Growing Open: The transition from QDA to Grounded Theory

Astrid Gynnild, Ph.D.

Abstract

Doing a PhD can principally be carried out in three ways;
firstly by applying existing theories on new data, secondly by
theoretically comparing existing theories and thirdly by
generating a new theory. Choice of approach of course depends on
awareness and accessibility of alternatives. In essence, most PhD
studies are exploratory journeys in a jungle of descriptive
methodologies based on very uniform data. In this paper, the
author elaborates the exploratory research process that
subconsciously, and later consciously, required a shift from the
initial QDA approach to grounded theory. The cutting point was
discovering the multifaceted implications of the all-is-data dictum
in GT.

Introduction

Data collection and data analysis is crucial for the way
research is conducted. It concerns research methods, research
settings, data sources, amounts of data collection and what to
look for in the data. The implications of ”all is data”, as
conceptualized by Barney Glaser, can therefore not be
overestimated. In practice, the ”all is data” statement brings us
right to the core of grounded theory methodology. Its power in
capturing change-in-process, which probably is the only steady
aspect of modern work life, is immense and incompatible with
any other methodology.

Like many other PhD candidates, I started out with a
qualitative approach intended to result in applying existing
theories on new data – and ended up with a grounded theory. The
area of study was news professionals in multimedia and crossmedia
companies in Norway, and how they coped with rapidly
changing conditions for work. Reflecting back on the exploratory
processing that lead to the sudden and definite switch in
methodology, it appears that the transition from QDA to a
grounded theory approach required a parallel process of growing
open.

After several months of concentrated qualitative efforts, I
had come to a point where I was unconsciously searching for a
methodology that could include a more diverse range of data
sources than the typical quantitative or qualitative approaches. It
was a troublesome period during which a main concern was loss
of time and lack of meaningful, productive progress in the study.
By the time, I did not know grounded theory. Consequently,
options for theory generation instead of descriptive verification of
existing theory were out of sight. In the ensuing paragraphs, I
will provide some of the reflections and questions that lead to the
transition from QDA to a grounded theory approach, followed by
a further elaboration of some all-is-data implications.

By the time of methodological shift, the data already
collected included hundreds of pages of statements illuminating
more facts and details than could possibly be handled in a highly
detailed, descriptive dissertation. My initial aim had been to
study the development of multimedia journalists in three large
Norwegian multimedia houses, descriptively comparing
similarities and differences. So far, all the data stemmed from
qualitative interviewing of news reporters in these news
corporations. The in-depth, semi-open interviews, as the genre is
called in qualitative research, were taped and transcribed
verbatim. Some of the interviews had been factor analyzed
according to q-methodological principles, a branch within
phenomenology. Two typologies had come out of analyzing the six
first interviews. Now the question was whether to continue on the
same track with the 14 next interviews.

Incubating

At this point, I had been through the preparing and
concentrating stages of exploratory processing, which is a basic
process in any kind of knowledge work. Now incubating was
reached, or rather, the chaos stage. I was in a state of confusion
and bewilderment, feeling overwhelmed by facts, data, and other
kinds of impressions waiting to be sorted. Therefore, I took a long
think break. The thought of another mathematical round of data
feeding based on forced choice did not contribute to raising my
levels of energy. Rather, it was accompanied by restlessness and
discomfort, a theoretical unrest. What was actually going on?
Was this the best way of handling the data? What data was there,
actually? The study had come to a crossroads. One option was
sticking to it; just continue with more qualitative interviews and
develop more “experimental designs”. The other option was a
more thorough analysis of existing data, combined with a search
for other analytic approaches within phenomenology. A third
option was to explore other theories and methods.

The period called for mental and physical withdrawal from
the PhD project. Later, when reading grounded theory, I realized
that it was a period of intensive subconscious processing. In this
phase, I was messing around “not doing proper work”. In reality, I
was waiting for the best idea or approach to become conscious. To
sort things out, I turned to reflection through wondering
(Grendstad 1986). A mind map was made, listing the issues that
up till then were registered on the broad topic of journalists and
journalism in multimedia settings. A number of thinkable empty
spots were uncovered, and finally a crucial question arose: What
is the potential of phenomenology in exploring further data in
this particular field?

When reading some of the interviews for a second or third
time, I was struck by certain patterns that repeatedly emerged.
Some traits appeared not to be restricted to the news corporations
under study; rather, they seemed to reflect more general patterns
in journalists’ everyday work. The impression became even more
evident as the interview data was supplemented with written
data from books, articles and relevant websites, and also,
informal conversations with journalists.

