Grounded theory
[read more=”Click here to Read More” less=”Read Less”]
pre-existing theory and practice. Grounded theory suggests that there is an overemphasis
on quantitative research and wishes to demote the idea that the discovery of relevant
concepts and hypotheses are a priori to research. For further see (Bryant and Charmaz, 2007a,
2007b; Charmaz, 1983, 2000, 2006; Clarke, 2002; Corbin and Strauss, 1990; Cresswell, 2007;
Glaser, 1978, 1992; Glaser and Strauss, 1967; Goulding, 2002; Hood, 2007; Strauss, 1987;
Strauss and Corbin, 1990, 1994). Grounded theory posits that theory is derived from data and
cannot be divorced from the process by which it is developed. Subsequently, questions,
hypotheses and concepts are generated through the data and worked out during the course of
the research (Glaser and Strauss, 1967). Strauss and Corbin (1998) argued that grounded
theory provided methodological lenses for seeing and comprehending the world and involved
the following characteristics: first, on the one hand, the means for critical analysis of events and
situations from an objective stand point, while on the other hand the capability for recognising
subjective tendencies. Second, abstract thinking and openness to useful criticism as well as
empathy and sensitivity when dealing with the actions and discourse of those involved with the
investigation. Finally, there should be immersion and absorption within the research, analysis
and data collection process. Fundamentally, grounded theory allows the direction of the
research and analysis of the data to be guided by an empowered researcher. Indeed, because
of the flexibility of the approach, grounded theory is used by researchers from a range of
different subject areas and disciplines and deployed in eclectic and distinct ways. It is used in
areas as diverse as business, nursing, political studies, psychology, sociology and many areas
of social science in general. Furthermore, when using grounded theory a number of
philosophical issues in terms of ontological and epistemological positions may be ascertained.
This chapter accepts that this methodological approach is primarily concerned with data
collection, analysis and theory construction but because of the philosophical underpinning as
well as eclectic areas that utilise grounded theory, it is worth identifying its roots and evolution
as a useful tool or mechanism for undertaking qualitative research.
Philosophical Underpinnings of Grounded Theory
Even though there is a phenomenological underpinning to interpretations of grounded theory
Glaser and Strauss (1967) primarily drew on the US pragmatist tradition of Dewey (1950) and
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 2 of 25 An Introduction to the Philosophy of Methodology
symbolic interaction approach of Mead (1962) and Blumer (1969). Dewey (1950) argued that
‘flowers can be enjoyed without knowing about the interactions of soil, air, moisture and seeds
of which they are the result. But they cannot be understood without taking just these
interactions into account and theory is a matter of understanding’ (1950: 12). Theory cannot
answer questions ‘unless we are willing to find the germs and roots in matters of experience’
(Dewey, 1950: 12).
Definition Box: Pragmatism
Pragmatism defines truth as those tenets that prove useful to the believer or user. The
verifiability of truth exists to the extent that actuality or things correspond with statements
and thoughts. Objective truth cannot exist because it needs to relate to practice; both
subjective and objective dimensions are necessary.
In the acquisition of knowledge, an essential element ‘is the perception of relations, especially
the relations between our actions and their empirical consequences’ (Scheffler, 1974: 197). In
such a way, the world around us and individuals take on deeper meaning; in this context,
humans need experience and the means of storing that experience. However, as with
phenomenology experience is more than ‘a passive registering or beholding of phenomena; it
involves deliberate interaction with environmental conditions, the consequences of which are
critically noted and fed back into the control of future conduct’ (Scheffler, 1974: 197).
Pragmatists considered that from ‘the child’s exploration of its environment to the scientist’s
theorising about nature the pattern of intelligent thought is the same: a problem provides the
initial occasion of inquiry … Experiment … is, experience rendered educative’ (Scheffler, 1974:
196).
Through these pragmatist foundations grounded theory can be understood in terms of
symbolic interaction, where the ‘individual enters as such into his own experience only as an
object not as a subject; and he can enter as an object only on the basis of social relations and
interactions’ (Mead, 1962: 225). Through the language and structure of roles we become a
generalised other; we attain a consciousness of self as a generalised other. This may allow the
individual to take an impartial and general standpoint in observing and evaluating one’s own
conduct when one becomes a generalised other or the object of one’s own reflection. At this
point one has become self, which reflects the phenomenological underpinnings of Hegelian
recognition and being in the world.
Blumer (1962, 1969) also built on the work of the pragmatist tradition and considered that,
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 3 of 25 An Introduction to the Philosophy of Methodology
‘ordinarily human beings respond to one another, as in carrying on a conversation, by
interpreting one another’s actions or remarks and then reacting on the basis of interpretation’
(Blumer, 1969: 71). Grounded theory builds on this understanding and considers that research
should be grounded out of reality and that the researcher should enter into the field and
discover/comprehend what is going on. People have an active role in shaping the world and
through interrelationships in terms of meaning, action and conditions the nature of experience
continually evolves, which creates continual re-interpretation of phenomenon (Corbin and
Strauss, 1990). Indeed, grounded theory is primarily inductive and pursues the interpretations
of those involved in the situation that is being researched and the interpretations of the
researcher in relation to the data. Through this process, grounded theory is enacted and
substantive theory constructed. Symbolic interaction is distinct to human beings, it is part of
what makes us human because we ‘interpret or define each other’s actions’ rather than simply
react to them (Blumer, 1962: 179). Furthermore, humans have and are able to act towards self.
Mead considered that the ability to react to self was the central mechanism of existence.
This mechanism enables human beings to indicate to themselves things in their surroundings
and thus guide action through what is noted. ‘Anything of which a human being is conscious …
he is indicating to himself … The conscious life of the human being … is a continual flow of self
indications’ (Blumer, 1962: 180). Such is similar to phenomenology in terms of being in the
world and intentionality. Fundamentally, ‘the formation of action by the individual through …
self-indication always takes place in a social context’ (Blumer, 1962: 183); as does the
background or environment (fore-structure) which indicates possible ways of questioning
(Heidegger, 1962/1984). Additionally, Stern considered that grounded theory was an
interpretative method and was underpinned by ‘phenomenology that is methods that are used
to describe the world of the person or persons under study’ (1994: 213).
