The Impact of Symbolic Interaction on Grounded Theory

By Barney G. Glaser, Ph.D., Hon. Ph.D.

(Chapter 10, The Grounded Theory Perspective III: Theoretical
Coding, Sociology Press, 2005)

As I stated in the introduction to chapter 9, GT is a general inductive
method possessed by no discipline or theoretical perspective or
data type. Yet the takeover of GT by Symbolic Interaction (SI)
and all the departments and institutes that SI informs and resides
in is massive and thereby replete with the remodeling of GT. The
literature on qualitative methodology is massive and replete with the
assertion that SI is the foundation theoretical perspective of GT. GT
is reported as a SI method. That GT is a general inductive method is

Sure, GT can use SI type data and its perspective, but as a general
method it can use any other type data, even other types of qualitative
data, as well as quantitative, visual, document, journalistic and in
any combination, and any other theoretical perspective, such as e.g.
systems theory, social structural theory, structural functional theory,
social organization theory, cultural theory etc. Thus, the takeover of
GT as an SI perspective methodology is just discipline-perspective
dominance, as discussed above, and nothing more. It, of course,
dominates with a set of TCs (process, strategies, conditions, context
etc) I have considered at length in chapters above.

Researchers, especially in nursing, just want a theoretical
perspective. SI institutionalizes GT as its own! Researchers
like it because it gives them an ontology (what is data) and an
epistemology (a philosophy of research). The takeover becomes
structurally induced by researchers, especially nursing, in their
research, since they want a theoretical perspective in advance. It
gives them a feeling of power, while they do not realize that the SI
takeover reduces the general method power of GT. The writers
on GT as a SI method use as their legitimating source because of
Strauss’s (my co-author of discovery of GT) training in SI. They
ignore the roots of GT in my training in concept-indicator index
construction in quantitative survey research.

In the following pages, I will discuss these issues at length. Much
has already been said in this book about SI and its set of TCs. This
chapter just focuses and adds some ideas. The goal of this chapter,
as in all the above chapters, is to free GT from this dominance so
GT analysts will have the fullest range of TCs – from any and all
perspectives– possible at his fingertips for emergence. No one
discipline with/and its theoretical perspective defines and possesses
GT, as I discussed at length in chapter 9. Obviously many GTs use
a SI perspective (as well as others), whether bounded or not by it.
Earned, emergent relevance is the TC of best choice.

Sources of SI Dominance

Obviously, the impact, dominance and possession of SI on GT came
from Anselm Strauss’s training in SI at University of Chicago. Many
authors assert this one source of SI. Carolyn Weiner (op. cit. page
6) says: “GT derived from the tradition of SI, this sociological stance
is based on the perspective of George Herbert Mead as developed
by the Chicago school of sociology and asserts that people select
and interpret meanings from their environment, formed in many
definitions of the situation. The individual acquires a commonality
of perspective with others as they learn and develop together the
symbols by which aspects of the world are identified. In other words
there is a social construction of reality.”

Marjorie MacDonald and Rita Schreiber (op cit page 42) “to begin,
we must explicate what we mean when we speak of grounded
theory. Central to our understanding is SI, a theoretical perspective
rooted in the philosophy pragmatism (Blumer 1960/86, Dewey, 1922,
mead,1934/67). Human action and interaction and the construction
and reconstruction of meaning within levels of context are central
phenomena of interest and foci for theory development. This is a
synergistic and dynamic process in which action/interaction changes
the context, which leads, in turn, to the construction of new meaning
and new action. In light of this, GT is concerned fundamentally with
the relationship between person and society.”

Marjorie MacDonald says (op cit, page 126) “For Strauss,
pragmatism was central to his thinking. Although people
acknowledge the theoretical origins of GT as being rooted in SI, it
seems that insufficient attention has been given to its pragmatist
underpinnings, as originally outlined by Dewey (1922) and later
articulated by Blumer.”

Janice Morse affirms in her article “Situating GT, (“Using GT in
Nursing, “op.cit. page 2. ) “The second point is that I treat GT
as a method– that is, as a particular theoretical perspective to
analyze data that originally evolved through a particular theoretical
perspective i.e. symbolic interaction”

P.Jane Milliken and Rita Schreiber recently wrote a chapter entitled
“Can You Do GT without SI? (GT in Nursing Research op cit, page
177 to189) They begin by quoting Ian Dey referring to the sources of
GT – quantitative (Glaser) and qualitative (Strauss) methods: In the
marriage of these two traditions , it was intended to harness the logic
and rigor of quantitative methods to the rich, interpretive insights of
the symbolic interactionist tradition.” (Dey, op cit 1999, page 25)
The ladies conclude “Thus, GT emerged from and is intrinsically
tied to symbolic interaction.” There follows this initial statement
of the ascendancy and claim on GT of SI ( at the start of their
chapter), After this assertion, 12 pages of close argument leads
to their conclusion, which is: .”To achieve this end the researcher
necessarily engages in symbolic interaction within her self or himself
and with the data, with participants and with the emerging theory.
Thus it is our view that symbolic interactionism is inherent in GT
research, whether the researcher is aware of it or not. If research
is truly grounded theory, it cannot occur in the absence of symbolic
interactionism, which is intrinsic to the process. This does not imply
that other theoretical perspectives – such as feminism, critical theory
or hermeneutics– may not be incorporated as well, but that these
other perspective are superimposed onto symbolic interactionism.”
(op cit page 188-9).