In some matters, patterns did not seem to depend on neither
structural conditions of work nor individual age. In other cases,
specifically structural conditions or age seemed to be at the fore. I
was repeatedly struck by the absence of specific issues when
interviewing news reporters. This issues-not-mentioned
phenomenon appeared to be a great puzzle. Work tasks and
questions taken into consideration, several issues that obviously
were on the respondents mind, simply lay “underneath” the
exposed levels of data.

Gradually, the awareness of other vague signs during
communication with interviewees grew. It was like tiptoeing on a
spiral; discovering one sign led to the discovery of the next. The
signs were exposed as restlessness, engagement, nonengagement,
vigorous contributing, active and passive resisting
and risk-taking, just to mention a few. They could be expressed in
a number of ways. For instance, when talking to journalists who
smoked, either at work or at leisure settings, their smoking
patterns became interesting to recall. When approaching issues
that were really on their minds, they instinctively picked up a
cigarette. At meetings in the newsroom, I recalled observing that
some people showed up physically every time, but rarely said
anything during the daily brainstorming sessions. Others always
managed, in some way or other, to position themselves at a
physical distance from the rest of the group. I consistently
wondered why, and also asked managers and the reporters
themselves. The problem was what to do with answers – since
they did not come out of any formalized interview situation.
Several times, journalists also called off-the-record after
interviews. They wanted to talk about things that were not easily
mediated in other people’s presence.

I was struck by the many layers of communication. One
aspect was body language, another aspect was verbal secrecy.
They both have to do with social exposure and social cover-ups,
with norms of expression and indirect communication. But where
could such observations fit in? In short, since a variety of social
and verbal signs emerged again and again, how should they be
treated? Would it be scientifically ethical to pretend that some
signs simply didn’t exist? If not, how could all this data be
integrated into the analysis?

While working on these thought experiments, it became clear
to me that clinging to the descriptive study of three multimedia
organizations would not only mean an overwhelming amount of
work; it would also limit the research scope and hence the
research results. The beginner approach had set me up for a
description capture based on very uniform data, a capture from
which I was now searching for ways to escape. What did I really
want to make a study of? At this point, the dilemma of
generalizability still seemed insolvable. How could the general
patterns I was on track of be mediated? Phenomenological
methods, at least q-methodology, appeared to be too narrow and
too rigid for the field of interest, so what could be done? Should
the q-findings simply be left behind in order to carry out a restart?

Eureka

The dilemma was presented to a friend working in another
academic field. He asked whether I had heard of grounded theory.
His short message was “In grounded theory you go back and forth
between data collection, analysis and sorting. Data decides where
to go next and you learn how to conceptualize. It’s all concerned
with data, what data to look for and how they are to be handled.
All is data, you know. Just start reading Glaser’s books.”

At the library, I opened a book where Glaser explains “all is
data” this way: “exactly what is going on in the research scene is
the data, whatever the source, whether interview, observations,
documents. It is not just what is being, how it is being and the
conditions of its being told, but all the data surrounding what is
being told (Glaser 2001, p. 145). Exactly this paragraph
contained all the information that I needed to overcome the data
overwhelm and theoretical restlessness manifested during the
incubating period.

In reality, by switching to a grounded theory study all the
data that so far did not “fit in” were usable and could be
incorporated just like other kinds of relevant data. Eureka stage
was finally reached. Eureka is known as the moment of discovery,
the moment when new insight breaks through. It is a mental
state associated with high spirits, exhilaration, relief, glow, and
energy. I experienced that when eureka moment is reached, the
rest of the task is done with more ease, since energy arousal is a
physical sign that one is intuitively heading in the right direction.
I was on my way to a grounded theory approach, which in
practice meant that I was heading towards the two last stages of
exploratory processing, namely elaborating and presenting.

Since a considerable amount of data was gathered before the
switch from QDA to grounded theory, the process of integrating
the data into the GT approach needs to be explained. First, data
sources in relation to the ‘all is data’ concept will therefore be
elaborated, followed by some practical aspects of the
interrelationship between data, research settings and research
methods.

