How to Write the Methodology Chapter (Without Writing a Textbook)

The short answer

A methodology chapter describes what you did and why, in enough detail that someone else could repeat it. It covers your research design and its justification, how you selected your sample or material, how you collected data, how you analysed it, and what the approach cannot tell you. Around 10 to 15 percent of the thesis is typical.

The single most common failure is writing a textbook summary of research methods instead of an account of your own study. Marks come from justified choices, not from demonstrating that you know what a survey is.

What the chapter is actually for

It does three jobs, and only the first is obvious:

• Transparency. A reader can see exactly how you reached your results, and could in principle repeat the study.
• Justification. You show your approach was appropriate to the question — not merely that it was one of the available options.
• Boundary-setting. You define what your results can and cannot support, which protects you in the discussion and at the viva. A stated limitation is far harder to attack than one an examiner discovers.

The structure that works

Conventions vary by discipline, but this order is defensible in most:

1. Research design. Qualitative, quantitative or mixed; exploratory or confirmatory. One paragraph, with the reason.
2. Justification. Why this design answers this question better than the obvious alternative. See below — this is the part that earns marks.
3. Sample or material. Who or what, how many, how selected, why that.
4. Data collection. Instruments, procedure, timeframe, setting. Enough that it could be repeated.
5. Data analysis.The specific procedure, named and referenced. “The interviews were analysed thematically” is not a procedure; how you coded, in what order, against what scheme, is.
6. Quality criteria. Reliability and validity, or credibility and transferability in qualitative work.
7. Ethics, where human participants are involved — consent, anonymisation, approval.
8. Limitations.Sometimes placed in the discussion instead; check your department's convention.

How much method theory to include

Less than you think. A good rule: for every paragraph explaining what a method is, you should have several describing what you did with it.

Cite the methodological source and move on. Your reader does not need to be taught what a semi-structured interview is — they need to know how many you conducted, how long they ran, what your guide covered, and how you decided when to stop.

The exception is a method that is unusual in your field or that you have adapted. Then explain it properly, because the adaptation is a contribution and needs defending.

How to justify a choice properly

A justification is a comparison, not an assertion. “A qualitative approach was chosen because it allows deeper insight” justifies nothing — it would be equally true of any study. The structure that works:

1. What the question requires.
2. What the alternative would have given you.
3. Why that is insufficient here.
4. What your choice gives instead, and at what cost.

Worked example: “Because the study asks how practitioners interpret the policy rather than how many follow it, a survey would have produced frequency data without access to the reasoning behind responses. Semi-structured interviews were therefore chosen, accepting that the resulting sample is too small to support claims about prevalence.”

That last clause is doing real work. Naming the cost of your own choice is what marks a methodologically aware writer, and it also pre-empts the most obvious criticism.

Writing limitations without undermining yourself

Students often either omit limitations or apologise through them. Neither reads well. State them plainly, then say what follows:

• Weak: “Unfortunately the sample was very small, which is a major problem.”
• Strong: “The sample of eight participants supports analytical rather than statistical generalisation; findings indicate patterns that warrant testing at scale rather than establishing prevalence.”

Same fact, and the second shows you understand what your data can carry. That is the competence being assessed.

Tense, voice and the small conventions

• Past tense for what you did— “participants were recruited”. Present tense for the method in general — “thematic analysis identifies patterns across a data set”.
• Passive voice is conventional here in most disciplines, more so than anywhere else in the thesis. Some departments now prefer active. Check recent theses from your own department.
• Be specific with numbers.“Several interviews” is not repeatable; “eleven interviews of 45 to 70 minutes” is.
• Put instruments in the appendix — interview guides, questionnaires, coding schemes — and reference them from the chapter.

Mistakes that cost marks

• Describing methods without justifying them. The most common, and the most expensive.
• A textbook chapter with your study bolted on. Four pages on epistemology, one paragraph on what you actually did.
• Analysis described in one sentence. How you got from raw data to findings is the part examiners scrutinise most, and it is routinely the shortest section.
• No account of deviations. If recruitment failed and you changed approach, say so — that is normal research, and concealing it is worse than the deviation.
• Limitations that are not really limitations.“More time would have been useful” is true of everything ever written.

One note on how this chapter reads

The methodology chapter is the most formulaic writing in a thesis — standardised procedure descriptions, passive constructions, repeated sentence patterns. That is correct and expected; the conventions exist to make studies comparable.

It is also the section AI detectors flag most often, because regular, predictable prose is precisely what those tools measure. If your university runs a detector, it is worth knowing what your own methods chapter looks like to one before someone else tells you. Our AI check is free for the first 1,500 words and publishes, beside every verdict, how often that verdict is wrong about verified human writing — the measurements are here. If it flags your methods chapter, that is usually a fact about the genre rather than about you.