Making Sense of
Qualitative Data: The
Grounded Theory
Approach to
Discovery
Karen Locke, Ph D
W. Brooks George Professor
Mason School of Business
College of William and Mary
Context
• What is the situation in which
we come to be participating in
a workshop on grounded
theory today?
Qualitative Research
in management &
Organization studies
• Prosecuted within an
increasingly differentiated
institutional context
• Without an infrastructure of
courses and apprenticeships
to match interest
• An expanding “umbrella
domain”
Where is the
grounded
theory
tradition
today
7,640,000
Contest &
evolution
GROUNDED THEORY IN
MANAGEMENT & ORG.
STUDIES TODAY
• GT as a signifier typifying non hypotheticodeductive research (the theory building)??
• GT as a ceremonial citation masking “common
sensing” your way through data??
• GT as site of contested intellectual property
rights
• GT as part of a new landscape of
methodological multilingualism??
• GT as part of researchers’ individually shaped
methodological vocabularies and practices??
• GT as one vocabulary and set of practical
heuristics evident in “discovery oriented”
research practice.
Workshop
Objectives
• Understand grounded theory’s
intellectual tradition
• Understand the general contours of
and logic underlying this foundational
tradition in qualitative research
• Learn and experience the practice of
its operational procedures (e.g.
theoretical sampling, category
development, constant comparison,
memoing etc)
• Appreciate their contribution to
“discovery”
Method Separates
From “Discovery”
• “Discovery concerns the origin,
creation, genesis, and invention of
scientific theories and hypotheses.
Justification concerns their
evaluation, test, defense, success,
truth, and confirmation. Discovery
is for description alone, for
psychology and the history of the
sociology of science. Justification,
however, is for the philosophy of
science and epistemology “
(Kordig, 1978; p. 110).
Yet… advantages of
discovery oriented
qualitative approaches?
• Allow the “real” world of work to
inform, press against and shape
theorizing (urges theoretical open
mindedness)
• Can capture complexity through
multi-faced accounts of action
• Identify the process or “how”
discovered outcomes are generated
• Serendipitous findings can spur new
research
…advantages of
discovery oriented
qualitative approaches?
• Investigate how participants
make sense of events and
behaviors at work
• Identify how local meanings
influence behavior
• Explore in detail how context
shapes meaning, experience, and
behavior -link to practice
• Investigate new developments in
the world of work
Discovery
Achievements
• Documenting Evolution
and Change
• Challenging Rational
Perspectives on Workplace
Behavior
• Troubling Existing
Theoretical Conceptions
(Locke, 2011 Academy of Management Annals)
Discovery Necessitates A
Recursive, Contingent
Research Design
• “Funnel Shaped” (Hammersley &
Atkinson, 1983)
• “Garbage Can II” (Grady and Wallston,
1988)
• “An iterative process that involves
“tacking” back and forth between
implications of components of
design” (Maxwell,2005)
Implications
• “The complete analysis isn’t” (Michael
Quinn Patton, 2002)
• Flexibility in research question, data
gathering, theoretical formulations
• The experience of ambiguity and instability
in the process
An Interactive
/Recursive Model of
Research Design
Purposes
Conceptual
Context
Research
Questions
Methods
Quality
(Maxwell, J. 2005. Qualitative Research Design. Sage)
Openness &
Emergence
Embarking on a project without
knowing where we will end up. That
is being unsure
• As to what our specific research
question will be (Barley, 1990)
• What data we will have gathered
(Van Maanen, 1998)
• What the project will end up being
‘a case of ’ (Becker/ Ragin, 1992)
• What concepts we will have
developed (Agar 2006)
Observed
Particulars
Dave is research scientist who owns
pants with duct-taped pockets to
accommodate a set of forty-odd keys
carried on two separate rings
connected by a spring clip.
