How to Write a Research Proposal for a PhD Scholarship (With Outline and Examples)

A PhD scholarship is not awarded for a degree. It is awarded for a project. The research proposal is where that project lives: it is the document that convinces a supervisor to attach their name to you, a university to offer you a place, and a funder — a government scheme, a research council or a university scholarship panel — to spend three or four years of money on your idea. Strong grades get your file opened. The proposal is what gets it funded.

This guide explains how to write a research proposal for a PhD scholarship from first idea to final draft: what reviewers are actually testing, the structure that universities themselves publish, realistic word counts (which vary more than applicants expect), how to write each section, and two complete original outline examples — one in the sciences, one in the social sciences — written for this guide and clearly labelled as samples. It closes with the ethics, timeline and budget elements applicants forget, the mistakes that sink proposals, and answers to the questions doctoral applicants ask most. When you are ready to find schemes to apply to, browse our scholarships directory and our guide to fully funded PhD scholarships for international students.

What a PhD Scholarship Proposal Must Prove

University guidance is remarkably consistent about what a proposal is for. The University of Warwick’s PhD proposal guidelines state that it should evidence your ability to plan and organise independent research and the relevance of your topic to a senior academic who could supervise you. Aston University’s guidance builds the same test into its structure: clear aims, a rationale grounded in recent literature and the gaps in it, a defensible method, and a timeline matched to the funding period. Reviewers — who may include a non-specialist panel member for scholarship competitions — are silently asking five questions:

1. Is there a real question? Not a topic (“climate change and farming”) but a question with an answer you do not already know.
2. Is it new? Does it fill a gap you can point to in published work, rather than repeat it?
3. Can it be done? Are the data, sites, equipment, participants and skills actually reachable within the scholarship period?
4. Can you do it? Does your background show the training and preparation this specific project needs?
5. Does it belong here? Does it fit the supervisor’s and department’s research — the people who must support it for years?

Every section of a good proposal exists to answer one of those questions. If a paragraph answers none of them, cut it.

How Long Should a PhD Research Proposal Be?

There is no universal length, and inventing one would mislead you. Published university guidance for application-stage proposals ranges widely:

The pattern behind the numbers: the more the proposal is a first conversation with a potential supervisor, the shorter it is; the more it is a formal scholarship or studentship application, the longer and more complete it must be. Government scholarship routes sit at the formal end — a CSC doctoral study plan, a DAAD research proposal, or a Commonwealth PhD application each expects the full argument. Your binding figure is always the one in your target programme’s or university’s current instructions. When no figure is given, 1,500–2,000 words plus references is a defensible default for a scholarship proposal. The Aston University proposal guidance and the University of Bristol’s research proposal page are worth reading in full before you draft, precisely so you can see how real universities phrase their expectations.

The Outline: A Structure That Works Across Disciplines

Working title

A proposal title should do three jobs: name the phenomenon, signal the approach or angle, and locate the study. “An investigation into education” does none. “Does adaptive tutoring software close the mathematics gap? A randomised study in rural secondary schools” does all three. Keep it under about 20 words, avoid jargon a neighbouring discipline would not understand, and remember it is a working title — supervisors expect it to evolve.

Abstract

Write it last. In 150–200 words: the problem, the gap, your question, your method in one sentence, and the contribution in one sentence. Scholarship panels often screen on the abstract before reading further, and a non-specialist member may read little else — this paragraph carries disproportionate weight.

Introduction and rationale

Open with the real-world or scholarly problem, sized honestly (who is affected, at what scale, at what cost — with sources for any figures you use). Then narrow: what is not yet understood about this problem, and why that ignorance matters. End the introduction by stating, in plain words, what your project will do about it. Resist the temptation to begin with the dawn of the discipline; begin with the problem.

Research questions and objectives

One primary question, two to four sub-questions, and objectives that mirror them (“To determine whether…”, “To measure…”, “To explain…”). Each sub-question must be answerable by the methods you later describe — reviewers check the join. If a question cannot be answered with reachable data, either redesign the method or drop the question now, not in your second year.

