Psychology dissertation case study research gives you a structured way to investigate a bounded person, group, programme, organisation, event, or setting in depth and in context. The method can reveal how psychological processes interact with relationships, institutions, culture, and time. Its strength does not come from studying one participant or collecting many details. It comes from defining a defensible case, linking the design to a focused question, integrating appropriate evidence, and explaining how interpretations were reached.
This guide takes you from the first design decision to a transparent dissertation report. It distinguishes a research case study from a clinical case report, explains single and multiple case designs, and shows how to build credible analysis without claiming statistical generalisability. The examples use globally applicable psychology settings and avoid assuming one national ethics system or university format.
What is a psychology dissertation case study?
A research case study is an in-depth examination of a contemporary phenomenon within its real-life context. The case is deliberately bounded. Its boundaries may involve people, place, activity, organisation, and a defined period. For example, a case could be one university counselling service during the first year of a peer-support programme, rather than “student mental health” in general.
The case study approach described by Crowe and colleagues is suitable when researchers need a detailed, multifaceted account of a complex issue in its natural context. It often answers how and why questions. A psychology dissertation might ask how staff and students understand a new trauma-informed practice, why a wellbeing programme works differently across settings, or how identity is negotiated within a specific community initiative.
A case study is not simply a small-sample interview project. Nor is it automatically a clinical case report about diagnosis or treatment. A rigorous research case study identifies the unit of analysis, sets boundaries, draws on a coherent methodological position, and usually considers more than one source or perspective. If your question concerns patterns across a broad population, prevalence, or causal effects under controlled conditions, another design may fit better.
| Research aim | Likely design | Why |
|---|---|---|
| Understand one programme in its institutional context | Case study | Context and process are central to the question |
| Estimate anxiety prevalence among students | Cross-sectional survey | The aim is a population estimate, not a bounded case |
| Test whether an intervention changes an outcome | Experimental or quasi-experimental design | The question concerns an effect or causal contrast |
| Explore lived experience across dispersed participants | Phenomenological or other qualitative design | Shared experience, rather than a bounded system, is central |
Table of Contents
Decide whether the design fits your question
Begin with the phenomenon, not with a preferred data collection technique. Case study research is useful when the phenomenon and its context cannot be separated easily. It may be appropriate when access to a distinctive setting offers a meaningful opportunity, when implementation differs across sites, or when an unusual case can challenge existing explanations.
Ask four questions before committing:
- Can I identify the case in one precise sentence?
- Can I state what falls outside the case?
- Does context form part of the explanation I need?
- Can I obtain enough ethical access to examine the case credibly?
If the answer to the first two questions is unclear, the proposed topic may be a general qualitative study rather than a case study. A project called “experiences of social media and self-esteem” has no obvious case. It becomes case-based when it examines, for example, how members of one youth arts collective negotiate social media use during a six-month exhibition project.
Define and bound the case
Boundaries prevent the dissertation from expanding into an unmanageable account of everything connected with the topic. They also tell readers what your conclusions refer to. Define the unit of analysis first. The unit could be an individual, team, service, classroom, online community, intervention, or event. Participants are not always the case. Staff interviewed about a service may be evidence providers, while the service remains the unit of analysis.
Use a boundary statement
A practical boundary statement specifies the case, location or context, activity, time, and evidence. For example: “The case is the implementation of a peer-led stress-management programme within one metropolitan university residence from September to May, examined through participant interviews, facilitator reflections, anonymised programme documents, and observations of training sessions.”
That sentence excludes other residences, later programme phases, and broader student mental health services. It also makes the evidence plan visible. Crowe and colleagues emphasise pre-defined boundaries concerning scope, time, social group, organisation or place, and evidence priorities. Boundaries can be refined after feasibility work, but changes should be documented and justified.
Distinguish the case from embedded units
An embedded design studies subunits within one case. In the residence programme example, facilitators, first-year participants, and residence staff could be embedded units. They allow comparison without turning the project into three unrelated studies. Keep the overall case question visible and explain how evidence from each unit contributes to it.
Embedded units can sharpen analysis, but they create a common risk: the dissertation may report separate groups and never integrate them at case level. Plan a final synthesis that explains the programme as a bounded system.
