restricted access Introduction
In lieu of an abstract, here is a brief excerpt of the content:

INTRODUCTION Human and social systems are complex. Understanding phenomena related to such systems demands a holistic approach, which can produce not only detailed descriptions of situations and events but also an in-depth understanding of the actors involved, their feelings and the interactions among them. Only qualitative research methods can provide a comprehensive view of this type (Benbasat, Goldstein and Mead 1983; Eisenhardt 1989; Patton 1982; Worthman and Roberts 1982). A case study, in particular, makes it possible to observe and analyze phenomena as a single, integrated whole (Bullock 1986). Quantitative methods, though useful, cannot do so, for their main tool, the questionnaire, is 2 The Case Study as Research Method based on respondents’ considered answers, and as we know 95% of human thoughts are unconscious (Fauconnier 1997; Schank 1998; Wegner 2002; Woodside and Wilson 2003; Zaltman 2003) and individuals have limited awareness of their own thought processes (Van Someren, Barnard and Sandberg 1994; Witte 1972; Woodside and Wilson 2000). Before going further, we should clarify what we mean by method. We will use Aktouf’s (1987: 20) definition of method as “the logical procedure employed by a science, i.e. the set of specific practices it uses to render the development of its demonstrations and theories clear, understandable and irrefutable.” As the case method provides an in-depth understanding of phenomena, their constitutive processes and the actors involved, some scholars believe it is best suited to theory building (Dyer and Wilkins 1991; Eisenhardt 1989; Gersick 1988; Harris and Sutton 1986; Woodside and Wilson 2003). But others, including some proponents of grounded theory, argue that it is equally appropriate for validating a theory (Anderson 1983; Eisenhardt 1989; Glaser and Strauss 1967; Pinfield 1986; Richards and Richards 1994; Strauss and Corbin 1990). Finally, some maintain that it is possible to develop a research design that combines theory building and verification. On this view, a case study can serve to generate a new theory, which can then be tested immediately using measurable constructs and falsifiable hypotheses (Eisenhardt 1989; Gladwin 1989; Howard and Morgenroth 1968; Woodside and Wilson 2003). More precisely, the case method is said to be appropriate for describing, explaining, predicting or controlling processes associated with a variety of phenomena at the individual, group and organizational levels (Woodside and Wilson 2003). The combination of these four functions is also possible. Describing means answering the questions who, what, when, how (Eisenhardt 1989; Kidder 1982); explaining means attempting to answer the question why; predicting means producing short-term and long-term forecasts of future psychological states, behaviours or events; and controlling means trying to influence cognition, attitudes and behaviours in an individual case (Hersen and Barlow 1976; Woodside and Wilson 2003). The main advantages of case research are that it can produce an in-depth analysis of phenomena in context, support the development of historical perspectives and guarantee high internal validity, Introduction 3 which is to say that the observed phenomena are authentic representations of reality. In short, the case study is adaptable to both the context and the researcher.1 But when using the case method for research purposes, we must always bear in mind that it also suffers from weaknesses. First, it is time-consuming for both the researcher and the subjects. Secondly , the external validity of the results is problematic, for it is difficult for another researcher to reproduce a case study. Finally, the case method has significant shortcomings when it comes to the generalizability of the results. There is little chance that comparable studies will be conducted to generalize the theory inferred from the case study or to make the results applicable to an entire population (Lecompte and Goetz 1982; Lucas 1974; McMillan and Schumacher 1984; Whyte 1963; Worthman and Roberts 1982). This is unsurprising, for the specificity , diversity and narrow focus of a case study are not readily compatible with attempts to achieve a universal scope. For one thing, the pursuit of generalizability can distract the researcher from specific features of the case at hand that could be important for a full understanding (Stake 1994). For another, excessive use of empirical data with a view to generalization will lead almost inevitably to an overly complex theory in which it is difficult to distinguish between general relationships and those that are particular to the specific case. It must therefore be accepted that the result of a case study is, more often than not, an idiosyncratic theory, which is to say that it applies to a particular phenomenon or a...