Research Memorandum
79  March 2009
Research Quality in
Qualitative Case Studies
ADAM LINDGREEN
Centre for Marketing, Communications and International Strategy
Business School, University of Hull
Cottingham Road, Hull, HU6 7RX, UK
Email: [email protected]
MICHAEL B. BEVERLAND
RMIT University
ISBN: 978-1-906422-08-0
2
 2009 A. Lindgreen, M.B. Beverland
All intellectual property rights, including copyright in this publication, except for those attributed to
named sources, are owned by the author(s) of this research memorandum. No part of this
publication may be copied or transmitted in any form without the prior written consent from the
author(s) whose contact address is given on the title page of the research memorandum.
3
Abstract
Business-to-business marketing (B2B) research has a long tradition of using
qualitative case studies. Industrial Marketing Management, for example, has
actively encouraged the use of qualitative case study methods, resulting in many
important theoretical advances in B2B marketing. However, debate still rages
over the nature of “research quality” in qualitative case studies. Based on an
analysis of 105 case-based articles appearing in the period 1971-2006 in
Industrial Marketing Management, we analyze how authors address research
quality. We identify three key areas that future qualitative case researchers need
to address explicitly: consideration of method and research quality, presentation
of raw data, and use of triangulation. As a result of this analysis, we recommend
that researchers and reviewers become more sensitive to different case study
methodologies and epistemological paradigms, be more explicit about their
methods, present more direct evidence for their claims, and improve their
practice of triangulation.
Keywords
Case research; research quality in qualitative case research; qualitative research
4
INTRODUCTION
The use of rich qualitative case studies1 to build theory characterizes B2B
marketing research (Dubois and Araujo, 2004, 2007; Easton, 2000; Harrison and
Easton, 2004). The value of case studies to B2B marketing theory is recognized
in the call for papers of all three specialist B2B marketing journals: Industrial
Marketing Management, Journal of Business & Industrial Marketing, and Journal
of Business-to-Business Marketing. However, the nature of “research quality” in
case studies (Dubois and Araujo, 2004) and its associated practices, as
demonstrated in our paper, vary widely. For example, reviewers diverge on how
to address research quality. Some reviewers are adamant that feedback is
sought on interpretations from all informants, while others seem unconcerned
when this is not explicitly stated in an article. Likewise, reviewers have different
views on whether articles benefit from the use of direct quotes, with some
reviewers seeking further examples and others suggesting such ones are
inappropriate for managerial articles. As well, views are divergent on the value of
single versus multiple cases, use of terms such as “validity” and “reliability”, and
considerations of generalizability. As regular reviewers ourselves, it is our
experience that standards of presenting case study information vary wildly, and
we remain surprised at how few authors explicitly consider issues of research
quality in their articles. Although debate about the role of case studies in the B2B
marketing discipline occurs sporadically (e.g., Dubois and Araujo, 2007;
Hillebrand, Kok, and Biemans, 2001), explicit consideration of how authors
should improve research quality of case studies is rare, especially in comparison
to the B2B management discipline where such issues are regularly debated (e.g.,
Academy of Management Journal, 2007, 50[1]; Dyer and Watkins, 1991;
Eisenhardt, 1989, 1991; Van Maanen, 1983).
It is important to address research quality in case studies for at least six reasons.
First, attention to research quality likely leads to better practices in the field (Kirk
and Miller, 1986). Second, being sensitive to how research quality is addressed
may result in richer insights and therefore better theory generation (Yin, 1994).
Third, active debate over research quality is a sign of a healthy research
community, and thus will improve the status of the case study method
(Silverman, 2004a). Fourth, having explicit standards of research quality will
improve the trustworthiness of case studies in the eyes of readers, thereby
improving the status of the B2B marketing discipline, and potentially increasing
the impact of case study research (Lindgreen, 2008). Fifth, debates on research
quality can alleviate concerns raised by other researchers over the value of
qualitative research (including cases) in marketing (Levy, 2005). Sixth, and
finally, having clear guidelines on how research quality can be addressed in case
For readability we will simply use the terms “case studies” or “cases” to refer to qualitative case
studies unless otherwise indicated in text. Although case studies can be wholly or partially
quantitative, we limit our focus to cases that employ predominantly qualitative methods to gather
primary data because those using predominantly quantitative data address research quality
through standard tests for validity and reliability.
1
5
studies is essential for doctoral candidates in B2B marketing (Carson et al.,
2001).
Drawing on 105 identified qualitative case studies published in Industrial
Marketing Management in the period 1971-2006, the present paper contributes to
literature in the following three areas. First, the paper identifies how authors in
B2B marketing have addressed—explicitly or implicitly—issues of research
quality in case studies. Second, the paper describes trends in how research
quality has been addressed over time. Third, by comparing theory and practice,
the paper suggests areas for future discussion. These contributions provide a
timely consideration of the “state of the art” in B2B marketing case research, and
contribute to debates within the discipline.
The remaining parts of the paper are organized as follows. First, we review the
two main approaches in B2B marketing for addressing research quality in case
studies. Second, we provide details of our methods. Third, we present our
findings, focusing on three key areas for case researchers to address. A
conclusion covering practical implications then follows.
RESEARCH QUALITY IN CASE STUDY RESEARCH
Questions of research quality cannot be divorced from considerations of
epistemology (Creswell, 1998). Several different approaches exist including
positivism, interpretivism, critical studies, and pragmatism, each of which has
differing positions on the notion of research quality (Easton, 2000). Due to space
limitations we will focus on how quality is operationalized in practice by
researchers from the two main schools of case research in B2B marketing:
positivism and interpretivism. These two schools were identified as the two
dominant methods of case study research based on an analysis of key citations
used by B2B marketing case researchers. Derived from our research (see
Section 3), we found that positivist citations often include Eisenhardt [1989],
Glaser and Strauss [1967], Strauss and Corbin [1998], and Yin [1994], whereas
those favoring interpretivism often quote Lincoln and Guba [1985]) and the
practices used.
The Positivist Approach to Research Quality in Case Studies
Positivists address research quality in case studies using tests similar to those of
quantitative researchers. That is, research quality is spoken of in terms of validity,
reliability, and to some extent generalisability (although positivists often limit case
studies to an exploratory role; Harrison and Easton, 2004). Researchers in this
tradition recognize they may only ever gain an approximate understanding of
reality; but nevertheless they do argue for a “right way” of interpreting case data
(cf. Kirk and Miller 1986, p. 11). Table 1 identifies how positivist researchers
address research quality in case studies. Our findings suggest that positivism is
6
the dominant school of thought in B2B case study research (although this is
judged primarily on how authors explicitly address case research).
For positivists the primary means of addressing concerns over validity is
triangulation. Triangulation involves “using different methods or sources to
corroborate each other” (Mason, quoted in Silverman, 2004a, p. 98), and it can
involve combining different data sources, theoretical perspectives, methods, and
researchers (Silverman, 2004b). Triangulation is seen as useful for overcoming
partiality, which may arise from reliance on a single source, and is important
because it ensures the convergence of data around key themes and also
improves comprehensiveness. Triangulation involves researchers looking for
different sources of evidence to support a single contention, while remaining
aware that each source of data has limitations due to its context-bound nature
(Silverman, 2004b).
A means of enhancing construct validity is to provide readers with a clear chain
of evidence including examples of raw data to back up the claims being made. In
contrast to interpretive research this data is presented to the reader with no
interpretation (Alvesson and Sköldberg, 2000). That is, passages from subjects
or interviewees (the term “informants” is used by interpretivists) or other forms of
data should be unambiguous to multiple researchers. To avoid claims of
anecdotalism,2 researchers should be prepared to provide readers direct access
to their case database, coding scheme, and case memos, which have not been
included in the article (Silverman, 2004b).
Another means by which researchers can achieve construct validity is to seek
direct feedback from the interviewees on whether their case memos and
emergent theory correctly capture reality (Yin, 1994). Although doubts have been
raised over the strength of such checks (Silverman, 2004a), member checking
can aid in error reduction, and is useful for improving validity.
Bryman (quoted in Silverman, 2004b, p. 34) describes the problem of anecdotalism as: “There is
a tendency towards an anecdotal approach to the use of data in relation to conclusions or
explanations in qualitative research. Brief conversations, snippets from unstructured interviews
[...] are used to provide evidence of a particular contention. There are grounds for disquiet in that
the representativeness or generality of these fragments is rarely addressed.”