Reworking Data

The switch to grounded theory methodology yielded a total
reworking of data and the preliminary draft. The existing slices of
data, such as the interviews and the many field observations that
up to this point had only been stored in my own mind, provided a
rather confusing picture of the research scene. It was a great
relief to find out that in ‘The Discovery of Grounded Theory’, it is
emphasized that a great variety of sources contributes to building
a dense and rich theory. Different kinds of data, or slices of data,
allow a multifaceted investigation of the research area.

The possible integration of all kinds of data made it clear
that all the data collected could be valuable in the generation of a
grounded theory. All the interviews, the q-sample and the
unwritten observations and questions could be good guides to
further work; they were simply different kinds of data. After
reading more about grounded theory, the first step in the “new”
data analysis was substantive coding of the verbatim interviews.
The systematic coding actually revealed what I had intuitively
sensed before switching to grounded theory: Much of the same
data appeared again and again in various facets.

I realized that now the initial “walking survey” tendencies
were grounded in a systematic data analysis. Specifically, this
systematic analysis and continuous grounding in a wide array of
relevant data is a fundament that separates conceptual research
from, for instance, journalism. In several of his books, Barney
Glaser points to the fact that with growing experience, most of us
are “walking surveys”; the missing link is the systematic analysis
of inherent data.

As a researcher, I now experienced that the most important
question one should continuously ask is: Where do I obtain the
most relevant data, and where should I go next? Which groups
and subgroups need to be visited now (Glaser and Strauss 1967)?
In practice, such theoretical sensitivity requires the analyst to be
constantly on the alert as to what emerges from the data, and
flexible enough to switch from one kind of data collection to
another as it becomes necessary. The analyst is continuously
challenged by the fact that the data decides: “As he collects data
his job is to deal with exactly what is happening, not what he
would want to happen, not what his own interest would wish the
data to be. The data is not ‘truth’ it is not ‘reality’. It is exactly
what is happening. The GT researcher has to be oriented to each
course of action having its own meaning. And once the GT
researcher lets this meaning emerge and sees the pattern, he/she
will feel ‘sure’ that this is what is going on. This sureness cannot
be known beforehand. It emerges conceptually through constant
comparison.” (ibid. p.146)

During the study, I experienced a growth in the awareness of
what was going on in the empirical field. Glaser points out how
theoretical sensitivity is used to uncover data that otherwise
might be overlooked: “Grounded theory is based on the systematic
generating of theory from data that itself is systematically
obtained from social research. Generating theory and doing social
research are two parts of the same process. How the analyst
enters the field to collect the data, his method of collection and
codification of the data, his integrating of the categories,
generating memos and constructing theory – the full continuum
of both the processes of generating theory and of social research –
are all guided and integrated by the emerging theory.” (Glaser
1978, p. 2)

A couple of times I was really tempted to “cover up” some
strategies that news people frequently turn to, both at an
individual and a structural level. The cover-up concept, as
developed by Argyris (1986), points to a widespread human
defense mechanism. When people feel embarrassed or
threatened, the tendency is to oversee the phenomenon that
causes such feelings. Both individuals and groups cover up.
According to Argyris, this simple fact accounts for much of the
counterproductive actions in organizations. For instance, if a
manager has set unrealistic time limits for an investigative
journalism project, it is likely that one or more journalists
involved will engage in strategies to cover up the manager’s
mistake so as not to embarrass him. They will therefore pretend
not to notice.

Like journalists, sociologists are consistently challenged by
cover-up actions taking place in various settings. However,
during the theory building, I realized that with grounded theory,
it is not possible to build a dense and credible theory if you are
not totally honest about your findings (Glaser 2001, 2006). So I
had to force myself to accept some of the data, although I was
quite astonished and actually did not personally welcome what
emerged.

Shortly before revising and restructuring the drafts into the
final version, most of the collected data was reviewed for a third
time. The review resulted in quite a few, new concepts. From this
experience I was reminded of the old dictum that ‘you have to
learn to see in order to see, and you have to learn to hear in order
to hear’ – and it all takes considerable amounts of focused time to
think and structured manual work. However, when carrying out
grounded theory, these restructurings are extremely energizing
and personally developing. Restructurings help the analyst to
uncover higher levels of abstraction in the data. This in turn is
practical evidence of how easily grounded theory can be modified
at any time. It connects grounded theory with the roots of
cognitive processing, the very essence of empirical research.