Literature review

This is not a summary of everything you have read; it is an argument that ends at your gap. Organise it by theme or debate, not author by author. For each strand: what the key studies found, what they assumed or missed, and how that limitation builds the case for your project. Cite recent work prominently — a review that stops five years ago tells the panel you have not entered the current conversation. End with a short paragraph: “What remains unknown is X. This project addresses X by Y.” That sentence is the hinge of the whole proposal.

Methodology

The section panels scrutinise hardest. Cover, in order: your overall design and why it fits the questions; your data — what it is, where it comes from, sample or corpus size and how you chose it; your collection procedures, including access (who has agreed, or how you will secure agreement); your analysis approach, including software or techniques; ethics — consent, confidentiality, data storage, and the approval process your university requires, especially for research with people, children, patients or sensitive data; and risks and mitigations — what could go wrong (access refused, dataset delayed, recruitment shortfall) and what you would do instead. A named contingency is a mark of a researcher; its absence is a mark of a student who has not yet imagined reality.

Timeline

Show a phased plan across the standard funded period (usually three years full-time, sometimes three and a half or four). Be honest about overlap — literature work, ethics approval, fieldwork and writing interleave — and include the fixed points: confirmation/upgrade review, data-collection windows, analysis, chapter drafting, and a realistic submission buffer. A plan that allocates the final year entirely to “writing up” with nothing drafted earlier is a known fantasy; show drafting throughout.

Expected outcomes and contribution

Separate the two. Contribution is what knowledge will exist afterwards that does not exist now — a tested explanation, a new dataset, a method adapted to a new context. Outcomes are the deliverables and beneficiaries: the thesis, papers, a policy brief, a tool, training for practitioners. Scholarship funders with development missions — DAAD EPOS, for instance, whose official database entry asks doctoral candidates to show development-related motivation — also read this section for what changes in your home context when you return.

Fit and preparation

Name the fit concretely: the supervisor’s recent publications your project extends, the department’s centre or dataset you will use, the training you will take. And your own preparation: the master’s dissertation, methods courses, languages, fieldwork or professional experience that make this project credible in your hands. If you have already corresponded with the supervisor, say so factually; our guide on how to email a professor for a PhD scholarship shows how to open that conversation months before the deadline.

Example 1 (Sample Outline — Science): Low-Cost Sensors for Flood Early Warning

Original sample written for this guide. It is an outline example for learning the structure — adapt it to your own genuine project; never submit copied text or a project you do not intend to carry out.

• Working title: Can community-maintained low-cost river sensors improve flood warning lead time in monsoon river basins?
• Abstract (skeleton): Flash and riverine floods in [region] are currently forecast from sparse official gauges. This project will deploy and validate a network of low-cost ultrasonic river-level sensors maintained by trained community volunteers, test whether their data — assimilated into a simple hydrological model — extends usable warning lead time, and identify the maintenance model that keeps such networks alive beyond pilot funding.
• Rationale: Official gauge density is low in the study basin; published pilots of low-cost sensing are short and rarely report multi-monsoon reliability or community-maintenance outcomes — the gap this project addresses.
• Research questions: (1) How accurate are low-cost sensors against reference gauges across two monsoon seasons? (2) By how much does their data improve forecast lead time in the basin model? (3) Which community maintenance arrangements sustain sensor uptime above an agreed threshold?
• Literature strands: low-cost environmental sensing accuracy studies; flood forecasting and data assimilation; citizen-science network sustainability — each strand ending in the specific limitation the project answers.
• Methodology: 24 sensors installed alongside 6 reference gauges; two monsoon seasons of paired data; error and drift analysis; model experiments with and without sensor data; semi-structured interviews with volunteer maintainers; ethics approval for interview data, with consent and anonymised storage.
• Timeline: Year 1 — literature, ethics, procurement, installation before monsoon 1. Year 2 — second season, modelling, interviews. Year 3 — analysis, thesis and a deployment guidance note for the national hydrology agency.
• Contribution: the first multi-season reliability evidence for this sensor class in this basin type, an open dataset, and a costed maintenance model local agencies can adopt.

Example 2 (Sample Outline — Social Science): Remittances and Girls’ Secondary Schooling

Original sample written for this guide, for structure-learning only.

