Few choices shape a dissertation more than whether it will be qualitative, quantitative or a mixture of both. The decision affects how you collect data, how many people you need, what software you use, how you write your findings and what you can claim at the end. It also affects your stress levels, because each approach carries different kinds of work and different kinds of risk.
Students often treat the choice as a matter of taste: “I am bad at maths, so qualitative”, or “numbers look more scientific, so quantitative”. Neither is a good reason. The best approach is the one that answers your research question most convincingly within your time, skills and access. This guide explains the differences, compares strengths and limits, gives examples, and offers a simple way to decide. For support across the whole project, see our main guide to dissertation help UK.
The short answer
- Choose quantitative if your question asks how much, how many, how often, to what extent, or what is the relationship between, and you can measure the things involved.
- Choose qualitative if your question asks how, why, or what is it like, and you need to understand meanings, experiences or contexts in depth.
- Choose mixed methods only if your question genuinely needs both, and you have the time and skills to do both well.
If your question does not clearly point one way, the question itself may need sharpening. See how to choose a dissertation topic and how to write a dissertation proposal.
What is quantitative research?
Quantitative research collects numerical data and analyses it statistically. It seeks to measure variables, test hypotheses, identify patterns and relationships, and, where the design allows, generalise from a sample to a wider population.
Typical methods:
- Structured questionnaires and surveys.
- Experiments and quasi-experiments.
- Analysis of existing datasets, such as national surveys or organisational records.
- Structured observation with coding schemes.
Typical data: numbers, scores, counts, ratings and categories.
Typical analysis: descriptive statistics, correlations, t-tests, ANOVA, chi-square, regression, factor analysis, using software such as SPSS, R, Stata or Excel.
Typical language: variables, hypotheses, significance, effect size, reliability, generalisability.
What is qualitative research?
Qualitative research collects non-numerical data, such as words, images and observations, and analyses it for meaning, patterns and themes. It seeks to understand experiences, perspectives and contexts in depth rather than to measure them.
Typical methods:
- Semi-structured or in-depth interviews.
- Focus groups.
- Case studies.
- Ethnography and participant observation.
- Document, policy or media analysis.
Typical data: transcripts, field notes, documents, images.
Typical analysis: thematic analysis, content analysis, framework analysis, grounded theory, interpretative phenomenological analysis, discourse analysis, often supported by NVivo, ATLAS.ti or manual coding.
Typical language: themes, meanings, experiences, context, credibility, transferability, reflexivity.
Side-by-side comparison
| Quantitative | Qualitative | |
|---|---|---|
| Main purpose | Measure, test, compare, predict | Explore, understand, interpret |
| Typical questions | How many? To what extent? What is the relationship? | How? Why? What is it like? |
| Data | Numbers | Words, images, observations |
| Sample | Larger, often hundreds | Smaller, often 8 to 30 |
| Sampling | Often aims at representativeness | Often purposive, aims at relevance and richness |
| Tools | Surveys, scales, experiments, datasets | Interview guides, observation notes, documents |
| Analysis | Statistical | Thematic or interpretive |
| Researcher role | Detached, minimising influence | Engaged, reflexive about influence |
| Typical philosophy | Positivist or post-positivist | Interpretivist or constructivist |
| Approach | Usually deductive | Usually inductive |
| Strength | Breadth, comparability, generalisation | Depth, nuance, context |
| Limitation | Limited depth; may miss meaning and context | Limited generalisability; time-intensive analysis |
| Quality criteria | Validity, reliability, generalisability | Credibility, dependability, transferability, confirmability |
For how this feeds into the chapter that explains your choices, see our guide to the dissertation methodology chapter.
Choosing by research question
Your question is the best guide. Here are example questions and the approach each suggests.
| Research question | Likely approach | Why |
|---|---|---|
| To what extent does daily social media use predict anxiety among UK first-year undergraduates? | Quantitative | Measures two variables and tests a relationship |
| How do first-year undergraduates experience social media use during exam periods? | Qualitative | Explores experience and meaning |
| Do patients who receive telehealth follow-up report higher satisfaction than those who attend in person? | Quantitative | Compares groups on a measurable outcome |
| How do patients with long-term conditions make sense of telehealth consultations? | Qualitative | Seeks understanding of interpretation |
| What factors influence employee turnover in UK retail SMEs, and how strongly? | Quantitative (or mixed) | Identifies and weighs factors |
| Why do experienced nurses decide to leave acute NHS wards? | Qualitative | Seeks reasons and context |
| How have UK courts interpreted “reasonable adjustments” in employment cases? | Qualitative (doctrinal) | Analyses texts and reasoning |
| How widespread are phishing attempts across small UK charities, and how do staff respond? | Mixed | Needs prevalence and understanding of response |
Notice that the topic often does not decide the approach. The wording of the question does.