In short in their view, GT is possessed by SI as ascendant, no matter
what other theoretical perspective may also be included in the
study. It is their “philosophical justification” for doing GT.; it is their
“epistemology that guides its unfolding.” (op cit page 189). Their
discipline dominance of GT by SI is wrong. No data type defines
GT. We must keep in mind that their misguided view, is because
(1) qualitative data, particularly long interviews are very suitable for
nursing research , (2) they are only trained in interview research
which is SI oriented, and (3) because of their training they are not
really aware of other theoretical perspectives or their TCs – such as
systems , social organization, social structural, phenomenological,
economic, etc, etc and other types of data.—survey, documents,
visual, experimental, library, observational etc, etc. Thus their focus
on GT as needing SI, or GT as an SI method, is a socially structured
vested fiction. It is vested in discipline dominance, departments and
careers with a specific SI research orientation in nursing. These
fictions do not overturn or disappear easily; there is too much vesting
at stack. All I am saying here is that there would be no threat to their
limited view by staying open to TCs from all theoretical perspectives.

The Strauss origination of GT using SI has a pretty heavy impact
and dominance, which given the above discussion, is hard to
resist., but GT is just not an SI in possession. But I, Barney G.
Glaser, was co originator, if not the originator of GT. I was clear in
Discovery of GT, Theoretical Sensitivity (pages 62-64) and Doing
GT that GT was “based in a concept-indicator model” leading to
conceptualization (page 62) taken from psychological research and
used extensively in quantitative research. I then added the constant
comparative method – comparing the indicators – to conceptualize
the categories and their properties. And I then added Lazarsfeld’s
notion of the interchangeability of indicators., which led to theoretical
saturation, so no more indicators need be attended. Thus GT came
straight from survey research analysis. And it came from Robert K.
Merton’s training his students in substantive and theoretical coding
– conceptualization.

GT is just a relatively simple inductive model that can be used
on any data type and with any theoretical perspective. It is just a
general inductive model, or paradigm, if you will, that is sufficiently
general to be used at will by any researchers in any field, any
department and any data type. . No one theoretical perspective can
possess it.

Thus SI perspective people can use it as a sensitizing perspective
,of course, but they are in error to say that it is an SI method and
SI must be used in all GT as its foundation. SI is not necessary to
legitimize the GT method. And in the bargain, this takeover stultifies
the emergence of a full range of possible TCs.

The GT model is used in survey research all the time. I read
many papers and monographs while at Columbia University.,
the most impressive being Lazarsfeld’s “the Academic Mind” a
study of apprehension during the McCarthy era.. My dissertation,
“Organizational Scientists: their Professional Careers” (1964), a
study of scientific recognition, which was published immediately, was
a quantitative GT. The GT model is used on qualitative data that is
not SI oriented.. See also Diane Vaughn: “The Challenger Launch
Decision.” , which book is done from documents. Strauss also gave
a long discussion in Discovery (chapter VII) on the use of library
materials. But keep in mind I am not arguing here for the use of GT
on whatever data, as I have said in my previous monographs: “all is
data” for GT. What I am discussing here is the use of a full range of
TCs for the emergent integration of GT, which means TCs should not
be restricted by one theoretical perspective like SI does.

In Theoretical sensitivity I list 18 TC families which come from
general sociology, not linked with any perspective, e.g. cutting points
are cutting points, ranges are ranges, binaries are binaries. Etc.
Their use has been squelched in large measure by the impact of SI.
In Doing GT I listed more TCs. The full range is awesome and fun
to learn, but SI curbs this knowledge. In Theoretical Sensitivity I
coined the terms Basic Social Process and Basic Structural Process
as TCs in trying to bring structure into SI oriented research.

MacDonald says that (op cit page 128) “ Strauss’s important
contribution in this era was that of SI with a concern for organization
and structural perspectives. In fact, Denzin, in his response to
charges of an a structural bias in SI, cites Strauss’s early work as
evidence that social structures were given their due in terms of their
role in human action.” Their brief allusion to increasing the range
of TCs was small and brought back into the control of SI research.
It was a brief acknowledgment at most. They ignored my 18 TC
families, and now we see they ignore the fullest range of TCs by
being shackled by SI. The biasing power of SI as ascendant is very

Using qualitative data does not make GT a qualitative method, it is a
method that can be used on all data and accuracy is not the issue,
conceptualization is for GT. Thus it follows that when SI researchers
use GT, it does not make it an SI method. The reverse is correct,
SI simply used a method that can be used by any theoretical
perspective from which many TCs can emerge. Yet the QDA
remodeling of GT is clearly aided by the impact of SI on GT. (See
GT perspective II)

A Paradigm

The quest for an ontology and epistemology for justifying GT is not
necessary. It will take these on from the type of data it uses for a
particular research FOR THAT RESEARCH ONLY. GT is simply
an inductive model for research. It is a paradigm for discovery of
what is going on in any particular arena. It provides a global view
by “providing a method of solving the puzzle of viewing human
experience and of structuring reality.” Wendy Guthrie, “Keeping
Client In Line” (1998).. Whether GT takes on the mantle for the
moment of prepositivist, positivist, postpositivist , postmodernism,
naturalism , realism etc, will be dependent on its application to the
type of data in a specific research. But in any case any TC that
emerges to integrate the theory should still come from the fullest
range possible. GT is not only appropriate for nursing research,
but also for education, medicine, public health, management,
organization, accounting, social work, business, finance, sociology,
social psychology etc etc. as many studies I have read can attest to.

