Theorising from Case Studies: Towards a Pluralist Future for International Business Research

Abstract

The literature on case studies, both in the field of international business (IB) and in the social sciences more generally, has tended to focus on the methods of data production and analysis suited to this research strategy. In contrast, in this paper we investigate methods of theorising from case studies. We seek to understand how case researchers theorise, and how future IB research might utilise case studies for theorising. By means of a qualitative content analysis of case studies published in Journal of International Business Studies, Academy of Management Journal and Journal of Management Studies, we construct a typology of theorising from case studies. Two dimensions of the case study, namely causal explanation and contextualisation, form the basis for our typology. We distinguish four methods of theorising – inductive theory-building, interpretive sensemaking, natural experiment and contextualised explanation – only the first of which has been widely used in JIBS in the period that we investigate. On the basis of our own qualitative analysis, we show the limitations of inductive theory-building, and argue that greater utilisation of the other methods of theorising would enhance the case study's explanatory power and potential for contextualisation. We argue for a more pluralist future for IB research.

Full text

Theorising from case studies: Towards a
pluralist future for international business
research
Catherine Welch
1
,
Rebecca Piekkari
2
,
Emmanuella Plakoyiannaki
3
and Eriikka
Paavilainen-Ma¨ntyma¨ki
4
1
Faculty of Economics and Business, Discipline of
International Business, The University of Sydney,
Australia;
2
Department of Management and
International Business, School of Economics,
Aalto University, Finland;
3
Department of
Business Administration, School of Economics,
Aristotle’s University of Thessaloniki, Greece;
4
Department of Marketing, Turku University,
School of Economics, Finland
Correspondence:
R Piekkari, Professor of International
Business, Department of Management and
International Business, Aalto University,
School of Economics,
PO Box 21230, FI-00076 Aalto, Finland.
Tel: þ358 9 4313 8734;
Fax: þ358 9 4313 8880
Received: 18 September 2009
Revised: 20 August 2010
Accepted: 25 August 2010
Online publication date: 16 December 2010
Abstract
The literature on case studies, both in the field of international business (IB) and
in the social sciences more generally, has tended to focus on the methods of data
production and analysis suited to this research strategy. In contrast, in this paper
we investigate methods of theorising from case studies. We seek to understand
how case researchers theorise, and how future IB research might utilise case
studies for theorising. By means of a qualitative content analysis of case studies
published in Journal of International Business Studies,Academy of Management
Journal and Journal of Management Studies, we construct a typology of theorising
from case studies. Two dimensions of the case study, namely causal explanation
and contextualisation, form the basis for our typology. We distinguish four
methods of theorising – inductive theory-building, interpretive sensemaking,
natural experiment and contextualised explanation – only the first of which has
been widely used in JIBS in the period that we investigate. On the basis of our
own qualitative analysis, we show the limitations of inductive theory-building,
and argue that greater utilisation of the other methods of theorising would
enhance the case study’s explanatory power and potential for contextualisation.
We argue for a more pluralist future for IB research.
Journal of International Business Studies (2011) 42, 740–762.
doi:10.1057/jibs.2010.55
Keywords: case theoretic approaches; secondary data source; theory–method
intersection
INTRODUCTION
The case study has an established place in qualitative international
business (IB) research.
1
A recent review of articles published in
four core IB journals over a 10-year period found case studies
to be the most popular qualitative research strategy (Piekkari,
Welch, & Paavilainen, 2009). This prevalence of the case study is
not surprising, given its potential to generate novel and ground-
breaking theoretical insights. Yet our contention in this paper is
that the theorising potential of case studies has not been fully
realised in the field of IB. We attribute this to the entrenched belief
that case research is suited only to inductive theory-building. In
this paper, we seek to challenge this belief by offering alternatives
to inductive theorising and broadening the possibilities in IB for
theorising from case studies. In order to do so, we consider how the
case study generates causal explanations and how it incorporates
context – two features of the case study that are often regarded as
Journal of International Business Studies (2011) 42, 740–762
&
2011 Academy of International Business All rights reserved 0047-2506
www.jibs.net

being incompatible. By challenging common
preconceptions about case studies, we see our audi-
ence as comprising not just qualitative researchers,
but also the wider community of scholars who
are often called upon to evaluate the theoretical
contribution of case research.
Our paper contributes to the debate over theoris-
ing in IB research, which recognises that alongside
the perennial epistemological dilemma faced by
social scientists – namely, how to develop robust
explanations about phenomena in the social world
– IB scholars have to contend with the question of
how to ensure that their theories are sensitive to
diverse national contexts. Consequently, we would
argue that IB is a highly appropriate field in which
to discuss the development of rigorous, yet context-
sensitive, theory. There is growing concern that, in
the pursuit of robust explanations, contextualisa-
tion has suffered. Greater use of qualitative research
has been suggested as a remedy for this imbalance,
thus placing approaches such as the case study
squarely on the agenda for IB theory. Yet our con-
tention in this paper is that, in IB research, the
dominant view of the case study as a tool solely for
inductive theory-building has restricted its theorising
potential, both in terms of generating causal
explanations and of contextualising theory.
In this paper, we challenge this dominant view by
constructing a typology that offers alternatives to
inductive theory-building. The first alternative
views the case study as a natural experiment for con-
firming or modifying existing theory. This method
attributes greater explanatory power to the case
study than does inductive theory-building. The
second alternative, by conceiving case research as a
form of interpretive sensemaking, affirms the value of
contextualisation to theorising. However, these two
alternatives are both potentially limiting, we argue,
because they set up a trade-off between the strengths
of internal validity on the one hand and thick
description on the other. In this paper, we outline a
third alternative – a recent development in the
methodological literature and new to IB – that
rejects this trade-off, and instead emphasises the
ability of the case study to generate contextualised
explanation. By comparing these four approaches to
theorising, we show that the differences between
each type are fundamentally epistemological and
philosophical (i.e., paradigmatic) in nature. Broad-
ening the possibilities for theorising from the case
study therefore requires an appreciation of how
these underlying paradigmatic assumptions both
enable and constrain empirical research.
We develop our arguments in this paper as
follows. First, we review growing concerns about
the decontextualised nature of theorising in IB
research, arguing that inductive theory-building
reinforces rather than resolves this dilemma. We
then detail our analytical approach: a qualitative
content analysis of case studies published in Journal
of International Business Studies ( JIBS),Academy of
Management Journal (AMJ) and Journal of Manage-
ment Studies ( JMS) that allowed us to examine how
researchers have conceptualised and utilised the
contextualising and explanatory potential of the
case study. This qualitative approach to textual
analysis, while rarely used in IB research to date
(Duriau, Reger, & Pfarrer, 2007), offers the strengths
of iterative conceptual development, contextua-
lised insights and access to the linguistic features of
texts. We then present the typology generated from
our content analysis, discussing first its foundations
in theory (i.e., the methodological literature) and
then its grounding in data (i.e., published case
studies). We conclude by specifying how future IB
researchers might enhance the theorising potential
of the case study by combining contextual richness
and explanatory rigour.
In this paper we will be using terms – notably
context, theory, explanation and causation – that
have been greatly contested in the social sciences,
yet whose meaning is too often taken for granted
by researchers. These terms will be developed
during the course of our discussion, but we will
offer our own definitions upfront. By context, we
are referring to the contingent conditions that, in
combination with a causal mechanism, produce an
outcome. Explaining a phenomenon we take to
mean showing what makes it what it is. Explana-
tion need not necessarily be causal (Ruben, 1990:
233), but causal explanations are our focus in this
paper given their centrality to the debate over the
theorising potential of case studies. An explanation
is causal if it makes claims about the capacities of
objects and beings to make a difference to their
world (adapted from Kakkuri-Knuutila, Lukka, &
Kuorikoski, 2008; Sayer, 1992). We take theory to
mean a form of explanation that offers a coherent,
examined conceptualisation of a phenomenon
(based on Sayer, 2000). Our subsequent discussion
in this paper will reveal that these definitions are
heavily influenced by critical realism, and that they
have profound implications for our understanding
of how to theorise from case research.
In the IB field, however, these foundational
elements of the scientific endeavour receive little
Theorising from case studies Catherine Welch et al
741
Journal of International Business Studies

scrutiny (for an exception, see Redding, 2005). We
contend that a reassessment of these fundamental
concepts is needed for the theorising potential
of case studies to be realised. As Sayer (1992) has
persuasively argued, methodology should not just
be regarded as a matter of choosing among diff-
erent methods of data production
2
and analysis;
rather, it is about choosing among competing
methods of theorising. Yet we contend that much
of the methodological literature on case studies
in IB, as well as in the social sciences generally,
has focused on methods of data production and
analysis rather than methods of theorising. In this
paper, we show that alternative methods of theoris-
ing from case studies are available, and argue for
their application to IB. Ultimately, by contrasting
different theorising methods for case researchers,
we are advocating paradigmatic pluralism (for
a similar argument, see Brannen & Doz, 2010;
Morgan, 1980; Van Maanen, 1995). We have been
influenced by contemporary philosophers of
science who argue that pluralism is a necessary
precondition for scientific discovery and theoreti-
cal advances (Kellert, Longino, & Waters, 2006);
accordingly, we argue that the IB field would bene-
fit from diversity in approaches to theorising.
To establish the need for greater diversity, we turn
to the growing concern that context has been
neglected in IB and management research.
CONTEXT AND THEORISING IN IB RESEARCH
We would argue that the ground we cover in this
paper – the tension between scientific explanation
and context – is a concern for any research, but that
it is particularly visible and pressing in IB, given
the field’s cross-border nature. How to account
for context has been a recurring, but unresolved,
question for IB scholars (Brannen & Doz, 2010;
Redding, 2005). For example, researchers on China,
facing an institutional environment very different
from the Western origins of most management
theories, have been conducting a lively debate on
the need for contextualised theories and research
processes (e.g., Child, 2000; Shapiro, Von Glinow, &
Xiao, 2007; Tsui, 2006). The internationalisation
of the general management community has also
sparked interest in contextualisation (see, e.g.,
Rousseau & Fried, 2001; Tsui, 2007). In this litera-
ture, context is typically defined as “the surround-
ings associated with phenomena which help
to illuminate that phenomena [sic]” (Cappelli &
Sherer, 1991: 56).
Advocates for “contextualized knowledge” (Tsui,
2004) concede that they face barriers to change in
the scholarly community. Above all, they point
to the entrenched belief that “context-free”, univer-
salist knowledge is superior to that of “context-
valid”, localised knowledge (Blair & Hunt, 1986;
Tsui, 2004; Whetten, 2009). Bamberger (2008: 844)
observes that dominant beliefs “may be forcing us
to overweight generalizability and, in the process,
underweight contextual sensitivity”. Suggestions of
how to encourage contextualised research range
from the modest – for example, adding context
effects as moderating variables (Whetten, 2009) –
to the more radical, such as Tsui’s (2006) call to
explore non-Western methodological tools. There
is an emerging consensus that “context-oriented”
qualitative research forms part of the answer
(Bamberger, 2008; Johns, 2006; Rousseau & Fried,
2001; Tsui, 2004).
However, proposals for modest change do not
confront the underlying reasons for the dominance
of decontextualised research: namely, the positivist
assumptions that are still taken for granted in
the IB field (Brannen & Doz, 2010; Jack, Cala
´s,
Nkomo, & Peltonen, 2008; Redding, 2005). Any dis-
cussion of contextualisation is necessarily embed-
ded in a complex web of beliefs concerning
the nature of theorising in the social sciences: the
question of how to contextualise is ultimately
about how to theorise, and the answer depends
on one’s philosophical orientation. The argu-
ments in favour of contextualising IB research
are therefore well established, but the solutions
are more contested, and ultimately highly value
laden.
By seeking to investigate how contextualising
and explaining can be brought together in the case
study, our paper addresses a gap in the existing
literature on contextualising IB research. To date,
this literature, while placing the need for more
qualitative research on the agenda, has not exam-
ined its theorising potential in any depth. Qualita-
tive research is referred to in very general terms,
with no differentiation among its many traditions,
and its contribution is simply assumed to lie in rich
description and exploratory, inductive theorising.
However, in relation to the case study, there are two
problems with these assumptions. The first is that
because the generalisability of case study findings is
low, its theorising potential is ultimately regarded
as inferior to that of hypothesis-testing research.
Case studies are therefore confined to the initial,
exploratory phase of research, and their potential
Theorising from case studies Catherine Welch et al
742
Journal of International Business Studies