Strengths and limitations in more detail
Quantitative: strengths
- Breadth and comparability. Many cases can be compared on the same measures.
- Statistical testing. You can say how likely a pattern is to be due to chance.
- Generalisation. With a suitable sample and design, findings may apply beyond the participants.
- Clear structure. Variables, hypotheses and tests give a defined path.
- Transparent reporting conventions. Markers know what to look for.
- Efficiency with large samples. Online surveys and secondary datasets can reach many cases quickly.
Quantitative: limitations
- Limited depth. Numbers can miss context, meaning and complexity.
- Measurement problems. Poorly designed or inappropriate scales weaken everything that follows.
- Self-report bias. People may not answer accurately or honestly.
- Sample demands. Small or unrepresentative samples limit what you can claim.
- Statistical skills. Choosing, running and interpreting tests correctly takes learning.
- Association is not causation. Unless the design supports causal claims, you must be careful.
Qualitative: strengths
- Depth and nuance. Participants explain in their own words, revealing things you did not anticipate.
- Context. You can understand how and why things happen in particular settings.
- Flexibility. The design can adapt as insights emerge.
- Good for new or poorly understood topics.
- Smaller samples are acceptable, provided they are justified.
- Rich quotations can bring findings to life.
Qualitative: limitations
- Limited generalisability. Findings are not statistically representative.
- Time-intensive. Interviewing, transcribing and coding take much longer than students expect.
- Researcher influence. Your presence and interpretation shape the data, so reflexivity is essential.
- Subjectivity concerns. You must show how themes were built from data, not from your assumptions.
- Access and ethics. Recruiting participants, especially on sensitive topics, can be hard.
- Volume of data. Hours of transcripts can overwhelm without a clear analytic plan.
Mixed methods: when and how
Mixed methods combine qualitative and quantitative elements in a single study. They can be powerful, but they are not an easy option. Doing two things properly takes more time, skill and word count than doing one.
Common mixed methods designs
| Design | Sequence | Purpose | Example |
|---|---|---|---|
| Explanatory sequential | Quantitative first, then qualitative | Use interviews to explain survey results | Survey shows low engagement; interviews explore why |
| Exploratory sequential | Qualitative first, then quantitative | Use interviews to build a survey | Interviews identify factors; survey tests them across a larger group |
| Convergent (parallel) | Both at the same time | Compare results to confirm or contrast | Survey and interviews about the same issue, merged in discussion |
When mixed methods make sense
- Your question has distinct parts that need different evidence (for example “how widespread” and “how experienced”).
- One method alone would leave an important gap.
- You have enough time, word count and expertise for both.
- Your supervisor and department support the design.
When to avoid it
- You are choosing it only to look more sophisticated.
- You have a short timeline or tight word limit.
- You cannot yet do either method well on its own.
A modest, well-executed single-method study usually scores higher than an ambitious mixed design done thinly.
Sample sizes in practice
There is no universal rule, and your department’s guidance comes first. These are common student-project ranges.
| Approach | Typical range | Notes |
|---|---|---|
| Quantitative survey | 100 to 400+ | Depends on the analysis; regression and factor analysis need more cases. Use a power calculation or recognised guidance where possible |
| Quantitative experiment | 30 to 100+ per group | Depends on effect size and power |
| Qualitative interviews | 8 to 20 | Justify with information power or saturation |
| Focus groups | 3 to 6 groups | Of 4 to 8 participants each |
| Case study | 1 to 4 cases | Depth matters more than number |
| Document or content analysis | Depends on corpus | Define selection criteria clearly |
Whatever you choose, explain why it is adequate for your purpose, and acknowledge limits.
How the work differs
Students are often surprised by where the effort goes.
| Stage | Quantitative | Qualitative |
|---|---|---|
| Design | Heavy: instrument choice, hypotheses, power | Moderate: interview guide, sampling logic |
| Data collection | Can be quick online; recruiting enough respondents can be hard | Slow and personal; scheduling and building rapport take time |
| Preparation | Cleaning, coding, checking assumptions | Transcribing (several hours per interview hour) |
| Analysis | Technical, rule-based; software does the calculations but you must choose and interpret | Interpretive, iterative, time-heavy |
| Writing | Tables, figures and statistical reporting | Themes, quotations and narrative |
| Biggest risk | Wrong test, weak measures, low response | Thin analysis that only describes, unclear theme development |
If you want help with the analysis stage, see our dissertation data analysis help page.
