How to Write a Research Proposal for a PhD or Research Fellowship

What a research proposal is for

A proposal has three jobs. First, it shows that you know your field well enough to see a gap, a problem or an open question. Second, it shows that you can plan a realistic method. Third, it helps the department decide whether anyone there can supervise the work.

Both the topic and the plan are expected to change after you begin. Supervisors know this. They are looking at your reasoning, your ability to write clearly and your awareness of what is possible in the time and with the resources available.

Requirements differ widely. Some programs want a short outline of one or two pages, others a detailed document of several thousand words, and some ask for a proposal only after admission. Always follow the instructions in the call, including word limits, headings and file formats.

Typical sections

Programs use different headings, but most proposals contain the following parts. If the call gives its own structure, use that instead.

• Title: specific and informative. A reader should understand the topic and the approach from the title alone. Avoid titles that are only a broad field name.
• Background and context: a short review of what is known, the key debates and the gap your project addresses. Cite the most relevant and recent work, not everything you have read.
• Research question or questions: one main question, perhaps with two or three sub-questions. Write them as clear questions or hypotheses that could be answered.
• Aims and objectives: what the project will achieve. Aims are the general purpose, objectives are the concrete steps.
• Method: how you will collect and analyse evidence, such as experiments, models, fieldwork, interviews, archives, surveys or text analysis. Explain why this method suits your question and mention the data, sources or equipment you will need.
• Timeline: a realistic plan by stage or by year, including reading, data collection, analysis, writing and time for revision.
• Expected outcomes and contribution: what the project could add to knowledge or practice, and what you might produce, such as articles, a dataset or a thesis chapter. Be honest about uncertainty.
• Ethics, risks and resources: if your work involves people, personal data or sensitive material, say how you will handle approval and consent. Mention main risks and a fallback plan.
• References: a short list of the main sources, in the citation style the program requests, or a consistent style if none is given.

How long it should be

Use the length stated in the call. When no length is given, a common range is about 1,000 to 3,000 words, depending on the field and the level, but this is only a general pattern and not a rule. Humanities and social science proposals are often longer than those for laboratory-based projects. A longer proposal is not a better one. If a call asks for 1,500 words, a text of 1,500 words that is clear and focused is stronger than one that goes over the limit.

As a rough division, the background and the question together might use a third of the text, the method a third, and the remainder can be shared between the timeline, outcomes and references. The method section is where many applicants are thinnest, so give it enough room.

Choosing a topic and finding a supervisor fit

A strong proposal starts from something you already know. Your thesis, a course project, a work problem or a question that came up in your reading can all be a starting point. Then narrow it. A topic such as climate change and agriculture is a field. A question about how changes in rainfall timing affect the planting decisions of small farmers in one region is a project.

The topic also has to fit the department. A good proposal for the wrong place will not succeed. To check fit:

• Read the profiles and recent publications of the academics in the department. Look for two or three whose work is close to your question.
• Check whether they are currently taking PhD students, whether the department advertises funded projects, and whether the call asks you to apply to a specific project or to propose your own.
• Write to a potential supervisor before applying, if the program allows or encourages it. Keep the message short: who you are, your question in two or three sentences, why you chose this person, and a request for a brief reply or a short call. Attach your CV and, if you have one, a one-page summary. Do not send a long proposal without being asked.
• Do not send identical emails to many professors. Each message should show that you read their work.
• In your proposal, mention the people or groups whose expertise matches your project, if the program expects this, and explain why in a sentence or two.
• Be ready for a lack of reply. Academics are busy, and silence is common. A polite follow-up after one to two weeks is acceptable. After that, move on to other options.

A worked mini-example outline

The following outline is for an imaginary project. It is not a real study and is shortened to show the logic. Your own proposal will need more detail and real references.

Title: Public library use among young adults in mid-sized towns after the introduction of digital lending services.

Background: Libraries have added e-books and online services, but it is unclear whether these services reach young adults who did not use the physical library. Existing studies mostly focus on large cities or on older readers. This project looks at mid-sized towns.

Research question: How does the introduction of digital lending change the ways in which young adults aged 18 to 25 use public libraries in mid-sized towns? Sub-questions: Who starts using the library because of digital services? Which users stop visiting in person? What do users say about their reasons?

Aims and objectives: To describe changes in use patterns (objective 1); to identify groups who are reached or missed (objective 2); to suggest practical changes to service design (objective 3).

Method: A mixed approach. Analysis of anonymised lending and membership data from three libraries, provided under an agreement. A short survey of young adult members and non-members. About twenty interviews to explain the patterns. The data analysis would compare periods before and after the service started. Reason for this choice: the numbers show what changed, and the interviews help to explain why.

Timeline: Year 1, literature review, ethics approval, access agreements, survey design. Year 2, data collection and first analysis. Year 3, interviews, second analysis and first articles. Year 4, thesis writing and revision.

Expected outcomes: A thesis, two journal articles, an anonymised dataset if permitted, and a short practical report for library managers.

Risks: The libraries may not provide data, so the survey and interviews could be extended, and a fourth library may be approached. Low survey response would be addressed by offering more than one way to answer.

Notice how the question is narrow, the method matches it, each stage has a time slot, and the risk section shows that the applicant has thought about problems.

Common mistakes

• A topic that is too broad to finish in the time of the degree. If you cannot describe it in one or two sentences, narrow it.
• No clear question. A list of themes or a description of a field is not a research question.
• A method that does not match the question, or no method at all. Saying that you will analyse data does not explain how.
• A literature review that lists authors one after another without showing the gap. The background should lead directly to your question.
• An unrealistic timeline, for example fieldwork in several countries in six months.
• Ignoring the department. If no one there works on the subject, the proposal is hard to accept, however good it is.
• Overclaiming, such as promising to prove something or to solve a major problem. Modest and precise claims are more credible.
• Weak writing: undefined terms, long sentences and missing structure. Clear writing is part of the evidence that you can write a thesis.
• Missing or inconsistent references, or citing only old sources when the field has recent work.

Getting useful feedback

• Give a draft to someone in your field, such as a former thesis supervisor or a lecturer, and ask specific questions: Is the question clear? Is the method realistic? What is missing?
• Ask someone outside your field to read it too. If they can follow the aim and the logic, a committee member from a neighbouring area probably can.
• Ask readers to summarise your project in two sentences. If their summary differs from your intention, the text needs work.
• Allow time between drafts. A text you read after a few days is easier to improve.
• Keep a version of the proposal that is easy to shorten, with the key points first, so that you can adapt it to different word limits.
• If you receive criticism, look for patterns. When two readers mention the same problem, it needs to be fixed.