Furthermore, with distinct linkage to Heidegger and later Merleau-Ponty, interpretation ‘is
grounded in something we have in advance … understanding operates in … an involvement
whole that is already understood’. Effectively the environment or background determines
possible ways of questioning. However, interpretation is also ‘grounded in something we see in
advance – in foresight’ (Heidegger, 2004: 191). Additionally, interpretation is made on the basis
of those interacting with the researcher; in this context, the individual is both subjective and
objective. The researcher is interpreted by self in relation to society and interpreted by society
in relation to self through self-indication. ‘Self-indication is a moving communicative process in
which the individual notes things, assesses them, gives them a meaning, and decides to act on
the basis of meaning’ (Blumer, 1962: 183). However, to accomplish this some foresight is
necessary, even if this is only a comprehension of language or culture. This foresight takes into
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 4 of 25 An Introduction to the Philosophy of Methodology
consideration the theories that already exist or more formal theories. Additionally, the
researcher brings to the analysis expertise, knowledge and theoretical sensitivity. Effectively,
objectivity is continually pursued through the recognition of subjective influences throughout
the research process. As with phenomenology there are relationships between reality and the
human mind. Indeed, the ‘research findings constitute a theoretical formulation of the reality
under investigation rather than consisting of a set of numbers or a group of loosely related
themes’ (Corbin and Strauss, 1990: 24).
‘One now regularly takes an impartial and general standpoint in observing and evaluating one’s
own conduct … (however) … The organised community or social group which gives to the
individual his unity of self may be called the generalised other which is the attitude of the whole
community’ (Blau, 1952: 162–3, original parentheses). Interpretations of situations change as
individuals change (this includes the researcher and the researched). Indeed, with the
knowledge that pure objectivity cannot be attained the research can become more objective, for
example, through accepting subjectivity one can become more objective. Grounded theory
attempts to understand social patterns and construct social theory. ‘Grounded theory involves
soliciting … emic viewpoints to assist in determining the meaning and purposes that people
ascribe to their actions’ (Guba and Lincoln, 1994: 110). ‘Grounded theory methodology
incorporates … assumption(s) … concerning the human status of actors whom we study. They
have perspectives on and interpretations of their own and other actors’ actions. As researchers
we are required to learn what we can of their interpretations and perspectives’ (Strauss and
Corbin, 1994: 280, authors’ brackets).
Reflection Box: Pragmatism, Phenomenology and Symbolic Interaction
Demonstrate where similarities exist between pragmatism, phenomenology and symbolic
interaction.
Indeed, such identifies aspects of being in the world and intentionality outlined by Heidegger
and Husserl. The rest of this chapter will outline what grounded theory entails and explain how
it may be applied. It then discusses the different perceptions of grounded theory displayed by
Barney Glaser and Anselm Strauss as well as the subsequent variants that evolved from these
distinctions.
Using Grounded Theory
Grounded theory is a way of collecting data and through comparative analysis and coding a
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 5 of 25 An Introduction to the Philosophy of Methodology
means of generating or building substantive theory. Grounded theory is about building theory
through the collection and analysis of rich data; that is data that exists below the surface of the
social entity being investigated. To achieve rich data some persistence and creativity in the
inquiry process is required. Initial questions the researcher may first attempt to develop include:
What research problem should be studied?
How may the project be undertaken?
Which methods should be used in the collection of rich data?
The research problem should emanate through work or ideas generated through previous
studies; issues that have become apparent following the writing of a report, essay, dissertation
or academic paper. Furthermore, when identifying the research problem, issues regarding
access and participant numbers requires some consideration. If the right participants are
located and access to these allowed, then the first hurdles for accumulating rich data have
been overcome. Such data will illuminate further detail regarding the research question or issue
and enable new insight into what may initially have seemed mundane concepts. Grounded
theory is social science based and involves research about and with people. Rich data enables
insight into the veiled and sometimes opaque feelings, ideas and beliefs of the individuals
involved in the investigation as well as those investigators will undoubtedly carry themselves.
Grounded theories are developed or built through the comparative analysis of rich diffuse data
(data collected from different places and through separate methods). Through writing thick
description based on field-notes, observations, personal diaries and accounts of events as well
as interviews and secondary data, detailed narratives may be formed.
Fundamentally, grounded theory attempts to improve theory as one can only replace existing
theoretical frameworks with improved or enhanced theories (Glaser and Strauss, 1967).
‘Grounded theory is based on the systematic generation of theory from data, and itself is
systematically obtained from social research (and) offers a rigorous orderly guide to theory
development that at each stage is closely integrated with a methodology of social research’
(Glaser, 1978: 2). Through comparative analysis, grounded theory creates theories made up of
general categories. It is not necessary to know the empirical or specific situation better than
those involved; the researcher simply wishes to develop theory that applies to relevant
behaviour. Theory is never complete but always under development. Theory generated from
data means that ideas or hypotheses are not only derived from this but worked out in relation to
the data as the research progresses.
While positivism seeks to verify deduced hypotheses, grounded theory implies steps prior to
discovering what concepts and hypotheses are relevant to the area of research should be
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 6 of 25 An Introduction to the Philosophy of Methodology
made explicit. Ultimately, the relationship between categories and sub-categories which are
discovered during the research should be as a result of information contained within the data or
from deductive reasoning which has been verified within the data, but not from previous
assumptions which have not been supported (Howell, 2000).
Grounded theory offers a rigorous approach to research in terms of:
interactive iterative nature of data collection and analysis;
comparative methods;
conceptual analyses through memo writing;
refinement of emerging ideas through sampling;
theoretical framework derived from and integrated with both data and theory.
Glaser (1978, 1992) identifies the following criteria:
Fit
Work
Relevance
Modifiability.
Theoretical categories must emerge and be developed from data analysis; they must fit.
Grounded theory should order the data so as to explain the phenomena; it should work. It
should have relevance in terms of dealing with actual problems and processes located in the
research setting. In addition, through accounting for variation grounded theory is durable and
flexible.
Glaser and Strauss (1967) challenged the division between theory and data as well as the
perceived lower importance of qualitative techniques in relation to quantitative. Furthermore,
they also synthesised data collection and analysis and critiqued ideas that qualitative research
was unsystematic and relied heavily on impressions. Overall, they argued that qualitative
research could go beyond descriptive case studies and engender theory development. Glaser
used his training in positivism to develop a rigorous form of qualitative analysis. Strauss
brought a pragmatic perspective that built on ideas relating to symbolic interaction. Glaser
(1992) clearly states that he used the statistical analytical method as a model for the qualitative
method in grounded theory. Consequently, a model for quantitative analysis provided the basis
for a unique qualitative approach.