Memoing as Parallel Processing

A process that took place parallel to the open coding of
interview data was the writing of memos. In the beginning, large
piles of memos were written; it was like a stream that had been
waiting to be released for a long time. The first memos were ideas
that came to my mind while coding the interviews. The next
memos were based on visual observations in the field, data which
so far had not fitted in anywhere in the study. Notes and
reflections were jotted down, for instance respondent comments
before and after the “official” interviews, gestures at meetings
and discussion topics during smoke breaks.

During the whole process of generating the theory, memos
have served as notes to myself on ideas and concepts and their
relationships. Ideas are like cats; suddenly they are everywhere,
and then they are gone and they don’t care about where you are,
what you are working on or whether it is day or night. So you
have to seize them at once. The good thing about ideas is that as
soon as you get them down on a piece of paper, they will not
vanish but are accessible for later analysis.

Memo writing is discussed here because it proved to be a
necessary tool in grasping several types of data and then keeping
the ideas for further analysis. Before ideas come to the surface
and can be stored in memos there is always a period of what
Glaser has termed preconscious processing. The flexibility of
memoing allows the analyst’s pre-consciousness to work on an
idea as long as necessary, although any emerging idea can be
taken care of when it pops up, irrespective of working hours or
other structural conditions. The first round of constantly
comparing data slices from interviews, observations, informal
talks and reading periodicals for journalists, journals,
newspapers, analytical articles and other relevant literature
clearly suggested that the interviews alone provided rather
uniform data. The division between interpreted or properline
data, baseline data and vague data in grounded theory is of great
value in understanding how uniform data can limit the
generation of a dense and rich theory.

Properline data tends to be the easiest accessible data,
particularly within preset research frames such as taped
interviewing. During interviews, people often say what they think
they are expected to say. Sources tend to give the analyst either
interpreted information or information they think is appropriate
to the situation. Baseline data refers to data gathered when
sources are more relaxed and do not have to worry about, for
instance, colleagues or managers, but feel free to express what is
really on their minds. Unexpected off-the-record phone calls that
I received after interviewing are examples of baseline data.

Observations of non-verbal communication, such as body
language, positioning at meetings, informal group divisions and
the like are examples of vague data. Vague data cannot bear a
theory alone, but it can contribute to an initial foothold on
theoretical sampling and where to go next. The various layers of
relevant data help the analyst to achieve as much diversity in the
emergent categories as possible and ensure that the hypotheses
are firmly grounded.

In parallel with data gathering in the field, I started
systematic readings of journalism magazines and of literature
that seemed relevant to the issues involved. The memo processing
speeded up. Several theoretical outlines based on the analyzed
data were made, but the outlines always stopped at some point.
There were holes in the data that could only be saturated by more
theoretical sampling.

Research Methods and Settings

The discussion of variety in data collection leads us to the
next issue on which we need to shed light, namely research
methods and research settings. As elaborated above, where to go
and subsequently where to go next are basic questions in
theoretical sampling. Grounded theory’s applicability to single
units as well as to any number of units makes it possible to
search for data in any relevant accessible setting. And, depending
on accessibility, a multiplicity of data collection methods can also
be used. The point is to be as flexible as possible in accordance
with variations in structural conditions (Glaser and Strauss
1967). In this particular study, methods and settings for data
gathering are so closely interrelated that they need to be
elaborated together.

The qualitative interviews at the beginning of the research
mentioned above were conducted in the newsrooms. To gain
admission to the newsrooms the research project was introduced
by e-mails to the editors. The interviewees were selected and
asked whether they wanted to participate before I entered the
newsroom. I suggested interviewing people who held differing
views, or at least were of different age and gender and possessing
diverse competences. Some names were provided by managers,
some by journalists known beforehand, and the final decision on
whom to interview was made by me.

Flexible Approaches

In the grounded theory phase of the study, a variety of
approaches were tested out to get as close to news reporters and
their daily concerns as possible. The abandonment of the research
unit meant that relevant data could be gathered in any type of
newsroom. It introduced a hectic period of moving in and out of
newsrooms during which small and large broadcast media were
visited and also online papers and newspapers. A variety of faceto-
face informal talks and newsroom observations were the
methods mostly used. Some talks could last for five minutes,
others for more than an hour.

A phone call or two to editors or managers was usually
enough to get free access to observing and talking to news
reporters in their daily surroundings in the organizations.
Instead of agreeing on dates beforehand, many visits to
newsrooms were made just to be “a fly on the wall”. The
observations proved important to reveal empty spots in the
understanding of what was going on. Such issues were pursued
by talking informally to people, in open office landscapes as well
as in individual offices.

