Choosing your dissertation methodology feels overwhelming mainly because it gets tackled backwards, picking between qualitative, quantitative or mixed methods before the research question is properly settled. Once that question is doing the work it should, the method it needs becomes a narrower, calmer decision, and the paradigm, sampling and ethics choices around it start to make practical sense rather than reading as jargon.
Key takeaways
- Your research question should drive your methodology, not the other way round; a clear question usually points towards qualitative, quantitative or mixed methods on its own.
- Qualitative methods explore meaning in depth with small, purposive samples; quantitative methods measure and test relationships with larger samples; mixed methods combine both when one alone will not do.
- Your research paradigm, how you view knowledge and reality, matters more as you move from undergraduate to masters by research and PhD level, where you must name and justify it.
- Sampling, data collection and analysis need to line up logically with your method and paradigm, not be chosen separately from a menu.
- Ethics approval from your university’s Human Research Ethics Committee is required before you collect data from people, and it takes longer than most students expect.
Start with the question, not a method you already like
Methodology overwhelm usually has one cause: the decision gets made backwards. A student picks a survey because it feels manageable, or interviews because they sound more interesting than statistics, then bends the question to fit. Reverse the order and it gets easier. Ask what you would need to know, and from whom, to answer your question. A question about experience or meaning points towards qualitative methods. A question about how much, how many, or whether one thing predicts another points towards quantitative methods. A question with a genuine numeric part and a genuine “why or how” part is a candidate for mixed methods, not a sign you have failed to narrow things down.
If the question itself keeps shifting, methodology will feel unstable no matter which method you pick, and that is a literature review problem more than a methodology one; see the guide to getting unstuck on your literature review. If the real problem is less “which method” and more “I cannot open the document”, that is closer to burnout, covered in the guide to dissertation burnout and motivation. This article sits inside the wider dissertation and thesis survival guide, worth a look if methodology is only one of several things weighing on you.
Qualitative, quantitative or mixed methods, in plain English
Strip away the jargon and the three approaches simply answer different kinds of questions. Qualitative research asks what something is like or why it happens, using words as data. Quantitative research asks how much, how many, or whether a relationship exists, using numbers as data. Mixed methods use both, side by side or in sequence, because the full answer needs a number and a narrative.
Qualitative methods
Qualitative methods suit questions about experience, meaning or process, especially where little is already known. Common methods include interviews, focus groups, a case study of one organisation or community, ethnographic observation, and analysis of existing documents. Data usually comes back as transcripts, field notes or documents, worked through with thematic, content or narrative analysis.
Quantitative methods
Quantitative methods suit questions about frequency, magnitude or the relationship between variables, where you want to generalise across a larger group. Common methods include structured surveys, true or quasi experiments, and secondary analysis of existing datasets. Data comes back as numbers, scores or rates, analysed with descriptive statistics and, often, inferential tests.
Mixed methods
Mixed methods combine a qualitative strand and a quantitative strand deliberately, not as a way to avoid choosing. A common design collects survey data first, then uses interviews to explain an unexpected result, or the reverse. Mixed methods means more work, not less, since both strands need planning, collecting and analysing before being brought together at interpretation, so choose it only when the question genuinely has two parts.
The table below lines the three approaches up side by side.
| Qualitative | Quantitative | Mixed methods | |
|---|---|---|---|
| Best for | Exploring experience, meaning or process in depth, especially where little is already known | Measuring frequency, magnitude or the relationship between variables across a group | Questions with both a numeric part and a why or how part |
| Typical data | Transcripts, field notes, documents, open ended responses | Scores, counts, rates, closed survey responses | Both of the above, collected together or in sequence |
| Typical methods | Interviews, focus groups, case study, ethnography, document analysis | Surveys and questionnaires, experiments, quasi experiments, secondary data analysis | A qualitative strand and a quantitative strand, combined by design |
| Analysis | Thematic, content or narrative analysis | Descriptive statistics, often with inferential tests | Each strand analysed separately, then integrated at interpretation |
| Sample size | Usually small, guided by depth and saturation rather than a target figure | Usually larger, often guided by a power calculation or target response rate | Varies by strand, a smaller qualitative sample alongside a larger quantitative one |
Read the “best for” row against your own question first; the rest usually falls into place once that match is right.
How much your research paradigm actually matters
A research paradigm, sometimes called a research philosophy or worldview, is your position on what counts as valid knowledge and how reality can be known. It sounds abstract, but it is the reasoning behind your method, not a separate hoop to jump through. Positivism assumes a single, measurable reality and sits under quantitative work. Interpretivism, or constructivism, assumes people construct meaning differently and sits under qualitative work. Pragmatism starts from the question and picks whatever methods answer it best, which is why it pairs so often with mixed methods. A critical or transformative worldview goes further, treating research as a way to challenge power imbalances, and turns up more in social work, education and health related topics.
How much this needs spelling out depends on your level of study. In an undergraduate or honours dissertation, a paragraph naming your paradigm is often enough. At masters by research and PhD level, examiners typically expect a dedicated section justifying the paradigm and connecting it explicitly to your method, sampling and analysis. A mismatch here, a quantitative method bolted onto interpretivist language, or the reverse, is a common reason a research proposal gets sent back at confirmation. If a supervisor keeps waving you towards a paradigm without explaining the reasoning clearly enough for you to defend it later, raise it directly; the guide to dissertation supervisor problems covers how.
