What is the difference between a case study and a case analysis?

A research case study is an inquiry you design and carry out on a bounded real case (Yin; USC's case-study page). A case analysis is a problem-based assignment on a packet someone else wrote, usually in the HBS tradition (USC's case-analysis page). USC's twelve-point tip is blunt: the researcher picks the case; the professor assigns the analysis. If you did not collect evidence outside the handout, you are writing an analysis.

How long should a case study be?

Long enough to bound the case, show the evidence, and finish the analysis — not long enough to dump every note. Course papers often run 2,000–5,000 words. A Yin research report can be article length. HBS says a modern teaching case is typically 10–20 pages before your analysis; Biddle's 1921 General Shoe case was one page. CARE reports follow the journal's limit, not a word-count myth.

Can I write a case study with one interview?

You can write an interview study. Yin’s method wants multiple sources that converge. The SAGE refresher lists six common sources; one interview can be a source inside a case. It is not, by itself, a completed case study.

Do I need a hypothesis?

Exploratory and intrinsic cases can start with questions. Explanatory cases need theoretical propositions or predicted patterns so pattern matching is possible. Teaching-case analyses need a decision, not a null hypothesis. CARE reports need a uniqueness claim, not a p-value.

How many cases do I need?

One is enough if the selection logic is strong (a critical, extreme, or revelatory case). Add cases only for stated replications. Yin's refresher rejects a power-analysis formula for case counts: more cases raise confidence, but you do not tally "five supported the proposition" the way you would tally a sample.

How do I cite an unpublished interview in the case study?

In APA 7 a personal interview that a reader cannot recover is cited in the text only, not on the reference list. A published or archived interview is cited as the container (podcast, article, transcript). See cite an interview in APA.

What belongs in a medical case report title?

CARE item 1 on the 2013 checklist: the diagnosis or intervention of primary focus, followed by the words case report. Do not write a witty title that hides the condition. Specialty extensions listed on EQUATOR as of 16 February 2026 (CARE-radiology 2024, CARE-acupuncture 2025, PhyCARE 27 January 2026, CARE-DN 18 January 2026) keep a title rule unless the journal says otherwise.

Is a case study qualitative or quantitative?

Either, or mixed. Yin is explicit that the method is not a synonym for qualitative research. Embedded surveys, budgets, and time series can sit inside a case. The design is defined by the bounded case and the how/why question, not by the absence of numbers.

Can I generalize from a single case?

You can generalize to theory (analytic generalization). You cannot treat one site as a random sample of a population. Yin's refresher uses Allison (1971): generalize the crisis-decision models, not U.S.–Cuban relations. Flyvbjerg (2006) is the paper to cite when a reviewer recites the "n = 1" objection.

What citation style should a case study use?

Whatever the course or journal requires. Psychology and education usually want APA 7. History often wants Chicago notes. Clinical case reports follow AMA or the journal's Vancouver/NLM house style. The which-style guide helps you choose; this page does not.

How to Write a Case Study (Yin 6th, HBS, CARE) [2026]

How to Write a Case Study: Quick Answer

Name the assignment first. If you designed the inquiry, write a research case study: bound the case, state a how or why question, collect at least two kinds of evidence, and analyze before you narrate. If the instructor handed you a packet, write a teaching-case analysis: decide what you would do with incomplete information. If you are reporting one patient's course for a journal, follow the 2013 CARE checklist and obtain consent. Do not invent a school, a firm, or a patient to pad the page.

Templates by assignment (not invented sources)

Research case-study sentence:

This single-case study asks how [bounded case] produced [outcome] in [place and dates], using interviews, documents, or observation that I triangulate against a rival explanation.

Teaching-case analysis sentence:

Standing in the protagonist's place at the decision point in [assigned case], I recommend [action] because of evidence from the packet, and I would measure success by [criterion].

CARE title template:

[Diagnosis or intervention] … case report

Yin's when-to-use test:

How/why question + little control over events + a contemporary phenomenon still tangled with its context.

