In academic and clinical research, case study research is a specific qualitative (and sometimes mixed-methods) design for investigating a bounded, real-world instance in depth — not a business-school teaching exercise and not a single-patient clinical write-up. This guide covers the case study as a research method: what qualifies as one, when to choose it, how it’s designed, and how it differs from adjacent formats and methods. If you’re looking for how to write up a completed case study paper section by section, see CASRAI’s guide to structuring a case study paper; if you’re trying to distinguish a research case study from a clinical case report, see the Case Study vs. Case Report comparison.
What Is Case Study Research?
The most widely cited methodological definition comes from Robert K. Yin, whose Case Study Research: Design and Methods frames a case study as an empirical inquiry that investigates a contemporary phenomenon in depth and within its real-world context — especially when the boundary between the phenomenon and the context is not clearly evident. Robert Stake’s complementary tradition treats the case study less as a method defined by technique and more as a choice of what to study: a single bounded system (a person, program, organization, event, or community) examined as an integrated whole.
Operationally, a piece of research qualifies as a case study when it:
- Focuses on one case, or a small, deliberately selected set of cases, rather than a large sample
- Treats the case as bounded — defined in time, place, and activity, so a reader can tell what is inside the case and what is context
- Draws on multiple sources of evidence (interviews, documents, observation, archival records) that are triangulated against each other rather than relying on a single instrument
- Asks primarily “how” or “why” questions about a process or mechanism, rather than “how many” or “how much” questions
A study that surveys 300 organizations with a fixed questionnaire is not a case study, even if it discusses one organization in the introduction. A single anecdote used to illustrate a general point, without systematic data collection or bounding, is not a case study either — it’s an example. The method requires deliberate design, not just a small N.
When to Use the Case Study Method
Yin’s own guidance is that case study is the preferred method when three conditions hold together: the research question is a “how” or “why” question, the researcher has little or no control over behavioral events, and the focus is on a contemporary phenomenon within a real-world context rather than a purely historical one. Typical uses in academic and applied research include:
- Explaining why a policy, program, or intervention produced the outcomes it did in a specific setting
- Investigating a phenomenon that is not yet well understood, where an in-depth, exploratory look precedes hypothesis-driven work
- Studying a setting where random assignment or large-sample survey methods are not feasible or ethical — a single organization undergoing a unique change process, for example
- Building or refining theory by examining how a mechanism operates in a real context, rather than testing an existing theory against a large sample
Case study is a poor fit when the research question is fundamentally about prevalence, frequency, or statistical association across a population — those questions call for survey, experimental, or large-N observational designs instead.
Types of Case Study Designs
By Purpose (Yin’s Typology)
Yin distinguishes case studies by the kind of research question they answer:
- Exploratory — used to develop hypotheses or a conceptual framework for further study, often as a precursor to a larger research program
- Descriptive — documents a phenomenon and the real-world context surrounding it in detail, without necessarily testing a causal claim
- Explanatory — used to explain presumed causal links in real-world interventions that are too complex for survey or experimental methods to capture, often tracing a causal chain through the case
By Structure: Single-Case vs. Multiple-Case
A single-case design is justified when the case is critical (it tests a well-formulated theory), extreme or unique, representative or typical, revelatory (previously inaccessible to research), or longitudinal (studied at two or more points in time). A multiple-case design studies several cases using the same protocol, treating each case as one experiment in a series — replication logic, not sampling logic, governs how many cases to include and why. Within either, a design can be holistic (one unit of analysis for the whole case) or embedded (multiple, nested units of analysis within the same case, such as several departments within one organization).
By Intent (Stake’s Typology)
Stake’s framework classifies case studies by why the case was chosen: an intrinsic case study is undertaken because the case itself is of inherent interest, independent of what it might reveal about other cases; an instrumental case study uses a case primarily to gain insight into a broader issue or to refine a theory; a collective (or multiple) case study examines several cases jointly to understand a phenomenon, population, or general condition.
Data Collection and Triangulation
Case study research is defined partly by its reliance on multiple, converging sources of evidence rather than a single instrument. Common sources include semi-structured interviews, direct and participant observation, organizational or archival documents, physical artifacts, and audiovisual material. Triangulation — checking whether independent sources of evidence point to the same conclusion — is the primary mechanism case study researchers use to strengthen construct validity, since no single source is treated as sufficient on its own. Maintaining a documented chain of evidence (a clear, auditable link from research question, to raw data, to case study database, to conclusions) is a related quality safeguard specific to this method, distinct from the statistical reliability checks used in quantitative designs.