Choose a defensible case study type
Design labels should clarify the logic of inquiry rather than decorate the methodology chapter. Case studies can be described by purpose, number of cases, and structure. Some traditions distinguish intrinsic, instrumental, and collective cases. Others distinguish explanatory, exploratory, and descriptive purposes. State the methodological tradition you are following and use its terms consistently.
| Design choice | Best used when | Main analytical responsibility |
|---|---|---|
| Intrinsic single case | The particular case is important in its own right | Explain its distinctiveness without treating uniqueness as proof |
| Instrumental single case | The case helps illuminate a wider issue or concept | Connect detailed findings to a clear conceptual question |
| Multiple or collective cases | Planned comparison can test or refine an explanation | Complete within-case analysis before cross-case comparison |
| Embedded case | Subunits reveal different parts of one bounded system | Integrate subunit findings at the case level |
A single case can be justified because it is critical, unusual, revealing, longitudinally accessible, or especially informative for theory. Convenience alone is weak justification. Multiple cases can strengthen comparison, but each case requires enough depth. A dissertation with limited time may produce more credible work from one well-bounded case than from several thinly studied sites.
Build the research question and propositions
Case study questions usually begin with how, why, or in what ways. They should name the phenomenon and case without assuming the answer. “How do peer facilitators and participants understand changes to support-seeking during the residence programme?” is more useful than “Does peer support reduce stigma?” if the available design cannot estimate a causal effect.
Propositions can focus data collection and analysis. A proposition is a theoretically informed expectation, not a result to be confirmed at all costs. For example, a researcher might propose that perceived similarity between facilitators and participants shapes trust, while remaining open to contradictory evidence. An interpretivist design may use sensitising concepts rather than formal propositions. Either choice should align with the project’s epistemology.
Write an alignment chain: aim, question, case, evidence, analysis, and intended claim. If an interview question, document, or observation does not contribute to that chain, reconsider it. The broader psychology dissertation methodology guide can help you articulate philosophical assumptions and justify the design.
Select the case and participants purposefully
Case selection and participant sampling are related but different. First select the case because it can illuminate the research question. Then recruit people, records, occasions, or artefacts that provide relevant perspectives within that case. The psychology dissertation sampling guide explains purposeful strategies in more detail.
A typical case may show how an ordinary setting operates. A deviant or negative case can expose mechanisms hidden in routine examples. A critical case can examine whether an explanation holds under strategically important conditions. Maximum-variation selection may help compare contrasting sites, while criterion selection ensures every case meets pre-specified requirements.
Define inclusion and exclusion criteria before recruitment. State who controls access, how gatekeepers may shape the sample, and whose perspective could be missing. If management identifies all interviewees, staff may feel observed or dissenting voices may be excluded. Mitigate this with private invitations, clear voluntariness, and alternative recruitment routes where ethical approval permits.
Plan multiple sources of evidence
Multiple evidence sources can illuminate different aspects of a case. They do not need to agree. Interviews reveal accounts and meanings; observations reveal interactions and practices; documents show formal expectations and historical decisions; routine quantitative data may describe change over time. The aim is not to collect every possible source but to select evidence that answers the question.
| Evidence source | What it can contribute | Limitation to address |
|---|---|---|
| Semi-structured interviews | Interpretations, experiences, explanations, and contested meanings | Retrospective accounts and perceived power in the interview |
| Observation | Practices, interactions, setting, and non-verbal context | Researcher presence, selective attention, and access limits |
| Documents or digital artefacts | Policies, timelines, communication, and institutional narratives | Documents were created for purposes other than your research |
| Routine or questionnaire data | Descriptive patterns that contextualise qualitative findings | Measures, missingness, and small numbers constrain inference |
Use a case study protocol to keep collection consistent. Include the question, boundaries, recruitment plan, interview or observation guides, document-request list, field-note structure, security procedures, and decision rules. Maintain a case database or organised audit trail that separates raw evidence, coded material, analytic memos, and final claims.
For interview design, use neutral prompts and ask for concrete episodes rather than general approval. The psychology dissertation interview questions guide covers question sequencing and follow-up prompts. For a broader evidence plan, consult the psychology dissertation data collection guide.
