On the Potentials and Problems of Secondary Analysis. An Introduction to the FQS Special Issue on Secondary Analysis of Qualitative Data

Abstract: Archived qualitative data are a rich and unique yet often unexploited source of research material that can be reanalysed, reworked, and compared with contemporary data. This issue aims to debate the methodological, ethical and theoretical considerations relating to the secondary analysis of qualitative data and to provide exemplars of applications of the method. Many of the papers present actual case studies based on re-using qualitative data, while others build on the growing body of published evidence that provide arguments relating to the strengths and weaknesses of particular approaches to the secondary analysis of qualitative data.

In the first section, the papers explore issues of context: how to best preserve context and challenges posed by decontextualised archived data. The second set of papers offers case studies of reuse in areas including class, medicine, history, and employment and consider reinterpretation of original findings, analytic strategies and ways of teaching secondary analysis data of qualitative data. In the final section the contributions cover more practical issues such as strategies for anonymisation and tools that address some of the deficiencies of current technological systems for handling qualitative data.

Although new resources to support secondary analysis are starting to appear, the need for more still exists, in particular for high quality and transparent exemplars of re-analysis. We hope this issue of FQS goes some way toward filling this need.

Key words: secondary analysis, qualitative data, data archives, context, anonymisation, teaching and learning, social research methods

Archived qualitative data are a rich and unique yet often unexploited source of research material that can be reanalysed, reworked, and compared with contemporary data. This issue aims to debate the methodological, ethical and theoretical considerations relating to the secondary analysis of qualitative data and to provide exemplars of applications of the method. The contributions presented here are divided into three broad categories: the first three address issues of context; the next eight cover approaches to re-use of data, in particular, asking new questions of old data; and the last five discuss procedures for archiving qualitative data, and cover issues such as strategies for dealing with confidentiality and technical matters to do with managing and classifying data. [1]

Many of the papers present actual case studies based on re-using qualitative data, while others build the growing body of published evidence that provide arguments relating to the strengths and weaknesses of particular approaches to the secondary analysis of qualitative data. Some of the contributions are drawn from papers presented at the RC33 Sixth International Conference on Social Science Methodology that took place in Amsterdam in August 2004. The Research Committee on Logic and Methodology (RC33) is a specialist thematic research interest group of the International Sociological Association (ISA) that helps to foster professional networking and collaboration within the field of logic and methodology in sociology. RC33 hosts its main conference every four years, and the 2004 event, on the theme of "Recent Developments and Applications in Social Research Methodology", attracted over 350, mostly European, delegates. [2]

Louise CORTI organised and chaired two sessions on the Secondary Analysis of Qualitative Data. Researchers from Germany, the Netherlands, the UK and Canada spoke on the merits and problems of re-visiting qualitative data by offering case studies of analysis of existing data sources. Themes covered were the presentation and organisation of data, through issues of anonymity and approaches to re-analysis, to teaching with qualitative data sources. This issue of FQS builds on the previous issue, Text-Archive-Re-use, published in 2000, which represented a bringing together of the first thematic collection of published articles that addressed issues of secondary analysis of qualitative data. Four years later we see a second set of papers that help to explore and promote this relatively new methodological approach. CORTI and THOMPSON (2004) argue that there is very little published research on issues relating to the practise of secondary analysis of qualitative data and this also applies to text books and the student market. While there is an ever-growing number of published texts describing different styles of qualitative interviewing, there are fewer on how to analyse and interpret interview material, and almost none on reuse. Thus where does the researcher, scholar or teacher turn for guidance on practise? Methodological handbooks need to incorporate such approaches into their texts to complement the existing guidance on research design, fieldwork practise, and methods of and tools for analysis and writing up. Fortunately, we have seen the rise of such readers since 2002, which have actively begun to appraise the secondary analysis of qualitative data, as cited in many of the papers that follow. These contributions are intended to build on this body of literature and hopefully encourage the take-up of this practise. [3]

In the first section, three papers explore issues of context. Harry VAN DEN BERG begins by presenting his experiences of working on data as part of a large multi-team collaborative project of thirteen researchers. He presents an exploration of some of the methodological arguments raised against secondary analysis when the collaborators were asked to analyse the same dataset, the main challenge being that of decontextualisation. He discusses different theoretical and methodological positions concerning the contextualization of interview discourse. The second paper deals with a similar matter. Here Val GILLIES `and Rosalind EDWARDS explore matters of context in conducting secondary analysis by drawing out a range of epistemological, methodological, practical and ethical challenges associated with the re-use of historically specific archived datasets. Focussing on data relating to change and continuity in family life, the authors outline in their work a method of evaluating social change through comparative historical analysis by comparing contemporary data with accounts collected in the 1960s. The final contribution in this section is written by Jo-Anne KELDER who presents a case study based on how she analysed data from a research project to ground the analysis of a new project using the technique of distributed cognition theory and cognitive ethnography. The paper addresses both theoretical considerations of the challenges of using multiple sources and types of data and methodological matters that raise a different set of ethical issues than typically confront the researcher when undertaking primary data collection. KELDER demonstrates the benefits of designing an ethics application to provide for data reuse. [4]