Yin restates the definition in a SAGE methods refresher drawn from Case Study Research (2009a, p. 18). The wording below is his. The same refresher opens by warning that its procedures — and the rest of the handbook — do not cover teaching-case development.

Yin, abbreviated definition (SAGE refresher, citing 2009a, p. 18)

An empirical inquiry about a contemporary phenomenon (e.g., a “case”), set within its real-world context—especially when the boundaries between phenomenon and context are not clearly evident.

Yin, research method versus teaching cases (SAGE refresher)

None of the tasks, nor the rest of this book, deals with the development of teaching case studies—frequently also referred to as the “case study method”—the pedagogical goals of which may differ entirely from doing research studies.

Which Assignment You Were Actually Given

USC keeps Yin's split on two pages: Writing a Case Study versus Writing a Case Analysis Paper. Medicine adds a third genre through the CARE guidelines (CAse REports), still the 2013 13-item checklist on EQUATOR (record last updated 16 February 2026). Mixing the three is the most common reason a draft fails.

Three assignments that share the words case study
Assignment	Who creates the case	Your job on the page	Authority
Research case study	You bound and study a real case	Answer a how/why question with triangulated evidence and analytic generalization	Yin 6th ed.; Stake 1995
Teaching-case analysis	The instructor or a publisher wrote the packet	Recommend a decision from incomplete information and defend it	HBS case method; USC case-analysis guide; Ellet 2018
Medical case report	You treated or observed one patient	Report the unique course, work-up, and take-away with consent	CARE 2013 checklist

If the prompt says "analyze the case we assigned," you are not designing Yin-style fieldwork. If it says "conduct a case study of X," you are. If it says "write a case report" in a clinical course, open CARE before you open Yin.

What Counts as a Research Case Study

USC's research-assignment guide defines the paper you are trying to write. The sentence is theirs.

USC Libraries, Writing a Case Study

A case study research paper examines a person, place, event, condition, phenomenon, or other type of subject of analysis in order to extrapolate key themes and results that help predict future trends, illuminate previously hidden issues that can be applied to practice, and/or provide a means for understanding an important research problem with greater clarity.

The case is a bounded entity: a person, organization, event, program, neighborhood, or decision, with spatial and temporal edges you can defend. Yin's refresher calls it "generally a bounded entity (a person, organization, behavioral condition, event, or other social phenomenon)" whose spatial and temporal edges may still be blurred — which is why context is data, not scenery. A survey that asks 400 people the same five items is not a case study. An experiment that assigns treatments is not a case study. A literature review that never leaves published papers is not a case study. Those designs belong in the methodology section of a different paper.

When a Case Study Is the Right Method

Yin gives three situations. The first is the research question. His refresher puts the rule in this wording:

Yin, when the method fits (SAGE refresher)

case studies are pertinent when your research addresses either a descriptive question—“What is happening or has happened?”—or an explanatory question—“How or why did something happen?”

Little or no control over events. You cannot assign the intervention or randomize the organization. If you can, an experiment or quasi-experiment is usually stronger for a what-works question.
A contemporary phenomenon. Histories reconstruct the finished past from documents. Case studies can still use archives, but they also collect interviews, observation, and current records while the context is live.
Context cannot be stripped off. If the answer changes when you remove the school, the ward, the city, or the year, you need a case study rather than a decontextualized variable model.

Yin contrasts the leftover questions: an initiative's effectiveness in producing a particular outcome is for experiments; how often something happened is for surveys. Use those methods when they fit. Use a case study when you need the rich description or the mechanism they will not give you. The SAGE 6th edition sharpens the same point by adding eleven applications — Application #1 is an exploratory case on how new organizational practices become routinized — so readers can see the method used, not only defined.

Types You Must Name in the Methods Section

Yin's purpose: descriptive, exploratory, explanatory

Name the purpose in one sentence so readers know what kind of claim you are allowed to make. Do not call an exploratory sketch explanatory just because the word sounds stronger.

