The methodology chapter is where many dissertations either earn the marker’s trust or lose it. It explains how you collected and analysed your data, and — crucially — why those choices were right for your research question. If you have read guidance on writing a methodology but still feel unsure what a strong one actually looks like, this collection of dissertation methodology examples is for you.
Below you will find 32 worked examples across quantitative, qualitative, mixed-methods and secondary research, covering business, psychology, nursing, education, marketing, law, criminology and more. Each example shows the key choices — philosophy, approach, design, sampling, data collection, analysis and ethics — so you can adapt the pattern to your own project. For step-by-step writing advice, pair this page with our dissertation methodology guide.
A dissertation methodology is the chapter that describes and justifies the research methods used to answer your research questions. It usually covers:
Many UK business and social science programmes structure this chapter around the “research onion” developed by Saunders, Lewis and Thornhill, moving from philosophy at the outer layer to data collection and analysis at the centre.
Abstract advice such as “justify your choices” is hard to apply until you see it in action. Studying dissertation methodology examples helps you to:
| Section | What to include | Typical length (15,000-word dissertation) |
|---|---|---|
| Introduction | Restate aims and research questions; outline the chapter | 100–200 words |
| Research philosophy | Positivism, interpretivism, pragmatism or critical realism, with justification | 250–400 words |
| Research approach | Deductive, inductive or abductive | 150–250 words |
| Research design and strategy | Survey, experiment, case study, interviews, review | 300–500 words |
| Sampling | Population, sampling technique, sample size, recruitment | 250–400 words |
| Data collection | Instruments, piloting, procedure | 300–500 words |
| Data analysis | Statistical tests or qualitative techniques; software | 300–500 words |
| Validity, reliability or trustworthiness | How quality was ensured | 200–300 words |
| Ethics | Consent, confidentiality, data protection, approval | 200–300 words |
| Limitations | Honest discussion of constraints | 150–250 words |
The methodology usually accounts for 10–15% of the total word count. See our guide to structuring a dissertation for how it fits with the other chapters.

Quantitative methodologies collect numerical data to test hypotheses and measure relationships. They usually follow a positivist philosophy and a deductive approach.
Philosophy: positivism. Approach: deductive. Design: cross-sectional survey. Sample: 180 office employees from three UK firms, selected by stratified random sampling. Data collection: online questionnaire using validated productivity and autonomy scales. Analysis: descriptive statistics, Pearson correlation and multiple regression in SPSS. Ethics: informed consent, anonymised responses and secure data storage.
Philosophy: positivism. Approach: deductive, testing hypotheses from Social Comparison Theory. Design: correlational survey. Sample: 220 undergraduate students aged 18–24 recruited through convenience sampling. Data collection: Rosenberg Self-Esteem Scale and a daily usage log. Analysis: correlation, t-tests by gender and hierarchical regression. Ethics: right to withdraw and signposting to wellbeing services.
Design: quasi-experimental pre-test/post-test with a comparison group. Sample: two Year 10 maths classes (n = 56). Data collection: standardised test scores before and after a six-week intervention. Analysis: independent and paired t-tests, effect sizes (Cohen’s d). Limitations: non-random allocation and a single school setting.
Design: survey based on the SERVQUAL model. Sample: 300 shoppers intercepted at four stores using quota sampling by age and gender. Analysis: reliability testing (Cronbach’s alpha), factor analysis and structural equation modelling. Justification: a quantitative design allows the relationships in the conceptual model to be tested statistically.
Design: retrospective cohort study using anonymised ward incident data. Sample: 12 months before and after rounding was introduced on two medical wards. Analysis: incidence rates per 1,000 bed days and chi-square tests. Ethics: NHS governance approval and use of anonymised secondary data only.
Design: secondary data panel study. Sample: FTSE 350 companies over five years. Data: ESG scores and share price data from financial databases. Analysis: panel regression with fixed effects, controlling for company size and sector. Limitations: differences in ESG rating methods.
