Guide
Choosing a research methodology: qualitative, quantitative or mixed
How to get from a research question to a design you can defend, and what examiners look for in the methodology chapter.
The question chooses the method
Method follows from the research question, not from preference or habit. Look at the verb in your question. "How many", "how much", "to what extent" and "does X affect Y" call for measurement. "How", "why" and "what is the experience of" call for interpretation. Questions that ask both need both.
The three approaches compared
| Quantitative | Qualitative | Mixed methods | |
|---|---|---|---|
| Purpose | Measure, test relationships, generalise | Explore, explain meaning and process | Combine breadth with depth |
| Typical data | Numbers from surveys, experiments, records | Words and observations from interviews, documents, fieldwork | Both, collected together or in sequence |
| Analysis | Statistical | Thematic, content, discourse or similar | Each strand analysed, then integrated |
| Main risk | Measuring precisely something that does not capture the concept | Claims that go beyond what the cases can support | Two shallow studies instead of one integrated design |
Doctrinal legal research, which analyses legislation, case law and commentary, and historical or archival research are established approaches in their own right. They need the same justification: which sources, selected how, and analysed by what method.
Five decisions to make and justify
- Design. Experiment, survey, case study, ethnography, doctrinal analysis, comparative study, or another recognised design.
- Data or sources. What counts as evidence for each research question.
- Selection. Population and sampling for empirical work; selection criteria for cases, jurisdictions or documents.
- Collection. Instruments and procedures, and how you tested them.
- Analysis. The specific technique for each question, chosen before you look at the results.
Check feasibility early
A design is only as good as your ability to carry it out. Confirm access to participants, records or archives; the time each stage needs; the skills and software required; and the ethical approval your institution demands before any data is collected.
What the methodology chapter must show
- The link between each research question and the data and analysis that answer it.
- The reason for each choice, with reference to the methodological literature.
- Enough detail that another researcher could follow what you did.
- How you addressed quality: validity and reliability, or credibility and trustworthiness, as your approach requires.
- Ethical considerations and how consent, confidentiality and data storage were handled.
- Limitations, stated plainly.
Common mistakes
- Choosing the method first and bending the question to fit it.
- Describing procedures without justifying them.
- Deciding on the analysis only after the data is collected.
- Calling a study "mixed methods" when one strand is a token addition.
- Omitting limitations, which examiners will then raise for you.
Sources and further reading
The EQUATOR Network lists reporting guidelines for most study designs, which are also a useful checklist at the design stage. Your institution's research ethics policy governs approval.
Common questions
What is the difference between methodology and methods?
Methods are the tools you use to gather and analyse data, such as interviews, surveys or document analysis. Methodology is the reasoning behind the overall approach: why this design and these methods are a sound way to answer this question.
Is mixed-methods research better than a single method?
Not by default. Mixed methods suit questions that need both breadth and depth, but they take more time and skill, and the two strands must be genuinely integrated. A well-executed single-method study is stronger than a thin mixed one.
How large should my sample be?
It depends on the design. Quantitative studies should justify sample size from the analysis planned. Qualitative studies justify it from the depth of data needed to answer the question. State the reasoning in the chapter.
Need to justify your design?
We give guidance on research design and on analysing your own data. Send your research questions with an enquiry.
See plans and feesAsk a question