GT as an inductive method, would seem to fit Klee’s notion of
paradigm offered in his book: “Introduction to the Philosophy of
Science:” (Oxford University Press (1996) page 135)

“A paradigm is an achievement that defines practice for a
community of researchers. It defines practice because the
achievement constitutes a model to be imitated and further
extended. Future research tries to fit itself to the same pattern
as the original achievement. This definition of practice brings
in its wake, the seeding of the fundamental principles of
the domain, the subsequent possibility of extremely precise
research, a pedagogical tradition that trains succeeding
generation in the use of the paradigm, a collection of institutions
designed to promote the paradigm (professional journals,
professional associations), a worldview with metaphysical

Yes GT is a paradigm as Klee portrays. But the reader should keep
in mind that in generating its subsequent dimensions of used and
institutionalization, that it does not become discipline, theoretical
perspective, and department dominated — as SI has done — and
then the openness of GT is lost to the power of the dominators. And
then subsequently, of course, the fullest range of TCs is lost in the

Klee recognizes this drift to dominance of a paradigm when he says
(op cit page 134) “A general approach to research come to dominate
a field – becomes a paradigm – when the practitioners working
under its direction score an amazing research achievement , an
achievement everyone recognizes as such, even those practitioners
committed to competing approaches.” To be sure this happened to
GT, based on its achievements, BUT in the bargain it became mixed
with those practitioners who are wedded to the SI perspective.
Then of course with SI dominance GT was remodeled. (See GT
Perspective II).

SI Possession Everywhere

As I said above, the methodological literature is replete with the
notion that GT is a SI methodology. Much GT to date, in nursing
and management, seems to use qualitative data, which tends to be
SI oriented. But bear in mind that neither the type of data or the
theoretical perspective of the SI TCs, which are many, define the GT
method. The GT method just uses them as it would any other type of
data or any theoretical perspective with its own TCs.

An immediate cause of SI possession of GT , is that nursing
research is usually highly interactional (though it need not be), so
nurse GT researchers and QDA researchers are highly and easily
drawn to SI thinking and use it for preconceived. frameworks to
model the research. SI provides a source of many TCs which have
grab for intense, long interview data, which detracts from the using
the fullest range of TCs that could be emergent. Those supervisors
of dissertations coming from departments that embrace GT as an SI
method, can easily require those SI TCs to be used as if that is all
there is to it in doing GT. It is not, there are many TCs that are lost
by this SI possession.

I can only give the reader a taste of this SI possession of GT– so
replete, so dominating, so remodeling and so ascendant and so
limiting to the full generality of GT. Here are a number of samples of
assertions to this effect. Keep in mind that they are simply followers
in style of the original, but not accurate, assumption that GT is an SI
method. They simply take for granted the SI foundation of GT. Also
keep in mind that researchers could say the same about systems
theory, social structural theory, cultural theory, social organization, or
any other theoretical perspective.

Karen Daley in her very good dissertation on Asthma and Decision
Making, asserts on page 32 “GT is based on the philosophic
perspective of symbolic interactionism. SI assumes that human
action depends upon meaning that people ascribe to their situations.
This assumption drives grounded theory research by allowing the
researcher to look closely at an individual’s interpretation of self and
their actions.”

Alvita Nathanial in her dissertation on Moral Reckoning of Nurses
says: (page 34) “Moral Reckoning” which emerged from the present
study, is supported by extant research and theory. It is congruent
with SI, a common theoretical foundation of GT. Inquiry from the SI
perspective is particularly appropriate for a study of nurse’s moral
distress. GT is a natural product of the postpositive movement and

Locke in her book of GT(page 30, op cit) asserts “A theoretical
perspective informs how we understand complex social realities and
what we direct our attention to when collecting and conceptualizing
data. (Becker 1986). In his guide for conducting fieldwork, William
Foote white (1984) underscores the importance of having an explicit
theoretical perspective….. a good theoretical perspective, such
as SI, helps qualitative researchers orient themselves to the worlds
they study, but it does not specify what they will find.”. To be sure
grounding her SI assertion using the history of QDA helps legitimate
the SI possession of GT and remodels it.

Janice Morse in her article on “Emerging From the Data” The
Cognitive Processes of Analysis in Qualitative Inquiry.”( in Critical
Issues in QDA, Janice Morse Editor, Sage 1994, page 34) simple
states in her chart that the conceptual basis of GT is SI, as if no
question is involved. She further keeps the same point in 2001,
seven years later, without budging, in her article “Situating GT”:
She says: “The second point is that I will treat the GT as a method
– that is as a particular theoretical perspective (i.e. symbolic
Interactionism).” Possession by SI of GT is sure and complete—, no
question, no further analysis among these authors.!

Alvesson and Skoldberg in their book “Reflexive Methodology:
New Vistas for Qualitative Research,( Sage 2000, page 11) states:
“Roots. Glaser and Strauss’s ‘GT” has dual roots in SI in the person
of Strauss, and the other in the statistically oriented positivism
that was part of Glaser’s intellectual luggage. General surveys
of SI such as Plummer (1991) usually mention GT as particular
orientation with this movement. Although we do not altogether
agree with this description, we do regard SI as the most important
source of inspiration for GT.” At least I +(Glaser) was mentioned
with roots before SI took over in their minds. SI does take over with
ease, because the roots of GT in inductive survey analysis are not
understood by many researchers.

