Does a Grounded Theory dissociated from its epistemological bases make sense?
The example of Charmaz's Constructivist Grounded Theory
Term Paper (Advanced seminar) 2007 19 Pages
II. What is GT and what are its epistemological groundings?
II.1. Strauss and Corbin's Systematic Design
II.2. Glaser's Emerging Design
II.3. The epistemological foundations of Strauss’s and Corbin’s Grounded Theory
III. Charmaz's constructivist Grounded Theory
III.1 Critique on Strauss & Corbin
III.2 Constructivist "Grounded Theory"
III.3 What is really new about Charmaz's approach?
III.4. How justified are Charmaz’s arguments against Strauss & Corbin?
Since the first publication of Grounded Theory1 (in the following GT) by Anselm L. Strauss and Barney Glaser in 1967, a lot of scientists and researchers have written about, applied and reformulated this theory-building method.2 What is lacking in many of these works is GT’s acknowledgement as a methodology. Many researchers have applied the procedures proposed by Strauss and Glaser,3 but they do not seem to be aware of the epistemological implications resulting from the use of the GT method. As Strauss together with his co-author Juliet Corbin stated:
“In this book (Basics of Qualitative Research, BD/CL), we are offering more than a set of procedures. We are offering a way of thinking about and of viewing the world that can enrich the research of those who choose to use this methodology.” (Strauss/ Corbin 1998: 4)
Kathy Charmaz appears to be an exception in this canon, since she is not only thinking about the philosophy of knowledge behind GT, but also attempting to renew the epistemological groundings of GT in the field of constructivism. This leads us to the main question of this thesis: Is this possible? Can the GT method be abstracted from the epistemology standing behind it and be founded on a new one? An affirmative answer would mean that GT from the beginning was simply a method and could be used by all researchers, no matter which tradition they belong to or which theoretical approach they pursue.4 Or as Charmaz put it: “…researchers starting from other vantage points – feminist, Marxist, phenomenologist – can use grounded theory strategies for their empirical studies.” (Charmaz 2000: 511) The negation of the question would imply that Charmaz’s constructivist approach cannot claim the term Grounded Theory for itself.
The following paper will examine this question in greater detail in order to find a tentative answer. Starting with a short description of the development, procedures and epistemological foundations of GT5, the second part will focus on Charmaz’s arguments brought forward against the variant of Strauss and her constructivist version of GT. At the end of this section, her propsals will be evaluated in terms of innovation. In a third section her arguments against Strauss & Corbin will be examined regarding their warrant and validity. Finally, the conclusion will draw together all arguments to solve the question about the exchangeability of the epistemological fundament of GT.
II. What is GT and what are its epistemological groundings?
Grounded Theory is a general methodology for deriving theories from the analysis of the patterns, themes, and common categories underlying observational data (Strauss 1987; Strauss/ Corbin 1994). The methodology “was developed by Glaser and Strauss in the early 1960s during a field observational study of hospital staffs' handling of dying patients” (Strauss 1987: 5). Their pioneering book, The Discovery of Grounded Theory (1967) laid the foundation for the discussion of GT today, and it became a procedural guide for numerous qualitative researchers claiming the use of the methodology to legitimize their research (Charmaz 2000).
In the years since Discovery, Strauss and Glaser have moved the methodology in somewhat conflicting directions while independently publishing several books that refine and explain their early methods (Glaser 1978, 1992; Strauss 1987). The late Strauss, with his more recent coauthor, Corbin, has proposed more formulaic and systematic coding (analytical) procedures in order to give rigor to studies and to provide practical guidelines especially for GT beginners (Strauss/ Corbin 1990, 1998). However, Glaser has emphasized a more flexible and open-ended form of Grounded Theory research to warrant fidelity to data and avoiding the influence of preconceived ideas and knowledge (Glaser 1978, 1992).