Getting the reasoning right is one thing, setting it out in the register examiners expect is another. Our Masters and PhD qualified writers regularly help postgraduate students turn a clear plan into a methodology chapter that reads that way on paper, and our editors can tighten a draft you have already written without changing what it says.
Sampling, data collection and analysis, in outline
Once the method is settled, three decisions follow in order: who or what you will sample, how you will collect data, and how you will analyse it. Qualitative studies typically use purposive sampling, deliberately selecting participants who can speak richly to the question, and stop recruiting once new interviews stop adding insight, often called saturation. Quantitative studies more often use random or stratified sampling when the aim is to generalise, with sample size set by a power calculation or a practical target response rate.
Data collection tools need to match the method: an interview guide with open prompts for qualitative work, a piloted questionnaire for quantitative work, a clear protocol for accessing existing datasets if that is your source. Plan analysis before data collection starts, since it partly shapes how you should be collecting data in the first place. If turning transcripts or a spreadsheet of responses into a coherent findings section is where you expect to get stuck, our data analysis and writing help works through the analysis with you and helps write it up clearly.
None of this needs to be elaborate to be defensible. A short, honest paragraph explaining why you chose this sample, this size and this analysis, tied back to your question and paradigm, does more for your mark than a long methods section padded with textbook definitions. If time is now the bigger issue than the method itself, the guide to finishing a dissertation when you are running out of time covers scoping a feasible study without abandoning rigour.
Ethics considerations you cannot skip
Ethics approval is not paperwork bolted onto the methodology, it is part of it. Most Australian universities will not let you collect data from people until a Human Research Ethics Committee, usually shortened to HREC, has approved your application. Many HRECs apply the principles in the NHMRC National Statement on Ethical Conduct in Human Research, covering informed consent, minimising harm, privacy and the extra care needed for children, patients or other vulnerable groups. Your own university’s ethics office holds the specific forms, categories and timeframes that apply to you, so check directly rather than assuming your project matches a friend’s elsewhere.
In practice, this means writing a plain language information sheet, a consent form, and a clear explanation of how data will be stored and protected. Build in real lead time: low risk applications can still take several weeks, and anything involving children, patients or identifiable sensitive data usually takes longer. If your design changes after approval, even a small tweak to your interview questions, check whether it needs a formal amendment before you keep collecting data.
Aligning your method to your question, and getting the chapter written
Before writing the chapter, test the whole chain out loud: does this method answer this question, does the paradigm explain why the method makes sense, does the sample fit what the analysis needs, and does your ethics approval cover what you are proposing. A chapter that survives that test is usually most of the way to being defensible at confirmation and, later, at a full thesis defence; see the guide to the PhD viva and thesis defence.
Most methodology chapters move through a similar sequence: a short restatement of the question, the paradigm and why it fits, the design, participants or data sources and sampling, data collection instruments, the analysis approach, ethics, and finally limitations. Limitations are not an admission of failure, every method has trade offs, and naming them shows the examiner you understand your design.
If you would rather talk it through than stare at a blank document, our dissertation coaches can work through your question, paradigm and method with you directly, then help turn that conversation into a structured chapter your supervisor can give real feedback on.
Frequently asked questions
How do I choose between qualitative and quantitative methods for my dissertation?
Choosing between qualitative and quantitative methods comes down to what your question is asking: experience or meaning points towards qualitative, frequency or relationships between variables points towards quantitative. If your question genuinely has both parts, consider mixed methods instead.
What is mixed methods research and when should I use it?
Mixed methods research combines a qualitative strand and a quantitative strand in one study, used when a single approach cannot fully answer your question. It suits questions with both a numeric part and a “why or how” part, but it means more work than choosing one approach alone.
What is a research paradigm and do I actually need to write about it?
A research paradigm is your underlying position on what counts as valid knowledge and how reality can be known. Undergraduate and honours dissertations often need only a short paragraph on it, while masters by research and PhD theses typically expect a dedicated, justified section.
How do I decide on my sample size?
Deciding on your sample size depends on your method: qualitative studies are usually guided by depth and saturation rather than a fixed number, while quantitative studies are often guided by a power calculation or a realistic response rate. Your supervisor can help you settle on a defensible figure.
What ethics approval do I need for dissertation research in Australia?
Ethics approval for dissertation research in Australia almost always means applying to your university’s Human Research Ethics Committee before collecting any data from people. Requirements and timeframes vary by institution and risk level, so confirm the process with your own ethics office early.
Can I change my methodology after my proposal or confirmation has been approved?
Changing your methodology after approval is usually possible but rarely simple, since it can affect your ethics approval and your candidature milestones. Talk to your supervisor as soon as you suspect a change is needed, and check your university’s policy on amendments.
How long should my methodology chapter be?
How long your methodology chapter should be depends on your degree level and your university’s own guidelines, so there is no single figure that applies everywhere. It should be long enough to justify your paradigm, design, sampling, data collection, analysis and ethics clearly, and no longer.
Can someone help me write or edit my methodology chapter?
Yes, our writers and editors can help you plan, structure, draft or edit a methodology chapter based on the question, paradigm and method you have already chosen. We work from your own thinking and never generate data or present the work as supervised research you did not do.