Application of Qualitative Research Methodology for Developing Social...
Application of
Qualitative Research Methodology for
Developing Social Work Practice Models
This article highlights the need to develop practice models in the field of social
work, so as to contribute to the overall development of social work knowledge. It
is based on the assumption that both qualitative and quantitative methodologies
rest on separate epistemological positions, which result in different goals of so-
cial research, as well as in divergence with regard to the manner in which the two
methodologies pursue social knowledge. The article also recognises the signifi-
cant role of qualitative research methodology in developing social work practice
models based on empirical research, and discusses the nature of research design
decisions required of this methodology. The essential features of qualitative re-
search methodology, resulting from its epistemology, are first presented, so as to
subsequently reflect on the essentially emergent nature of qualitative research de-
sign decisions, and their application to the development of social work practice
models. Prior to discussing the nature of qualitative research design decisions,
this paper examines the essential features of qualitative research methodology,
and outlines the nature of its contribution to knowledge development.
Dr. Rajshree P. Mahtani is Reader, Unit for Social Policy and Social Welfare
Administration, Tata Institute of Social Sciences, Mumbai, Maharashtra, India.
This article highlights the need to develop practice models in the field
of social work, so as to contribute to the overall development of social
work knowledge. It recognises the significant role of qualitative re-
search methodology in developing social work practice models based
on empirical research, and discusses the nature of research design de-
cisions required of this methodology. The essential features of quali-
tative research methodology, resulting from its epistemology, are first
presented, so as to subsequently reflect on the essentially emergent
nature of qualitative research design decisions, and their application

50 Rajshree P. Mahtani
to the development of social work practice models. The specific na-
ture of research design decisions, along with examples, are examined
in relation to each stage of the qualitative research process.
• developing the initial conceptualisation of the research concern
pertaining the practice area focused upon,
• selecting the research setting and the researched,
• choosing data collection methods, and
• selecting and developing approaches and procedures for
analysis to arrive at the concluding conceptualisation of the
resulting empirically developed practice model.
This paper concludes with a discussion on the scope of social work
practice models developed empirically within the framework of qual-
itative research methodology.
Those in the field of social work are continuously confronted with
complexities within their specific areas of practice. These complexi-
ties result from several factors. Collectively, social work practitioners
in India work in a variety of field situations, characterised by diver-
gence in terms of the
• constituencies in whose interests they work, which are typically
segmented in a variety of ways: for example, a constituency may
be perceived as an individual, or a community, or even as a
particular vulnerable and marginalised group to which the
individual/community belongs; alternatively, a constituency
may be perceived as a particular issue of concern (environment,
education, availability of basic amenities, etc.);
• wide range of ideological approaches adopted for problem-
solving and social change, and the micro/macro levels at which
the change-efforts are directed;
• multiple types of interventions used; and
• several other groups and constituencies, as well as the socio-
political environment intervened with in order to benefit the
particular constituency of concern.
All the above-stated factors interact together to create specific con-
texts of social work practice, each with its own unique features. At
present, many of these contexts of social work practice are still in the
stage of exploration in India, as the problems/issues addressed are, in
themselves, large and complex. Complexities in social work practice
are further compounded as the process of conceptualising the problems/

Application of Qualitative Research Methodology... 51
issues addressed is largely interwoven with the process of problem-
solving, that is, as social work practice intervenes in its areas of con-
cern, it further specifies and even expands its conceptualisations of
these areas of concern. Further, the multi-disciplinary nature of social
work knowledge contributes to the complexity in its practice. Often,
problem-solving efforts result from attempts to integrate and
contextualise diverse social science theories with knowledge derived
from practice experiences. All these complexities, inherent in social
work practice in India are reflected equally within social work educa-
tion, with the result that the social work profession in India is still
struggling to define itself.
Over the years, through cumulative experiences, social work prac-
titioners have developed conceptualisations, which guide their work
within their particular contexts of field practice. These conceptua-
lisations pertain to both the problems/issues they address, as well as
the interventions they use to bring about change. These practice-
based conceptualisations may be termed as practice 'wisdom', as they
are in the nature of loosely-held guiding principles, which result di-
rectly from practice experiences. Unfortunately, such practice wis-
dom has largely not found its way into the formal body of social work
knowledge. It is only when practice wisdom is located within, and in-
tegrated with the formal body of social work knowledge that it is
transmitted through education and extension training to social work
practitioners working in other contexts, and to future generations of
social workers. The development and dissemination of practice-based
knowledge, and its integration into the formal body of social work
knowledge is required for the vibrant and cumulative growth of the
field of social work.
For practice wisdom to be assimilated into the formal body of so-
cial work knowledge, experiences pertaining to a specific, localised
field situation need to be conceptually integrated with other such situ-
ations to develop a practice model applicable to a class of similarly
defined field situations, which symbolise a context of social work
practice. For example, in the context of social work practice pertain-
ing to disaster management, practice wisdom can be found in the
work of a particular agency working on a specific disaster manage-
ment programme. However, a practice model on disaster manage-
ment has the potential to be applicable to multiple agencies working
on the management of various types of disasters. Thus, the develop-
ment of practice models requires that empirical experiences

52 Rajshree P. Mahtani
pertaining to a context of social work practice are conceptualised,
viewed as a dynamic process, and integrated analytically with the ex-
isting knowledge.
There is a need to formulate empirically-based practice models for
social work (Grasso and Epstein, 1992), as they contribute towards
the development of formal theories rooted in social work practice.
Practice models have the potential to enrich academic theories drawn
from the various social science disciplines, which are utilised in the
field of social work, as practice models are anchored in field experi-
ences. A theory can be perceived as an integrated network of generali-
sations, anticipated hypotheses, and assumptions (Hebsur, 1985).
Practice models, themselves, can be perceived to be theoretical state-
ments, which integrate assumptions, generalisations, and anticipated
hypotheses in relation to the conceptualised social phenomenon, the
interventions used, and their outcomes within a specified context of
practice. Research plays an essential and integral role in the develop-
ment of such practice models (Grasso and Epstein, 1992). However, it
is recognised that, in fact, there continues to be a gap between practice
and research in the field of social work (Riessman, 1994). Research
findings are seldom used to either conceptualise problem areas, or to
select interventions, or to evaluate them.
Historically, social work, like the various social science disci-
plines, has relied on structured, pre-determined research designs,
which are today classified as belonging to the 'quantitative methodol-
ogy of research'. Over the last 35 years or so, the various social sci-
ence disciplines and allied fields, such as social work, and the
management sciences, have recognised the growth of an alternative,
and evolving paradigm of research, namely 'qualitative research
methodology' (Bryman, 1989; Das, 1983; Morey and Luthans, 1984;
Van Maanen, 1982). The historical development of qualitative re-
search methodology has witnessed polarised arguments about the rel-
ative value of each type of methodology, that is, qualitative and
quantitative (Crompton and Jones, 1988; Das, 1983; Kirk and Miller,
1986; Miles, 1979; Reissman, 1994; Turner, 1988; Van Maanen,
1982). It is not the purpose of this article to focus on these polarised
arguments. This article rests on the assumption that both qualitative
and quantitative methodologies rest on separate epistemological posi-
tions, which result in different goals of social research, as well as in
divergence with regard to the manner in which the two methodologies
pursue social knowledge (Bryman, 1988b; Leininger, 1994). In brief,

