teoria ugruntowana, antropologia kulturowa
[ Pobierz całość w formacie PDF ]
Volume 3, No. 3—September 2002
Constructivist Grounded Theory?
Barney G. Glaser
Abstract
: I refer to and use as scholarly inspiration
CHARMAZ's excellent article on constructivist
grounded theory as a tool of getting to the
fundamental issues on why grounded theory is
not constructivist. I show that constructivist data,
if it exists at all, is a very very small part of the
data that grounded theory uses.
Key words:
grounded theory, constructivism,
constructivist data
References
Author
Citation
Constructivist Grounded Theory (GT) is a misnomer. GT can use any data; it remains to be
figured out what it is. In my book "The Grounded Theory Perspective" (GLASER, 2001) I wrote
a chapter (11) that dealt with "all is data." I said:
"'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 interview,
observations, documents, in whatever combination. It is not only what is being told, how it is
being told and the conditions of its being told, but also all the data surrounding what is being
told. It means what is going on must be figured out exactly what it is is to be used for, that is
conceptualization, not for accurate description. Data is always as good as far as it goes,
and there is always more data to keep correcting the categories with more relevant
properties." (p.145) [1]
"All is Data" is a GT statement, NOT applicable to Qualitative Data Analysis (QDA) and its
worrisome accuracy abiding concern. Data is discovered for conceptualization to be what it is—
theory. The data is what it is and the researcher collects, codes and analyzes exactly what he
has whether baseline data, properline
1
data or objective data or misinterpreted data. It is what
the researcher is receiving, as a pattern, and as a human being (which is inescapable). It just
depends on the research. [2]
Remember again, the product will be transcending abstraction, NOT accurate description. The
product, a GT, will be an abstraction from time, place and people that frees the researcher from
the tyranny of normal distortion by humans trying to get an accurate description to solve the
worrisome accuracy problem. Abstraction frees the researcher from data worry and data
doubts, and puts the focus on concepts that fit and are relevant. [3]
One major worry in QDA research, which does—but should not—effect GT, is a different take
on the personal predilections of interviewer and interviewee. According to QDA interview data
yields the construction of data that represents the mutual interpretation of the interviewer and of
the interviewee as the interview proceeds. This constructivist orientation is that data is
constructed with interacting interpretations. [4]
This orientation, as written, never seems to see it as a characteristic of the type of interviewing.
It probably applies to lengthy, in-depth interviews where mutuality can grow based on forcing
1
See GLASER, 1998a, p.9.
Copyright © 2002
FQS
FQS
- Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (ISSN 1438-5627)
FQS
3(3) Barney G. Glaser: Constructivist Grounded Theory?
type interview guides (see Kathy CHARMAZ, 2000). But this type of interviewing is a small
piece of GT interviewing, although it happens and one can do GT from it. Much GT interviewing is a
very passive listening and then later during theoretical sampling focused questions to other
participants during site spreading
2
and based on emergent categories. It is hard for mutual
constructed interpretations to characterize this data even though the data may be interpretive:
for example psychotherapists telling the interviewer how to see a psychiatric facility or a
supervisor telling how to understand his foremen. [5]
GT is a perspective based methodology and people's perspectives vary. And as we showed in
"Awareness of Dying" (GLASER & STRAUSS, 1965) participants have multiple perspectives
that are varyingly fateful to their action. Multiple perspectives among participants is often the
case and then the GT researcher comes along and raises these perspectives to the abstract
level of conceptualization hoping to see the underlying or latent pattern, another perspective.
This becomes complex, which core variable analysis organizes to reduce the confusion to an
integrated complexity. Further complexifying the data is the type of data the GT researcher is
obtaining—baseline, properline (confirm usage), interpretive, vague—and its varying sources.
[6]
Thus it is just too, too simple a statement when Kathy CHARMAZ (2000, p.510) says:
"I add … another vision for future qualitative research: constructivist grounded theory.
