A case study paper follows a different structural logic than a standard empirical article — it is organized around bounding, describing, and analyzing a real-world instance in depth, not around a single hypothesis test. Case study research in the social sciences (political science, sociology, education, public administration, management, and related fields) has its own methodological literature, led by Robert K. Yin’s Case Study Research: Design and Methods and Robert Stake’s work on intrinsic and instrumental case studies, and journals in these fields expect a paper to show that lineage in how it is organized. This guide walks through that structure section by section. For the discipline-general empirical skeleton, see CASRAI’s IMRaD Structure entry and the IMRaD Structure Variations by Academic Discipline guide — this page covers where case study papers follow that pattern and where they depart from it.
What makes a case study paper structurally different
Yin defines 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 its context is not clearly evident. That definition has two direct structural consequences that a standard IMRaD paper does not need to handle:
- The case itself has to be defined and bounded before findings can be presented — readers need to know what the “case” is (an organization, a policy episode, a community, a single individual’s trajectory), and where its boundaries in time, place, and activity sit. A methods section that never states this is a common and diagnosable weakness.
- The case selection itself has to be justified as a methodological choice, not just described. Why this case, and not another, given the research question? A case study paper that skips this reads as if the case was chosen by convenience rather than design, even when it wasn’t.
A case study research paper is also a distinct genre from a clinical or biomedical case report, which typically describes a single unusual patient presentation without a generalizable research question or an explicit methodology section. CASRAI’s Case Study vs. Case Report comparison sets out that distinction in detail; this guide is about the social-science research case study, which is closer in ambition to a full empirical study than to a descriptive report.
Decide and justify the case study design before drafting
The methodology section of a case study paper has to answer design questions that an IMRaD methods section doesn’t face in the same form. Deciding these before drafting, rather than backfilling them into a written-up methods section, produces a cleaner paper:
- Purpose type (Yin): is the case study exploratory (developing hypotheses or a framework for further study), descriptive (describing a phenomenon and its context in depth), or explanatory (testing or building a causal explanation for an outcome)? State this explicitly — it shapes what the discussion section is allowed to claim.
- Single-case vs. multiple-case design: a single case is justified when it is critical, extreme/unusual, typical/representative, revelatory, or longitudinal (Yin’s single-case rationales). A multiple-case (comparative) design instead follows a replication logic — each additional case is chosen to either confirm the same pattern (literal replication) or produce a contrasting result for predictable reasons (theoretical replication), not simply to add more data points.
- Holistic vs. embedded design: is there a single unit of analysis (the whole organization, the whole policy), or multiple embedded sub-units within the case (several departments within one organization, several implementation sites within one program)?
- Intrinsic vs. instrumental purpose (Stake): is the case worth studying for its own sake (intrinsic), or chosen primarily because it illuminates a broader issue or theoretical question (instrumental)? A collective case study extends the instrumental logic across several cases studied jointly to understand a phenomenon.
The core sections, in order
1. Title and abstract
The title should signal both the case and the analytical question — not just name the site or organization studied. Most journals in this space expect a structured or semi-structured abstract stating the research question, the case and why it was selected, the data sources, and the central finding, typically in 200–250 words.
2. Introduction
States the research question, situates it in the relevant body/policy problem, and — distinct from a standard IMRaD introduction — explicitly justifies why a case study approach fits this question rather than a survey, experiment, or large-N comparative design. Reviewers in this tradition specifically look for this justification; its absence is one of the most common desk-reject reasons for case study submissions.
3. Literature review / theoretical or conceptual framework
Case study papers usually foreground the theoretical framework more explicitly than a standard IMRaD paper does, because the framework is what the case is ultimately read against in the discussion. This section should make clear what existing theoretical propositions or conceptual categories will structure the data collection and analysis — Yin’s approach treats these propositions as central to the whole design, not just background reading.
4. Methodology
This is the section reviewers scrutinize hardest in case study submissions. It should cover, in this order:
- Design type and rationale — the choices from the section above, stated explicitly.
- Case selection and bounding — why this case (or these cases), and the case’s boundaries in time, place, and activity.
- Data sources and triangulation — case study research typically draws on multiple data sources (semi-structured interviews, documents and archival records, direct or participant observation, and sometimes quantitative data) specifically so findings can be corroborated across sources rather than resting on one. State what was collected and how the sources were combined.
- Analysis approach — common analytic strategies include pattern matching (comparing an empirically observed pattern against a predicted one), explanation building (iteratively refining an explanation across the case evidence), time-series analysis for process-tracing designs, and cross-case synthesis for multiple-case designs. Name the strategy actually used; “thematic analysis” alone is often too generic for reviewers who expect a case-study-specific analytic logic. See CASRAI’s qualitative research methodology section guide for how to write this section when the underlying data collection is interview- or document-based, and the narrative analysis guide where the case is built primarily from narrative accounts.
- Rigor and trustworthiness — qualitative case study papers commonly report against Lincoln and Guba’s four criteria (credibility, transferability, dependability, confirmability) in place of the quantitative validity/reliability language a reviewer might otherwise expect, and increasingly state a positionality or reflexivity statement addressing the researcher’s relationship to the case.
- Ethics — where the case involves human participants, this section (or a linked ethics statement) should cover consent, anonymization of the site/participants where confidentiality was promised, and IRB/ethics board approval. See CASRAI’s research ethics in qualitative research guide.