In the 4 years since Dave configured his
keys in this way, “the only times Dave
has actually used the spring clip feature to
separate his keys were to temporarily loan
out one side or the other, take a leisure trip
to Europe, and once while he moved house”
(Nippert-Eng, 1996; p. 49).
Karen Locke Ph D College of
William and Mary
Conjectural
Work
“There is nothing but imagination that can
ever supply [us] an inkling of the truth.
[We] can stare stupidly at phenomena; but
in the absence of imagination they will not
connect themselves together in any rational
way” (C. S. Peirce CP 1:46)
The ability to engage the confounding
array of talk, action, feelings, images,
and texts that is field data. And, in the
context of this engagement, [without
recourse to a prior theoretical
specification] to imagine processes,
structures, or characterizations such
that were they operative, they would
render the data intelligible.
GT practices
drive this
interplay
Observed
Particulars
Ideas
Existing
theoretical
elements
The Grounded
Theory
Approach
An approach to qualitative analysis
that develops abstract
conceptualizations through
microscopic data analysis.
Exercise: For
your selected
study…
• What “phenomenon” was the
researcher/s interested in exploring –
what was the study’s purpose?
• What did they do – what was their
research design?
• What references to GT did they make?
• What practices did these point to?
• How well did the study’s design and
data gathering approach access the
phenomenon?
• What did the study yield? What
categories? What theoretical narrative?
What practices
make a theory
grounded?
• approach your study leary of the ways in which
the phenomena has been framed and explained
• organize your data gathering so that you can
access in detail the perspectives of the actors
involved in the phenomenon (how they
understand their world, what they are doing and
why)
• sample theoretically
• Conceptualize and code microscopically
• analyze the data systematically and comparatively
• continually enrich your thinking by reading widely
around the phenomena
• grow your theory by going back and forth
between progressively more focused data and
successively more abstract
conceptualizations of them
• integrate the emergent categories around a
narrative
Analysis: “Heads
Up”
• Assigning meaning to unstructured and
ambiguous data, within a flexible recursive
process
• Begin analysis as you gather data NOT
after
• Progressively narrow to one process
within the data (this will not be linear!)
• Narrow to one part of that process
• Data gathering and analysis are
iterative (expect analysis to impact data
collection)
THE DEVELOPMENT OF
CONCEPTUAL CATEGORIES
(c.f. Grounded Theory)
Theoretically
Driven Data
Sampling
Fracture, Chunk and
Name Data
Compare, Compile
& Rename
Data Fragments
Generate Provisional
Conceptualizations
with their Indicators
Develop Robust Working
Conceptualizations
Process is “finished” when core categories are
theoretically saturated – (in theory)
Character and
Experience of
Grounded Theory Data
Sense Making Process
• Creative opened-ended
dimension
• Mechanical procedural
dimension
• Individual analyst is central
agent
• Meaning making is richer
when the process is a social
one
I. Sampling in
Grounded
Theory
Information not statistical
generalization
NOT random … never, never, never,
never …
BTW, did I mention sampling is never
random?
Purposive & Theoretical
Sampling
-Driven by Information Needs
• The logic of theoretical or purposive
sampling is that you select units which will
provide you rich information relative to your
orienting research questions
• Theoretical sampling to expand, check and
refine conceptual categories. Conceptuallydriven sequential sampling- usually not
wholly pre-specified, but can evolve once
fieldwork begins.
• Sampling reflects researcher interests in the
evolving fit between gathered data and the
emerging theory. You sample for the purpose
of developing your emerging theory.
Ensuring Sampling
Focus: Clarifying units of
Social Organization
Expressing Our Phenomena
• Social Practices – recurring categories of
normal talk or action
• Episodes – significant events in the life of…
• Encounters – 2 or more persons mutually
involved
• Roles – categories of person (formal roles,
informal roles, social types)
• Relationships – parties interacting over period
of time, viewing themselves as connected in
some way (vary in a gazillion ways,
interdependence, power, trust, information
regarding each other, etc)
• Groups
• Organizations – goal pursuing entities with
formal and informal strategies
• Processes
• May have interpretive, emotional as well as
agency aspects
Data Forms
• Interview transcripts – e.g. accounts of people’s
experiences
• Field notes – observations of what people do,
say, and understand
• Documentary evidence – policy documents,
texts, memos, emails, accident reports, etc.