Application of Qualitative Research Methodology... 53
qualitative research methodology is primarily directed towards the
goal of exploration and theory generation, while quantitative research
methodology is primarily directed towards the confirmation of
pre-formulated theory. Further, the epistemology of each methodol-
ogy indicates the philosophic assumptions made with regard to the re-
lationship between the researcher and the known, as well as
assumptions pertaining to the nature of knowledge about the social
world (Bryman, 1988b; Gould, 1999). The epistemological position
held shapes methodology by explicating the manner in which the so-
cial world can be known, or studied.
It is necessary to recognise epistemological differences in qualita-
tive and quantitative methodologies in order to prevent errors in their
usage. For example, every research endeavour should reflect a corre-
spondence between the epistemological position held, and the spe-
cific research methods used. Each type of methodology offers a
variety of specific research methods, which are based on its unique
epistemology. Therefore, it must be recognised that the decision to
mix research methods across the two methodologies within the same
research endeavour essentially disregards the intent and philosophic
assumptions of each methodology (Leininger, 1994). Further, recog-
nition of the divergent epistemological roots of the two methodolo-
gies is needed to prevent errors expressed during research-related
discussions which occur outside the realm of pure academics. One
such common error is that the terms 'qualitative research' and 'quali-
tative data' have, at times, been used interchangeably. This is a seri-
ous error as 'qualitative research' represents a distinct methodology,
while 'qualitative data' may be collected within the framework of ei-
ther qualitative research, or quantitative research. The same miscon-
ception exists with regard to the terms 'quantitative research', and
'quantitative data'.
Thus, it is crucial to recognise that qualitative and quantitative
methodologies present two distinct approaches to social inquiry. Fur-
ther, in addition to pursuing different goals of social research, each of
these two types of methodologies offers its own characteristic ap-
proach towards arriving at research design decisions for a particular
research endeavour. Qualitative research methodology emphasises
the need to view research design decisions as dynamic and emergent,
to suit the needs of the evolving conceptualisation of the research con-
cern. Viewing qualitative research designs as being essentially emer-
gent in nature is a significant departure from the conventional view of

54 Rajshree P. Mahtani
research designs within quantitative methodology, which views them
as blueprints, wherein research design decisions are finalised prior to
the implementation of a particular study. Prior to discussing the na-
ture of qualitative research design decisions, this paper examines the
essential features of qualitative research methodology, and outlines
the nature of its contribution to knowledge development. This is nec-
essary so as to locate the essentially emergent nature of qualitative re-
search methodology to empirically formulate social work practice
models, and thereby contribute to the development of social work
knowledge. While discussing the essential features of qualitative
methodology, key differences between the qualitative and quantita-
tive methodologies have been briefly referred to wherever necessary.
Qualitative research methodology encompasses a family of research
methods based on an interpretive, non-positivist approach to the study
of social phenomenon (Denzin and Lincoln, 1998). The interpretive
approach does not take for granted an external, fixed world of social
phenomena, and instead, it examines the processes by which the so-
cial world is constructed by its social actors. The interpretive ap-
proach is in the contrast to the positivist, natural science approach
underlying quantitative methodology (Bryman, 1988b; Denzin and
Lincoln, 1998). Quantitative methodology tends to view social reality
as external to social actors, and thus, it views social reality as an ob-
ject of study. On the other hand, the interpretive position inherent in
qualitative methodology focuses on discovering the processes by
which social actors perceive, give meaning to, and enact social phe-
A variety of rich intellectual streams have contributed to the devel-
opment of qualitative methodology as a distinctive approach to in-
quiry (Denzin and Lincoln, 1998). Though some differences exist
between these streams of philosophic thought, which contribute to
qualitative methodology, it is important to recognise that all the
streams adopt an interpretive position with regard to social inquiry. A
level of integration across the various streams has occurred in the field
of qualitative methodology, whereby this approach to social inquiry
can be described in terms of its essential, characteristic features,
which also distinguish it from quantitative methodology. Six key fea-
tures characterise qualitative methodology. Each of these six

Application of Qualitative Research Methodology... 55
characteristics is related to the others, and they are built on paradig-
matic assumptions, which set them apart from quantitative methodol-
ogy. Thus, it is necessary to view these six characteristics in totality,
rather than individually so as to assess the scope of qualitative meth-
odology. Each of these six characteristics is outlined below.
The first characteristic of qualitative methodology refers to its fo-
cus on uncovering the meaning of the phenomenon being studied
(Crompton and Jones, 1988; Dabbs, 1982; Das, 1983; Kirk and
Miller, 1986; Morse, 1998c; Van Maanen, 1979a). Qualitative meth-
odology seeks to do this by describing the phenomenon of interest in a
holistic manner and by locating it within its context (Bryman, 1988b;
Cassell and Symon, 1994; Denzin and Lingoln, 1998; Leininger,
1994; Morse, 1992; Noblit and Engel, 1992; Padgett, 1998). This ho-
listic search for meaning seeks 'thick descriptions' (Gilgun, 1992)
which not only provides detailed descriptions, but serves to convey a
complex network of processes by which meaning is created and en-
acted by multiple actors (Downey and Ireland, 1979). The search for
contextualised holistic meaning results in understanding a phenome-
non in its entirety from the viewpoint of multiple perceptions. This
emphasises the need to place individual actions, events, behaviours
and perceptions within the frame of a larger social and historical con-
text. In this way, qualitative methodology seeks an interpretive, un-
derstanding of human experience, and the socially constructed nature
of reality, as well as an understanding of the situational factors that
shape inquiry. It is necessary to emphasise that the term 'context' is
not used in a limited sense in qualitative methodology. Contexts are
viewed as layered influences, which shape the phenomenon of inter-
est. This characteristic of qualitative methodology leads to an
ideographic approach to analysis, wherein meaning is understood
within the boundaries created by defining both the space and time
within which the phenomenon is located. In contrast, quantitative
methodology emphasises the measurement of a phenomenon, and its
frequency of occurrence. Analysis in quantitative methodology is fo-
cused on identifying causal relationships between selected variables,
which are usually viewed as de-linked from other variables in their
contexts (Denzin and Lincoln, 1998). Such analysis is monothetic in
nature, which seeks to establish findings applicable outside the frame
of particular contexts. The ideographic context-specific approach, in-
herent in qualitative methodology, facilitates the dynamic study of

