. San Francisco. M. Lofland. 1989 ‘Focus groups on sensitive topics: how to 74. Brown.. C.)and Quantitative Research . Johnson. Grodin social L Glantz Children as Research Subjects: Science. & Keith-Spiegel. Canberra. 1993 Australian National University. J. 1994 ‘What comes (just) after post? The case of 67. 68. Faculty of Arts. H. & Scott. New Women York. Jossey-Bass. H. no. L. P. 563– Mariampolski. 1995 ‘Incorporating approaches into clinical child psychology Journal of qualitative research’Child Psychology . 57– 75. Wadsworth. 16. 3rd edn. D. G. M. Kelly. pp. ethnography’ in N. M. Lincoln Handbook of Qualitative (eds) Research.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 195 References Johnson. Penguin. 433– Marcus. pp. L. New York. Project H. New 1988 York. ment . A. 4th edn. Doing Field Research . The Free Press. Oxford University Press. 1992 ‘Multiple methods in the study of CA. Concepts. pp. 1967 Boston. L. Writing a Thesis . 47– La w 80. 1987 ‘Qualitative and quantitative London. 27– Social Research. 1988 ‘The least-adult role in studying Journal children’ of Contemporary Ethnography . to Laurie. G. York. The Illness Narratives: Suffering. Liebow. Mandell. J. no. 1999 Domestic Violence Matters: An Evaluation of a DevelopMA. Basic Books. & Lofland. ——1993 Tell Them Who I Am: The Lives of Homeless . Koocher. Aldershot. 2. K. (eds). household resource allocation’ in J. pp. 48. G. & Fine. Avebury. L. 1989 Massachusetts Institute of Technology Press. G. 29. Tally’s Corner . 1997 ‘Confessional tales: an exploration of the self and other in two Australian Journal of ethnographies’ vol.

Longman Cheshire. B. R. Mason. Sage. vol. K. S. 1997 ‘The “inside” and the “outside”: realities in interviews’ in D. M. Matocha. Hood. 8. 18 (Supplement 1). 1990 ‘PerformanceSocial Problems. pp. no. CA. K. M. Thousand Oaks. vol. Sage. Daly & G. 196 . 25– Social Work 31. (eds) Multiple Methods in 1987 Program Evaluation: New Directions for Program Evaluation 35 . E. L. McCall. F. students’ e 106. J. L. Minichiello. Qualitative Data Analysis: An 1994 Expanded Sourcebook. pp. pp. F. pp. R. & Shotland. In-depth 1990 Interviewing: Researching People . Silverman finding Qualitative (ed. & Glassner. B. 1994 ‘Research policy and child altreatment: developing the scientific foundation for m effective of children’ protection Child Abuse and Neglect . 1999 ‘Children and childhood’ in S. Miller (eds) Doing Qualitative Research: Research Methods for W Primary . Melton. 1. 3– Care 28. 11– vol. Sage. M. London. Mayall. pp. A. San Francisco. pp. M. vol. 1996 Qualitative Researching . Sage. P. funded nonprofits and accountability’ Nonprofit Management and Leadership . Handel Qualitative (eds) Methods in Family Research . L. 1992 ‘Primary care research: a multimethods typology and qualitative road map’ in B. e 24. Buckingham. organisations’ pp. & Meshejian. G. Crabtree & . 117– science’ 32. Mayall & S. pp. no. London. D. 1992 ‘Sources of stress and coping responses of high school Adolescenc . McDonald. 461–74.CA. B. & Huberman. 10– dic . Miles. no. 3. 4. no. 1992 ‘Case study interviews: caring for persons with AIDS’ in J.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 196 Qualitative practice research in Mark. 1999 ‘Damned if we do and damned if we don’t: exposing ethical tensions in field Australian research’ . & Allison. J. and Crabtree. pp. B. D. McAuliffe. R. W. 1.. Newbury Park. Gilgun. Mates. Becker. Melbourne.) Research: Theory. pp. Miller. vol. ——1999 ‘Internal control and accountability in nonprofit human service Australian Journal of Public Administration . 22. Aroni. vol. 37. M. 52. Jossey-Bass. 1996 ‘Institutionalised organisations? A study of nonprofit human service organisations’ unpublished PhD Dissertation. 1– 28. M. B. C. 66– 84. ——1997 ‘Government. & Alexander. K. Oliver Critical Issues in Social Research: Power and Preju(eds) Open University Press. no. vol. & Flood.. 27. 51– 64.. H. A. 58. B. The University of Queensland. V. 99– 112. CA. Method and Practice . Newbury Park. 1. Sage. Miller. 2nd edn. & Coleman. M. Timewell.