• Visual elements, e.g. drawings, photographs,
graphics, etc.
(We end up with a large amount of
unstructured and uncategorized data)
Ensuring Data
Quality
• Is it the right data for the job?
• Does the data fit with your phenomenon?
• Is it “thick” enough?
• Do you have a detailed record of what those
you are researching, do, see, hear, experience
and understand
• Do you have examples, histories, stories,
explanations
• Do you only have “opinions” … I think
Creating “Data”
and “Making
Meaning”
•Interactive, Recursive, and
Interdependent
•Relying on multiple “meaning
making” activities
Articulating What
Your Data Might
Mean
• Meaning Making Activities
• Developing contact summaries
• Creating data units
• Creating ideas about /Inventively naming data
chunks (developing categories)
• Conceptually developing categories through
constant comparison
• Analytic Memoing
• Creating an overarching organization for your
data (Action strategy models,
conditions/consequences models, stage models,
typologies, etc.)
• Framing theoretical implications – contributing to
the literature
A Different Analytic
Procedure: Domain
Analysis
DOMAIN
Professor
(Cover Term)
Is a kind of
Semantic Relationship
Motivator
Decision maker
Included Terms
Spradley, J. (1980) Participant Observation
Other Potential
Semantic
Relationships
• Spatial X is a place in Y / X is a part of Y
• Cause-effect X is a result of Y
• Rationale X is a reason for doing Y
• Location for action X is a place for doing Y
• Function X is used for Y
• Means-end X is a way to Y
• Sequence X is a step (stage) in Y
• Attribution X is a characteristic of Y
Articulating
What Your Data
Might Mean
How does identity expanding
learning (learning that
transforms who you are) occur?
II. Contact
Summaries
Begin actively thinking about data
Create a general indexing system
Miles & Huberman 1994
Contact Summary-How does
identity expanding learning occur?
• What people, events or situations were
involved?
• What were the main themes or issues?
• What research questions did the contact bear
most closely on?
• What new speculations or guesses were
suggested by the contact?
• Where should you place most energy during the
next contact, what sorts of information should
be collected?
[Miles & Huberman, 1994]
III. Chunking
Observations to
create “data”
How do you take streams of
observations (field notes, archival
documents, transcripts) and turn
them into little pieces of data that
can be worked with analytically?
The Development of Conceptual Categories
(c.f. Grounded Theory)
Theoretically
Driven Data
Sampling
Fracture, Chunk and
Name Data
Compare, Compile
& Rename
Data Fragments
Generate Provisional
Conceptualizations
with their Indicators
Develop Robust Working
Conceptualizations
Process is “finished” when core categories are
theoretically saturated – (theoretically)
How would you
chunk this data?
SUBJECT: So right away, I know I'm not
3
making a decision, I'm making a
4
recommendation which is always easy to waffle
5
on, apologize for, or mitigate.
6
This also allows me to make, sort of –
7
I won't feel as constrained. I can probably
8
actually make a little bit more, the kind of
9
decision I want to make here, which is more
10
gut based than factual.
11
I can already tell this is going to get
12
political. I can already discern this is
13
going to be a powder keg.
Developing Conceptual
Categories: Chunking
Fracturing and Naming
Data
• Convention is line by line
• In practice this may be a few words, a
sentence, several sentences
• It should be a meaningful, discrete
unit
• Name what’s happening (keep it
active and dynamic)
* You are trying to avoid generating
impressionistic “themes” rather than
data specified grounded categories
IV. Meaning
Making / Open
Coding
Active Naming
Imagining
Comparing
Data as an
indicator – “Seeing
as…”
Ceci n’est pas la soupe!