Constructivist grounded theory celebrates first hand knowledge of empirical worlds, takes a
middle ground between postmodernism and positivism, and offers accessible methods for
taking qualitative research into the 21st century. Constructivism assumes the relativism of
multiple social realities, recognizes the mutual creation of knowledge by the viewer and the
viewed, and aims toward interpretive understanding of subjects' meanings." [7]
If this is the way the data come down, then fine, BUT it is a bare small piece of the GT research
action and it does not help "doing" for those doing the research. It just remains to be clear
about the data that obtains and that is whatever it is. She is trying to solve the worrisome
accuracy problem of QDA by trying to ascertain the data emerging in the deep, long (hour or
so) interview situation. This kind of interviewing is characteristic of her "pet" substantive areas
requiring depth, again a small piece of the GT action. Her quest is not to take the data as it
comes, but to be sure it is accurate, so she gets to mutual interpretation as the answer. When I
say that some data is interpreted, I mean the participant not only tells what is going on, but tells
the researcher how to view it correctly—his/her way. I do not mean that they are mutually built
up interpretations. Adding his of her interpretations would be an unwarranted intrusion of the
researcher. [8]
The constant comparative method discovers the latent pattern in the multiple participant's
words, such as, for example, pain leveling provided by dental clinics undermines repair work.
Her miss in that the GT focus is on conceptualization of latent patterns, and GT is about a
concept, e.g. cautionary control, and not about the accuracy of story talk. In fact, in a recent
study of "talk story," by Bay JONES (2002), how the stories were built was irrelevant. They
were efforts at sharing, mutual affirmations and support and camaraderie to reduce the
bewilderment of the lonely ongoing world and to exert shared control by perspective over it.
The competitive parlance was a one-upmanship control to preempt the descriptive scene that
all could share. Thus CHARMAZ talks the talk of conceptualization, but actually walks the talk
of descriptive capture. Accordingly GT is remodeled to a QDA method from its origination of
conceptual core variable analysis of "whatever" data is involved—baseline, properline (confirm
usage), interpreted or vague. Her understanding of abstractions involved in theoretical coding,
substantive coding, delimiting, theoretical sampling etc, etc, are missed, neglected or quashed
in favor of QDA methods and descriptive capture
3
. [9]
2
"Site spreading" is discussed at length in GLASER, 2001, Chapter 12.
3
"Descriptive capture" is a main theme in "Grounded Theory Perspective," and introduced in Chapter 1 of this book
(GLASER, 2001).
Copyright © 2002
FQS
FQS
3(3) Barney G. Glaser: Constructivist Grounded Theory?
So we can see that constructivism—joint build of an interactive, interpreted, produced data—is
an epistemological bias to achieve a credible, accurate description of data collection—sometimes.
But it depends on the data. If the data is garnered through an interview guide that forces and
feeds interviewee responses then it is constructed to a degree by interviewer imposed
interactive bias. But, as I said above, with the passive, non structured interviewing or listening
of the GT interview-observation method, constructivism is held to a minimum. [10]
It appears that constructivism is an effort to dignify the data and to avoid the work of confronting
researcher bias. Remember bias is just another variable and a social product. If the researcher
is exerting bias, then this is a part of the research, in which bias is a vital variable to weave into
the constant comparative analysis. It happens easily in "hot" or "passionate position" issue
oriented research, such as political, feminism, or abuse type research or in research on
inviolate control structures, which cannot tolerate implicit subversion. This aspect of default
remodeling, that is covering bias up for what it is—another variable—is a vital loss to GT. [11]
CHARMAZ (2000, p.522) comes close to what I am saying but descriptive capture of QDA
subverts it. She says: "Like wondrous gifts waiting to be opened, early grounded theory tests
imply that categories and concepts inhere within the data, awaiting the researcher's discovery
… Not so." This statement is unbelievably wrong. Categories, which are concepts, are not
wondrous gifts, they come from the tedium of the constant comparative method linked with
sensitive theoretical sampling and are constantly fitted to the data. [12]
Compounding this wrong thinking, CHARMAZ continues:
"Glaser (1978, 1992) assumes that we can gather our data unfettered by bias or biography.
Instead, a constructivist approach recognizes that the categories, concepts and theoretical
level of an analysis emerge from the researcher's interactions within the field and questions
about the data." [13]
As I have said, to the degree a researcher's personal predilection biases the data, it is a
variable to consider, for example "she thinks that way because she is a feminist." But as I have
also said, the constant comparative process reveals these biases. AND I am also quite gratified
to see that most researchers, I have worked with, take great pains to not intrude there own
views in the data. In addition, the abstractions that emerge become independent of the
researcher bias that CHARMAZ worries about. For example credentializing, cultivating, spiritual
power abusing or pseudo-friending just go on, no matter the bias take on them that may
emerge. For example when a researcher hears "I do not need a degree or certificate, I know it
all anyway," this structurally impossible bias does not do away with the general process of
training. And furthermore, GT is about concepts not accurate descriptions as CHARMAZ
worries about. Descriptive capture remodels GT. [14]
Continuing her descriptive capture, CHARMAZ (2000) says, yet again:
"The grounded theorist's analysis tells a story about people, social processes, and
situations. The researcher composes the story; it does not simply unfold before the eyes of
an objective viewer. The story reflects the viewer as well as the viewed." [15]
Again, absolutely NO, the GT researcher does not "compose" the "story." GT is not description, and
the unfolding is emergent from the careful tedium of the constant comparative method and
theoretical sampling—fundamental GT procedures. These are not story making, they are
generating a theory by careful application of all the GT procedures. The human biasing
whatever is minimized to the point of irrelevancy in what I have seen in hundreds of studies.