Design: A/B experiment. Sample: visitors to an online store randomly served a standard or optimised page. Data: analytics on load time, bounce rate and conversions. Analysis: logistic regression and conversion-rate comparison. Ethics: no personal data collected beyond anonymised analytics.
Design: longitudinal diary study over one term. Sample: 90 first-year students. Data: weekly sleep diaries and module grades. Analysis: repeated-measures analysis and correlation. Strength: captures change over time rather than a single snapshot.
Design: secondary analysis of police.uk crime data before and after a council lighting scheme. Sample: 40 streets with upgraded lighting and 40 comparison streets. Analysis: difference-in-differences comparison. Limitations: reporting variations and displacement effects.
Design: survey with 150 employees across eight hotels. Measures: training satisfaction, organisational commitment and turnover intention. Analysis: mediation analysis testing whether commitment explains the link between training and retention.
Qualitative methodologies explore meanings, experiences and perceptions. They usually follow an interpretivist philosophy and an inductive approach, with smaller, purposefully selected samples.
Philosophy: interpretivism. Approach: inductive. Design: phenomenological study. Sample: 12 postgraduate international students recruited by purposive and snowball sampling. Data collection: semi-structured interviews of 45–60 minutes. Analysis: interpretative phenomenological analysis (IPA). Trustworthiness: member checking and a reflexive diary.
Design: qualitative descriptive study. Sample: 10 registered nurses from acute wards. Data: online semi-structured interviews. Analysis: Braun and Clarke’s six-phase thematic analysis using NVivo. Ethics: distress protocol and confidential reporting.
Design: multiple case study of three family-owned firms. Data: interviews with owners and staff, plus company documents. Analysis: within-case and cross-case thematic analysis. Justification: case studies allow in-depth understanding of leadership in its real-world context.
Design: focus groups with 18 secondary teachers across three schools. Analysis: thematic analysis. Strength: focus groups reveal shared views and debate between participants.
Design: qualitative content analysis. Sample: 300 posts from six UK fashion brands over six months. Analysis: coding frame developed from the literature and refined inductively.
Design: narrative inquiry. Sample: six young adults who were carers as teenagers. Data: life-story interviews. Analysis: narrative analysis. Ethics: safeguarding procedures and support signposting.
Design: ethnography within one organisation over three months. Data: participant observation, field notes and informal interviews. Analysis: thematic and cultural analysis. Ethics: gatekeeper consent and anonymised organisation name.
Design: semi-structured interviews with 14 officers. Analysis: framework analysis. Limitations: potential social desirability bias.
Design: qualitative interviews with 15 patients using remote consultations. Analysis: thematic analysis guided by the Technology Acceptance Model.
Design: grounded theory. Sample: 20 consumers aged 18–25, sampled theoretically until saturation. Analysis: open, axial and selective coding to develop a model of trust.
Mixed-methods research combines quantitative and qualitative data, usually under a pragmatist philosophy. The key is to explain how the two strands are integrated.
Design: explanatory sequential mixed methods. Phase 1: survey of 250 students on wellbeing and service use. Phase 2: interviews with 12 respondents to explain the survey findings. Integration: joint display comparing statistical and thematic results.
Design: exploratory sequential. Phase 1: interviews with eight SME owners. Phase 2: survey of 120 SMEs built from the interview themes. Analysis: thematic analysis followed by regression.
Design: convergent parallel. Data: adherence questionnaire (n = 100) and interviews (n = 10) collected at the same time. Integration: results compared in the discussion chapter.
Design: survey testing the Theory of Planned Behaviour (n = 200) plus a focus group exploring barriers. Analysis: SEM and thematic analysis.
Design: learning analytics data from a VLE combined with student interviews. Strength: combines behavioural data with student explanations.
Design: case study using company performance data and manager interviews. Analysis: descriptive statistics and thematic coding.