for generating causal explanations is overlooked.
Second, while all qualitative research is commonly
assumed to be context sensitive, a strong trend
towards decontextualisation has in fact prevailed in
much case research. The rich context that is the
essence of a case study is ultimately regarded as a
hindrance to theorising. Since to theorise is to
generalise away from context, “explaining” and
“contextualising” are regarded as being fundamen-
tally opposed. In this paper, we challenge these
perceptions about theorising from case studies, and
suggest how context and explanation might be
reconciled. In this way, we are contributing not just
an enhanced understanding of the theorising
potential of the case study, but also a means of
enriching the context orientation of IB research. In
the next section, we detail how we ourselves used a
qualitative approach to develop our conceptual
understanding of the theorising potential of case
studies in IB.
QUALITATIVE CONTENT ANALYSIS OF
PUBLISHED CASE STUDY ARTICLES
We commenced this study with a broad research
question, namely “How do IB case researchers
theorise from case studies?” In this section, we
detail how and why we took the approach of a
qualitative content analysis, and how we selected
our dataset. Our approach to qualitative content
analysis would best be termed “directed”: that is,
the analysis commences with an initial coding
scheme (Hsieh & Shannon, 2005). In contrast to a
quantitative content analysis, even though the
codes are selected in advance, they do not remain
fixed during the analysis, but rather are refined
through successive iterations between theory and
data (Berelson, 1971; Ryan & Bernard, 2000). One
strength of this approach is therefore that it allows
for fresh conceptual understanding that is also
grounded in empirical data.
A key task in any form of textual analysis is to
decide on the appropriate sample; in other words,
which texts to analyse (Krippendorff, 2004). Our
journal selection followed the qualitative principle
of purposeful sampling, which allows the content
analyst to select the units of investigation relevant
to the study (Krippendorff, 2004). We initiated our
analysis with JIBS; as the leading journal in IB,
we can expect that the case studies it publishes,
although few in number (see Appendix), will
influence research standards in the field (Clark &
Wright, 2007). We examined the period 1999–2008
in order to capture the most recent case study
practices. We found little diversity in the methods
of theorising from case studies in JIBS. At this point,
we added a research question, namely: “What are
the alternatives to theorising from case studies, and
what is their potential contribution to IB research?”
Consequently, we expanded our dataset to
include two management journals – AMJ and JMS
– that are comparable with, yet contrasting to,
JIBS. Like JIBS, they are highly ranked journals
that influence, and also publish articles on, IB; yet
they have had different editorial policies towards
qualitative research. In contrast to JIBS,AMJ has
published numerous editorials (e.g., Gephart, 2004;
Pratt, 2009; Suddaby, 2006) encouraging and
providing advice on qualitative submissions. JMS,
the most highly ranked European-based manage-
ment journal,
3
has also promoted discussion on
qualitative research standards (e.g., Shah & Corley,
2006). The inclusion of these two journals provided
us with greater diversity of theorising practices,
while at the same time still limiting ourselves to
publications that are of relevance to IB scholars.
A central challenge in assembling our dataset was
to identify all case studies in the three journals. We
categorised articles as case studies if they met the
definition proposed by Piekkari et al. (2009: 569):
“a research strategy that examines, through the use
of a variety of data sources, a phenomenon in its
naturalistic context, with the purpose of ‘confront-
ing’ theory with the empirical world”. In order to
identify a case study, we read the entire paper, not
just its title and abstract. All articles were cate-
gorised independently by at least two members of
the research team, and differences in opinion led us
back to the “raw data”, the case study articles
themselves. Classification of some of the articles
was hampered by the omission of essential details –
even a methodological section. Our analysis also
confronted the issue that “case study” is a contested
term, and difficult to distinguish from other quali-
tative approaches (Wolcott, 2001), so our categor-
isation of articles did not always agree with that of
their authors.
4
Having settled on a final dataset of 199 case
studies (see Appendix), we then proceeded by
analysing their contents qualitatively. In contrast
to quantitative approaches, whose concern is the
enumeration of categories, the aim is a holistic
interpretation of the text that goes beyond its literal
meaning. This enabled us to remain consistent
with the objective of our paper, namely to analyse
methods of theorising “in context” rather than
“away from context”. A more quantitative content
Theorising from case studies Catherine Welch et al
743
Journal of International Business Studies

analysis would not have been meaningful, because
some authors used methodological terms very
loosely. For example, “grounded theory” was com-
monly mentioned as a technique for data analysis
and coding, but only rarely as a methodology for
inductive theory-building. Therefore a frequency
count of this term would have misrepresented the
popularity of inductive theory-building approaches
in the dataset. Instead of frequency counts, we used
other techniques to aid our categorisation of texts
(Berelson, 1971), chiefly intra-content comparison
(i.e., comparing different parts of the same text,
which allowed us to detect common themes as well
as inconsistencies), comparisons between different
texts (i.e., comparing across articles), and compar-
ison of the textual content with a standard (in our
case, our evolving typological categories).
As well as classifying the case studies according
to their method of theorising, we wrote an
analytical memo about each article, which helped
us to proceed systematically and consistently (Miles
& Huberman, 1994). The memo addressed the
following questions.
Do the authors of the case articles state the
theoretical objectives of the study and, if so,
how?
How do they relate theory to empirical data?
Do they integrate the research context into the
theoretical explanation of the case?
Do they refer to methodological sources, and
which ones?
Do they generalise from case data?
Do they make causal claims?
Do they analyse the case holistically, or construct
process explanations?
What theoretical language do they use?
Given the focus of this paper, we did not analyse
other aspects of the case study design and write-up,
such as the methods of data production or analysis.
In sum, the memos encouraged us to take advan-
tage of the strengths of qualitative research to focus
on the linguistic elements of the texts, representa-
tions of the theorising process, and the assump-
tions made by their authors.
At least two of us coded each article, indepen-
dently first and then jointly, and we conducted two
separate rounds of coding. The repeated double-
coding of all units (i.e., articles) is not a common
practice in content analysis, because of the time
and cost involved (see Kolbe & Burnett, 1991;
Potter & Levine-Donnerstein, 1999). We took this
step because our objective, consistent with our
qualitative approach, was to enrich our analysis
with multiple interpretations and achieve an inter-
subjective understanding across coders (for a simi-
lar argument, see Barbour, 2001; Krippendorff,
2004; Yardley, 2000). Any divergence in coding
was talked through, as qualitative content ana-
lysis values “the content of disagreements and the
insights that discussion can provide for refining
coding frames” (Barbour, 2001: 1116). Articles that
seemed to challenge our existing coding scheme
led us to further scrutinise and elaborate on our
evolving categories (for a similar approach to
coding, see Locke & Golden-Biddle, 1997).
Accordingly, our approach to content analysis
enabled successive iterations between theory (i.e.,
the methodological literature) and our dataset.
Each iteration led to a modification to and enrich-
ment of our conceptual understanding. Our initial
review of the methodological literature generated
two broad categories: positivist approaches to case
studies vs alternative traditions. Close textual read-
ing of the case studies in our dataset challenged
this dichotomous view, and at one stage of our ana-
lytical process we worked with six distinct catego-
ries of theorising methods. These were eventually
collapsed into four categories, which then required
us to elucidate the commonalities and differences
between them. The typology we present in this
paper was progressively developed in the course of
our analysis. After developing the first version of
our typology, we conducted a final round of coding
in order to refine the key dimensions of the
typology and attributes of each category.
Content analysis faces the challenge that inter-
pretivists and critical realists term the double
hermeneutic (Giddens, 1984; Noorderhaven, 2004):
all research is an interpretation of an already inter-
preted world. In our study, we were interpreting
published studies without additional information
about the authors’ original motivations or the
modifications they made during the review process.
Moreover, the four methods of theorising are, of
necessity, ideal types: in actual research practice, we
found that authors do not necessarily adhere to a
consistent philosophical position, methodological
approach or even research design. We neither claim
that ours is the only possible classification of these
articles, nor that our categorisation necessarily
matched authorial intentionality.
Having discussed the qualitative, interpretive
nature of our analysis, we now present the insights
we gained into the four methods of theorising. We
turn first to the support for our typology that we
Theorising from case studies Catherine Welch et al
744
Journal of International Business Studies

found in the methodological literature (i.e., the
theoretical foundations of our typology), followed
by an analysis of how the four methods were used
in research practice (i.e., the empirical foundations
of our typology).
CONSTRUCTING THE TYPOLOGY:
THEORETICAL FOUNDATIONS
In this section we turn to the typology generated
from our content analysis, and trace the foun-
dation for each method of theorising to the extant
methodological literature on case studies, and to
distinctive philosophical traditions. Three of the
methods – inductive theory-building, natural
experiment and interpretive sensemaking – are well
established, while the fourth – which we label
contextualised explanation – is a more recent
addition to the methodological literature. We
provide an overview of each method and its
underlying philosophical orientation (see Table 1
for a summary), paying particular attention to
how explanation and context are framed. Two
dimensions of the case study, namely contextuali-
sation and causal explanation, form the basis
of our typology, which we bring together in a
two-by-two matrix.
Case Study as Inductive Theory-building
Proponents of this method identify the main
potential of the case study as lying in its capa-
city to induce new theory from empirical data.
Eisenhardt (1989), the methodological authority
most closely associated with this position (see also
Bonoma, 1985; Leonard-Barton, 1990), explicitly
grounds her defence of the case study’s inductive
strengths in a “positivist view” of science, whose
aim is “the development of testable hypotheses and
theory which are generalizable across settings”
(Eisenhardt, 1989: 546; see also Eisenhardt &
Table 1 Comparing four methods of theorising from case studies
Dimension Inductive theory building Natural experiment Interpretive
sensemaking
Contextualised explanation
Philosophical
orientation
Positivist (empiricist) Positivist
(falsificationist)
Interpretive/
constructionist
Critical realist
Nature of
research
process
Objective search for
generalities
Objective search for
causes
Subjective search
for meaning
Subjective search for causes
Case study
outcome
Explanation in the form of
testable propositions
Explanation in the
form of cause–effect
linkages
Understanding of
actors’ subjective
experiences
Explanation in the form of causal
mechanisms
Strength of
case study
Induction Internal validity Thick description Causes-of-effects explanations
Attitude to
generalisation
Generalisation to population Generalisation to
theory (analytic
generalisation)
“Particularisation”
not generalisation
Contingent and limited generalisations
Nature of
causality
Regularity model: proposing
associations between events
(weak form of causality)
Specifying cause–
effect relationships
(strong form of
causality)
Too simplistic and
deterministic a
concept
Specifying causal mechanisms and the
contextual conditions under which they
work (strong form of causality)
Role of
context
Contextual description a first
step only
Causal relationships
are isolated from the
context of the case
Contextual
description
necessary for
understanding
Context integrated into explanation
Main
advocate
Eisenhardt Yin Stake Ragin/Bhaskar
Theorising from case studies Catherine Welch et al
745
Journal of International Business Studies

