What are the types of case study?

Case studies can be single-case or multiple-case, and holistic (one unit of analysis) or embedded (several). By purpose they are often described as intrinsic (the case itself is of interest), instrumental (the case illuminates a broader issue), or collective/multiple (several cases compared). Yin and Stake offer the two most-cited frameworks.

How do I bound a case study?

Define the case explicitly: what is in and what is out, in terms of the unit of analysis, the time period, and the setting. A clearly bounded case keeps the study focused and answerable; an unbounded 'case' that drifts into everything is a common reason case studies lose rigour.

Home › Guides › Research Design › Case study research

Case study research

Q: What is case study research?

Case study research is an in-depth investigation of a single case or a small number of cases — a person, group, organisation, event, or programme — within its real-world context. It uses multiple data sources to build a rich, holistic understanding, and is well suited to 'how' and 'why' questions about contemporary phenomena you cannot control.

Q: Can you generalise from a case study?

Not in the statistical sense. Case studies support analytical or theoretical generalisation — extending or testing theory — and transferability, where rich description lets readers judge whether findings apply to their own context. Framing your contribution this way, rather than claiming statistical generalisability, answers the most common critique.

By Dr. Rafiq Muhammad, MD, PhD · Updated June 2026

A case study is an in-depth investigation of one case — or a few — in its real-world context: a person, a team, an organisation, an event, a programme. It trades breadth for depth, drawing on multiple data sources to understand how and why something happens where you can’t control the conditions. Done well it’s rigorous; done loosely it’s “I looked at one thing.”

When a case study is the right choice

Your question is how or why, not how many.
The phenomenon is contemporary and you can’t manipulate it.
Context matters — you can’t sensibly separate the case from its setting.

The types

Two axes you’ll be asked about: single vs multiple case, and holistic vs embedded (one unit of analysis vs several within the case). By purpose, cases are often intrinsic (the case itself matters), instrumental (the case illuminates a wider issue), or collective (several cases compared). Yin and Stake are the two frameworks examiners expect you to cite.

Bound the case — or lose rigour

Define what is in and what is out: the unit of analysis, the time period, the setting. A clearly bounded case stays answerable; an unbounded “case” that drifts into everything is the most common way case studies lose their footing.

Strength of the method: case studies use multiple data sources — interviews, documents, observation, records — and converge them. That triangulation is what turns a single case from anecdote into evidence.

Handling the “you can’t generalise” critique

You can’t generalise statistically from one case — and you shouldn’t claim to. Case studies offer analytical generalisation (extending or testing theory) and transferability (rich description lets readers judge relevance to their own setting). Framing your contribution that way — not as a representative sample — answers the critique before it’s raised.

Deciding between a case study and another design? The free Methods Checker tests whether your question, design, and methods actually fit together.

Get the free Research Design toolkit

Design-selection worksheets, a case-boundary template, and worked examples from Research Design Simplified. We’ll email you the download link.

Get Research Design Simplified

Frequently asked questions

What is case study research?

An in-depth study of one or a few cases in real-world context, using multiple data sources — suited to how/why questions.

What are the types?

Single vs multiple, holistic vs embedded; by purpose intrinsic, instrumental, or collective. Yin and Stake are the key frameworks.

How do I bound a case?

Define the unit of analysis, time period, and setting — what is in and what is out. Unbounded cases lose rigour.

Can you generalise from a case study?

Not statistically — but analytically (testing theory) and via transferability (rich description). Frame it that way.

Types of research design → Open the Methods Checker →