I also joined lunches with journalists individually or in
groups, and of course coffee breaks and other intermissions.
Wherever and whenever they had time to talk, it was okay with
me. Some people were interviewed by phone, others at cafes
before or after work, and sometimes the conversations took place
at bars in the evening and at night. There was no longer any need
for accurate description; what I needed was data that could help
to conceptualize patterns of behavior among the people
concerned. Consequently, there was no longer a dependency of
either notebooks or tapes. As long as a sheet of paper or a napkin
was within reach, the memos that were needed could be written.
The liberation from accurate description thinking obviously has a
favorable practical aspect. It frees the analyst to handle larger
amounts of data, and data of all kinds, without experiencing data
overwhelm.

A few times during theoretical sampling, I arranged
particular dates with multivariate groups of news reporters. The
topic that I wanted the respondents to reflect on was introduced,
for instance journalistic creativity. The principle of the “pedagogic
sun” (Grendstad 1986) was used as a guide to reflections in
writing, followed by a plenary discussion on each news reporter’s
contributions.

I continuously switched between theoretical sampling,
memoing, coding, constant comparison and more theoretical
sampling. As the piles of memos grew, the conceptual sorting
became more complex. The memos were re-sorted many times
during the process; as more and more data was accumulated, the
categories and properties that emerged early in the process
needed to be modified. In grounded theory, modifications are
usually necessary for rich and dense theory generation. The goal
is always to conceptualize empirical data through constant
comparison of a variety of data.

Since grounded theory is conceptual, the interrelated set of
hypotheses that constitutes the theory is independent of time,
space and people. Yet it is emphasized that the theory should be
grounded in all possibly relevant data. The emphasis on grounded
generation rather than verification means that the only testing
possible is experienced relevance and fit in the substantive area.
In other words: A grounded theory is only good as far as it goes in
explaining what’s going on in an area (Glaser 1996). The
statement implies that the generation of grounded theory is both
a very abstract and a very practical task, and it is always possible
to keep correcting, or rather modifying, the categories with more
relevant properties.

Illustrations

During the write-up stage of the study I experienced that in
conceptual theorization, which is by nature an abstract piece of
writing, it is not easy to find the right balance between
conceptualization and its illustrations. The first draft of the
initial chapters, based on the sorting of conceptual memos, was
written very theoretically and with a minimum of illustrations.
Feedback from layman co-readers suggested that the theory
would be more easily accessible if more illustrations were added.
The accessibility aspect is of course very central. A stated goal in
grounded theory is that it should be found useful by laymen as
well as by experts. The elimination of all illustrations might
make the theory very dense and accessible only to theoretical
insiders. But also for theoretical insiders, it might be more
difficult to understand than necessary. The point is that a lack of
illustrations undoubtedly reduces the size of the audience who
might find the theory useful.

In reality, the richness and denseness of a generated theory
depends to a large extent on successful switching between
abstractions and concrete illustrations. Another discovery made
when working and reworking the drafts was that the more
extroverted the research process became, the more options for
illustrations to earn a place in the theoretical outline. I realized
that illustrations provide the reader with conceptual breaks.
They add a data dimension which makes the theory more
meaningful simply because illustrations are what they are,
namely slices of empirical data which ground concepts in concrete
facts. But illustrations are not examples from reality intended to
prove that the theory is correct.

Barney Glaser explains it like this: “The credibility of the
theory should be won by its integration, relevance and
workability, not by illustration used as if it were proof. The
assumption of the reader, he should be advised, is that all
concepts are grounded and that this massive grounding effort
could not be shown in a writing. Also that as grounded they are
not proven; they are only suggested. The theory is an integrated
set of hypotheses, not of findings. Proofs are not the point.”
(Glaser 1978, p. 134)

Problems with Method Mix

After this discussion of the practical application of grounded
theory methodology, I will share some other cognitive discoveries
made during the transition from phenomenological to a grounded
theory approach. Theoretically, phenomenological research can
reach the same levels of abstraction as a grounded theory. A
dilemma is that phenomenology, with its emphasis on narratives
and rich descriptions, invites the researcher to stay at the
description level. No matter conceptualizations that come out of
phenomenological studies, they are not systematically grounded
in the data nor constantly compared and coded. Additionally,
narratives and rich descriptions are the basis for
phenomenological research.

