- Recognize the recent change in opinions about the logic behind case study (inductive versus abductive).
- Analyze the logical fallacy involved in the classic “Jonesville-is-the-USA” argument.
- Identify the reasons why a case study cannot be conflated with an “action research” study, despite common temptations to do so in study findings.
Case Study Logic: Induction or Abduction?
As the history in Module 2 shows, case-based logic has been closely associated with inductive reasoning dating back to antiquity. Both case reasoning and early scientific reasoning (ca. 19th century) shared the belief that observing particulars could lead to a collection of cases, and that a collection of cases – once analyzed – could produce inductive hypotheses about particulars, and ultimately a general theory. Recent philosophical treatment of this logic, however, casts a negative light on the relationship between case logic and induction, arguing instead that abduction plays the dominant role (Thomas & Myers, 2015; Thomas, 2010; Gomm, Hammersley, & Foster, 2000[b]). Abduction differs from induction in that it credits the thinker with an ability to imagine likely explanations based on limited observations (and hence knowledge of cases), and then weigh those hypotheses over the course of continued experiences:
Gary Thomas, for example, proposes that case logic is actually a special kind of abductive logic that unfolds for “practitioners” as they weave together practical examples of a certain problem, and attempt to arrive at a holistic narrative that summarizes their varied accounts of the problem. Thomas views case study findings as a kind of practical “example-based” knowledge that is similar to Michael Polanyi’s notion of “tacit knowledge” and “Donald Schön’s “practitioner knowledge,” in that it keeps its context dependency rather than becoming inductive generality, or so-called “theory” (Thomas, 2010, pp. 577-80).
Case Findings Fallacy (part 1): The Claim that “Jonesville-is-the-USA!”
Gomm, Hammersley, and Foster (2000[b], p. 99) point to one of the more egregious errors in writing-up case findings: the claim that one’s case is actually a microcosm of some larger life situation. The microcosm argument is a form of what literary theorists call “synechdoche”; the habit of using a part of something to stand for the whole. It usually tempts those researchers who are desperate to claim generalizable results and assure the relevance for their work. In rebuke of this habit, Gomm and company quote from Clifford Geertz, the famous anthropologist, who criticized his colleagues for defending the “Jonesville-is-the-USA” scenario:
The notion that one can find the essence of national societies, civilizations, great religions, or whatever summed up and simplified in so-called typical small towns and villages is palpable nonsense. What one finds in small towns and villages is (alas) small-town or village life. (Gomm, Hammersley, & Foster, 2000[a], p. 262; Geertz, 1973, p. 22, italics added).
As noted, qualitative case studies lack the statistical sampling that would be necessary to extend generality to the larger population levels, and thus do not achieve the microcosm possibility. Furthermore, it would be naïve to believe that a miniaturized version of such a large entity could exist in the form of a manageable case unit, given the many complex components. Geertz’ point is that cases with localized contexts matter precisely because of their localized context, not the context of some higher level of U.S. population. British social psychologist Hans Eysenck explained that “sometimes we simply have to keep our eyes open and look carefully at individual cases – not in the hope of proving anything [about the larger population], but rather in the hope of learning something” (Flyvbjerg, 2006, p. 224; Eysenck, 1976, p. 9)!
Case Findings Fallacy (part 2): The Claim of “Action” Research
Novice researchers will occasionally introduce the term “action research” in their report on findings, hoping to emphasize the practicality of their conclusions or their intensive involvement in the social system as part of the research (e.g., in the role of “participant observer”). This verbiage should be discouraged, however, unless the researcher has already planned and executed a formal “action research” design of some kind, which is different from the sort of qualitative or mixed-methods case studies considered here (as part of this training). The following criteria are important distinguishers of “action” researcher:
- The word “action” refers specifically to an occupational function, a departmental program, a health regimen/medical routine, an organizational protocol, or some other systematic process-based phenomenon in which the investigator has a vested interest (due to his or her practitioner status).
- The research is conducted by (or for) someone currently taking the action, overseeing the action, or manipulating the action in any way.
- The main goal of the research is to generate findings that assist the social actor, the actor’s institution/agency, or the actor’s community of practice – especially by improving the action itself (as part of the research phase).
- The data collection coincides with some part of the action or overlaps the entirety of the action, such that the action’s outcomes are also data in research.
If these criteria are present, then the researcher has transcended the bounds of normative case study research and migrated into action research. Industry-based case studies share with action research the goal of delivering professionally relevant and practice-related findings, but they diverge in the manner of this delivery. Case studies do not seek to induce change and then model this process in real-time, such that study results carry immediate and direct implications for the industry practice. There are, however, historical overlaps between the two approaches that might account for some of their confusion. For instance, action research pioneer Chris Argyris trained under sociologist William Foote Whyte, a case study authority during the middle of the 20th century, and thus used case study terminology in his early writings. Case study and action research also played an instrumental role in the rise of program evaluation during the 1980s and 90s and were seminal in showing the usefulness of qualitative data for testing “theories-in-use” (Rogers, 2000, p. 209; Yin, 2000, p. 192; Argyris, 1991; 1979).
Argyris, C. (1979). Essay Review: using qualitative data to test theories. Administrative Science Quarterly 24 (4): 672-79.
Argyris, C. (1991). The use of knowledge as a test for theory: the case of public administration. Journal of Public Administration & Theory 1 (3): 337-54.
Eysenck, H. J. (1976). Introduction. H. J. Eysenck (Ed.), Case Studies in Behavior Therapy (pp. 1-15). London: Routledge.
Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative Inquiry 12 (2): 219-45.
Geertz, C. (1973). The Interpretation of Culture. New York: Basic Books.
Gomm, R., Hammersley, M., & Foster, P. (Eds.) (2000[a]). Case Study Method: Key Issues, Key Texts. Thousand Oaks: SAGE.
Rogers, P. J. (2000). Program theory: not whether programs work but how they work. In D. L. Stufflebeam, G. F. Madaus, & T. Kellaghan (Eds.), Evaluation Models: Viewpoints on Education and Human Services Evaluation (pp. 209-32). Dordrecht: Kluwer Academic.
Thomas, G. (2010). Doing case study: abduction, not induction, phronesis not theory. Qualitative Inquiry 16 (7): 575-82.
Thomas, G., & Myers, K. (2015). The Anatomy of the Case Study. Thousand Oaks: SAGE.