Graebner, 2007: 28).
5
The dilemma is that the
small-Ncase study would seem to be incompatible
with this objective, which aspires to uncover regu-
larities or laws of behaviour by emulating the
methods of the natural sciences. Eisenhardt (1989)
does not question this “law-explanation ortho-
doxy” (Outhwaite, 1987: 7) – namely, that to explain
an event or phenomenon means to subsume it
under a general law – rather, she creates a legitimate
role for the case study by casting it as the “natural
complement” to deductive theory-testing. While
case studies cannot provide nomothetic, law-like
generalisations, they can generate the theoretical
propositions upon which large-scale quantitative
testing is based.
Eisenhardt’s model of the theorising process is
strongly positivist in terms of its empiricism, in
that she regards observation as the basis for theory
development, and theory induced from data is
likely to be more valid as “it closely mirrors reality”
(Eisenhardt, 1989: 547). She assumes that this pro-
cess of observation can be objective, with the res-
earcher achieving validity and reliability through
the minimisation of bias. Eisenhardt distances
her inductive theory-building approach from other
qualitative traditions that avoid generalisability
and universal claims in favour of “rich, complex
description” (Eisenhardt, 1989: 546). She regards
“contextual description” as “a necessary first step”
in case research, but on its own it does not lead to
generalisable theory (Eisenhardt, 1991: 626). Rather,
case researchers need to escape the “idiosyncratic
detail” of individual cases and conclude with “only
the relationships that are replicated across most or
all of the cases” (Eisenhardt & Graebner, 2007: 30).
Thus there is a shift from context-bound detail to
context-free propositions.
Eisenhardt’s acceptance that explanation takes
the form of law-like generalisations affects her view
of causality, as well as of context. While she notes
that qualitative data can provide insight into “why
or why not” particular relationships occur, this is
not the primary focus of enquiry, which is rather to
identify generalisable patterns for further testing.
She avoids the use of terms such as “causal” or
“causation”, instead simply referring to “relation-
ships” between variables and constructs. This evokes
the regularity model of scientific explanation com-
monly traced back to the philosopher David Hume:
namely, that the goal of scientific explanation is
to uncover “constant conjunction” or covariation
between variables (Brady, 2008). Positivist philoso-
phers of science have frequently been uncomfortable
with the notion of causality, given that it is
ultimately unobservable and therefore non-
empirical: thus Eisenhardt’s avoidance of the con-
cept of causality and preference for covariational
terms is in keeping with this tradition (see Abbott,
1998, on the same trend in sociology). The main
aim of inductive theory-building research is to
propose associations between constructs and vari-
ables that can then be tested. This can be seen as a
weak form of causality, in that it seeks to establish
regularities rather than the reasons behind them.
Case Study as Natural Experiment
Yin (2009) agrees that case studies are well suited to
exploratory theory-building, but unlike Eisenhardt
he does not confine case studies to this early stage
in the theorising process. Much of his influen-
tial book on case studies ( Yin, 2009) is devoted to
an account of how case studies can be used for
“explanatory” rather than “exploratory” purposes.
In fact, he regards case studies as best suited to
“how and why” questions that “deal with opera-
tional links needing to be traced over time” (Yin,
2009: 9). “Explanatory” case studies use deductive
logic to test propositions, adjudicate among rival
explanations, revise existing theories and establish
causal relationships; in other words, they are suited
to verification and not just discovery of new theory
(see also Eckstein, 2000). Flyvbjerg (2006: 227) goes
so far as to claim that case studies are “ideal” for
falsification, which Karl Popper regarded as central
to theory development.
Although Yin (2009) is not explicit about his
philosophical assumptions, he does not question
the goals of generalisability, validity and reliability.
However, despite sharing these core philosophical
commitments with Eisenhardt, he nevertheless
regards the possible contribution of the case study
very differently (Table 1). In his view, the explana-
tory logic of the case study shares many features
with the laboratory experiment. As a “natural
experiment” (Lee, 1989), the strength of the case
study lies in its high degree of internal validity, so
long as appropriate procedures are followed in its
design and implementation. Many of the proce-
dures that Yin (2009) advocates – such as replica-
tion logic, pattern matching and time-series
analysis – are adaptations of experimental techni-
ques. Similarly, his reply to concerns about the
case study’s generalisability is to argue that, like
the experiment, the case study generalises to
theoretical propositions and not to populations
(Yin, 2009). Lack of statistical generalisability does
Theorising from case studies Catherine Welch et al
746
Journal of International Business Studies

not preclude case studies from having a strong
explanatory contribution to offer.
Yin (2009: 143) concedes that the process for
developing causal explanations with a case study
“has not been well documented in operational
terms”. He also notes that “causal links may be
complex” (2009: 141) and involve multiple inde-
pendent and even dependent variables. Yet while
case evidence is holistic and complex, attention
to the research design and proper application of
analytical techniques enable the researcher to
converge on a set of causal relationships, isolating
them from the broader context of the case. Other
cases can then be investigated to establish whether
the causal patterns occur as predicted, just as multi-
ple experiments are used to refine and test theory.
Given this experimental logic, Yin is comfortable
with the use of explicit causal language (see also
Hillebrand, Kok, & Biemans, 2001).
Case Study as Interpretive Sensemaking
The notion that case studies are a form of
interpretive sensemaking is part of a rich tradition
of “idiographic” rather than nomothetic social
science; in other words, a social science that seeks
to understand the particular rather than generate
law-like explanations. Unlike positivist epistemol-
ogy, which insists on the unity of the social and
natural sciences, interpretive approaches
6
empha-
sise the uniqueness of the social sciences, in which
subjects ascribe meaning to their own behaviour,
and researchers are part of the world they study (see
Table 1). Given that human activity can be under-
stood only by accessing how it is intended and
experienced, case researchers in this tradition –
echoing a controversy that dates back to the 19th
century – argue that in the social sciences, the
scientific ideal of erkla¨ren (explaining an action by
attributing it to exogenous causal factors) needs to
be replaced by verstehen (understanding an action
through the actor’s subjective experience of it)
(Johnson & Duberley, 2000). Stake (1995: 38), a
prominent advocate of interpretive sensemaking
(see also Feagin, Orum, & Sjoberg, 1991), insists on
“the difference between case studies seeking to
identify cause and effect relationships and those
seeking understanding of human experience” (see
also Prasad, 2005, for a similar view in manage-
ment). Case studies are well suited to the latter, as
Lincoln and Guba (1985) argue, because they
enable the rich contextual description essential to
understanding.
Given this philosophical commitment, Stake
(1995) proceeds to question the ideals upheld in
positivist case traditions, including generalisability,
causality and objectivity. In his view, “particulari-
zation” is the goal of case studies: that is, an under-
standing of the uniqueness of the case in its entirety.
In contrast to researchers aiming at generalisable
explanations, who seek “to nullify context” and “to
eliminate the merely situational”, researchers in the
interpretive tradition embrace context, narra-
tives and personal engagement on the part of the
researcher (Stake, 1995: 39, 40). Establishing cause–
effect relationships is regarded as “simplistic” in the
face of this complexity (Stake, 2005: 449); instead,
the aim is “thick description” – in other words, an
appreciation of how the social context imbues
human action with meaning (Table 1). Stake also
disputes the notion that objectivity on the part of
the researcher is possible, and argues that, when
adjudicating among competing interpretations,
“there is no way to establish, beyond contention,
the best view” (Stake, 1995: 108).
Case Study as Contextualised Explanation: An
Emerging Alternative
In this section, we introduce contextualised
explanation to the IB field (see Table 1). Given this
method is a more recent development, it exhibits
less consistency and uniformity than can be found
in the other three methods we have profiled.
Critical realism forms the ontological basis for this
method, but social scientists are still debating how
to apply this philosophy in practice. Meanwhile,
researchers have pioneered analytical procedures
for generating contextualised explanations, such as
process tracing and qualitative comparative analy-
sis. However, these scholars do not necessarily show
an explicit or consistent philosophical commit-
ment. Accordingly, we discuss the philosophical
and methodological innovations associated with
contextualised explanation separately.
How to explain in context: philosophical insights from
critical realism. The philosophical foundation for
contextualised explanation is distinct from the
other methods of theorising, as it lies in critical
realism (see Table 1). There are multiple variants of
critical realism, so the focus in this paper will be on
the most influential: Roy Bhaskar (e.g., 1998), and
those who have introduced his philosophies to
practising social scientists (for applications to
management see, e.g., Ackroyd & Fleetwood, 2000;
Reed, 2005; Tsoukas, 1989). Bhaskar is realist in the
Theorising from case studies Catherine Welch et al
747
Journal of International Business Studies

sense that he acknowledges the existence of a
reality that is independent of our perceptions of
it, but he also regards our comprehension of reality
as theory-laden and subjective, and social pheno-
mena as concept-dependent (in other words, con-
stituted by the meanings we attach to them).
Bhaskar regards explanation of social phenomena
as being “both causal (as does the positivist) and
interpretive (as does the hermeneuticist)” (Collier,
1994: 167). In other words, Bhaskar provides a
way to reconcile explanation (erkla¨ren) and under-
standing (verstehen).
Bhaskar rejects the empiricist assumption that
sensory observation is the only basis for explana-
tion, instead arguing that causality can be under-
stood only with reference to “transcendental”,
or unobservable, causal mechanisms. In Bhaskar’s
philosophy, the concept of “causal mechanism”
refers to the causal powers (or liabilities) of objects,
structures and entities. Objects (whether physical,
human or social) have causal powers by virtue of
their intrinsic nature: an object and its causal
powers are necessarily or internally related (Sayer,
1992). However, in an open system such as that of
the social world, the relationship of causal mechan-
isms to their effects is contingent and external,
rather than necessarily and internally related. That
is to say, whether a causal mechanism is activated
depends on the conditions in which it operates:
mechanisms are tendencies that may not be
actualised, and even if actualised, may not be
empirically observable. Only in a closed system,
which is carefully manufactured in an experimental
situation, can a causal mechanism potentially be
isolated from other generative processes, and
regular effects produced and observed. In open
systems, in contrast, there can be no symmetry
between explanation and prediction: “The same
causal power ycan produce different outcomes y
[or] different causal mechanisms can produce the
same result” (Sayer, 2000: 15). This means that
explanation needs to account for the spatio-tem-
poral context in which causal mechanisms operate.
As a result, causation is not about the search for
event regularities: social scientists need to go
beyond events to understand the nature of objects,
and cause–effect relationships do not consistently
produce regularities in an open system. Causal
explanation lies rather in understanding the con-
stituent nature of objects: in other words, what
objects are capable of doing. Causal explanations
are developed not by collecting observations, but
rather by digging beyond the realm of the observable
to understand the necessity inherent in objects
(Collier, 1994). The appeal to empirical obser-
vation – either through inductive theory-building
or through deductive theory-testing – does not
satisfy a critical realist.
Bhaskar’s critical realism rejects the determinism
and reductionism that are inherent in the regularity
model. He ascribes causal power to human agency:
that is, an actor’s reasons for acting can play a role
in causing that action (Collier, 2005; Outhwaite,
1987). Yet, at the same time, explanations cannot
be reduced solely to human intentionality and
agency, because human actors operate within
already existing social structures. Social structures
condition our actions, yet through our actions we
(re-)produce these very social conditions. Explana-
tory accounts therefore need to encompass human
intentionality – the articulated reasons of social
actors – as well as an actor’s position in the social
structure. Therefore, while human action is inher-
ently meaningful and purposeful, a causal explana-
tion cannot be built solely from actors’ own
understandings and interpretations.
In addition, critical realism challenges the possi-
bility of a purely inductive or deductive process of
theory development. Lawson (2003) proposes that
an explanation often starts with a surprising con-
trast, triggered by the realisation that an observed
outcome is different from what had been antici-
pated (provoking the question, “why not X?”). This
suggests that a new causal factor is in operation, or
the observation domain was not as well understood
as initially thought, or existing understandings of
causal mechanisms need to be refined. This is
essentially an abductive process: the starting point
is a perceived mismatch between an empirical
observation and an existing theory, leading to a
“redescription” or “recontextualisation” of the
phenomenon (Danermark, Ekstro
¨m, Jakobsen, &
Karlsson, 2002). In this view, theorising is a process
not of discovery but of conceptualisation (Sayer,
1992).
While critical realism offers a distinctive ontology
and epistemology, it does not align itself to a
specific research methodology. However, Sayer
(1992: 243) argues that enquiries into causes (as
opposed to regularities) – typified by questions such
as “What produces a certain change?” – require an
“intensive” research strategy, typically involving a
qualitative, in-depth study of “individual agents
in their causal contexts”. Accordingly, case studies
are well suited to developing causal explanations
and “exposing” generative mechanisms (Danermark
Theorising from case studies Catherine Welch et al
748
Journal of International Business Studies