• Working title: Do international remittances keep girls in school? Evidence from migrant-sending households in [country/region].
• Abstract (skeleton): Remittances raise household income, but their effect on girls’ secondary schooling is contested: income effects may fund schooling while a parent’s absence may withdraw it. Using household survey data linked with a new qualitative study, this project estimates the net effect and explains the household decisions behind it.
• Rationale: Existing studies use older data, treat remittances as uniform, and rarely disaggregate by which parent migrated or who receives the money — distinctions policy can actually act on.
• Research questions: (1) What is the association between remittance receipt and girls’ secondary enrolment and completion, after accounting for household wealth? (2) How does it differ by migrant and recipient identity? (3) Through what household decision processes do these effects operate?
• Literature strands: migration–education economics; intra-household bargaining and gender; qualitative studies of left-behind children — converging on the disaggregation gap.
• Methodology: secondary analysis of the national household panel (most recent waves), with careful treatment of selection into migration (matching and sensitivity analysis); 30 household case studies in two high-migration districts; ethics approval covering interviews with caregivers and, where included, assent-plus-consent procedures for minors; anonymised data handling throughout.
• Timeline: Year 1 — literature, ethics, quantitative analysis begins. Year 2 — fieldwork, transcription, integrated analysis. Year 3 — writing, policy workshop, thesis submission.
• Contribution: recipient- and migrant-disaggregated estimates for a current policy debate, plus a decision-level explanation that survey coefficients alone cannot give.

A Realistic Three-Year Timeline (Template)

Adapt the phases to your discipline — laboratory sciences front-load procurement and ethics for animal or clinical work; archival humanities front-load archive access and language training. The non-negotiables are an ethics milestone before any human data exists, drafting that starts in year one, and a buffer you do not spend in advance. The University of Warwick’s proposal guidelines make the same underlying point: around 2,000 words plus a timetable and bibliography, written to show the project is planned, not merely imagined.

How the Proposal Connects to Scholarship Routes

The same core proposal serves several competitions with different packaging. For CSC Type B, it sits inside your study plan for one specific Chinese university and should name the laboratory or supervisor group whose work it extends — see our CSC Type B guide. For DAAD routes, it must additionally carry the development rationale the funder scores — our DAAD EPOS scholarship guide details that programme’s expectations, including at least two years of professional experience. For university studentships in the UK, Australia or Canada, the proposal usually travels with a supervisor’s supporting statement, which is why contacting the supervisor early — and sending them a tight two-page version first — is standard practice. And if a scheme asks for a “study plan” instead, our study plan guide shows how the two documents differ and how to convert between them without submitting the wrong genre.

Budget and Resources: The Section Applicants Forget

Not every competition asks for a budget, but every serious project has costs, and naming them accurately is part of feasibility. Work through your design and list what the project consumes beyond your stipend and tuition: fieldwork travel and subsistence; equipment and consumables; dataset licences or archive fees; specialist software; participant incentives where your ethics framework permits them; transcription or translation; and conference travel for dissemination. For each item, note whether the university, the funder or a partner institution normally covers it — many universities hold small research-training grants for exactly these costs, and saying “data collection travel estimated at [range], to be met from the department’s research training support” shows you have investigated rather than guessed. Two warnings matter. First, do not invent precise prices for things you have not checked; ranges with a stated basis (“based on the current return fare and 30 nights of field accommodation”) are more credible than false precision. Second, never build a design whose essential costs nobody will pay: if the core dataset is unaffordable and unobtainable, the honest response is to redesign the project at proposal stage, when redesign is free, rather than in year two, when it is not. Where a funder publishes eligible-cost rules, follow them literally — ineligible items in a budget are read as evidence that you did not read the call.

From First Idea to Submittable Draft: A Working Sequence

Most strong proposals take four to six weeks part-time. Week one: read twenty recent papers in your area and write the one-paragraph gap statement — if you cannot, read ten more. Week two: draft the research questions and interrogate them against reachable data; kill or narrow any question that fails. Week three: write the methodology and timeline first (they discipline everything else), then the literature review as an argument ending at your gap. Week four: draft the introduction, rationale, contribution and abstract, in that order. Then stop for several days. Return with a reader’s eye: send the draft to your prospective supervisor if you are in contact, and to one informed non-specialist; ask each the same three questions — what is this project, why does it matter, and what worries you about whether it can be done? If either reader cannot answer the first two from the abstract and introduction alone, rewrite those before touching anything else. Final pass: check every citation exists and is listed, every promise in the questions is kept in the methods, the word count matches the call, and the title still describes the project you actually designed — it often will have sharpened along the way.