Denzin & Y. Canberra. pp. Longman. J. 47– Disability 57. 220– Sage. NZ. pp. Barnardos New Zealand. no. CA. C. Livingstone. E. Longman. & Spoonley.. M.) Education. R. J. ——2000a ‘Ethical issues in qualitative research with families’ in M. Buckingham. ——2000b ‘Through the eyes of the actors: three challenges for qualitative research with International Journal of Qualitative families’ Health Research . Tollich Ethical Issues for Qualitative Research . 1994 ‘Designing funded qualitative research’ in N. Palmerston North.. K. 197 . ofAustralian Government Publishing Service. A. A. Robinson. 1981 ‘Interviewing women: a contradiction in terms’ in H. K. Sanders. 30– Kegan Paul.. M. S. Research and Practic . pp. Techniques.J. Sage. Violence and Male Power: Feminist Activism. 1999 National Statement on Ethical Conduct in Research Involving Human . J. Open University Press. Analysis . Family Australian Government Publishing Service. & Sanders.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 197 References ——1995 In-depth Interviewing: Principles. 841– 53. Munford. Commonwealth Department of Health and s Services. Hester. R. Oakley. C. Lincoln Handbook of Qualitative Research . Henare. Morse.. Sanders. and Medical Research Council National Health Ethics 1995 Qualitative Methods in Health Research: An Information Paper for Aspects of Institutional Ethics Committees . Moran-Ellis. Newbury Park. Dunmore 1999 Press. 2nd edn. 35. Mulder. P. Tisdall. CA. Barnardos New Zealand. Pearson (ed. J. 10. Roberts Doing Feminist Research . Auckland. vol. Minkes. (eds) Thousand Oaks. Murphy. 61. L. M. Routledge & (ed. 1. Canberra. You Can Write: A Do-it-yourself Manual . D... 1996 Working Successfully with Families: Stage 1 .) London. 1985 Longman Cheshire. Radford (eds) Women. & Spoonley. pp.. 1996 ‘Close to home: the experience of researching child sexual abuse’ in M. Supporting Families . Jack.. A. Wellington. P. Morgan. 9. 2nd 1997 edn. J. vol. 176– e 87. 6. pp. pp. Wellington. & Weston. Munford. Kelly & J. Melbourne. Melbourne. Focus Groups as Qualitative Research . 99– 112. no. Tisdall. J. R. Munford. L. Commonwealth Department Health and Family Services.. 1994 ‘Consulting the children: interviews with children using residential respite care services’ and Society . 1998 Working Successfully with Families: Stage 2 .

