Graham Gibbs (2009). Analysing Qualitative Data (Series: The SAGE Qualitative Research Kit). London: Sage, 176 pages, ISBN: 9780761949800, £20.99

Abstract: The reviewed book concentrates on the issue of qualitative data analysis. It primarily provides practical information for students and social researchers on how to deal with textual data starting with data preparation until writing the actual report. The book proposes coding as a convenient and effective way to process data. But, what is additionally valuable, it suggests multiple comparative analysis techniques which would allow a researcher to go beyond simple "coding and retrieving" in order to achieve more plausible and in-depth interpretations. Apart from this, it introduces a reader to the analysis of biographies and narratives, as well as to computer-assisted qualitative data analysis software (CAQDAS). The book is full of practical advice, with a lot of how-to-do suggestions and useful examples, as well as indications of the possible difficulties in the analysis. Still, even if the book seems to be kept on an elementary level, it would have helped readers to better apply proposed practical advice to their research, if, before reading the book, they had acquired more general knowledge about the specificity of qualitative analysis and the various approaches used within it.

The book "Qualitative Data Analysis" was published as a part of the Sage Qualitative Research Kit, edited by Uwe FLICK. The Kit includes positions discussing topics such as qualitative research design, data gathering, data analysis—both textual and visual, and quality of research. The Kit focuses on "how-to-do problems and on how to solve such problems in the research process" (p.xii), and therefore it takes on a very pragmatic character. The book considered in this review, written by Graham GIBBS, concentrates on the issue of qualitative data analysis, primarily providing basic information for students and social researchers on how to deal with textual data. The book focuses on giving practical advice on how to process qualitative data in order to get a valuable interpretation, suggesting the use of computer-assisted qualitative data analysis software (CAQDAS) as a tool to ease data management and processing, but also as a way to assure a systematic approach to data. The book gathered "the best of" from many other, canonic books on data analysis (e.g., STRAUSS & CORBIN, 1998; MILES & HUBERMAN, 1994; SILVERMAN, 1997), providing the reader with advice, "good practices" and "good hints," as well as a lot of technical information and illustrative examples. [1]

The book gives very clear and straightforward information about the preparation, reduction, and sorting of gathered data. Overall, it presents the research process as built of five steps: data preparation, data extension, coding, comparing, and, finally, writing a report. The author points out that the process of interpretation cannot be separate from the data gathering and its preparation: each step undertaken by the researcher is based on the interpretation of the previous steps of the research process (compare MILES & HUBERMAN, 1994). [2]

The first step in the described process is data preparation: as the book deals with textual data, Chapter 2 concentrates on the issues connected to transcribing, seeing it already as a first step in analysis of the data. For GIBBS, even the seemingly technical action of transcribing is not free from interpretation. Chapter 3, on writing, encourages novice researchers to expand their material by keeping a research diary, and doing field notes and memos from the first day of their fieldwork. Using the motto "writing is thinking" (p.25), GIBBS encourages researchers to write extensively about each aspect of the research (actions, thoughts, secondary material one has read, even researcher's emotions), in the hope that this will produce innovative ideas. He also describes memos in this chapter, even though they seem to better fit into later chapters (4 and 6) on coding and comparative analysis. These chapters constitute the core of the book and will be discussed further. The chapters are full of hints on not just how to code but also how to relate codes to each other and to the attributes of researched cases. [3]

The reflection on coding is accompanied by the material about the usage of CAQDAS (such as ATLAS.ti, NVivo and MAXqda) and how these pieces of software support a researcher in his/her work. The book additionally contains a chapter about analytic quality and ethics, which reminds the reader of very crucial aspects of every study—researcher reflexivity, research reliability and validity—as well as bringing to light important ethical questions that might emerge while analyzing qualitative data. Last but not least, the chapters on coding and comparative analysis are split up by a chapter on the analysis of biographies and narratives. In this chapter, GIBBS stresses certain aspects of narrative data, mainly structural, that should be considered while analyzing narratives. [4]

The focus of the book is on coding and on code's utility in more advanced comparative analysis. GIBBS presents the issues of coding in a very extensive way showing how a comparative approach to coding might help a researcher in getting interesting interpretations. It is very valuable that the author tries to go beyond the "coding & retrieving" techniques prevailing in qualitative data analysis and asks the questions: How can a researcher get something more than just descriptive results? What procedures can help us to relate concepts to each other, to the specificity of the case, and to its context? The author believes that intensive coding, code-comparisons, constructing and working on the code hierarchy, building tables representing the data (compare MILES & HUBERMANN, 1994), and case-by-case comparisons will all allow a researcher to move from the descriptive level of interpretation to the analytic one. GIBBS also presents memos as another way of escaping from simply descriptive results, stressing that memos should be kept at a conceptual level. [5]

