INTRODUCTION: THE NATURE OF CASE STUDY RESEARCH
According to Stake (2000: 345), « case studies have become one of
the most common ways to do qualitative inquiry ». Case studies are the
source of some of the foundational work in organization theory. Classic examples include Patterns of Industrial Bureaucracy (Gouldner,
1954), Union Democracy (Lipset, Trow and Coleman, 1956) and Men
Who Manage (Dalton, 1959). These case studies are considered
exemplary by contemporary writers holding quite different epistemological preferences (e.g., Eisenhardt, 1991; and Dyer and Wilkins,
1991). After a decline in the late 1960s, there has been renewed interest in case study research since the 1980?s both in the US and Europe
(Stablein, 2006). Despite renewed popularity, case study « is a well-used term that has many meanings » (Stablein, 2006: 359). The notion
of case itself is multiple and debated (e.g., Dyer and Wilkins, 1991;
Eisenhardt, 1991) and has even motivated a book What Is a Case?
(Ragin and Becker, 1992).
In the context of this special issue, we use a broad definition of case
study research as the study of at least one case, a case being a
bounded system. This definition does not exclude any data collection
techniques and is less restrictive than that proposed by Yin (2003) who
excludes archival analyses and historical studies because they do not
include data from living people. Thus, from our perspective, case studies can be ethnographies, as well as historical studies (Vaughan,
1992). Moreover, although qualitative inquiry dominates (Stake, 2000),
case study research can also include quantitative data from surveys or
archival records. Although the use of multiple methods is considered
one of the strengths of case study research (Yin, 2003), a single data
collection method is also possible. The definition of a case as a bounded system simply requires a researcher to focus on the details of a
case and to analyze its context —it does not a priori restrict the methods used to achieve this.
Case study research can also be conducted within a positivist (e.g.,
Eisenhardt, 1989, 1991; Yin, 2003) or an interpretive (e.g., Lincoln and
Guba, 1985) tradition and may serve a variety of purposes. Stake
(2000) distinguishes between intrinsic and instrumental case studies.
An intrinsic case study aims to provide a better understanding of one
particular case. An instrumental case study is designed to provide
insight into an issue or to develop generalizations. When several
instrumental case studies are involved, Stake (2000) labels the design
a collective case study. Thus, cases can be studied for intrinsic or
instrumental purposes, and they can be aimed at interpretive understanding of multiple social realities (Guba and Lincoln, 1994), at theory building (Eisenhardt, 1989; Yin, 2003) or theory testing (Campbell,
1975; Yin, 2003; Bitekhtine, forthcoming).
The key element of our definition, like that of Stake (2000), is the focus
on the case —the specific entity under study. This can be a person, a
group, an organization, a relationship, an event, a process, a problem
or any other specific entity. It is a bounded system (Stake, 2000)
although it may not always be easy to determine where the case ends
and the context begins (Goode and Hatt, 1952, in Stake, 2000).
Whether multiple or not, aimed at theorizing or not, the logic of case
study research is idiographic. Even for multiple case studies aimed at
theorizing, the logic involves drawing inferences from similarities or differences in patterns between pairs of cases. This differs from both survey and experiments where inferences are made by comparing central
tendencies. The logic also differs from qualitative research that uses
data unrelated to a specific context. One consequence of this type of
research is that each case can be at least partially described as a
whole through one or more stories or narratives that will constitute
either the main published output of the study (Stake, 2000), or at the
very least, an unpublished step in the analysis (Yin, 2003).
Finally, our definition does not require that all cases within a collective
set be conducted in exactly the same way. Some cases can be included in the research design for specific purposes and conducted with
less intensity than others. For example, Leonard-Barton (1990) suggests the use of retrospective cases to complement a longitudinal
ethnographical approach. Glaser and Strauss (1967) also recommend
partial investigation to enrich some categories.
Thus, our definition of case study research does not aim to hide the
variety in strategy, design, purpose or epistemology, nor the possible
conflict between perspectives. Neither does it aim for unification. Yet,
we believe that there is sufficient commonality in practices and interests
among all case study researchers that they can learn from and appreciate other work that falls under the broad umbrella we have defined.