et al., 2002), while conversely the “explanatory
penetration” of “extensive” large-Nstudies is likely
to be weak. Yet the application of critical realism to
case studies remains “underdeveloped” (Elger, 2010:
256). In the meantime, recent years have seen
methodological innovations in case research that
question positivist forms of explanation and can be
seen as consistent with a critical realist approach
(Ragin, 2009).
How to explain in context: Methodological insights.
Consistent with critical realism as a philosophy,
methodological approaches to contextualised
explanation are concerned with accounting for
why and how events are produced. Understanding
how the outcome in a particular case was brought
about (e.g., “A led to E through steps B, C, D”)
entails working backwards from events (causes-of-
effects explanations) rather than estimating the net
effects of causes (effects-of-causes explanations)
(Mahoney & Goertz, 2006). The technique of
working backwards – of identifying the inter-
vening causal process between two “variables” –
has been termed process tracing (George &
Bennett, 2004; Gerring, 2007b; Hall, 2006). It
involves a careful construction of a causal chain
of evidence from observations that (unlike much
data used in the social sciences) are non-
comparable, because they are not from a uniform
population (Gerring, 2007a). Such an approach to
causality has been defended as providing stronger
explanatory power than the “weak” correlational
form (“if X changed by a certain amount, then Y
will have changed by a related amount”) offered by
the regularity model (Elliott, 2005; George &
Bennett, 2004; Roberts, 1996).
Reconstructing causal chains of events suggests
a historical approach, yet history is open to the
charge that it only provides an explanation of the
particular. Proponents of contextualised explana-
tion question this neat separation between the
particular/historical and general/theoretical: “Case
studies typically partake of both worlds” (Gerring,
2007a: 76). As George and Bennett (2004) contend,
generalities are routinely used – and refined – to
make sense of the particular (see also Hall, 2006).
Researchers make sense of particular events by
classifying them as belonging to a class or broader
phenomenon, and by making reference to existing
theories, generalities and known patterns in order
to “connect the discrete steps in an explanatory
narrative” (Roberts, 1996: 54). In the process of
iterating between the particular and the general,
theories can be refined and reassessed, or even rival
explanations proposed. Equally, just as researchers
require an understanding of the general to make
sense of the particular, so too is the latter essential
to explanatory accounts. History, then, is not oppo-
sed to general theory; rather, “theory cannot escape
history” (Calhoun, 1998: 860), in that explanations
of actions require them to be situated in “social
time” and “social place” (Abbott, 1998).
So far, our discussion has assumed that there is a
single causal chain or pathway to be investigated.
However, case-oriented researchers question the
assumption of causal homogeneity made by posi-
tivist research traditions; in other words, “the idea
that causal factors operate in the same way for
all cases” (Ragin, 2000: 51). Instead of regarding
causation as uniform, Ragin (2000; Rihoux & Ragin,
2009) proposes a “multiple conjunctural” view as the
foundation for case-based research. By “conjunc-
tural”, he means that case researchers explain by
factoring in the combination of conditions found
in the case rather than seeking to measure the net
effect of an isolated variable. This is because a single
variable may have a very different effect, depending
on the configuration of variables with which it is
combined in a case. Thus, for example, in combina-
tion with A and C, B may cause Y, but in other
circumstances Y may occur only in B’s absence
(expressible in Boolean algebra as Y¼(AANDBANDC)
OR ((NOT B) AND D AND E). Understanding the
effect of B therefore requires putting it in its spatial-
temporal context. Because B may produce one
effect in a particular context, but a different effect
in another situation, “it is not useful to generalize
about the overall effect of B without saying some-
thing about the context (i.e., other variable values)
in which B appears” (Mahoney & Goertz, 2006:
235). Ragin (e.g., Rihoux & Ragin, 2009) has pro-
posed a formal method based on set theory, quali-
tative comparative analysis, in order to analyse
cases holistically as combinations of conditions.
Ragin (2000) positions this “configurational view”
as having a fundamentally different explanatory
logic from that of the positivist approach, which
assumes away causal heterogeneity.
As well as being “conjunctural” in nature,
causality is “multiple”, given that the same out-
come may be produced by different causal path-
ways (also known as equifinality) (Rihoux & Ragin,
2009). Again, Ragin argues that causality should be
conceived in set-theoretic rather than probabilistic
terms. In Boolean algebra, multiple causation can
be expressed in terms such as Y ¼(A AND B) OR
Theorising from case studies Catherine Welch et al
749
Journal of International Business Studies

(C AND D) (Mahoney & Goertz, 2006). Given the
heterogeneous nature of causation, case researchers
have concluded that generalisations are not uni-
versalities; they are always necessarily limited.
Generalisation therefore involves “careful setting
of scope” (Byrne, 2009: 9): causal explanations
require an understanding of the conditions under
which they do – and do not – operate. Researchers
can aim for no more than “contingent general-
izations”; in other words, propositions such as
“if circumstances A, then outcome O” (Gerring,
2007a). In this way, what Eisenhardt regards as
“idiosyncratic detail”, to be removed from the
explanation, now becomes part of the causal fabric
of an explanatory account.
The Four Methods Compared
Figure 1 is a visual representation of our typology of
case study theorising. The four methods of theoris-
ing are positioned in relation to each other on
the basis of whether their emphasis is on causal
explanation ( y-axis) and/or contextualisation (x-axis).
The figure encapsulates our argument as to why
contextualised explanation offers potential. In the
inductive theory-building approach (Quadrant 1),
the emphasis on both causal explanation and
contextualisation is weak. This avoidance of causal
claims and context can be traced back to the
pursuit of nomothetic, law-like generalisations,
which privileges the search for regularities rather
than causes, and for context-free rather than
context-sensitive knowledge. The established alter-
natives to inductive theorising redress these limi-
tations, but only by accepting the traditional
trade-off between causality and contextualisation.
The method of the natural experiment (Quadrant
2) is a welcome development, in that it provides a
defence of the case study’s ability to generate
causal, internally valid explanations. However, as
we have seen, the emphasis on contextualisation is
still weak. In the sensemaking tradition (Quadrant
3), the “rich story” that troubles Eisenhardt and
Graebner (2007) is transformed into the case study’s
main strength, but at the cost of any claims to
causal explanation. All three methods reinforce
rather than question longstanding divisions in
the social sciences: between erkla¨ren and verstehen,
explanation and understanding, nomothetic and
idiographic, objective and subjective, inductive
and deductive, general and particular, context-free
and context-specific.
The fourth method of theorising, contextualised
explanation (Quadrant 4), represents an escape
from the explanation–contextualisation trade-off.
This method of theorising is based on the assertion
that case studies can generate causal explanations
that preserve rather than eradicate contextual rich-
ness. Proponents insist that explanatory accounts
are necessarily context-bound: as Sayer (1992: 60)
has written, “making sense of events requires that
we ‘contextualize’ them in some way”. While posi-
tivist traditions abstract away from time and place,
contextualised explanation is a way of explaining
“without laws” (see, e.g., Abbott, 1997). The regu-
larity model of causation is rejected in favour of a
more complex understanding that recognises the
contingent nature of cause–effect relationships.
In this section, we have confined our attention to
how the four methods of theorising have been
developed in the methodological literature on case
studies, as well as their philosophical foundations.
The question still remains, however, as to how the
four methods in our typology are used in research
practice. Accordingly, we now shift our focus
from the ideal types discussed by methodologists
and philosophers (how researchers should use case
studies to theorise) to theorising practices (how
researchers do use case studies to theorise). We will
argue that grounding our typology in research
practice allows for a more nuanced understanding
of each method and its potential in future IB
research.
CONSTRUCTING THE TYPOLOGY: EMPIRICAL
FOUNDATIONS
In this section, we will discuss the insights from our
case study dataset into each method of theorising,
reporting on findings about each category in our
typology (see Figure 1). Inductive theory-building
was clearly the most popular of the four methods of
theorising, as Table 2 shows. JMS was the exception
among the journals, in having both the highest
number of case studies and the highest percentage
4. Contextualised explanation
2. Natural experiment 1. Inductive theory-building
3. Interpretive sensemaking
Emphasis on causal explanation
Emphasis on contextualisation
Weak
Weak
Strong
Strong
Figure 1 Four methods of theorising from case studies.
Theorising from case studies Catherine Welch et al
750
Journal of International Business Studies

of case studies that used alternatives to theory-
building. For each method of theorising we high-
light the key elements of the theorising process that
have been central to our discussion throughout the
paper: philosophical assumptions (while authors’
philosophical orientation was typically not made
explicit, it was reflected in how they represented
the theorising process), causality, context and
generalisability.
Quadrant 1: Inductive Theory-building (N¼96)
The case studies that we classified as falling into
this quadrant were all positivist in their assump-
tions, and identified their theoretical contribution
as being exploratory. Despite these similarities,
this category was the most diverse of the four. In
particular, articles differed in the extent of theory
development that they reported had occurred prior
to entering the field. At one extreme, we identified
articles characterised by a grounded theory app-
roach, in which the introduction would often be
followed by the methods section instead of the
literature review to underline the inductive nature
of theorising (e.g., Pratt, Rockmann, & Kaufmann,
2006). While most papers in this category were
written up deductively, authors would demonstrate
the inductive nature of their work by presenting a
model or a set of propositions as an outcome rather
than as a starting point of the research (e.g., Harvey,
Pettigrew, & Ferlie, 2002; Maitlis & Lawrence,
2003). At the other extreme were those, such as
Danis and Parkhe (2002), who while positioning
their contribution as theory-building, nonetheless
included a priori propositions. In between were
authors such as Gilbert (2005), who refined existing
theory. While he adhered to the Eisenhardt app-
roach in a faithful way, it was more common for
authors to “cherrypick”: that is, apply some guide-
lines from Eisenhardt, but not use her model as a
template. While Yin (2009) was also heavily cited
by Gilbert (2005) and others in this quadrant, the
references were to his exploratory (rather than
explanatory) case study.
We would argue that most articles in this
quadrant used the term “inductive” very loosely.
Moreover, only a few articles made the process of
“inductive” theory-building explicit and transparent.
We found an exception in Tolich, Kenney, and
Biggart (1999: 594), who “from their interviews y
were able to draw out inductively what [they]
believe were four variables that had a significant
effect [on X] y” Similarly, Denis, Langley, and
Pineault (2000) distinguished between the “skeletal
conceptual framework” with which they com-
menced the study, and the insights they gained
from their fieldwork. Otherwise, it seemed that the
majority of the authors were not explicit in
specifying which of their insights were inductively
derived from their fieldwork and which were more
theory-driven.
While inductive theory-building articles typically
did not explicitly aim to study causal chains or
relationships, they nevertheless used causal voca-
bulary, as can be judged from expressions such as
“influenced”, “interacts with”, “critical determinants
of”, “centrally facilitative in”, “leads to”, “trigger”
(e.g., Co
ˆte
´, Langley, & Pasquero, 1999; Faems,
Janssens, Madhok, & van Looy, 2008; Harvey,
Pettigrew, & Ferlie, 2002; Tolich et al., 1999). This
causal vocabulary typically conformed to that of
the regularity model, with authors using terms such
as “associated with” and “moderating influence”
(Wilkinson, Gamble, Humphrey, Morris, & Anthony,
2001; Wong & Ellis, 2002), while propositions were
framed in correlational terms such as “the higher
ythe greater” (Bu
¨chel, 2000).
The articles in this category were united in their
descriptive treatment of the research context as a
first step before analysing data. Thus these authors
tended to have a separate section outlining the
research setting, but the discussion of it remained
at a descriptive rather than analytical level (e.g.,
Boxall & Steeneveld, 1999). Context is a feature of
these studies, but not as a means of providing
explanation. Despite this dominant trend, we did
identify authors who showed greater sensitivity
towards context. They tended to pursue process
research (e.g., Co
ˆte
´et al., 1999) or grounded
theorising (e.g., Ravasi & Schultz, 2006; Rodrigues
& Child, 2008). However, in these studies the rich
process data did not translate into process theoris-
ing: for example, Maritan (2001) concludes with
propositions that are correlational rather than
processual in nature.
A common concluding point for articles in this
quadrant was the standard warning about the lack
Table 2 Number of case studies in AMJ,JIBS and JMS, 1999–2008
Theorising typology AMJ JIBS JMS Total
Inductive theory-building 23 12 61 96
Natural experiment 1 5 21 27
Interpretive sensemaking 8 0 44 52
Contextualised explanation 5 2 17 24
Total 37 19 143 199
Theorising from case studies Catherine Welch et al
751
Journal of International Business Studies