5. Case description / context
A section (sometimes folded into the start of Findings) that orients the reader to the case itself before analysis begins — the organization’s history, the policy’s timeline, the community’s demographics, whatever is needed to understand what follows. This is where case study papers commonly use thick description (a term from anthropologist Clifford Geertz, now standard vocabulary in qualitative methods) — enough contextual and sensory detail that a reader unfamiliar with the setting can follow the analysis, not a data dump of every observation collected.
6. Findings / analysis
For a single-case design, this is typically organized around the theoretical propositions or research sub-questions set out earlier, not chronologically or by data source. For a multiple-case design, most journals expect within-case analysis first (each case presented and analyzed on its own terms) followed by a distinct cross-case synthesis section that explicitly compares patterns across cases — collapsing straight to cross-case comparison without ever presenting each case on its own is a common structural weakness reviewers flag.
7. Discussion
Connects the findings back to the theoretical framework from section 3, and is where the paper should explicitly address what kind of generalization it is claiming. Case study findings do not generalize to a population in the statistical sense a survey or experiment does; Yin’s term for what they do offer is analytic generalization — generalization to theoretical propositions, not to other cases or populations. Stake’s parallel term is naturalistic generalization, generalization a reader makes based on recognizing similarities to their own experience. A discussion section that overstates what a single case can support (implying representativeness it doesn’t have) is one of the most common reasons case study papers draw skeptical review, particularly from reviewers trained in quantitative traditions.
8. Limitations
Beyond generic limitations, case study papers should specifically address transferability (what about this case is or isn’t likely to hold elsewhere), any access constraints that shaped what data could be collected, and rival explanations that were considered and ruled out — Yin’s recommended practice of explicitly addressing rival explanations, rather than presenting only the favored interpretation, strengthens the paper’s claim to explanatory rigor.
9. Conclusion, declarations, and references
A short conclusion restating the contribution, followed by standard declarations (funding, conflicts of interest, data availability, ethics approval) and references. Where the case site or participants were promised confidentiality, confirm the final manuscript doesn’t inadvertently re-identify them through overly specific contextual detail in the case description section.
Reporting checklists reviewers may expect
Where a case study paper is built primarily on interviews or focus groups, some journals and reviewers expect it to follow, or at least reference, the COREQ (Consolidated Criteria for Reporting Qualitative Research) checklist — a 32-item checklist covering the research team, study design, and analysis/findings reporting, published in the International Journal for Quality in Health Care in 2007 and widely adopted well beyond health research as a general qualitative-reporting standard. It is not case-study-specific, but completing and citing it (or a discipline-specific equivalent) is an efficient way to preempt several of the methodology-section gaps described above.
How the structure shifts across social-science fields
- Political science and international relations: case studies are frequently built around process tracing — tracing the causal mechanism linking a hypothesized cause to an outcome step by step — which typically produces a more chronologically organized findings section than the proposition-organized structure described above.
- Sociology and anthropology: case studies drawing on ethnographic fieldwork lean more heavily on thick description and often integrate findings and discussion rather than separating them strictly, reflecting the discipline’s narrative conventions.
- Business, management, and public administration: closest to Yin’s original template, with an explicit methodology section, named propositions, and (for multiple-case designs) a formal cross-case synthesis table.
- Education: often follows Sharan Merriam’s qualitative case study tradition, which places more emphasis on the case as a bounded system studied for the meaning participants make of their experience within it.
Common structural mistakes
- Never explicitly stating the case’s boundaries (time, place, activity), leaving readers to infer what is and isn’t part of the case.
- Describing the case selection as convenience without addressing why the case is theoretically appropriate to the research question.
- Skipping straight to findings organized by data source (“what the interviews said,” “what the documents showed”) instead of by theoretical proposition or research sub-question.
- For multiple-case papers, presenting only a cross-case table with no within-case narrative to support it.
- Claiming statistical generalizability for a single-case design instead of the analytic/naturalistic generalization the method actually supports.
- Omitting a rival-explanations discussion, leaving the paper’s causal claims looking asserted rather than tested.
Frequently asked questions
How is a case study paper different from a case report?
A case report (most common in clinical and biomedical fields) describes a single unusual instance, typically without an explicit research question, theoretical framework, or generalizable methodology. A social-science case study paper is a full research design with a stated research question, a justified case-selection rationale, systematic data collection and triangulation, and an explicit analytic framework. See CASRAI’s Case Study vs. Case Report comparison.
How many cases should a case study paper include?
There is no fixed number. Single-case designs are justified on specific grounds (critical, extreme, typical, revelatory, or longitudinal cases); multiple-case designs add cases for literal or theoretical replication, not simply for volume. Journals rarely specify a minimum — what they expect is an explicit rationale for the number chosen.
Does a case study paper need a separate literature review section?
Almost always, yes, though it is often framed as a “theoretical framework” or “conceptual framework” section rather than a standalone literature review, because in Yin’s design tradition the framework directly shapes the propositions the case is analyzed against.
Can a case study paper use quantitative data?
Yes. Case study research is a research design, not a synonym for qualitative research — it can and often does combine qualitative and quantitative data sources (survey data, administrative records, financial statements) as part of triangulation, particularly in business, management, and public-policy case studies.
What does “analytic generalization” mean, and why does it matter for structure?
It is Yin’s term for generalizing case study findings to theoretical propositions rather than to a population, in contrast to the statistical generalization a survey or experiment supports. It matters structurally because the discussion section should frame the contribution in these terms explicitly — claiming population-level generalizability from a case study is a common and avoidable reviewer objection.
For the broader scholarly-writing conventions this guide assumes — abstract structure, reference formatting, section sequencing generally — see CASRAI’s Scholarly Writing cluster hub.