56 Rajshree P. Mahtani
The second characteristic of qualitative methodology aims to re-
duce distance between field data and theory by not starting a study
with a rigid, tightly structured, a priori framework. Instead, it aims at
evolving a theoretical framework as a result of the study (Bryman,
1988b; Cassell and Symon, 1994; Glaser and Strauss, 1967; Gilgun,
1992; Jauch, Osborn, and Martin, 1980; Lincoln, 1985; Miles, 1979;
Morse, 1992; 1997b; Padgett, 1998; Turner, 1983; Van Maanen,
1979a). Qualitative methodology is inductive in nature, which per-
mits empirical data to shape emerging concepts and theories. In order
to develop theories rooted, or grounded in data, there is minimal use
of pre-formulated analytical labels and hypotheses. Some qualitative
researchers do advocate the use of semi-structured conceptual frame-
works to initiate the search process, but they emphasise that these are
held tentatively and are continuously revised during data collection
and analysis (Eisenhardt, 1989; Miles and Huberman, 1994). How-
ever, the dominant view within qualitative methodology is that struc-
tured a priori frameworks are likely to constrain the generation of
insights, and result in findings, which exhibit an incomplete fit with
the phenomenon being studied. Hence, qualitative methodology ad-
vocates that the research process should emphasise the inductive de-
velopment of theories based on field observations. In contrast, in
quantitative methodology, pre-formulated concepts and theories are
usually the starting point for inquiry (Bryman, 1988b).
The third characteristic of qualitative methodology is related to the
first two as it is directed towards developing an emic (insider) per-
spective, that is, an understanding of meaning, from the perspective of
the researched (Denzin and Lincoln, 1998; Morse, 1992; Padgett,
1998). Thus, this third characteristic aims to reduce the distance be-
tween the researcher and the researched (Cassell and Symon, 1994;
Kirk and Miller, 1986; Miles, 1987; Mintzberg, 1979; Van Maanen,
1979a; 1982; Walizer and Wienir, 1978). Qualitative methodology is
used in naturalistic settings so as to link observations to the context of
the phenomenon being studied (Cassell and Symon, 1994; Denzin
and Lincoln, 1998; Padgett, 1998). The researcher is required to en-
gage with the research setting intensely, for a prolonged period of
time, and to enter into an active dialogue with the researched, so as to
gain a deeper understanding of the phenomenon of interest from the
view of multiple perspectives, and develop contextually relevant find-
ings. The expected role of the researched is that they should be
proactive in terms of defining their key issues in relation to the study.

Application of Qualitative Research Methodology... 57
Thus, the qualitative research process emphasises the need to be
transparent, and there is recognition that the research, itself, is a social
process, influenced by the choices and decisions made (Casell and
Symon, 1994). Underscoring the third characteristic is the acceptance
of the inherent subjectivity of all social processes, including the re-
search process. The argument here is that the search for objectivity is
largely misguided given multiple stakeholder perspectives, and it is
these varying perspectives and interpretations which are of value in
understanding behaviour. Thus, the researcher needs to actively con-
sider the reflexive character of the research process (Cassell and
Symon, 1994). In contrast, quantitative methodology adopts an etic
(outsider) perspective, where the researcher looks in on the social
world being studied (Bryman, 1988b). On the other hand, qualitative
researchers attempt to become insiders to the experiences of the re-
The fourth characteristic refers to the fact that qualitative method-
ology typically results in data that are open-ended and descriptive
(Das, 1983; Miles and Huberman, 1994; Mintzberg, 1979; Van
Maanen, 1982), as is implied by the absence of rigid, structured, ^pri-
frameworks. The purpose of description is to disclose the concepts
and complex patterns of relationships between the concepts observed,
and to explain why things happen as they do. Qualitative methodol-
ogy views social reality in processual terms, and thus, focuses on the
interlinkages between events, behaviours, perceptions, and actions
(Bryman, 1988b). From this perspective, qualitative methodology
also explores the interpretations of those researched with regard to the
factors that result in such interconnections. In contrast, quantitative
methodology tends to view social reality in static terms (Bryman,
The fifth characteristic of qualitative methodology refers to its ten-
dency to access multiple sources of data, which may require the use of
multiple methods of data collection (Cassell and Symon, 1994; Glaser
and Strauss, 1967; Kirk and Miller, 1986; Padgett, 1998; Sanday,
1979; Van Maanen, 1982; Walizer and Wienir, 1978). For example,
qualitative studies may use multiple methods to gather data from var-
ied sources, such as individual or group interviews, content analysis
of documents, and through observation. Accessing multiple sources
of data by using multiple qualitative methods within the larger frame
of qualitative methodology serves the purpose of triangulation. The
concept of triangulation within a single qualitative study refers to the