2
7
Table 1: Positivist Research Quality Criteria for Case Studies
Design test
Construct
validity
Theoretical explanation of the concept
To secure that correct operational measures have been
established for the concepts that are being studied
Operationalized through
1. Triangulation through multiple sources of data or
interviews
2. Provision of a chain of evidence using cross-case
tables or quotes from informants
3. Interviewees review the draft case and give feedback
Internal validity
To make sure that a causal relationship—certain
conditions lead to other conditions—has been
established. Internal validity is a concern of explanatory
or causal cases studies but not for exploratory or
descriptive cases that do not attempt to make causal
statements
1. Pattern matching through cross-case analysis
2. Search for negative cases, ruling out or accounting for
alternative explanations
3. Time series analysis
External validity
To prove that the domain, to which a case study‟s
findings belong, can be generalized
1. Specification of the population of interest
2. Replication logic in multiple case studies
Reliability
To demonstrate that a case study‟s findings can be
replicated if the case study procedures are followed
1. A standardized interview protocol
2. Constructs well defined and grounded in extant
literature
3. Audit trail by providing access to data
8
Internal validity is important for reliability because it allows researchers to
replicate results. Positivists address this issue in three ways. First, they go
beyond within-case analysis to look for patterns across cases (pattern matching).
This can involve the use of multiple case studies (in which case researchers seek
to replicate their findings much like quantitative researchers do; Harrison and
Easton, 2004) or multiple interviews / observations within a single case
(Eisenhardt, 1989; Yin, 1994). The aim of this practice is to ensure theoretical
saturation of emerging constructs by identifying multiple instances of the same
phenomena (Glaser and Strauss, 1967).
Second, and related to the first practice, positivists seek out negative cases or
instances. This practice is a form of falsification whereby researchers subject
emerging theories to strict tests. Negative cases can play another important
role—if the emerging theory can predict their occurrence then validity is
enhanced (Yin, 1994).
Third, positivists seek longitudinal data to look for causal patterns and drivers
(Eisenhardt, 1989). That is, researchers following this practice may conduct
multiple interviews over time, enhance interview data with historical information,
build up a case history from primary and secondary data, or ask interviewees to
reflect on the past.
Although positivists often downplay the ability of case studies to generalize to a
wider population, nevertheless they encourage researchers to engage in
practices that enhance generalizability (see Eisenhardt, 1989). Two primary
methods are used to address external validity: specifying the population of
interest and replication logic. The first practice involves selecting cases that are
representative of the wider industry population, or providing a representative
case. The second practice involves using multiple cases to “test” emerging theory
much in the way experimental researchers seek to replicate emerging results.
Using this practice, researchers deliberately seek out cases where they think
their initial findings will be replicated or negated for predictable reasons (Perry
1998).
In case research, external reliability is difficult because findings are temporally
and contextually bound (Harrison and Easton, 2004). As such, case researchers
primarily focus on enhancing internal reliability, or “The extent to which, given a
set of previously generated concepts, new researchers could match these
concepts with the data in the same way as the original researchers” (Murphy et
al. 1998, p. 176).
Internal reliability is addressed by positivists in three ways: a standardized
interview protocol; well-defined constructs that are grounded in the extant
literature; and providing an audit-trail for future researchers. The first method is
controversial because even in cases where replication logic is employed,
researchers may quickly need to depart from a prepared set of questions to
9
probe specific leads to derive new insights (Strauss and Corbin, 1998).
Nevertheless, researchers should at least provide some information about the
questions asked at each case and the justification for them. The second method
allows other researchers to more easily replicate results and add incrementally to
theory. Finally, the third method requires the researcher to provide a detailed
exposition of the research process because data is shaped by the researcher
and depends on the context. Researchers should provide enough information so
that future researchers can replicate their methods, even though they may not
gain the same result due a lack of insight or ability (Silverman, 2004a). A detailed
explanation of how the data was collected and analyzed should be provided
(Silverman, 2004b). Importantly, researchers could ensure reliability by providing
direct quotes or examples of data in their articles for others to check and make
assessments regarding objectivity, and even provide interested readers with
access to the raw transcripts.
The Interpretivist Approach to Research Quality in Case Studies
A summary of interpretivist approaches to improving research quality in
qualitative case studies is provided in Table 2. The interpretivist view calls on
researchers to address issues of research quality, but suggests different
standards—standards that are not simply taken from one domain (quantitative)
and applied to another (qualitative case studies). Such an approach has been
used in recent B2B marketing research (Beverland and Lockshin, 2003; Flint,
Woodruff, and Gardial, 2002). Interpretivist researchers address research quality
through methods that help ensure the reported findings provide a trustworthy
account.
Interpretivists differ from positivists in a number of ways. First, interpretivists
focus on understanding “informants” (not distant “subjects” or “interviewees”) and
their reality (Silverman, 2004a). Second, although interpretivist researchers
attempt to maintain some form of distance from the subject matter (in that they
remember they are researchers and not advocates), they nevertheless play a far
more active role in constructing the theoretical account (Stake, 2005). Third, the
theoretical account is the result of an interaction between a researcher and the
informant (Perry, 1998). Fourth, some interpretivists challenge the notion of
quality in research, arguing that each interpretation is unique, and that replication
is therefore impossible (Easton, 2000), although others suggest that
interpretations can be improved and that some are more trustworthy than others
(Lincoln and Guba, 1985). Fifth, the desire for trustworthiness is assessed by a
case‟s confirmability, credibility, transferability, and dependability.
10
Table 2: Interpretivist Research Quality Criteria for Case Studies
Design test
Confirmability
Theoretical explanation of the concept
The extent to which interpretations are the result of the
informants and the phenomenon as opposed to
researcher bias
Operationalized through
1. Multiple perspectives on phenomenon of study
2. Multiple coders and interpretations
3. Interpretations presented to colleagues and informants
/ population members
4. Use of “grand tour” questions to allow informant
control
5. Trust between informant and researcher
Credibility
The extent to which the findings appear to be acceptable
representations of the data
1. Triangulation (data, informants, and methods)
2. Review of emergent findings by colleagues and
informants
Transferability
The extent to which findings from one case study in one
context will apply to case studies in other contexts
1. Theoretical sampling
2. Relating findings back to wider industry and market
context to identify boundary conditions
3. Presenting findings to population members or
informants in other similar populations
Dependability
The extent to which a case study‟s findings are unique to
time and place; the stability or consistency of the
explanations
1. Multiple entries to site of interest
2. Informants asked to reflect on current and past
practices
11
Interpretivists often argue that despite claims of objectivity and distance, the
practices used by positivists often result in the very bias they seek to avoid
(Alvesson and Sköldberg, 2000). Thus, positivists may miss important
insights, ignore rival explanations, or confirm their own theoretical
predispositions (Lincoln and Guba, 1985). Instead, interpretivists seek to be
aware of and minimize personal bias in the research process (importantly,
they should openly state their biases in their articles and to informants—a
practice called reflexivity; Van Maanen, 1983). To do so, researchers seek to
ensure confirmability through openly inviting multiple perspectives, defending
their interpretation to informants, colleagues, and members of population from
which the informants came (including a comprehensive explanation for why
things have been defined in a certain way and why certain methods were
employed). As well, they can reduce bias through multiple coders and
interviewers (Alvesson and Sköldberg, 2000).
Importantly, confirmability (and credibility) is enhanced by an interviewer
allowing informants to speak on their own terms. This can be achieved
through “grand tour” questions (broad questions that allow the informant
plenty of scope) rather than more directive questions (McCracken, 1988). In
doing so, it is more likely that a genuine or authentic informant‟s account will
emerge from the interview. As well, the interviewer focuses primarily on
listening and understanding the informant, thus minimizing the interviewer‟s
active role in the interview. As such, the interviewer builds up trust with the
informant who will be more likely open up to a non-judgmental and directive
interviewer. Trust can also be built up through being non-judgmental, allowing
informants to speak about things they believe are important, interviewing the
same informant several times, and conducting the interview in a place of the
informant‟s choice (Flint et al., 2002).
Triangulation is central to research quality in case studies per se, including
interpretivist case research. For interpretivists, triangulation is achieved by
using multiple informants, multiple sources of data, where possible multiple
researchers, and seeking out multiple views of the interpretation. These
means help to build credibility. Also important in assessing credibility is the
presentation of informant passages, field notes, or images along with an
interpretation of such data by the informant. Thus, reviewers and readers can
assess whether the interpretation offered is reasonable and therefore
credible. Credibility can also be improved by allowing informants to review
emergent findings (Lincoln and Guba, 1985).
Rather than consider generalizability, interpretivist researchers consider the
extent to which their findings may transfer across contexts (Creswell, 1998).
