- A case study investigates a bounded phenomenon in depth, within its real-world context, using multiple sources of evidence
- There are six main types: exploratory, descriptive, explanatory, intrinsic, instrumental, and collective: each suited to a different research purpose
- Data is collected through interviews, observations, document analysis, surveys, and physical artifacts: used in combination to triangulate findings
- Analysis moves through organization, coding, pattern identification, cross-source comparison, and theoretical interpretation
- Case studies are best suited to “how” and “why” questions; they complement rather than replace surveys, experiments, and other methods
- Rigor is achieved through construct validity, internal validity, transferability, and reliability: supported by triangulation and transparent documentation
When Should You Use a Case Study?
The case study design is most appropriate in situations where context is everything and where separating a phenomenon from its environment would strip away the very meaning you are trying to understand. Specifically, use a case study when:
| Situation | Why a case study fits |
| You want to explore a new or understudied phenomenon | No established theory yet; depth over breadth needed |
| You need to explain “how” or “why” something occurred | Causal mechanisms require contextual tracing |
| Your research question begins with “how” or “why” | These are the classic triggers for case study design |
| The case is unique, rare, or extreme | Only one or very few instances exist to study |
| You want to test or refine an existing theory | A single well-chosen case can challenge or extend theory |
| You are evaluating a program or intervention | Real-world conditions matter to the assessment |
| You want to compare across contexts | Multiple cases illuminate variation and similarity |
Researchers in health sciences, education, social work, business, psychology, environmental studies, and political science all regularly use case study methodology. It is particularly valued when quantitative data alone would not capture what is most important.
Types of Case Studies
One of the most important decisions in case study research is choosing the right type of case study. Different types serve different research purposes. The most widely used classification: developed by Yin (2018) and elaborated by Stake (1995): covers the following categories.
By research purpose
Exploratory case study
An exploratory case study is conducted when little is already known about a topic. The goal is to define questions, identify variables, and develop hypotheses for future research: not to draw firm conclusions.
- Used in the early stages of inquiry
- Focuses on discovery and question generation
- Often serves as a pilot for a larger study
Example: Examining how a newly remote-first company manages team cohesion when no prior research exists on that specific organizational context.
Descriptive case study
A descriptive case study provides a complete, detailed account of a phenomenon within its context. The researcher is not trying to explain why something happened but to document what happened fully and accurately.
- Based on an established theoretical framework
- Focuses on the “what” rather than the “why”
- Produces rich, contextually detailed documentation
Example: Documenting the full process of a hospital’s transition from paper records to electronic health records, including all stakeholder responses and operational changes.
Explanatory case study
An explanatory case study aims to establish causal relationships: to explain how or why a particular outcome occurred. This is the most analytically demanding type and is frequently used in social sciences.
- Useful for causal analysis in complex real-world settings
- Seeks to identify mechanisms and processes
- Often involves theory testing
Example: Investigating why a particular school’s standardized test scores improved significantly after adopting a new pedagogical method, while neighboring schools using the same method saw no change.
By the nature of the case itself
Intrinsic case study
In an intrinsic case study, the researcher’s primary interest is in the case itself: not because it represents something broader, but because of its own inherent interest or importance.
- Driven by the unique value of the specific case
- Not designed to generalize to other contexts
- Common in biography, history, and clinical research
Example: Studying the leadership style of a particularly unconventional CEO because that individual’s approach is itself fascinating and significant.
Instrumental case study
An instrumental case study uses a specific case as a vehicle for understanding a broader issue or refining a theory. The case is secondary; the insight it enables is primary.
- The case serves as a tool for broader understanding
- Findings are intended to apply beyond the specific case
- Used for theory building and theory testing
Example: Studying one city’s response to a flooding event to understand general principles of municipal crisis management.
Collective (multiple) case study
A collective or multiple case study involves studying several cases simultaneously or sequentially to investigate a general phenomenon. The cases may be similar (replication logic) or contrasting (maximizing variation).
- Produces more robust and generalizable findings than a single case
- Cases can be analyzed individually then compared
- Stronger basis for theoretical conclusions
Example: Studying five different schools that implemented the same reading intervention to determine which contextual factors predict success or failure.
