How to plan a research project: from topic to finished plan

A good research project is one where the question, the method, and the evidence line up, and where you can finish it in the time you have. The size of the topic has little to do with it. Students who choose a large, interesting topic and a loose plan are the ones whose supervisors send the proposal back.

Planning is the part of the project that is cheapest to fix. A weak plan costs you an afternoon to rewrite. A weak plan discovered after data collection costs you months.

1. Choose a topic you can finish

Interest matters, but access matters more. Before you commit, answer four practical questions.

• Can you reach the data? If you plan to interview hospital managers, someone has to agree to be interviewed. If you plan to use records, you need permission to see them.
• Do you have the time? Count the weeks between now and your submission date and work backward.
• Does your supervisor know the area, or know someone who does?
• Does your university require ethics approval for this kind of data collection, and how long does its committee take?

A topic that fails any of these is a risk, and it is better to find that out in week one.

2. Read until you can name what is already known

Read widely enough to answer two things: what researchers have already established about your topic, and what remains unanswered. The unanswered part is your opening. If you cannot name it yet, keep reading. Our guide to organizing sources in a literature review shows how to keep that reading from turning into a pile of summaries.

3. Turn the topic into one question

A topic is an area. A question can be answered. "Student mental health" is a topic. "How do first-year nursing students at Kenyatta University describe the stress of their first clinical placement?" is a question.

Write the question as a single sentence, then test it. Could one person with your resources answer it within your deadline? If the answer is no, the question is too broad. We explain what that means, and how to fix it, in what to do when your professor says your research question is too broad.

4. Choose a method that answers that question

The question decides the method. The method does not decide the question.

• A question about whether two things are related, or whether groups differ, calls for a quantitative design, such as a survey analyzed with correlation or a t-test.
• A question about how people experience or describe something calls for a qualitative design, such as interviews analyzed for themes.
• A question that needs both numbers and explanation can call for a mixed methods design, but mixed methods doubles the work and your plan must show why you need both.

John Creswell and J. David Creswell's Research Design: Qualitative, Quantitative, and Mixed Methods Approaches sets out these choices clearly, and your methods course probably lists it.

5. Check feasibility against the calendar

Write each task and give it a date: ethics application, recruitment, data collection, analysis, each chapter draft, supervisor review, and final formatting. Leave time for the steps that depend on other people. Ethics committees, participants, and supervisors all take longer than you expect, and they do not work to your schedule. Our guide to IRB approval, step by step lists what that application usually requires.

6. Put the plan on one page

Before you write anything longer, write one page that holds the following:

1. The research question.
2. The purpose of the study, in one sentence.
3. The method and the reason it fits the question.
4. Who or what you will study, and how many.
5. How you will analyze the data.
6. The timeline.

Show it to your supervisor. A one-page plan takes ten minutes to read, so the feedback comes back fast and the objections come out early.

Why plans get sent back

Four problems account for most of it. The question is too broad. The method cannot answer the question. The sample is too small or cannot be reached. The ethics approval was never planned for. All four show up on a one-page plan if you look for them.

Want a second reader before you submit your plan?

The Two Dissertation Moms offers a Research Design and Methodology Review that checks whether your question, design, sample, and analysis fit together before your supervisor or committee sees them.