Mistakes That Sink PhD Proposals

• A topic instead of a question. If your proposal cannot state what it will find out, it is a reading list, not a project.
• A literature review that summarises rather than argues. No gap, no project.
• Methods chosen by fashion. Naming techniques (“big data”, “ethnography”) without data, access or skills to operate them.
• No ethics section for research involving people — panels read the omission as inexperience.
• A three-year plan for a five-year project. Over-scope is the most common feasibility failure; narrowing is a strength, not a concession.
• Claiming a supervisor’s support you do not have, or naming a supervisor whose published work your proposal contradicts without noticing.
• Submitting the identical proposal everywhere — fit sections, lengths and funder criteria differ, and panels can tell.

Frequently Asked Questions

How long should my PhD research proposal be?

Follow your target’s instructions first. Published university guidance ranges from 600–1,000 words (Bristol, for an initial application proposal) through 1,000–1,500 words excluding references (Aston) to about 2,000 words plus timetable and references (Warwick). For a formal scholarship application with no stated limit, 1,500–2,000 words plus references is a sound working target.

Do I need a supervisor before I apply for a PhD scholarship?

For most university and Type B-style routes, effectively yes — or at least a supervisor who has read your proposal and agreed in principle to supervise. Some centralised schemes allocate supervision after selection. Either way, evidence of fit with a named research group strengthens the proposal; never claim an agreement that does not exist.

Can my proposal change after I win the scholarship?

Yes — proposals are expected to evolve as you read, pilot and collect data, and supervisors plan for that. What cannot change casually is the funded project’s core: a wholesale switch of topic or field usually needs the funder’s and university’s approval. Write a proposal you genuinely want to spend three years on.

What is the difference between a research proposal and a study plan?

A study plan covers your whole programme — background, courses, timeline, career goals — while a research proposal argues one project in depth: question, literature gap, methodology, ethics and contribution. Scholarship forms sometimes use the labels loosely; doctoral applicants should make sure the research core is proposal-strength whatever the document is called.

Should I include a budget in my proposal?

Include one when the funder or university asks, or when your project has obvious non-standard costs (fieldwork travel, equipment, participant payments, specialist software). Aston University’s guidance lists budget and resources as a standard closing section. Keep it itemised and modest — inflated budgets signal inexperience as clearly as missing ones.

How many references should a proposal have?

Enough to prove you know the current conversation — typically 15–30 for a 1,500–2,000-word proposal, weighted toward recent peer-reviewed work, formatted consistently. Every reference must be cited in the text, and every citation must appear in the list.

I have a strong idea but no research experience. Can I still win a PhD scholarship?

Yes, if the proposal shows preparation honestly: a relevant master’s dissertation, methods training, professional or volunteer research exposure, and a feasible, well-supervised design. Panels fund trajectory as well as track record — but the methodology and timeline must be realistic for a first major project, with training built in.

Sources & Verification

• Aston University — “Writing your PhD research proposal” (official guidance): structure of title, aims, literature review, data and methodology, timeline, budget, references; 1,000–1,500 words excluding references.
• University of Bristol — research proposal guidance (official): rationale, issues and research question, intended methodology, expected outcomes, timetable; 600–1,000 words for an initial application proposal.
• University of Warwick — PhD research proposal guidelines (official): approximately 2,000 words plus timetable and references; proposal as evidence of ability to plan independent research and of fit with a prospective supervisor.
• DAAD EPOS official scholarship database entry (daad.de) for the development-motivation criterion applied to doctoral applicants.
• Both outline examples in this article are original samples written for StudyGraduate and are labelled as such; they describe fictional projects for teaching purposes.

Disclaimer: StudyGraduate is an independent information portal, not an agency or a university, and cannot guarantee admission, supervision or scholarship selection. Proposal requirements differ by funder, university and cycle — always follow the current official instructions for the scheme you are applying to.