58 Rajshree P. Mahtani
convergence of multiple perspectives that can strengthen research
findings by presenting a comprehensive picture of complex social
processes involving multiple stakeholders (Cassell and Symon, 1994;
Padgett, 1998).
The sixth characteristic of qualitative methodology highlights that
it is not a neat and sequential research process, as it needs to be re-
sponsive to the continuously identified situations and circumstances
in the field setting (Bresnen, 1988; Mintzberg, 1979; Popay, Rogers,
and Williams, 1998; Vas Maanen, 1983). The research process inher-
ent in qualitative methodology is characterised by the interweaving of
data collection, analysis, and the emergent conceptualisation
(Bresnen, 1988; Eisenhardt, 1989; Miles, 1979; Miles and Huberman,
1994; Turner, 1998; Van Maanen, 1979a). Thus, this characteristic
emphasises the open, unstructured, flexible, and iterative research
process of qualitative methodology, where emergent conceptuali-
sations are modified and refined during overlapping data collection
and analysis (Bryman, 1988b; Cassell and Symon, 1994; Noblit and
Engel, 1992; Padgett, 1998). Many analytical decisions (such as what
specific lines of inquiry should be pursued) are made by the re-
searcher in the research setting during data collection. Additionally,
as analysis proceeds, the tentative findings are checked, modified and
confirmed in the field setting. The qualitative research process en-
hances the possibility of encountering the unexpected and unantici-
pated domains of inquiry related to the phenomenon being studied,
and also facilitates appropriate responses to changes required in the
research strategy as it unfolds. In contrast, the quantitative research
process in structured, sequential and pre-planned.
The six essential features, which characterise qualitative research
methodology, provide the framework within which research design
decisions are made in qualitative studies. The nature of qualitative re-
search design decisions are also indicative of the iterative, and re-
sponsive process of research in qualitative studies, which have the
potential to develop social work practice models. These are discussed
Within the framework of qualitative research methodology, research
design decisions are made on the basis of heuristics, or guidelines,
which provide direction to the research process. The qualitative

Application of Qualitative Research Methodology... 59
research process is iterative and flexible in nature because qualitative
research design decisions evolve during the research process, and
they are not rigidly-held a priori blueprints of decisions. The
interactive, flexible qualitative research process facilitates the devel-
opment of empirically relevant concepts and their interrelationships,
so as to generate theories anchored in data. The nature of qualitative
research design decisions are examined below in relation to the four
key stages of the research process, namely:
• conceptualising the research concern,
• selecting the researched,
• selecting the mix of data collecting methods, as well as shaping
the process of data collection, and
• selecting and developing procedures for analysis to facilitate the
development of field-based practice models.
Since qualitative research design decisions are not pre-determined
blueprints, and instead, take shape during the research process, it is
necessary for qualitative studies to make the research process trans-
parent by explicating not only the content of decisions, but also the
process resulting in these decisions.
Conceptualising the Research Concern
Qualitative researchers unanimously agree that the strength of quali-
tative studies is derived from the fact that they do not begin with struc-
tured, rigidly-held a priori conceptualisations of their research
concerns, or subjects of study (Bryman, 1988b; Creswell, 1998).
However, contemporary debate within qualitative research methodol-
ogy does address the issue pertaining to whether or not qualitative
studies should make explicit their a priori conceptualisations, even if
these are unstructured and tentative in nature (Miles and Huberman,
It is necessary to recognise that all research endeavours begin with
some a priori conceptualisations, which are either implicitly held, or
explicitly stated. Even if a research starts without an explicitly stated
conceptualisation, the implicitly held conceptualisation influences
the nature of research inquiry (Hebsur, 1985). Thus, both implicit and
explicit conceptualisations guide the search for data, the manner in
which concepts are identified, and the nature of interrelationships be-
tween concepts. The epistemology of qualitative research methodol-
ogy recognises that construction of meaning of the social world, as
well as the labels used to describe it, result from selective perception

60 Rajshree P. Mahtani
of experiences (Gould, 1999; Reissman, 1994). It is crucial to recog-
nise that this applies not only to how the researched ascribe meaning
to social phenomenon, but also to how researchers conceptualise their
research concerns. The conceptualisation of a research concern can-
not be separated from the researcher's experiences and perceptions
(Mahtani, 1999). Given the above reasoning, it is suggested that a pri-
ori conceptualisations should be made explicit, even if they are tenta-
tive in nature. Explicitly stated a priori conceptualisations will make
the search for data, as well as the development of concepts and their
interrelationships, a transparent process.
In qualitative studies, explicitly stated a priori conceptualisations
are unstructured, flexible, and held loosely, so as to facilitate system-
atic exploration of the research concern. Such a priori conceptu-
alisations typically integrate available knowledge regarding the
research concern, including both academic literature and field experi-
ences. These conceptualisations comprise of concepts and the interre-
lationships between these concepts. At this initial stage, a priori
concepts serve as sensitising concepts, which guide the initial search
for data (Clarke, 1997). For example, Mahtani (1999) demonstrates
how criteria were developed in the a priori conceptualisation of two
studies to identify and discern strategies of NGOs. These criteria
were, in fact, sensitising concepts, and they were used as heuristics to
identify strategic decisions taken by the NGOs studied. Relationships
between the sensitising concepts serve as lines of inquiry, which facil-
itate comparison of data. Qualitative research methodology recog-
nises that the unstructured, tentatively stated a priori conceptualisa-
tion undergoes change as a result of data collection and analysis. In
the inductive research process, analysis of data is directed towards de-
veloping the a priori conceptualisation into a structured, well-defined
theoretical statement of field-based experiences. Thus, the conclud-
ing conceptualisation incorporates multiple stakeholder views, and
integrates them holistically to represent the context of the field stud-
ied through qualitative research methodology.
Research design decisions pertaining to the a priori conceptuali-
sation focus upon the extent to which it is extensive in terms of the
sensitising concepts identified, and the lines of inquiry specified. As
data are collected, the a priori conceptualisation undergoes changes
to represent the data, till it emerges as a tight, structured conceptuali-
sation of field experiences, with clearly defined concepts and their in-
terrelationships. For example, Mahtani's (1992) study demonstrates

