Volume 4, No. 1, Art. 15 – January 2003
A Constructive/ist Response to Glaser
Abstract: Recent articles on the Grounded Theory Method (GTM) have started to analyze its conceptual and philosophical foundations. In particular it has been argued that the early characterizations by GLASER and STRAUSS exhibit a scientistic and positivist orientation that is no longer tenable. In her recent contribution to the GTM literature, CHARMAZ distinguished between objectivist GTM and constructivist GTM. This drew a response from Barney GLASER in an earlier issue of FQS. What follows is a rejoinder to GLASER, offering some clarification of developments in people's understanding of this important and widely-used qualitative approach.
Key words: grounded theory, qualitative research, constructivism
Table of Contents
Having quite coincidentally, and initially unknowingly, written an article (BRYANT, 2002) that deals with similar issues to those raised by CHARMAZ in her contribution to the Handbook of Qualitative Research (CHARMAZ, 2000), I was pleased to note that Barney GLASER had responded to her in a recent contribution to FQS (GLASER, 2002). 
As I read GLASER's response, however, my pleasure turned to dismay. Instead of a coherent response to a provocative and well-reasoned argument, I found an incoherent and inconsistent article formatted like a poor piece of tabloid journalism. (I have no reason to believe that GLASER did not endorse publication of the paper in its final version.) 
CHARMAZ's argument can be stated in fairly simple terms. She distinguishes between objectivist and constructivist concepts of the Grounded Theory Method (GTM). The former assumes the reality of an external world, takes for granted a neutral observer, and views categories as derived from data. The latter "recognizes that the viewer creates the data and ensuing analysis through interaction with the viewed" (CHARMAZ, 2000, p.523), GTM is then a tool rather than a prescription. 
CHARMAZ is not prescribing the constructivist view as the only valid one, but she is making the case for a full and proper consideration of issues of constructivism as they impact upon some of the central—and oft-repeated—defining phrases of GTM. Many of these founding formulations make up the mantra of GTM, used by an overwhelming majority of articles by researchers reporting their use of GTM. Here are three representative examples—one from the original GLASER and STRAUSS monograph, and one each respectively from later works by GLASER, and STRAUSS and CORBIN:
"[t]he basic theme in our book is the discovery of theory from data systematically obtained from social research." (GLASER & STRAUSS, 1967, p.2)
"The first step in gaining theoretical sensitivity is to enter the research setting with as few predetermined ideas as possible—especially logically deducted [sic], a prior [sic] hypotheses. In this posture, the analyst is able to remain sensitive to the data by being able to record events and detect happenings without first having them filtered through and squared with pre-existing hypotheses and biases." (GLASER, 1978, pp.2-3)
"A researcher does not begin a project with a preconceived theory in mind (unless his or her purpose is to elaborate and extend existing theory). Rather, the researcher begins with an area of study and allows the theory to emerge from the data." (STRAUSS & CORBIN, 1998, p.12) 
These statements, like many others, including those in GLASER's recent response to CHARMAZ, are objectivist—i.e. positivist—in the sense that representation is seen as ultimately unproblematic once a neutral point of reference can be assured for the researcher. In the 1960s such assurances could be couched in the terms of accepted ideas about rigorous scientific method, and it is not surprising to note that the founding statements of GTM were couched in a clear and deliberate positivist view of scientific research. This was understandable given the dominance of such ideas in the 1960s, but it has become less comprehensible since then, given the extensive critiques of positivism that have emerged in the last 40 years. Any "guarantees" of neutrality these days can only be given once objectivist GTM can be seen to have engaged with the constructivist arguments. 
Now it may be that GLASER has addressed these issues elsewhere, but I am unaware of any such work by him or any other objectivist GT people. GLASER's response to CHARMAZ continues to indicate a position uninformed by what are now acknowledged to be key arguments about science, claims to knowledge, and representation that must be taken into account even if only to be challenged and undermined or entirely refuted. If there is a persistent refusal to engage with such issues, then perhaps it indicates a fundamental conceptual weakness in GTM itself; something which the constructivist reinterpretations seek to remedy. 