Five questions to help you decide
Work through these in order.
- What does my research question actually ask? Look at its verbs. Measure, compare, test, predict point to quantitative. Explore, understand, interpret, experience point to qualitative.
- What kind of evidence would convince a sceptical reader? Numbers and statistical patterns, or rich accounts and context, or both?
- What can I access? Do you have access to enough respondents or a usable dataset? Can you reach people willing to be interviewed? See also dissertation ethics approval in the UK.
- What are my skills, and what can I learn in time? Be honest about statistics, interviewing and software.
- What fits my time and word count? Count backwards from your deadline, allowing for ethics, data collection, analysis and writing.
If the answers point in different directions, talk to your supervisor. See working with your dissertation supervisor.
Common myths
“Quantitative is more rigorous.” Rigour means doing a method well and honestly, whichever it is. A badly designed survey is less rigorous than a carefully conducted interview study.
“Qualitative is easier.” It avoids statistics but demands different skills: interviewing, coding, interpretation and sustained reflexivity. Many students underestimate the analysis workload.
“Quantitative means no thinking about meaning.” Good quantitative work still needs careful concepts, measures and interpretation.
“Qualitative cannot be generalised, so it is weak.” It aims for transferability and theoretical insight, not statistical generalisation, which is a different and legitimate goal.
“Mixed methods always impress markers.” Only when the design is justified and well executed.
“I can decide later.” The approach shapes your ethics application, instruments and timeline. Decide early.
Examples across disciplines
These illustrate how question and approach fit together. They are models of scope and logic, not suggestions to copy.
Business and management
- Quantitative: To what extent does perceived organisational support predict employee engagement in UK retail firms?
- Qualitative: How do middle managers in UK financial services interpret and enact hybrid working policies?
Nursing and healthcare
- Quantitative: Is there an association between nurse staffing levels and reported medication errors in one NHS trust?
- Qualitative: How do newly qualified nurses describe their first six months on acute wards?
Psychology
- Quantitative: Does sleep quality predict academic self-efficacy among UK undergraduates?
- Qualitative: How do first-time parents experience social support during the first year?
Education
- Quantitative: Is there a relationship between homework time and attainment among Year 9 pupils in one school?
- Qualitative: How do secondary teachers in England perceive workload policies?
Law
- Doctrinal (qualitative): How have UK courts interpreted “reasonable adjustments” under the Equality Act 2010?
- Mixed or empirical: How do employment tribunals’ reasoning and outcomes vary across a sample of reported decisions?
Computer science
- Quantitative: How do two machine-learning classifiers compare in detecting phishing emails across a standard dataset?
- Qualitative: What usability barriers do small charity staff report when adopting an open-source CRM?
How to justify your choice in writing
Markers want to see reasoning, not preference. A convincing justification follows a pattern.
- State what the question requires. “The research question seeks to understand how employees experience motivation, which is subjective and context-dependent.”
- Name the approach and design. “A qualitative design using semi-structured interviews was adopted.”
- Explain why it fits. “Interviews allow participants to describe their experiences in their own words, enabling depth of understanding (Author, Year).”
- Acknowledge the alternative. “A survey would allow a larger sample but would limit depth, which is central to the question.”
- Admit the limits and how you managed them. “The small sample limits generalisation; the study therefore seeks transferable insight rather than statistical representation.”
Support methodological claims with credible methods literature, and reference accurately. Cite Them Right is a trusted guide to citation formats, and our guide to Harvard referencing for UK dissertations covers common slips.
Quality and ethics in both approaches
Whichever approach you choose, you must show quality and handle ethics carefully.
- Quantitative: validated scales, reported reliability (for example Cronbach’s alpha), pilot testing, assumption checks, appropriate tests and honest reporting of non-significant results.
- Qualitative: reflexive journal, audit trail, triangulation, member checking where suitable, clear account of how themes were developed, and thoughtful use of quotations.
- Both: informed consent, confidentiality, secure data storage in line with UK GDPR and the Data Protection Act 2018, and approval from your university ethics committee.
Useful references include the BERA ethical guidelines, the British Psychological Society, the UK Research Integrity Office and, for health research, the NHS Health Research Authority.
Common mistakes when choosing
- Choosing by comfort instead of by question.
- Mismatch. For example, a survey to answer a question about lived experience.
- Overreaching with mixed methods and doing both thinly.
- Underestimating qualitative analysis time.
- Underestimating quantitative recruitment. Reaching enough respondents is often harder than expected.
- Claiming generalisability from small or non-random samples.