Application of Qualitative Research Methodology... 61
the manner in which the initial abstract concept of strategy was struc-
tured into 35 dimensions. It is in this sense that interaction between
conceptualisation, data collection and analysis was referred to while
discussing the essential features of qualitative research methodology.
Selecting the Researched: Theoretical Sampling and Saturation
The sampling process in qualitative research methodology differs sig-
nificantly from the sampling process in quantitative research method-
ology, which uses statistical sampling. Qualitative research
methodology presents an alternate perspective on sampling through
the process of theoretical sampling (Glaser and Strauss, 1967). Theo-
retical sampling is a systematic process in which heuristics are used to
select the researched for a particular study. One mistaken view of
qualitative research is that it is adequate to select 'a few' researched,
without specifying the criteria used to select them. This is a serious er-
ror. Qualitative researchers recognise that the generation of theories
grounded in data can only be achieved through the process of theoreti-
cal sampling and saturation (Bresnen, 1988; Bryman, 1988a;
Dunkerley, 1988; Glaser and Strauss, 1967; Turner, 1988). Special-
ised academic journals for qualitative research specify that theoretical
sampling and saturation should be essential aspects of the qualitative
research process (Morse, 1997b).
In contrast to statistical sampling, theoretical sampling does not
permit generalisation of findings to a larger population. The issue of
generalisation of findings is perceived differently in qualitative re-
search, where it is understood in theoretical rather than statistical
terms (Bresnen, 1988; Bryman, 1988a; Dunkerley, 1988; Glaser and
Struss, 1967; Turner, 1988). In attempting theoretical generalisation,
the researcher's efforts are directed towards ensuring that a particular
set of findings completely represent the emergent field-based theory
with regard to its concepts and their interrelationships. In other words,
theoretical sampling attempts to ensure that the concepts discovered,
as well as their interrelationships, completely represent the context of
the phenomenon studied, that is, the research concern. Thus,
heuristics pertaining to theoretical sampling decisions focus on ob-
taining divergent data to completely saturate the emergent,
field-based theory. Since the concepts and their interrelationships
comprising the emergent theory are shaped by empirical data, it fol-
lows that iteration occurs between the processes of theoretical sam-
pling, data collection, and analysis (Glaser and Strauss, 1967; Popay,

62 Rajshree P. Mahtani
Rogers and Williams, 1998). Thus, the exact number of researched in-
cluded within a particular qualitative study cannot be specified at the
start of the study. When the emergent field-based theory is bounded,
and additional data does not develop it further, then it is accepted that
saturation is complete (Glaser and Strauss, 1967). Until saturation oc-
curs, the emergent theory is incomplete, and concepts and their inter-
relationships are not clearly formed (Morse, 1997b).
Mahtani (1999) reports on the specific steps undertaken, and the
heuristics used to study the strategies of NGOs. Here, the process of
theoretical sampling and saturation were followed at two levels,
namely to select the NGOs, and to select the participants within each
NGO. Both groups comprised the researched. Theoretical sampling
aimed to maximise the differences among the NGOs included in the
study so as to obtain as many dimensions of the content of strategy, as
well as variations in the strategy formation process. Heuristics were
developed to maximise differences in strategy among the selected
NGOs, and these included divergence in terms of the:
• social/societal problems addressed;
• services offered, campaigns and programmes undertaken; and
• the constituency in whose interests they worked.
Saturation occurred when the addition of NGOs did not alter either
the dimensions identified to describe the NGOs' strategies, or the
emergent theory pertaining to the strategy formation process within
the NGOs. A similar process was followed for theoretically sampling
participants within each NGO.
Selecting Data Collection Methods
As discussed earlier in the section on the essential features of qualita-
tive research methodology, a mix of methods is used in qualitative
studies to incorporate diversity and multiple stakeholder views. Qual-
itative research methodology offers a variety of methods, such as nar-
ratives or oral histories, content analysis of documented data,
unstructured interviews, focus group discussions, and participant ob-
servation (Denzin and Lincoln, 1994; Hoffman, 1996; Morgan, 1988;
Morse, 1997a; Riessman, 1994; Walizer and Wienir, 1978). The
scope of data collection and analysis of each method is clearly speci-
fied, as is the manner in which they are to be used. Qualitative studies
need to choose a mix of methods suited to the needs of the concep-
tualisation of the study. Some research design decisions pertaining to
choosing specific methods for a particular qualitative study are:

Application of Qualitative Research Methodology... 63
i) What type of data can be collected by using a specific method?
ii) What are the scope and limitations of the chosen methods in
relation to the particular study, and how do they complement
one another?
iii) Are the methods to be used sequentially, or simultaneously?
iv) What methods will be used with the individual researched,
and what methods will be used with the researched as a
The answers to research design questions like those stated above,
and the reasons for the same indicate the heuristics used to arrive at
these dimensions.
Cross-checking of data obtained across methods is an essential
component of the qualitative data collection process. Recording the
data obtained is also an integral part of the qualitative data collection
process. Field notes of the various methods used form the basic raw
data (Bresnen, 1988; Glaser and Strauss, 1967; Miles and Huberman,
1994; Turner, 1988). Field notes indicate gaps in data collection, and
provide direction for further lines of inquiry.
Qualitative studies (for example, Bresnen, 1988; Sanday, 1979;
Van Maanen, 1982), are characterised by a responsive approach to
data gathering from multiple sources. The use of such multiple
sources of data enhances the validity of the findings (Snow and
Hambrick, 1980) and enables the researcher to cross-check the infor-
mation obtained so as to explore varying stakeholder perspectives
(Sanday, 1979). The iterative research process between data collec-
tion and analysis, characterised by continuously comparing the
emerging concepts and relationships with new data, facilitates the
identification of discrepancies between varying interpretations (Kirk
and Miller, 1986). Qualitative methodology, with its unstructured
mode of data collection facilitates achieving construct and contextual
validity as theoretical constructs and hypotheses, emerge from empir-
ical observations (Das, 1983; Glaser and Strauss, 1967; Kirk and
Miller, 1986; Miles, 1979; Miles and Huberman, 1994).
Conversely, the responsive nature of data collection in qualitative
studies makes it difficult to assess the reliability of the data obtained.
For example, the choice of specific participants is linked to their in-
volvement in particular events, and during interviews, the specific
questions asked are linked to the nature of responses received for ear-
lier questions. The result is that unique questions are asked of differ-
ent participants.