They achieve this through three means: theoretical sampling, identifying
boundary conditions, and member checking. First, theoretical sampling
involves selecting cases that are most likely to yield rich theoretical insights
(Yin, 1994). Sampling decisions are considered of utmost importance for case
researchers (Dubois and Araujo, 2007). Since interpretivists seek to
generalize to theory rather than a wider population, discussions of sampling
decisions are critical (Creswell, 1998). Second, and related to the previous
point, is the need to be sensitive to the boundary conditions of the case study
12
findings. For example, in presenting their theory of commercial friendship,
Price and Arnould (1999) explored the specific nature of their relationship
context, thus identifying important moderators that may influence the extent to
which their findings can be transferred across different theoretical contexts.
They were able to do this by being sensitive to the nature of the industry and
relationship context in which they developed their case studies. Thus, they
suggested limits to how far friendships and relationship metaphors could be
transferred to commercial contexts, opening up a range of possible future
avenues for research. Third, member checking involves presenting findings to
other members of the population (rather than the informants) of interest or
similar populations. For example, in seeking to achieve transferability of their
findings on managing advertising-client relationships Beverland, Farrelly, and
Woodhatch (2007) presented their results to account managers in other
ongoing business service contexts such as design, market research, and IT
provision.
Interpretivists note that there are several barriers to reliability in case
research, including the difficulty of exactly replicating research that is carried
out in a natural setting, and the inability to control subject behavior. Also,
since qualitative case researchers assume that phenomenon under study will
change, reliability is difficult (Lincoln and Guba, 1985), and thus dependability
should be used. To ensure dependability, researchers should take into
account the inherent instability of the situation (Harrison and Easton, 2004;
Murphy et al., 1998) by presenting evidence of variability and change from the
interviewees. This can be achieved through spending extended time in the
field, multiple entries to the site of study (including multiple interviews with the
same informant), and asking informants to reflect on past and current
practices (Creswell, 1998). In this way, dynamism can be captured and an
explanation for it offered (Perry, 1998).
Judging Research Quality in Qualitative Case Studies: Summary and
Synthesis
The two previous sections identify the differences between positivists and
interpretivists in addressing research quality in case studies. First, although
many researchers often mix the two traditions (see Silverman 2004a for detail
on the misuse of founding texts or references by qualitative researchers), our
position—or our bias—is that researchers should adopt a more purist position,
and that their cases should therefore be assessed against the standards of
their chosen paradigm (a position also suggested by Lincoln and Guba, 1985;
Strauss and Corbin, 1998). Second, it is important to note that the methods
identified above must be demonstrated in an article, rather than just stated.
That is, research quality must be shown in the writing up of the article, as
methods are a mode of inquiry rather than a series of checks conducted after
the data has been collected (Silverman, 2004a, 2004b). Based upon this
review, we believe that central to both approaches to research quality are the
following three practices:
13
1. An explicit consideration of epistemological stance and issues of research
quality.
2. The provision of clear detail about the methods used, including sampling
decisions, data collection tools, and analysis.
3. Triangulation of data, methods, and researchers. This can include some
form of informant or collegial feedback, or member checking.
Evidence for these three practices was sought across the 105 case studies
published in the period 1971-2006 in IMM.
METHODS
Data for this paper were drawn from published issues of IMM in the period
1971-2006. Besides space considerations, we chose IMM because the journal
is recognized as the leading journal in industrial marketing and is a top-10
journal by influence within marketing over a 30-year period (Baumgartner and
Pieters, 2003). As such, IMM was judged to provide preeminent examples of
case research within the sub-discipline of B2B marketing. For the purposes of
this paper we define case studies as “an exploration of a „bounded system‟
[bounded by time and place] or a case (or multiple cases) over time through
detailed, in-depth data collection involving multiple sources of information rich
in context” (Creswell, 1998, p. 61; see also Stake, 2005, p. 485).
Cases were identified in a number of ways. First, we conducted a keyword
search (looking for the terms “case study” or “qualitative”) of articles published
in IMM. A typical example of this was “Its practical application is demonstrated
through a case study in industrial engineering and construction.” (Mühlbacher,
Dreher, and Gabriel-Ritter, 1994, p. 287) This process resulted in a dataset of
125 articles. Second, we read each article carefully in order to assess if the
qualitative articles met the definition provided above, and to see that case
studies were, in fact, based primarily on qualitative data. We removed articles
that were theoretical discussions of the case method (n=3), purely quantitative
cases (n=16), and one example of a case study that had been printed twice.
Third, as this process failed to identify articles we knew were case studies, we
then conducted a further search reading through the abstracts and, if
necessary, the method sections of each published article to identify further
case studies. As a result of this process, we identified a total of 105 articles
that focused on qualitative case studies. Although we report on many of these
cases directly, space considerations preclude from providing a full list;
however, we are happy to provide this database should other researchers
desire it. Figure 1 depicts a trend analysis of case publication vs. total number
of articles and issues during the period 1971-2006. The final dataset
contained examples of all three types of case studies, including intrinsic
(highly descriptive, non theory-driven cases), instrumental (theory-building
cases), and multiple or collective case studies (Stake, 2005).
The analysis occurred in three phases. First, we analyzed each published
article on the basis of its explicit consideration of case research quality. The
criteria we used were drawn from the literature on quality in case research
14
(see Tables 1 and 2). Second, given that relatively few researchers explicitly
addressed even rudimentary issues of research quality, we analyzed each
published case study for evidence of quality criteria being used. This process
was done by both authors and involved two stages—within-case analysis and
cross-case analysis (Eisenhardt, 1989). The first stage involved a careful
reading of each individual article (within-case analysis). Following this, both
authors wrote memos on each article, identifying key issues and practices
(Strauss and Corbin, 1998). Second, both authors analyzed the cases
according to four time-periods (1971-1979, 1980-1989, 1990-1999, and 20002006). These periods were relatively arbitrary, but were useful devices in
breaking up the data into manageable chunks and also in identifying trends in
how authors of the articles dealt with issues of research quality in practice.
This process of cross-case analysis involved looking for patterns across timeperiods, and enabled the identification of exemplar cases, and problems and
outcomes in relation to how authors addressed case research quality.
The third and final round of analysis involved repeating the above steps on
multiple case studies (publications of multiple case studies in IMM increased
dramatically from 1990). We analyzed these articles separately because
presenting information from several cases gives rise to particular problems
(Dyer and Watkins, 1991; Easton, 2000). Our final interpretation was
presented to colleagues in two seminars to gain further feedback, and two
anonymous reviewers for this paper.
Figure 1. Cases and total papers published in IMM, 1971-2006
15
FINDINGS
In this section we will examine how authors have addressed research quality
in articles published in IMM. Since this task involves critically analyzing
colleagues‟ work we must approach the task with a degree of sensitivity. First,
authors publishing case research in the 1970s and early 1980s had few
guidelines to work from to address issues of research quality. Second, it is
possible that due to space considerations authors removed parts of their
article that addressed issues of research quality. For example, information
that may have helped reviewers positively judge the article may have been
removed prior to publication in order to reduce page space or improve
readability. Third, we are examining articles from a particular point of view—
the impression an article leaves on us when reading it with research quality in
mind. That is, we are not stating that the article is unimportant, or poor quality
per se. Fourth, we are also sensitive to the fact that authors must make
choices in what they report, and that these choices will be determined by the
aim of the article—an article focusing on building theory or exploring new
practices may therefore focus on this contribution rather possibly at the
expense of reported quality. As a result, when we use the terms “research
quality” or “case quality” we use these terms to refer to the extent authors
addressed the criteria in Tables 1 and 2.
Nevertheless, we will interpret passages from articles and identify the
impression they leave upon readers, positive or negative. Passages are
contained in Exhibit 1. Based on our analysis, we identified the following three
themes.
Exhibit 1: Sample Passages from Published Case Studies in Industrial
Marketing Management, 1971-2006
Passage 1: “The process of negotiation, from two in-depth case studies involving Swedish
firms and firms in India and Nigeria, are compared with the process involving two Swedish
firms. Case I dealt with the negotiation process between Defibrator, a Swedish supplier of
Pulp Mill and the Hindustan Paper Corporation (HPC) of India. Case II dealt with the
negotiation process between two Swedish firms ASSI (Statensskogs Industrier), the buyer,
and Sunds AB, the supplier. In Case III, the seller was Power, supplier of electric power
system, and the buyer was a state enterprise, Tender Board (TB) in Nigeria” (Ghauri, 1988, p.
49).
Passage 2: “This article is based on longitudinal observations of 5 to 15 years of five new
product cases dealing with small companies” (Sarin and Kapur, 1990, p. 301).
Passage 3: “The case details and summary tables in the appendix help to illustrate a certain
number of phenomena relative to the key accountization process” (Pardo, Salle, and Spencer,
1995, p. 127).
Passage 4: “The decision system analysis of this article was initiated by performing an indepth interview with the person(s) involved in the buying process. The researcher then
obtained a detailed description of the basic sequence of behaviors exhibited in procurement.