“SEEING AS” INVENTIVE
NAMING OR
CONCEPTUALIZING
We] “can stare stupidly at
phenomena; but in the absence
of imagination they will not
connect themselves together in
any rational way.” Peirce, 1896
[
This work of imagining what
might be – accounting for, and
at the same time remaining
accountable to the data - is our
concern
It was more than fifteen years ago that I entered the
laboratory of Professor Agassiz, and told him I had
enrolled my name in the scientific school as a student
of natural history. He asked me a few questions about
my object in coming, my antecedents generally, the
mode in which I afterwards proposed to use the
knowledge I might acquire, and finally, whether I
wished to study any special branch.
To the latter I replied that while I wished to be well
grounded in all departments of zoology, I purposed
to devote myself specially to insects.
"When do you wish to begin?" he asked.
"Now," I replied.
This seemed to please him, and with an energetic "Very
well," he reached from a shelf a huge jar of
specimens in yellow alcohol.
"Take this fish," he said, "and look at it; we call it a
Haemulon; by and by I will ask what you have seen."
With that he left me, but in a moment returned with
explicit instructions as to the care of the object
entrusted to me.
"No man is fit to be a naturalist," said he, "who does
not know how to take care of specimens."
Generative questions to
pose to your data that
facilitate
conceptualization
• What is happening?
• What might this be a case of / what can I think
of this as being about?
• What is the actor doing?
• What is the basic problem faced by the actor/s
here?
• What do these actions and statements take for
granted?
• What question about my topic does this data
suggest?
• What imagery or metaphor does this data
evoke?
How is this fragment the same as and
different from others?
V. Meaning
Making
Developing Conceptual
Categories
The Development of Conceptual Categories
(c.f. Grounded Theory)
Theoretically
Driven Data
Sampling
Fracture, Chunk and
Name Data
Compare, Compile
& Rename
Data Fragments
Generate Provisional
Conceptualizations
with their Indicators
Develop Robust Working
Conceptualizations
Process is “finished” when core categories are
theoretically saturated
Developing Categories
From Data Chunks
Category Working Label:
Data
Source:
Interview #
Interview #
Data Samples Constituting Category:
Constant
Comparison
How is this fragment the same as and
different from others?
http://www.youtube.com/watch?v=nxI
ErzX3aQQ&feature=related
Constant
comparison: How are
these the same and
different?
• “Slowly I drew forth that hideous fish,
and with a feeling of desperation again
looked at it.”
• “When I had finished, he waited as if
expecting more, and then, with an air
of disappointment:”
• “…so I walked home by Charles River
in a distracted state, with my two
perplexities.”
Constant Comparison:
How are these the same
and different?
• Finally, just to start some kind of
action, I began copying. In my slow,
painstaking ragged handwriting, I
copied into my tablet everything
printed on that first page, down to
the punctuation marks.
• I pushed my fingers down its throat
to see how sharp its teeth were. I
began to count the scales in the
different rows until I was convinced
that that was nonsense. At last a
happy thought struck me --- I would
draw the fish; and now with surprise
I began to discover new features in
the creature.
The Development of Conceptual Categories
(c.f. Grounded Theory)
Theoretically
Driven Data
Sampling
Fracture, Chunk and
Name Data
Compare, Compile
& Rename
Data Fragments
Generate Provisional
Conceptualizations
with their Indicators
Develop Robust Working
Conceptualizations
Process is “finished” when core categories are
theoretically saturated
Out of his comfort
zone /floundering
• I turned it over and around; looked it in the face -- ghastly;
from behind, beneath, above, sideways, at a three-quarters
view -- just as ghastly. I was in despair; at an early hour, I
concluded that lunch was necessary; so with infinite relief…
• Slowly I drew forth that hideous fish, and with a feeling of
desperation again looked at it.