II. 1. Strauss and Corbin's Systematic Design
The systematic design is the detailed GT method that Strauss & Corbin identified in their 1990 coauthored book, Basics of Qualitative Research: Grounded Theory Procedures and Techniques and on which they elaborated in their second edition in 1998 (Charmaz, 2000). In this book, the GT is much more prescriptive than the original conceptualization in 1967, but as Strauss & Corbin highlight at the beginning of the second edition:
“There are procedures to help provide some standardization and rigor to the process. However, these procedures were designed not to be followed dogmatically but rather to be used creatively and flexibly by researchers as they deem appropriate.” (Strauss/Corbin 1998: 13)
Their systematic design emphasizes the use of data analysis steps of open, axial, and selective coding for developing a theory that is thus grounded in data.
Open coding is the first analytic process of “conceptualizing, defining categories, and developing categories in terms of their properties and dimensions“. (Strauss/ Corbin 1998: 121) In conceptualizing, data are segmented into discrete incidents, ideas, events, happenings, or acts.6 These are given abstract names – codes, as Strauss & Corbin call them (Strauss/ Corbin 1990, 1998). The conceptual name or code can be placed on the objects by the analyst or may be taken from the words of respondents; the latter are usually referred to as “in vivo codes” (Strauss/ Corbin 1998). “Once concepts begin to accumulate, the researcher should begin the process of grouping them or categorizing them under more abstract explanatory terms, that is, categories.“ (Strauss/ Corbin 1998, 114) After a category is identified, the analyst can begin to develop it in terms of its specific properties and dimensions in order to provide a more detailed category (Strauss/ Corbin, 1990, 1998).
Axial coding is a process in which “data are put back together in new ways after open coding by making connection between categories”. (Strauss/ Corbin 1990: 96) In axial coding, the grounded theorist selects one open coding category, positions it at the center of the process being explored, and then relates it to other categories (Strauss/ Corbin 1990, 1998). This is done by utilizing a coding paradigm involving casual conditions, context, intervening conditions, strategies and consequences (Strauss/ Corbin, 1990, 1998).
Selective coding is “the process of integrating and refining the theory“ (Strauss/ Corbin 1998: 161). In integration, the categories in the axial coding model are organized around an identified (selected) central idea indicating what “this research is all about” (Strauss/ Corbin 1998). In this way, the researcher can provide a theoretical explanation for the process being studied in the research. Once a theoretical scheme is outlined, the analyst refines the theory which consists of “reviewing the scheme for internal consistency, and for gaps in logic, filling in poorly developed categories, trimming excess ones and validating the scheme” (Strauss/ Corbin 1998: 156). “Poorly developed categories are saturated through further theoretical sampling. The theory is validated by comparing it to the data or by presenting it to respondents for their reactions.” (Strauss/ Corbin 1998: 157) Theoretical Sampling is, according to Strauss and Corbin, “sampling on the basis of concepts that have proven theoretical relevance to the evolving theory” (Strauss/ Corbin 1990: 176). For them, theoretical sampling is a pivotal part of generating a good grounded theory. Whenever researchers find gaps in their data and holes in their theories whilst in the process of developing and refining them, Strauss and Corbin strongly recommend conducting continuous theoretical sampling. This can be achieved by going back to the field and collecting additional data to fill the conceptual gaps, resulting in “theoretical saturation” (Charmaz 2000; Strauss/ Corbin, 1990, 1998).
1 The authors of this paper prefer spelling the methodology with capital letters, while grounded theories derived through such method with small letters.
2 For a different example than the one of Charmaz’s constructivist version, see e.g. Hildenbrand 2004.
3 … and later Strauss & Corbin and Glaser by himself. On the development of GT and the split between Strauss and Glaser see below.
4 Actually, Strauss himself is claiming such flexible employment of GT (Strauss 2004: 429). However, the authors of this paper view this issue to be important and questionable, that is why they decided to examine it in detail.
5 Since the split between Glaser and Strauss resulted in a development of their approaches into two different directions and the authors of this paper find Strauss’s concept in its whole more sound and comprehensible, we will mainly focus on Strauss’s approach.
6 … depending on the kind of data which is used in the study.