At this stage I do not wish simply to repeat the sorts of argument made in CHARMAZ's and my contributions—any interested readers will, I am sure, follow these up for themselves. The essential issues are that the positivist stance of a neutral observer, gathering data about the world, from which theories somehow emerge is now so severely discredited that one of the few places in which one can find such unreconstructed positivism is in the work of some of those claiming adherence to GTM—including, but not restricted to, Barney GLASER. 
Here is an example from the field of informatics, a paper by a group of researchers studying Group Support Systems [GSS] in cross-cultural contexts. De VREEDE, JONES, and MGAYA offer their version of the GTM mantra as follows—and it is in no way exceptional in the GTM literature.
"This approach [GTM] aims to develop inductively derived grounded theories about a phenomenon. A grounded theory is not built a priori; rather, it emerges during study as data collection, analysis, and theory development occur in parallel." (De VREEDE, JONES, & MGAYA, 1999, p.205) 
What they mean by "in parallel" is not clear; but they almost immediately undermine the statement about emergence by noting that "data-collection activities may be guided by relevant existing theories". They try to offer some explanation and justification for what they mean by being "guided by ... relevant existing theories", arguing that it might be thought that their "data-collection efforts could have been prestructured using research domain relevant theories such as the cultural theories of Hofstede or the TAM". Given that they are obviously well aware of such theories, the authors reassure their readers that they took a conscious decision not to allow this knowledge to affect their work, "in order to avoid a standard way of thinking about the phenomena observed". How they managed this feat of cognitive evasion is not clear. 
Here is a view of cognition that is determinedly objectivist. Other theories, known to the observer can simply be discarded, assumptions can be reduced or dispensed with altogether. The phenomena can be observed from a totally neutral position by a dispassionate, passive observer. Cognitive reservoirs of previous experience and knowledge can be dammed, blocked or diverted—the imagery is theirs. The flow of "raw data" can be turned on and off like a tap; and categories and theory emerge from this neutral, passive observational practice—"we closely examined all collected data, broke them into discrete parts, and labeled these parts". How the parts were identified is never explained. 
As a statement of GTM this is, unfortunately, unexceptional; as a statement about observation and theoretical insight it is naïve and misleading. Couched within a perspective that allows for theory to be "inductively derived", it begs far too many questions that have been at the center of philosophy of science for at least the past 30 years. De VREEDE et al. give no explanation for what counts as "data" in their work. They see no problem with "induction", despite the fact that it has been largely discredited in such a simplistic form. Their observational role is largely passive, yet they fail to explain how the data can be broken into "discrete parts" and how "categories could be identified". 
GLASER in his response to CHARMAZ echoes the same themes. He states that "data is discovered"; and he may understand what is meant by "It just remains to be clear about the data that obtains and that is whatever it is". Unfortunately he does not make it clear to the reader, and so at best the statement stands simply as an article of faith—"it is what it is". 
So why then do the anti-objectivists bother with GTM? Why not simply jettison the whole approach, leaving it to GLASER and his colleagues? The argument for the revising and reconceptualizing of GTM is best put by TURNER who characterizes GTM as an
"approach to qualitative data [that] promotes the development of theoretical accounts which conform closely to the situations being observed, so that the theory is likely to be intelligible to and usable by those in the situations observed, and is open to comment and correction by them. ... The approach also directs the researcher immediately to the creative core of the research process and facilitates the direct application of both the intellect and the imagination to the demanding process of interpreting qualitative research data. It is worth noting that the quality of the final product arising from this kind of work is more directly dependent upon the quality of the research worker's understanding of the phenomena under observation than is the case with many other approaches to research." (TURNER, 1983, pp.334-335, stress added) 
Ultimately GTM is far too valuable a method to leave to the objectivists. If we look at what GLASER and STRAUSS actually did, rather than what they claimed—and continued to claim—they were doing, there is the basis for a powerful research approach. BASZANGER and DODIER (1997) term the method of GLASER and STRAUSS' as one of "constant comparison". They characterize it as a method "consisting of accumulating a series of individual cases, of analyzing them as a combination between different logics of action that coexist not only in the field under consideration, but even within these individuals or during their encounters". The aim of such methods is generalization rather than totalization, with the objective of producing "a combinative inventory of possible situations". 