The other side of the equation that is symbolic interaction involved a number of underpinnings
for grounded theory. These include:
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 7 of 25 An Introduction to the Philosophy of Methodology
pragmatism (close relationship between theory and practice);
idiographic as opposed to nomothetic research;
qualitative research as primary rather than secondary;
exploration as a central function;
creation of worldviews through sensitising concepts (empathy and judgment are central);
social action rather than structuration as the main focus;
symbols are relayed between actors through agency;
intersubjectivity is central;
interaction between self and other is central for humanity;
self is a world of meaning not an external structure (primarily cognitive).
Open and axial coding examines phenomena through comparing and categorising data. This
text will break down coding procedures and explain the processes and activities involved in
each of the procedures. The initial objective for grounded theory is to identify categories and
properties which are relevant to the theory and allow a level of integration. ‘The goal of the
analyst is to generate an emergent set of categories and their properties which fit, work and are
relevant for integrating theory. To achieve this goal the analyst begins with open coding’
(Glaser, 1978: 56). Attention should be fixed on a category and the properties that emerge
continually coded and analysed as the initial basic steps. Fundamentally, the researcher
constantly compares and continually categorises data and concepts.
Theoretical Sampling
Theoretical sampling is undertaken on the basis that ‘concepts have proven theoretical
relevance to the evolving theory’ (Strauss and Corbin, 1990: 176). Theoretical sampling involves
three processes: open sampling, which relates to open coding; relational and variational
sampling, which is associated with axial coding; and discriminate sampling, which is linked to
selective coding (coding processes are discussed below). Proven theoretical relevance
identifies concepts that are significant enough to be considered categories:
they are deemed significant because (1) they are repeatedly present or notably absent
when comparing incident after incident (2) through coding procedures they earn the
status of categories … The aim of theoretical sampling is to sample events, incidents,
and so forth, that are indicative of categories, their properties and dimensions, so that
you can develop and conceptually relate them. (Strauss and Corbin, 1990: 177)
The sampling is undertaken purposefully which encompasses choosing individuals and
documentation that demonstrated variations in the categories and what happened when
change occurred. As with the coding (see below) the distinction between relational and
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 8 of 25 An Introduction to the Philosophy of Methodology
variational sampling and discriminate sampling became unclear. Discriminate sampling is direct
and deliberate. ‘In discriminate sampling, a researcher chooses the sites, persons and
documents that will maximise opportunities for verifying the story line and relationships
between categories’ (Strauss and Corbin, 1990: 187). Sampling in grounded theory studies is
concerned with the ‘representativeness of concepts in their varying forms. In each instance of
data collection, we look for evidence of its significant presence or absence, and ask why?’
(Strauss and Corbin, 1990: 190) (see Table 9.1). Grounded theory studies look ‘for incidents
and events that are indicative of phenomena’ (Strauss and Corbin, 1990: 187). They pursue
density and:
the more interviews, observations and documents obtained, then the more evidence
will accumulate, the more variations will be found, and the greater the density will be
achieved. Thus there will be wider applicability of the theory, because more and
different sets of conditions affecting phenomena are uncovered. (Strauss and Corbin,
1990: 190–1)
Open, Axial and Selective Coding
Grounded theorist code data as it is collected. Coding defines and categorises data. Codes are
created as the data is studied. Data should be continually interacted with and questions
continually posed as the analysis develops. It should not fit into pre-conceived codes but codes
that have emerged through the researcher’s interpretation of the data. Coding is iterative and
interactive and line by line coding can ensure that the researcher’s beliefs are not imposed on
the data and interpretation.
Glaser (1978, 1992) stressed continual comparative methods, that is, comparing:
views, action environments, narratives, discourse, beliefs and stories;
data from the same individuals at different points in time;
situation with situation;
incident with incident;
data with categories;
categories with categories. (Charmaz, 1983; Glaser, 1978, 1992)
Strauss and Corbin (1990) are more structured and developed new procedures:
dimensionalisation;
axial coding;
conditional matrix.
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 9 of 25 An Introduction to the Philosophy of Methodology
These are supposed to make emerging theory thicker or denser, complex and rigorous.
Through memo writing, coding provides the very bedrock of grounded theory; the essential
relationship between data and theory is a code, which conceptualises the underlying patterns
of the data. Consequently through ‘generating a theory by developing the hypothetical
relationships between conceptual codes (categories and their properties), which have been
generated from the data as indicators, we discover a grounded theory’ (Glaser, 1978: 55).
Charmaz (1983) considered that coding incorporates ‘the initial phase of the analytical method
(and) is simply the process of categorising and sorting data. Codes then serve as shorthand
devices to label, separate, compile and organise data. Codes range from simple concrete, and
topical categories to more general, abstract conceptual categories for an emerging theory’
(Charmaz, 1983: 111). Indeed, in ‘grounded theory research there are three basic types of
coding; open, axial and selective’ (Corbin and Strauss, 1990: 9). Through memo-writing, coding
requires researchers to re-assess what may seem like an obvious perspective of the
phenomenon under investigation; the data should undergo analysis from different standpoints
and it should be continually critiqued and questioned. Comparison is a central technique and
the researcher should continually compare memos with codes and data with data and codes.
Through coding interviews, documentation, observations, focus groups and secondary data,
categories emerge through data segmentation on which labels may be placed. Coding involves
interpreting the data and moving from concrete to more abstract conceptualisations; through
coding one takes the first tentative steps towards theory generation. Coding involves the very
framework of the analytical process and identifies the bridge between the data and emergent
substantive theory. Coding involves microanalysis (detailed close line-by-line attention) and
involves a dynamic fluids process; categories and properties are created and relationships
assessed through open and axial coding.
Definition Box: Microanlysis
Microanalysis can involve word by word coding of the data which forces the researcher to
concentrate on specific meanings incorporated in the text; this means the way the
sentences are structured and flow as well as distinct contexts. Line-by-line coding which
comprises of labelling each line read (not sentences but each line); this throws up
unexpected themes and ideas that one may not detect when coding sentence by
sentence. When dealing with empirical difficulties accounted through observations, focus
groups, interviews or documentation, line-by-line coding is particularly useful as it details
and reveals opaque issues, problems and situations. This coding procedure investigates
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 10 of 25 An Introduction to the Philosophy of Methodology
underlying meanings and actions as well as clarifies the significance of ideas through
identifying and comparing deficiencies in the data.