The behaviors were “transcribed and then broken down into a sequence of short phases,
each phase corresponding to a single task relevant statement.” [Reference] These statements
were assembled in flowchart form to summarize the whole process and to reveal any
extraneous factors that may have affected the procurement process. A follow-up interview
was conducted to insure the accuracy of the findings. The subject of this study is the
purchasing process of creosote distribution poles” (Wilson, 1984, p. 195).
16
Passage 5: “.... a seasoned resin technologist cynically suggested that no further interviews
were needed, but that merely a plot be made of pentaerythritol consumption over the last few
years and an “absolutely straight line be drawn right through to 1970 or 1975, according to
which is desired.”.... 1. Chemical Age, April 1961: When dyeing problems are solved
3
polyproylene will find wide textile usage, quoting ICI spokesman” (Kratschmar, 1972, p. 271).
Passage 6: “These two investigations carried out within a week quickly established that: 1.
The physicist‟s estimates of outsiders‟ usage of CMS equipment were too high; 2. That the
type of service offered to potential users would be more costly that originally conceived. This
was because a need was uncovered for advisory services for potential users…” (Cowell and
Blois, 1977, p. 332).
Passage 7: “The following 22 success stories are drawn from a larger sample of contest case
studies conducted by WTCA headquarters among NETWORK users during 1988-1990. In
total, they illustrate well the marketing uses and advantages of NETWORK for prize-winning
small firms on a variety of industrial sectors. Alcoholic beverages. A trading company posted
an item in offers to sell offering “Alcoholic beverages—all types and brands.” They received
more than 40 responses from all over the world. According to WTCA‟s last information, more
than half of those had resulted in firm contacts. Antifreeze. Petroil Industries found Taroko
Enterprises through NETWORK. This led to a sale by Petroil of 9,000 gallons of antifreeze, a
$57,000 transaction” (Holden, 1991, p. 165).
Passage 8: “…the attitude to business development can be assessed from the following
statement: “The continuing development of sales of casting to Japan, the United States, and
Germany will be dependent upon future movements in exchange rates. At levels around
US$0.80, little additional business will be achieved from these countries; however any drop to
around US$0.75 will result in some significant orders.” This view indicates a cost, price-taker,
product oriented approach, rather than a market, demand-oriented approach of the successful
businesses” (Harker, 1998, p. 323).
Passage 9: “Secondary information in the form of company reports, product brochures and
marketing literature was collected. These informed the researchers with background
information on the companies, their size, activities, and involvement with larger customers.
Company and industry web sites were visited before conducting interviews to establish how
the companies viewed and presented themselves” (Johnsen and Ford, 2006, p. 1008).
Passage 10: “To aid in data collection and ensure consistent interviewing procedures, we
drafted an interview guideline rooted in our intention to explore key issues and problems of
supplier involvement in NPD (Miles & Huberman, 1994). The interviews lasted 30-60 min with
the first part focusing on supplier involvement on the organizational level and the second part
on the project level. The interviewer used follow-up questions to further explore constructs,
patterns, interrelations, and particular situations of the firm, its collaboration with suppliers or
the NPD project. Data recording involved transcriptions based on the interview guideline
components” (Wagner and Hoegl, 2006, p. 938).
Passage 11: “In terms of addressing the study‟s validity, the issue of access is again central.
Remenyi, Williams, Money and Swartz (1998) for instance argue that validity results from
gaining full access to the knowledge and meaning of the respondents. More rigorous tests of
validity were addressed through the use of multiple sources of evidence and data
triangulation—particularly in relation to subjective and controversial issues…” (Salonen,
Gabrielson, and Al-Obadi, 2006, p. 745).
Passage 12: “However, some of the actors involved in the two projects have also been
interviewed. When this was not possible we have used reports written by these actors as well
as internal project documents and memos from meetings with these actors to include their
perspectives as well. Despite this, a limitation of our study is the predominant focus on the
focal actor Alpha” (Windhahl and Lakemond, 2006, p. 810).
3
Please note all passages from published articles have been transcribed verbatim, and
include any grammatical and typographical errors in the original.
17
Too Little Detail on Methods and Quality Criteria
As identified in the literature review, researchers need to be explicit about
their epistemological position (positivist or interpretivist) and in how they
address issues of research quality. Also, details need to be provided about
the nature of the research design, and justification for such choices. In
particular, significant details must be provided about sampling strategies since
sample choices are central to quality in case study research (Dubois and
Araujo, 2007). Such explicit consideration assists readers in judging the
quality of the research.
Although considerations of quality are explicitly addressed by quantitative
researchers, as Table 3 identifies, less than half of all articles published in
IMM explicitly justify their choice of method in terms of research questions,
prior research, or direct appeals to the literature. Fewer still address issues of
research quality including issues of validity4 (22.8%), reliability (16.2%), and
generalizability (23.8%) or their interpretivist equivalents. Furthermore,
researchers often only provide limited detail on sampling decisions, data
gained, analysis, and questions asked at interviews or time spent in field. Our
analysis indicates, however, that reporting this detail has improved from 2000
although this could be a function of an increase in the number of published
case studies.
Table 3: The Nature of Case Studies and Research Quality Published in
Industrial Marketing Management, 1971-2006
Number of cases:
Single
Multiple
Explicit justification of case method:
Yes
No
Methods:
More than one
Secondary analysis
Observation
Interviews
Reliability explicitly addressed:
Yes
No
Interview and coding process described:
Yes
No
Number of coders:
Single
Multiple
Validity explicitly addressed:
Yes
No
4
Frequency
Percentage
54
51
51.4
48.6
49
56
46.7
53.3
65
18
14
8
61.9
17.1
13.3
7.6
17
88
16.2
83.8
43
62
41
59
86
19
89.9
18.1
24
81
22.8
77.2
So few researchers considered issues of validity that we had to lump all considerations of
validity (construct and internal) into one umbrella category “validity”. We choose to focus on
external validity, as a separate issue as more authors considered this, but at the end of their
articles.
18
External validity (generalizability) explicitly
considered:
Yes
No
25
80
23.8
76.2
Typical examples of the limitations identified above from this period are
contained in passages 1-7 and 9-12 in Exhibit 1. Passage 8 provides the
basis to discuss a counter example. Passages 1-3 represent the “methods”
section of each article. All authors address important areas (influences on
business negotiation, causes of new product failure, and key account
management practices respectively), and identify key themes, but at no time
do readers get a sense of the actual data gained, the experiences or views of
any informants (assuming there were some), or any sense of triangulation
(despite the claims of the author in passage 3 no case details are forthcoming
in an appendix—one only gets a sense that the case is an old French firm). In
these cases, it is virtually impossible to form a judgment of research quality.
As a result, readers are asked to simply trust the researchers on three
important areas of B2B research (this is especially so since readers never
gain any access to raw data—see section 4.2).
Passage 7 is also taken from a case that suffers from similar problems. The
first part of this passage contains the entire information about the method.
Positivists might immediately and legitimately ask: “How large was the original
sample from which these 22 cases were drawn?” “What was the success rate
vs. failure rate for these firms?” “Was this rate any different to what they had
experienced prior to the use of NETWORK?” and “What did the firms involved
report about NETWORK and its benefits?” Interpretivists would wonder why
information was presented in such a passive summary form, wonder why so
little rich information was gained from this data set, and suspect that
information may have been taken out of context.
In contrast, passages 4 and 8 come from cases containing information that
would assist with judgments of research quality. Passage 4 contains a
number of details necessary to assist with quality judgments. First, it details
the informant (although provides little information about this individual), the
firm, the focus of the interview, method of analysis (including a reference to a
previous article detailing how to analyze information of this nature), and
included a follow-up interview to check to accuracy of findings. But, much is
also left out—little information is provided about the firm, the industry, and
even the geographic location of the sampled case (useful for positivistic
assessments of generalizability and transferability), no detail on questions or
coding is really provided, and we gain no information on the result of the
follow-up. Therefore, when coupled with a lack of informant quotes and no
triangulation, it is difficult to assess the overall quality of this case.
Passage 8 is taken from Harker‟s (1998) article. This article in many ways is
an exemplar positivist case study. Harker provides the following information: a
direct link between the research questions and choice of method, a defined
unit of analysis, samples from a defined population (Australian firms
employing between 300-500 people), detail on the environment in which these
19
firms competed (highly dynamic), a defined sample (firms that had
experienced two years of decline followed by two years of success in absolute
and relative terms), clear descriptions of data sources and the time frame
which the data encompassed, information on three sources, and detail on
analysis and coding. The only fault in this article (from a research quality point
of view) was a lack of detail about the interviews conducted and the questions
asked. Such detail and the subsequent presentation of data (see Passage 8
and section 4.2) clearly reinforce the positivistic approach taken in the article.