With such ideas in mind, I want to look at GLASER's response in a little more detail—a sort of small-scale example of constructivist GTM. A close reading of the piece reveals a number of categories—which do not magically emerge, but arise from my particular reading of the article—just as GLASER's reading of CHARMAZ's article leads him to (re)conceptualize her views in a particular way. (I accept the issue and inevitability of idiosyncratic readings; GLASER appears not to.) 
The categories that emerge are as follows. 
GTM is held up in contrast to a largely undefined category of QDA—qualitative data analysis; whereas GTM is based around the statement "all is data". Whatever QDA actually is, it is not based on this; but whether this means that QDA is based on something along the lines of "all is not data", or "not all is data" is not clarified. The two categories, carefully placed in mutual opposition, seem to fill out largely from the data—i.e. GLASER's own words—with the use of concepts such as "worrisome accuracy". 
"Worrisome accuracy" is such a slippery concept—never defined or defended—that its interest arises largely from the use made of it by GLASER, rather than any intrinsic meaning. Is GLASER arguing that GTM need not bother about accuracy? If so, what is the meaning of the key GTM terms of fit and relevance? How can they be judged other than on the basis of accuracy or propriety with regard to some criteria or baseline? It seems that GLASER uses the concept of worrisome accuracy simply in order to be able to drive a conceptual wedge between GTM and QDA—a category forced on the reader by GLASER purely to act as a contrast to GTM. Furthermore this allows GLASER to cast aside the critiques mentioned by CHARMAZ, since they apply only to "descriptive capture" and "descriptive methods", and have no engagement with GTM. 
A further category that emerges from GLASER's response is that of abstract agency. GLASER, correctly, sees the constructivist orientation as one of active involvement in the research process; although GLASER characterizes this in a disparaging fashion, using terms such as bias, passion, personal predilections and so on. GLASER sees this as no part of GTM, and he allows himself the cheap jibe at CHARMAZ that "she thinks that way because she is a feminist". (GLASER may counter that he only included this as an example—but it is telling that it is his only example in an under written response to a contribution from a woman. I leave it to the reader to judge if that is a forced conclusion or one that emerges.) 
The point about abstract agency, however, is crucial. GLASER makes several points against QDA, constructivist GTM or whatever, along the lines that such approaches do not confront researcher bias. So how does GTM deal with this issue? GLASER states that GTM provides a method for discovering or conceptualization of latent patterns; but this begs the question that lies at the root of the contructivist critique of GLASER's position—how does this discovery or conceptualization take place? The constructivist position would argue that there is a dialogue between the researcher and the research subject—in both senses of the word "subject"—i.e. the person who is the concern of the research, as well as the research area itself. GLASER neatly evades this with fairly consistent use of forms of grammar that preclude or conceal this issue. Thus he gives abstract nouns the power of agency or action—e.g. "GT can use any data", "Categories, which are concepts, ... are constantly fitted to the data." (stress added) The researcher herself, however, only has agency if it is decided that such an issue "has relevance"—who makes this decision is left unstated. 
This leads on to the category of "data" itself—perhaps the pivotal issue. The constructivist position, like all those emanating from an understanding of the profound weakness of classic positivism, cannot accept anything along the lines of data as it is characterized by the original texts of GTM, and in particular by GLASER in his response. GLASER again sidesteps the issue with some complex and often strangely ungrammatical convolutions—thus he states that "data is discovered"; but by whom? He then argues that "[I]t just remains to be clear about the data that obtains and that is whatever it is." If this can be excused as part of a rushed response, then the same cannot be said for the quote GLASER himself selects from his own book.
"'All is data' is a well known GLASER dictum. What does it mean? It means exactly what is going on in the research scene is the data, whatever the source, whether the interview, observations, documents, in whatever combination. It is not only what is being told ... but also all the data surrounding what is being told." 