Protect participants and the case
Case studies can make confidentiality difficult because contextual detail is analytically valuable. Even after names are removed, combinations of role, location, event, and quotation may reveal a person or organisation. This is deductive disclosure. Risk is especially high in small teams, unusual cases, and connected communities.
Complete a disclosure-risk review before collection and again before submission. Consider whether role titles can be broadened, dates shifted within an honest range, non-essential details removed, quotations paraphrased with approval, or cases combined at a higher level. Do not alter details that change meaning. Explain any protective transformations in the dissertation.
Consent should cover the actual evidence and reporting plan. Observation, archived messages, organisational records, and screenshots may require permissions beyond interview consent. Clarify whether a gatekeeper can consent for the setting but not for individual participants. Plan for distress, safeguarding disclosures, withdrawal, data access, storage, and deletion according to your institution and jurisdiction. The psychology dissertation ethics guide provides a structured starting point, but formal approval must come from the relevant ethics body.
Analyse the case systematically
Start analysis during data collection when your approved design allows iterative work. Early memos can reveal missing perspectives, unclear boundaries, and emerging rival explanations. Do not change recruitment or data collection informally; follow the amendment process required by your ethics approval.
Complete within-case analysis first
Build a case description, timeline, or evidence map. Identify actors, events, contextual conditions, and turning points. Code evidence using a strategy aligned with the question. Thematic analysis may organise patterns of meaning, while content analysis may examine defined categories across documents. These approaches are analytic tools, not substitutes for case-level integration. See the guides to psychology dissertation thematic analysis and psychology dissertation content analysis.
Create a claim-evidence matrix. For each provisional claim, record supporting sources, contradictory evidence, contextual conditions, researcher interpretation, and confidence. This exposes claims that rely on one vivid quotation and helps distinguish participant accounts from your synthesis.
Use pattern matching and explanation building carefully
Pattern matching compares an observed pattern with one predicted by theory or a rival explanation. Explanation building develops and revises an account of how or why events occurred. Neither technique proves causality by itself. State alternative explanations and test them against the available evidence.
For example, participants may attribute increased support-seeking to peer facilitators. Programme documents may show that confidential booking was introduced at the same time, and observation may reveal that staff actively promoted it. A credible analysis considers the combined explanation rather than crediting the most prominent narrative.
Compare cases without flattening context
In a multiple-case design, conduct a complete analysis of each case before building a cross-case matrix. Compare the same constructs, identify similarities and differences, and ask which contextual conditions might explain them. Do not pool people from different cases and call the result a multiple-case analysis. The cases, not just participants, are the primary units of comparison.
Strengthen trustworthiness
Triangulation means examining the same issue through different evidence, perspectives, researchers, or theoretical lenses. It is not a vote in which two sources defeat one. Disagreement can reveal role-based meanings, change over time, or limits in the evidence. Document how you handled convergence, complementarity, and contradiction.
Search deliberately for negative cases and rival explanations. Keep reflexive notes about your role, access, assumptions, relationships, and decisions. If you are an insider, describe both access advantages and pressures on disclosure. If you are an outsider, consider what participants may assume you cannot understand.
Other useful strategies include a documented audit trail, peer debriefing, transparent code development, rich contextual description, and participant engagement suited to the design. Participant checking is not automatically required or decisive. Participants can comment on factual accuracy or interpretations, but they may disagree because perspectives differ. Explain what was checked, by whom, and how feedback affected the account.
The critical review by Hyett and colleagues found that published qualitative case studies often lacked methodological clarity. Avoid that weakness by naming your case study tradition, linking every method to the question, and showing how the analysis produced the case-level explanation.
Make claims that fit the design
A case study normally does not support statistical generalisation from a small number of cases to a population. Its contribution may be analytical or theoretical generalisation: findings refine a concept, illuminate a mechanism, or show how an explanation operates under stated conditions. Readers then judge transferability using the contextual detail you provide.
Use calibrated language. Write “the evidence suggests,” “within this case,” or “participants described” when appropriate. Avoid “proved,” “all students,” or “the programme caused” unless the design genuinely supports those claims. Separate descriptive findings, participant interpretations, your analytic explanation, and implications for future research.