In the transition phase from QDA to GT, a question that
started buzzing around in my head was: How can
phenomenology, concerned as it is with individual experiences in
time and space, be successfully paired with the abstracts of
grounded theory? Once again I had to reflect on my main concern
in this dissertation. How did I want to generate data and how did
I want to display my findings? “Researchers not clear on the
distinction between conceptual and descriptive get easily
confused on whether the theory describes a unit or conceptualizes
a process within it,” as Glaser writes (Glaser 2001, p 15).

One of the most important aspects of conceptualization is
that concepts last forever, whereas descriptions are tied to time
and people and are only of value within a concrete setting. When
reflecting on these topics, the phenomenological philosophy
appeared strong and clear, whereas phenomenology as a research
method appeared surprisingly unclear compared to grounded
theory. In the area of description versus conceptualization, the
paths became blurred.

After reading Barney Glaser’s book on descriptive
remodeling of grounded theory (Glaser 2003), it became obvious
that a method mix would not work out very well. Mixing QDA
with grounded theory would most probably downgrade the goal of
conceptual theory into a remodeled version of a qualitative
descriptive version. While putting these arguments forward,
however, I am perfectly aware that I have tested out only a tiny
part of phenomenological research approaches. There are
certainly a number of other ways to go both within
phenomenology and qualitative methodologies as a whole that
could have been profitable in a study like this. But it would have
ended up as a different type of study.

Contrary to QDA, grounded theory is basically free of
epistemological categorizations. While methodology researchers
have tried to classify grounded theory as symbolic interactionism,
Barney Glaser himself resists all attempts at labeling grounded
theory as part of any “ism”. He points out that grounded theory is
free of ties to any theory of science: Since grounded theory is a
hypothesis of the interrelationship between a set of categories, it
does not deal with philosophical conceptions of what is “truth”
(Glaser 2004).

Researchers start with an area of interest, but with no
preconceived view of what problems they will study or how the
participants deal with the problems. They are open to what may
occur. In grounded theory, the analysts “let the problems and
their continued resolving emerge. They trust the fact that the
world goes on whether or not they know how and the research
issue is to discover a core variable and ensuing theory that
accounts for what they are finding is the main concern of the
participants.” (Glaser 1996, p. xiii)

Now the reader will probably assert that in this particular
study, the analyst could not be totally open and ignorant about
the area of interest, since nearly twenty qualitative interviews
and a preliminary q-sample were carried out before grounded
theory was brought into the research arena. True, a lot of data
was collected and partly analyzed. However, the uniform data
collection alone suggests that the research frame was still wide
and unfocused and that I had little systematic empirical
knowledge about news reporters’ main concerns.

Use of Literature

During the study, literature has been used in three ways:
firstly as a preview to sensitize the analyst to the research
domain, secondly to provide a theoretical background to the
theory, and thirdly as data during theory generation. Some of the
references included in the introductory part of the dissertation
are integrated into the theory of creative cycling; others only
provide a historical backdrop with respect to news reporters’
position in media research during the last decades. An overall
principle has been to refer to relevant literature successively,
instead of devoting a particular chapter to a literature review.
According to Glaser (2001), when generating a grounded theory,
existing theory and other theory should be treated like any other
data, and it should only be used to the extent that it earns its way
into the developing theory.

The best way to learn about grounded theory and its
implications is by reading the original literature. Whenever I
have had a problem, I have turned to Glaser’s books and found
the answers and explanations that were needed. Since all his
books are thoroughly grounded in data, the information gives a
feeling of “deja-vu” which is instructive and energizing.

A principle applied throughout the study has been to check
the original sources of grounded theory first, and to keep this
information in mind when subsequently reading other
researchers’ writings on grounded theory. This way of exploring
grounded theory theoretically has provided invaluable insight
into analyst accuracy variables and also, the troublesome path
methodology researchers start on when their data is not well
grounded. Moreover, as a former journalist, I have experienced
that misunderstandings and misinterpretations can easily start
to live a life of their own if you trust second-hand sources too
much. One of grounded theory’s great advantages is that the
methodology is thoroughly explained from start to end in the
original literature.

As for integrating other literature into the theory, I have
found it fruitful to seek out outstanding works in fields as diverse
as organizational theory, healthcare, personal development,
psychology and scientific theory. The wide variety of impulses
provided by these sources has been more than necessary in the
theory generation process. Data from the journalistic domain has
been constantly compared to relevant data from other domains of
working life, and it has undoubtedly made the theory richer and
more general.