64 Rajshree P. Mahtani
Reliability, as understood in quantitative studies, is achieved when
it is assured that the specific questions asked and the operations in-
volved in the study can be repeated to arrive at the same findings
(Kirk and Miller, 1986; Krippendorff, 1980; Scott, 1984). It has been
suggested by Kirk and Miller (1986) that in qualitative studies reli-
ability can be enhanced by describing the detailed procedures adopted
so as to facilitate a repetition of these procedures. Hoffman (1996) of-
fers another viewpoint on the issue of reliability in qualitative studies,
which is defined as the consistency of responses received about the
same events on a number of different occasions by the same individ-
ual. Thus, within the context of these studies, the question of reliabil-
ity pertains to confirming whether the data obtained are indeed
factual, and the question of validity pertains to examining the proxim-
ity between the conceptual labels used to describe the data, and the
data itself.
Data collection experiences reported by Mahtani (1999) indicate
that there is immense possibility by which methods can be mixed in
qualitative studies to achieve triangulation within the framework of
qualitative research methodology. The need to mix methods results
from the recognition that research is a social process, involving vari-
ous actors. A mix of methods facilitates capturing varying perspec-
tives pertaining to the same research concern.
Selecting Procedures for Analysis
Data collection within the frame of qualitative methodology is la-
bour-intensive, and yields a large volume of descriptive data. Analy-
sis of such non-standardised, open-ended data calls for procedures
which are in accordance with the assumptions and characteristics of
qualitative methodology, and which permit exploration of meaning
(that is, meaning created by varying perceptions) in their context.
The purpose of qualitative data analysis is to disclose concepts and the
complex patterns of their interrelationships, so as to explain why
things happen as they do. Analysis in qualitative methodology is es-
sentially interpretive in nature, and follows a set of heuristics rather
than specific algorithms. Such algorithms are characteristic of the
wide range of computing techniques used in quantitative data analy-
Some qualitative researchers report descriptive data, without re-
sorting to any type of content analysis, or interpretation of data. How-
ever, specialised academic journals on qualitative research promote

Application of Qualitative Research Methodology... 65
qualitative research that synthesises and interprets data to develop
conceptual models or theories (Morse, 1999). This paper discusses
those procedures, which are in accordance with the interpretive pro-
cess of analysis in qualitative methodology, and which facilitate the
development of conceptual models or theories.
Qualitative researchers recognise that analytical decisions are
shaped by the initial, flexible conceptualisation of the research con-
cern, and by the emerging conceptualisation as it is moulded by data
(Eisenhardt, 1989; Glaser and Strauss, 1967; Miles, 1979; Miles and
Huberman, 1994; Mintzberg, 1979; Morse, 1994). Though analysis is
initiated and focused by the a priori conceptualisation, it is simulta-
neously closely interwoven with the data collection process. The
close intertwining of data collection and analysis is advocated by
qualitative researchers, as this serves to anchor the emerging concepts
and their interrelationships in empirical data (Glaser and Strauss,
1967; Morse, 1994).
Iteration between data collection and analysis involves the follow-
ing procedures. All information collected should be recorded in the
form of field notes, which represent the raw data. If voluminous data
are collected for each unit of analysis (that is, individual, family,
group, community, organisation, or any system), then these varied
data from multiple sources needs to be integrated into a holistic, de-
scriptive case report for each unit of analysis, so as to facilitate analy-
sis. The case report on each unit of analysis should be verified by the
researched to whom it applies. However, Morse (1998b) disputes the
necessity of this step in qualitative analysis, as she asserts that verifi-
cation of data occurs continuously during the data collection process.
Each piece of data gathered is cross-checked during the subsequent
process of data collection. While verification of data does occur con-
tinuously during the data collection process, it is still suggested that
the researched should be involved in the verification of the case re-
ports, as these integrate raw data, which is an analytic process. In fact,
the feedback received on the case reports serves to further expand the
data collected (Mahtani, 1999). Hence, it is recommended that verifi-
cation of data, in the form of such case reports, by the participants
should be perceived as a logical and natural termination of the data
collection process.
Once the case reports are completed, the, within-case analysis initi-
ated during data collection with the raw data, is continued with the
case reports. It is now possible to develop the coding categories to a

66 Rajshree P. Mahtani
higher level of abstraction. Procedures for removing and specifying re-
flective remarks have been described by Miles and Huberman (1994)
to enable the development of field-based concepts and their labels.
Qualitative researchers recommend that the conceptual labels used
for emergent concepts should correspond closely to the terminology
of the raw, narrative, descriptive data (Glaser and Strauss, 1967;
Miles, 1979; Morse, 1994). As discussed earlier, concepts included in
the a priori conceptualisation function as 'sensitising concepts'
(Clarke, 1997), and during data collection, they serve to guide the
search for data. During analysis, these sensitising concepts alert the
researcher to promising avenues of investigation pertaining to con-
cepts and their interrelationships. As data are collected, they are used
to refine, modify, or even replace the sensitising concepts, so that the
conceptual labels which emerge through inductive analysis closely
match the field context studied (Mahtani, 1999). As the density of
data in relation to the different concepts and their categories in-
creases, saturation becomes visible. Linkages between various con-
cepts are then established to identify the themes, or patterns of
relationships between concepts. Emergent themes are refined and
bounded at this stage.
Cross-case analysis is carried out after completion of the
within-case analysis for all the units of analysis included in each
study. During cross-case analysis, the concepts and themes identified
during the with-case analysis for all the units of analysis are examined
to ascertain the extent to which they are similar across the cases stud-
ied (Eisenhardt, 1989; Miles and Huberman, 1994). For this purpose,
coding categories need to be standardised across all the NGOs, wher-
ever applicable, and a systematic comparison is made of the emerging
themes (complex hypotheses) with evidence from each case in order
to assess the extent to which it fits the data. Cases which confirm the
emergent themes enhance confidence in the findings, while those
which do not support the emergent themes provided insights into how
the emerging theory can be modified and extended. This guiding prin-
ciple in analysis in qualitative methodology has been discussed by
Eisenhardt (1989) and Miles and Huberman (1994). The themes re-
sulting from cross-case analysis are at a Sager level of abstraction as
those formulated from the within-case analysis of each case.
Eisenhardt (1989) suggests that the evidence and procedures used in
qualitative analysis must be reported clearly so that readers may be
able to judge the strength and consistency of the relationship found.

