What is a case study?

Thoughts on what it means to "do" case study research

Case studies are one of the most common forms of qualitative research. But what exactly is a case study? A case study is a type of study, which distinguishes it from an example, exemplar, or anecdote. While case studies absolutely contain examples and granular details, they are neither examples nor stories alone. Certainly, high-quality case studies involve stories and a narrative. What distinguishes a case study from example or story is that a case study implies a multi-source collection of data, such as historical archives, surveys, interviews, observations, and mapping spaces. Case studies are consequently informed by data and first-hand research. This distinction helps to separate case study research from the more popular and less rigorous version of the term “case study” that tends to accompany digital marketing and online retail wherein a case study is an illustration of a phenomenon—just an example.

At the heart of case study research is a case. A case is a source of consternation for researchers, who typically note that a case has no unifying definition and is responsive to domain knowledge, disciplinary conventions, and practical constraints (e.g., Bartlett and Vavrus, 2017; Bromley, 1986; Dyson and Genishi, 2005; Gerring, 2017; Jocher, 1928; Ragin and Becker, 1992; Tellis, 1997; Yin, 2009). A decided lack of consensus about a “case” creates friction between domains and disciplines.

There are, however, recurring patterns and Wittgensteinian “family resemblances” regarding case study definitions. To examine a case more closely, I want to describe the case study through four foundational elements: case studies are (1) epistemically open-ended, (2) but do not draw clear demarcations between subject of analysis and its context. Case studies are demarcated through (3) a bounded system aimed at holistic investigations, and (4) are based on multiple types of data to better understand the multiple realities of a case.

First and second, the case study is an open-ended in-depth investigation (Dyson and Genishi, 2005; Merriam, 1998; Morgan, 2012; Stake, 1995; Yin, 2009), where the delineation between the subject of analysis and context is not clear or inseparable (e.g., Flyvbjerg, 2006, p. 222; Duff , 2007, p. 30; Lillis, 2008, p. 374-77; Morgan, 2012, p. 668). Case studies are, more specifically, epistemically open-ended with a detailed level of density and holism. Researchers such as Crowe et al. (2011), Dyson and Genishi (2005), Gerring (2017), Kenny & Grotelueschen (1984), Stake (1995), Swanborn (2010) argue that cases are holistic and intense due to their comprehensive and fully elaborated nature.Colloquially, case studies examine “up close.” For this reason, researchers should come to case study research understanding their investigations will not “pin down” their inquiries. Maxwell (2009) reflects this epistemic positioning by framing the case study as about process of research, not the product of a research investigations, writing that case study research questions “…involve an open-ended, inductive approach to discover what these meanings and influences are and how they are involved in these events and activities— an inherently processual orientation” (p. 83). The case study approach necessarily sees multiple meanings in an investigation, reflecting a multiplicity of contexts, some of which are contradictory, paradoxical, and messy. Case study knowledge is therefore context-dependent rather than “ground truth.” The implications here can include: the role of generalizability (either intrinsic or extrinsic) when analyzing and writing up case studies, the extent to which case study research is preliminary (leading to further study and studies in terms of design), and the purpose of a case itself (to learn about phenomena or to transfer knowledge from one context to another).

Third and fourth, case studies are holistic, bounded systems (Dyson and Genishi, 2005; Gerring, 2017; Stake, 1995; Yin, 2009) that draw on multiple sources of data, that is triangulation (e.g., Baxter and Jack, 2008; Blatter & Haverland, 2012; Gerring, 2017; Miles, 2015; Simons, 2012; Yin, 2009). A bounded system is identified by the researcher through their own research questions, disciplinary background, tools, life histories, and practical constraints. A bounded system is a flexible approach for demarcating when, where, and how to begin and end conducting research. Merriam writes that a case can be “an intensive, holistic description and analysis of a bounded phenomenon such as a program, an institution, a person, a process, or a social unit” (xiii).