Lamborg and Kirkevold in their article on “Truth and Validity of GT”,
Nursing Philosophy 4, pp. 189-200 write: “From a SI perspective,
Glaser and Strauss provided the interaction perspective of GT. GT
relies heavily on Mead’s and Blumer’s SI (Milliken and Schreiber,
2001). Inspired from SI, GT is concerned with the dynamic
relationship between person (individual or collective) and society.”
The mutual quoting among author of GT being an SI method is

Here are two more affirmations; Karen Locke in her book on GT
in Management Research (op.cit., page 20) says: “American
pragmatism and , in particular, sociology’s SI school of thought
constitute the disciplinary traditions that help to inform GT. SI can
best be understood as working through of the pragmatist world view.”
Locke also says on page 25: “GT’s association with SI school of
thought is repeatedly articulated by its originator and their students.
It is particularly important for those organization researchers who
work outside sociology’s disciplinary boundaries to appreciate this
link between GT and SI.” “And often as a research product, GT
reflects SI’s theoretical and methodological presuppositions about
the nature of the social world and the way it can be studied.” These
statements certainly shut down the openness of GT to a full range of
TCs, by outright assuming of SI’s ascendance over GT.

Phyllis Stern, a noted GT teacher and researcher ordains GT as
possessed by SI in her famous article “Eroding GT” in “Issue in
Qualitative Research Methods” (sage 1994 page 215). She says:
“the framework for the GT theorist is rooted in SI (Blumer, Mead, op
cit). wherein the investigator attempts to determine what symbolic
meaning artifacts, clothing, gestures, and words have for groups of
people as they interact with one another.”

The ordaining of GT as SI, the worldview is purports to project, and
theoretical capitalism involved is unending. GT will be possessed
and taken off the general level for specific people and field
purposes, irrespective of its general power. And in the bargain the
full range of TCs is lost – there is no staying open in the face of the
takeover. But this is not necessary once the reader gets the idea:–
openness is not a threats, it helps.

Two senior researcher, Schrieber and Marjorie MacDonald state
flatly with confidence : “this pragmatic view flows directly from the
philosophical foundations of GT in SI and pragmatism.” (Using
GT in Nursing, op cit., page 43) In the same chapter they say:
“Central to our understanding of GT is SI, a theoretical perspective
rooted in the philosophy of pragmatism (Blumer, 1969, Dewey,
1922,Mead 1934,67).”(op cit page 42 ) Their confidence has roots
in the past, always a legitimating approach. But these roots provide
present blind unthoughts also like, they say,: “SI is a theoretical
perspective that illuminates the relationship between individuals
and society, as mediated by symbolic communication, which…looks
beyond the behavioral component to the underlying meaning that
motivates it.”(page 178).” Good jargon, but so… what has this
to do with GT discovering the latent patterns in any kind of data
from interchangeable indices. This confident sureness strangles
GT!, especially the constant comparative approach to generating
emergent substantive and theoretical codes !

Rita Schreiber and Jane Milliken were shocked by my statement: “GT
can be done outside the theoretical framework of SI (Glaser, 1999)
(Schrieber and Milliken “Using GT in Nursing” page 177). They then
devote a whole chapter trying to handle this rather simple statement,
not realizing GT is a general inductive method.. It is called “Can you
‘Do” GT Without SI” They engage in a heavy, erudite analysis of SI
always concluding with sentences as this: “Clearly, in our view, the
epistemology of GT is steeped in SI;” (page 180) or “Clearly, in our
view GT is a methodology in the sense that Harding (1987) uses the
term, in that it bridges the philosophical underpinnings of SI and the
conduct of GT research endeavor.” (page 181).

They continue “What follows is a discussion of key elements of
technique used in GT as they relate to SI.” These elements fall into
their line since they deal with only one qualitative data type: high
interactional, interpretive data, as if all qualitative data is this type. In
fact there are many types of qualitative data., which they leave out
and which would undermine their contention. :”SI penetrates even
the technical level of GT so that, in our view, an adequate GT cannot
be divorced from SI” (page 187) . They concluded firmly: “Thus it is
our view that SI is inherent in GT research, whether the researcher
is aware of it or not. If research is truly GT, it cannot occur in the
absence of SI, which is intrinsic to the process” .

They say lastly “other perspectives (used in GT research) are
superimposed onto SI.” (page 188) This is a “bit” of a nod to
other perspectives by a perpetually ascendant, impenetrable,
possessive SI perspective. This remodels GT to a QDA method,
with all its negative consequences, but here I simply point out that
they endorse only a small set of TCs. They are not staying open
to the sensitivity of the fullest range of TCs that can emerge. Their
possession closes them down to TC possibilities and stultifies GT.

Their colleagues Crotteau, Bunting and Draucker, in their article
“GT in Hermeneutics” (Using GT in Nursing, op cit ,page 193) echo
Schreiber, Mac Donald, and Milliken, They say: “GT is rooted in
SI, which focuses on the meaning of events to people in natural
settings.” This is the same sureness about the SI takeover of GT.
The echo is everywhere. The roots of GT in general induction and
concept-indicators analysis are lost to these SI owners of GT and
with it is lost emergent TCs.

In a solo chapter Marjorie MacDonald does register a small doubt
about the SI takeover and possession of GT (Using GT in Nursing
op cit page 121) She says: “There are good reasons why critics
have charged both SI and GT with an astructural bias… It is difficult
not to make the judgement that GT does indeed ignore issues of
power, culture, social organization, economics, gender and other
structural influences on human action.” Anti critiques she says
have “concluded that many interactionists have not neglected
social structure, especially since the mid-1970’s. They argue that
the notion of astructural bias as a defining feature of SI should be
reconsidered.(Page 119). MacDonald explains this argument or
controversy (page 117): “Many social theorists saw SI as being in
distinct opposition to the classical European sociological perspective
which was concern with macro-social analyses of societal structures
(e.g. economy, polity, culture) as the primary determinants of human
action. Thus SI emerged as a reaction to the dominance of structural
functionalist perspective in sociology.”