Write the dissertation transparently
The methodology chapter should justify the case study approach, epistemological position, case definition, boundaries, design type, selection, access, evidence sources, protocol, analysis, reflexivity, ethics, and quality procedures. The findings chapter may combine a concise case narrative with themes, explanatory patterns, timelines, or cross-case comparisons. Preserve an evidence trail through carefully selected quotations, observations, documents, and descriptive data.
Reporting standards provide useful checks. The Standards for Reporting Qualitative Research offer 21 items covering the title, abstract, problem, purpose, approach, researcher characteristics, context, sampling, ethics, data collection, analysis, trustworthiness, synthesis, links to evidence, integration, limitations, and conflicts. If interviews or focus groups are central, the COREQ checklist provides 32 reporting items. Use a checklist to improve transparency, not to replace methodological reasoning.
Common case study mistakes
- Calling any small sample a case study. Define the bounded case and explain why context matters.
- Confusing participants with the case. State the unit of analysis and each participant’s evidential role.
- Using convenience as the only selection logic. Provide a substantive reason the case can answer the question.
- Collecting several sources but analysing only interviews. Integrate each planned source or explain why it was excluded.
- Treating agreement as the only sign of quality. Analyse contradictions and rival explanations.
- Reporting themes without a case-level synthesis. Reconnect themes to boundaries, context, sequence, and the central question.
- Overclaiming generalisability or causality. Match conclusions to the design and evidence.
- Protecting names but exposing identities through context. Review combinations of details for deductive disclosure.
Frequently asked questions
Can one person be a psychology dissertation case study?
Yes, if the individual is a clearly justified, bounded case and the research question requires contextual depth. However, one interview with one person is not automatically a case study. Consider what contextual evidence is needed and whether detailed reporting could identify the participant.
How many participants do I need?
There is no universal number. The appropriate sample depends on the case, embedded units, evidence sources, question, access, and analytical depth. Justify who can provide relevant perspectives and show that the plan is feasible. Do not claim that a numerical threshold guarantees saturation or quality.
Do I need more than one data source?
Multiple sources are often valuable because case studies aim to understand a phenomenon in context. They can also reveal contradictions. A single-source design needs a particularly clear rationale and should not imply triangulation that did not occur.
Can I use quantitative data in a qualitative case study?
Yes. Descriptive questionnaires, attendance records, or routine outcomes can contextualise qualitative evidence. Explain their role, quality, missing data, and limitations. If quantitative and qualitative strands have equal methodological weight and planned integration, a mixed methods psychology dissertation framework may be more accurate.
Is a case study generalisable?
Usually not through statistical inference from the case to a population. A case can support analytical insights, refine theory, or offer findings that readers may judge transferable to sufficiently similar contexts. Provide the contextual and methodological detail needed for that judgement.
Should participants approve my interpretation?
Participant feedback can help check factual accuracy or explore interpretations, but it is not an automatic truth test. People within one case may hold different legitimate perspectives. Describe the purpose, process, responses, and analytical consequences of any participant checking.
Conclusion
A strong psychology dissertation case study begins with a question that genuinely requires contextual depth. It defines and bounds the case, selects it for an explicit reason, gathers relevant evidence, protects identities, and integrates analysis at case level. Its credibility rests on transparent reasoning, reflexivity, attention to contradictory evidence, and claims proportionate to the design.
Before collecting data, create a one-page design map linking your question, case, boundaries, embedded units, evidence, ethics, analysis, and intended claim. Ask your supervisor and ethics reviewers to challenge the weakest link. If you need methodological support, seek guidance that strengthens your own decisions and preserves authorship, confidentiality, and academic integrity.
Authoritative references
- Crowe S, Cresswell K, Robertson A, et al. The case study approach. BMC Medical Research Methodology. 2011;11:100.
- Baxter P, Jack S. Qualitative Case Study Methodology: Study Design and Implementation for Novice Researchers. The Qualitative Report. 2008;13(4):544-559.
- Hyett N, Kenny A, Dickson-Swift V. Methodology or method? A critical review of qualitative case study reports. International Journal of Qualitative Studies on Health and Well-being. 2014;9:23606.
- O’Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for Reporting Qualitative Research. Academic Medicine. 2014;89(9):1245-1251.
- Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research. International Journal for Quality in Health Care. 2007;19(6):349-357.