The six articles in this special issue reflect our broad definition of case
study research. Some authors are more positivist, while others belong
rather to interpretative or constructivist traditions. The six articles do
not cluster around a particular theme, nor do they provide a comprehensive panorama of current case study issues. This is rather a collection of articles that highlights particular issues that are often faced,
but that are perhaps less frequently discussed explicitly in methods
texts. In full compatibility with the case study tradition, many of these
articles build on unique personal experiences to draw their lessons.
We believe that these articles will resonate with the experiences of
senior organizational researchers, and we hope they will also offer
assistance to newer researchers.
In the remainder of this paper, we will first briefly introduce the six articles individually. We will then draw attention to some common themes
that emerge from a consideration of the entire collection before concluding with some suggestions for future thinking on doing case study
research in organizations.
The first article by Simon Down, Karin Garrety and Richard Badham
and the second article by Veronika Kisfalvi both analyse the consequences of researchers? emotions. In their paper « Fear and Loathing
in the Field: Emotional Dissonance and Identity Work in Ethnographic
Research », Down and colleagues discuss the varying reactions of different members of the research team to people and episodes encountered on their research site. The analysis of the researchers? field notes
and personal reflections provides striking evidence of the influence of
identity and emotions on fieldwork, and of the advantages of surfacing
and discussing these issues.
Despite a shared interest in emotions, Kisfalvi comes to this issue from
a different angle. Whereas Down and colleagues build on the sociology and social psychology literatures to examine emotional dissonance
during field work, Kisfalvi builds on literature from psychoanalysis and
neuroscience to establish how feelings interact positively with reasoning. In a lively « confessional tale » (Van Maanen, 1988: 73), she then
describes how anxiety and later mourning accompanied her ethnographic journey from first access to final draft. Kisfalvi argues that subjectivity and emotions, once examined, bring a deeper level of understanding and objectivity to her findings. Together, these two articles
provide very interesting and complementary contributions on the role
The article by Attila Bruni also draws attention to the researcher?s position in the field but focuses exclusively on the process of entry. In his
article, « Access as Trajectory: Entering the Field in Organizational
Ethnography », Bruni builds on an unusual experience to reframe the
negotiation of access. He first argues that the way in which the
researcher plays along with site expectations is more important than
image per se for obtaining access. He notes further that the process of
negotiating access can itself become an important moment of observation that reveals organizational processes since it reflects actors?
logics and practices. Access is thus considered as a trajectory in which
each encounter is both an occasion for new negotiations and an
opportunity for observation.
Geneviève Musca?s paper on longitudinal embedded case study
designs also describes access as an ongoing process, but adopts a
different perspective on the researcher?s position in the field. While
Bruni immediately slipped into the job of medical secretary that he did
not know, Musca avoided intervening in an activity she knew very well.
Although adopting a different perspective from previous authors,
Musca also reflects on her research practice and provides some interesting descriptions of her fieldwork experience. She offers some useful guidelines for conducting longitudinal embedded case study
research, covering the advantages and limits of this design, the negotiation of access, the definition of units of analysis, data collection, data
analysis and presentation, and approaches to ensuring validity.
The last two articles focus on the potential of unusual data sources for
organizational research: visual data in the article by Aylin Kunter and
Emma Bell, and secondary qualitative data in the article by Didier
Chabaud and Olivier Germain. The articles present the advantages of
these data sources, provide typologies of ways of using them, and
offer recommendations for their appropriate use. For example, Kunter
and Bell point out that visual data are able to capture elements of culture that are very difficult to express in words. However, they identify
several challenges relating to access, ethical considerations and
modes of analysis. They further show that visual data can be used for
reflexive practice and illustrate this with photographs from their own
The lack of a direct relationship with the field is one concern raised by
critics of the reutilization of qualitative data, the approach proposed by
Chabaud and Germain. These authors first show that the boundaries
between the reutilization of qualitative data and more well-established
practices can be ambiguous (e.g., the delegation of data collection to
assistants), and that a large variety of forms of reutilization of qualitative data exist, including some that are rarely debated (e.g., the development of theory from data collected for another purpose by the
researcher). They conclude that from most epistemological perspectives, the reutilization of qualitative data is a legitimate practice. However, a detailed evaluation is required to determine whether the available data is relevant and complete enough to address the research
The set of articles in this special issue offer a range of perspectives on
doing case study research in organizations. They review different
sources of data, they offer ideas about obtaining access to that data,
and they reflect on how the researcher?s positioning in the field may
influence the quality of that data. Doing case study research also raises a number of other questions that are perhaps less well-covered by
the present collection but that warrant further discussion. We will
briefly draw attention to three of these here.