’ Review 20. Hood. vol. & Gardner. & Emler. 4. D.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 198 Qualitative practice research in ——‘People’s ways of knowing: gender and methodology’ in S. Pe-Pua. Padgett. 6. 41– tives 60. Bureau of Immigration. Owen. pp. 136– 55. 1. 1998 ‘Writing for Australian publication’ of Public Health—Interchange . 2nd edn. 57– Reicher. Sage. 53. no. no. S. J. s CA. 1981 ‘Knowledge for Social Service practice’ . no.) Getting into Life . 90– 9. Qualitative Methods in Social Work: Challenges and 1998 Reward . M. H. Allen & Unwin. 42. E. Oliver Critical Issues in Social (eds) Research: Power and Prejudice . 20. no. ——1990 Qualitative Evaluation and Research Methods . P. pp. 2. Multicultural and Population Research. vol. CA. Polkinghorne. Patton. 1989 ‘Phenomenological research methods’ in R. 1988 ‘The future of Evaluation Practice . J. M.. Australian Government Publishing Service. S. Canberra. Paget. L. Rein. M. M. Thousand Oaks. vol. pp. vol. Plenum Press. R. 1. 41. 198 . 22– world’ ——1988 ‘Secrecy and status: the social 7. B. 1990 ‘Performing the Journal of Contemporary text’ Ethnography . S. New York. Women’s situation”?’International Forum . Thousand Oaks. Sage. Other Woman: Contemporary Single 1985 Women in Affairs with Married Men . pp. 13– H. D. New York. Valle & S. M. The New 92. Halling Existential–Phenomenological Perspec(eds) in Psychology . ——1986 ‘Another Psychology Today . Buckingham. no. Pirkis. & Rogers. A. London. pp. 1986 63. Newbury Park. 209– . 2. 71– Journal 6. Patton. pp. M. 12. vol. 579– Studies Richardson. Program Evaluation: Forms and 1999 Approaches . 1989 ‘Interviewing—an “unnatural J. CA. Methuen. pp. vol. Sydney. & White. 3rd edn. evaluation’pp. Open University Press. Conversation Analysis: The Study of Talk-in-interaction . Tex t CA. 12. J. 1– Revie w Ribbens. pp. G. Sage. 154– 70. 55. 4. no. Psathas. construction of forbidden relationships American Sociological vol. 1996 We’re Just Like Other Kids: Street-frequenting Youth of Non-English-speaking Background . 3. K. pp. ‘Managing reputations in adolescence: of delinquent and non-delinquent identities’ the pursuit in Beloff (ed. Free Press. pp. 1989 ‘Some ethical implications of qualitative research’ Research in Nursing and Health . Q. no. Thousand Oaks. 19. N. 1995 Sage. 2. Ramos. Utilization-focused Evaluation: The New Century 1997 . Mayall & S.

T. Action London. pp. Etzioni A Sociological Reader on Complex Organizations . Newbury Park. Writing by Degrees: A Practical Guide to 1996 Writing Theses and Research Papers . R. 1998 The Ethnograph v5. 6. vol. Schon. (ed. vol. 558– Holt. CA.. vol. Working Success1999 fully with Families: Stage 3 . Narrative Analysis . C. Scott. 42– Quarterly Rosenhan. 5. Sage. Scott. ——1995 ‘Child protection assessment: an ecological perspective’ unpublished PhD Dissertation. L. 359– 76. CookGumperz. vol. D. abuse’ 16. Temple Smith. 1969 ‘Field methods in the study of organizations’ in A.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 199 References ——1990 Writing Strategies: Reaching Diverse Audiences . University of Melbourne . pp. pp. R. Corsaro & J. The Reflective Practitioner: How Professionals Think 1983 in . 1. CA. D. Social Work . New York. C. no. no. Sage.) Rinehart & Winston Inc. 564– 8. 39– Service 51. Berlin. & Richards-Ward. places’ 179.W. Barnardos New Zealand. no. CA . Scienc . 35. Wellington. A. 28–37. 199 . & Laing. 250– e vol. pp. K. Auckland . 345–58. vol. 76. D. ——1992 ‘Early identification of maternal depression as a strategy in the prevention of child Child Abuse and Neglect . Rountree. A. School of Social Work. Sanders. Mouton de Gruyter. M. W. T. of measuring social competence through interviewing’ in J. University of Melbourne. 1973 ‘On being sane in insane L. K. 1986 ‘Adult elicited child behavior: the paradox Siegert. ——1987b ‘Maternal and child health nurse role in postpartum depression’ Australian Journal of Advanced Nursing . pp. Rist. Thousand Oaks. 8. 1977 ‘On the relations among educational research paradigms: from disdain to Anthropology and Education detente’ . Streeck Children’s Worlds and (eds) Children’s Language . 1.0 . Newbury 1993 Park. Munford. ——1990 ‘Practice wisdom: the neglected source of practice research’. J. pp. 1987a ‘Primary intervention in maternal depression’ unpublished Master of Social Work Thesis. pp. Riessman. 8.9. Addison Wesley Longman. Seidel. R. Qualis Research Associates. 63. J. ——1989 ‘Meaning construction and social work Social practice’ Review . pp..

