10.1177/1077800405284363 Qualitative Inquiry Flyvbjerg / Case-Study Research Misunderstandings Five Misunderstandings About Case-Study Research Bent Flyvbjerg Aalborg University, Denmark This article examines five common misunderstandings about case-study research: (a) theoretical knowledge is more valuable than practical knowledge (b) one cannot generalize from a single case, therefore, the single-case study cannot contribute to scientific development (c) the case study is most useful for generating hypotheses, whereas other methods are more suitable for hypotheses testing and theory building (d) the case study contains a bias toward verification and (e) it is often difficult to summarize specific case stud- ies. This article explains and corrects these misunderstandings one by one and concludes with the Kuhnian insight that a scientific discipline without a large number of thoroughly executed case studies is a discipline without systematic production of exemplars, and a discipline without exemplars is an ineffective one. Social science may be strengthened by the execution of a greater number of good case studies. Keywords: case study case selection critical cases validity in case studies Wtrying hen I first became interested in in-depth case-study research, I was to understand how power and rationality shape each other and form the urban environments in which we live (Flyvbjerg, 1998). It was clear to me that to understand a complex issue such as this, in-depth case-study research was necessary. It was equally clear, however, that my teachers and colleagues kept dissuading me from employing this particular research methodology. ���You cannot generalize from a single case,��� some would say, ���and social science is about generalizing.��� Others would argue that the case study may be well suited for pilot studies but not for full-fledged research schemes. Others again would comment that the case study is subjective, giving too much scope for the researcher���s own interpretations. Thus, the validity of case studies would be wanting, they argued. At first, I did not know how to respond to such claims, which clearly formed the conventional wisdom about case-study research. I decided, there- 219 Qualitative Inquiry Volume 12 Number 2 April 2006 219-245 �� 2006 Sage Publications 10.1177/1077800405284363 http://qix.sagepub.com hosted at http://online.sagepub.com

fore, to find out where the claims come from and whether they are correct. This article contains what I discovered. The Conventional Wisdom About Case-Study Research Looking up case study in the Dictionary of Sociology as a beginning, I found the following in full citation: Case Study. The detailed examination of a single example of a class of phe- nomena, a case study cannot provide reliable information about the broader class, but it may be useful in the preliminary stages of an investigation since it provides hypotheses, which may be tested systematically with a larger number of cases. (Abercrombie, Hill, & Turner, 1984, p. 34)1 This description is indicative of the conventional wisdom of case-study research, which if not directly wrong, is so oversimplified as to be grossly misleading. It is correct that the case study is a ���detailed examination of a sin- gle example,��� but as we see below, it is not true that a case study ���cannot pro- vide reliable information about the broader class.��� It is also correct that a case study can be used ���in the preliminary stages of an investigation��� to generate hypotheses, but it is misleading to see the case study as a pilot method to be used only in preparing the real study���s larger surveys, systematic hypotheses testing, and theory building. According to the conventional view, a case and a case study cannot be of value in and of themselves they need to be linked to hypotheses, following the well-known hypothetico-deductive model of explanation. Mattei Dogan and Dominique Pelassy (1990) put it like this: ���One can validly explain a par- ticular case only on the basis of general hypotheses. All the rest is uncontrol- lable, and so of no use��� (p. 121 see also Diamond, 1996, p. 6). In a similar manner, the early Donald Campbell did not mince words when he relegated single-case studies to the methodological trash heap: Such studies have such a total absence of control as to be of almost no scientific value. ...Any appearance of absolute knowledge, or intrinsic knowledge about singular isolated objects, is found to be illusory upon analysis. . . . It seems well-nigh unethical at the present time to allow, as theses or dissertations in education, case studies of this nature (i.e., involving a single group observed at one time only). (Campbell & Stanley, 1966, pp. 6-7) If you read such criticism of a certain methodology enough times, or if you hear your thesis advisers repeat it, you begin to believe it may be true. This is 220 Qualitative Inquiry

what happened to me, and it made me uncertain about case-study methodol- ogy. As I continued my research, however, I found out that Campbell had later made a 180-degree turn in his views of the case study and had become one of the strongest proponents of this method. I eventually found, with the help of Campbell���s later works (e.g., Campbell, 1975) and other works like them, that the problems with the conventional wisdom about case-study research can be summarized in five misunderstandings or oversimplifica- tions about the nature of such research: Misunderstanding 1: General, theoretical (context-independent) knowledge is more valuable than concrete, practical (context-dependent) knowledge. Misunderstanding 2: One cannot generalize on the basis of an individual case therefore, the case study cannot contribute to scientific development. Misunderstanding 3: The case study is most useful for generating hypotheses that is, in the first stage of a total research process, whereas other methods are more suitable for hypotheses testing and theory building. Misunderstanding 4: The case study contains a bias toward verification, that is, a tendency to confirm the researcher���s preconceived notions. Misunderstanding 5: It is often difficult to summarize and develop general propo- sitions and theories on the basis of specific case studies. These five misunderstandings indicate that it is theory, reliability, and valid- ity that are at issue in other words, the very status of the case study as a scien- tific method. In what follows, I focus on these five misunderstandings and correct them one by one. First, however, I outline the role of cases in human learning. The Role of Cases in Human Learning To understand why the conventional view of case-study research is prob- lematic, we need to grasp the role of cases and theory in human learning. Here two points can be made. First, the case study produces the type of context- dependent knowledge that research on learning shows to be necessary to allow people to develop from rule-based beginners to virtuoso experts. Sec- ond, in the study of human affairs, there appears to exist only context- dependent knowledge, which, thus, presently rules out the possibility of epistemic theoretical construction. The full argument behind these two points can be found in Flyvbjerg (2001, chaps. 2-4). For reasons of space, I can give only an outline of the argument here. At the outset, however, we can assert that if the two points are correct, it will have radical consequences for Flyvbjerg / Case-Study Research Misunderstandings 221