The astructural criticism and complaint of SI is MOOT for GT,
unless, as these researchers would have it, that GT is possessed
by SI. But GT is not possessed by SI, so let the this SI controversy
be argued any which way, it is irrelevant for GT. It is GT’s general
method power that puts it outside this argument. But these
researchers do not realize the distinction and difference between SI
and GT, thus they do not realize the full implications for remodeling
GT by SI notions and the loss of TCs.

Perhaps to counteract this critique of SI and GT, caught in the
bargain, Strauss offered the conditional matrix to bring in a bit of
social structure to GT research. But this gesture is mere tokenism
to maintain the takeover of GT by SI. It barely touches the fullest
range of TCs, which bring all kinds of structure, culture, systems,
social organization. etc…into GT, once is it allowed to stay open
and be free of the SI possession. Once GT is seen as a general
, inductive method that can use any kind of data, since as I have
said in many places: for GT “all is data”, this controversy is truly
irrelevant. The researcher will be sensitive to only those categories,
both substantive and theoretical that emerge as relevant.

This distinction is seen clearly by one of my former students: Dr
Barry Gibson. In excerpts from an unpublished paper posted to
the online forum of the GT Institute website.(Sep 2000): He says
referring to the SI takeover of GT: “This coupling process risks the
possibility that the data become pre-conceived in relation to the
distinctions that are received within existing theory.. The argument
is that such distinctions must not be preconceived as relevant
before the process of theory generation commences “(Glaser, 1992,
page 116.) . I see this preconception as SI TC based in interaction
studies. But bear in mind that GT is conceptual, not interactional.
Further Barry says, quite rightly, “The epistemological implications
of using the incident as the basis of data analysis are clear that
there can be no subjects (participants) in data analysis and theory
generation should be guided by the concept of constant comparison
and theoretical saturation.. In terms of the method of GT, there are
subjects and these are constituted as both the research and the
informant. These both disappear, however, during the process of
data analysis whereby the incident (as conceptualized) becomes
the unit of analysis and all ‘individuals’ in the research process fade

The SI takeover is clearly a remodel with many negative
consequences. GT is simply a general inductive method that
conceptualizes into a generated theory, which explains the latent
patterns in any type of data of a general area, whether substantive or

Naomi Elliot, a Ph.D. candidate in the School of Nursing. St James
Hospital Dublin, Ireland, and a student in my seminar saw this
takeover instantly when as she says (email Feb. 2003): “I was
challenged by a colleague who considers that an exploration of how
SI underpins GT is critical to any discussion on GT methodology.
Having read your Chapter 2, “The Roots of GT”, I am left thinking
that to restrict such a discussion to SI alone is to ignore the
influences such as survey design analysis, explication de text, or
qualitative math etc on GT. So my questions are, to what extent does
SI underpin GT?” My answer to Naomi is above: to no extent. It
depends on the type data used, of which only ONE kind is SI data,
even within the scope of qualitative data.

I could go on but the pattern is clear. In the literature everywhere
SI possesses and GT! And this possession is reinforced by senior
researchers, junior researchers, departments, different fields that like
qualitative research and writers on qualitative methods. A robust full
range of possible emergent TCs is lost to the domination of SI. GT’s
generality is lost to the drive for ontology and epistemology which
discussions have turned to SI for foundation. Now let us look at
some consequences of this possession.

Consequences of SI’s impact on GT

My colleague, Judith Holton, (email 10/2004) expresses quite clearly
the general consequence of SI’s take over of GT. She says, in her
words, “GT has been co-opted by the critical mass of those working
within SI and constructivist theoretical frameworks (esp. nursing
and related health services). The result is a remodeling and eroding
of classic GT as a general methodology; its being lost by sheer
dint of their numbers and by their mutual echoing in peer-reviewed
journals.. In this echoing of one another, they lost the ability to see
beyond their own preconceived worldview. Instead, they continually
reinforce mutually held perceptions thereby blocking out their ability
to remain open to GT as a general methodology that works with any
theoretical framework, as appropriately emergent and with any data
as available.”

There follows here several more specific consequences. The SI
takeover of GT transfers to GT all its problems in the bargain “as if”
they are problems with GT. For one problem, as we have just seen
above,. SI is seen as astructural which deflects focus from relevant
structural categories and structural sensitivity. And, of course, it
is SI which has the astructural problem, not GT. Many feel this
problem with no idea how to solve the lack of fitting TCs from other

SI as the foundation of GT leads to a theoretical capitalism that
admits of no other theoretical perspective, hence search for
other TCs outside the SI perspective. Staying open is hard if not
impossible., when facing a theoretical capitalist as supervisor or just
teacher. Their identity is based on an SI orientation! Training in their
department will be based on SI, not other theoretical perspectives.
Resulting TCs are prioritized to an SI orientation. For the few who
may wish to break out to staying open and finding relevant, emergent
TCs, the wrestle is often just too hard. to take on, especially for

Why? Because SI which is mixed in with QDA and then remodels
GT, as part of the takeover, to suit its data type (interaction), blinds
GT researchers to being sensitive to other theoretical perspective’s
TCs. Staying open to extricate GT and oneself from this replete
massive, possessive, takeover is too much to expect of the average
researcher. Especially if he/she wishes a career and publications
in the social structures that enforce this thought. Taking GT
outside the SI orbit is too contrary, even subversive. If fact there
is certainly nothing wrong with the SI perspective and its use in
QDA. They are ascendant to be sure in the literature. They are
just NOT characteristics of GT as a general, inductive method,
as SI devotees would co-opt it. The loss of this takeover is, of
course, a loss of tremendous theoretical power brought to GT by
discovering emergent TCs (from any field) that fit and integrate the
generated theory. Instead we often get from the takeover a retreat
to a descriptive, tiny topic research, devoid of but a few concepts
and lacking conceptual generality. It is then called GT which is
nothing more than a legitimating label. (See several journal issues of
Qualitative Health Research).