of generalisability due to the small-Nnature of the
study and the context-specific nature of the find-
ings. Like Eisenhardt, many authors in this quad-
rant portray their contribution as being a “bridge”
to theory-testing, and end with a series of proposi-
tions (e.g., Coviello, 2006; Orr & Scott, 2008). In
such articles, the setting of the study can be por-
trayed as a limitation: thus Noda and Collis (2001)
describe the “context yof the study” as potentially
introducing bias (see also Beverland, 2005). Some
case researchers were less apologetic, and merely
made the transferability of their findings a matter
for discussion (e.g., Boxall & Steeneveld, 1999).
However, few case researchers in this quadrant
specified the contexts to which the findings could
be transferred, or why. An exception can be found
in Pratt et al. (2006: 259), who, while presenting the
“unique nature” of the setting as a weakness in
their quest for generalisability, nevertheless argue
that “careful” transferability is possible, and specify
the settings that offer “easy-to-see parallels” to their
own study.
Quadrant 2: Natural Experiment (N¼27)
Authors in this quadrant subscribed to positivist
assumptions, but they positioned their theoretical
contribution as being theory-testing, applying an
established theory or providing rival explanations
(although they typically did not explicitly use the
metaphor of a natural experiment). Buck and
Shahrim (2005) test a causal proposition by using
a “least likely” case (as recommended by George &
Bennett, 2004), which they selected “in order to
maximise the possibility of refutation” (Buck &
Shahrim, 2005: 58). At the same time, Marko
´czy
(2000) is a clear example of articles that offer a rival
explanation to that favoured by existing theory.
She concludes with a very bold statement – namely,
that cultural differences are overrated as an expla-
natory factor – and offers what she terms “novel
factors” in accounting for beliefs. A “rival explana-
tion” study can also take the form of a re-evaluation
of a previously reported case that questions how it
has been interpreted in existing literature (e.g.,
Howells, 2002). These different forms of challen-
ging existing theory were not widely used in our
dataset. Nor were articles that sought to apply
existing theory in order to provide a causal account,
Collinson and Rugman (2008) being an exception.
Accordingly, we would argue that the ability of the
case study to modify, verify, test and challenge
existing theory and offer rival explanations has
been underexploited.
In this quadrant we detected the frequent use of
causal language, which was often more explicit
and pronounced compared with inductive theory-
building articles. For example, Taylor (1999: 858)
states as his objective “to determine what affects
the degree of control yand to what extent control
is related to [X] y”. The correlational associations
between variables are the dominant approach to
causal relationships in this paper, but the author
also concedes that “there is a complex pattern” at
play, “one in which a combination of factors y
affect the outcome” (Taylor, 1999: 866). While
Taylor’s (1999) variable-oriented approach to caus-
ality was typical of this category, the authors of one
paper developed a more nuanced argument regard-
ing the level of “path-effects”, concluding that
while “historical patterns of development” were
relevant, they “do not fully explain the present”
(Hill, Martin, & Harris, 2000).
Authors whose studies were classified into this
category dealt with context in different ways. In
some articles, the empirical context was seen to
offer the advantage of a “natural” laboratory setting,
as suggested by our label for this quadrant. In light
of this aim, researchers took great care to select the
naturally occurring conditions that were the most
appropriate test of the theory, given that, unlike in
an artificial laboratory setting, the environment
cannot be controlled by the researcher. For exam-
ple, De Boer, Van den Bosch, and Volberda (1999:
389) chose publishing firms moving into multi-
media because they are “right at the heart of the
evolutionary process in which technologies stem-
ming from various industries are converging into
hybrid forms”. While we found almost decontex-
tualised case studies in this category (e.g., Stiles,
2001), we identified others which drew on context-
ual factors to generate explanation. For example,
Ogbonna and Harris (2002) succeed in enriching a
case study of two change initiatives by offering a
context-sensitive account as to why the cross-case
differences had occurred. In their study, context
enhances the internal validity of the study, since it
controls for environmental effects and promotes a
replication logic. Yet, even in these articles, context
was decomposed into a set of variables that had an
effect on the phenomenon under study, rather than
capturing the influence of context more holistically.
Moreover, context tended to be de-emphasised or
even isolated from the findings and conclusions,
given that authors made generalising claims and
provided context-free models as their contribution
(e.g., Johnson-Cramer, Cross, & Yan, 2003).
Theorising from case studies Catherine Welch et al
752
Journal of International Business Studies

Quadrant 3: Interpretive Sensemaking (N¼52)
Researchers in this quadrant, particularly those
adhering to a social constructivist approach
(Hodgson, 2002; McCabe, 2000; Watson & Watson,
1999), were typically very explicit about their
philosophical stance and their rejection of positi-
vist assumptions. A distinct approach to reflecting
on and representing the theorising process could
often be found in these papers. While some authors
used descriptors such as “exploratory” and “theory-
building”, in many other articles, a rigid distinction
between theory and evidence was not upheld;
instead these two elements were interwoven
throughout the paper, in keeping with the belief
that theory and observation cannot be separated
(Chreim, 2005; McInerney, 2008). The authors of
one paper explicitly described the theorising pro-
cess as “a mixture of both deductive and inductive
methods” (Noon, Jenkins, & Lucio, 2000: 504). More
commonly, authors did not use either “induction”
or “deduction” to denote their theorising. How-
ever, there were articles that claimed to be inter-
pretive but nevertheless revealed traces of the
positivist theory-building tradition, for example
by acknowledging the shortcomings of a single
case study in developing generalisable theory (e.g.,
Coupland & Brown, 2004).
Authors presented their theoretical objectives in
terms of illuminating and providing insight, for
example: “we are interested in the worldviews of
organizational members” (Maguire & Phillips,
2008: 380). Portraying worldviews is not just a
descriptive effort; Ram (1999) provides a rich
narrative which is infused with theoretical con-
cepts. These authors would often start their article
with a vignette or a personal encounter from the
field. Researchers with a more social constructivist
approach sought not only to understand partici-
pants’ meaning, but also how these meanings were
constructed. For example, Yakura (2002) shows
how time is “constructed” in multiple ways in a
consulting firm, while Lindgren and Packendorff
(2006: 841) view project work “as an ongoing
construction of patterns of femininity and mascu-
linity in society”.
We found no explicit reference to causes, unless
authors were referring to the causal models
employed or constructed by research participants.
However, we encountered extensive use of causal
and explanatory language (e.g., Heracleous &
Barrett, 2001), despite the absence (in the main)
of positivist language. Instead, another vocabulary
was in use: managers “enact”, power has “effects”,
hegemony is “produced” and meaning is
“constructed” (Barry, Chandler, & Clark, 2001;
Benjamin & Goclaw, 2005). Salaman and Storey
(2002: 163) conclude that the managers they
studied “are both producers as well as products of
the corporate culture”: in other words, they address
the agency-structure question discussed above in
relation to critical realism. Narratives, which were a
commonly used device, were also much more than
descriptions; rather they had deep, explanatory
purposes, as Ng and de Cock (2002: 40) state
explicitly: “Story interpretation requires an answer
to the deceptively simple question: ‘Why did things
turn out the way it did?’ ” In this way, as Kakkuri-
Knuutila et al. (2008) have observed, “to under-
stand” and “to explain” are not as opposed as they
may seem.
Researchers in this quadrant tended to include
themselves as part of the context, rather than
taking an objective stance. In particular, authors
adopting a social constructivist approach often did
not just seek to examine how research participants
constructed meaning, but also scrutinised their
own sensemaking: “we unashamedly present our
paper as our own construction and are happy to
make explicit the discursive resources which we
bring both to our research design and data analysis”
(Watson & Watson, 1999: 485). Similarly, Dick and
Cassell (2002: 958) add that “the researcher’s own
role in the production of knowledge needs to be
accounted for”. Authors who took this approach
did not seek to claim neutrality: “we rejected the
idea that an objectively verifiable reality can be
accessed through research” (Dick & Cassell, 2002:
960–961), and presented theirs as just one possible
“reading” (Chreim, 2005: 589). Generalisability was
often not discussed or even, in the case of Ng and
de Cock (2002: 43), rejected explicitly: they state
there is no need “to provide law-like theories
with their attractive elegance and highly glossed
accounts”.
Quadrant 4: Contextualised Explanation (N¼24)
Overall, case studies that emphasised causal expla-
nation (Quadrants 2 and 4 in Figure 1) were in the
minority. In our content analysis, we paid attention
to how authors in this quadrant were able to
combine the inherent strength of the case study
to contextualise with its explanatory potential.
Compared with the “natural experiment” quad-
rant, these articles aimed to generate explanation,
but without strong adherence to positivist assump-
tions. While most of the authors did not explicitly
Theorising from case studies Catherine Welch et al
753
Journal of International Business Studies

reveal their philosophical assumptions, we found
evidence of social constructivism/interpretivism
(e.g., Ogbonna & Wilkinson, 2003), and mild
positivism (e.g., Wilkinson et al., 2001), as well as
three papers whose authors were explicitly drawing
on the critical realist tradition (Chung, 2001;
O’Mahoney, 2007; Pajunen, 2006).
Although there was variation in terms of how
authors presented their theorising process, they
tended not to separate theory-building and theory-
testing. For example, Jacobides (2005: 486) – one of
the few articles with the ambitious scope of directly
challenging an established theory – searched for
new analytical insights inductively, “without being
bound by existing theory”. Sminia (2003) aims to
explain why a TV channel failed. This seemingly
atheoretical purpose is countered by a discussion of
how existing explanations are insufficient to shed
light on failed ventures. The author takes a more
deductive approach in combining insights from
existing theoretical perspectives to arrive at a novel,
more convincing explanation. Several papers fall in
between these two extremes, as they recognise both
inductive and deductive elements in theorising.
Denis, Lamothe, and Langley (2001: 812), for
example, write that “[o]ur approach was and is also
partly deductive (theory inspired) and partly induc-
tive (data inspired)”.
In this quadrant, authors were more open about
the explanatory aims of their paper (e.g., Sminia,
2003; White & Liu, 2001). Clark and Soulsby
(1999: 537) set out to “offer a feasible and credible
explanation of the spread of the [multidivisional
organisation] in the Czech Republic”. In a similar
vein, Farjoun (2002: 848) builds “an explanatory
model of institutional development”. The explana-
tory aim is also made clear by West (2008: 1508),
who “sought to explain the commercialization of
Shannon’s theory during its first quarter century”.
In other articles “explanation” was not used, with
authors (Mota & Castro, 2004; O’Mahoney, 2007)
instead referring to “opposing and driving forces”,
“nonlinear relationships” and “multiple pathways” –
all terms suggesting a complex view of causality.
Despite these differences, what typifies the authors’
language is a very particular view of causality as a
complex and dynamic set of interactions that are
treated holistically. For example, Perlow, Okhuysen,
and Repenning (2002) introduce “mutual causality”
and “causal loop diagrams” to capture the connec-
tions between speed and decision-making in an
Internet start-up. Jacobides (2005: 492), in turn,
states that “[t]his is a study of a particular industry
and yit focuses on understanding the causal
dynamics of a particular setting”. White and Liu
(2001) offer “alternative transition trajectories” for
firms operating within the industry under study. As
we have shown, there is a well-established causal
vocabulary in the critical realist tradition, but refer-
ences to this literature are rarely made. O’Mahoney
(2007: 1345) is one of the few to adhere explicitly
to the critical realist tradition in developing an
“explanatory theory”. Another author, Chung (2001),
uses Ragin’s Boolean algebra to systematically
compare cases.
For the authors of these articles context was
a necessity, not a problem, in constructing rich
explanations. As Perlow et al. (2002: 949) write,
“[o]ur findings suggest the importance of examin-
ing decisions and their relationship within the
context in which they happen”. In this category
theorising was viewed primarily as a localised
explanation. Similarly, Clark and Soulsby (1999:
555) weave context into their theoretical interpre-
tation: “the roles of institutional and strategic
choice factors could only be understood in their
mutual interaction”. In their paper, explaining in
context took the form of a real-time processual
analysis that was very different from the process
studies found in the inductive theory-building
quadrant. The context was used to generate an
explanation for the motives of the managers in the
study. They argue that factors typically treated in
other studies as “independent variables” should
rather be understood as a “recursive process” (Clark
& Soulsby, 1999: 556). In this quadrant, history and
process become essential to developing a causal
account. In defending her historical methodology,
Farjoun (2002: 871) argues that “by its nature, his-
torical analysis particularly attends to continuity
and process, to diverse influences and to context”.
Denis et al. (2001: 815) emphasise that the time
periods they identify are not predictable stages, but
allow for more complex explanations such as
“multidirectional causality”.
Based on their in-depth knowledge of the con-
text, authors in this category discuss the scope
of the generalisable claims they are making, and
identify specific contextual aspects that would
expand – rather than reduce – the transferability
of their findings. For example, White and Liu
(2001: 122) conclude that “[d]eregulating or priva-
tizing industries and breaking up monopolies
represent environmental contexts that are concep-
tually similar to China’s transitional economy in
which the central plan has been discarded”. Other
Theorising from case studies Catherine Welch et al
754
Journal of International Business Studies

