Qualitative data analysis - classmatandread.net

1 Qualitative data analysisLearning how to analyse Qualitative data by computer can be fun. That is oneassumption underpinning this new introduction to Qualitative analysis , whichtakes full account of how computing techniques have enhanced and transformedthe field. The book provides a practical and unpretentious discussion of themain procedures for analysing Qualitative data by computer, with most of itsexamples taken from humour or everyday life. It examines ways in whichcomputers can contribute to greater rigour and creativity, as well as greaterefficiency in analysis . The author discusses some of the pitfalls and paradoxes aswell as the practicalities of computer-based Qualitative perspective of Qualitative data analysis is pragmatic rather thanprescriptive, introducing different possibilities without advocating oneparticular approach. The result is a stimulating, accessible and largely discipline-neutral text, which should appeal to a wide audience, most especially to arts andsocial science students and first-time Qualitative Dey is a Senior Lecturer in the Department of Social Policy and SocialWork at the University of Edinburgh, where he regularly teaches researchmethods to undergraduates.

2 He has extensive experience of computer-basedqualitative analysis and is a developer of Hypersoft, a software package foranalysing Qualitative data . Qualitative data analysisA user-friendly guide for socialscientistsIan DeyLONDON AND NEW YORKF irst published 1993by Routledge11 New Fetter Lane, London EC4P 4 EESimultaneously published in the USA and Canadaby Routledge29 West 35th Street, New York, NY 10001 Routledge is an imprint of the Taylor & Francis GroupThis edition published in the Taylor & Francis e-Library, 2005. To purchase your own copy of this or any of Taylor & Francis or Routledge s collection of thousands ofeBooks please go to 1993 Ian DeyAll rights reserved. No part of this book may be reprinted or reproducedor utilised in any form or by any electronic, mechanical, or other means,now known or hereafter invented, including photocopying and recording,or in any information storage or retrieval system, without permission inwriting from the Library Cataloguing in Publication DataA catalogue record for this book is available from the British LibraryLibrary of Congress Cataloging in Publication DataA catalog record for this book is available from the Library of CongressISBN 0-203-41249-4 Master e-book ISBNISBN 0-203-72073-3 (Adobe eReader Format)ISBN 0-415-05851-1 (hbk)ISBN 0-415-05852-X (pbk) Contents List of figures, illustrations and tables vi Preface xi Acknowledgements xiv1 Introduction 12 What is Qualitative data ?

3 103 What is Qualitative analysis ? 314 Introducing computers 575 Finding a focus 656 Managing data 777 Reading and annotating 878 Creating categories 1009 Assigning categories 12010 Splitting and splicing 13711 Linking data 16112 Making connections 17713 Of maps and matrices 20114 Corroborating evidence 22715 Producing an account 24516 Conclusion 272 Appendix 1: If the Impressionists had been Dentists 277 Appendix 2: Software 281 Glossary 283 References 285 Index 288vFigures, illustrations and The steps involved in data analysis chapter by chapter Describing a bit of data as a ripple in the flow of experience Category relating two similar observations Categorizing using inclusive categories Nominal variable with mutually exclusive and exhaustive values Ordinal variable indicating order between observations Interval variable with fixed distance between values Quantitative and Qualitative data in dynamic balance Qualitative analysis as a circular process Three aspects of description in Qualitative analysis Categorizing as a method of funnelling data Derivation of nominal variables with exclusive and exhaustive values Formal connections between concepts Formal and substantive connections between building blocks Connections between chronological or narrative sequences Causal connections between concepts Qualitative analysis as a single sequential process Qualitative analysis as an

4 Iterative spiral A link between text held in separate locations Deriving hypotheses about humour from the literature Main themes for analysing humour Integrating themes around issues of style and substance Case documents kept in a hierarchical file system data stored in fields on a card-based filing system Relating data to key themes Mapping ideas to data within and across cases Relating two ideas Alternative category lists for analysing female stereotypes Weighing up the degree of refinement in initial category set Developing a more refined category list Categorizing data 1 Categorizing data 2 Categorizing data 3 Levels of subclassification of the subcategory suffering Initial relationships between categories Incorporating categories, and distinguishing more and less importantlines of analysis Reassessing relationships between categories 1 Reassessing relationships between categories 2 Reassessing position of categories in analysis Revising analysis with minimum disturbance Comparing subcategories of substance Shifting the analytic emphasis Single hyperlink between two bits of data stored separately Multiple hyperlinks between bits of data stored separately Linking dentists and patients Observing the link debunked by between databits Linking and categorizing complement each other Linking two databits An explanatory link between two databits Linking and categorizing two databits Inferring an explanatory link between two databits Explaining Mrs Sol Schwimmer s litigation Conditional and causal links in the tale of Kaufman and Tonnato Connecting incongruous and cathartic humour Linking data and connecting categories The difference between associating and

5 Linking events Association and linking as mutually related means of establishingconnections Following a trail of links through the data Two trails of links through the data Following a trail of different links through the data A chain of causal links in the data Retrieving chronological links in the Claire Memling story Vincent s explanations linked to chronology of events in the ClaireMemling story Textual and diagrammatic displays of information Map of relationship between two concepts Map of complex relationships between four variables The history of the universe through time A small selection of symbols based on computer graphics Differentiating concepts through different shapes and patterns Incorporating detail by including subcategories Adjusting for the empirical scope of categories Mapping relationships for all cases Comparing differences in scope through a bar chart Using overlaps to indicate scale Adjusting for scope in presenting classification scheme Adjusting scope of most refined categories Distinguishing exclusive and inclusive relationships Making relationships between categories more explicit Representing strength of different causal relationships Comparing strength of relationships between categories Integrating connections between categories Representing reciprocal connections between categories Identifying positive and negative categories Representing concurrence between categories Using space to represent time Concurrence between categories Two routes through the data .