Yin purposes with a real published example or an explicit template
Purpose	What you owe the reader	Example
Descriptive	A detailed account of what occurred inside the bound case	Walo, Bull, and Breen (1996) describe the direct and indirect economic costs and benefits of a local sports event in Lismore, New South Wales — a typical event, not a mega-event (USC cites this study).
Exploratory	Questions or propositions for a later study, not a causal verdict	Template (not a published study): a first look at a new clinic workflow before a theory is worth testing. Yin's 6th-edition Application #1 is the published counterpart: how new organizational practices become routinized.
Explanatory	How or why the outcome happened, with rival explanations entertained	Allison's Essence of Decision (1971) explains the Cuban Missile Crisis through competing models of governmental action. Yin's refresher uses that book as the classic analytic-generalization example.

Stake's purpose: intrinsic, instrumental, collective

Robert E. Stake, The Art of Case Study Research (SAGE, 1995), sorts cases by why you care about this one. An intrinsic case is studied because the case itself is of special interest. An instrumental case is studied to understand something else — a theory, a policy, a disease course. A collective case study coordinates several instrumental cases. Stake also used and annotated a real school case, Harper School, to show selection, interpretation, and reporting. If your only reason for picking the site is "it was nearby," you do not yet have a Stake purpose.

Yin's design matrix: single or multiple, holistic or embedded

Decide two more things before you write. One case or several? Holistic (the case is the only unit) or embedded (you also collect data on nested units, such as employees inside one firm)? Multiple cases follow replication logic, not sampling logic: pick the next case because you predict a similar result (direct replication) or a contrast for a stated reason (theoretical replication). The refresher is blunt that there is no power-analysis formula for "how many cases." "I interviewed five friends" is not a multiple-case design.

How to Select the Case

USC, citing Seawright and Gerring (2008), says practical access cannot be the only justification. Yin names the usual strategies: a typical case, a critical case that can test a claim, an extreme or deviant case, or a revelatory case that opens a setting others could not enter. Flyvbjerg's correction of the "no generalization" myth depends on that strategic choice. USC's own atypical-case example is Azerbaijan rather than a more obvious Middle Eastern nation if the problem is girls' access to secondary education in predominantly Muslim countries — a selection rationale, not a finished study you should invent details for.

Real selection decisions, not invented vignettes

Typical case (Walo, Bull, and Breen 1996):

USC summarizes the choice: the authors studied a local sports event in Lismore, Australia, because prior work had measured mega-events and could not speak to ordinary local events.

Critical case (Goldthorpe, Lockwood, Bechhofer, and Platt, the affluent-worker study):

Flyvbjerg (2006) recounts the logic: Luton was chosen as fertile ground for the claim that high wages were dissolving working-class identity. If the claim failed in a favorable case, it was unlikely to hold in intermediate cases.

Teaching case (HBS inaugural case, 1921):

Baker Library's exhibit on Clinton P. Biddle's The General Shoe Company (1921) is the inaugural HBS case: a one-page labor problem in a hypothetical shoe plant, written for class, not a Yin field study of a real firm.

Write the selection rationale in the methods section: why this case, why not the more obvious neighbor, and what you will not claim. USC's place example is the right shape: if you study homeless encampments of veterans, say why Echo Park rather than Griffith Park.

Structure of a Research Case Study

USC says the skeleton looks like a college research paper, with two differences: the introduction must justify the case, and findings often share a section with discussion. The introduction has to answer four questions. The wording is USC's.

USC Libraries, four questions the introduction must answer

Who or what is being studied?

Why is this topic important to investigate?

What did we know about this topic before I did this study?

How will this study advance new knowledge or new ways of understanding?

Research case-study sections and what each one must do
Section	Job	Failure mode
Introduction	Link the research problem to this case and preview why the design fits	A topic paragraph that never names the bound case
Literature review	Show what prior studies of similar cases missed, including conflicts you can settle	A generic topic review that could sit on any paper
Method	Define the case, the selection logic, the sources, and the analytic strategy	"I used qualitative methods" with no chain of evidence
Case description	Enough context and chronology that a stranger can follow the action	Either a data dump or a story with no dates, actors, or documents
Analysis / findings	Patterns, mechanisms, and quotes tied back to the questions	Description only; the "so what" never arrives
Discussion	Meaning, rivals, limits, and what transfers analytically	A leap from one school to "all schools"
Conclusion	A synthesis, not a second discussion. Restate the claim the case can support	New evidence that should have appeared earlier

Keep the literature review aimed at the case. USC's test is whether the review explains why this subject of analysis can fill a gap. If a citation would survive after you swapped the case for a different city, it probably belongs in a different paper.