Indeed, open and axial coding are undertaken together in the formation of categories and
properties, however for the purpose of explanation we will now separate these activities and
explain the processes individually.
Memo Writing and Open Coding: A Practical Example
Initially, grounded theory is inductively derived from the study of the phenomena it represents.
Data collection, analysis and theory are reciprocal; one does not start with a theory, which is
then tested but with an area of study from which what is relevant to the area is derived through
data collection. Induction is grounded in social phenomena or observations and experience;
hence the strong link between inductive procedures and grounded theory. Glaser and Strauss
(1967) and Charmaz (1983) proposed that data collection and analysis were undertaken
simultaneously and interpretations formed through data discovery. This process which
simultaneously led back to the data (the researcher resides in the field of discovery) allows for
emerging ideas because it provides for further data collection. A main strength of the grounded
theory approach is that data and ideas are derived through the research rather than through a
priori deduction. Verification is secondary to understanding, not simply understanding the
phenomenon, but an understanding of social life as process.
As such, theoretical analyses may be transcended by further work either by the
original or later theorists. In keeping with their foundations in pragmatism, then,
grounded theorists aim to develop fresh theoretical interpretations of the data rather
than explicitly aim for any final or complete interpretation of it … Although every
researcher brings to his or her research general preconceptions founded in expertise,
theory, method, and experience, using the grounded theory method necessitates that
the researcher look at the data from as many vantage points as possible. (Charmaz,
1983: 111–14)
Data is analysed as it emerges or collected and through coding, ‘order created’ (Charmaz,
1983: 114). It is given that researchers have pre-understandings through their expertise,
experience, theoretical frameworks, and for grounded theory these are important elements of
analysis and important perspectives to be taken into consideration when dealing with the data;
it is important that data be considered from many different perspectives or positions
With grounded theory, data collection procedures and strategies are continually refined; flexible
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 11 of 25 An Introduction to the Philosophy of Methodology
guidelines are outlined rather than dogmatic prescriptions. Methods that deal with research
questions should be targeted and followed in an investigative manner; methods of data
collection affect the extent phenomenon will become important and how when and where this
will be observed. Indeed, methods of data collection determine how we make sense of the
phenomenon investigated. Coding for grounded theory involves posing analytical questions.
Memos provide a record of the research process and include modes of ‘analysis, thoughts,
interpretations, questions, and directions for further data collection’ (Strauss and Corbin, 1998:
110).
Strauss and Corbin (1998) give an example of an interview they undertook with a drug user
which homed in on the term ‘use’ as it is specifically used by people when speaking of drugs.
Usually when the term used is employed it means that something is employed in a specific way
‘that an object or person is used for some purpose’ (Strauss and Corbin, 1998: 110). When one
uses something one considers that the user is in control and when one reassesses the term
drug user in this context, even though it specifically means to consume, remnants of these
ideas may also be included in the term; ‘for example, being used for some reason, having
control over what one does’ (Strauss and Corbin, 1998: 111). Rather than simply taking drugs
the term drug user takes on a number of connotations in terms of ‘self-control over usage …
purposeful … directed act that serves as an end and has a desired effect’ (Strauss and Corbin,
1998: 111) Consequently, even though such issues may not be explicit in the data during the
initial stages, this code or idea is something that may be kept in mind and used as the analysis
and further data collection progresses (ibid). So the memo ‘user’ arises and directs the
researcher to think in more detail about this term, for instance, one does not usually employ the
term an ‘alcohol user’ and if one did what would this actually mean? Think of the reasons you
may use alcohol; do you use it to cook, get to sleep, get drunk, take to dinner parties or provide
for individuals invited to your home, do you use alcohol in bars, the park, at home, what type of
alcohol do you use spirits or wine, do you drink alone or only with company, does it relax you
and assist sleep, for how long does someone and how often do they use alcohol? Each of
these questions raise certain properties regarding alcohol use ‘such as frequency, duration,
type, purpose, way of using and place of use’ (Strauss and Corbin, 1998: 111) Consequently, in
subsequent interviews issues regarding these properties may be further explored in terms of,
reasons for usage, place, type, environment, levels of usage. Indeed, the category use or user
is beginning to emerge and the properties relating to this may be placed on a ‘dimension’ with
say ‘usage’ at one point and ‘reason’ at the other with the other terms placed on this dimension
in relation to these two extremities. User emerges into a category with properties representing
property locations along the continuum. For example, alcohol usage in terms of alcoholism and
social drinker may be identified in terms of frequency and type (the number of times one gets
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 12 of 25 An Introduction to the Philosophy of Methodology
drunk and strength of alcohol) will distinguish between these properties within this category.
However, we may wish to clarify what we mean by social drinker and identify distinctions
regarding drinking alone or not and the type of alcohol consumed. Further delineations may
emerge in terms of binge drinking and how this may differ from alcoholism or social drinking;
through such a process, property patterns begin to emerge and congregate on the dimension.
Dimensions involve the ‘range along which general properties of a category vary, giving
specification to a category and variation to the theory’ (Strauss and Corbin, 1998: 101).
Obviously, through further data collection and analysis this dimensionalisation may be further
developed and/or transformed.
Axial and Selective Coding
Axial coding involves the re-structuring of the whole process by finding connections between
the data. Axial coding pulls the analysis together and provides a means of unifying the data
into a coherent whole. It indicates how the categories created by open coding fit together and
how they congregate around a core category. This type of coding provides a means of
organising large amounts of data as the analysis details and understands the development of a
major or core category. Fundamentally axial coding links categories and sub-categories through
diagrams as it pursues the development of substantive theory. It uses a set of terms to denote
the process which include: causal conditions and phenomenon, which involve the environment
of the researched in terms of circumstance or specific situation and the entity under analysis as
well as the context or the more general historical and social/cultural situation within which the
research or investigation is undertaken. The intervening conditions which pull together answers
relating to questions regarding why where, what, when, how? This portrays the need for
action/interaction between entities in terms of the reason certain ideas and events occurred, for
example, whether these be strategic or routine responses made to certain conditions,
phenomenon or questions; action and interaction correspond with questions relating to whom
and how. Finally, the consequences are the outcome of the process especially the success or
failure of the action/interaction between entities. Consequences are denoted through questions
relating to the outcome of actions and interactions, for example what would happen if x or y
occurred? In ‘axial coding we continue to look for additional properties for each category and to
note the dimensional location of each incident, happening or event’ (Strauss and Corbin, 1990:
114–15).