M.. W. J. Sage. W. CA. Breaking Out: Feminist Consciousness and 1983 Feminist Research . A. S. R. A. Macmillan. J. Taylor. 1987 Cambridge University Press. 200 . Bogdan. vol. 95–115. pp. 25– ’ 37. R. 1. 1993Implications of developmental research for ‘ interviewing children Child Abuse and Neglect . Spradley. New York. 2nd edn.. Tammivaara. 11. J. R. & White. pp. J. A. N. pp. A. no. & Heshusius. N. 17. Curtis. L. 1. Thousand Oaks. 218– Quarterly . 1986 ‘On eliciting information: dialogues with child Anthropology and Education informants’ vol. H. J. Beckford. J.. Thousand Oaks. pp. Adopted 1999 Children Speaking . C. 1997 ‘Family friendliness of community-based services for children and adolescents with emotional and behavioral disorders and their families: an observational Journal of study’ Emotional and Behavioral Disorders . Basics of Qualitative Research: 1998 Grounded Theory Procedures and Techniques . L. C. G. John Wiley & Son. S. Thomas. Participant Observation . & Saywitz. Research. New York. N. Wechsler. Bussey. CA. no. R. S. Cambridge. & Wise. vol. Stanley. K. 1986 ‘Closing down the conversa. Lowe. & Corbin. Strauss. Qualitative Research: Analysis Types and Software Tools . Women’s ethnography?’ Studies International Forum . Tesch. J. 3rd edn. pp. 82– 92. inquirers’ 1. & Murch. 2. S. no. pp. Steward. L. Basics of Qualitative Research: Tech1990 niques and Procedures for Developing Grounded Theory . Holt.the end of the quantitative–qualitative debate tion: among educational Educational Researcher. Strunk. vol. & Corbin. S. E. & Enright. P.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 200 Qualitative practice research in Singh. Strauss. V. Routledge & Kegan Paul. & Cohen. Smith. Ellis. Thomas. K. 4th 2000 edn. 15. Qualitative Analysis for Social Scientists . J. British Agencies for Adoption and Fostering.. Goodman. 1978 ‘Generating innovation in social work: the paradigm of developmental Journal of Social Service research’ 2. K. M. New York. 21– 7... The Elements of Style . 1988 ‘Can there be a feminist Stacey. J. Sage. 5. E. New York. vol. London. & Introduction to Qualitative Research 1998 Methods . 4–12. no. D. 38.. 1990 The Falmer Press. B. Strauss. Rinehart 1980 & Winston. J. London . L. 17.

United Nations Centre for Human Rights Convention on the 1989 Rights of the Child . 134– context’ Trinder. & Simons. pp.The tation. ——1997b ‘Resources. pp. pp. P. Weiss. Intending to ask lineal. K. 4. Seen and Heard: Involving Disabled Children and Young 1997 People in Research and Development Projects . physically abused children: relationships with the childrearing Journal of Family Psychology . Sydney. Trickett. 7. Walker Applied Qualitative Research . Chicago. vol. The Contact Project 2001 First Year: Interim Report for the Advisory Committee . J. University of East Anglia. 201 . 1– Process vol. 121– 30. Buckingham. pp. Beek. Tomm. C. Wade. K. J. vol. Eardley States.) Aldershot. ——1997a ‘Family advocacy roles and highly dependent residents in nursing Australian Journal on Ageing . vol. Experiencing Special Education: What 1993 Young People with Special Education Needs Can Tell Us . of the Field: On Writing Ethnography. 1990 ‘Evaluation for decisions: Is anybody there? Does anybody care?’ in M.. Ward. (ed. & Connolly. Hants. Saunders & T. strategic or reflexive Family . L. 1994 ‘Long term marriage and long term Australian care’ Journal on Ageing . 16. 13. no. 3– 26. H. B. Newbury Park. C. 1993 ‘Maladaptive development of school-aged. pp. Geneva. 47. 26. 128– Learning 31. Alkin Debates on Evaluation . no. CA. Gower.) Sage. 4. no. (ed. & Moore. circular. Ward. University of Queensland. L. Open University Press. (eds) Markets and Communities: Remapping the Boundaries Volume 2 . pp. rights and relationships: family participation in residential care’ in P. United Nations. M. 1998 ‘Practising partnership: involving people with learning difficulties in British Journal of research’ Difficulties . 27. 1985 ‘An introduction to applied qualitative research’ in R. Tales 1988 University of Chicago Press. 15. Norwich. C. questions?’ 1. Joseph Rowntree Foundation.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 201 References Tilse. M. 171– 84. homes’ 1. K. vol. goodbye: the experience of placing and visiting a long term partner in a nursing home’ unpublished PhD Disser. Social Policy Research Centre. no. York. 20–3. R. L. van Maanen. Walker. 1988 ‘Interventive interviewing: part III. 172– ——1996 ‘The long 4. pp. pp.