How to Write a Teaching-Case Analysis

Harvard Business School's MBA case-method page describes the packet you are analyzing and the four-step classroom process (read, discuss, engage, reflect). The sentences below are HBS's. The live page no longer publishes a case-count for the two-year MBA; do not invent one.

Harvard Business School, How the Case Method Works

Each case is a 10-20 page document written from the viewpoint of a real person leading a real organization.

In addition to background information on the situation, each case ends in a key decision to be made. Your job is to sift through the information, incomplete by design, and decide what you would do.

There’s no right answer, but you have to confront the complexities of the choice you make.

The case method prepares you to be in leadership positions where you will face time-sensitive decisions with limited information.

The inaugural packet is older and shorter than that 10–20 page modern form. Baker Library still exhibits Clinton P. Biddle's 1921 General Shoe Company: workers leaving a hypothetical plant 45 minutes early while orders piled up. Jan Rivkin's line on that exhibit — also quoted on the HBS case-method page — is the skill the analysis paper is testing.

Jan Rivkin on General Shoe (HBS / Baker Library)

How do you go into an ambiguous situation and get to the bottom of it? That skill – the skill of figuring out a course of inquiry, to choose a course of action – that skill is as relevant today as it was in 1921.

USC's case-analysis page is the writing counterpart. Its definition is the one instructors mean when they say "case study" in a business, social-work, or public-administration course and then hand you a packet. USC also points students to William Ellet, The Case Study Handbook: A Student's Guide, revised edition (Harvard Business School Publishing, 2018).

USC Libraries, Writing a Case Analysis Paper

Case analysis is a problem-based teaching and learning method that involves critically analyzing complex scenarios within an organizational setting for the purpose of placing the student in a “real world” situation and applying reflection and critical thinking skills to contemplate appropriate solutions, decisions, or recommended courses of action.

USC's writing tip, "Case Study and Case Analysis Are Not the Same!," lists twelve differences. Four of them are the ones that still fail drafts.

USC Libraries, four of the twelve case-study versus case-analysis differences

Case study is a method of in-depth research and rigorous inquiry; case analysis is a reliable method of teaching and learning.

The researcher is responsible for identifying the case to study; a case analysis is assigned by your professor.

Case study is fact-based and describes actual events or situations; case analysis can be entirely fictional or adapted from an actual situation.

A case analysis assignment rarely has a single correct answer because one of the goals is to force students to confront the real life dynamics of uncertainly, ambiguity, and missing or conflicting information within professional practice.

That last USC sentence keeps the page's own spelling ("uncertainly"). Treat it as a reminder that teaching packets may include red herrings. Biddle's General Shoe is the historical proof: a one-page hypothetical plant can still train the same decision skill as a modern 10–20 page case about a real organization.

Read once for the situation, then again for contradictions. USC's first two steps: get the overview, then reread for circumstances, stakeholder behavior, and data that look misleading.
Name stakeholders, issues, and events, and how they connect. Separate symptoms from core problems. A late shipment can be a symptom; a broken incentive system can be the problem.
Generate more than one option. HBS is explicit that there is no single right answer. USC says you must weigh short-term versus long-term effects and fairness.
Recommend one course and say how you will know it worked. Criteria and measurable indicators belong in the paper, not only in the last slide.
Stay inside the packet plus assigned readings. Do not "solve" the case with facts you looked up after the decision date unless the instructor asked for outside research.