These chapters represent what I like about the book the most: the fact that the interpretation of the qualitative data is seen more as a technical process rather than a magical illumination. Of course neither the discovery nor the idea is ever free from some form of an illumination, but the book shows (e.g., to students looking for the content of their term final report) that it is not just by reading-reading-reading that they will finally discover what the data "says" to them. But, it shows that analysis can be a process consisting of clear, sometimes technical, steps, which can lead to answers to the posed research questions. It answers the question one might have been asked by her/his students: "what we are suppose to DO with this data!?" This point also appears relevant in the discussion about the reliability of the qualitative research. And, as GIBBS writes at the beginning of the book, "creating a consistent and perceptive analysis, that remains grounded in those data [...] requires good organization and a structured approach to the data" (p.2). In fact, the book provides a lot of tools that might help to achieve this task, but something that would be a good addition to its later editions is a chapter on planning the analysis: how to decide which tools use and how to manage the process of analysis as a whole: to which extend a researcher should remain faithful to a predefined research design or to which extend one can be flexible to modify the process in order to answer better posed research questions. [6]

The chapter on comparative analysis ends with the proposal to create a model (p.86), which for GIBBS is a framework explaining key aspects of the studied phenomenon in terms of other aspects and/or elements of the situation (p.86). He refers to grounded theory methodology sensu STRAUSS and CORBIN (1998), and proposes to follow its procedures: identify codes, link them—preferably looking for causal relationships, find the central category, and, finally, after doing some "manipulation" with the codes, establish a model that "identifies causes, intervening conditions, actions and consequences" (p.88). This is a very good step—even if in the book this topic did not take up much space—to reach beyond descriptive results and try to build a framework allowing an explanation of social phenomenon. But, if we check the book of Anselm STRAUSS and Juliet CORBIN (1998, p.156), we can see that they did not stop at this point, but went further. The chapter following one on the creation of a model (for them a diagram) is "Refining the Theory." [7]

This decision of GIBBS of having used some of the grounded theory heritage but avoiding to refer to theory creation, made me think about two issues always present for the qualitative analysts: the impact of the grounded theory on qualitative methodologies, and, consequently, the ability or, even responsibility, of qualitative research for developing theory. [8]

This book is an interesting exemplification of how grounded theory methodology has actually influenced general methodological procedures in qualitative social research. Grounded theory methodology promoted a more structured approach to data analysis, proposing to the researcher a defined process to follow. Today it often happens that its tools are used selectively, or it is a nice name given to any inductive research no matter if it aims at theory creation or not. In the book sometimes proposals of the grounded theory methodology are presented as an element of general methodology, and sometimes grounded theory methodology appears as a parallel approach to "just doing research." Instead of assuming that the reader knows grounded theory methodology, I would add a box discussing its place within qualitative methodology and the selective usage of its tools outside a general theoretical framework. [9]

Still, the concern about multiple pathways taken by grounded theory does not seem so much an issue. In the discussions regarding qualitative research, more problematic appears its superficiality and the limited attention given to theory building. Both of these topics were explicated by Juliet CORBIN (interviewed by CISNEROS-PUEBLA, 2004, para. 2):

"There are now many versions of the [grounded theory] method and other than the fact they all share a desire to build theory from data, I don't know exactly what they have in common. I also find that researchers are combining methods, which are parts of Grounded Theory with some other method [...] but not for the purpose of actually building theory. So I would say that Grounded Theory has taken a path of its own. […] But what Grounded Theory becomes doesn't concern me as much as what qualitative research has become. There is more emphasis on alternative methods and little interest in theory development. Students don't want to put in the long hard work that goes into theory building. […] there are those who seem to want fast solutions to doing data analysis. They are satisfied to pull out a few good themes without having to put the effort into doing an in-depth analysis that will lead to theme or concept development. The result is superficial work; which in turn gives qualitative research a bad name." [10]

And even if GIBBS's book puts much emphasis on the systematic character of the researchers' work, and in-depth knowledge of one's data, it does not clearly answer the question about the actual aim of the qualitative analysis. What can and should we actually achieve by the analytic procedures presented? Should it be an in-depth description of social phenomenon? Should our research "develop and refine hypotheses" (p.xi)? Should we create a model that explains determinants of the researched? And, finally, should we actually aim at the creation or the development of a theory? [11]

GIBBS writes close to the end of the book that, "only excellent qualitative work can bear theory or suggest implication of an existing one" (p.145). I believe that the author should challenge the reader and discuss the feasibility of theory development through qualitative analysis, as well as provide hints on how to achieve it with the procedures described in the book. [12]