- Quantifying qualitative data carelessly, such as reporting percentages from ten interviews.
- Using a method without understanding its analysis. Plan the analysis before collecting data.
- Leaving the decision too late, after ethics deadlines have passed.
- Not justifying the choice in the methodology chapter.
Quantitative or qualitative for international and Nigerian students
Many students from Nigeria and across Africa study with UK universities, in person or through transnational programmes. A few points are worth special attention.
- Think about access. Surveys can be distributed online across borders, but response rates and sampling frames can be uncertain. Interviews by video call need reliable connectivity and agreed time zones.
- Consider language and translation. If interviews will be conducted in another language, plan for translation and how you will check meaning.
- Check local data sources. If you plan secondary data analysis, verify the quality, availability and licensing of Nigerian or African datasets before committing.
- Plan ethics for two contexts, your UK university and any local gatekeepers or authorities.
- Be explicit about positionality if researching your own community or workplace.
- Justify your choice clearly. UK markers reward reasoning, not just description.
Our online dissertation support services and Masters dissertation support are designed with these situations in mind.
A quick decision checklist
- My research question is clear and I have underlined its key verbs.
- I know whether it asks for measurement, understanding or both.
- I have checked what my department permits and expects.
- I can access a suitable sample or dataset.
- I have the skills, or a realistic plan to learn them in time.
- My timeline allows for design, ethics, data collection, analysis and writing.
- I can explain why this approach beats at least one alternative.
- I have discussed the choice with my supervisor or a mentor.
- I have planned the analysis before collecting data.
- I understand the limitations and can state them honestly.
Academic integrity when choosing and reporting your approach
Your methodological choices should be your own reasoned decisions, reported honestly. Do not describe procedures you did not carry out, and cite every framework or technique you draw on. Seeking feedback from your supervisor, an academic skills adviser or a mentor is sensible, as long as you follow your university’s rules on outside help. For more, see how to avoid plagiarism in your dissertation, and for the standards UK institutions work to, the QAA’s UK Quality Code.
Frequently asked questions
What is the main difference between qualitative and quantitative research?
Quantitative research measures variables with numbers and tests relationships statistically. Qualitative research explores meanings and experiences in depth using words, observations and documents.
Which is easier for a dissertation, qualitative or quantitative?
Neither is simply easier. Quantitative work demands statistical skills and enough respondents. Qualitative work demands interviewing skill and time-intensive analysis. Choose the approach that fits your question and strengths.
Which is better for a dissertation?
The one that best answers your research question. Neither is inherently superior. Markers reward fit, justification and rigour.
Can I do a dissertation without statistics?
Yes, if your question suits a qualitative, doctrinal or literature-based approach and your department permits it. Check your programme’s requirements.
Can I use both qualitative and quantitative methods?
Yes, as a mixed methods design, but only if your question genuinely needs both and you have the time, skills and word count. A single well-done method often scores higher than a thin mixed design.
How many participants do I need?
It depends on design and analysis. Quantitative surveys often need 100 or more, while qualitative interviews commonly range from about 8 to 20. Always justify your number and follow your department’s guidance.
Can qualitative research be generalised?
Not statistically. It aims for transferability: giving enough context that readers can judge whether findings apply elsewhere, and for theoretical insight.
What software do I need?
Quantitative projects commonly use SPSS, R, Stata or Excel. Qualitative projects often use NVivo or ATLAS.ti, or manual coding. Check what your university provides.
Can I change my approach after I start?
Small adjustments are normal. A major change usually needs supervisor agreement and may require a revised ethics application, so act early and report the change honestly.
Can I use secondary data?
Often yes, where your university permits and the data fit your question. Check licensing, documentation and variable definitions, and explain the limits of using data collected for another purpose.
Can someone help me choose and justify my approach?
Yes, within your university’s rules. Supervisors, academic skills advisers and mentors can help you test your reasoning. Writers.ng offers guidance and feedback so you can reach your own decision.
Continue through the dissertation process
Here is the usual order of the chapter guides.
- How to structure a dissertation
- How to write a dissertation proposal
- How to choose a dissertation topic
- How to write a dissertation literature review
- Dissertation methodology chapter
Explore our wider support:
- Dissertation help UK: expert support from topic to submission
- Online dissertation support services UK
- Dissertation data analysis help UK (SPSS, NVivo, R, Stata)
- Masters dissertation support UK
- Free dissertation resources UK
Not sure which approach fits your question?
Share your research question, your programme and your deadline. We will give honest, specific feedback on which approach fits, how to justify it and what it will take to do well, and leave the decision in your hands.