The GT reflections of the researcher are his/her skill at doing GT. This remodeling by CHARMAZ of
GT is clearly just not correct and is implicitly supporting the QDA requirements for accuracy.
CHARMAZ has not considered the properties of conceptualization in her offer of a constructivist GT.
[16]
CHARMAZ asserts that the abstract terms and dense writing GLASER (1978) employed in
"Theoretical Sensitivity" rendered the book inaccessible to many readers. This statement is just
not true. "Theoretical Sensitivity" has sold over 3,000 copies. It is used in many many dissertations
and letters to me lauding it are legion. CHARMAZ's assertion legitimizes the default remodeling
Copyright © 2002
FQS
FQS
3(3) Barney G. Glaser: Constructivist Grounded Theory?
of GT down to some conceptual description. It appears that most of her undergraduate
students cannot or hardly can conceptualize, so most do QDA. This is very real, but no reason
to remodel GT. [17]
CHARMAZ constantly pursues, over and over in her article, this constructionist tack on QDA
while using it to remodel GT. She compounds her error by saying, irrespective of their differences:
"Both Glaser and Strauss … assume an external reality that researchers can discover and
record … Glaser and Strauss (1967) imply that reality is independent of the observer and
the methods used to produce it. Because both Glaser and Strauss …follow the canons of
objective reportage, both ... write about their data as distanced experts …, thereby
contributing to an objective stance." (CHARMAZ, 2000, p.513) [18]
I said compounding her error because she neglects the carefulness of the GT method which
makes the generated theory as objective as humanly possible. BUT also she neglects that the
product is conceptual which provides an abstract distance from the data. Thus the
conceptualizations are distant, objectifications if she wishes to use these terms. But more to the
point, she is caught by descriptive capture and is remodeling GT to QDA story talk, while
neglecting the fundamental properties of abstraction analysis. [19]
Using constructivism as a justification in reverse CHARMAZ engages in a recidivism which
makes the researcher's interactive impact on the data more important than the participants.
Constructionism is used to legitimate forcing. It is like saying that if the researcher is going to
be part of constructing the data, then he/she may as well construct it his way. Again the
properties of abstraction are ignored and GT is remodeled. Listen to what CHARMAZ says:
"Glaser assumes that data become transparent, that we researchers will see the basic
social process in the field through respondents' telling us what is significant. However, what
researchers see may be neither basic nor certain (Mitchell and Charmaz, 1996). What
respondents assume or do not apprehend may be much more important than what they talk
about. An acontextual reliance on respondents' overt concerns can lead to narrow research
problems, limited data and trivial analyses." (CHARMAZ, 2000, p.514) [20]
This statement is so untrue and so descriptive captured. She uses constructivism to discount
the participant's main concern, which is always relevant to ongoing resolving behavior, in favor
of the researcher's professional concern, which is most often irrelevant to behavior in the
substantive area (see GLASER, 1998a, Chapter 8, pp.115-132). I have seen this over and over
in research. Then her descriptive capture leads her to totally ignore that the researcher by
constant comparisons conceptualizes the latent pattern—core category—in the resolving of the
main concern, which conceptualization the participants may not be aware of since it
conceptualizes their incidents. So an incident which may have appeared trivial can actually be
a vital indicator of the core category that resolves the main concern. [21]
CHARMAZ is also unaware that the conceptualization of the core category based on incidents
has a generality that may easily inform and be related to the professional problem. Thus Amy
CALVIN, in her dissertation (2000), got nowhere trying to study end of life directives,
particularly organ donations. When she listened to the participants she discovered a theory of
personal preservation under a condition of a deteriorating physical life—an irreversible illness.