86– 106. Sage. F. & Bucuvalas. F. Sydney. pp. CA. pp. Van Zuuren. J. New York. M. Hale & Iremonger. London. J. & van Zuuren. J. 9. Berwyn. 33. The Falmer Press. J.) . 202 . Ashgate. Wertz & B. pp. Mook Advances in Qualitative Psychology: Themes and Variations (eds) . pp. Aldershot.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 202 Qualitative practice research in Weiss. 1987 ‘Qualitative research: educational considerations’ in F. Yuen-Tsang. Woods. 3– 23. J. Mason Child Welfare Policy: Critical Australian Perspec(ed. Wilkinson. J. G. K. Social Science Research and 1980 Decision-Making . 1993 ‘Children’s rights: debates and dilemmas’ in J. Wertz. Columbia University Press. F. no. 453– vol. PA. 143– tives 56. H. G. M. 1985 ‘New songs played skilfully: creativity and technique in writing up qualitative research’ in R. C. 1988 ‘On the nature of professional communication: publication for Social Work . F. W. P.) in Educational Research: Qualitative Methods Issues . 1990 Writing Up Qualitative Research . H. & Hopps. Wolcott. practitioners’5. F. Williams. A. Swets North America. L. Towards a Chinese Conception of Social 1997 Support: A Study on the Social Support Networks of Chinese Working Mothers in Beijing . Burgess (ed. Newbury Park.

28. recruiting impac t the childre 109–10 n agency permission 29 anthropology 2. Verna 107 behaviourism 5. 106. 9. in a 7. group discussions/focus groups 99. 30. 151–2. 24 Adopted Children Speaking xvii . . analysis 183–4 . of staff s articulateness 3251–2. 92. 111–12. of discussion 61. 104–5. data 155–6 . 111– study 12 Booth. ethical/political 35–9 issues children: adopted 27–8. 103. 180. involvement in research 101– 2. 9. 111. as participants 26– 7. 74 anxiety: of client 29. research study xvii. 93. 103 authorial voice 3 Beckford. 36–7 recording 59. 39. to services 25 accountability 23. of new mother 15 . 103. 146– team 7 Coleman. 107–8. interviewing 95– 110–11 . 173. 6 beneficence 21 book publishing 164–5 Tim xi. explaining research to 27– 8. 72. 2 case analysis.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 203 Inde x Page numbers interviews in italics refer to case studies 29 child 11–12. Booth. 111–12. 181 academic 38 freedomto research setting 32– access: 3. 76. 98 ‘Child protection assessment: anecological perspective’ xvii. 171–2. 118. 111–12. 81. Anne xii. co-observers 76 coding 144– 149–50 . negative 52 203 . participant observation with 99–101. 171– 8–9. 118. use of drawings 108–10 97 client permission 29 clinical audit 23– 4 clinical hypotheses 4 clinical judgement 4 closure. 107–8. research with 76. 104–5. . 173. 39. 69. 106. 118. 94–102. researchxvii. abuse/neglectinvestigations 68. 80. Wendy xi. 71 63. 8–9. research study xvii.

