PhD Guidance
How to Structure a Strong Research Proposal
A research proposal is an argument, not a form. This guide walks through the sections committees expect and how to make them work together.
A research proposal has one job: to convince a committee that your question matters, that it has not already been answered, and that you have a credible plan to answer it. Every section should contribute to that argument.
This guide follows the order most committees expect, explains what each section has to achieve, and ends with a checklist you can use before you submit. Your university’s template always takes priority, so map these sections onto the headings it asks for.
Start with the problem, not the topic
Topics are broad; problems are specific. “Employee engagement in the IT sector” is a topic. “Why do engagement interventions in Indian IT firms show weaker retention effects than the international literature predicts?” is a problem. Committees approve problems.
A useful test: can you state the problem in two sentences, one describing what is known and one describing what is not? If you cannot, keep narrowing. A well-defined problem also tells you which literature to review and which data you will need.
The sections committees expect
| Section | What it must do | Typical length |
|---|---|---|
| Introduction and background | Set the context and show why the problem matters now | 1 to 2 pages |
| Problem statement | State the specific gap your study addresses | Half a page |
| Objectives and research questions | Turn the problem into precise, answerable questions | Half a page |
| Literature review | Show what is known, what is contested and where the gap sits | 3 to 6 pages |
| Methodology | Explain the design, sample, instruments, analysis and ethics | 3 to 5 pages |
| Expected contribution | Say what will be different once the study is complete | Half a page |
| Timeline and resources | Show a realistic plan the committee can believe | 1 page |
Write research questions that can be answered
Good research questions are specific, few in number and matched to a method. Three to five is usually enough. Each question should name the concepts involved, the population or context, and the kind of answer you expect: a description, a comparison, a relationship or an explanation.
- Too broad: “What affects employee engagement?”
- Answerable: “How does perceived supervisor support relate to engagement among early-career software engineers in Bengaluru?”
- If your design is quantitative, hypotheses should follow directly from the questions and the literature.
Position your study in the literature
The literature review in a proposal is shorter than the one in your thesis and has a narrower job: to show that you know the main debates and that your gap is real. Organise it by themes or arguments rather than by author, and end it by stating the gap in a sentence that matches your problem statement.
Justify every methodological choice
Committees scrutinise methodology more closely than any other section. For each choice, give the reason: why this design answers your questions, why this sample is appropriate and reachable, why these instruments are valid, and how you will analyse the data.
- Design: quantitative, qualitative or mixed, and why it fits the questions.
- Sampling: who, how many, how they will be selected, and whether you can reach them.
- Instruments: validated scales, interview guides or the data sources you will use.
- Analysis: the technique you will use for each research question.
- Ethics: consent, confidentiality and the approvals you will need.
Make the sections talk to each other
The most common weakness we see is disconnection. The gap identified in the review should lead directly to the research questions, and the methodology should be justified by those questions. If a reader could remove a section without the argument changing, that section needs rewriting.
A proposal is read in sequence but judged as a whole. Consistency across sections is what signals readiness.
Plan a realistic timeline
Break the study into phases, such as literature review, instrument development, data collection, analysis and writing, and give each one a duration. Allow time for ethics approval and for supervisor feedback, which usually take longer than expected. A simple Gantt chart shows the whole plan on one page.
Before you submit
- Check that every research question has a matching analysis method.
- Confirm the sample size or participant plan is achievable within the timeline.
- Align terminology: the same construct should have the same name throughout.
- Format references and headings to your institution’s template.
- Ask someone outside your field to read the problem statement. If they cannot explain it back to you, simplify it.
- Read the whole proposal once from start to finish, checking that the gap, questions and methods tell one story.
A strong proposal does not need to be long. It needs to be coherent: a clear problem, questions that follow from it, and a plan that can answer them.