How to Write a Research Proposal Methodology Section: Justify Every Choice

# How to Write a Research Proposal Methodology Section: Justify Every Choice

Your methodology section answers one question: why did you choose to do it this way? A supervisor reads it looking for gaps. Every design decision, every sampling choice, every question on your instrument, and every statistical test needs a reason attached. If you cannot give one, the supervisor will ask for it — and you will be revising under pressure.

This guide walks through the decisions that attract the most questions, in the order you will make them.

1. Start with the design, and name it precisely

Do not write "the study will use a quantitative approach." That says nothing. Name the design: cross-sectional survey, quasi-experiment, case study, phenomenological inquiry, mixed-methods explanatory sequential.

Then justify it against your research question. A design is justified when it can actually answer the question you asked. If you want to know how common something is, you need a design that measures prevalence in a defined population. If you want to know how people experience something, you need a design that captures meaning, not frequency.

The test: Can you complete this sentence? "A [design] is appropriate here because the research question asks [X], which requires [type of data]."

If you cannot, the design is not yet justified.

[unsupported — needs source]

2. Define your population, then your sampling frame

Two different things, often confused.

• Population: everyone the finding is meant to apply to. Be specific: "second-year undergraduate students enrolled in the Faculty of Education at [university] in the 2024/2025 academic year."
• Sampling frame: the actual list you will draw from. A class register, a staff database, a patient log.

Your supervisor will ask why you chose this population and not another. Answer in terms of your research question and access. If you are studying a programme offered at three universities and you can only reach one, say so, and say what that limits.

The test: Could another researcher, given your description, identify exactly who was eligible and who was not?

[unsupported — needs source]

3. Justify the sampling technique and the sample size

Technique

State whether the sampling is probability or non-probability, and which specific method: simple random, stratified, cluster, purposive, snowball, convenience.

Then justify it. Convenience sampling is not automatically wrong. It is wrong when you present it as representative. If you used it because you had access to one campus and no sampling frame, write that. Then state the limitation it creates.

Size

Give a number and a reason. Reasons include:

• A calculation (for example, a sample size formula with stated parameters).
• A rule of thumb from a methods text, cited.
• Practical constraints — time, cost, access — stated openly.

Do not write "as many respondents as possible." That is not a sample size.

The test: Can you explain, in two sentences, why this number and not half or double it?

[unsupported — needs source]

4. Describe your instrument, item by item

If you use a questionnaire, an interview guide, or an observation checklist, the reader needs to know:

• What it measures. Each section should map to a research question or objective.
• Where it came from. Adapted from an existing instrument, or developed for this study.
• Whether it was validated. If you adapted a published instrument, say what you changed. If you built your own, say how you checked it made sense.
• How it was piloted. Who tested it, how many people, and what changed as a result.

If you developed the instrument yourself, expect the question: "How do you know it measures what you think it measures?" Have an answer ready. A pilot with a small group of people similar to your sample is a common one.

The test: Can you point to each item and say which research question it serves?

[unsupported — needs source]

5. Explain your data collection procedure step by step

Write this as a sequence a stranger could follow.

1. How participants were approached.
2. What they were told before consenting.
3. How consent was recorded.
4. Where and when data was collected.
5. How long each session took.
6. How data was stored.

This section is where supervisors look for ethical gaps. If your study involves human participants, you will likely need clearance from a recognised body before collecting data. In Uganda, the Uganda National Council for Science and Technology (UNCST) is described as the national body that registers research involving human participants. Treat that description as unverified here — confirm the current requirement and procedure with UNCST directly or with your supervisor before you rely on it.

[unsupported — needs source]

6. Match your analysis to your data and your question

Do not list tests you have heard of. List the ones your data can actually support.

• Quantitative: name the specific test — descriptive statistics, chi-square, t-test, regression — and say what it will tell you. State your software (for example, SPSS, R, Stata).
• Qualitative: name the approach — thematic analysis, content analysis, discourse analysis — and describe the steps. Transcription, coding, theme development, member checking.
• Mixed methods: explain how the two strands connect. Which comes first, and how does one inform the other?

The test: For each research question, can you name the analysis that answers it?

[unsupported — needs source]

7. State your limitations before your supervisor does

A limitations paragraph is not an admission of failure. It shows you understand your own design.

Write plainly:

• What your sample cannot tell you.
• What your instrument cannot capture.
• What your design cannot prove.

Then say how you mitigated each one, where you could. If you could not, say that too.

This is the paragraph that most often turns a defensive supervision meeting into a productive one. You have already named the weakness. The supervisor's job shifts from finding it to helping you handle it.

[unsupported — needs source]

A final check before you submit

Read your methodology section and ask, for each paragraph: "Why?"

If the answer is in the text, you are done. If the answer is only in your head, write it down.

Related reading

• How to Write a Research Proposal: A Practical Guide for University Students in Uganda
• How to Structure a University Dissertation, Chapter by Chapter
• How to Find Credible Academic Sources for a Literature Review
• APA 7 Referencing Explained: In-Text Citations, Reference Lists, and Common Errors
• Using AI in Academic Research Responsibly: Accuracy, Citations, Originality, and Academic Integrity

Note on sources: No sources were supplied for this piece, so no citations appear. Every claim that would normally carry a reference is marked [unsupported — needs source]. These markers are accurate and should remain visible. To add citations, supply a research methods textbook, your department's research handbook, UNCST's published guidance, and any Uganda-specific source on language and translation in data collection.