To plan a case study, define exactly what the case is, explain its boundaries and selection, then connect each research question to evidence you can obtain and analyse appropriately. Keep the surrounding context visible. Adding “a case study of” to a title does not supply that plan.
Your report should let a reader follow the case from question to evidence to interpretation. UniDraft helps organise your confirmed method, permitted source material and actual findings into an editable project document. Start your case-study project write-up, then check every account and conclusion against your records and supervisor's instructions. You conduct the study and analysis; the software does not create access, participants or findings.
What makes a study a case study?
The case is the defined entity or process you investigate in context. In their methodological paper, Crowe and colleagues explain the importance of defining and selecting the case, considering appropriate evidence sources and making interpretation transparent. Their healthcare examples explain research principles; they are not a required template for Nigerian undergraduate projects.
A case might concern a bounded programme or organisational process. It is not automatically the same as a participant, a building or the place where you distributed questionnaires. Start with the research-design decision guide if the approach is still undecided. Different case-study traditions use different assumptions; agree the approach and rationale with your supervisor.
Write the boundary before collecting evidence
Use this original planning sheet. These are prompts to discuss, not a compulsory institutional form.
Baxter and Jack's methodological overview also identifies defining and bounding the case, research questions and evidence sources as design decisions. Use these questions to make your own plan explicit rather than copying a study's choices.
| Decision | What your plan must make clear |
|---|---|
| Research question | What you want to understand about the case |
| Case and unit of analysis | The entity or process about which the main explanation will be made |
| Place and activity | Which setting and activity are included |
| Time boundary | The period examined and why it fits the question |
| Selection rationale | Why this case can illuminate the question; access alone is not the full explanation |
| Evidence scope | Which accounts, observations or records are relevant and obtainable |
| Exclusions | What the study will not examine and the consequence for its claims |
| Required permissions | Applicable supervisor, institutional, site, ethics and participant arrangements |
If the case concerns a process within one office, the staff interviewed may provide evidence about that process. Do not silently redefine the case as each staff member when writing the findings. If the plan involves several cases, explain the selection and comparison rather than treating them as one undifferentiated group.
A fictional example: a library request process
This entire example is invented for teaching. No university, permission, interview, observation or finding is being reported. The choices illustrate planning and are not a validated research protocol.
Suppose the question is: “How is a student request for a reserved item handled within one departmental library service during the selected teaching period?” The proposed case is that request-handling process in its setting. The study would need a defensible reason for selecting it and a feasible access plan.
The boundary might include how requests are received, checked and communicated during that period. It might exclude borrowing services elsewhere and long-term changes in student attainment. Those exclusions limit the claims: this plan cannot establish whether library provision improves grades nationally.
Before choosing instruments, write the evidence connection:
| Part of the question | Possible permitted evidence | What it would not establish alone |
|---|---|---|
| What is the stated process? | The applicable procedure document, with its date and version | That everyone followed it in practice |
| How do relevant people describe it? | Appropriately selected staff or student accounts | The frequency of every event across the whole period |
| What happens in selected instances? | An approved observation record of specified steps | Every unobserved request or a person's private reason |
These are possible sources, not instructions to collect all three. Use the data-collection guide to justify the records needed. Check confidentiality and permission before seeking internal logs or observing interactions; access to a service is not permission to research its users.
Keep an evidence register you can trace
Give each permitted item a stable reference and document its origin, relevant date or version, access conditions and connection to the question. Keep factual notes separate from your interpretation. Do not publish names or private records merely to make the report look well evidenced.
An original working register can use these columns:
| Evidence ID | Source and context | Relevant question | Location of useful material | Limit or uncertainty |
|---|---|---|---|---|
| Fill from your actual record | Document version, observation conditions or coded account | State the question it helps address | Page, permitted timestamp or note reference | Missing context, restricted access or competing explanation |
The data-management plan helps connect storage, authorised access and retention decisions. Use the interview-guide or observation-checklist workflow only where that method fits the agreed design. A register organises evidence; it does not replace analysis.
What if the sources disagree?
Do not make agreement the objective. Different records can describe different periods, perspectives or parts of a process. Investigate the difference and preserve what remains unresolved rather than selecting the most convenient account.
Fictional interpretation exercise, not collected findings: imagine a written procedure says that a confirmation is sent after a request. An invented account says confirmations are sometimes absent. One invented observation includes a confirmation.
That observation does not disprove the account or establish that the procedure is always followed. Questions to investigate would include whether the records concern the same version, channel and period, and what other permitted evidence could address the difference. If access cannot resolve it, explain the uncertainty.
Use this original analysis worksheet:
| Proposed interpretation | Supporting record | Complicating record or alternative | What can fairly be concluded? |
|---|---|---|---|
| Write a specific claim | Trace it to the actual item and context | Consider another explanation or relevant absence | Keep the conclusion within the available evidence |
Choose and explain an analytic approach that fits the study. The thematic-analysis guide addresses one way of interpreting qualitative material; the content-analysis guide covers explicit coding of communication material. Neither label automatically analyses a case for you.
Turn the study into a readable report
Your methodology chapter should explain the case boundary, selection, sources, actual collection and analysis, permissions and deviations from the plan. Use the tense and headings your supervisor requires; a proposal describes intended work, while the final report must identify what actually happened.
Present enough context to understand the findings while respecting confidentiality. Explain a finding, show the relevant permitted evidence and discuss alternatives and limits. Organising one subsection per source can leave the reader with disconnected accounts: show how the material addresses the question instead.
Before defence, practise explaining:
- What is the case, and what is outside it?
- Why was this case selected?
- Which evidence supports one important conclusion?
- What disagreed, remained unavailable or changed during the study?
- What would you need before extending the claim to another setting?
Clear boundaries make the argument assessable. They do not guarantee approval or turn one case into evidence about every university. UniDraft can help organise the written account around the work you confirm. Prepare your project document using your actual records and applicable instructions.
Methodological sources
- Crowe et al. (2011), The case study approach, BMC Medical Research Methodology, 11:100, DOI 10.1186/1471-2288-11-100. Its case-study principles inform the guide; the example and worksheets above are original teaching aids.
- Baxter and Jack (2008), Qualitative Case Study Methodology: Study Design and Implementation for Novice Researchers, The Qualitative Report, 13(4), 544–559, DOI 10.46743/2160-3715/2008.1573. The linked journal abstract identifies the design decisions discussed above.