authors warn against “overgeneralizing”, such as
Jacobides (2005), who emphasises the industry-
specificity of his study. Finally, some authors in this
category did not seek generalisability at all. Instead,
they “sought to embrace all the richness and
complexity of a real ysetting” to generate a
localised explanation and invite the reader to
evaluate the applicability of their results in other
situations (Ogbonna & Wilkinson, 2003: 1159).
Overall, our content analysis revealed consider-
able versatility in theorising practices, and has pointed
to variations within each typological category. We
identified articles with greater context sensitivity,
even in the positivist quadrants that typically have
a weak emphasis on contextualisation. Equally, case
studies using inductive theory-building and inter-
pretive sensemaking methods might make causal
claims, despite not acknowledging this explicitly,
and despite differences in the causal vocabulary in
use. Overall, our findings suggest that case research-
ers lack an established vocabulary to express the
theorising process or its outcome. In the following
section we will consider the wider implications of
these findings.
CONCLUSION AND FUTURE DIRECTIONS
In this paper, we have sought to expand the
possibilities for theorising from case studies in IB
research by constructing a typology of theorising
methods. To enrich the future of qualitative
research in IB, we have employed qualitative
research ourselves – an in-depth qualitative content
analysis of 199 case study articles published in three
journals during 1999–2008. Our typology – which
contrasts inductive theory-building, interpretive
sensemaking, natural experiment and contextua-
lised explanation – was developed by iterating
between the existing methodological literature
and a dataset of published case studies. We would
argue that this typology enriches the potential of
the case study, both for contextualising and for
producing causal explanations. In this concluding
section we draw out these possibilities, and the
implications for future case research in IB. Con-
sistent with the aims of our paper, our focus is on
theorising, even though we recognise there are
other dimensions to the case studies we analysed.
We commenced this paper by arguing that the
current dominance of inductive theory-building in
JIBS (as indeed, in the other journals we analysed)
may be hindering the potential for case studies to
contextualise theory and generate causal explana-
tions. As we have discussed, the call for greater
contextualisation of IB theory has been repeatedly
made by scholars in the field (Brannen & Doz,
2010). However, we have argued that the dominant
method of inductive theorising places little empha-
sis on context: articles in this tradition treated
context descriptively rather than analytically.
In this method, context is seen as a limitation,
given that the goal is law-like explanation. Simply
conducting more case studies – which has been
advocated in the literature on contextualising IB
theory – would therefore not necessarily lead to
more context-sensitive theorising. Rather, scholars
need to consider the implications of their choice of
theorising method carefully, because these methods
differ in their emphasis on contextualisation.
As we have seen, the decontextualised nature of
inductive theory-building studies is not the only
concern. In addition, the widespread assumption
that the role of the case study lies only in the explo-
ratory, theory-building phase of research down-
plays its potential to propose causal mechanisms
and linkages, and test existing theories. The danger
is that these legitimate uses of the case study are
underutilised or even questioned. In IB, Yin (2009)
has been used largely to justify the exploratory role
of case studies, overlooking the strong emphasis he
places on explanatory case studies. In this paper
we have highlighted this neglected dimension of
Yin’s work, and have argued that the case study has
an important role to play in refining, verifying,
testing and challenging existing theory. Our con-
tent analysis revealed examples of case studies that
effectively performed this role and placed a strong
emphasis on causal explanation, although they
were in the minority. This application of the case
study is worth further examination, given its
potential to interrogate existing theories.
Set against this background, we have proposed
that the method of contextualised explanation,
while rarely found in our dataset of published
case studies, holds promise in that it offers a high
degree of contextualisation without sacrificing the
goal of causal explanation. We would argue that,
above all, the value of this approach lies in its
different view of how to generate theories about the
social world: the rejection of the regularity model
of causation, scepticism towards the possibility of
meaningful law-like generalisations, and a defence
of context as being an essential component of,
rather than a hindrance to, explanation. As a result
of redefining the theorising process in this way,
proponents of contextualised explanation seek
to explain “without laws”. They offer a way of
Theorising from case studies Catherine Welch et al
755
Journal of International Business Studies

reconciling context and explanation by acknow-
ledging the complexity of the social world, the
bounded scope and contingency of causal relation-
ships, and the simultaneous operation of multiple
interaction effects. The possibility of such a recon-
ciliation is also an abiding theme in IB, given that
as a field its raison d’e
ˆ
tre is to explain phenomena
in diverse national, cultural and institutional
contexts.
While we suggest that greater application of
contextualised explanation would benefit the IB
field, we are not, however, advocating it as the sole
method for theorising from case studies. We are not
seeking to replace one method (i.e., inductive
theory-building) with another (i.e., contextualised
explanation). Rather, we have adopted a pluralist
stance: that a field benefits from the diversity of,
and even tension between, different approaches.
A broad paradigmatic consensus restricts methodo-
logical innovation and limits the range of methods,
research problems, choice of topics – and ultimately
theories. Yet while the research phenomena in IB
are diverse in nature, we have not detected the
same degree of diversity in theorising about these
phenomena. The pursuit of law-like explanations
remains the taken-for-granted approach to theoris-
ing in IB, leading to decontextualised research
methods and consequently decontextualised the-
ories (Brannen & Doz, 2010). In this paper we have
argued that the goal of more contextualised theo-
ries requires IB scholars to rethink their assump-
tions about the role of the case study. The search for
greater pluralism led us beyond IB to examine key
journals in the general management field, allowing
us to gain insights into theorising from other
research traditions.
Our paper contributes to a more pluralist future
for IB by offering alternative methods of theorising
from case studies. We have demonstrated that case
researchers have a choice about how to theorise,
just as they have a choice about how to produce
and analyse data. Our typology encourages res-
earchers to reflect critically upon their own and
others’ preconceived views of how to theorise from
case studies; to explore possibilities for theorising
that go beyond that of inductive theory-building;
and to escape the trade-off between internal
validity and thick description that is found in
positivist and interpretive paradigms. From the
perspective of those reviewing case research, the
typology provides the means to evaluate theo-
retical contributions. By articulating and more
explicitly specifying their method of theorising,
researchers can foster greater mutual understanding
of the theoretical purpose of their studies.
Yet at the same time as illustrating the flexibility
that the case study methodology offers with respect
to theorising, we have also suggested the need for
coherence when reporting the theorising process.
Each method of theorising entails distinct philoso-
phical assumptions, research objectives and out-
comes, as well as a vocabulary for describing the
theorising process and articulating a study’s theo-
retical contribution. By scrutinising the linguistic
elements of texts, we found that case researchers
were not always clear and consistent in the way
they wrote up their theorising purpose and process.
Methodological rigour has traditionally been
reflected in selecting methods and research designs
that fit the research question of the study. Based on
our content analysis, we would argue that metho-
dological rigour is also evidenced by methodo-
logical self-awareness, transparency and careful
linguistic choices in reporting the theorising pro-
cess.
7
We hope that one contribution of our paper is
to encourage case researchers to (re)consider their
own approach to theorising from case studies, and
be consistent in following their choice throughout
their study.
In a pluralist field of research, greater methodo-
logical consistency and adherence to a particular
method of theorising need not restrict authors from
engaging with, learning from and being influenced
by other methods. Thus, while we have highlighted
the benefits for IB of exploring critical realism, our
content analysis has implications for researchers
operating within positivist and interpretive tradi-
tions as well. For those researchers taking a more
positivist approach, we would suggest that con-
siderable potential lies in exploiting the range of
“natural experiments” that we identified in our
content analysis: testing theory, proposing rival
explanations, reanalysing cases, and applying or
challenging existing theory. Such case studies go
beyond the posing of covariational propositions, to
providing explanations for causal relationships. In
addition to strengthening the emphasis on causal
explanation, there is also potential for researchers
operating within positivist traditions to be sensi-
tised to context when theorising from case studies.
In our content analysis we found that authors
achieved this by detailing how contextual factors
produced the outcome, and how their findings
might be transferred to other settings. These authors
were able to use context to specify the boundary
conditions of their explanations.
Theorising from case studies Catherine Welch et al
756
Journal of International Business Studies

For researchers taking an interpretive approach,
we would stress the advantages to recognising
and making more explicit the explanatory fabric
that permeates their contributions. In our content
analysis we found that, while researchers in this
tradition used a causal language different from that
associated with the other methods of theorising,
interpreting and understanding the social world
also involves offering an explanation as to why
events occur in the way they do. Some authors
did succeed in sensitively combining contextua-
lised explanation with a range of philosophical
traditions, including interpretivism and even a
moderate form of positivism that seeks limited,
contingent generalisations rather than universal-
ities. Therefore we feel there is potential in explor-
ing how insights from contextualised explanation
could inform research more broadly.
In the years since the publication of Eisenhardt’s
(1989) article on theorising from case studies,
greater appreciation has emerged in the methodo-
logical literature as to how authors can explain in
context. At its best, contextualised explanation can
provide novel theoretical accounts that incorporate
rather than deny complexity. While IB scholars
have so far not contributed to the emerging
methodological debate on case studies, we would
hope that the next ten years of JIBS will see a
growing diversity and innovation in approaches
to theorising. In a more pluralist field, case res-
earchers would approach theorising differently.
They would move beyond the conformity to
the inductive theory-building that prevailed in
the JIBS case studies of the previous decade. At the
same time, they would question the trade-off
between internal validity and thick description
that characterises both positivist and interpretive
paradigms. In this more pluralist scenario, case
researchers, regardless of their paradigmatic stance,
would be able to combine context sensitivity
with explanatory rigour in their theorising. By
arguing that contextualisation and rigorous expla-
nation can be complementary rather than contra-
dictory outcomes, we have proposed a future for
the case study in IB that stands in contrast to the
limited role to which it has traditionally been
assigned.
ACKNOWLEDGEMENTS
We thank Snejina Michailova, Marja-Liisa Kakkuri-
Knuuttila, Kalle Pajunen, Kristina Rolin, Ben Tipton,
Denice Welch, Lawrence Welch, Ian Wilkinson and
Yorgos Zotos for their helpful comments on earlier
versions of this manuscript; as well as students in our
PhD case research classes at Aalto University and
University of Sydney. Particular thanks are due to Geoff
Easton and Ricardo Morais, who have been responsible
for our interest in critical realism. Early versions of the
paper benefited from feedback at the 2009 European
International Business Academy annual conference,
and at Academy of Management symposia in 2009
and 2010. We are extremely grateful for the con-
structive and thoughtful guidance provided by Mary
Yoko Brannen as special issue editor, as well as the
anonymous JIBS reviewers.
NOTES
1
Although we recognise that case studies can be
mixed and even quantitative (for a discussion in IB see,
e.g., Nummela & Hurmerinta-Peltoma¨ki, 2006), in
this paper we are concerned with case studies as a
qualitative research strategy.
2
In this paper we follow interpretivists and
critical realists in acknowledging that research is an
act of interpretation. We use terms such as data
“production” and typology “construction”, rather
than seeking to conceal the role of the researcher.
3
We decided to include a journal originating in
Europe because it has been suggested that case study
traditions are more firmly established there than in the
US (Bengtsson, Elg, & Lind, 1997).
4
An illustrative example is the article by Nutt (2000),
who positioned his study as a multiple case investiga-
tion (N¼376) of strategic decisions. This paper was not
included in our analysis, since it treated the cases as
observations, rather than investigating the phenom-
enon in its natural setting.
5
There are many variants of positivism (Halfpenny,
1982), including logical positivism, logical empiricism
and falsificationism. The similarities rather than the
differences among these traditions are our focus in this
paper. However, it is worth noting that we would
characterise Eisenhardt’s empiricism as distinct from
the assumptions behind the “natural experiment”
approach, which rather follows a falsification logic, as
advocated by Karl Popper (for a discussion of empiri-
cism vs falsificationism, see Johnson & Duberley,
2000).
6
In this paper we will use “interpretive” in a broad
sense to refer to research traditions that include
postmodernism, postcolonialism, critical theory and
social constructivism.
7
We thank an anonymous reviewer for this
insight.
Theorising from case studies Catherine Welch et al
757
Journal of International Business Studies