Definition Box: Axial Coding
Axial coding is illustrated through:
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 13 of 25 An Introduction to the Philosophy of Methodology
causal conditions;
phenomenon;
context;
intervening conditions;
action/interaction;
consequences. (Corbin and Strauss, 1990: 96–7)
Axial coding can be used as a framework to encourage the emergence of categories which
provide insight into the core category and the substantive theory. Glaser considered that axial
coding ‘undermines and confuses the very method that he (Strauss) is trying to build’ (1992:
61). This process forces the data and negates theoretical coding. The grounded theorist should
code categories and properties and allow theoretical codes to emerge where they will. Strauss
and Corbin (1990) argued that axial coding allows a more focused means of discovering and
relating categories. Strauss and Corbin develop categories (phenomenon) in relation to the
underlying conditions that enable its development, and through identification of properties the
location of this phenomenon on a dimension (dimentionalisation), the context and the
action/interaction strategies used to ‘handle, manage, and respond to this phenomenon’ (1990:
61).
Selective coding illustrates how the phenomenon fits around a core category and involves the
process by which emerging categories are organised and unified around a core category
(Corbin and Strauss, 1990). Core categories incorporate central phenomenon of research
projects as they are identified through questions such as: ‘What is the main analytical idea
presented in this research? What does all the action/interaction seem to be about? The
selection of data and the creation of other categories are processed with the core category in
mind which are identified and unified through axial coding.
Between these three coding procedures exists ‘process’ through which changes to the data are
monitored and made explicit. Process is also built into the theory. Process analysis involves
‘breaking phenomenon down into stages, phases, or steps. Process may also denote
purposeful action/interaction that is not necessarily progressive, but changes in response to
prevailing conditions’ (Corbin and Strauss, 1990: 10). Even though initially inductive, grounded
theory also involves a deductive component, which is primarily used during the identification
and pursuit of process. ‘As you have probably noticed while coding we are constantly moving
between inductive and deductive thinking’ (Strauss and Corbin, 1990: 111). When process is
difficult to identify, the researcher may turn to deductive analysis so as to identify, possible
situations of change, ‘then go back to the data or field situation and look for evidence to
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 14 of 25 An Introduction to the Philosophy of Methodology
support refute or modify that hypothesis’ (Strauss and Corbin, 1990: 148).
Theoretical Sensitivity
Grounded theory research should be approached with an objective demeanour, pursed with an
open mind and primarily through inductive processes. That said, one must acknowledge that
objectivity, an open mind as well as a completely inductive position are difficult to attain.
Everyone is instilled with subjective tendencies and pre-conceptions, that said recognising
these tendencies is part way to dealing with them. To overcome these tendencies one may use
one’s theoretical sensitivity, which allows the researcher to remain sensitive and ‘record events
and detect happenings without first having them filtered through and squared with pre-existing
hypotheses and biases’ (Glaser, 1978: 3). As identified in the phenomenology of Heidegger and
Merleau-Ponty individuals have pre-determined ideas and different levels of sensitivity, which
depend on ‘previous reading and experience with or relevant to the area … Theoretical
sensitivity refers to the attribute of having insight, the ability to give meaning to data, the
capacity to understand, and the capability to separate the pertinent from that which isn’t … It is
theoretical sensitivity that allows one to develop a theory that is grounded conceptually dense’
(Strauss and Corbin, 1990: 40–1). Glaser expanded on his earlier definition of theoretical
sensitivity and argued that it ‘refers to the researcher’s knowledge, understanding, and skill
which foster his generation of categories and properties’ (1992: 27).
Question: Theoretical Sensitivity
What is theoretical sensitivity? How does this relate to ideas outlined by certain
phenomenological positions?
Surveys in Grounded Theory
Within the grounded theory methodology the role surveys and quantitative data play is
ambiguous. ‘The sociologist whose purpose is to generate theory may of course collect his own
survey data, but, for several reasons, he is more likely to analyse previously collected data
called secondary (data)’ (Glaser and Strauss, 1967: 187). However, the researcher should give
him/herself the ‘freedom in the flexible use of quantitative data or he or she will not be able to
generate theory that is adequate … (and) in taking this freedom … be clear about the rules he
is relaxing (Glaser and Strauss, 1967: 186). Such flexibility will allow the richness of qualitative
data to become apparent:
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 15 of 25 An Introduction to the Philosophy of Methodology
and lead to new styles and strategies of quantitative analysis, with their own rules yet
to be discovered … For example, in verification studies cross-tabulations of
quantitative variables continually and inadvertently lead to discoveries of new social
patterns and new hypotheses that are often ignored as not being the purpose of the
research. (Glaser and Strauss, 1967: 186)
Grounded theory relaxes rules of verification and accuracy of evidence to enable further theory
generation ‘the way they are relaxed for purposes of generating theory could apply to many
styles of analysis’ (Glaser and Strauss, 1967: 187). ‘One might use qualitative data to illustrate
or clarify quantitatively derived findings; or, one could quantify demographic findings. Or, one
could use some form of quantitative data to partially validate one’s qualitative analysis’ (Strauss
and Corbin, 1990: 18–19). This outcome may be realised through triangulation. Consequently,
although surveys are not grounded theory techniques in the purest sense, used in certain ways
they may benefit theory generation. In this context, they may be utilised in a grounded theory
study and eventually they may be considered as part of the technique. This, one may
speculate, is the direction in which Glaser and Strauss (1967) pointed toward and Glaser
(1992) makes clear. ‘To repeat, qualitative analysis may be done with data arrived at
quantitatively or qualitatively or in some combination’ (Glaser, 1992: 11) All methods are
acceptable and how they may be used ‘together effectively … depends on the research’
(Glaser, 1992: 12).
Example: Using a Survey
To supplement categorisation and provide further coding a survey can be conducted to
investigate further perceptions. The survey questions can be formulated in relation to …
categories dimensions and properties (developed through previous coding) and
emphasised basic questions such as When? Where? What? and How Much? (Howell,
2000).
Substantive and Formal Theory
Glaser and Strauss (1967) argued that grounded theory was concerned with two types of
theory: substantive and formal. They emphasised that theory generation was accomplished
through the collection, coding and analysis of data and that these three operations, as far as
was possible, were undertaken together. Collection, coding and analysis should interact
throughout the investigation as their separation hinders theory generation and set ideas stifle it.
