A research proposal can have an interesting question and still leave a reader unsure how the study would work. The methods section closes that gap. It explains what evidence you need, how you would obtain it and how the analysis would address the question. For a PhD application, the aim is a credible plan with clear reasoning, while acknowledging details that will develop through supervision.
Begin with the evidence you need
Write your main question above the methods draft. Then describe what you would need to observe to answer it. A question about how people interpret an experience requires different evidence from a question about the association between measured variables. This does not mean one type of evidence is better. It means the proposed design must fit the particular uncertainty you are investigating.
For a hypothetical project on postgraduate feedback, you might want to understand how students use comments when revising drafts. Interviews could explore their explanations, while comparing document versions could show changes in the writing itself. Explain what each source would contribute. Do not add multiple methods merely to make the proposal appear sophisticated.
Explain the setting and selection process
Identify the population, setting, materials or cases you intend to study. State the selection criteria and explain why they are appropriate. If you propose interviews, describe the relevant participant characteristics and recruitment route. If you propose analysing documents, explain which documents qualify and how you would find them. A clear selection process makes the scope easier to assess.
Be careful with precise sample numbers that have no justification. Depending on the design, an appropriate size may depend on statistical assumptions, the depth of qualitative work, available cases or resource constraints. Explain your reasoning at the level possible in the application. Where a calculation or refinement is still needed, acknowledge it rather than presenting an arbitrary figure as settled.
Connect collection with analysis
Describe the sequence from obtaining evidence to producing findings. Name the proposed analytical approach and explain why it helps answer the question. “The data will be analysed” is too vague. A more useful plan explains whether you will compare patterns, develop themes, estimate relationships or test a model, and how those operations connect to the research objective.
Distinguish software from methodology. Naming a coding package or statistical language does not explain the analysis. State what decisions the researcher will make, how variables or categories would be defined and how interpretations would be checked. Keep the explanation accessible enough for a reader who understands the field but does not use your preferred tool.
Address feasibility and research responsibilities
Identify access requirements, training needs and likely dependencies. If the project needs an organisation’s cooperation, say whether access is confirmed or exploratory. Explain how you would manage a delay or unavailable dataset. A realistic alternative can make the proposal stronger when it preserves the central question rather than replacing the project with something unrelated.
Consider consent, confidentiality, secure handling of information and any relevant review processes. Do not promise approval before it has been granted. At the application stage, you can show awareness of the issues and describe the process you expect to follow. Avoid including identifiable participant information in an illustrative proposal.
Show the limits of the design
Explain what the study could establish and what it could not. A narrow sample may support a detailed account of one setting without representing every institution. An observational association does not automatically establish causation. Clear limits help reviewers see that the proposed claims will match the evidence rather than exceed it.
Finish by checking alignment across the proposal. Each major question should have a corresponding source of evidence and analytical step. Remove methods that do not serve a clear purpose, and identify questions that the design cannot yet answer. A concise, coherent plan is easier to defend in an interview and easier to develop into a practical doctoral project.
Academic Guide Series 300 | Item 006