Although passages 1-3 are taken from cases published prior to 2000, trend
analysis (see Figure 2) indicates that only a handful of researchers in recent
years (2000 onwards) consider issues of research quality explicitly. Although
the trend is improving, just as many authors publishing cases in IMM after
2000 are not explicitly addressing quality criteria, despite the widespread
existence of positivist and interpretivist criteria for quality in case research.
Also, the trend may simply reflect the large increase in published cases in
IMM since 2000. Of those articles that did address quality issues explicitly,
preference was given to descriptions of the interviewing and coding process
(which would assist with reliability) and, to a lesser extent, the use of multiple
coders (18.1%).
Figure 2: Trend Analysis for Explicit Use of Research Quality in IMM,
1971-2006
20
Passages 9-10 are taken from articles with many positive aspects from the
point of view of research quality—adequate justification for the use of case
methodology, background detail on the case(s), and some method detail. Yet,
both passages, in the context of the total article, are also problematic on
quality grounds. In regards to passage 9, readers have little insight into the
amount and quality of material. Although this form of reporting the use of
secondary information is replicated in many cases, in the context of this
article, little insight is gained into what type of companies were studied
(reference is made to small and some large European firms that did business
with UK suppliers).
Passage 10 has similar problems and is typical of many recently published
cases—some method details are given, but not enough to aid judgments of
quality. For example, although it is stated in passage 9 that an interview guide
was used, it was not provided in an appendix. Despite some broad categories
of issues focused on, we gain no insight into the actual questions asked, how
many follow-up interviews were conducted, and whether the initial findings
were checked with informants. As well, the last sentence is all the information
we ever gain on how the data was analyzed (although transcription is useful, it
hardly counts as a form of analysis). In fact, descriptions of coding in most
cases consisted of references to grounded theory procedures such as open,
axial, and selective coding (see Strauss and Corbin, 1998) rather than clear
descriptions of codes and analysis procedures.
This lack of description of procedures makes replication, as well as judgments
of validity, dependability, and trustworthiness difficult (although not
impossible), thus undermining perceptions of research quality.
Increased Access to Raw Data
The problems identified in Section 4.1 are compounded when authors provide
little or no direct access to data in the form of informant passages or
secondary data—a practice central to both positivist and interpretivist schools
of quality. Typical of many published case studies is the lack of raw data
presented to readers. Instead, data is often presented in summary form, with
(in rare cases) some small snippets of text from informants or secondary
sources used as a narrative device. Passages 1-7 and 9-12 in Exhibit 1 reflect
this issue in various ways (passages 5 and 8 are reflective of a contra
example).
Passages 5, 8, and 6-7 are worth comparing on this point. Passage 5 (notably
from 1972) examines the research process behind breakthrough products by
drawing on the experiences of seven cases. The first part of the passage
provides supportive evidence for the overall point that the case in question is
explaining—that projections for breakthrough product success are often
simplistic. The second part provides evidence from secondary sources that
reinforce the point that desk research can be a useful source to justify
investing in breakthrough product development. However, this direct quote
from an informant is the only one used in the entire article while the secondary
source is one of just four reported. Such limited reporting of data can give rise
21
to impressions of anecdotalism (Silverman, 2004a), although tradeoffs are
necessary when presenting multiple case studies.
In contrast, passage 6 represents part of an article aiming to demonstrate the
value of different research methods for predicting potential demand. Passage
6 contains information to suggest that demand projection is very difficult
because buyers are uncertain of the value of the product offered. However,
this passage (which is representative of the article) effectively asks us to trust
the authors on this point—there is no direct evidence from the informants or
even any provision of information as to how the investigators ascertained this
finding. Passage 7 comes from a case with similar problems.
One outcome of providing limited access to data (especially when little detail
on method is also forthcoming) is that readers could perceive that data has
been “forced” to fit theory.5 Passage 7 is an example of how this perception
can arise. The author structures the article in the following way: a long
introduction followed by a literature review that develops a new model or
advocates a new approach, followed by a brief statement on method, followed
by a short findings section (relative to the length of the article) without any raw
data to allow readers to judge how the authors came to their conclusions. As
identified in passage 7, the findings are presented in summary form and
support the authors‟ original contentions.6 Many of the cases we analyzed
(e.g., Brennan and Turnbull, 1999; Ford and McDowell, 1999; Håkansson,
Havila, and Pederson, 1999; Loeser, 1999; Möller and Rajala, 1999) were
structured similarly and suffered from this problem.
In contrast, other articles were structured in such a way as to allow the reader
to form an independent judgment of the authors‟ claims. For example,
passage 8 is one of several examples presented by Harker (1998) where the
author regularly moves between interpretation and raw findings (in fact, one
quote is often provided for each important theme that Harker derives),
allowing the reader to judge whether the author‟s interpretation is valid or
reasonable. In contrast to passages 6-7, passage 8 starts with a short
description of the firm‟s attitude to business development. Note, Harker does
not tell readers what the attitude actually is, but provides them with the
informant‟s statement, and then an interpretation, which is a reasonable view
of the firm‟s strategic outlook (especially when coupled with sample detail
information about the case and industry environment). This passage allows
readers to do several things. First, the readers can interpret the data directly,
thus allowing for replication and validity. Second, the readers can make up
their own mind about the firm‟s strategic stance rather than having to trust the
author. Third, the readers can make a total judgment about the validity of the
findings.
5
Although we are not saying that these authors did force data to fit their preconceived
theories, a reader could reasonably draw such a conclusion based on how these articles are
structured, and on the lack of information to make a judgment to contrary.
6
This form of presentation represents the positivistic influence of Eisenhardt (1989) on case
research whereby cases effectively test propositions (Easton, 2000). However, the lack of
direct access to raw data and the structure of the articles make judgments of quality difficult—
similar concerns would be raised by interpretivist readers.
22
Presenting data from multiple case studies poses a number of problems
because trade-offs must be made in terms of richness and generalizability
across the cases (Dyer and Watkins, 1991). Authors publishing multiple case
studies in IMM adopted three practices (or mixes of each) to data
presentation: cross-case tables (see, for example, Matthyssens and Faes,
1985), presenting summaries of each single case and then discussing key
themes (see article containing passage 2), and using snippets from a range of
cases to support propositions or key themes (passage 8 and Alam, 2006;
Beverland, 2005; Mason, Doyle, and Wong, 2006).
By themselves, each practice can present problems. For example, reliance on
cross-case tables means that readers have access only to summaries of data
thus making judgments of quality difficult (in fact, this method comes closest
to treating case information as quantitative frequencies). As well, any richness
associated with the individual case is lost.
The second practice—presenting summaries of each case and then
discussing key themes—gets unwieldy as the number of cases increases,
thereby decreasing readability. Nevertheless, such an approach goes some
way to retaining the richness of each individual case. However, problems can
arise because in summarizing information from multiple cases it is often the
direct quotes that are lost. For example, the article from which passage 2 is
taken provides summary information about new product failure from three
cases and then synthesizes the key themes across the cases. However, at no
time do readers gain access to raw data and information, and this means that
despite the increased sample size any judgment of quality remains difficult.
The third practice tries to balance the requirements for richness and
generalizability across cases (as well as theoretical saturation; Strauss and
Corbin, 1994), but can give rise to claims of anecdotalism because readers
only get presented with select snippets of case material that appear to support
the authors contentions. As well, such an approach also comes close to
replicating quantitative research in that multiples quotes can be used as
“proof” of an author‟s contention. For example, passage 7 comes from an
article using this approach. In this article (admittedly an extreme version of
this practice), the entire sample (22 cases) are summarized and presented as
proof for the superiority of an early e-procurement model. Yet, in doing so, not
only are readers left without any raw data to form their own judgments about,
the author provides no real interpretation of his data (in fact, one wonders why
a survey design was not used in this instance), instead presenting the
summarized information as hard evidence (when, in fact, the process of
summarizing has involved authorial choices).
Demonstrating Triangulation in Writing-Up
Tables 1 and 2 identify the importance of triangulation to ensuring research
quality. We found relatively few authors discussed triangulation or used it
effectively. Confusion seems to exist on what triangulation involves in
practice. Silverman (2004a, 2004b) suggests that too many researchers rely
23
on the existence per se of multiple sets of data or researchers to represent
triangulation. Although many authors report the use of multiple data collection
methods (see Table 3) that usually involve authors drawing on primary and
secondary data in practice these authors often used different sources of data
to “flesh out” details of a case. Likewise, in relation to the uses of multiple
methods it was rare for authors to discuss how much information was
accessed, what it covered, how it related to other data from other methods,
and how this information was recorded and coded. As well, authors rarely
identified how multiple coders operated or dealt with disagreements (or
whether disagreements existed). Finally, triangulation can also be achieved
via informant feedback. Yet, evidence for member checking among reported
cases in IMM is rare and, when done, the results of such a process are not
reported.