How to Write a Medical Case Report (CARE)

A clinical case report is not a social-science case study. The CARE guidelines are a 13-item consensus checklist first published in 2013 (Gagnier et al., simultaneous in seven journals, including Journal of Medical Case Reports 7:223) and explained in Riley et al., Journal of Clinical Epidemiology 89 (2017): 218–235. The EQUATOR Network still lists that 2013 checklist as the reporting guideline (record last updated 16 February 2026). The same EQUATOR record now also lists specialty extensions: CARE-radiology (Wang et al., 2024), CARE-acupuncture (Duan et al., 2025), PhyCARE for physiotherapy case reports (Naqvi et al., BMJ Open, 27 January 2026), and CARE-DN for dry-needling case reports (Tzigkounakis et al., Cureus, 18 January 2026). Use an extension only if the target journal asks for it. CARE-writer is the project's own online organizer for preprints or journal submission. The title rule, the uniqueness test, and consent are CARE's wording.

CARE homepage, why a case report is worth writing (quoting Vandenbroucke 2001)

Good case reporting demands a clear focus, to make explicit to the audience why a particular observation is important in the context of existing knowledge

2013 CARE Checklist (official wording from care-statement.org/checklist)

The diagnosis or intervention of primary focus followed by the words “case report”.

What is unique about this case and what does it add to the scientific literature?

The primary “take-away” lessons from this case report (without references) in a one paragraph conclusion.

The patient should give informed consent. (Provide if requested.)

The CARE homepage cites Driggers 2016 as a clinical case that generated a testable hypothesis. Rita W. Driggers and colleagues reported a pregnant woman and fetus infected with Zika virus in the 11th gestational week: maternal ZIKV RNA was still detectable at 16 and 21 weeks, and fetal brain abnormalities appeared on ultrasound and MRI before microcephaly or intracranial calcifications. That is a real published case report, not a composite patient.

Real published case report (CARE homepage example)

Driggers et al., NEJM 2016:

Driggers, R. W., Ho, C.-Y., Korhonen, E. M., Kuivanen, S., Jääskeläinen, A. J., Smura, T., Rosenberg, A. Z., Hill, D. A., DeBiasi, R. L., Vézina, G., Timofeev, J., Rodríguez, F. J., Levanov, L., Razak, J., Iyengar, P., Hennenfent, A. K., Kennedy, R. D., Lanciotti, R. S., du Plessis, A., & Vapalahti, O. (2016). Zika virus infection with prolonged maternal viremia and fetal brain abnormalities. New England Journal of Medicine, 374(22), 2142–2151. https://doi.org/10.1056/NEJMoa1601824

Title and keywords. End the title with "case report" and include that phrase among 2–5 keywords.
Abstract. CARE wants background, key points from the case, and the main lessons — written after the report, not before.
Patient information and clinical findings. De-identify. Report concerns, history, relevant past interventions, and the physical examination.
Timeline. A table or figure of this episode of care. Do not bury dates in prose only.
Diagnostic assessment and therapeutic intervention. Methods, challenges, other diagnoses considered, then type, dose, duration, and changes.
Follow-up, discussion, and patient perspective. Outcomes, adherence, adverse events, strengths and limits, the take-away paragraph without references, and the patient's view of treatment when you can obtain it.

Do not invent a rare presentation. If you do not have a real patient and consent, you do not have a case report. For citation of the clinical literature around the case, use the medical citation guide (AMA 11 / ICMJE / NLM).

Collect Evidence From More Than One Source

Yin's design requires multiple sources that converge. One long interview is a source, not a case study. The SAGE refresher lists six common sources (Exhibit 1.1). Name each source in the methods section and show, in analysis, where two sources agree or clash.