In the second set of papers, the authors tell of their experiences in re-using data for new research purposes, and raise some interesting and informative practical and methodological matters for consideration. Mike SAVAGE and David KYNASTON discuss their work on revisiting data from classic UK qualitative sociological studies housed at the University of Essex, including Elizabeth BOTT's 1950s study on Family and Social Networks and John GOLDTHORPE and David LOCKWOOD's Affluent Worker Studies. Mike SAVAGE explores analytic strategies and methodological issues, paying special attention to their nature as "classics" and their influence on the canon of theoretical and methodological scholarship. He argues for the re-analysis of the published work itself in addition to the original data, and for a consideration of reading data "against the grain". Savage acknowledges the complexity of issues about secondary analysis, but moves quickly into a "feasibility study", arguing convincingly that we need to augment critique with actual test cases. Savage uses Elizabeth's BOTT's original fieldnotes from Family and Social Network (1956) to trace the development and use of social network analysis in this project. Then he uses previously unanalysed responses to an open-ended query about class to disentangle informants' responses from the framework used by John GOLDTHORPE and David LOCKWOOD (1968) in their Affluent Worker study. Without a doubt, the feasibility of insightful secondary analysis is demonstrated. [5]

Historians often demonstrate greater openness to secondary analysis than do social scientists. They are, by necessity, dependent on secondary materials for their research. David KYNASTON demonstrates this openness in his current project, a social history in mid-20th century Britain. He describes how reanalysis of real raw, unmediated and often very textured historical evidence from primary studies can reveal insights into the original methodology and to help reveal some of the assumptions and prejudices that may have been at work during the fieldwork period. His paper gives numerous examples of the enhanced "texture" that original data provide relative to the smoother, more polished published accounts. As he notes, an interviewer's opinions about the "tastefulness" of the wallpaper in an informant's home reveals much about the home, the informant, and perhaps most of all, the interviewer. [6]

Joanna BORNAT speaks of her reanalysis of an archived dataset on geriatric medicine from research conducted in the 1960s. Two different approaches, reconstructive oral history and digitised analysis, are discussed to highlight the ethical and methodological issues raised by analysing data generated at another time and by another researcher for a different purpose. Likewise, Petra NOTZ follows with an outline of the methodological steps she took, and challenges she confronted along the way, for a secondary analysis of primary interviews from an application-oriented perspective focussing on the work/life balance of middle managers. [7]

Irena MEDJEDOVIC and Andreas WITZEL use an interview from a panel-study about young adults entering the job market to highlight the opportunities, constraints and circumstances involved in secondary analysis. They propose that codes and category schemes used in the computer-assisted analysis of the primary data are also useful for the secondary analyst, and do not necessarily prejudice or contradict fresh coding of the data. The fact that it is difficult to retain coded data in a non-proprietary way, that is, outside of the particular CAQDAS software, makes this a challenge for those archiving qualitative data. A possible solution to this largely technical, but also cultural, problem is addressed by LEGEWIE et al. in their contribution (introduced in the next section). Mike WEED proposes a method that he calls meta-interpretation, which focuses on the interpretive synthesis of qualitative research. In his paper he reviews and evaluates eight research methods or approaches that include some form of synthesis, including literature review, cross-case comparison and interpretive phenomenological analysis. In his view, his meta-interpretation approach and procedures he presents retain an interpretive epistemology that is congruent with primary research analyses. [8]

Finally, in this section presenting exemplars of data re-use, Louise CORTI and Libby BISHOP reflect on the current published literature and existing training provision for secondary analysis of qualitative data and describe the various approaches to support and training taken by the ESDS Qualidata at the UK Data Archive at Essex. They describe the preparation of potentially useful kinds of training materials that have helped to support teachers and learners to incorporate secondary analysis methodology into everyday research culture and practice. [9]