The first issue concerns analysis and interpretation. Among the articles
in this issue, Musca looks at the analysis strategies available for
embedded longitudinal studies, and Kunter and Bell examine the challenges of analyzing visual data. In a more personal vein, Kisfalvi
describes how the computer-aided coding process she engaged in distanced her from the emotional impact of her field observations. She
found that she needed to recapture those emotions in order to translate the codes into a valuable interpretation.
However, there is certainly room for more reflection about the role and
impact of computerized aids in qualitative data analysis. Increasingly,
case study articles are reporting the use of tools such as NVivo and
Atlas.ti. Yet, with rare exceptions (e.g., Malina and Selto, 2001), these
reports do not provide much detail on whether and how the tools contributed distinctively to the interpretation. The assumption seems to be
that computer aids merely mechanize and simplify manual coding but
that the underlying cognitive and interpretive processes of analysis are
unchanged. This may perhaps be the case. On the other hand, literature on the use of technology for other tasks (e.g., Bilda and Demirkan
 for computer-aided design) suggests that the impact could be
more profound. One might ask for example whether the structures and
capabilities of software programs alter the way in which analysts consider their data. While there have been a number of comparative evaluations of different software tools published (Weitzman and Miles,
1995; Bournois, Point et Voynnet-Fourboul, 2002), there has been little study of the effects of their use on the interpretations that are generated.
Second, the articles in this special issue pay relatively little direct attention to the challenges of writing up case study research. The need to
present enough data to carry conviction while respecting normal page
limits poses particular problems for publication in traditional academic
journal outlets. These are compounded with multiple case study
designs. Golden-Biddle and Locke?s (2007) book, Composing Qualitative Research, offers a very interesting and instructive analysis of successful exemplars of qualitative research articles published in the
major management journals. While case study researchers can learn
from these practices, there is room for innovation beyond the models
established by others. For example, as Kunter and Bell suggest, electronic media (like this journal) may offer scope for creativity unavailable
from more traditional outlets 
Note however that in an encouraging
development, the.... Kunter and Bell give the example of
Ruby?s (2006) web-based ethnographic study of the Oak Park community that innovatively combines text and visual materials such as
photography and video.
Finally, a preoccupation that is only indirectly touched on in the articles
in this special issue concerns the appropriate criteria for evaluating the
quality of case study research. The journal Organizational Research
Methods has a forthcoming special issue on this topic that will be very
welcome. The concern about criteria is delicate in a domain where
multiple epistemological paradigms coexist. Our own experience suggests that some of the traditional difficulties that case study
researchers may have had in publishing their work in the more highly
rated journals may be partly due to disagreement among case study
researchers themselves about what criteria are relevant.
This is not to argue that case study researchers should all agree. However, it would be helpful if they were more explicit about what their
epistemological beliefs are and how these beliefs affect their understanding of appropriate quality criteria. The journal MIS Quarterly
(Markus and Lee, 1999, 2000a, 2000b) innovated by publishing a
series of qualitative research articles in which authors were asked to
pay particular attention to defining the evaluation criteria they considered appropriate in the light of their perspective. Indeed, the consistency between epistemological perspectives, quality criteria and
research methods is probably more important than a given perspective
or method per se. A broad variety of approaches to case study
research can thus be recognized as legitimate if done well in their own
terms, including those offering rich and insightful stories (Dyer and
Wilkins, 1991) as well as those aiming to generate strong theoretical
propositions (Eisenhardt, 1991).