disaggregating data 29. document analysis 136–7. and deception 77. 53. 127. in public in staff evaluation 46 32. 181–2 . 70. guidelines 24. 25. findings 179. giving meaning to 3. research 173–5 . 26 empiricism 5 enticement. 22. 13. study choosing observatio 81–2. 30. 26 ethics 21–47. transcribing 143– 148–9 4. secondary observations 29–30. places and homelessness in Fortitude Valley’ xvii. 162–4 ‘Five star motels: spaces. data analysis 142–57. 43 feminis 132. 61. s 151. adapting to suit participants 92–118. 73. when to 83–4 stop debriefing: interviewees 28. within organisations 31–4. of findings 17. document 151 . contradictory 126–7. use 30. 34. people with learning disabilityto children 27– 116–17 . 81. review see committee board s s evaluation research 32. 25. 28. background . s background to 10–11 . 49. 18. 9 evidence-based practice 5 experiences. 118. 180 204 . of n interest/ concern 54. computer 8 programs see also 147 NUD*IS . difficulties limits 26. researcher 61 deception. familiarising 144. readings timing/recordin 84–5 g focus groups 64. 59 follow-up interviews 70–2 68. 44 g 30. 2. informal interview s insider/outsider dichotomy 39–41 . impact 82–3 . 23–4. n communicating with population 85–6 . coding 144–7. to 8. research 113–14 . 103 family support services 13–14. 9. generalisability. 62. approach deriving categories 7. 62. in observation 77 deinstitutionalisation 32–3 dementi 16 a developmental research 180– 1 diaries 50 distress. dealing with 56. Yvonne xii. 11. undue 25 ethical research. 148. analysing observation 152–3 . 68. structuring 159–60. deductive 154–5 . 31. 120. 63. fundin 14. interview skills 57. 32. 38–9. flexible 105–6. 29–30. up mixed impact 167–70 methods duty of care 21. 36. 178– 158. writing 128–9 . 146 159–60. 111 72. 57. ownership domain 30. children with 99. 53. question design recording 63. mixed methods 49. document-based research 29 Domestic Violence Matters xvii 127– 169 . 161–3. 119– 41. The Ethnograph T confidentiality 24. study 184–5 . research study xvii. 162. 56. 53. data collection 48. committees 22– 3. 21. professional 21. research m 5 70–1 traditions . 30 correspondence 2 counselling skills 57 Darlington. of questionnaires transcribing 132. 165. 39. 75. 80. 60. 151 2. 161–2. when 144 consen see informed t consent consumer opinion 2 contracts 26. computer programs 147. 33 communication: with homeless populatio 85–6. 61.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 204 Qualitative practice research in Commonwealth Department of Human Services and Health 26.

style 75. human experimentation 22 human services 1. 61. with 73. 31. style 75. informal unstructured 68–9 56. to longstudy term outcome 186–8 s in-depth interviewing 2. 167. recording 59–60. 65–6. 37. 12. meaning-making process 49– 50. interview process 50. 127. background14–16. 12. 95–6. 49– 48. 112–13 . 93. therapeutic 27 medical sociology 2 205 . Catherine xiii. 117. initial contact 55– 6. 37. 133. 47 ‘Identification of postpartum depression by child health nurses’ xviii. 12. 72 57. of colleagues management 32. 21 literature search 18– 19. people with: flexible data collection 105– 6. 133. retrospective elements 50. of parents 35–6. research study xvii. 26. 126. 111– with 76. Liz xii–xiii. people with intellectual 103– disability 112–17 . 69–70 . 112–17 . 153 management. 6. . research in 21–2. 154–5 . 127. 118 interviewees. flexibility 105– 110–11 . organisations 31.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 205 Inde x ‘getting in/getting on/getting out’ 31–2. ending 60– children 95– 9. and use of public space 10–11 . gaining consent of32 market research 2. interviewing 103–5. therapeutic 72. selection 55. deductive approach 134–5 . research study xvii. follow-up 68. 