• When I had finished, he waited as if expecting more, and
then, with an air of disappointment: "You have not looked
very carefully; why," he continued, more earnestly, "you
haven't seen one of the most conspicuous features of the
animal, which is as plainly before your eyes as the fish itself.
Look again; look again!" And he left me to my misery
• I was piqued; I was mortified. Still more of that wretched
fish…
• professor inquired “Do you see it yet?" "No," I replied. "I
am certain I do not, but I see how little I saw before."
• This was disconcerting; not only must I think of my fish all
night, studying, without the object before me,…
• "Do you perhaps mean," I asked, "that the fish has
symmetrical sides with paired organs?“
• I became increasingly frustrated at not being able to express
what I wanted to convey
• It was sad; I couldn’t even write a straight line
• I spent two days just rimming uncertainly through the
dictionary’s pages
Learning to see / be involves stepping out of your comfort
zone—voluntarily or involuntarily— how tentative and unsure
of himself he is (bolded words)
Portraying a divided
literature
Unfortunately, the relevant literature offers contradictory
assessments of the ability of the social sciences to
account for these phenomena. One the one hand we
are told that “Max Weber was the first to consider
bureaucracy as the problem of industrial society () that
he articulated the “classical theory of bureaucracy ()
…All this clearly conveys the impression that Weber’s
work on bureaucracy constitutes …a paradigm...On
the other hand the same literature contends that all
theories of bureaucracies, including Weber’s are
“underdeveloped” ()..
As with any new perspective, its [culture’s] vagaries,
disjunctions and confusions must be carefully
addressed for the research tradition to advance. One
of the most depressing disputes organizational culture
research is what, exactly, constitutes a culture…
Although the dichotomy is not exact, two major
research traditions are emerging …functionalist…
interpretive ()
Stress has been studied from a number of perspectives,
but rather than producing consensus, this research has
portrayed stress variously…
Previous Labels: Research results confused
Links with: Negative findings, Literature progressing
Recommendations
when developing
categories
• Generate multiple category names for data
fragments (c.f. brainstorming)
• Do assign data fragments to more than one
category (discreteness will come later, don’t
force it)
• Do not use technical jargon, ie prior
specified theory as category names (bracket
it out)
• Category names are flexible, use images,
phrases, sentences
• Category names should be a good fit for
the data incident being described
• When a category has 6-12 data indicators,
write a formal definition of that category
Recommendations
when developing
categories
• Do be on the look out for
developing major or “core”
categories
• Be aware how many conceptual
categories you have in
development
• Too many categories? Make a
list of all the category names;
ask yourself, which ones are
similar which ones are different?
VI. Theoretical
Sampling
In light of your developing
categories, what kinds of data do
you need to develop them further?
The Development of Conceptual Categories
(c.f. Grounded Theory)
Theoretically
Driven Data
Sampling
Fracture, Chunk and
Name Data
Compare, Compile
& Rename
Data Fragments
Generate Provisional
Conceptualizations
with their Indicators
Develop Robust Working
Conceptualizations
Process is “finished” when core categories are
theoretically saturated
VII. Memoing
Central to emergence, i.e. growing
your theory
A methodological practice
An exploration
Analytic
Memoing
• Central analytic procedure
• Any writing that is done in relationship to
the research other than actual field notes,
transcription or coding.
• A way of getting things down on paper to
facilitate reflection and insight
• A way of helping you understand and make
sense of your topic and your data
• Prose that tell what your conceptual
categories, matrices, contact summary
elements are about
• Connect ideas, data, and theory
Memoing
• p. 34 it is imperative to engage in a written
conversation with yourself. Imagine yourself
sitting at a desk in a year or so , actually
beginning to write, ore confident then than
now, having a firmer sense of the patterning
of things, perusing all the notes, transcripts
and documents that you have gathered.