Yin, six common sources of evidence (SAGE refresher, Exhibit 1.1)

Direct observations (e.g., human actions or a physical environment)

Interviews (e.g., open-ended conversations with key participants)

Archival records (e.g., student records)

Documents (e.g., newspaper articles, letters and e-mails, reports)

Participant-observation (e.g., being identified as a researcher but also filling a real-life role in the scene being studied)

Physical artifacts (e.g., computer downloads of employees’ work)

Interviews. Semi-structured conversations with people who were there. Recoverable published interviews are cited; unpublished personal interviews are often in-text only — see citing interviews in APA 7 and personal communications.
Documents. Reports, emails, minutes, policies, lesson plans, electronic health-record extracts you are allowed to use. Quote the document; do not paraphrase a memory of it.
Direct observation. Field notes with date, place, and what you could actually see.
Physical or digital artifacts. Products, screens, layouts, tools.
Archival records. Enrollment files, budgets, incident logs, census tables.
Optional surveys. Embedded quantitative data can sit inside a case. They do not turn the study into a survey.

Keep a case-study database Yin would recognize: raw notes, documents, and a chain of evidence from a claim in the paper back to a dated source. If a reviewer asks "where did this number come from?" you should be able to point to a file, not a feeling. The refresher also wants a protocol whose questions are aimed at you, the researcher — a mental framework, not a verbatim script you read to participants.

Yin, case-study protocol (SAGE refresher)

Importantly, the questions in the protocol are directed at the researcher, not at any field participant.

Analyze the Case; Do Not Only Narrate It

The old student failure mode is still the main one: ten pages of story and half a page of "this shows that leadership matters." Yin’s analytic strategies are the fix. Use at least one and name it.

Pattern matching. Compare what the data show with a pattern you predicted from theory before you fell in love with the story. The refresher's community example is Bradshaw (1999): predicted "catastrophic" pre–post trends after a military-base closure, then compared those sectors with rival towns. The 6th edition reprints that comparison as Application #7.
Explanation building. Iterate an explanation of how or why, then test it against the next document or interview.
Time-series analysis. Track a few indicators across the bound dates. A timeline is not automatically an analysis; the analysis is what changed and what did not.
Cross-case synthesis. Only if you actually designed a multiple-case study. Do not "synthesize" five news stories you found last night.
Rival explanations. Write the best competing account in good faith. If you cannot state a rival, you are still describing.

Yin, true rivals (SAGE refresher)

True rivals compete directly with each other and cannot coexist.

When you quote participants, keep the quote short and spend the next sentence saying what it is evidence of. Paraphrase when the wording is not the point. Present numbers and timelines the way a reader can check — see how to present data.

The Four Quality Tests Yin Expects

Yin's refresher points to construct, internal, external, and reliability concerns and sends readers to the handbook (2009a, pp. 40–45; the 6th edition keeps the same four tests). Treat them as writing jobs, not slogans.

Yin's four tests, as they appear on the page
Test	Question a skeptic asks	What you put in the paper
Construct validity	Are you measuring the thing you named?	Multiple sources, a chain of evidence, and (when ethical) informant review of the draft
Internal validity	Could something else explain this?	Pattern matching, explanation building, and written rivals — explanatory cases only
External validity	Why would this matter beyond the case?	Theory in the design, replication logic, and analytic — not statistical — generalization
Reliability	Could another researcher follow your steps?	A protocol, a case-study database, and an audit trail

Can You Generalize From One Case?

Yes, but not the way a random sample does. Yin calls it analytic generalization: the case speaks to a theory, not to a population count. His refresher's worked example is Allison (1971; later Allison and Zelikow, 1999). The wording below is Yin's.

Yin, analytic generalization (SAGE refresher, on Allison 1971)

The authors do not generalize their findings and theoretical framework to U.S.-Cuban relations—or to the use of missiles. They use their theoretical propositions to generalize their findings to the likely responses of national governments when involved in superpower confrontation and international crises.

Bent Flyvbjerg, "Five Misunderstandings About Case-Study Research," Qualitative Inquiry 12, no. 2 (2006): 219–245, lists the five textbook errors. USC reprints them. Quote Flyvbjerg's own numbering, then his correction of the second misunderstanding — the one your discussion section has to survive.

Flyvbjerg (2006), Qualitative Inquiry 12(2), the five misunderstandings

Misunderstanding 1: General, theoretical (context-independent) knowledge is more valuable than concrete, practical (context-dependent) knowledge.

Misunderstanding 2: One cannot generalize on the basis of an individual case; therefore, the case study cannot contribute to scientific development.

