Multiple Case Study — Multiple Case-Based Case Study Design
Multiple Case Study Design · Also known as: multiple-case design, collective case study, multi-site case study, multi-case study
A multiple case study (also called a multiple-case design or collective case study) is a qualitative research design in which two or more bounded cases are examined together to pursue a common research question. By studying several instances of a phenomenon in parallel, the researcher can compare patterns, identify convergences and divergences, and build more robust, transferable conclusions than a single case could support. The design draws principally from Robert Yin's case-study methodology and Robert Stake's collective case study tradition.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Choose a multiple case study when your research question asks how or why a phenomenon occurs across different contexts, and when analytic generalization — extending a theoretical proposition to new settings — matters more than statistical generalization to a population. The design is particularly strong when you want to test whether a pattern holds under varied conditions, or to explain why outcomes differ across sites. It is appropriate in education, organizational research, public policy, health services, and social sciences. Do not choose this design when you need statistical representativeness, when resources allow only one case and depth matters most (prefer a single case study), when the cases cannot be clearly bounded, or when the research question calls for frequency or magnitude rather than mechanism and process.
Strengths & limitations
- Replication logic across cases strengthens confidence in findings beyond what a single case can provide.
- Multiple sources of evidence within each case enable rich triangulation of data.
- Suited to explanatory and descriptive questions about complex, context-embedded phenomena.
- Findings support analytic generalization to theory, broadening applicability beyond the specific sites studied.
- Divergent cases can be used to test and refine theoretical propositions rather than simply illustrate them.
- Resource-intensive: collecting and analyzing several full cases demands considerable time, access, and fieldwork effort.
- Analytic generalization is not statistical generalization; findings cannot be extrapolated to a population in a probabilistic sense.
- Quality depends on the researcher's consistency in applying the same protocol across all cases; researcher skill is a major variable.
- With more than 8–10 cases the design risks becoming superficial, losing the depth that distinguishes case study from survey research.
Frequently asked
How many cases do I need for a multiple case study?
Yin recommends thinking in terms of replication logic rather than a minimum number: even two or three cases can be powerful if they are chosen to produce literal replication (same result expected) or theoretical replication (contrasting result for identifiable reasons). In practice, studies commonly use between 2 and 10 cases. Fewer than two is by definition a single case study; more than about ten risks sacrificing the depth that defines a case study.
What is the difference between a multiple case study and a comparative study?
Both compare across units, but a comparative study (especially in political science or sociology) often selects cases to test causal hypotheses and may use structured focused comparison. A multiple case study in the Yin tradition emphasizes thick, in-depth description within each case and uses replication logic to build or test theory. The two overlap considerably; the distinction is partly disciplinary convention.
Can I use quantitative data in a multiple case study?
Yes. Case study is a research design, not a data type. Yin explicitly allows quantitative evidence as one of several data sources within a case. The defining feature is that data from multiple sources are triangulated within and across bounded cases — not that only qualitative data are used.
What is the difference between Yin's and Stake's approaches?
Yin's approach is more positivist-leaning: it emphasises a formal protocol, replication logic, pattern-matching against propositions, and rigour criteria analogous to construct validity and reliability. Stake's collective case study is more interpretive and naturalistic: it foregrounds the researcher's role as interpreter, prioritises thick description, and values understanding over explanation. Both are legitimate; the choice should match your epistemological stance.
How do I ensure quality in a multiple case study?
Yin identifies four quality criteria: construct validity (use multiple data sources and member checking), internal validity for explanatory studies (use pattern-matching and rival explanations), external validity/transferability (use replication logic and thick description), and reliability (maintain a case study protocol and database so another researcher could replicate the data collection process).
Sources
- Yin, R. K. (2018). Case Study Research and Applications: Design and Methods (6th ed.). Sage. ISBN: 978-1506336169
- Stake, R. E. (2006). Multiple Case Study Analysis. Guilford Press. ISBN: 978-1593852481
How to cite this page
ScholarGate. (2026, June 3). Multiple Case Study Design. ScholarGate. https://scholargate.app/en/qualitative/multiple-case-based-case-study
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Comparative Case StudyQualitative↔ compare
- EthnographyQualitative↔ compare
- Grounded TheoryQualitative Research↔ compare
- PhenomenologyQualitative↔ compare
- Single-Case StudyQualitative↔ compare