Many UK dissertations, especially in health, law and business, do not involve primary data collection. A secondary or literature-based methodology must be just as rigorous and transparent.
Design: systematic review following PRISMA guidance. Search: MEDLINE, CINAHL and PsycINFO for 2015–2025. Criteria: randomised controlled trials in adults. Quality appraisal: CASP checklists. Synthesis: narrative synthesis.
Design: integrative literature review. Sources: 40 peer-reviewed articles. Analysis: thematic synthesis of benefits, risks and ethical issues.
Design: doctrinal analysis of legislation, case law and academic commentary, with comparative analysis of a second jurisdiction. Referencing: OSCOLA.
Design: document analysis of government strategies and committee reports. Analysis: qualitative content analysis using a policy framework.
Data: a national longitudinal survey dataset. Analysis: logistic regression on predictors of loneliness. Strength: large, representative sample without primary data collection.
Design: archival research using newspapers, government records and letters. Analysis: source criticism and thematic interpretation.
This study adopts an interpretivist philosophy because it seeks to understand how international students make sense of their experiences in UK higher education. Interpretivism assumes that reality is socially constructed and that meaning is best understood through participants’ own accounts. A positivist approach was rejected because it would reduce complex experiences to measurable variables and overlook context.
A stratified random sample of 180 employees was drawn from three organisations, with strata based on department to ensure all functions were represented. A minimum sample size of 150 was calculated to detect a medium effect size at a 5% significance level with 80% power. Participants were invited by email and the survey remained open for three weeks.
Interview transcripts were analysed using Braun and Clarke’s six-phase approach to reflexive thematic analysis: familiarisation, generating initial codes, constructing themes, reviewing themes, defining and naming themes, and producing the report. NVivo 14 was used to organise codes, and a reflexive journal was kept to record analytical decisions.
Ethical approval was obtained from the university research ethics committee before data collection. Participants received an information sheet, gave written informed consent and were reminded of their right to withdraw. Data were anonymised, stored on a password-protected university drive and handled in line with UK GDPR.
The use of convenience sampling limits the generalisability of the findings, and self-reported measures may be affected by social desirability bias. These limitations are considered when interpreting the results in Chapter 5.
| Your research question asks… | Likely methodology | Typical analysis |
|---|---|---|
| “What is the relationship between X and Y?” | Quantitative survey | Correlation, regression |
| “Does X cause a change in Y?” | Experiment or quasi-experiment | t-tests, ANOVA |
| “How do people experience X?” | Qualitative interviews | Thematic analysis, IPA |
| “How and why does X happen in this organisation?” | Case study | Thematic, cross-case analysis |
| “What does existing evidence say about X?” | Systematic or literature review | Narrative or thematic synthesis |
| “How much and why?” | Mixed methods | Statistics plus thematic analysis |
| Technique | Type | When to use |
|---|---|---|
| Simple random | Probability | You have a full list of the population |
| Stratified | Probability | You need sub-groups represented |
| Cluster | Probability | Population is spread across locations |
| Convenience | Non-probability | Time and access are limited (acknowledge limits) |
| Purposive | Non-probability | You need participants with specific experience |
| Snowball | Non-probability | Hard-to-reach groups |
| Quota | Non-probability | You want proportions similar to the population |
Quantitative methodologies should discuss validity (are you measuring what you intend to measure?) and reliability (would the results be consistent?). Qualitative methodologies usually discuss trustworthiness using credibility, transferability, dependability and confirmability. Practical techniques include piloting questionnaires, using validated scales, checking Cronbach’s alpha, member checking, triangulation and keeping an audit trail.
Every methodology needs an ethics section. Cover informed consent, confidentiality, anonymity, data storage, the right to withdraw, risks to participants and how approval was obtained. Research involving NHS patients, children or vulnerable adults usually needs additional approvals. The UK Research Integrity Office publishes useful guidance on good research practice.
Your methodology should connect clearly to your literature review and introduction, and its key choices should be summarised in your abstract.