Misunderstanding 3: The case study is most useful for generating hypotheses; that is, in the first stage of a total research process, whereas other methods are more suitable for hypotheses testing and theory building.

Misunderstanding 4: The case study contains a bias toward verification, that is, a tendency to confirm the researcher’s preconceived notions.

Misunderstanding 5: It is often difficult to summarize and develop general propositions and theories on the basis of specific case studies.

Flyvbjerg (2006), correction of misunderstanding 2

One can often generalize on the basis of a single case, and the case study may be central to scientific development via generalization as supplement or alternative to other methods. But formal generalization is overvalued as a source of scientific development, whereas "the force of example" is underestimated.

Write the transfer you are making ("this mechanism should appear when the same incentive is present") and the transfer you are not ("this is not a rate for all districts"). Flyvbjerg allows analytic generalization and the force of example. He does not allow "therefore 80% of hospitals."

Ethics You Cannot Leave for the Appendix

Informed consent. Participants should know what you will publish. CARE requires patient consent for case reports; many journals will ask for the form.
De-identification. Pseudonyms, shifted minor details, and no recoverable record numbers. Do not promise anonymity if a principal can still be identified from the town plus the job title.
IRB or ethics review. Required for most studies of living people. Teaching-case analyses of published HBS packets usually are not human-subjects research; your own interviews are.
Member checking. Yin lists informant review as a construct-validity tactic. Offer a draft of their quotations, not a veto over your analysis.
Harm. A published case can cost someone a job or a diagnosis. If the insight requires exposing a person, you do not yet have a publishable case.

How to Cite the Sources Inside a Case Study

Every recoverable document, article, and published interview belongs on the end list your style names. APA 7 titles that page References; MLA 9 titles it Works Cited; Chicago notes use a Bibliography. See how to write a bibliography. Field interviews you cannot recover stay out of an APA reference list and appear as personal communications. Do not invent a DOI or a docket number to make a source look official.

Real works cited on this page (APA 7 shape)

Yin 6th edition:

Yin, R. K. (2018). Case study research and applications: Design and methods (6th ed.). SAGE.

Stake 1995:

Stake, R. E. (1995). The art of case study research. SAGE.

Flyvbjerg 2006:

Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative Inquiry, 12(2), 219–245. https://doi.org/10.1177/1077800405284363

Walo, Bull, and Breen 1996 (as USC cites it):

Walo, M., Bull, A., & Breen, H. (1996). Achieving economic benefits at local events: A case study of a local sports event. Festival Management and Event Tourism, 4, 95–106.

Allison 1971:

Allison, G. T. (1971). Essence of decision: Explaining the Cuban Missile Crisis. Little, Brown.

Ellet 2018 (teaching-case handbook USC cites):

Ellet, W. (2018). The case study handbook: A student's guide (Rev. ed.). Harvard Business School Publishing.

Driggers et al. 2016 (CARE homepage example):

Format the recoverable sources

Once the analysis is honest, generate the end-list entries and check them against the handbook. Interviews, reports, and journal articles use different slots.

Open the citation generator How to write the bibliography

Failure Modes That Still Show Up in Drafts

You wrote a teaching-case memo as if it were fieldwork. If the only "data" is the assigned packet, use the HBS/USC analysis structure and do not claim triangulation.
You invented a vivid school or patient. An earlier version of this page used a fictional Lincoln Elementary mentoring program. That is fiction, not a case. Use Walo et al., Allison, Luton, General Shoe, Driggers, or a labelled template.
You treated General Shoe as a Yin field study of a real firm. Baker Library is explicit: Biddle's 1921 case is a one-page hypothetical plant. Modern HBS cases are usually 10–20 pages about a real organization. Both are teaching packets, not research you conducted.
You cited a live case-count for the HBS MBA. The current HBS case-method page does not publish one. Do not restore an old "500 cases" figure.
Description without analysis. For every page of narrative, write analysis that answers how or why. Cut the colorful detail that does not earn a claim.
Cherry-picking. Actively collect the email or interview that ruins your favorite explanation, then account for it.
Statistical overclaim. One case does not yield a prevalence. Flyvbjerg allows analytic generalization; he does not allow "therefore 80% of hospitals."
Unbounded case. "Technology in education" is a topic. "How School X implemented a 1:1 laptop program from August 2023 to June 2024" is a case.
Missing ethics. No consent, no de-identification plan, no statement of your role. Readers should know whether you were the consultant who designed the program you then "found" succeeded.