REFERENCES
Abbott, A. 1997. Of time and space: The contemporary
relevance of the Chicago school. Social Forces, 75(4):
1149–1182.
Abbott, A. 1998. The causal devolution. Sociological Methods
Research, 27(2): 148–181.
Ackroyd, S., & Fleetwood, S. (Eds) 2000. Realist perspectives on
management and organisations. London: Routledge.
Bamberger, P. 2008. Beyond contextualization: Using context
theories to narrow the micro-macro gap in management
research. Academy of Management Journal, 51(5): 839–846.
Barbour, R. 2001. Checklists for improving rigour in qualitative
research: A case of the tail wagging the dog? British Medical
Journal, 322(May 5): 1115–1117.
Barry, J., Chandler, J., & Clark, H. 2001. Between the ivory tower
and the academic assembly line. Journal of Management
Studies, 38(1): 87–101.
Bengtsson, L., Elg, U., & Lind, J.-I. 1997. Bridging the
transatlantic publishing gap: How North American reviewers
evaluate European idiographic research. Scandinavian Journal
of Management, 13(4): 473–492.
Benjamin, O., & Goclaw, R. 2005. Narrating the power of non-
standard employment: The case of the Israeli public sector.
Journal of Management Studies, 42(4): 737–759.
Berelson, B. 1971. Content analysis in communication research,
(2nd ed.) Glencoe, IL: Free Press.
Beverland, M. B. 2005. Crafting brand authenticity: The case
of luxury wines. Journal of Management Studies, 45(5):
1003–1029.
Bhaskar, R. 1998. The possibility of naturalism: A philosophical
critique of the contemporary human sciences, (3rd ed.) London:
Routledge.
Blair, J. D., & Hunt, J. G. 1986. Getting inside the head of the
management researcher one more time: Context-free and
context-specific orientations in research. Journal of Manage-
ment, 12(2): 147–166.
Bonoma, T. V. 1985. Case research in marketing: Opportunities,
problems, and a process. Journal of Marketing Research, 22(2):
199–208.
Boxall, P., & Steeneveld, M. 1999. Human resource strategy and
competitive advantage: A longitudinal study of engineering
consultancies. Journal of Management Studies, 36(4): 443–454.
Brady, H. E. 2008. Causation and explanation in social science.
In B. Steffenmeier, H.E. Brady, & D. Collier (Eds), The Oxford
handbook of political methodology: 217–270. Oxford: Oxford
University Press.
Brannen, M. Y., & Doz, Y. 2010. From a distance and detached
to up close and personal: Bridging strategic and cross-
cultural perspectives in international management research
and practice. Scandinavian Journal of Management, 26(3):
236–247.
Bu
¨chel, B. 2000. Framework of joint venture development:
Theory-building through qualitative research. Journal of
Management Studies, 37(5): 637–661.
Buck, T., & Shahrim, A. 2005. The translation of corporate
governance changes across national cultures: The case
of Germany. Journal of International Business Studies, 36(1):
42–61.
Byrne, D. 2009. Case-based methods: Why we need them; what
they are; how to do them. In D. Byrne & C.C. Ragin (Eds), The
Sage handbook of case-based methods: 1–10. London: Sage.
Calhoun, C. 1998. Explanation in historical sociology: Narrative,
general theory, and historically specific theory. American
Journal of Sociology, 104(3): 846–871.
Cappelli, P., & Sherer, P. D. 1991. The missing role of context in
OB: The need for a meso-level approach. In B.M. Staw (Ed.),
Research in organizational behavior, (Vol. 13): 55–110.
Stanford: JAI Press.
Child, J. 2000. Theorizing about organization cross-nationally. In
J.L.C. Cheng & R.B. Peterson (Eds), Advances in international
comparative management, (Vol. 13): 27–75. Stanford: JAI Press.
Chreim, S. 2005. The continuity–change duality in narrative
texts of organizational identity. Journal of Management Studies,
42(3): 567–593.
Chung, C. 2001. Markets, culture and institutions: The emer-
gence of large business groups in Taiwan, 1950s–1970s.
Journal of Management Studies, 38(5): 719–745.
Clark, E., & Soulsby, A. 1999. The adoption of the multi-
divisional form in large Czech enterprises: The role of
economic, institutional and strategic choice factors. Journal
of Management Studies, 36(4): 535–559.
Clark, T., & Wright, M. 2007. Reviewing journal rankings
and revisiting peer reviews: Editorial perspectives. Journal on
Management Studies, 44(4): 612–621.
Collier, A. 1994. Critical realism: An introduction to Roy Bhaskar’s
philosophy. London: Verso.
Collier, A. 2005. Critical realism. In G. Steinmetz (Ed.), The
politics of method in the human sciences: Positivism and its
epistemological others: 327–345. Durham, NC: Duke University
Press.
Collinson, S., & Rugman, A. M. 2008. The regional nature of
Japanese multinational business. Journal of International
Business Studies, 39(2): 215–230.
Co
ˆte
´, L., Langley, A., & Pasquero, J. 1999. Acquisition strategy
and dominant logic in an engineering firm. Journal of
Management Studies, 36(7): 919–952.
Coupland, C., & Brown, A. D. 2004. Constructing organiza-
tional identities on the web: A case study of Royal Dutch/Shell.
Journal of Management Studies, 41(8): 1325–1347.
Coviello, N. E. 2006. The network dynamics of international
new ventures. Journal of International Business Studies, 37(5):
713–731.
Danermark, B., Ekstro¨m, M., Jakobsen, L., & Karlsson, J. C. 2002.
Explaining society: Critical realism in the social sciences. London:
Routledge.
Danis, W. M., & Parkhe, A. 2002. Hungarian–Western partner-
ships: A grounded theoretical model of integration processes
and outcomes. Journal of International Business Studies, 33(3):
423–455.
De Boer, M., Van den Bosch, F. A. J., & Volberda, H. W. 1999.
Managing organizational knowledge integration in the emer-
ging multimedia complex. Journal of Management Studies,
36(3): 379–398.
Denis, J.-L., Lamothe, L., & Langley, A. 2001. The dynamics
of collective leadership and strategic change in pluralistic
organizations. Academy of Management Journal, 44(4):
809–837.
Denis, J.-L., Langley, A., & Pineault, M. 2000. Becoming a leader
in a complex organization. Journal of Management Studies,
37(8): 1063–1099.
Dick, P., & Cassell, C. 2002. Barriers to managing diversity in a
UK constabulary: The role of discourse. Journal of Management
Studies, 39(7): 953–976.
Duriau, V. J., Reger, R., & Pfarrer, M. D. 2007. A content analysis
of the content analysis literature in organization studies:
Research themes, data sources, and methodological refine-
ments. Organizational Research Methods, 10(5): 5–34.
Eckstein, H. 2000. Case study and theory in political science. In
R. Gomm, M. Hammersley, & P. Foster (Eds), Case study
method: 119–164. London: Sage.
Eisenhardt, K. M. 1989. Building theories from case study
research. Academy of Management Review, 14(4): 532–550.
Eisenhardt, K. M. 1991. Better stories and better constructs: The
case for rigor and comparative logic. Academy of Management
Review, 16(3): 620–627.
Eisenhardt, K. M., & Graebner, M. E. 2007. Theory building
from cases: Opportunities and challenges. Academy of
Management Journal, 50(1): 25–32.
Elger, T. 2010. Critical realism. In A.J. Mills, G. Durepos, &
E. Wiebe (Eds), Encyclopedia of case study research, (Vol. 1):
253–557. Los Angeles, CA: Sage.
Theorising from case studies Catherine Welch et al
758
Journal of International Business Studies

Elliott, J. 2005. Using narratives in social research. London: Sage.
Faems, D., Janssens, M., Madhok, A., & van Looy, B. 2008.
Toward an integrative perspective on alliance governance:
Connecting contract design, trust dynamics, and con-
tract application. Academy of Management Journal, 51(6):
1053–1078.
Farjoun, M. 2002. The dialectics of institutional development in
emerging and turbulent fields: The history of pricing conven-
tions in the on-line database industry. Academy of Manage-
ment Journal, 45(5): 848–874.
Feagin, J. R., Orum, A. M., & Sjoberg, G. 1991. A case for the
case study. Chapel Hill, NC: University of North Carolina Press.
Flyvbjerg, B. 2006. Five misunderstandings about case-study
research. Qualitative Inquiry, 12(2): 219–245.
George, A. L., & Bennett, A. 2004. Case studies and theory
development in the social sciences. Cambridge, MA: MIT Press.
Gephart, R. 2004. From the editors: Qualitative research and the
Academy of Management Journal.Academy of Management
Journal, 47(4): 454–462.
Gerring, J. 2007a. Case study research: Principles and practices.
New York: Cambridge University Press.
Gerring, J. 2007b. The mechanismic worldview: Thinking inside
the box. British Journal of Political Science, 38: 161–179.
Giddens, A. 1984. The constitution of society: Outline of the theory
of structuration. Berkeley, CA: University of California Press.
Gilbert, C. 2005. Unbundling the structure of inertia: Resource
versus routine rigidity. Academy of Management Journal,
48(5): 741–763.
Halfpenny, P. 1982. Positivism and sociology: Explaining social life.
London: George Allen & Unwin.
Hall, P. A. 2006. Systematic process analysis: When and how to
use it. European Management Review, 3(1): 24–31.
Harvey, J., Pettigrew, A., & Ferlie, E. 2002. The determinants of
research group performance: Towards mode 2? Journal of
Management Studies, 39(6): 747–774.
Heracleous, L., & Barrett, M. 2001. Organizational change as
discourse: Communicative actions and deep structures in the
context of information technology implementation. Academy
of Management Journal, 44(4): 755–778.
Hill, S., Martin, R., & Harris, M. 2000. Decentralization,
integration and the post-bureaucratic organization: The case
of R&D. Journal of Management Studies, 37(4): 563–585.
Hillebrand, B., Kok, R. A. W., & Biemans, W. G. 2001. Theory-
testing using case studies: A comment on Johnston,
Leach, and Liu. Industrial Marketing Management, 30(8):
651–657.
Hodgson, D. 2002. Disciplining the professional: The case of
project management. Journal of Management Studies, 39(6):
803–821.
Howells, J. 2002. The response of old technology incumbents to
technological competition: Does the sailing ship effect exist?
Journal of Management Studies, 39(7): 887–906.
Hsieh, H.-F., & Shannon, S. E. 2005. Three approaches to
qualitative content analysis. Qualitative Health Research, 15(9):
1277–1288.
Jack, G. A., Cala
´s, M. B., Nkomo, S. M., & Peltonen, T. 2008.
Critique and international management: An uneasy relation-
ship? Academy of Management Review, 33(4): 870–884.
Jacobides, M. G. 2005. Industry change through vertical
disintegration: How and why markets emerged in mortgage
banking. Academy of Management Journal, 48(3): 465–498.
Johns, G. 2006. The essential impact of context on organiza-
tional behavior. Academy of Management Review, 31(2):
386–408.
Johnson, P., & Duberley, J. 2000. Understanding management
research. London: Sage.
Johnson-Cramer, M. E., Cross, R. L., & Yan, A. 2003. Sources
of fidelity in purposive organizational change: Lessons from
a re-engineering case. Journal of Management Studies, 40(7):
1837–1870.
Kakkuri-Knuutila, M. L., Lukka, K., & Kuorikoski, J. 2008.
Straddling between paradigms: A naturalistic philosophical
case study on interpretive research in management
accounting. Accounting, Organizations and Society, 33(2–3):
267–291.
Kellert, S. H., Longino, H. E., & Waters, C. K. (Eds) 2006.
Scientific pluralism. Minneapolis, MN: University of Minnesota
Press.
Kolbe, R. H., & Burnett, M. S. 1991. Content analysis research:
An examination of applications with directives for improving
research reliability and objectivity. Journal of Consumer
Research, 18(2): 243–250.
Krippendorff, K. 2004. Content analysis. An introduction to its
methodology, (2nd ed.) Thousand Oaks, CA: Sage.
Lawson, T. 2003. Reorienting economics. London: Routledge.
Lee, A. S. 1989. Case studies as natural experiments. Human
Relations, 42(2): 117–137.
Leonard-Barton, D. 1990. A dual methodology for case studies:
Synergistic use of a longitudinal single site with replicated
multiple sites. Organization Science, 1(3): 248–266.
Lincoln, Y. S., & Guba, E. G. 1985. Naturalistic inquiry. Newbury
Park, CA: Sage.
Lindgren, M., & Packendorff, J. 2006. What’s new in new forms
of organizing? On the construction of gender in project-based
work. Journal of Management Studies, 43(4): 841–866.
Locke, K., & Golden-Biddle, K. 1997. Constructing opportunities
for contribution: Structuring intertextual coherence and
“problematizing” in organizational studies. Academy of Man-
agement Journal, 40(5): 1023–1062.
Maguire, S., & Phillips, N. 2008. Citibankers at Citigroup: A
study of the loss of institutional trust after a merger. Journal of
Management Studies, 45(2): 372–401.
Mahoney, J., & Goertz, G. 2006. A tale of two cultures:
Contrasting qualitative and quantitative research. Political
Analysis, 14(3): 227–249.
Maitlis, S., & Lawrence, T. B. 2003. Orchestral manoeuvres in
the dark: Understanding failure in organizational strategizing.
Journal of Management Studies, 40(1): 109–139.
Maritan, C. A. 2001. Capital investment as investing in
organizational capabilities: An empirically grounded process
model. Academy of Management Journal, 44(3): 513–531.
Marko
´czy, L. 2000. National culture and strategic change in
belief formation. Journal of International Business Studies,
31(3): 427–442.
McCabe, D. 2000. Factory innovations and management
machinations: The productive and repressive relations of
power. Journal of Management Studies, 37(7): 931–953.
McInerney, P.-B. 2008. Showdown at Kykuit: Field-configuring
events as loci for conventionalizing accounts. Journal of
Management Studies, 45(6): 1089–1116.
Miles, M. B., & Huberman, M. 1994. Qatative data analysis: An
expanded sourcebook, (2nd ed.) Thousand Oaks, CA: Sage.
Morgan, G. 1980. Paradigms, metaphors, and puzzle solving in
organization theory. Administrative Science Quarterly, 25(4):
605–622.
Mota, J., & Castro, L. M. 2004. A capabilities perspective on the
evolution of firm boundaries: A comparative case example
from the Portuguese moulds industry. Journal of Management
Studies, 41(2): 295–316.
Ng, W., & de Cock, C. 2002. Battle in the boardroom: A
discursive perspective. Journal of Management Studies, 39(1):
23–49.
Noda, T., & Collis, D. J. 2001. The evolution of intraindustry firm
heterogeneity: Insights from a process study. Academy of
Management Journal, 44(4): 897–925.
Noon, M., Jenkins, S., & Lucio, M. M. 2000. Fads, techniques
and control: The competing agendas of TPM and TECEX
at the Royal Mail (UK). Journal of Management Studies, 37(4):
499–520.
Noorderhaven, N. G. 2004. Hermeneutic methodology
and international business research. In R. Marschan-Piekkari
& C. Welch (Eds), Handbook of qualitative research methods
for international business: 84–108. Cheltenham: Edward
Elgar.
Theorising from case studies Catherine Welch et al
759
Journal of International Business Studies