The usage of CAQDAS is certainly a way to ease analysis process, as well as a way to assure that researcher work will avoid "selective anecdotalism," which is using atypical or spectacular cases to make a general point (p.100). From the technical point of view, the programmes allow a researcher to work with a large amount of data, to store them in one file as well as guarantee easy access to it. Reliability-wise, they might help to control analytical process by providing an action scheme to follow, and to assure that all the data material has been considered in the analysis. [13]

These chapters might be an inspiration for researchers to actually get started using CAQDAS in their work. Besides its explanation of the simple procedures of coding and retrieving, writing memos, and using the search functions, the book shows how the software helps to relate codes, case attributes, or segments of text to each other. Nevertheless, it would be even more inspiring if the chapters about CAQDAS matched the analytical level of the previous chapters on comparative analysis. The software programmes propose fascinating tools for researchers. They can, for example, help us in verifying our hypothesis through an analysis of co-occurrences of codes/concepts, visual representation of the concepts/codes/texts relations, or through more advanced text-mining techniques (e.g., in QDAMiner, clustering, multidimensional scaling, correspondence analysis). Finally, the programmes give an opportunity to construct concept maps or networks, which is something that is really useful while trying to find a theoretical model grounded in the data. Many researchers (MILES & HUBERMAN, 1994; BRIGHTMAN, 2003; WOREK & PEREK-BIALAS, 2006) stress the importance of non-linear structuring of ideas in qualitative data analysis, which would lead to exploration of causal and other relationships between actions and their consequences, as well as between variables, case attributes, and/or case contexts. This structuring can involve the researcher’s own ideas and the ideas of his/her subjects, as well as the final structure of a theoretical framework. [14]

The book's two chapters on CAQDAS usage give a very clear account of how to proceed in using ATLAS.ti, MAXqda and NVivo basic procedures in order to get to know your data. GIBBS shows small differences between these three software options, but does not discuss their general approach to dealing with data—something that would be very valuable. What has to be always stressed is the fact that it must not be programmes that define a researcher's approach to the data. Rather, because the researcher should choose the programme that would best support his/her data approach, the knowledge of available programmes is indispensable (CARVAJAL, 2002; LEWINS & SILVER, 2006). [15]

The book is full of practical advice, a lot of how-to-do suggestions, and useful examples, as well as the indications of the possible difficulties and dangers of mistakes (e.g., how to avoid bad coding or what could be the perils of using CAQDAS). All this makes it an interesting option for the teachers of a qualitative research course as well as for novice researchers. Still, it would be useful if beforehand its readers were introduced to the specificity of qualitative analysis as such and various approaches used within it. This would then constitute a theoretical basis for the practical choices they would have to make in the process of data analysis. [16]

Brightman, Jenny (2003). Mapping methods for qualitative data structuring (QDS). Presented at "Strategies in Qualitative Research: Methodological Issues and Practices using QRS Nvivo and NUD*IST" conference, Institute of Education, London, May, 8-9, http://www.ebooknetworking.net/view-cat-creswellresearchdesign-id-1535086.html [Date of Access: March 8, 2010].

Carvajal, Diógenes (2002). The artisan's tools. Critical issues when teaching and learning CAQDAS. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 3(2), Art. 14, http://nbn-resolving.de/urn:nbn:de:0114-fqs0202147 [Date of Access: October 15, 2008].

Miles, Matthew B. & Huberman, A.Michael (1994). Qualitative data analysis: A sourcebook of new methods. Beverly Hills, CA: Sage.

Silverman, David (Ed.) (1997). Qualitative research: Theory, method and practice. London: Sage.

Strauss Anselm L. & Corbin, Juliet M. (1998). Qualitative analysis for social scientists. Techniques and procedures for developing grounded theory (2nd ed.). Cambridge: Cambridge University Press.

Worek, Barbara & Perek-Białas, Jolanta (2006). Tworzenie map pojęciowych. Jakościowa technika rekonstrukcji procesów kognitywnych. In Józef Garczarczyk (Ed.), Ilościowe i jakościowe metody badania rynku. Pomiar i jego skuteczność (pp.165-178). Poznań: Wydawnictwo Akademii Ekonomicznej w Poznaniu.

Ewa KRZAKLEWSKA, MA, is a PhD candidate at the Jagiellonian University of Krakow, in the Institute of Sociology. Her interests include youth sociology, transition to adulthood, research methods, and academic mobility.

Jagiellonian University of Krakow
Institute of Sociology
ul. Grodzka 52, Krakow
Poland

Krzaklewska, Ewa (2010). Review: Graham Gibbs (2009). Analysing Qualitative Data [16 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 11(3), Art. 4, http://nbn-resolving.de/urn:nbn:de:0114-fqs100341.

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

ISSN 1438-5627

Creative Commons Attribution 4.0 International License