Passages 11 and 12 come from articles typical of those giving some explicit
consideration of quality—the authors may even state they engaged in
triangulation although it is not clear from the reported findings how they did.
Passage 11 explicitly tells the reader that validity has been addressed
(although no raw data or rich quotes are provided). As well, although the
authors note triangulation was achieved through informant checking, there is
little evidence of using multiple data sources in the article. Passage 12 suffers
from similar problems. This practice (using multiple datasets) can lead to
researchers avoiding recognizing the limitations in one dataset, and move
quickly to another in order to cover up gaps in the initial data collection
(Silverman, 2004b). Although there may be positive reasons for gathering
multiple forms of data, triangulation in Silverman‟s view represents a way of
conducting a research project rather than just a series of post-hoc checks. For
example, B2B case researchers sometimes specify that secondary data were
used to enhance information obtained in interviews as a means of dealing
with validity. Yet, this does not in itself represent triangulation (although it may
enhance the comprehensiveness of data collected and ensure against recall
bias; Yin, 1994).
Triangulation involves trying to get a fix on what actually happened
(Silverman, 2004b, p. 177). For example, in their study of implementing
market orientation in business markets, Beverland and Lindgreen (2007)
suggested that the tone of internal communications changed through three
stages of implementation. To demonstrate this, the authors sought evidence
from marketing managers, internal stakeholders, as well as viewed examples
of speeches made at previous industry conferences, and internal marketing
documents to identify the changing nature of “tone” over time. Such a practice
went beyond drawing on multiple sources to find new information to identifying
a broad range of examples and experiences associated with the tone of
communication, thereby reinforcing the authors‟ views that the tone of internal
communications was reflective of different levels of cultural acceptance of
market orientation.
We suggest authors of cases in B2B marketing need to follow Silverman‟s
(2004a) advice that triangulation is a mode of inquiry, and not simply an
afterthought (in the same way that quantitative researchers design their
24
studies with quality considerations in mind from the beginning). As well,
authors should specify in detail how much data they gained, the sources and
number of each, and then draw on these multiple sources through their
findings section to demonstrate triangulation in practice, allowing readers to
make judgments on quality.
CONCLUSIONS
The findings have a number of implications for B2B marketing case
researchers, reviewers, and for the IMM review process. First, the three
themes presented above indicate that case researchers need to pay far
greater attention to issues of method, research design, and quality (especially
given the importance attributed to case research in IMM, and the greater
incidences of published cases since 2000). Specifically, researchers need to
be far more explicit about their bias (paradigmatic choices), sampling
decisions (believed to be the most critical issue for case research; Dubois and
Araujo 2004, 2007), technical method details, and quality procedures. As well
as giving explicit attention to these issues, researchers need to be more
aware of how quality is demonstrated in the writing-up of findings. Specifically,
readers need to get some first-hand access to raw data, need to see
triangulation being used, and negative case examples being dealt with. These
issues need to be addressed regardless of the epistemological preferences of
authors (although the execution of quality criteria will differ).
Second, reviewers need to become more aware of issues of research quality,
be more demanding of authors, and show greater sensitivity in reviewing
cases. In regard to issues of quality, reviewers need to have an understanding
of different case research paradigms and the implications of each for how
authors address research quality. In this regard, the review form for IMM
should be adjusted to allow reviewers to give ratings on these issues for
(qualitative) case research (currently, this feedback is only given to
quantitative articles). As well, this requires reviewers to hold authors to
account in one sense, but also to be more sensitive to different approaches.
For example, more positivistic-minded reviewers may prefer multiple case
studies to a rich single case. However, multiple case studies are not the gold
standard of case research and give rise to just as many problems (if not more)
than rich single cases. Likewise, Silverman (2004a) notes that interview data
is often treated as the “gold standard” in case research. Yet, good cases can
be developed from different sources of data. As well, triangulation requires the
use and presentation of multiple sources. Again, reviewers need to demand to
see multiple sources of evidence if they were gained and be prepared to
accept them—both authors of this article once got a review claiming that
secondary data (used in the context of rich primary data) was second-rate
data!
Third, editors and reviewers need to be sensitive to issues of article length
(subject to concerns over contribution-length ratio). Until greater attention is
paid by all case researchers to issues of method and quality (i.e., at which
point widespread understanding of different standards will allow researchers
25
to give less explicit attention to these issues in articles), authors will need
more space to attend to the issues outlined here.
Given concerns raised over perceptions of forcing or anecdotalism we
suggest two important changes to how articles are structured. First, the great
value of any case study lies in the rich context and findings (Easton, 2000)—
or the story—and thus the majority of the article should be devoted to this
information. Thus, as long as authors clearly ground their choice of method in
an extant literature, reviewers should demand authors of case study give the
greatest weight to their findings, discussion, and method (in that order), and
downplay the role of a lengthy literature review.
Second, questions must be raised about whether there is an effective upper
sample size for multiple cases. The analysis above identified that authors
struggle to retain the context of each individual case when presenting multiple
case studies. Importantly, attempts to present such information can decrease
readability, raise a range of issues around quality (forcing and anecdotalism),
and, once a certain sample size is reached, read more like a quantitative
paper (see the article from which passage 7 is taken for an example). There
are many benefits to be had from multiple case studies, but at a certain
sample size, readers are entitled to ask “why wasn‟t a survey used” (cf.
Silverman, 2004a), given that for large sample sizes readers either need to
trust authors that themes were consistent across all cases (often unlikely), or
authors need to reduce information about each case into a bit-sized summary
form in order to reduce word count—at which point the benefits from case
research are being lost.
The trend analysis in this paper can be read as either a cause for despair or a
cause for optimism. We prefer the later, and hope that the paper, as well as
others within this issue, stimulates established and emerging case
researchers to give greater attention to the issues raised here. In this way,
rich qualitative cases will continue to remain a valuable tool for B2B marketing
researchers.
26
REFERENCES
Alam, I. (2006). Removing the fuzziness from the fuzzy front-end of service
innovations through customer interactions. Industrial Marketing Management,
35(4), 468-480.
Alvesson, M. and Sköldberg, K. (2000). Reflexive methodology: New vistas for
qualitative research. London: Sage.
Baumgartner, H. and Pieters, R. (2003). The structural influence of marketing
journals: A citation analysis of the discipline and its subareas over time. Journal
of Marketing, 67(2), 123-139.
Beverland, M. B. (2005). Adapting within relationships to adapt to market-led change:
Does relationship success lead to marketplace inertia? Industrial Marketing
Management, 34(6), 577-589.
Beverland, M. B., Farrelly, F. J. and Woodhatch, Z. (2007). Exploring the dimensions
of proactivity within advertising-client relationships. Journal of Advertising,
36(4), 49-60.
Beverland, M. B. and Lindgreen, A. (2007). Implementing market orientation in
industrial firms: A multiple case study. Industrial Marketing Management, 36(4),
430-442.
Beverland, M. B. and Lockshin, L. (2003). A longitudinal study of customers‟ desired
value change in business-to-business markets. Industrial Marketing
Management, 32(8), 653-666.
Brennan, R. and Turnbull, P. W. (1999). Adaptive behavior in buyer-supplier
relationships. Industrial Marketing Management, 28(5), 481-495.
Carson, D., Gilmore, A., Perry, C. and Gronhaug, K. (2001). Qualitative Marketing
Research. London: Sage.
Cowell, D. W. and Blois, K. J. (1977). Conducting market research for high
technology products. Industrial Marketing Management, 6(5), 329-336.
Creswell, J. W. (1998). Qualitative inquiry and research design: Choosing among five
traditions. Thousand Oaks (CA): Sage.
Dubois, A. and Araujo, L. (2004). Research methods in industrial marketing studies.
In Håkansson, H., Harrison, D. and Waluszewski, A. (Eds.). Rethinking
Marketing: Developing a new understanding of markets. Chichester: Wiley.
Dubois, A. and Araujo, L. (2007). Case research in purchasing and supply
management: Opportunities and challenges. Journal of Purchasing & Supply
Management, 13(3), 170-181.
Dyer, W. G. and Watkins, A. L. (1991). Better stories, not better constructs, to
generate better theory: A rejoinder to Eisenhardt. Academy of Management
Review, 16(3), 613-619.
Easton, G. (2000). Case research as a method for industrial networks: A realist
apologia. In Fleetwood, S. and Ackroyd, S. (Eds.). Realist Perspectives in
Management and Organizations. London: Routledge.