Application of Qualitative Research Methodology... 67
The findings resulting from the cross-case analysis should be com-
pared with existing literature in order to identify similarities and dif-
ferences. This is an essential feature in order to generalise the findings
theoretically to a similar class of phenomenon (Eisenhardt, 19789;
Glaser and Strauss, 1967; Miles and Hubrman, 1994; Popay, Rogers
and Williams, 1998; Turner, 1983).
This paper began by recognising the need to develop social work practice
models, so as to contribute to the development of social work knowledge.
The significant role of qualitative research methodology in developing
social work practice models was also recognised. This paper now con-
cludes by discussing the potential applicability of qualitative research
methodology to empirically formulate social work practice models.
Concept development is the first, fundamental aspect of develop-
ing practice models. This involves the identification of configuration
of concepts identified from field data. These configurations of con-
cepts pertain to the analytical perspective held regarding the problem
of concern and its context, as well as to the nature of interventions uti-
lised and the range of changes effected. Further, conceptualisations of
practice models need to be process-oriented, as they demonstrate the
manner in which interventions are utilised to effect social change.
Thus, in addition to identifying configurations of concepts, practice
models also depict the inter-linkages between concepts, indicating the
manner in which concepts affect each other. A third core aspect of de-
veloping practice models is the need to incorporate multiple stake-
holder perceptions in relation to the problem, its context, and the
change process. These three core aspects of social works practice
models require that an exploratory stance is adopted to search for new
knowledge applicable to a particular context of field practice.
Qualitative methodology is intrinsically suited to exploratory stud-
ies aimed at concept development, because of its unstructured, induc-
tive, and flexible nature, and also because, it promotes greater
researcher-researched interaction (Dabbs, 1982; Das, 1983;
Dunkerely, 1988; Kirk and Miller, 1986; Morse, 1997b; Nobilt and
Engel, 1992; Padgett, 1998; Van Maanen, 1982). The nature of quali-
tative research design decisions facilities a process of inquiry, which
begins with a flexible, a priori conceptualisation of field practice.
This a priori conceptualisation provides the sensitising concepts and
the exploratory lines of inquiry, which initiate the process of data

68 Rajshree P. Mahtani
collection. The iterative characteristic of qualitative methodology
through the processes of data collection and analysis results in a re-
vised and refined conceptualisation in the form of a practice model
grounded in empirical field experiences. A crucial feature of the re-
sulting practice model is that it is located in its field context, as the
grounded conceptualisation intrinsically links the process of change
to the context in which the change occurs.
The process-orientation of practice models indicates the need to
ask 'how' and 'why' questions historically over time as well as over a
range of complex concepts, many of which are likely to be identified
during the course of data collection. Qualitative methodology is pre-
ferred while studying process and also when it is desired that the phe-
nomenon being studied should not be de-linked from its context
(Cassell and Symon, 1994; Das, 1983; Glaser and Strauss, 1967;
Jauch, Osborn and Martin, 1980; Noblit and Engel, 1992; Padgett,
1998; Van Maanen, 1979a). This methodology is particularly suited
to incorporating emic perspectives and explanations, as well as multi-
ple stakeholder views, as it provides a framework for using multiple
methods to collect data from varied sources. Preservation of chrono-
logical flow enables identification of the actual events and situations
that lead to specific outcomes. This permits the inductive analysis of
change as a process.
Thus, qualitative methodology facilitates the exploration of com-
plex phenomena in a manner that is holistic, and context-specific.
Qualitative methodology is conducive to the development of complex
constructs so that they closely correspond to the field situation. Fur-
ther, it permits the analysis of process, and it preserves the chronolog-
ical sequence in which such process unfolds. It is these features of
qualitative methodology, with the flexible nature of its research de-
sign decisions, that gives it the potential to be utilised to develop so-
cial work practice models. The development of such field-based
practice models will result in reducing the distance between practice
and research, and thereby strengthen social work knowledge.
Bresnen, M.
Insights on Site: Research into Construction Project Or-
ganizations. In A. Bryman (Ed.), Doing Research in Or-
ganisations, London: Routledge.

Application of Qualitative Research Methodology... 69
Bryman, A.
Introduction: 'Inside' Accounts and Social Research in
Organizations. In A. Bryman (Ed.), Doing Research in
Organizations, London: Routledge.
Quantity and Quality in Social Research, London: Unwin
Research Methods and Organization Studies, London:
Unwin Hyman.
Cassell, C. and
Qualitative Research in Work Contexts. In C. Cassell and
Symon, G.
G. Symond (Eds.), Qualitative Methods in Organiza-
tional Research: A Practical Guide, London: Sage Publi-
Clarke, A.E.
A Social Worlds Research Adventure: The Case of Re-
productive Science. In A. Strauss and J. Corbin (Eds.),
Grounded Theory in Practice, Thousand Oaks: Sage
Cresswell, J.W.
Qualitative Inquiry and Research Design: Choosing
among Five Traditions, Thousand Oaks: Sage Publica-
Crompton, R. and
Researching White Collar Organizations: Why Sociolo-
Jones, G.
gists should not stop doing Case Studies. In A. Bryman
(Ed.), Doing Research in Organizations, London:
Dabbs, J.M., Jr.
Making Things Visible. In J.V. Maanen, J.M. Dabbs, Jr.
and R.R. Faulkner (Eds.), Varieties in Qualitative Re-
search, Beverly Hills: Sage Publications.
Das, H.
Qualitative Research in Organizational Behaviour, Jour-
nal of Management Studies, 20(3), 301-314.
Denzin, N.K. and
Part III: Strategies of Inquiry. In N.K. Denzin and Y. Lin-
Lincoln, Y.
coln (Eds.), Handbook of Qualitative Research, Thou-
sand Oaks: Sage Publications.
Introduction: Entering the Field Qualitative Research. In
N.K. Denzin and Y.S. Lincoln (Eds.), The Landscape of
Qualitative Research: Theories and Issues, Thousand
Oaks: Sage Publications.
Downey, H.K. and
Quantitative Versus Qualitative: Environmental Assess-
Irleand, R.D.
ment in Organizational Studies, Administrative Science
Quarterly, 24(4), 630-637.
Dunkerley, D.
Historical Methods and Organisation Analysis: The Case
of a Naval Dockyard. In A. Bryman (Ed.), Doing Re-
search in Organizations, London: Routledge.
Eisenhardt, K.M
Building Theories from Case Study Research, Academy
of Management Review, 14(4), 532-550.
Gilgun, J.F
Definitions, Methodologies, and Methods in Qualitative
Family Research. In J.F. Gilgun, K. Daly, and G. Handel
(Eds.), Qualitative Methods in Family Research,
Newbury Park: Sage Publications.