Those materials are, in effect, messages to a
future self, and they will lose some of their
immediacy and context. What is clear
NOW will not be so clear later on. Patiently
explain to a future self what you are doing,
why you think it is interesting, why you have
chosen to record what you have, what
relevance it will have
Rock: Handbook of Ethnography
Types of Memo:
From simple to
complex
• Simple, elemental memos – a rendering of
emerging thinking on a specific category on
research question in more general terms.
• How you personally relate to actors and/or the
phenomenon
• Your code/ category choices and working
definitions/understandings
• More complex memos – render thinking on
categories’ relationships, include exploration of
possible links to existing theory.
• Relationships among categories and your research
questions
Excerpt from early
memo
The second clearest category comprises those papers that
perhaps create an intellectual imperative for what is to
follow. There seem to be two textual acts that are key
here. The first is to launch a serious critique of the
literature, (even though it –the lit- is used to establish
consensus on the importance of the topic for
organization studies) They seriously impugn the existing
literature literature as demonstrating that it is not up to
the task of dealing with the investigatory domain
identified. So, Barley suggests that existing approaches
haven’t got at the social construction of meaning,
Gephart suggests that existing approaches haven’t got at
how succession is produced, Hirschhorn and Gilmore
suggest that existing organizational change traditions
cannot deal with organizations in crisis type situations.
There seem to be various ways in which the literature is
shown to be not up to the task, one might be that it
cannot address certain phenomena, another might be that
it is contradictory or in confusion, another may be that it
is not studying what it thinks it is studying, etc. The
second act is to introduce and/or legitimate the new and
more fruitful tradition
Ideas about what you think is happening, contact with the
categories you are developing, contact with literature?
Excerpt from Memo –
Embodiment in Org.
Studies
• Last night’s class focused on meditation
and mantra got me thinking about two
things. First how much mantras or
particular kinds of practices are present for
example the sports psychology stuff. See
the current issue of running journal.
Second, discussion of the state one was
trying to achieve in meditation reminded
me of Czsenkmihalyi’s work on flow … a
particular kind of mind/body
consciousness disposition. What about
mindfulness? Think about the edited
volume on embodied … and it is an
example of such practices already entering
organization studies. The question is
where else does the embodied piece enter
organization studies… the practice work –
not really have practices disconnected
from bodies (*possible contribution).
VIII. Growing
Theories
More than a collection of categories –
how do the categories relate to each
other?
What stories/ narratives can be
developed from observations and in
process conceptualizations?
Some Vocabulary
Questions/Theoretical
Elements- what’s, how’s and
why’s, when’s
• Types – exclusive categories of social units,
aspects
• Structure – constituent dimensions, detailed
organization
• Process – operations over time expressed as
phases or cycles …
• Agency – strategies used to deal with particular
situation
• Causes – conditions and contexts that
engender
• Consequences – of and for what …
Adapted from Glaser, 1973 & Strauss & Corbin, 1998
Open Coding to
Strauss & Corbin’s
Axial Coding Paradigm
Open Coding
Categories
Axial Coding Paradigm
Context
Category
Category
Causal
Category Conditions
Core
Category Strategies Consequences
or
Phenomenon
Category
Category
After Creswell 2e
Intervening
Conditions
One open coding
category as core
phenomenon
Key Terms
• Axial coding- Intense analysis focused on one
category at a time that treats the category as an
‘axis’ around which further coding and category
building is done.
• Categorizing – an analytic conceptualization
• Coding – process of naming what data is about
• Coding families or coding paradigm – a fund of
abstract theoretical terms to aid thinking about
categories and their relations to each other
• In-vivo or substantive coding – captures
substantive aspects of research setting, often in
research participants’ own terms
• Theoretical sampling – Ongoing sampling
specifically for the purpose of developing the
theory
• Theoretical saturation – the point at which
gathering data yields no new properties nor any
new theoretical insights