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 rigour
and comparative logic. Academy of Management Review, 16(3), 620-627.
Flint, D. J., Woodruff, R. B. and Gardial, S. F. (2002). Exploring the phenomenon of
customers‟ desired value change in a business-to-business context. Journal of
Marketing 66(4), 102-117.
Ford, D. and McDowell, R. (1999). Managing business relationships y analyzing the
effects and value of different actions. Industrial Marketing Management, 28(5),
429-442.
Ghauri, P. N. (1988). Negotiating with firms in developing countries: Two case
studies. Industrial Marketing Management, 17(1), 49-53.
27
Glaser, B. G. and Strauss, A. L. (1967). The Discovery of Grounded Theory. New
York (NY): Aldine de Gruyter.
Håkansson, H., Havila, V. and Pedersen, A.-C. (1999). Learning in networks.
Industrial Marketing Management, 28(5), 443-452.
Harker, M. (1998). The role of marketing in the company turnaround process.
Industrial Marketing Management, 27(4), 315-327.
Harrison, D. and Easton, G. (2004). Temporally embedded case comparison in
industrial marketing research. In Fleetwood, S. and Ackroyd, S. (Eds.). Critical
Realist Applications in Organisation and Management Studies. London:
Routledge.
Hillebrand, B., Kok, R. A. and Biemans, W. G. (2001). Theory-testing using case
studies: A comment on Johnston, Leach, and Liu. Industrial Marketing
Management, 30(8), 651-657.
Holden, A. C. (1991). How to locate and communicate with overseas customers.
Industrial Marketing Management, 20(3), 161-168.
Johnsen, R. E. and Ford, D. (2006). Interaction capability development of smaller
suppliers in relationships with large customers. Industrial Marketing
Management, 35(8), 1002-1015.
Kirk, J. and Miller, M. L. (1986). Reliability and Validity in Qualitative Research.
London: Sage.
Kratschmar, T. (1972). Marketing research and the “technological breakthrough”:
Some case studies. Industrial Marketing Management, 1(3), 269-277.
Levy, S. J. (2005). The evolution of qualitative research in consumer behavior.
Journal of Business Research, 58(3), 341-347.
Lincoln, Y. S. and Guba, E. (1985). Naturalistic Inquiry. Beverly Hills (CA): Sage.
Lindgreen, A. (2008). Managing Relationship Marketing: Methodological and
empirical insights. Aldershot: Gower Publishing.
Loeser, B. O. (1999). How to set up a cooperation network in the production industry:
Example of the Huber + Suhner AG. Industrial Marketing Management, 28(5),
453-465.
McCracken, G. (1988). The Long Interview. Newbury Park (CA): Sage.
Mason, K., Doyle, P. and Wong, V. (2006). Market orientation and quasi-integration:
Adding value through relationships. Industrial Marketing Management, 35(2),
140-155.
Matthyssens, P. and Faes, W. (1985). OFM buying process for new components:
Purchasing and marketing implications. Industrial Marketing Management,
14(3), 145-157.
Möller, K. and Rajala, A. (1999). Organizing marketing in industrial high-tech firms:
The role of internal marketing relationships. Industrial Marketing Management,
28(5), 521-535.
Mühlbacher, H., Dreher, A. and Gabriel-Ritter, A. (1994). MIPS-managing industrial
positioning strategies. Industrial Marketing Management, 23(4), 287-297.
Murphy, E., Dingwall, R., Greatbatch, D., Parker, S. and Watson, P. (1998).
Qualitative research methods in health technology assessment: A review of the
literature. Health Technology Assessment, 2(16), 1-278.
Pardo, C., Salle, R. and Spencer, R. (1995). The key accountization process of the
firm: A case study. Industrial Marketing Management, 22(2), 123-134.
Perry, C. (1998). Processes of a case study methodology for postgraduate research.
European Journal of Marketing, 32(9/10), 785-802.
Price, L. and Arnould, E. J. (1999). Commercial friendships: Service provider-client
relationships in context. Journal of Marketing, 63(4), 38-56.
Salonen, A., Gabrielson, M. and Al-Obadi, Z. (2006). Systems sales as a competitive
response to the Asian challenge: Case of a global ship power supplier.
Industrial Marketing Management, 35(6), 740-750.
28
Sarin, S. and Kapur, M. G. (1990). Lessons from new product failures: Five case
studies. Industrial marketing Management, 19(4), 301-313.
Silverman, D. (2004a). Doing Qualitative Research: A practical handbook. London:
Sage.
Silverman, D. (2004b). Interpreting Qualitative Data: Methods for analysing talk, text
and interaction. London: Sage.
Stake, R. E. (2005). Qualitative case studies. In Denzin, N. K. and Lincoln, Y. S.
(Eds.). The Sage Handbook of Qualitative Research, 3rd ed. Thousand Oaks
(CA): Sage.
Strauss, A. and Corbin, J. (1998). Basics of Qualitative Research, 2nd ed. Thousand
Oaks (CA): Sage.
Van Maanen, J. (1983). Qualitative Methodology. Newbury Park (CA): Sage.
Wagner, S. M. and Hoegl, M. (2006). Involving suppliers in product development:
Insights from R&D directors and project managers. Industrial Marketing
Management, 35(8), 936-943.
Wilson, E. J. (1984). A case study of repeat buying for a commodity. Industrial
Marketing Management, 13(3), 195-200.
Windhahl, C. and Lakemond, N. (2006). Developing integrated solutions: The
importance of relationships within the network. Industrial Marketing
Management, 35(7), 806-818.
Yin, R. K. (1994). Case Study Research: Design and methods, 2nd ed. Thousand
Oaks (CA): Sage.
29
The Business School
Research Memorandum Series
Please note that MANY of these are available at:
http://www.hull.ac.uk/hubs/05/research/rm.htm
78
77
76
75
74
73
72
71
70
69
68
67
66
65
64
63
62
61
60
59
MB Beverland, A Lindgreen, M W Vink (2008) Projecting Authenticity through
Advertising: Consumer Judgments of Advertisers‟ Claims
X Bian, L Moutinho (2008) The Role of Product Involvement, Knowledge, and
Perceptions in Explaining Consumer Purchase Behaviour of Counterfeits:
Direct and Indirect Effects
A Lindgreen, R Palmer, M Wetzels, M Antioco (2008) Do Different Marketing
Practices Require Different Leadership Styles? An exploratory study
S J Armstrong (2007) Rethinking Management Education: From Cognition, To
Action, To Learning
J Simon (2007) Concept Maps and Accounting Curriculum Development
J Atkins, K Bhattarai and S Trotter (2007) Modelling the Impact of LowCarbon Electricity
T Campbell and S J Armstrong (2007) An Empirical Study of the Relationship
between Organisational Learning and Organisational Effectiveness using
Causal Cogitive Mapping
M Antioco, R K Moeneart and A Lindgreen (2007) Reducing On-going
Product Design Decision-Making Bias
F Maon, V Swaen, A Lindgreen (2007) Corporate Social Responsibility at
Ikea: commitment and communication
P Drake, S Clarke (2007) Embedding a Systems/Lifeworld Approach in
Information Security (Risk) Management and Broadening its Application
A Lindgreen, V Swaen, W S Johnston (2007) Corporate Social Responsibility:
an empirical investigation of US organisations
J Headlam-Wells, J Craig, J Gosland, J Holdsworth (2007) Mentoring for
Management Development: An evaluation of new communication models for
E-mentoring
A Dobele, A Lindgreen, M. Beverland, J Vanhamme, R van Wijk (2007) Why
Pass on Viral Messages? Because they connect emotionally
M Beverland, J Napoli, A Lindgreen (2007) Industrial Global Brand
Leadership: a capabilities view
K R Bhattarai (2007) Capital Accumulation, Growth and Redistribution:
General equilibrium impacts of energy and pollution taxes in the UK
A Rashid Malik (2007) The WTO Agreement on Textile and Clothing (ATC):
An Impact Analysis with Reference to Pakistan
L Han (2007) Signalling Process and Self-Selection Mechanism in
Entrepreneurial Finance
K R Bhattarai (2007) Is There Any Evidence on Unemployment-Inflation
Trade-off in OECD Countries in Recent Years?