This condition makes the researcher feel like a part of the SI, QDA
takeover while it lets him off the hook of studying the GT method
to the extant its precise procedures of induction can be followed.
Impressionism creeps in place of careful constant comparison of
indices until their interchangeability reaches saturation. It changes
the use of literature to the degree of preconception of its relevance
and use, and it messes up the nature of memoing. It preconceives
theoretical sampling by its directing to, not letting emerge,
substantive and theoretical categories.

This takeover is so assumed, so automatic, so natural and so wrong
for GT. It is unscholarly and bias educated. It harps on accuracy,
verification, constructivism, interpretation and naturalism as binding
a GT product instead of conceptualization which is abstract of all
these SI/QDA concerns as I have written about at length l(See
GT Perspective I and II,2001,2003) . They are lost in meaning
making patterns based on evidence based impressions , not as GT
requires latent behavioral patterns arrived at by careful constant
comparisons. In doing so they force TC frameworks on data and
conceptualizations before hand,– usually a pet TC. The full range of
TCs from whatever perspective has no chance of emerging.

Again the prominent (heavy) QDA methodologists by citing the
prominent SI theorists (especially Blumer and Mead) hang tight to
GT as subordinate to SI’s possession. It is their worldview in the
center of a professional network., which kills curious transcendence.
GT looks for the latent patterns which explain what is going on as
people resolve their main concern, not the meaning, interpretive
patterns that are exchanged, which itself is a small part of the variety
of data that GT uses. SI’s possession of GT would have it as all the
data of GT, which is so wrong. And to be sure the consequence is to
close down the TC range available to integrate a generated theory.

Just because Awareness of Dying. Glaser and Strauss, which
gave rise to the “Discovery of GT”, dealt with qualitative data did
not make is an SI book. We discovered many latent patterns from
the interviews and observations that were simple fact that upon
comparison generated conceptual latent patterns (categories)
abstract of any joint interpretation and meaning analysis of
interviewer and respondent as SI would have it. Also the main
concern of the nursing staff — managing awareness of dying – was
an abstraction the nurses did not voice directly and clearly but it
was acted out all the time around dying people. SI data was but a
bear 25% of the study, if that. It is not 100% of all GT, which is an
overstatement the SI methodologists want to make as they wed it to
concerns of constructionism, accuracy, long guided interviews and
non-latent meaning patterns. Actually, SI accounts for only a small
part of the data that GT is used with.

SI is a natural type of data that occurs in interaction, but one can
question whether or not it is a discipline or perspective at all. But
this argument is for the espousers of it as a theoretical perspective..
Here I am just asserting that for GT “all is data”., no matter the
implicit theoretical perspective involved. And this openness allows
for the emergence of a full range of TCs. Latent patterns are
everywhere, in all types of data. GT is the method for their discovery
and conceptualization.

Taking the same type position as the SI methodologists, I could
just as easily say that GT is a social structural functional founded
methodology, or a systems founded method, or a cultural founded
method dealing with norms and values, or a social organization
founded method, a positivist founded method, etc, etc on and on,
And each would have its own ontology and epistemology. Every
grand scheme or perspective is implicit and sometimes explicit in
social action. And then off they run with an epistemology to justify
it. Fine, but GT is abstract of all this as it just depends on what data
and data mix is used to see the epistemology(s) involved. The open
GT researcher on choosing an area of interest will use whatever
data there is to generate his/her theory. He will let the main concern
emerge (the research problem) and generate a theory as “whatever”
level seems appropriate, whether middle range, grand, substantive
or formal. The credibility of his work is based on the careful
comparing of incidents to generate concepts that fit (with validity).
And his theory will be integrated by whatever TC emerges from the
fullest sensitivity he may have unstultified by SI.

For all the supposed positive functions to SI methodologists of
possessing and taking over GT, it just does not work. GT is simply a
general inductive method. The SI methodologists have in GT terms
not earned relevance for their assertion. They only have the vested
social fiction of socially structured power of their organizations:
institutes, departments, journals, careers etc. And it is no news to
sociologists that vested social fictions have great power.

The SI takeover of GT limits the data, the substantive categories and
subsequent TCs that the researcher will use or hopefully let emerge.
What constitutes appropriate data and what the SI researcher is
open to seeing in the data – how they construct meaning, means
that the researcher will miss much that upon constant comparison
will yield the latent patterns necessary to explain is going on. The
overwhelming SI perspective blinds the researcher to all the data
types, based on other perspectives, that are going on that are part
of it, in favor of just some meaning making. There are always social
structural features, social organization features, system features,
cultural features, economic features, etc, going on that will be
missed or slighted.

SI focuses on intense long interaction or conversational interviews,
often based on preconceived interview guides,. as meaning is
developed. The GT researcher listens to verbal quips, sees much
structural and cultural constraints etc and listens to participants
spilling their concerns, briefly or at length. He reads associated
documents , journals and newspaper articles, always comparing to
generate categories. SI drastically curtails theoretical sampling as a
framework and/or the problem is perceptively preconceived.. SI is
identity and legitimacy forming and in the bargain reduces sensitivity
to other data for research. The bargain of the SI takeover is not
good for GT as it stultifies it as a general method.

