Method Applied – Current Practice

To answer the last research question empirical data on current practices is needed – and to that end a research strategy must be carefully chosen.

Choosing a Research Strategy

One of the characteristics of the research situation determining the appropriateness of a research strategy is the existing body of knowledge (Yin, 1994; Ghauri, Grønhaug, &

Kristianslund, 1995 etc.). As pointed out several times already, a certain degree of skepticism is felt towards the current knowledge of SCM. Coupling this uncertainty with the lack of knowledge of the practices on SCRM, the research design for the empirical part has to enable an understanding of the phenomenon, instead of testing existing hypotheses.

In Ghauri, Grønhaug, & Kristianslund (1995) it is emphasized that the understanding of the problem is a classifying variable:

“Based on the problem structure, we may distinguish between three main classes of research design:

Research design Problem structure

1. Exploratory Unstructured

2. Descriptive Structured

3. Causal Structured” (p. 27).

According to Ghauri, Grønhaug, & Kristianslund an exploratory research design is appropriate when the research problem is badly understood, whereas both descriptive and causal research design are appropriate when the problem is structured and well understood.

They consider the case study method of special relevance when the area of interest is poorly understood:

“In relatively less-known areas, where there is little experience and theory available to serve as a guide, intensive study of selected examples is a very useful method of gaining insight and suggesting hypotheses for further research. … The main focus is on seeking insight rather than testing: instead of testing existing hypotheses we seek insight through the features and characteristics of the object being studied.” (pp. 87-88).

This position is challenged by Yin (1994) who claim case studies can be exploratory as well as exploratory/descriptive:

“A common misconception is that the various research strategies should be arrayed hierarchically. We were once taught to believe that case studies were appropriate for the exploratory phase of an investigation, that surveys and histories were appropriate for the descriptive phase, and that experiments were the only way of doing explanatory or causal inquiries. … This hierarchical view, however, is incorrect. … The more appropriate view of these different strategies is a pluralistic one. Each strategy can be used for all three purposes – exploratory, descriptive, or explanatory.” (pp. 3-4).

To Yin the case study research strategy is well suited in a very specific situation:

“In general, case studies are the preferred strategy when ‘how’ or ‘why’

questions are being posed, when the investigator has little control over events, and when the focus is on a contemporary phenomenon within some real-life context. Such ‘explanatory’ case studies also can be complemented by two other types – ‘exploratory’ and ‘descriptive’ case studies.” (p. 1).

The case study research strategy thereby fits this specific combination of research question, control over events, and “contemporariness” of events to be studies. ‘How’ and ‘when’

questions can also be studied by applying research strategies ‘experiments’ or ‘history’, but the former requires control over events, the latter focuses on non-contemporary events²⁰. Albeit Yin does not agree with Eisenhardt (1989b) (and others) that case study research is especially well-suited when performing research of more exploratory nature, he does not disqualify the research strategy neither.

But others have criticized the research strategy. In a special issue of the journal Administrative Science Quarterly on qualitative methods, Miles (1979) criticizes the case study method (or rather qualitative methods) for producing results that can not be analyzed, neither with-in case nor cross-case. Furthermore Miles points out that the frequency of objections from participants to research results far exceeds the frequency from other types of studies. In a reply published two years later in the same journal, Yin (1981) reciprocates by questioning the link between qualitative data and case study research (case study research can

20 Summary in Figure 1.1 in Yin (1994), p. 6.

use both qualitative and quantitative data) and by suggesting methods of performing with-in case and cross-case analysis.

A Model for Theory-building in Case Study Research

This line of development was continued in Eisenhardt (1989b) where a method for theory development from case study research is presented. Referencing other researchers performing case study research it is claimed the a priori definition of research question and possibly even constructs is beneficial:

“… definition of research question within a broad topic permitted these investigators to specify the kind of organization to be approached, and, once there, the kind of data to be collected. … A priori specification of constructs can also help shape the initial design of theory-building research.”

(Eisenhardt, 1989b, p. 536).

Knowing state-of-art of the subject through stringent literature review is an important requirement for defining the initial research question and preliminary constructs, but it is emphasized the researcher should be willing to alter these as appropriate during the study:

“Although early identification of the research question and possible constructs is helpful, it is equally important to recognize that both are tentative in this type of research. No construct is guaranteed a place in the resultant theory, no matter how well it is measured. Also, the research question may shift during the research.” (p. 536).

Differentiating the knowledge available in the literature from theoretical perspectives it is also emphasized that theory should delineate neither research question nor constructs, but:

“… theory building research is begun as close as possible to the ideal of no theory under consideration and no hypotheses to test. … [A]ttempting to approach this is important because preordained theoretical perspectives or propositions may bias and limit the findings. Thus, investigators should formulate a research problem and possibly specify some potentially important variables, with some reference to extant literature. However, they should avoid thinking about specific relationships between variables and theories as much as possible, especially at the outset of the process.” (p. 536).

Progressing to the case selection Eisenhardt further describes the difference in the sampling method for theory testing versus theory building studies:

“The cases may be chosen to replicate previous cases or extend emergent theory, or they may be chosen to fill theoretical categories and provide examples of polar types. While the cases may be chosen randomly, random selection is neither necessary, nor even preferable. As Pettigrew (1988) noted, given the limited number of cases which can usually be studied, it makes sense to choose cases such as extreme situations and polar types in which the process of interest is ‘transparently observable’. … In contrast, traditional within-experiment hypothesis-testing studies rely on statistical sampling, in which researchers randomly select the sample from the population. In this type of study, the goal of the sampling process is to obtain accurate statistical evidence on the distributions of variables within the population.” (p. 537).