Generalizability: Analytic vs. Statistical Generalization
The most common methodological objection to case study research is that findings from one case (or a handful of cases) cannot be generalized to a population. Yin’s response, now broadly accepted within the qualitative-methods literature, is that case studies generalize to theory, not to populations — a distinction usually labeled analytic generalization versus statistical generalization. A well-designed case study can extend or refine a theoretical proposition (analytic generalization), even though it cannot claim that its findings hold with a known probability across a defined population (statistical generalization, the kind survey research targets). Reviewers unfamiliar with this distinction sometimes mistakenly evaluate a case study against statistical-generalization criteria it was never designed to meet — stating the intended form of generalization explicitly in the methods section heads this off.
Strengths and Limitations
Strengths: case study research can capture contextual complexity that a survey instrument strips away; it is well suited to studying novel or poorly understood phenomena where the relevant variables aren’t yet known well enough to operationalize into a questionnaire; and it can trace process and mechanism (how something happened) rather than only association (that something happened).
Limitations: findings are harder to generalize statistically; the method is vulnerable to researcher bias in case selection and in what evidence gets emphasized during analysis, which is why an explicit, justified case-selection rationale and a documented chain of evidence matter so much; and rigorous case study research is often more time- and access-intensive per case than a survey wave, which constrains how many cases a study can realistically include.
Case Study Research vs. Other Qualitative Methods
Case study is frequently confused with, or treated as interchangeable with, other qualitative approaches that actually answer different kinds of questions:
- Ethnographic research focuses on culture and lived experience within a group over sustained fieldwork, typically through participant observation, rather than on a bounded case defined primarily by its real-world context. See CASRAI’s ethnographic research method guide.
- Grounded theory aims to build theory inductively from data across many instances using constant comparison, rather than bounding and describing a single case in depth.
- Phenomenological research investigates the structure of lived experience of a phenomenon across multiple participants, not a single bounded case. See CASRAI’s phenomenological research guide.
- Thematic analysis is an analysis technique that can be applied within a case study (or many other qualitative designs) to identify patterns across interview or document data — it is a method of analysis, not a study design in its own right. See CASRAI’s thematic analysis guide.
- Action research is organized around cycles of intervention and reflection intended to produce change within the setting studied, rather than around bounding and describing an existing case. See CASRAI’s action research guide.
These methods are not mutually exclusive in practice — a case study can incorporate thematic analysis as its analytic technique, and a single organization can be studied ethnographically or as a case study depending on the research question asked.
Case Study vs. Case Report
Outside the social sciences, “case study” is sometimes used loosely to mean the same thing as a clinical case report — a short, published account of a single patient’s unusual presentation, course, or outcome, typically following the CARE guidelines. The two are methodologically distinct: a case report generally describes one patient without a formal research methodology or an explicit generalization claim, while a case study research design is built around a stated methodology, a justified case-selection rationale, multiple triangulated data sources, and an explicit form of generalization (usually analytic, not statistical). See CASRAI’s full Case Study vs. Case Report comparison for a side-by-side breakdown.
Reporting a Case Study
Because a case study paper is organized around bounding, describing, and analyzing a case rather than around a single hypothesis test, it doesn’t map cleanly onto the standard IMRaD skeleton used across most empirical research (see CASRAI’s IMRaD Structure entry). Sections specific to case study reporting typically include an explicit statement of the case boundaries, the case-selection rationale, the unit(s) of analysis, and how triangulation was used to strengthen the findings. CASRAI’s guide to structuring a case study paper in social science research walks through that structure section by section, including how case study papers depart from IMRaD and how to write the case-selection justification.
Frequently Asked Questions
Is case study research qualitative or quantitative?
Case study research is most often qualitative, but it is not exclusively so — many case studies are mixed-methods, combining qualitative sources (interviews, documents, observation) with quantitative data drawn from the same case (organizational records, survey data collected within the case, financial or operational metrics). What defines the method is the bounded, in-depth, multi-source investigation of a case, not the exclusion of numbers.
How many cases should a case study include?
There’s no fixed rule. A single-case design is justified on specific methodological grounds (the case is critical, extreme, typical, revelatory, or longitudinal). A multiple-case design typically includes enough cases to satisfy replication logic for the theoretical proposition being examined — often a handful, since each case functions like one experiment in a series rather than one row in a statistical sample.
Can case study findings be generalized?
Case study findings generalize analytically, to theory, rather than statistically, to a population. A case study can extend, refine, or challenge a theoretical proposition; it cannot claim that its findings hold with a known probability across a broader population the way a random sample can.
What’s the difference between a case study and a case report?
A case study is a research method built around an explicit methodology, a justified case-selection rationale, multiple triangulated data sources, and a stated form of generalization. A case report is a clinical publication format describing a single patient’s unusual presentation or outcome, typically without a formal research methodology. See the full comparison.
What’s an example of a research question suited to the case study method?
Questions phrased as “how” or “why” about a specific, bounded, contemporary situation fit best — for example, “why did this particular data-sharing policy succeed at this institution when similar policies failed elsewhere?” A question phrased as “what percentage of institutions have adopted this policy?” is not a case study question; it calls for a survey design instead.