The strategy is recommended in Theoretical Sensitivity
(Glaser, 1978) because it “maximizes the avoidance of preempting,
preconceived concepts which may easily detract from
the input and the drugless trip (aha moments, eureka, authors’
note). It is hard enough to generate ones own ideas without the
‘rich’ derailment provided by the literature in the same field. The
analyst should not worry about coverage in the same field since
this literature will always be there.” (p.31)

Self Pacing

The process of finding the right balance between theoretical
sampling in the field and the reading of literature took time. It
challenged me to test out a variety of working methods and being
extremely flexible when planning the daily work. When
concentrating only on field data for longer periods of time, I
became locked in my own thinking. And vice versa; if I became
caught up in reading other theorists’ works, my own analytical
process was blocked. Sometimes I just became overwhelmed by
the endless amount of relevant research, and really had to fight
to get sufficiently grounded in empirical data to be able to
proceed.

The testing out exemplifies the necessity of theoretical
pacing, self-pacing and the development of a personal pacing plan
when generating grounded theory (Glaser 1978). Grounded
theory not only requires joint action when collecting, coding and
analyzing data. It requires that the analyst knows his own
temporal pacing and manages to develop a personal plan that
takes his research into consideration as well as his temperament
and private life.

As a methodology, grounded theory provides the analyst
considerable autonomy and freedom to pursue his own study
under a great variety of structural conditions. The experience
during this particular study is that theory generation is such an
absorbing and time-consuming project that it needs to be well
paired with other aspects of the analyst’s daily life in order to
work out. Grounded theory takes the time it takes, and it is hard
to make an accurate estimate of the time needed. Restrictions on
outer frames must of course always be taken into consideration.
But within such frames the analyst is dependent upon finding his
own plan so that he can establish realistic deadlines and make
continuous progress.

Since grounded theory is above all what Glaser terms a
delayed action phenomenon, I experienced that it is very
important to set aside enough time for subconscious processing.
When data is sampled, coded and analyzed, memoing is the
helper that attends immediate to all kinds of ideas that might
arise as a result of the previous work. Sometimes a concept
appears several months or even years after the analyst started
working on it. At other times, conceptualization just goes on and
on. The essence is that these creative aspects of theory generation
demand analytic flexibility and trust in emergence in order to
handle the outside world’s expectations. Taking breaks, doing
things other than studying, developing rituals to enable you to
continue when you get stuck in your own thinking – these are all
aspects of self pacing which I have found necessary during the
PhD process.

For instance, the first three or four categories of the theory
emerged very quickly, and the thought was that finding the core
category would happen just as fast. However, it took many
rounds of theoretical outlining and a long period of subconscious
processing before the interrelationship between the core and its
categories was discovered.

After many months of structured work and intensive
subconscious processing, I woke up at four o’clock one morning
and knew that the core category had emerged. I jumped out of
bed, picked a pen and a notepad and wrote memos continuously
for several hours. The moment was extremely energizing. It sent
her on a drugless trip that lasted for a long period of time and
gave me the confidence and power that I needed in order to
continue with theoretical sampling, coding and analysis of
enough data to generate the theory of creative cycling.

Author

Astrid Gynnild, Ph.D.
University of Bergen
Krohnengsgt. 10, N-5031 Bergen
Phone: + 4755560495/+4747028524
E-mail: agynnild@start.no

References

Argyris, Chris (1990): Overcoming Organizational Defenses.
Facilitating Organizational Learning. Boston, Allyn and
Bacon

Glaser, Barney G. (1996): Gerund Grounded Theory: The Basic
Social Process Dissertation. Mill Valley, Sociology Press

Glaser, Barney G. (1978): Theoretical Sensitivity. Mill Valley,
Sociology Press

Glaser, Barney G. (2001): The Grounded Theory Perspective:
Conceptualization Contrasted with Description. Mill
Valley, Sociology Press

Glaser, Barney G. (2003): The Grounded Perspective II.
Description’s Remodeling of Grounded Theory
Methodology. Mill Valley, Sociology Press

Glaser, Barney G. (2004): The Grounded Perspective III:
Theoretical Coding. Mill Valley, Sociology Press

Glaser, Barney G. and Strauss, Anselm (1967): The Discovery of
Grounded Theory. Strategies for Qualitative Research.
Hawthorne New York, Aldine

Grendstad, Nils Magnar (1999): Å lære er å oppdage. Oslo,
Didakta