Single-case vs. multiple-case design
| Feature | Single-case design | Multiple-case design |
| Best for | Unique, critical, or revelatory cases | Theory testing, comparison, replication |
| Depth | Very high | Moderate to high |
| Generalizability | Lower | Higher |
| Resource demand | Lower | Higher |
| Example | Phineas Gage’s brain injury case in neuroscience | Comparing educational reforms across five countries |
How to Conduct a Case Study: Step-by-Step Process
Step 1: Define your research question
Everything in a case study flows from the research question. A well-formulated case study question typically:
- Begins with “how” or “why”
- Is focused enough to be answerable with one or a few cases
- Is complex enough to warrant an in-depth approach rather than a survey or experiment
Step 2: Select your case or cases
Case selection is not arbitrary. Your case should be chosen because it is particularly likely to illuminate your research question. Strategies for selection include:
| Selection strategy | When to use it |
| Typical case | To illustrate what is ordinary or common in a category |
| Extreme or deviant case | To understand what happens at the outer edges |
| Critical case | To test a theory: if it doesn’t hold here, it doesn’t hold anywhere |
| Revelatory case | To study something previously inaccessible or unexamined |
| Maximum variation | To capture the full range of diversity in a phenomenon |
Also at this stage: define the boundaries of your case clearly. What is included? What time period? Which people, locations, or events fall within the case and which fall outside it? This process is called bounding the case.
Ethical considerations must also be addressed here: informed consent, confidentiality, anonymization, and access to sensitive data.
Step 3: Build a theoretical framework
Before collecting data, situate your case study within existing theory and literature. This theoretical framework serves several purposes:
- It grounds your analysis in established knowledge
- It guides what data to collect and what to look for
- It makes your findings meaningful to your field
Step 4: Determine your data collection methods
Case studies are inherently multi-method. The strength of the design comes from triangulation: gathering evidence from multiple sources and checking that they converge. The main data collection methods are described below.
- Structured interviews use a fixed list of questions, making responses easy to compare
- Semi-structured interviews have a topic guide but allow follow-up and exploration
- Unstructured interviews are open-ended conversations guided by the participant
- Participant observation: the researcher joins the activity being studied
- Non-participant observation: the researcher watches from outside without involvement
- Document analysis of reports, memos, meeting minutes, emails, policy documents
- Archival records and organizational histories
- Media coverage, published accounts, public records
Step 5: Collect your data
With your protocol established, data collection should be:
- Systematic: follow your protocol consistently
- Documented: keep a case study database of all raw evidence
- Iterative: allow emerging findings to prompt new questions while staying within your research boundaries
- Ethical: maintain confidentiality and honor consent agreements throughout
Keep a research journal or reflexivity log to note your decisions, reactions, and any potential sources of bias as they arise.
Step 6: Analyze your data
Case study analysis is typically conducted through the following sequence:
| Stage | What happens |
| Organization | Sort data by source, time period, or theme; transcribe interviews |
| Coding | Assign labels to segments of data that represent concepts or themes |
| Pattern identification | Look for recurring themes, contradictions, or surprising findings across sources |
| Cross-source comparison | Check whether interview data, documents, and observations tell a consistent story |
| Theoretical interpretation | Relate your patterns to the theoretical framework established in Step 3 |
| Verification | Re-check the analysis for accuracy and consistency; seek peer review or member checking |
If you are using multiple cases, add a cross-case synthesis step: analyze each case individually first, then compare across cases to identify what varies and what is consistent.
Step 7: Write and report your findings
How you report a case study depends on your purpose and audience. Two broad approaches exist:
Structured (scientific paper) format:
- Introduction and research question
- Literature review and theoretical framework
- Methodology (case selection, data collection, analysis approach)
- Findings
- Discussion and implications
- Limitations
- Conclusion
Narrative format:
- Story-driven, chronological account
- Weaves data into a coherent narrative
- More common in education, social work, and humanities research
Regardless of format, your report must include:
- Background and context of the case
- A clear description of how data was collected and analyzed
- Evidence-supported findings with direct references to the data
- Discussion of how findings relate to existing theory
- Honest acknowledgment of limitations
Case Study vs. Other Research Methods
Choosing between methods is one of the most consequential decisions a researcher makes. The table below compares case studies to other common approaches like ethnography, surveys, and experimental research.
| Feature | Case study | Survey | Experiment | Ethnography |
| Primary question type | How, why | How many, how often | Does X cause Y? | What is it like to live/work in X? |
| Sample size | 1 to a small number | Large | Varies (often moderate) | Small, immersive |
| Depth of data | Very high | Low to moderate | Moderate | Very high |
| Generalizability | Low to moderate | High | High (if randomized) | Low |
| Control over variables | None | None | High | None |
| Best for | Complex phenomena in context | Patterns across populations | Causal claims | Cultural and social meaning |
| Time and resource demand | High | Low to moderate | Moderate to high | Very high |
The case study is not better or worse than these alternatives: it is appropriate for a different type of question. When your question requires understanding mechanisms in context rather than measuring patterns across populations, case study is the right choice.