Because case studies are in-depth and attempt to understand the system under investigation in rich ways representing complexity, triangulation is necessary. Of triangulation, psychologist Bromley (1986) writes, “The navigational metaphor suggests that if several independent sources of evidence point to a common conclusion, then one’s confidence in that conclusion is strengthened” (6). Duff (2007) elaborates on Bromley’s claims when she writes:

Data, methods, perspectives, theories, and even researchers can be triangulated in order to produce either converging or diverging observations and interpretations. Although the notion of triangulation may have originally had positivist undertones (multiple sources of information leading to one “truth” to be discovered by the researcher), it can also be used to ascertain multiple forms of interpretation (or multiple realities) at work in order to “clarify meaning by identifying different ways the case is being seen (p. 30)

Baxter and Jack (2008), too, write, “Triangulation of data sources, data types or researchers is a primary strategy that can be used and would support the principle in case study research that the phenomena be viewed and explored from multiple perspectives” (p. 556). With these perspectives in mind, researchers can, consequently, understand that a case study may combine interviews, surveys, observations, fieldwork, user experiences, and many other sources to complement rich analyses. The use of multiple data sources underscores, again, the epistemic open-endedness of the case study as a methodology.

Typically, these data sources are combined to reflect the “multiple-realities” (Dyson and Genishi, 2005, p. 18; Stake 1994, p. 43; Yin, 2009, p. 18) of a case, meaning that case studies do not undercover—or create depending on the researcher’s epistemology—a single unifying reality but sets of historically contingent activity. Indeed, Baxter and Jack (2008) note that, “…researchers should…plan for opportunities to have either a prolonged or intense exposure to the phenomenon under study within its context so that rapport with participants can be established and so that multiple perspectives can be collected and understood…” (p. 556). Due to these multiple realities and perspectives, the bounded system of a case study can be iterative in nature, meaning that bounded systems are recreated over time if the inquiry needs the boundaries to be redesign or rethought. In fact, because case studies are meant to be detailed in nature, they probably should be iterative in nature, although some case study researchers, i.e., Gerring (2017) and Yin (2009), tend to disagree with this due to their disciplinary associations with more quantitative social sciences, political science and psychology respectively—and are subsequently less opened ended.

In addition to these four descriptive elements, there are two common themes around the case study as a research approach but not necessarily definitional. First, case studies yield complex results that are often unable to be distilled into summarized bullet points. Philosopher of science Mary S. Morgan (2012), for instance, writes, “The outcome is a complex, often narrated, account that typically contains some of the raw evidence as well as its analysis and that ties together the many different bits of evidence in the study” (p. 668). Likewise, Bent Flyvbjerg argues that the case study should result in a narrative, one that does not close the inquiry but opens it to possibility. Flyvbjerg (2006) writes:

I try to leave scope for readers of different backgrounds to make different interpretations and draw diverse conclusions regarding the question of what the case is a case of. The goal is not to make the case study be all things to all people. The goal is to allow the study to be different things to different people. I try to achieve this by describing the case with so many facets—like life itself—that different readers may be attracted, or repelled, by different things in the case. Readers are not pointed down any one theoretical path or given the impression that truth might lie at the end of such a path. Readers will have to discover their own path and truth inside the case. (p. 238)

Here, Flyvbjerg echoes the open-endedness of case study research but with an emphasis on writing up the study. The narrative of a case study, in my view, can be scaled up or down to account for the inquiry and allows the multiple realities to be reflect directly in the writing itself.

The second non-epistemic theme is to understand the history of the case study as a maligned form of research—generally before the 1990s and especially within the fields of psychology and sociology—and then to argue for the strengths of the case study approach (Gerring, 2004; Stoecker, 1991). To be clear, I have not read many articles published since the 2000s that malign case study research. Most researchers appear to accept the case study a legitimate approach, although there is considerable debate over its status as an approach versus a method versus a methodology. Gerring (2017) and Bartlett & Vavrus (2017) articulate some of these issues, although the discussions tend to be ensconced in disciplinary conversations, notably how a disciplinary mistreated the case study approach historically.

Sources:

http://publish.illinois.edu/johnrgallagher/files/2021/08/case-study-bib.pdf