6 Arriving at different results The whole is greater than the sum of the parts 1 The whole is greater than the sum of the parts 2 Tree diagrams representing different analytic emphases Tree diagrams indicating different analytic emphases Different writing strategies sequential and dialectical Decision-making laid out in algorithmic form Procedures for assigning categories in algorithmic form The two aspects of generalization Linear representation of analysis Loop representation of analysis analysis as an iterative process Different approaches to Qualitative research Structured and unstructured responses to the question What are themain advantages and disadvantages of closed questions in an interview? Example of a grading and marking system Grades with different mark bands Personal ads The library Comments on feminist humour Two attendants at a Turkish Bath Recording data fully but inefficiently Filing reference information questions and sources data filed efficiently In the Office Using memos to open up lines of enquiry Linking memos and data Preliminary definitions of categories Developing a more extensive category list Two ways of identifying bits of data Overlapping bits of data A preliminary category list Checking memos prior to categorizing data Contrasting definitions of the category temperament Inferring an emotional state from behaviour data stored following categorization of a databit Categorizing Vincent s first letter Comparing databits assigned to different categories Databits assigned to the category suffering Subcategories of suffering

7 Subcategorized databits for the category suffering Subdividing databits between subcategories Comparing databits between categories Possible links Information held on linked databits Implicit classifications in everyday life Alternative category lists Result of linking and categorizing two databits Multiple links between databits Linking non-sequential databits Concurrence between categories Comparing databits between the different cells List of indexed databits Boolean operators for category retrievals Retrieval based on categories assigned to proximate bits of data Retrieval based on categories temperament and suffering assigned toproximate bits of data Categories analysed as case variables Cross-tabulating categories as case variables: temperament and suffering in Vincent s letters (N=0) Identifying connections between categories for databits assigned tocategory suffering and databits linked to these by the link caused by Connecting X categories transposing and temperament to Y category suffering through causal links between the databits Comparing information across cases Matrix with non-exclusive values Using a matrix to explore variation in the data Databits by case and category data indices by case and category The number of assignations of each category by case Recoding the data to express more meaningful values Analysing subcategories as separate variables Recategorizing variables as values of suffering Frequencies for the variable suffering Cross-tabulating occupation and suffering Databits assigned to categories active and passive Passive and active responses by gender Distribution of responses by case 267xPrefaceA new book on Qualitative data analysis needs no apology.

8 By comparison with thenumerous texts on statistical analysis , Qualitative data analysis has been is some irony in this situation: even a single text might suffice for thestandardized procedures of statistical analysis ; but for Qualitative analysis , oft-notedfor the diffuse and varied character of its procedures, we might reasonably expect amultiplicity of texts, not just a few. Teaching a course on methods makes oneespecially aware of this gap. This book is my contribution to filling it, and I hope itwill encourage or provoke others to do the contemporary text on Qualitative data analysis has to take account of thecomputer. The days of scissors and paste are over. While those steeped in traditionaltechniques may still harbour suspicions of the computer, a new generation ofundergraduates and postgraduates expects to handle Qualitative data using the newtechnology. For better or worse, these students will not give Qualitative analysis thesame attention and commitment as quantitative analysis , if only the latter iscomputer-based.

9 This book is written primarily for them. I hope it may also be ofsome interest to other researchers new to Qualitative analysis and to those usingcomputers for this purpose for the first the methods presented here assume the use of specialist software tosupport Qualitative analysis , those seeking an introduction to individual softwarepackages must look elsewhere (for example, Tesch 1990). My intention is toindicate the variety of ways in which computers can be utilized in qualitativeanalysis, without describing individual software applications in detail. No oneapplication including my own package, Hypersoft will support the whole rangeof procedures which can be employed in analysing Qualitative data . The researcherwill have to choose an application to support a particular configuration ofprocedures, and one of my aims is to permit a more informed choice by identifyingthe range of analytic tasks which can be accomplished using one software package challenge of developing a software package to analyse Qualitative data hasbeen a useful stimulus to clarifying and systematizing the procedures involved inqualitative analysis .

10 It has also allowed me to write a text informed by what we cando with the computer. In my view, the advent of the computer not only enhances,but in some respects transforms traditional modes of book is based on my experiences as a researcher and teacher as well as a softwaredeveloper. My research has involved a variety of Qualitative methods, includingobservation, in-depth interviewing and documentary analysis ; and through it I havelearnt some of the procedures and paradoxes of Qualitative analysis . As a teacher, Ihave become convinced of the merits of learning by doing , a perspective which hasinformed the skills-based methods course I have taught over the last few years withmy colleague, Fran Wasoff. For those interested in skills acquisition, a text whichprovides a variety of task-related exercises and small-scale projects for studentswould be an invaluable asset. But this is not my aim in this book. Experience ofteaching Qualitative methods has also persuaded me of the value of a clear anduncomplicated introduction providing essential background knowledge and helpingto structure the learning experience.