Benefits of Case Studies
- Contextual richness: captures the complexity of real-world situations that cannot be replicated in a controlled setting
- Deep understanding: produces detailed knowledge about mechanisms, processes, and relationships within a specific case
- Methodological flexibility: can incorporate qualitative data, quantitative data, or both, adapting to the nature of the case
- Theory development: well-conducted case studies are a primary source of new theoretical propositions and conceptual frameworks
- Practical relevance: findings directly reflect real-world conditions, making them highly applicable to practitioners and policymakers
- Rare phenomena: the only viable method for studying cases that occur too infrequently to be captured by surveys or experiments
- Holistic perspective: treats the case as a whole rather than decomposing it into isolated variables
Limitations of Case Studies
| Limitation | Explanation | How to mitigate it |
| Limited generalizability | Findings from one or a few cases may not apply elsewhere | Use multiple cases; be explicit about the scope of claims |
| Researcher bias | The researcher’s perspective can shape analysis and interpretation | Use reflexivity logs, member checking, and peer review |
| Participant bias | Participants may behave differently when observed (Hawthorne effect) | Use multiple methods; compare observed behavior against self-reports |
| Time and resource intensity | In-depth data collection and analysis takes significant time | Plan carefully; scope the case boundaries tightly |
| Replication difficulty | The specific context makes exact replication nearly impossible | Document the protocol fully; be transparent about context |
| Subjectivity in analysis | Coding and interpretation involve judgment calls | Establish coding frameworks in advance; use multiple coders where possible |
Acknowledging these limitations explicitly in your report strengthens, not weakens, the credibility of your findings.
Ensuring Rigor and Quality in Case Study Research
Quality in case study research rests on four criteria, adapted from Yin (2018):
| Quality criterion | What it means | How to achieve it |
| Construct validity | You are measuring what you claim to measure | Use multiple data sources; have key informants review draft findings |
| Internal validity | Your causal conclusions are justified by the evidence | Pattern-match across sources; address rival explanations |
| External validity (transferability) | Readers can assess whether findings apply to their context | Provide thick description of the case context |
| Reliability | Another researcher following your protocol would reach similar conclusions | Document all procedures in a case study protocol |
Triangulation (the practice of cross-checking findings across different data sources, methods, or investigators) is the single most important tool for establishing rigor in case study research.
Frequently Asked Questions
Can a case study be quantitative?
Yes. While case studies are predominantly associated with qualitative research, they can incorporate quantitative data. A mixed-methods case study might combine patient outcome statistics with clinical interview data, or use survey responses alongside ethnographic observation. What defines a case study is the in-depth, contextualized focus: not the type of data.
How many cases do you need for a case study?
A case study can involve a single case or multiple cases. Single-case designs are appropriate for unique, critical, or revelatory cases. Multiple-case designs (typically 4 to 10 cases) are used when the researcher wants more robust, transferable findings or needs to compare across contexts.
What is the difference between a case study and a case report?
A case report (common in medicine) is a factual description of an individual patient’s symptoms, diagnosis, and treatment. A case study is broader and more analytical: it investigates a phenomenon systematically using a defined methodology, a theoretical framework, and multiple data sources. Case reports describe; case studies analyze.
How is a case study different from an experiment?
An experiment manipulates an independent variable under controlled conditions to establish causation. A case study observes a phenomenon as it naturally occurs, without manipulation or control. Experiments ask “does X cause Y?”; case studies ask “how and why did this outcome occur in this particular context?”
How long does a case study take?
There is no fixed duration; a focused single-case study might take a few weeks from design to write-up. A longitudinal multiple-case study examining change over time could take months or years. The timeline depends on the complexity of the case, the number of data sources, and the depth of analysis required.
Can case studies be used to develop theory?
Yes, case studies are one of the primary routes to theory development in social science. By uncovering mechanisms and patterns not previously described, a well-conducted case study can propose new theoretical concepts or extend existing frameworks. This is distinct from theory testing, though case studies can do both.
What is the difference between an intrinsic and an instrumental case study?
An intrinsic case study focuses on a case because the case itself is interesting: the researcher wants to understand that specific person, organization, or event. An instrumental case study uses the case as a means to understand something beyond it: a broader issue, pattern, or theoretical proposition. The distinction, developed by Stake (1995), reflects whether the case is an end or a tool.
References
Creswell, J. W., & Poth, C. N. (2017). Qualitative inquiry and research design: Choosing among five approaches (4th ed.). SAGE Publications.
Stake, R. E. (1995). The art of case study research. SAGE Publications.
Yin, R. K. (2018). Case study research and applications: Design and methods (6th ed.). SAGE Publications.
This article was originally published on June 27, 2024, and updated on June 6, 2026.