J B Simon (2006) A Framework for Business Research: Using the Knowledge
Vee
G Mengesha and R Common (2006) Civil Service Reform in Ethiopia:
Success in Two Ministries
30
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
D Kumala (2006) The Achievements and Challenges of the New East African
Community Cooperation
T Butcher (2006) The Socio-technical Impact of RFID Technologies in Supply
Chain Management
J G Cegarro, J R Córdoba-Pachón (2006) Assessing and Developing egovernment use by SMEs
R Common (2006) Globalisation and the Governance of Hong Kong
C Hammond (2006) Technical Efficiency and Organisational Change in UK
Public Library Systems: A Stochastic Distance Function Approach
K R Bhattarai, E Okyere (2005) Welfare and Growth Impacts of Taxes:
Applied General Equilibrium Models of Ghana
K. R. Bhattarai, M K Armah (2005) The Effects of Exchange Rate on the
Trade Balance on Ghana: Evidence from Cointegration Analysis
A. Aritzeta, S. Swailes, B. Senior (2005) Team Roles: Psychometric
Evidence, Construct Validity and Team Building
G F Lanzara, M Morner (2005) Making and Sharing Knowledge at Electronic
Crossroads: The Evolutionary Ecology of Open Source
W Robson, J R Córdoba-Pachón (2005) What Research Methodology Suits
Collaborative Research
K R Bhattarai (2005) Economic Growth: Models and Global Evidence
P Mould, T Boczko (2004) Standard-setting and the Myth of Neutrality:
Boundaries, Discourse and the Exercise of Power
P Murray (2005) „Learning‟ with Complexity: Metaphors from the New
Sciences
D Kamala (2004) The Voices of the Poor and Poverty Eradication Strategies
in Tanzania
M R Ryan (2003) Vertical Integration Results and Applications to the
Regulation of Supermarket Activity
M A Nolan, F R Fitzroy (2003) Inactivity, Sickness and Unemployment in
Great Britain: Early Analysis at the Level of Local Authorities
K R Bhattarai (2003) An Analysis of Interest Rate Determination in the UK
and Four Major Leading Economies
K R Bhattarai (2004) Macroeconomic Impacts of Consumption and Income
Taxes: A General Equilibrium Analysis
J Paez-Farrell (2003) Business Cycles in the United Kingdom: An Update on
the Stylised Facts
J Paez-Farrell (2003) Endogenous Capital in New Keynesian Models
P Kitchen, L Eagle, L Rose, B Moyle (2003) The Impact of Gray Marketing
and Parallel Importing on Brand Equity and Brand Value
M Afanassieva (2003) Managerial Responses to Transition in the Russian
Defence Industry
J R Córdoba-Pachón, G Midgley (2003) Addressing Organisational and
Societal Concerns: An Application of Critical Systems Thinking to Information
Systems Planning in Colombia
J Evans, R Green (2003) Why Did British Electricity Prices Fall After 1998?
C Hemingway (2002) An Exploratory Analysis of Corporate Social
Responsibility: Definitions, Motives and Values
P Murray (2002) Constructing Futures in New Attractors
S Armstrong, C Allinson, J Hayes (2002) An Investigation of Cognitive Style
as a Determinant of Successful Mentor-Protégé Relationships in Formal
Mentoring Systems
31
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
N O'Neill (2001) Education and the Local Labour Market
A Zakaria, Z Osemy, B Prodhan (2001) The Role of Accounting Information
Systems in Rationalising Investment Decisions in Manufacturing Companies
in Egypt
A Mejía (2001) The Problem of Knowledge Imposition: Paulo Freire and
Critical Systems Thinking
P Maclagan (2001) Reflections on the Integration of Ethics Teaching into the
BA Management Degree Programme at The University of Hull
N R Romm (2001) Considering Our Responsibilities as Systemic Thinkers: A
Trusting Constructivist Argument
L Hajek, J Hynek, V Janecek, F Lefley, F Wharton (2001) Manufacturing
Investment in the Czech Republic: An International Comparison
Gerald M, J F Gu, D Campbell (2001) Dealing with Human Relations in
Chinese Systems Practice
Z Zhu (2000) Dealing with a Differentiated Whole: The Philosophy of the
WSR Approach
W J Gregory, G Midgley (1999) Planning for Disaster: Developing a MultiAgency Counselling Service
L Pinzón, G Midgley (1999) Developing a Systemic Model for the Evaluation
of Conflicts
G Midgley, A E Ochoa Arias (1999) Unfolding a Theory of Systemic
Intervention
M Brown, R Packham (1999) Organisational Learning, Critical Systems
Thinking and Systemic Learning
G Midgley (1999) Rethinking the Unity of Science
P Keys (1998) Creativity, Design and Style in MS/OR
G Midgley, A E Ochoa Arias (1998) Visions of Community for Community OR
G Midgley, I Munlo, M Brown (1998) The Theory and Practice of Boundary
Critique: Developing Housing Services for Older People
J Clayton, W Gregory (1997) Total Systems Intervention or Total Systems
Failure: Reflections of an Application of TSI in a Prison
L Pinzon, N Valero-Silva (1996) A Proposal for a Critique of Contemporary
Mediation Techniques - The Satisfaction Story
R L. Flood (1996) Total Systems Intervention: Local Systemic Intervention
R L. Flood (1996) Holism and Social Action „Problem Solving‟
P Dudley, S Pustylnik (1996) Modern Systems Science: Variations on the
Theme?
W Ulrich (1996) Critical Systems Thinking for Citizens: A Research Proposal
G Midgley (1995) Mixing Methods: Developing Systems Intervention
J Thompson (1995) User Involvement in Mental Health Services: The Limits
of Consumerism, the Risks of Marginalisation and the Need for a Critical
Systems Approach
P Dudley, S Pustylnik (1995) Reading the Tektology: Provisional Findings,
Postulates and Research Directions
W Gregory, N R Romm (1994) Developing Multi-Agency Dialogue: The
Role(s) of Facilitation
N R Romm (1994) Continuing Tensions between Soft Systems Methodology
and Critical Systems Heuristics
P Dudley (1994) Neon God: Systems Thinking and Signification
M C Jackson (1993) Beyond the Fads: Systems Thinking for Managers
32
2
1
S M Green (1993) Systems Thinking and the Management of a Public Service
Organisation
M C Jackson and R L Flood (1993) Opening of the Centre for Systems
Studies
33
Centre for Economic Policy Research Papers
(prior to merging with the Business School in August 2002)
283 Naveed Naqvi, Tapan Biswas and Christer Ljungwall (2002). Evolution of
Wages and Technological Progress in China‟s Industrial Sector
282 Tapan Biswas, Jolian P McHardy (2002). Productivity Changes with Monopoly
Trade Unions in a Duopoly Market
281 Christopher Tsoukis (2001). Productivity Externalities Tobin‟s Q and
Endogenous Growth
280 Christopher J Hammond, Geraint Johnes and Terry Robinson (2000). Technical
Efficiency under Alternative Regulatory Regimes: Evidence from the Inter-War
British Gas Industry
279 Christopher J Hammond (2000). Efficiency in the Provision of Public Services:
A Data Envelopment Analysis of UK Public Library Systems
278 Keshab Bhattarai (2000). Efficiency and Factor Reallocation Effects and
Marginal Excess Burden of Taxes in the UK Economy
277 Keshab Bhattarai, Tomasz Wisniewski (2000). Determinants of Wages and
Labour Supply in the UK
276 Taradas Bandyopadhay, Tapan Biswas (2000). The Relation between
Commodity Prices and Factor Rewards
275 Emmanuel V Pikoulakis (2000). A “Market Clearing” Model of the International
Business Cycle that Explains the 1980‟s
274 Jolian P McHardy (2000). Miscalculations of Monopoly and Oligopoly Welfare
Losses with Linear Demand
273 Jolian P McHardy (2000). The Importance of Demand Complementarities in the
Calculation of Dead-Weight Welfare Losses
272 Jolian P McHardy (2000). Complementary Monopoly and Welfare Loss
271 Christopher Tsoukis, Ahmed Alyousha (2000). A Re-Examination of Saving –
Investment Relationships: Cointegration, Causality and International Capital
Mobility
270 Christopher Tsoukis, Nigel Miller (2000). A Dynamic Analysis of Endogenous
Growth with Public Services
269 Keshab Bhattarai (1999). A Forward-Looking Dynamic Multisectoral GeneralEquilibrium Tax Model of the UK Economy
268 Peter Dawson, Stephen Dobson and Bill Gerrad (1999). Managerial Efficiency
in English Soccer: A Comparison of Stochastic Frontier Methods
267 Iona E Tarrant (1999). An Analysis of J S Mill‟s Notion of Utility as a
Hierarchical Utility Framework and the Implications for the Paretian Liberal
Paradox
266 Simon Vicary (1999). Public Good Provision with an Individual Cost of
Donations
265 Nigel Miller, Chris Tsoukis (1999). On the Optimality of Public Capital for LongRun Economic Growth: Evidence from Panel Data
264 Michael J Ryan (1999). Data Envelopment Analysis, Cost Efficiency and
Performance Targeting
263 Michael J Ryan (1999). Missing Factors, Managerial Effort and the Allocation of
Common
34