In other words, data is the data; and is also the process that goes on in capturing the data; and is also anything in addition to the data itself—hardly the most helpful formulation. 
If this sounds as if I am being petulant and somewhat outlandish at GLASER's expense, then it is largely a reflection of the frustration that I feel at GLASER's comments. He provides the reader with very little to counter or clarify the arguments put forward by CHARMAZ. In fact what he does offer are a series of disjointed slogans that satisfy no one—"conceptualization not accurate description", "conceptual reality does exist", "GTM is not about descriptive capture", "the data is what it is". 
What we need from GLASER is a sustained engagement with the ideas that saturate the category of constructivism. What we have got so far is a response that reads as if GLASER is more intent on establishing "The One True Church of GTM", than he is in clarifying the conceptual foundations of the method. STRAUSS has been long cast out as an apostate. Anyone who uses GTM in a manner that GLASER finds "incorrect" will suffer a similar fate. GLASER will be the arbiter of what counts as GTM and who is best able to use it—blessed are those who can "conceptualize", they will inherit GTM, the rest will have to be satisfied with QDA. 
Now if this is what GLASER really wishes to do, then so be it; but I would like to be able to rescue the key ideas of the method. GLASER may feel proprietorial about GTM, he has a certain right to do so as far as the initial statements are concerned; but he now has to acknowledge that GTM has outgrown his grasp. We have GLASER's view of GTM—well-documented in his books and papers, but we also have several other views, including what I consider to be a far more potent and coherent one that is well exemplified by the article by CHARMAZ. GLASER's version of GTM is not the only game in town. 
Baszanger, Isabelle & Dodier, Nicolas (1997). Ethnography: Relating the Part to the Whole. In David Silverman (Ed.), Qualitative Research: Theory, Method and Practice (pp.8-23). London: Sage.
Bryant, Antony (2002). Re-grounding Grounded Theory. The Journal of Information Technology Theory and Application, 4(1), 2002, 25-42.
Charmaz, Kathy (2000). Grounded Theory: Objectivist & Constructivist Methods. In Norman Denzin & Yvonna S. Lincoln (Eds.), Handbook of Qualitative Research, 2nd edition (pp.509-535). Thousand Oaks: Sage.
De Vreede, Gert-Jan; Jones, Noel & Mgaya, Rabson J., (1998). Exploring the Application and Acceptance of Group Support Systems in Africa. Journal of Management Information Systems Winter 1998-99, 15(3), 197-234.
Glaser, Barney (1978). Theoretical Sensitivity: Advances in the Methodology of Grounded Theory. Mill Valley, CA: Sociology Press.
Glaser, Barney G. (2002, September). Constructivist Grounded Theory? [47 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [On-line Journal], 3(3), Art. 12. Available at: http://www.qualitative-research.net/fqs-texte/3-02/3-02glaser-e.htm.
Glaser, Barney & Strauss, Anselm L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine De Gruyter.
Strauss, Anselm L. & Corbin, Juliet (1998). Basics of Qualitative Research: Grounded Theory Procedures and Techniques, 2nd edition. Thousand Oaks: Sage.
Turner, Barry (1983). The Use of Grounded Theory for the Qualitative Analysis of Organizational Behaviour. Journal of Management Studies, 20(3), 333-348.
Antony BRYANT is Professor of Informatics at Leeds Metropolitan University, UK, and ASEM Chair at the Asia-Europe Institute, University of Malaya. Originally a social scientist, he retrained as a software engineer, and worked as a software developer before moving into university teaching. His research interests include software specification, information management and research methods. He is currently involved with several projects using GTM along the lines described in some of his recent writing.
Professor of Informatics
School of Information Management
Leeds Metropolitan University
LEEDS LS6 3QS, UK
Bryant, Antony (2003). A Constructive/ist Response to Glaser [25 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 4(1), Art. 15, http://nbn-resolving.de/urn:nbn:de:0114-fqs0301155.