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 16 of 25 An Introduction to the Philosophy of Methodology
Definition Box: Substantive Theory
Substantive theory necessitates four central criteria. Fit, comprehension, generality and
control: first, theory should be induced from diverse data and be faithful to reality (it
should fit). Second, the fit should be comprehensible; third, the data should be
comprehensive and interpretations conceptually wide (there should be generality).
Finally, in relation to generality, it should be made clear when conditions apply to specific
situations and phenomenon (there should be control) (Corbin and Strauss, 1990).
Grounded theory generates substantive theory through comparative analysis and coding. It
does not attempt to undermine theory but improve it ‘a theory’s only replacement is a better
theory’ (Glaser and Strauss, 1967: 28).
Grounded theory is based on the systematic generation of theory from data, and itself
is systematically obtained from social research. Thus, the grounded theory method
offers a rigorous orderly guide to theory development that at each stage is closely
integrated with a methodology of social research. (Glaser, 1978: 2)
Through the general method of comparative analysis, grounded theory wishes to create a
theory made up of general categories.
A formal theory is composed of a model plus an indefinite number of interpretations,
and there is a sharp distinction between model and interpretation. A model is not
affected by any of its interpretations, but can be understood and studied in abstraction
from all of them … A substantive theory … is … about something in the real world.
(Diesing, 1972: 31)
Consequently, substantive theory needs to be verified and if changes to the theory are to be
made there must be references to empiricism. The formal theory ‘can be understood and
studied in abstraction … one can … make deductions, search for inconsistencies, study the
effects of changes in those postulated, and add new terms without referring to anything
empirical’ (ibid). Formal theories are more abstract and may be broken down into meso (middle
range), grand or meta theories (philosophies); as discussed in more detail above. Substantive
theory relates to a practical situation and through accumulation of substantive theory can
develop and become meso, grand and eventually meta theory.
This chapter acknowledges that it is easy to find a problem with a theoretical concept by
identifying that certain data is missing. This could be the charge against most analyses.
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 17 of 25 An Introduction to the Philosophy of Methodology
However, as Glaser and Strauss put it: ‘If each debunker thought about the potential value of
comparative analysis … he would realise that he has merely posed another comparative datum
for generating another theoretical property or category’ (1967: 22). Despite what those
concerned with evidence may say, nothing has been disproved, only another comparison
created (Glaser and Strauss, 1967). Theory is never a finished product, but always under
development. Generating theory from data means that most of the ideas or hypotheses are not
only derived from the data but are worked out in relation to the data as the research
progresses.
Question: Grounded Theory
Identify why and then how you might undertake a grounded theory approach for an
upcoming research project. What would you consider the major strengths and
weaknesses of this approach?
Theory building is an important element within the phenomenological tradition or approach.
Made explicit in grounded theory qualitative research allows substantive frameworks to emerge
through the categorisation of data. Substantive theory is concerned with a specific domain or
area of inquiry and closely related to practical situations. Substantive theories are developed to
illuminate nuances of human interaction; they are embedded in the relationship between
theories and practice (praxis).
Categories are continually synthesised to form a theoretical framework through the continual
attention to the relationship between memos and existing theoretical ideas. Memos and ideas
are continually coded and through weaving these together substantive theory gradually
emerges. Coding data and memo-writing are central to the grounded theory technique of
constructing substantive theory; memos may be written representations, diagrams, tables,
matrices or vignettes. Through such a process substantive theory will gradually emerge.
Furthermore, a core category may be identified around which peripheral categories revolve.
Core categories require the following dimensions: they need to relate to other categories in
some form or another in qualitative (depth of relations) and quantitative (number of relations)
contexts, be recurrent in the data, maximise variations and build theory with implications for
pre-existing formal theory. In relation to these approaches substantive theory may also involve a
model or a diagram of the framework identified through the core category and related
peripheral categories. The substantive theory should identify how the separate categories are
integrated within the theory; it needs to be conceptually dense and integrated. However,
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 18 of 25 An Introduction to the Philosophy of Methodology
density and integration require intimacy with the data and close proximity to those being
researched and analysed which may fall into the trap of simply stating the obvious.
Consequently, the substantive or emergent theory needs to relate to formal or pre-existing
theoretical frameworks. ‘A substantive theory generated from the data must be formulated, in
order to see which of diverse formal theories are, perhaps, applicable for furthering additional
substantive formulations’ (Glaser and Strauss, 1967: 34).
Substantive theory is developed in relation to an empirical situation whereas formal theories
either pre-exist or are built with regard for a formal or conceptual area. Glaser and Strauss
(1967) argued that both substantive and formal theories exist between models and grand
theory but in the main they take the form of meso theories and allow a bridge between lower
range theories and philosophical perspectives. Formal theory may be generated directly from
the data but it is more conducive to develop substantive theory and from this move onto more
formal applications.
Disputation in Grounded Theory: Barney Glaser and Anselm Strauss
For many years there existed a dispute between Glaser and Strauss, which revolved around
differences regarding the emergence and forcing of data. Glaser considered that Strauss and
Corbin’s ‘pet theoretical code violates relevance and forces data’ (Glaser, 1992: 28). He
contended that such a structured outlook undermines the emergent, empirical and endless
ways of relating substantive codes. ‘The researcher must be aware of the vast array of
theoretical codes to increase his sensitivity to their emergence in the data’ (Glaser, 1992: 28).
However, in their work Strauss and Corbin address their book to those ‘who are about to
embark on their first qualitative analysis research project and who want to build theory at the
substantive level’ (Strauss and Corbin, 1990: 8). In other words, it is a simplification, which may
lead the researcher into the more difficult nuances of grounded theory.
Verification also seems to be a sticking point between the scholars. However, on closer
examination neither is pursuing pure verification; each wishes for it to add to theory generation
not to negate or disprove ‘but add variation and depth of understanding’ (Strauss and Corbin,
1990: 109). Glaser argued that the ‘two types of methodologies should be seen in sequential
relation. First we discover the relevance and write hypotheses about them, then the most
relevant may be tested for whatever use may require it’ (1992: 30). Whereas, Strauss and
Corbin saw it as an aspect of the grounded theory method; they considered that statements
should be verified against data, not to ‘necessarily negate our questions or statements, or
disprove them, rather … add variation and depth of understanding’ (Corbin and Strauss, 1990:
108–9). It is just as important to ‘find differences and variation as it is to find evidence that
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 19 of 25 An Introduction to the Philosophy of Methodology
supports our original questions and statements. The negative or alternative cases tell us that
something about them is different so we must move in and take a closer look’ (Strauss and
Corbin, 1990: 109). However, each considers that it is possible to utilise verification as part of
theory generation, the latter as part of grounded theory and the former as a methodology in its
own right (see Strauss and Corbin, 1990: 107–9, Strauss, 1987: 11–15 and Glaser, 1992: 27–
30).
