Single-Case Study — Single-Case Study Design

In a single-case study, one bounded instance is investigated in depth; in a multiple-case study, the same protocol is applied to two or more cases and findings are compared across them. Multiple cases provide stronger grounds for theoretical generalization because convergent findings across independent cases are more convincing than findings from one case alone. However, a single case is justified when it is genuinely unique, extreme, critical, or revelatory — conditions where replication across cases is neither possible nor the point.

Not exactly. Case studies can use qualitative data, quantitative data, or both. What defines the method is the focus on a bounded instance in its real-world context, not the type of data collected. A single-case study that traces an organization's quarterly performance over five years using archival records is still a case study even though the data are numeric. That said, most published single-case studies are primarily qualitative because the 'how' and 'why' questions they address call for interpretive rather than statistical analysis.

Yin's framework maps four quality criteria onto the design: construct validity (using multiple data sources and having key informants review draft reports), internal validity (applying pattern-matching or explanation-building to rule out rival interpretations), external validity (articulating the theoretical domain to which findings generalize and why), and reliability (maintaining a case study protocol and a case study database so another researcher could repeat the data collection and reach the same findings).

Not statistically — but analytically, yes. A single well-chosen critical case can confirm or disconfirm a theoretical proposition in the same way that a single experiment can. Yin calls this 'analytic generalization': the researcher generalizes findings to theory, not to a population. The claim is not 'this is how all hospitals work' but 'this case provides evidence that challenges / supports theory X under conditions Y and Z.'

Choose a case study when the primary goal is to understand a bounded real-world instance — an event, program, decision, or organization — and you want to explain how and why it unfolded as it did using multiple data sources. Choose ethnography when the goal is to describe the shared culture, meanings, and practices of a group over a prolonged period of immersion. Choose phenomenology when the goal is to reveal the essential lived-experience structure of a phenomenon as reported by participants. Case study is the most flexible of the three: it imposes fewer restrictions on data type and analytic tradition.