At undergraduate level, markers mainly look for a clear, feasible design that matches the research question. A typical example is a short online survey of 80–150 participants analysed with descriptive statistics and simple inferential tests, or 6–10 semi-structured interviews analysed thematically. Literature-based dissertations are also common, especially where ethics approval for human participants would take too long. Philosophy sections can be brief but should still explain the reasoning behind your choices.
At master’s level, the methodology must show deeper critical understanding. Markers expect you to compare philosophies, justify the design against alternatives, explain your sampling calculation or saturation strategy, and discuss validity or trustworthiness in detail. Mixed-methods designs, case studies and structural equation modelling are more common. You should also reflect on your position as a researcher, particularly in qualitative or workplace-based studies.
Doctoral methodologies are usually a full chapter of 8,000–15,000 words. They engage critically with methodological literature, may include pilot studies, and often justify an original combination of methods. Examiners will question your methodological decisions in the viva, so every choice must be defensible.
| Subject | Common designs | Typical analysis | Key ethical issues |
|---|---|---|---|
| Business and management | Surveys, case studies, interviews with managers | Regression, SEM, thematic analysis | Organisational consent, commercial confidentiality |
| Nursing and healthcare | Systematic reviews, qualitative interviews, audits | Narrative synthesis, thematic analysis, descriptive statistics | NHS approvals, patient confidentiality |
| Psychology | Experiments, correlational surveys, IPA | t-tests, ANOVA, regression, IPA | BPS ethics, debriefing, deception |
| Education | Action research, quasi-experiments, focus groups | Paired t-tests, thematic analysis | Working with children, parental consent |
| Law | Doctrinal, comparative, socio-legal research | Legal analysis, case analysis | Usually minimal for doctrinal work |
| Criminology | Secondary crime data, interviews with practitioners | Descriptive statistics, thematic analysis | Sensitive topics, safeguarding |
| Computing | Design science, experiments, usability testing | Performance metrics, usability scores | Data protection, informed consent for testing |
| Marketing | Surveys, content analysis, netnography | SEM, coding frames | Use of public social media data |
In a research proposal, the methodology describes what you plan to do and is written in the future tense. In the final dissertation, it reports what you did, including any changes from the plan and why they happened — for example, a lower response rate or a switch from face-to-face to online interviews. Explaining these changes honestly strengthens rather than weakens your chapter.
If your methodology feels weak or your supervisor has asked for major changes, our UK specialists can help you choose and justify the right design. Explore our dissertation help, dissertation proposal services, PhD dissertation writing services and dissertation editing service. For more on what top-band work looks like, read our guide to a first-class dissertation.
A simple example is a quantitative survey of 180 employees, using stratified random sampling and a validated questionnaire, analysed with correlation and regression in SPSS, under a positivist philosophy and deductive approach, with informed consent and anonymised data.
Typically 10–15% of the total word count — around 1,500–2,500 words for a 15,000-word master’s dissertation.
Yes. Systematic reviews, literature-based dissertations, document analysis and secondary dataset analysis are all valid if you explain your search strategy, selection criteria and analysis clearly.
Use the future tense in a proposal (“will be collected”) and the past tense in the final dissertation (“were collected”).
Most UK business, social science and health programmes expect one, especially at master’s level. Check your module handbook.
Many undergraduate and master’s projects use 6–15 interviews, aiming for depth and data saturation rather than a large number.
Methods are the specific tools you use, such as questionnaires or interviews. Methodology is the overall strategy and justification, including your philosophy, approach and design.
Yes — place full instruments in the appendices and refer to them in the methodology chapter.
Strong methodologies are not about complex designs; they are about coherent, justified choices. Use these dissertation methodology examples to see how philosophy, approach, design, sampling, data collection, analysis, ethics and limitations fit together, then adapt the closest example to your own research questions. A clear, honest methodology gives markers confidence in your findings — and that confidence shows in your final grade.