Nummela, N., & Hurmerinta-Peltoma¨ki, L. 2006. Mixed methods
in international business research: A value-added perspective.
Management International Review, 46(4): 439–459.
Nutt, P. C. 2000. Decision-making success in public, private
and the third sector organizations: Finding sector depen-
dent best practice. Journal of Management Studies, 37(1):
77–108.
Ogbonna, E., & Harris, L. C. 2002. Organizational culture: A ten
year, two-phase study of change in the UK food retailing
sector. Journal of Management Studies, 39(5): 673–706.
Ogbonna, E., & Wilkinson, B. 2003. The false promise of
organizational culture change: A case study of middle
managers in grocery retailing. Journal of Management Studies,
40(5): 1151–1178.
O’Mahoney, J. 2007. The diffusion of management innovations:
The possibilities and limitations of memetics. Journal of
Management Studies, 44(8): 1324–1348.
Orr, R. J., & Scott, W. R. 2008. Institutional exceptions on global
projects: A process model. Journal of International Business
Studies, 39(4): 562–588.
Outhwaite, W. 1987. New philosophies of social science: Realism,
hermeneutics and critical theory. Basingstoke: Macmillan.
Pajunen, K. 2006. Stakeholder influences in organizational
survival. Journal of Management Studies, 43(6): 1261–1288.
Perlow, L. A., Okhuysen, G. A., & Repenning, N. P. 2002. The
speed trap: Exploring the relationship between decision
making and temporal context. Academy of Management
Journal, 45(5): 931–955.
Piekkari, R., Welch, C., & Paavilainen, E. 2009. The case study as
disciplinary convention: Evidence from international business
journals. Organizational Research Methods, 12(3): 567–589.
Potter, W. J., & Levine-Donnerstein, D. 1999. Rethinking validity
and reliability in content analysis. Journal of Applied Commu-
nication Research, 27(3): 258–284.
Prasad, P. 2005. Crafting qualitative research: Working in the
postpositivist traditions. Armonk, NY: M.E. Sharpe.
Pratt, M. 2009. From the editors: For the lack of a boilerplate:
Tips on writing up and (reviewing) qualitative research.
Academy of Management Journal, 52(5): 856–862.
Pratt, M. G., Rockmann, K. W., & Kaufmann, J. B. 2006.
Constructing professional identity learning cycles in the
customization of identity among medical residents. Academy
of Management Journal, 49(2): 235–262.
Ragin, C. C. 2000. Fuzzy-set social science. Chicago, IL: University
of Chicago Press.
Ragin, C. C. 2009. Reflection of casing and case-oriented
research. In D. Byrne & C.C. Ragin (Eds), The Sage handbook
of case-based methods: 522–534. London: Sage.
Ram, M. 1999. Managing consultants in a small firm: A case
study. Journal of Management Studies, 36(6): 875–897.
Ravasi, D., & Schultz, M. 2006. Responding to organizational
identity threats: Exploring the role of organizational culture.
Academy of Management Journal, 49(3): 433–458.
Redding, G. 2005. The thick description and comparison of
societal systems of capitalism. Journal of International Business
Studies, 36(2): 123–155.
Reed, M. 2005. Reflections on the “realist turn” in organization
and management studies. Journal of Management Studies,
42(8): 1621–1644.
Rihoux, B., & Ragin, C. C. (Eds) 2009. Configurational
comparative methods. Thousand Oaks, CA: Sage.
Roberts, C. 1996. The logic of historical explanation. University
Park, PA: Pennsylvania State University Press.
Rodrigues, S., & Child, J. 2008. The development of corporate
identity: A political perspective. Journal of Management
Studies, 45(5): 885–911.
Rousseau, D. M., & Fried, Y. 2001. Location, location, location:
Contextualizing organizational research. Journal of Organiza-
tional Behavior, 22(1): 1–13.
Ruben, D. H. 1990. Explaining explanation. London: Routledge.
Ryan, G. W., & Bernard, H. R. 2000. Data management
and analysis methods. In N. K. Denzin & Y. S. Lincoln (Eds),
Handbook of qualitative research, (2nd ed.) 769–802.
Thousand Oaks, CA: Sage.
Salaman, G., & Storey, J. 2002. Managers’ theories about the
process of innovation. Journal of Management Studies, 39(2):
147–165.
Sayer, A. 1992. Method in social science: A realist approach,
(2nd ed.) London: Routledge.
Sayer, A. 2000. Realism and social science. London: Sage.
Shah, S., & Corley, K. G. 2006. Building better theory
by bridging the quantitative-qualitative divide. Journal of
Management Studies, 43(8): 1821–1835.
Shapiro, D. L., Von Glinow, M. A., & Xiao, Z. 2007. Toward
polycontextually sensitive research methods. Management and
Organization Review, 3(1): 129–152.
Sminia, H. 2003. The failure of the Sport7 TV-channel:
Controversies in a business network. Journal of Management
Studies, 40(7): 1621–1649.
Stake, R. E. 1995. The art of case study research. Thousand Oaks,
CA: Sage.
Stake, R. E. 2005. Qualitative case studies. In N. K. Denzin &
Y.S. Lincoln (Eds), Handbook of qualitative research, (3rd ed.)
443–466. Thousand Oaks, CA: Sage.
Stiles, P. 2001. The impact of the board on strategy:
An empirical examination. Journal of Management Studies,
38(5): 627–650.
Suddaby, R. 2006. From the editors: What grounded theory is
not. Academy of Management Journal, 49(4): 633–642.
Taylor, B. 1999. Patterns of control within Japanese manufactur-
ing plants in China: Doubts about Japanization in Asia. Journal
of Management Studies, 36(6): 853–873.
Tolich, M., Kenney, M., & Biggart, N. 1999. Managing the
managers: Japanese management strategies in the USA.
Journal of Management Studies, 36(5): 587–607.
Tsoukas, H. 1989. The validity of idiographic research explana-
tions. Academy of Management Review, 14(4): 551–561.
Tsui, A. S. 2004. Contributing to global management knowl-
edge: A case for high quality indigenous research. Asia Pacific
Journal of Management, 21(4): 491–513.
Tsui, A. S. 2006. Contextualization in Chinese management
research. Management and Organization Review, 2(1): 1–13.
Tsui, A. S. 2007. From homogenization to pluralism: Interna-
tional management research in the academy and beyond.
Academy of Management Journal, 50(6): 1353–1364.
Van Maanen, J. 1995. Fear and loathing in organization studies.
Organization Science, 6(6): 687–692.
Watson, T., & Watson, D. 1999. Human resourcing in practice:
Managing employment issues in the university. Journal of
Management Studies, 36(4): 483–504.
West, J. 2008. Commercializing open science: Deep space
communications as the lead market for Shannon Theory,
1960–73. Journal of Management Studies, 45(8): 1506–1532.
Whetten, D. A. 2009. An examination of the interface between
context and theory applied to the study of Chinese organiza-
tions. Management and Organization Reviews, 5(1): 29–55.
White, S., & Liu, X. 2001. Transition trajectories for market
structure and firm strategy in China. Journal of Management
Studies, 38(1): 103–124.
Wilkinson, B., Gamble, J., Humphrey, J., Morris, J., & Anthony, D.
2001. The new international division of labour in Asian
electronics: Work organization and human resources in Japan
and Malaysia. Journal of Management Studies, 38(5): 675–695.
Wolcott, H. F. 2001. Writing up qualitative research, (2nd ed.)
Thousand Oaks, CA: Sage.
Wong, P. L.-K., & Ellis, P. 2002. Social ties and partner
identification in Sino-Hong Kong international joint ventures.
Journal of International Business Studies, 33(2): 265–289.
Yakura, E. K. 2002. Charting time: Timelines as temporal boun-
dary objects. Academy of Management Journal, 45(5): 956–970.
Yardley, L. 2000. Dilemmas in qualitative health research.
Psychology and Health, 15(2): 215–228.
Yin, R. K. 2009. Case study research, (4th ed.) Thousand Oaks,
CA: Sage.
Theorising from case studies Catherine Welch et al
760
Journal of International Business Studies

APPENDIX
ABOUT THE AUTHORS
Catherine Welch is an Australian citizen with a
PhD from the University of Western Sydney. She is
currently a senior lecturer at the University of
Sydney, Australia. Her current research interests lie
in the areas of qualitative research methodology
and the internationalisation of the firm. She can be
contacted at catherine.welch@sydney.edu.au.
Rebecca Piekkari is a Finnish citizen with a
PhD from the Helsinki School of Economics.
She is Professor of International Business at the
Aalto University, School of Economics, Finland.
Her research interests focus on qualitative case
studies and language issues in multinational man-
agement. She can be contacted at rebecca.piekkari@
aalto.fi.
Table A1 Categorisation of journal articles 1999–2008
Journal Year Quantitative Mixed Qualitative
a
Case Cases % of empirical
b
Cases % of total
b
Articles per volume
c
NE
d
AMJ 1999 40 0 0 0 0.0 0.0 43 3
AMJ 2000 60 4 4 2 2.9 2.8 72 2
AMJ 2001 57 2 3 9 12.7 12.7 71 0
AMJ 2002 59 1 1 8 11.6 11.6 69 0
AMJ 2003 44 0 1 2 4.3 4.2 48 1
AMJ 2004 50 1 0 3 5.6 5.5 55 1
AMJ 2005 52 0 0 5 8.8 6.9 72 15
AMJ 2006 47 0 5 3 5.5 4.9 61 6
AMJ 2007 52 1 3 3 5.1 3.5 85 26
AMJ 2008 49 1 3 2 3.6 3.6 55 0
Total 510 10 20 37 6.4 5.9 631 54
JIBS 1999 22 0 0 2 8.3 6.5 31 7
JIBS 2000 26 3 0 1 3.3 3.3 30 0
JIBS 2001 27 1 1 1 3.3 2.6 39 9
JIBS 2002 25 0 0 4 13.8 12.5 32 3
JIBS 2003 26 0 0 1 3.7 2.9 35 8
JIBS 2004 16 0 0 2 11.1 8.7 23 5
JIBS 2005 24 0 0 2 7.7 6.1 33 7
JIBS 2006 35 1 0 1 2.5 1.9 52 12
JIBS 2007 47 1 1 2 3.7 3.0 66 12
JIBS 2008 56 0 2 3 5.0 3.9 76 16
Total 304 6 4 19 5.6 4.6 417 79
JMS 1999 9 0 4 13 50.0 30.2 43 17
JMS 2000 13 1 8 9 29.0 19.1 47 16
JMS 2001 15 2 5 11 33.3 22.9 48 15
JMS 2002 14 0 4 13 41.9 27.1 48 17
JMS 2003 20 1 11 25 43.9 31.3 80 23
JMS 2004 16 0 8 15 38.5 22.4 67 28
JMS 2005 19 1 5 19 43.2 27.9 68 24
JMS 2006 21 2 5 13 31.7 20.0 65 24
JMS 2007 20 0 7 12 30.8 17.1 70 31
JMS 2008 30 1 5 13 26.5 22.0 59 10
Total 177 8 62 143 36.7 24.0 595 205
Grand total 991 24 86 199 15.2 12.1 1643 338
a
Excluding case studies.
b
Average values.
c
Excluding editorials, commentaries and notes.
d
Non-empirical.
Theorising from case studies Catherine Welch et al
761
Journal of International Business Studies

Emmanuella Plakoyiannaki is a Greek citizen
with a PhD from the University of Strathclyde,
Glasgow. She is an assistant professor of Marketing
at the Aristotle’s University of Thessaloniki, Greece,
and was a visiting researcher at Aalto University,
School of Economics, Finland, when this paper was
written. Her research interests lie in the areas of
qualitative case studies, relationship marketing,
entrepreneurship and advertising. She can be
contacted at emplak@econ.auth.gr.
Eriikka Paavilainen-Ma
¨ntyma
¨ki is a Finnish
citizen with a PhD from the Turku School of
Economics. She is currently a senior research
associate/assistant professor at the Turku University
School of Economics, Finland. Her current research
interests lie in qualitative research methodology
and visual, longitudinal and narrative methods, the
internationalisation processes of the firm and entre-
preneurship. She can be contacted at eriikka.paavi
lainen-mantymaki@utu.fi.
Accepted by Mary Yoko Brannen, Guest Editor, 25 August 2010. This paper has been with the authors for two revisions.
Theorising from case studies Catherine Welch et al
762
Journal of International Business Studies