the conventional view of the case study in research and teaching. This view would then be problematic. Phenomenological studies of human learning indicate that for adults, there exists a qualitative leap in their learning process from the rule-governed use of analytical rationality in beginners to the fluid performance of tacit skills in what Pierre Bourdieu (1977) called virtuosos and Hubert Dreyfus and Stuart Dreyfus (1986) called true human experts. Here we may note that most people are experts in a number of everyday social, technical, and intel- lectual skills such as giving a gift, riding a bicycle, or interpreting images on a television screen, whereas only few reach the level of true expertise for more specialized skills such as playing chess, composing a symphony, or fly- ing a fighter jet. Common to all experts, however, is that they operate on the basis of inti- mate knowledge of several thousand concrete cases in their areas of exper- tise. Context-dependent knowledge and experience are at the very heart of expert activity. Such knowledge and expertise also lie at the center of the case study as a research and teaching method or to put it more generally still, as a method of learning. Phenomenological studies of the learning process there- fore emphasize the importance of this and similar methods: It is only because of experience with cases that one can at all move from being a beginner to being an expert. If people were exclusively trained in context-independent knowledge and rules, that is, the kind of knowledge that forms the basis of textbooks and computers, they would remain at the beginner���s level in the learning process. This is the limitation of analytical rationality: It is inade- quate for the best results in the exercise of a profession, as student, researcher, or practitioner. In a teaching situation, well-chosen case studies can help the student achieve competence, whereas context-independent facts and rules will bring the student just to the beginner���s level. Only few institutions of higher learn- ing have taken the consequence of this. Harvard University is one of them. Here both teaching and research in the professional schools are modeled to a wide extent on the understanding that case knowledge is central to human learning (Christensen, 1987 Cragg, 1940). At one stage in my research, I was invited to Harvard to learn about case methodology ���in action.��� During my stay, it became clear to me that if I were going to aspire at becoming an expert in my field of expertise, and if I wanted to be an effective help to my students in their learning processes, I would need to master case methodology in research and teaching. My stay at Har- vard also became a major step forward in shedding my uncertainties regard- ing the conventional wisdom about cases and case studies. At Harvard, I found the literature and people who effectively argued, ���Forget the conven- 222 Qualitative Inquiry

tional wisdom, go ahead and do a case study.��� I figured if it is good enough for Harvard, it is good enough for me, and I suggest others might reason like this, including whole institutions of learning. There is much to gain, for instance, by transforming the lecture format still dominant in most universi- ties to one of case learning (Christensen, 1987). It is not that rule-based knowledge should be discounted: It is important in every area and especially to novices. But to make rule-based knowledge the highest goal of learning is regressive. There is a need for both approaches. The highest levels in the learning process, that is, virtuosity and true exper- tise, are reached only via a person���s own experiences as practitioner of the relevant skills. Therefore, beyond using the case method and other experien- tial methods for teaching, the best that teachers can do for students in profes- sional programs is to help them achieve real practical experience for exam- ple, via placement arrangements, internships, summer jobs, and the like. For researchers, the closeness of the case study to real-life situations and its multiple wealth of details are important in two respects. First, it is impor- tant for the development of a nuanced view of reality, including the view that human behavior cannot be meaningfully understood as simply the rule- governed acts found at the lowest levels of the learning process and in much theory. Second, cases are important for researchers��� own learning processes in developing the skills needed to do good research. If researchers wish to develop their own skills to a high level, then concrete, context-dependent experience is just as central for them as to professionals learning any other specific skills. Concrete experiences can be achieved via continued proxim- ity to the studied reality and via feedback from those under study. Great dis- tance to the object of study and lack of feedback easily lead to a stultified learning process, which in research can lead to ritual academic blind alleys, where the effect and usefulness of research becomes unclear and untested. As a research method, the case study can be an effective remedy against this tendency. The second main point in connection with the learning process is that there does not and probably cannot exist predictive theory in social science. Social science has not succeeded in producing general, context-independent theory and, thus, has in the final instance nothing else to offer than concrete, context-dependent knowledge. And the case study is especially well suited to produce this knowledge. In his later work, Campbell (1975) arrived at a simi- lar conclusion, explaining how his work had undergone ���an extreme oscilla- tion away from my earlier dogmatic disparagement of case studies��� (p. 179), which was described above. In logic that in many ways resembles that of the phenomenology of human learning, Campbell now explained, Flyvbjerg / Case-Study Research Misunderstandings 223