In the final section of this special edition, we present papers that deal with practical issues, such as strategies for anonymisation and technical methods that address some of the deficiencies of current technological systems as they apply to qualitative data management and analysis. Diane OPITZ and Reiner MAUER present findings from a survey undertaken as part of a feasibility study for establishing an infrastructure for an archive of qualitative data (University of Bremen) in cooperation with the German Central Archive for Empirical Social Research (ZA) based at the University of Cologne. The authors focus on different aspects of the practical experiences with the secondary use of qualitative data, and consequences for the acquisition and use of data for secondary analysis. [10]

Moving on to the central issues of consent and confidentiality for secondary data analysis of qualitative data, Denise THOMSON et al. present a case study of working with anonymised data on a Canadian multi-team research project, Knowledge Utilization and Policy Implementation. The authors provide a set of practical recommendations that address some of the central questions of anonymisation and consider the strengths and weaknesses of the anonymisation process. Henning PATZOLD then goes on to describe a range of technical procedures that can be used to modify audio material so that it reduces the possibility of identifying speakers. In doing so, he also considers the boundaries of anonymisation. [11]

Stefan HAUPTMANN also uses technology to show how to create a Clickable Table of Contents (C-TOC) by using Atlas.ti that can offer a quick and easy overview of qualitative research material. HAUPTMANN proposes that by transforming primary raw audio material into a useful arrangement, the need for the lengthy process of fully transcribing qualitative data may be reduced. Finally, Heiner LEGEWIE et al. introduce the use of a prototype called QUESSY.ti. This is an interface between some of the most popular and powerful data bases, or relational data base management systems, in qualitative data management and storage and the CAQDAS, ATLAS.ti. The importance of this system is immediately obvious—it allows for direct import of retrieved data resulting from a database query into a QDA program (currently, ATLAS.ti). [12]

We hope that these papers will provide stimulating reading and positive consideration of some of the challenges involved in doing secondary analysis. We further hope that the papers will encourage readers, including researchers, teachers and students to approach the secondary analysis of qualitative data with a more positive outlook, reduced scepticism and a confidence that they now have a greater intellectual and methodological toolset with which to confront someone else's qualitative data. [13]

Corti, Louise & Thompson, Paul (2004). Secondary analysis of archive data. In Clive Seale, Giampietro Gobo, Jaber F. Gubrium, & David Silverman (Eds.), Qualitative research practice (pp.327-343). London: Sage.

Goldthorpe, John, Lockwood, David, Bechhofer, Frank & Platt, Jennifer (1968). The Affluent worker: Industrial attitudes and behaviour. Cambridge: Cambridge University Press.

Louise CORTI is currently an Associate Director at the UKDA at the University of Essex, where she is Head of ESDS Qualidata, Outreach and Promotion. In the past she has carried out research in the fields of health, education and survey and panel study methodology. She has taught sociology, social research methods, statistics, qualitative software packages, research & data management at undergraduate, postgraduate and professional level. Her recent publications include various articles on archiving and re-using qualitative data and a virtual tutorial for social research methods. Her current research expertise and interests involve enhancing access to digital qualitative data, XML applications for qualitative data, consent and confidentiality in social research, and the application, use and promotion of social science data in teaching and learning. She holds/has recently held both JISC and ESRC research grants across these areas.

Andreas WITZEL, Dipl.-Psych., is director of the Life-Course-Archive on the Graduate School of Social Sciences (GSSS), University of Bremen. His major research interests include qualitative methodology and methods, occupational socialization, biography and the life course, especially the transition from school to work.

Universität Bremen
Graduate School of Social Sciences (GSSS)
Archiv für Lebenslaufforschung
FVG-West, Wiener Straße/Ecke Celsiusstraße
Postfach 330 440
28334 Bremen, Germany

Libby BISHOP is currently a Manager at ESDS Qualidata where she plans, oversees and coordinates the day-to-day work of the section. She also manages the development of ESDS Qualidata Online, especially by selecting collections for enhancement and by writing materials to promote reuse and secondary analysis. She is also helping to develop a framework (DTD), methodology and tools for providing online access to enhanced archived qualitative data resources and has given workshops and conference presentations in these areas. She has over fifteen years experience in qualitative research and research management in academic and non-academic organisations. She has taught sociology, including introductory qualitative methodology. Her current research project uses secondary analysis of several Qualidata collections to explore recent historical shifts in the use and status of processed foods.

Corti, Louise; Witzel, Andreas & Bishop, Libby (2005). On the Potentials and Problems of Secondary Analysis. An Introduction to the FQS Special Issue on Secondary Analysis of Qualitative Data [13 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(1), Art. 49, http://nbn-resolving.de/urn:nbn:de:0114-fqs0501495.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License