50 media coverage 30 medical anthropology 2 medical research 27. 3. silences strengths/weaknesses 48– 51. 70–2 82–3 . 135–6. 32. 167. 18. Kelley 31–2. 10 4 interviewin see ing interviewin depth g interviewing 68–70 . of staff innovation 32 transplant 180 insider/outsider dichotomy 39–41. timing indigenous 40–1 Australians informed consent 24. to stud 12–13 . 123. 57 intuition 4 Johnson. 179. negotiating 21. 31. 28–9. surve use of y mixed method 133–4 s intellectual disability. entry to Beijing 42–3 community departments government 24 group discussions. focused 56–7. establishing rapport 54– 5. 7. 39. interview y analysis. research 92. 33. terminology 19 McDonald. finding/selecting participants 51–4. 43 ‘Institutionalised organisations?’ background xvii. 23. 18 1 knowledge 1–2. interviewing 68–70 . 24. interview guide 57. 82–3 9. 25–8. reflectiv 69 e interviews: client-centred 8. 32–3 Kelly. 24 hermeneutics 7 homelessness 39. of 37–8. 61 Massey University 14 (NZ) meaning construction 10. 12. 1. 58– 65. intrusiveness 24. 124–5. with children 99 group interaction 62 Growing up with Parents who have Learning Difficulties 11 2 health research 23. 102–6. 9. 51– 66. debriefing 28 interviewers 57. 108–11 . 133. 48– 28. 5.

111 . 118. 70–1. of events 51. with interviews 76. data children 145–6. generalisability of findings 17. 5. 111– 12. strengths/limitations 75–6. 123. levels of participation 29. methods 2. 93. 43. 43 non-profit 12–13 sector note90–1 taking NUD*IST 155–6 nursing home 16–17 . parental permission 26– 35–6. perspective 50. 118. paternalism. timing unstructure 64–5 d observer 75–6. intrusiveness 28. with children 99– 101 participants: children 26– 36–7. 138–40 .Qualitative Research – TEXT 22/11/2001 3:26 PM Page 206 Qualitative practice minors 26–7. with 61. with intellectual disability 8–9 participant observation. 156– analysis 7. 73. 92. 102. 181. reporting 29. analysis 151. sample size 17– 36. getting on with 32–3. Robyn xiii. recording 75. 29. 55. writing 112–13 . through one-way screen 37. naive 76 oral history 2 organisational theory 2rganisations o 33. 76alternative forms 165– m 173–5 7. 50. g background to 9. research study xvii. privacy 29. mixed methods 119– 129–33 . 7. 6–8. 71. level of participation 78. 8–9. study individualised approach 114–15 . 8. in human services 2– 18. placement 86 research in observation 2. up 172–3 parents. 80. 89–90 . 8. research issues 21– 47 paraphrasin 38 g parental permission 26– 35–6 . 91. 101 Parenting Under Pressure: Mothers and Fathers with Learning Difficulties xvii. with diminished 26. 74– 91. 56 . with 108–9 . 7. 36. reporting to 173–5 . family support services 13. approaches to interviewin 115–17 . 91. 36 justified performance anxiety 32 photographs. ways of protecting 30. capacity 102–6. 24. 91. 77. 81–2. of stakeholders 22 practice 4. back 60. 138 pitfalls Munford. 4. 79– 84–5 . see also 18 research quality assurance 23– 4 206 . safeguarding participants 24. data 3. 102– 138–40 . 51. 181 National Health and Medical Research Council 23 negative case analysis 52 neo-institutional theory 12. 114. safeguarding 21. 166– 7 proposal 23 psychology 5 qualitative research 1–2. 84–5. 37. 78. vulnerability 24. 41. 37 7. main purposes 121. ongoing contact 60. use in interview 116–17 politics: of research 22. 86–7. 135–7 New Zealand. co-observers combined 76. within organisations 24. 171–2. deception 77. 95–6. 117. 31– 4 positivism 1 post-partum 14–16 depression power: researcher/participant51. imbalance 5. 43. 90–1. 49. 13. mixed approach 125–6. 71 51. 152–3. roles 77– 87–9 . 133–4. 25. 3. finding/selecting 51–4. associated words 4– 5 practice 16 wisdom presentation 158– passi . 13.