This bore heavily on the professional problem and explained why organ donations were not
forthcoming and suggested avenues of potential resolutions to this problem. As I have said in
"Doing Grounded Theory" (GLASER, 1998a), only people who can conceptualize should do
GT. [22]
CHARMAZ continues:
"Most grounded theorists write as if their data have an objective status ... 'The data do not
lie.' ... [But d]ata are narrative constructions. ... They are reconstructions of experience; they
are not the original experience itself. ... Whether our respondents ply us with data in
interview accounts they recast for our consumption or we record ethnographic stories to
reflect experience as best we can recall and narrate,
data remain reconstructions
." (2000,
p.514, my emphasis, B.G.) [23]
Copyright © 2002
FQS
FQS
3(3) Barney G. Glaser: Constructivist Grounded Theory?
Let us be clear, researchers are human beings and therefore must to some degree reify data in
trying to symbolize it in collecting, reporting and coding the data. In doing so they may impart
their personal bias and/or interpretations—ergo this is called constructivist data. But this data is
rendered objective to a high degree by most research methods and GT in particular by looking
at many cases of the same phenomenon, when jointly collecting and coding data, to correct for
bias and to make the data objective. This constant correction succeeds in both QDA methods
and in GT's methodology especially so because the corrections are conceptualized into
categories and their properties, hence become abstract of researcher interpretations. The
latent patterns—categories—hold as objective if the GT researcher carefully compares much
data from many different participants. Personal input by a researcher soon drops out as
eccentric and the data become objectivist not constructionist. Thus, for example, no matter
what are nurses responses to being required to go back to school to get a more advance
degree, the latent pattern emerges is that they are being credentialized. And this substantive
theory has much generality in explaining responses in any field, when its members are being
forced, to go back to a school to get a license, certificate or credential. Credentialzing theory
emerges as real, it is not constructed (see "Gerund Grounded Theory," GLASER, 1998b, for
many examples). Clearly CHARMAZ's formulations are for QDA worrisome accuracy problems, NOT
for GT abstractions, unless, of course, she remodels GT to a QDA method. [24]
CHARMAZ cites several "critical challenges to grounded theory." All the critiques she cites
reflect descriptive capture and a QDA approach, thus are misapplied critiques regarding GT.
GT is a conceptual method, not a descriptive method, as we know. Thus descriptive critiques
which are all about worrisome accuracy do not apply to GT. She cites several authors who
state that GT methods were insufficient to respect their interviewees and portray their stories.
She says: Grounded theory "authors choose evidence selectively, clean up subjects' statements,
unconsciously adopt value-laden metaphors, assume omniscience and bore readers." (2000,
p.521) GT authors are challenged with respect to "their authority to interpret subjects' lives."
These criticisms imply that GT methods gloss over meanings with respondents stories. She
continues:
"Grounded theory research might limit understanding because grounded theorists aim for
analysis rather that the portrayal of subjects experience in it fullness ... fracturing the data
imply that grounded theory methods lead to separating the experience from the
experiencing subject, the meaning from the story, and the viewer from the viewed.
Grounded theory limits entry into the subjects worlds and thus reduces understanding of
their experience." [25]
These criticisms do not apply as they all remodel GT into a QDA method devoted to careful,
full, voice and meaning description of the participant's story, in short a QDA DESCRIPTION.
This is exactly what GT is not—a QDA meaning, story description. GT is a theory about a
conceptualized latent pattern—e.g. cultivating, credentializing, covering, client control, ritual
loss ceremonies ... etc, etc. Criticizing it for not doing what it does not purport to do, is an
authors' error on CHARMAZ's part. It is in essence a default remodeling of GT to a poor QDA
method, and thus a block on good GT research to achieve a conceptual theory: such as a
theory on desisting residual selves. [26]
CHARMAZ's error is compounded by her concluding from her misapplication:
"A constructivist grounded theory assumes that people create and maintain meaningful
worlds though dialectic processes of conferring meaning on their realities and acting within
them ... By adopting a constructivist grounded theory approach, the researcher can move
grounded theory methods further into the realm of interpretation social science ... [with]
emphasis on meaning, without assuming the existence of a unidimensional external reality.
A constructivist grounded theory recognizes the interactive nature of both data collection
and analysis, resolves recent criticisms of the method, and reconciles positivist
assumptions and postmodernist critiques. Moreover, a constructivist grounded theory
fosters the development of qualitative traditions through study of experience from the
standpoint of those who live it." (pp.521-522) [27]
Copyright © 2002
FQS
[ Pobierz całość w formacie PDF ]