Case Study Checklist

Before you submit

The assignment type is named: research case study, teaching-case analysis, or CARE report
The case is bounded in place and time
A how, why, or (for CARE) uniqueness question is stated
Case selection is justified with a named strategy
Research cases use at least two of Yin's six evidence sources that can converge
An analytic strategy is named (pattern matching, explanation building, rivals)
Teaching-case papers recommend a decision and a success measure
CARE reports include the words case report, a timeline, and consent
Limitations and rival explanations are written down
Transfer claims are analytic, not a fake sample percentage
Participants are de-identified; recoverable sources are cited
No invented organizations, patients, or quotations

Case study FAQs

What is the difference between a case study and a case analysis?: A research case study is an inquiry you design and carry out on a bounded real case (Yin; USC's case-study page). A case analysis is a problem-based assignment on a packet someone else wrote, usually in the HBS tradition (USC's case-analysis page). USC's twelve-point tip is blunt: the researcher picks the case; the professor assigns the analysis. If you did not collect evidence outside the handout, you are writing an analysis.
How long should a case study be?: Long enough to bound the case, show the evidence, and finish the analysis — not long enough to dump every note. Course papers often run 2,000–5,000 words. A Yin research report can be article length. HBS says a modern teaching case is typically 10–20 pages before your analysis; Biddle's 1921 General Shoe case was one page. CARE reports follow the journal's limit, not a word-count myth.
Can I write a case study with one interview?: You can write an interview study. Yin’s method wants multiple sources that converge. The SAGE refresher lists six common sources; one interview can be a source inside a case. It is not, by itself, a completed case study.
Do I need a hypothesis?: Exploratory and intrinsic cases can start with questions. Explanatory cases need theoretical propositions or predicted patterns so pattern matching is possible. Teaching-case analyses need a decision, not a null hypothesis. CARE reports need a uniqueness claim, not a p-value.
How many cases do I need?: One is enough if the selection logic is strong (a critical, extreme, or revelatory case). Add cases only for stated replications. Yin's refresher rejects a power-analysis formula for case counts: more cases raise confidence, but you do not tally "five supported the proposition" the way you would tally a sample.
How do I cite an unpublished interview in the case study?: In APA 7 a personal interview that a reader cannot recover is cited in the text only, not on the reference list. A published or archived interview is cited as the container (podcast, article, transcript). See cite an interview in APA.
What belongs in a medical case report title?: CARE item 1 on the 2013 checklist: the diagnosis or intervention of primary focus, followed by the words case report. Do not write a witty title that hides the condition. Specialty extensions listed on EQUATOR as of 16 February 2026 (CARE-radiology 2024, CARE-acupuncture 2025, PhyCARE 27 January 2026, CARE-DN 18 January 2026) keep a title rule unless the journal says otherwise.
Is a case study qualitative or quantitative?: Either, or mixed. Yin is explicit that the method is not a synonym for qualitative research. Embedded surveys, budgets, and time series can sit inside a case. The design is defined by the bounded case and the how/why question, not by the absence of numbers.
Can I generalize from a single case?: You can generalize to theory (analytic generalization). You cannot treat one site as a random sample of a population. Yin's refresher uses Allison (1971): generalize the crisis-decision models, not U.S.–Cuban relations. Flyvbjerg (2006) is the paper to cite when a reviewer recites the "n = 1" objection.
What citation style should a case study use?: Whatever the course or journal requires. Psychology and education usually want APA 7. History often wants Chicago notes. Clinical case reports follow AMA or the journal's Vancouver/NLM house style. The which-style guide helps you choose; this page does not.