70 Rajshree P. Mahtani
Glaser, B.G. and
The Discovery of Grounded Theory, Chicago: Aldine
Strauss, A.L.
Publishing Company.
Gould, N.
Qualitative Practice Evaluation. In I. Shaw and J.
Lishman (Eds.), Evaluation and Scoial Work Practice,
London: Sage Publications.
Grasso, A.J. and
Introduction. In A.J. Grasso and I. Epstein (Eds.), Re-
Epstein, I.
search Utilization in the Social Services: Innovations for
Practice and Administration, New York: The Haworth
Hebsur, R.K.
Theory Construction in Social Sciences: A Few Diffi-
culties and Limitations, The Indian Journal of Social
Work, 46(3), 347-358.
Hoffman, A.
Reliability and Validity in Oral History. In D.K.
Dunaway and W.K. Baum (Eds.), Oral History: An Inter-
disciplinary Anthology, Walnut Creek: Alta Mira Press.
Jauch, L.R.,
Structured Content Analysis of Cases: A Complementary
Osborn, R.N. and
Method for Organisational Research, Academy of Man-
Martin, T.N.
agement Review, 5(4), 517-525.
Kirk, J. and
Reliability and Validity in Qualitative Research, Beverly
Miller M.L.
Hills: Sage Publications.
Krippendorff, K.
Content Analysis: An Introduction to its Methodology,
Beverly Hills: Sage Publications.
Leininger, M.
Evaluation Criteria and Critique of Qualitative Research
Studies. In J.M. Morse (Ed.), Critical Issues in Qualita-
tive Research Methods, Thousand Oaks: Sage Publica-
Lincoln, Y.S.
The Substance of the Emergent Paradigm: Implications
for Researchers. In Y.S. Lincoln (Ed.), Organisational
Theory and Inquiry, Newbury Park: Sage Publications.
Mahtani, R.P.
The Strategies of Voluntary Organisations: A Study of
the Content of Strategy and the Process of Strategy For-
mation (Volume I and II), unpublished Doctoral Disserta-
tion, Ahmedabad: Indian Institute of Management (an
unpublished doctoral dissertation).
Application of Qualitative Research Methods to the Study
of Strategies in NGOs. Paper Presented at the Interna-
tional Conference on Advances in Qualitative Research
Methods at Edmonton, Canada (February 18-20, 1999).
Miles, M.B.
Qualitative Data an Attractive Nuisance: The Problem of
Analysis, Administrative Science Quarterly, 24(4),
Miles, M.B. and
Qualitative Data Analysis (Edition 2), Thousand Oaks:
Huberman, A.M.
Sage Publications.

Application of Qualitative Research Methodology... 71
Mintzberg, H.
An Emerging Strategy of Direct' Research, Administra-
tive Science Quarterly, 24(4), 582-589.
Morey, N.C. and
An Emic Perspective and Ethnoscience Methods for Or-
Luthans, F.
ganisational Research, Academy of Management Review
9(1), 27-36.
Morgan, D.L.
Focus Groups as Qualitative Research, Newbury Park:
Sage Publications.
Morse, J.M.
The Characteristics of Qualitative Research. In J.M.
Morse (Ed.), Qualitative Health Research, Newbury
Park: Sage Publications.
'Emerging from the Data': The Cognitive Processes of
Analysis in Qualitative Inquiry. In J.M. Morse (Ed.),
Critical Issues in Qualitative Research Methods, Thou-
sand Oaks, California: Sage Publications.
Editorial: 'Perfectly Healthy but Dead': The Myth of
Inter-rater Reliability, Qualitative Health Research, 7(4),
Editorial: The Pertinence of Pilot Studies, Qualitative
Health Research, 7(3), 323-324.
Editorial: Validity by Committee, Qualitative Health Re-
8(4), 443-445.
Editorial: What's Wrong with Random Selection? Quali-
tative Health Research,
8(6), 733-735.
Silent Debates in Qualitative Inquiry, Qualitative Health
Research, 9(2), 163-165.
Noblit, G.W. and
The Holistic Injunction: An Ideal and a Moral Imperative
Engel, J.D.
for Qualitative Research. In J.M. Morse (Ed.), Qualita-
tive Health Research, Newbury Park: Sage Publications.
Padgett, D.
Qualitative Methods in Social Work Research, Thousand
Oaks: Sage Publications.
Popay, J. Rogers, A.
Rationale and Standards for the Systematic Review of
and Williams, G.
Qualitative Literature in Health Services Research, Qual-
itative Health Research, 8(3), 341-351.
Riessman, C.K.
Preface: Making Room for Diversity in Social Work Re-
search. In C.K. Riessman (Ed.), Qualitative Studies in
Social Work Research, Thousand Oaks: Sage Publica-
Sanday, P.R.
The Ethnographic Paradigm(s), Administrative Science
Quarterly, 24(4), 527-538.
Scott, W.E., Jr.
The Development of Knowledge in Organisational Be-
haviour and Human Performance. In T.S. Bateman and
G.R. Ferris (Eds.), Method and Analysis In Organisa-
tional Research, Reston: Reston Publishing Company.
Shirley, R.C.
Limiting the Scope of Strategy: A Decision Based Ap-
proach, Academy of Management Review, 7(2), 262-268.

72 Rajshree P. Mahtani
Snow, C. and
: Measuring Organisational Strategies: Some Theoretical
Hambrick, D.
Methodological Problems, Academy of Management Re-
view, 5(4), 527-538.
Turner, B.A.
: The Use of Grounded Theory for the Qualitative of Or-
ganisational Behaviour, Journal of Management Studies,
20(3), 333-348.
: Connoisseurship in the Study of Organisational Cultures.
In A. Bryman (Ed.), Doing Research in Organisations,
London: Routledge.
Van Maanen, J.
: Reclaiming Qualitative Methods for Organisational Re-
search: A Preface, Administrative Science Quarterly,
24(4), 520-526.
: Introduction. In J. Van Maanen, J.M. Dabbs,Jr. and R.B.
Faulkner (Eds.), Varieties in Qualitative Research,
Beverly Hills: Sage Publications.
: Epilogue: Qualitative Methods Reclaimed. In J. Van
Maanen (Ed.), Qualitative Methodology, Beverly Hills:
Sage Publications.
Walizer, M.H. and
:- Research Methods and Analysis: Searching for Relation-
Wienir, P.L.
ships, New York: Harper and Row Publishers.
THE INDIAN JOURNAL OF SOCIAL WORK, Volume 65, Issue 1, January 2004