58. leading reflexive 58 see also research questions rapport 54– 66 5. analysis 182–3 . 188–9. 46 service delivery 21 service providers 29. lineal 58. 14. 170. 130–1. clarificatory 69 . 146. 47 role clarity 82. sample size 17–18. risk 37. within organisations 31– 4 Research Code of Ethics 24 research see participants participant s research questions 3. 126–7. coding 146– team 7 teamwor 43–5 k terminology 19 Ethnograph 148 . in human services 21–2. samplin 134–5 . 165–6. generating 4–5. transcribing 87–9. gaining consent of 32 staff evaluation 46 32. n forms of different 170–1 . 175. Jackie xiii–xiv. 34. as researcher 25 social sciences 5– 6 sociology 2. 13. 24. 59– 65 60. 61. theoretical g 67–8 52. background xviii. as enlightenment 159. and staff evaluation 46. 58. 16 . interviews:impact analysing 149–50 . 181 Scott. as theatre 32. as 8. 46. study interviewing 68–70. perspective 20. y choosing observatio 86–7. style participants’ 57 experience ‘The long goodbye: the experience and visiting longof placing term partners in a nursing home’ 138. 113–15 s 54. client debriefing 43. 47. 115–16. centrality of 18. 28. stories 50–1. 180 secondary data analysis 29– 30 self-disclosure 61 self-reflection 18. 32. analysing recordin 90–1. 44. research study xvii. as 173–5 feedback recommendations 177–9. 43. 23. 18. g 19. 22. observation role observations: 151. 154 ‘The experience of childhoodabuse’ sexual xvii. role clarity self-reflection 87–9. 27. 150 . capacity for harm words 31. methods 6–8 questionnaires 49. source of interference 32. timing g of 89–90 . 8. 87–9 78. see 73 researcher-practitioner split 5– 6 of harm 27. 158– 60 reports. to stud 16–17 . 8. circular 58. research studies xvii–xviii. gaining of consent 32. 74 staff. 68. 13. generalisability 17. impact 177. conceptualisin 12. 18. 59 subjectivit 131–2 y tape recording 55. associated 4. see also 63. reporting 29. 18. Dorothy xiv. in public domain 30 research 4. 43. 148–9 . The 153. background to 11–12 . Sanders. 180 referral 71–2 source reflexivit see selfy reflection relationship 46. 132–3 questions 57–8. 52– 130 4. readings. 86. 4. 181. offs transferability 180. trade36. 51. question s research team 43–5 researcher 71–2. 180. 20. 55. fulltime honesty/integrity 32. developmental 180–1.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 207 Inde x quantitative research 3. fraudulent 30. ownership of 188. use of multiple methods 138–40 207 . 23. 46. 152. 18. flexibility 78. 8. as whistle38 blower researcher-participant relationship also power 54. presentation document 151 . 46. 7. 181.

115 United Nations Convention on the Rights of the Child 94 universities 24.Qualitative Research – TEXT 22/11/2001 3:26 PM Page 208 Qualitative practice theoretical sampling 67–8 52. 155. y observatio participant 64–5. 63– 4. 73.research study xviii. 13. 30. 41– 2. 181 Towards a Chinese Conception of Social Support xviii. 13–14 issue ethical/organisationa 45–6 . 25 research in working hypotheses 4 Working Successfully with Families xvii. 127. 106. interview 65–6. 185–6 . 76. for 76different m 170–1 . 42. background to study . voice 160– 159. 80. audiences inarticulate 172–3 . 86. Thomas. 16. 2 Yuen-Tsang. structuring findings 159–60. l s research team impact 43–5 writing up 158– passi . 62–3. 138. Angelina xv. research study xviii. Caroline xiv. 114 . 67 65. study quotations 170 . 138. 107–8. trust 54. 107–8. 118. informants methods a mixed 167–70 . 148. 166–7. Cheryl xiv–xv. negotiatin s g entr 42–3 . research study xvii. 28. 179. 162–4. 181 Tilse. 208 . n sampling 67–8 transcribing 143– 148–9 4. 43. 42. 161. 170. 63– 4.

Sign up to vote on this title
UsefulNot useful