Strauss (1987) makes his position clear where he contends that induction, deduction and
verification are the very basis of grounded theory. ‘Because of our earlier writing in Discovery
(1967) where we attacked speculative theory – quite ungrounded in bodies of data – many
people mistakenly refer to grounded theory as “inductive theory” (however), as we have
indicated all three aspects of inquiry … are absolutely essential’ (Strauss, 1987: 12 author’s
brackets). ‘Grounded theory is of course inductive; a theory is induced or emerged after data
collection starts. Deductive work in grounded theory is used to derive from induced codes
conceptual guides as to where to go next for which comparative group or sub-group, in order to
sample for more data to generate the theory’ (Glaser, 1978: 37–8).
Glaser and Strauss’ disagreements are based around their emphasis on deductive and
inductive processes; Strauss considers that induction, deduction and verification are essential
elements of grounded theory. Induction is primarily based on experience with the same kind of
phenomena at some point in the past. It may be apparent because of personal experiences,
exploratory research into phenomenon, previous research or because of theoretical sensitivity
(knowledge of technical literature). ‘As for deduction: Success at it rests not merely on the
ability to think logically but with the experience in thinking about the particular kind of data
under scrutiny’ (Strauss, 1987: 12). This means drawing on experience as well as thinking
about the phenomenon and may include comparative analysis to further the deductive powers
(Strauss, 1987: 12). He also indicated that experience and learned skills are very important for
verification. ‘If … experience and associated learned skills at verification, deduction and
induction are central to successful enquiry, do not talent-gifts-genius contribute to that
success?’ (Strauss, 1987: 13).
Strauss and Corbin (1990) proposed that in grounded theory there is a continual movement
between inductive and deductive thinking and that their statements are deductively proposed
and verified. There is a continual comparison of incidents ‘there is a constant interplay between
proposing and checking. This back and forth movement is what makes our theory grounded’
(Strauss and Corbin, 1990: 111). Glaser posits that it is at this point that Strauss and Corbin
indulge in ‘full conceptual description by forcing the data and leaving the emergence of
grounded theory out completely’ (Glaser, 1992: 71). The sticking point is the confusion between
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 20 of 25 An Introduction to the Philosophy of Methodology
induction and deduction. Glaser charged that he (Strauss) ‘confuses induction with testing
deductive hypotheses which are forced on the data (and) that it is not inductive to say the data
disproves a hypothesis, it is simply a verification’ (1992: 71). However, Strauss and Corbin
contended that it is necessary to continually verify ‘concepts and relationships arrived at
through deductive thinking must be verified over and over again against actual data … we are
building grounded theory and it is the grounding or verification process that makes this mode of
theory building different’ (Strauss and Corbin, 1990: 111–12). In response to this Glaser argued
that Strauss and Corbin are developing a verification method. ‘It simply tests forced conceptual
hypotheses’ (Glaser, 1992: 92). It is not a method that generates theory but one that verifies; it
is a theory that forces the data rather than allowing it to emerge. For Glaser it was crystal clear
that Strauss and Corbin had created a verification method.
Glaser argued that ‘grounded theory is multivariate. It happens sequentially, simultaneously,
serendipitously and scheduled’ (1998: 1). Grounded theory is suitable for dealing with many
research problems and requires that the researcher let go (pursue induction) and not attempt
to force models onto the data. ‘Grounded theory requires a tolerance for feeling out of control
while generating the beginning of a relevant main concern, a core category and sub-categories’
(Glaser, 1998: 11). It is a revolving step-method that starts the researcher from being ‘know
nothing to becoming an expert who will later become a theorist with a publication and with a
theory that accounts for most action in a substantive area’ (ibid, p 13).
Glaser (2001) further argued that as the most widely used methodology in the social sciences
grounded theory provides a set of steps that are closely linked or underpinned by the rigours of
good science. Grounded theory was identified by Glaser (2001) as an approach to both
qualitative and quantitative data collection and analysis in response to positivistic attempts to
force theory on data which created misfit and irrelevance. Glaser (2005) reassesses grounded
theory analysis and identifies the differences between data gathered in everyday life and that
collected for scientific study. Glaser and Holton (2007) are predominantly concerned with what
they label ‘exampling’ (that is providing case studies or examples of research where grounded
theory was used) and how this offers a learning experience no matter how experienced in
grounded theory the researcher may be. Exampling identified ‘the power and scope of classical
grounded theory … the global reach of the methodology and the varying levels of
methodological maturity of the authors … exampling … provides a rich range of theories that
have emerged largely from novice efforts at applying the methodology. They are theories from
which all may learn about the application of grounded theory’ (Glaser and Holton, 2007: 1).
Question: Glaser and Strauss
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 21 of 25 An Introduction to the Philosophy of Methodology
Discuss the main differences between Glaser and Strauss. Which approach to grounded
theory would best suit your research project?
Charmaz (2000) considered that even though Glaser and Strauss were not quantitative, each of
their perspective involved a post-positivist ontological position. Even though both Glaser and
Strauss were deeply divided regarding a number of issues relating to grounded theory, both
identified an external objective reality and neutral observer. Each argued for unbiased data
collection, differing levels of technical procedures and means of representing respondents as
accurately as possible. That said, both Glaser and Strauss have similarities with critical theory
and constructivism when they wish to give voice to their respondents and acknowledge how
their views differ from those of the researcher. They also consider that art as well as science
should be utilised as an analytical tool.
Constructivist grounded theory assumes a relativism of multiple realities and the co-creation of
knowledge. It emphasises natural settings and develops a non-post-positivistic stance. To
enable this grounded theory, research should not be rigid or prescriptive, should focus on
meaning to intensify interpretive understanding and desist from using a purely post-positivist
approach. Charmaz (2000) juxtaposes constructivist and post-positivist perspectives to
grounded theory but qualifies this when she acknowledges that this methodological approach
may be used with elements of both paradigms of inquiry and that distinctions between the two
exist on a continuum.