The case selection method thereby determines the type of conclusion reached – the exploratory case study might naturally be limited to theory-generation if not supplemented by studies of a more explanatory nature.

But also the theory generation requires a stringent process for analysis of the often staggering volumes of data. Eisenhardt proposes to start with the with-in case analysis “to become intimately familiar with each case as a stand-alone entity” (p. 540) followed by a cross-case analysis to identify patterns of similarity. Various strategies for the cross-case analysis are proposed, all aiming at increasing the quality of the analysis and:

“… to force investigators to go beyond initial impressions, especially through the use of structured and diverse lenses on the data. These tactics improve the likelihood of accurate and reliable theory, that is, a theory with a close fit with the data. Also, cross-case searching tactics enhance the probability that the investigators will capture the novel findings which may exist in the data.” (p.

541).

The fit between the data and the emerging theory improve as the iterative process of systematically comparing data with hypotheses progresses. Between iterations constructs are refined and measures are developed, and multiple sources of evidence might be introduced to further improve construct validity. In conclusion the strengths of the case study research strategy are:

¾ the likelihood of generating novel theory,

¾ the emergent theory is likely to be testable with constructs that can be readily measured and hypotheses that can be proven false, and

¾ the resultant theory is likely to be empirically valid.

Conversely, some of the characteristics which lead to the strengths also may lead to some of the weaknesses when building theory from case studies. These include:

¾ the intensive use of empirical evidence may lead to overly complex theory, as well as the opposite, namely

¾ the creation of idiosyncratic theory stemming from the bottom-up approach of case study theory creation.

These shortcomings obviously are quite severe.

An Ongoing Discussion

Two years later a critique by Dyer & Wilkins (1991) was published objecting to:

“…three critical areas: (a) the in-depth study of a single case (context) versus the study of multiple cases (contexts), (b) deep versus surface description, and (c) the telling of good stories versus the creating of good constructs.” (p. 613).

The basis for their criticism was a defense for the “classic case study” where a single (or possibly two or three) cases are included in the study. They argue that the method put forward by Eisenhardt is cases study instead of case study and go on to argue that the “rich story” is an important exception to the argument made by Eisenhardt:

“With fewer than 4 cases, it is often difficult to generate theory with much complexity, and its empirical grounding is likely to be unconvincing, unless the case has several mini-cases within it.” (Eisenhardt, 1989b, p. 545).

To Eisenhardt knowledge is obtained and verified through cross-case examination whereas Dyer & Wilkins argue getting close to the case is a valid source of knowledge, resulting in

“deep insights”. They argue that the ultimate goal is to “provide a rich description of the social scene, to describe the context in which events occur, and to reveal … the deep structure of social behaviour.” (p. 615). Dyer & Wilkins further claim the main advantage of the classic case study is the “quality” of the theory generated:

“Theory that is born of such deep insights will be both more accurate and more appropriately tentative because the researcher must take into account the intricacies and qualifications of a particular context. … Eisenhardt’s method [is] necessarily constrained by the number of cases that will be studied, and description will be rather ‘thin’, focusing on surface data rather on deeper dynamics.” (p. 615).

Dyer & Wilkins admit Eisenhardt’s method might provide “flashes of insight” but insist the method will:

“… neglect the more tacit and less obvious aspects of the setting under investigation. [It] is more likely to provide a rather distorted picture or no picture at all, of the underlying dynamics of the case.” (p. 615).

Dyer & Wilkins end their criticism by insisting on the advantages of the classic case study (the rich story of a single or a few cases) and argue:

“… that the classic case study approach has been extremely powerful because these authors have described general phenomena so well that others have little difficulty seeing the same phenomena in their own experience and research.

We return to the classics because they are good stories, not because they are merely clear statements of a construct.” (p. 617).

In a response Eisenhardt (1991) first rejects the critiques put forward and continues to repeat the claim that rigor and stringent methods are necessary in order to create generalizable theory. In the concluding comments the relevance of good story telling is emphasized:

“… the similarities between single- and multiple-setting research are vastly more important then the differences. For both, storytelling is an essential first step, but good theory is fundamentally the result of rigorous methodology and comparative, multiple-case logic. This is as evident in the classic case studies as it is in contemporary multiple-case research.” (p. 627).

The authors mentioned above are but a few of the participants in the ongoing debate over the relevance and applications of case study research. Many other could have been mentioned but the examples described above clearly describe the intensity of the debate, and give some indications on the advancements within case study research.

Case Study Research for Exploratory Studies

The use of case study research is now accepted within Management/Organization Theory (e.g.

Woodside & Wilson, 2003), Operations Management (e.g. Meredith, 1998; Lewis, 1998),

Marketing (e.g. Johnston, Leach, & Liu, 1999; Hillebrand, Kok, & Biemans, 2001) and Logistics/SCM (e.g. Gammelgaard, 2004; Frankel, Naslund, & Bolumole, 2005) albeit debate on the application and contingencies surrounding case study research is still ongoing. In this dissertation the case study research strategy has been chosen due to its capability to unravel and make possible description of complex organizational phenomena of which little is known.

The detailed design of the case studies is described in Chapter 6.