SI’s takeover of GT and it consequent data limiting then generates
complaints that SI is astructural or too empirical. MacDonald
says: “Grounded theory has an enduring respect for the empirical
world and the perspectives of the people being studied. But
because theory is linked so closely to empirical reality, Layder
argues that GT is limited to what can be observed or recorded
about human behavior and the action/interaction among people:
(op.cit. page 120)”. This is, of course, descriptive capture (that I
discussed at length in “The GT Perspective I”), which itself stultifies
conceptualization required for GT. And SI itself is very compatible
and useful in descriptive qualitative research. Thus SI’s takeover
of GT tends to remodel it from conceptual theory to descriptive
capture and its accuracy concerns., so the resulting product looks
very empirically descriptive, NOT conceptually abstract of time, place
and people. The result is that Layder will say, “The entire thrust of
GT is tied to the empirical world as it appears to our senses.(Layder
1989a)” This statement is based on reading studies of QDA barren
of conceptualization., and steeped in SI data., as if it was GT, and it
is not.

The SI perspective is very capturing as it is easy in comparison
to other more difficult perspectives such as structural-functional,
systems theory etc… It just requires making meaning out of action/
interaction which we all do readily in everyday life. Everything is seen
as interpretive, which is only one minimal type of qualitative data.
And SI’s tendency to result in preframed ,descriptive capture leading
to routine QDA, Not GT, means that TCs lose relevance, since they
only integrate theory, not description. They are not necessary for
description or conceptual description, except maybe to keep saying
“process” as a catch all term.

The self-fulfilling effect of SI on nursing research is replete: virtually
all routine researchers see their research work as SI. This capture
is very desensitizing to the researchers, so they see no TCs that
emerge that come from other related qualitative and quantitative
data. that comes from other perspectives. For example, variation in
work shifts, hospital nursing turnover, authority structures, nursing
career effects cultural ethics, etc. As Carolyn Weiner says (op.cit.,
page 8) “All this can only happen if one has a view of humans as
shaping their worlds to some extent – but in the face of inevitable
structural constraints.” I would take her comment further saying
humans are immercifully shaped by structural processes, cultural
norms, systems, etc, etc. All this is out of view to the SI devotee,
hence the emergent TCs that come with it. Again SI limits the data
for research, the TCs used if any, and when calling it GT, it denudes
GT from a general method to a SI possess method.

The socially structured vested fictions about SI in nursing research
will not change by my words, or the words of others. Choosing the
SI perspective allows peer reviewers to critique routinely research
with discipline dominance of department and journals. And the
critique will often include the conjectural, elaborated, logical think
up theory of the grand SI theorists. And GT is co-opted into this
critique. It is time for GT to be used for what it was generated: a
general method.

I only wish the GT researcher to stay general and open by studying
other perspectives and their TCs so as to be open to their kind of
data and emergence of the best fit TC form the fullest range. Central
to our understanding of GT IS NOT SI , I affirm strongly and correctly.
This is in contrast to the MacDonald, Schreiber statement: “central
to our understanding of GT is SI, a theoretical perspective rooted in
the philosophy of pragmatism.” (op. cit. page 42).

There is a growing dissatisfaction, however, with the limits that
SI puts on type data, data processing and by consequence the
TCs used if the research climbs out of description and becomes
conceptual. For example MacDonald (op cit page 122) makes
the avant-garde statement: “Nursing, however, and other health
disciplines, are moving away from an individualist perspective,
particularly with the increasing emphasis on the importance of
health promotion practice. More and More, health promotion
practitioners are becoming concerned with societal level concerns
and the way social structures and institutions influence health and
health behavior”. She also quotes Pendergast and Knotternerus
(op.cit., page 135) as “having suggested that SI must deal with
the astructural bias if it is not to be increasingly marginalized
in sociology. Health promotion insists that social and structural
influences on health must be taken into account.” These statements
apply to all SI oriented research, no matter the area or the problem.

There will be no threat to SI, if it advocates allow other types of data
(structural, cultural, system, organizational, etc ) to be used and
its consequent TCs to emerge. It will likely enrich the SI aspects
of a research even as it loses the possessive takeover. And in the
bargain it will release its possession of GT as a SI method, so GT
can be seen for what it is : a general , inductive method suitable to
all types of data , whether qualitative (baseline, proper line, vague,
interpreted or conceptual (see Doing GT, page 42) or quantitative.
And then the true inductive roots of GT in survey analysis can be
seen and used carefully in the GT procedures.

In short SI severally limits type data collection among all qualitative
data possible, with consequent lack of openness, lack of sensitivity,
and the lack of TCs. Releasing researcher from these limits will not
undermine SI. It will enrich its use combined with other data dealing
with other perspectives (e.g., nursing and management go on usually
in highly bureaucratic organizations which yield many structural,
cultural and system variables) Then GT is raised to its true general
level and the researcher can be open to all possible TCs. GT
IS NOT as MacDonald asserts (op cit page 116) “an interpretive
research methodology such as grounded theory.”