Identifying the Nature of Grounded Theory: Further Distinctions
Charmaz (2006) invites the reader to accompany her as she ascends the levels of analysis and
theoretical conceptualisation while ensuring one is firmly adhered to the data on the ground or
in the field. She identifies the main distinction between Glaser and Strauss in terms of
positivism and pragmatism; Glaser more aligned with the former and Strauss closer to the
latter. Glaser incorporated empiricism and rigorous coding and emphasised emergent
properties and discoveries, whereas, Strauss emphasised human agency, subjectivity and
emergent processes. That said, both believed in the emergence of theory and relationships
between objectivity and subjectivity; Glaser was more inductive and less rigid or more
phenomenological and Strauss and Corbin more prescriptive and positivistic in process.
However, Strauss based his perspective on symbolic interaction (see above) which as
discussed incorporates elements of pragmatism and phenomenology. Most grounded theories
are substantive because they deal with specific issues in delimited areas. However, these can
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 22 of 25 An Introduction to the Philosophy of Methodology
become formal theories ‘through generating abstract concepts and specifying relationships
between them to understand problems in multiple substantive areas’ (Charmaz, 2006: 8).
Following the conflict between Glaser and Strauss and Corbin regarding the nature of
grounded theory, others involved themselves in the debate and devised their own
understanding of grounded theory. For example, Denzin and Lincoln (1994) considered that
grounded theory can be post-positivist and/or constructivist. Annells contended that our
understanding of grounded theory is based on an ‘awareness of the method’s ontological,
epistemological and methodological perspectives’ (1996: 379) and that these may be broken
down into four paradigms of enquiry; positivism, post-positivism, critical theory and
constructivism. Indeed, that understanding of methodology and consequently grounded theory
are determined by one’s metaphysical assumptions. Through their own epistemological
positions scholars formulated their own understandings of grounded theory with most agreeing
that it was not a unified framework. Indeed grounded theory should be seen as a family of
methodologies that are distinct from other qualitative methodologies and share common
characteristics (Bryant and Charmaz, 2007a, 2007b; Charmaz, 2006; Creswell, 2007; Goulding,
2002; Hood, 2007). These differences between grounded theory involve what has been
identified as the holy trinity; that any grounded theory study requires theoretical sampling,
emerging categories through the constant comparison of data and the development of
substantive theory through the theoretical saturation of categories (Hood, 2007)
Definition Box: Grounded Theory
A grounded theory methodological approach requires:
theoretical sampling;
emergence of categories through comparative analysis;
substantive theory development.
A number of perspectives exist but the three main influences in the new streams of
grounded theory emanate from the social constructivist position (Charmaz, 2000, 2006),
postmodernist situational analysis (Clarke, 2002) and anti-post-positivist position (Bryant,
2002; Bryant and Charmaz, 2007a, 2007b). Each transcends the post-positivist
dimensions of both Glaser and Strauss; each places a premium on the relationship
between the researcher and researched and the multiplicity of those involved in the
construction of reality or theory (Charmaz, 2000, 2006). Furthermore situational analysis
focuses on social situations, the co-construction of knowledge and interaction between
researchers and researched (Clarke, 2002).
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 23 of 25 An Introduction to the Philosophy of Methodology
Conclusion
This chapter has dealt with grounded theory and identified that it involves developing theory
through data collection and analysis; it mainly uses qualitative data and recognises the close
connection between theory and practice. It has also outlined that some disputation exists
regarding the nature of grounded theory and identified that Glaser and Strauss considered that
‘grounded theory be flexibly interpreted and that researchers should use it in their own way that
is, as it fits their investigation’ (1967: 9). Charmaz viewed grounded ‘as a set of principles and
practices … (which) can complement other approaches to qualitative data analysis, rather than
stand in opposition to them … grounded theory serves as a way to learn about the worlds we
study and as a method for developing theories to understand’ these worlds and phenomenon
(2006: 9). Glaser and Strauss (1967) argued that theory is discovered as it emerges through
the data. For Charmaz (2006) neither ‘theories nor data are discovered they are part of the
same world as ourselves and we construct our grounded theories through our past and present
involvements and interactions with people, perspectives and research practices’ (2006: 10).
Theoretical perspectives offer an interpretive portrayal of the world under study and analysis
not an extant picture (2006: 10). Substantive theories are ontological constructions and
grounded theory involves phenomenological, interpretivist positions with pragmatist
underpinning. Charmaz (2000; 2006) deals with re-assessing meanings or interpretations of
theory in relation to methodology: positivist and constructivist types of grounded theory are
juxtaposed to illustrate how different modes of analysis stem from contrasting starting points.
Finally, reflection is dealt with in terms of grounded theory processes.
Furthermore, grounded theory raises difficulties regarding precise definitions of induction and
deduction and the point where the former begins and the latter ends (and vice versa) and the
grey area between the two. Alfred Marshall argued that ‘(y)ou make all your contrasts rather too
sharply for me. You talk of the inductive & deductive methods: whereas I contend that each
involves the other & that historians are always deducing, & that even the most deductive writers
are always implicitly at least basing themselves on observed facts’ (Marshall, cited in Coase,
1995: 169). Ultimately, Marshall wished to emphasise the mutual dependency of induction and
deduction. Strauss and Glaser aimed to do the same but each with different weightings. As
with the problems of delineation between induction and deduction the same may be said in
respect of emergence and forcing. Ultimately, one may consider that grounded theory should
be interpreted as it was by Glaser. ‘By its very nature grounded theory produces ever opening
and evolving theory on a subject as more data and new ideas discovered. This nature also
applies to the method itself and its methodology’ (Glaser, 1978: ix). Grounded theory is a
SAGE SAGE Research Methods
©2013 SAGE Publications, Ltd.. All Rights Reserved.
Page 24 of 25 An Introduction to the Philosophy of Methodology
flexible methodological approach and should be applied to specific research projects. Ideas will
emerge from data through interpretation but theoretical sensitivity and pre-conceptions will also
ensure a level of forcing; as with inductive and deductive approaches, pure emergent or forced
studies are impossible. Because of the nature of human existence and ontological and
epistemological relationships with phenomenon, a certain level of forcing as well as subjectivity
in relation to interpretation and emerging ideas necessarily must exist. That said, grounded
theory takes into consideration these difficulties and provides a methodological approach that
may be used in a post-positivist, critical theory or constructivist/participatory fashion.
Further Reading
http://dx.doi.org/10.4135/9781473957633.n9
[/read]