Students being trained in GT research for generating and emergent
theory need to be trained in the TCs of many fields so they are
open and sensitive to all data which may be involved in a theoretical
completeness of a GT. This will stop the incessant rhetorical
wrestle that tries to link GT with only one theoretical perspective
(SI or Systems) since a good GT will likely have data from many
theoretical perspectives, which data when compared result in the
conceptualization of latent patterns leading to an explanation of
continued resolving of the main concern in an area. There are many,
many TCs that possibly have emergent fit. The reader (teachers and
students alike) should study the 18 TC families I listed in “Theoretical
Sensitivity” and the subsequent ones I listed in “Doing GT”, as
a start to learning the theoretical perspectives from which they
derive, their roots in many other fields and then study TCs of other
theoretical perspectives and fields. The resulting sensitivity to a
fuller range of TCs will enhance seeing their possible emergence to
integrate a generated theory.
In closing this chapter and this book, I wish to talk briefly of the
results of GT research which I used as I have tried to explicate
above and in previous books. First of all I hope I have answered
Naomi Elliott’s question, a Ph.D. candidate at a School Nursing
in Dublin, Ireland, to me in Feb 2004 email – a question many
others have asked: “The reason I am writing you is to ask for
clarification on the philosophical and theoretical underpinnings of
GT methodology. Just recently I was challenged by a supervisor
who considers that an exploration of how SI underpins GT is critical
to any discussion on GT methodology. So my question is to what
extent does SI underpin GT. Does it do it alone or is GT also based
on other influences?” I trust this chapter and book answers this
frequent question.

Vivian Martin emailed me in Oct, 2003, her trouble in trying to force
SI limits on her Ph.D. dissertation research. She wrote: “ The big
TC that has knitted together my conceptualizing is that to think of
these interpenetrating processes as being part of an autopoietic
system. I do not have a deep grasp of systems theory, but I do
know that Purposive Attending is autopoietic, sort of a mix between
SI and structuration. A big problem with SI is that it does not deal
with structure effectively and it does not really tell how things come
to take on meaning.” TCs such as awareness contexts, boundary
work, normative culture and structural identity, helped he also handle
Purposive Attending. She settled on awareness structuration. SI
analysis could not be forced, as we see above that some would

I have been involved in many dissertations that go way beyond
SI by combining it with other types of qualitative data and other
TCs. Wendy Guthrie’s (University of Strathclyde, Scotland) superb
dissertation on Keeping Clients in Line, deals with four types of
ascendant relationship control of professional clients. Barry Gibson’s
( Ireland) dissertation on Cautionary Control gives a typology on
compliance with cautionary rules in dentistry regarding HIV patients.
Brene Brown’s (University of Texas, Houston) dissertation on
professional accompanying relationships in social work focuses on
staying relevant , binary deconstruction and binary retreat. Amy
Calvin, (University of Texas, Houston) received the best dissertation
award for her dissertation on personal preservation while dying
from a deteriorating disability. Berit Brinchman (Bodo, Norway)
dissertation of proximity ethics and patient vitality. Hans Thulesius,
(MD, Ph.D., Faculty of Medicine, Lund, Sweden) on balancing
cure vs. comfort care in Palliative care. Judith Holton,(University
College Northampton, England) dissertation handles the problem
of rehumanizing in the knowledge workplace by joining fluctuating
support networks. Maria De Hoyos Guarjardo (University Warwick)
has accomplished an excellent dissertation on solutioning in
problem solving in higher mathematics. Hans Lehman (university
of Christchurch, New Zealand) did his dissertation of the conflict
between undue control and utility and is structuring in the arena of
international information networks. Walter Fernandez (Queensland
University, New Zealand), did his thesis on Metateams in Major
Information Technology Projects. Alvita Nathaniel, (University of
West Virginia, Nursing) a professor of nursing, did her dissertation on
Moral Reckoning of personal values, professional ethics, norms and
organizational constraints for nurses in situational binds on wards.
Tom Anderson, (University of Manchester, England) amazed us with
his GT dissertation on the visualizing of patient deterioration based
on soft data on intensive care wards.

And there are many more GT dissertations and post doctorate work,
that have come my way. The point is that, as I said in “Doing GT”
chapter 3, the rhetorical wrestle is a waste of time regarding ontology
and epistemology. There are too many different types of data
involved, GT being possessed by no one theoretical perspective for
any data type. These authors have dealt with whatever is going on
in their areas of concern with the “all is data” principal in mind, while
doing laudable work as far as they can go. and being available to as
many TCs as their current studies have allowed. Teasing out one
particular perspective in these complex GT’s would be a waste of
time with futile results. Product proof is the goal for GT as a general
inductive, rigorous method.

So the answer to Schreiber and MacDonald is two-fold. Yes, I am
far more experienced than they are and their connection of SI to
GT is flawed. These two answers are to their paragraph (op. cit.,
179) “For us, the connection between SI and GT was obvious.
Nonetheless, we recognized our own inexperience relative to
Glaser and found forced to rethink our assumptions, as we face the
possibility that our beliefs about the relationship between SI and GT
could be flawed.” Instead of their immaculate 12 page super-rethink
on why GT is an SI method, they need only do a GT of the dozens
of extant GT dissertations, to realize my points made ad nauseam in
this chapter.

Works by Barney G. Glaser

The Discovery of Grounded Theory (1967) with Anselm L. Strauss
Theoretical Sensitivity (1978)
Basics of Grounded Theory: Emergence vs. Forcing (1992)
Doing Grounded Theory: Issues and Discussions (1998)
The Grounded Theory Perspective: Conceptualization Contrasted
with Description (2001)
The Grounded Theory Perspective II: Description’s Remodeling of
Grounded theory Methodology (2003)
The Grounded Theory Perspective III: Theoretical Coding (2005)

All books available through:

Sociology Press
P.O. Box 400
Mill Valley, CA, U.S.A.
(415) 388 8431 (phone)
(415) 381 2254 (fax)