mappy

Research question mind map: from topic to proposal

The short answer

How do I use a mind map to turn a broad topic into a research question and a proposal outline?

Put the broad topic in the center and branch out what is known, what is disputed, and the gap stated in one sentence. List candidate questions, test each against a checklist such as FINER, keep one, and build the proposal outline from it. Write the reasoning and exclusions in each node's note.

Researched and drafted by an AI agent that works for Mappy. The method draws on standard research-methods teaching and the worked example is a fictional study scope with no findings. Mappy does not suggest topics or check your design.

Why a topic is not yet a question

Search for a research proposal mind map, a research question mind map or a research topic mind map and you will mostly find diagrams that start from a topic and fan out into subtopics. That helps you see the territory. It does not tell you what to study, because a topic such as "remote work and health" can be explored for years without ever being answered.

A researchable question is narrower. It names what you will look at, in what setting, and how an answer could be recognized. This guide covers the step before a literature review: topic, then problem, then question, then proposal outline. Once you have a question, the literature review guide takes it from there, and this guide does not repeat that work.

The skeleton: one topic, four branches

Put your broad topic in the center. In Mappy, Tab adds a child node and Enter adds a sibling. Make four main branches:

  • What is known. Two or three starting sources, such as a recent review article, a textbook chapter or a widely cited study. This is a starting point, not a full review.
  • What is disputed or missing. Where sources disagree, where a study covers only one setting, or where a question is simply untested.
  • The problem or gap. One sentence.
  • Candidate questions. Three or more, each a child node you can test and compare.

Add a fifth branch, Proposal outline, once you have chosen a question. Each node opens into a full note, so put the reasoning there: why this node exists, what it leaves out, and which assumption it rests on. Connections in Mappy are unlabeled, so write why two nodes are linked in the note. Press F1 to see the shortcuts, and Cmd+Shift+F focuses one branch when you want to work on only the candidate questions.

Known, disputed and the gap

What is known. For each starting source, make a child node with the citation and one line on what it supports. In the note, record the setting and method, since a claim may hold only for the group studied. Stop at a few sources. Reading widely comes later, in the literature review.

What is disputed. Write each disagreement as a pair of claims and, in the note, one possible reason for the difference, such as different populations, measures or time periods. Possible reasons are guesses to check, not conclusions.

The problem or gap. Write it as one sentence in this shape: "We know X in setting A, but not Y in setting B, which matters because Z." If you cannot finish the sentence, the topic is not narrowed enough, or the reading is not yet sufficient. Record in the note what you assumed about the existing research, for example "I assume no study has looked at this in my region", and check that assumption with a search before you commit.

Candidate questions and how to test them

Write three or more candidate questions as sibling nodes, so you are choosing rather than settling on the first idea. Then test each one. A well-known set of criteria is FINER, from Hulley and colleagues in Designing Clinical Research: a good question is Feasible, Interesting, Novel, Ethical and Relevant. It was written for clinical research, but the five checks are useful in other fields. If your field gives other criteria, use those.

A plain checklist works alongside it:

  • Answerable. Can it be answered with the data, time and skills you have?
  • Specific scope. Are the population, place and period named?
  • Not trivial. Is the answer more than a yes or no that the textbook already gives?
  • Definable terms. Can you say how each key term will be measured or identified?

Make the checks child nodes of each candidate, or a checklist in its note. Write the reason for each answer, not only a tick, because the reasons are what you will discuss with your supervisor. A candidate that fails one check may need a narrower scope rather than rejection. Keep rejected candidates in the map, collapsed, with a note on why, so you do not rediscover them.

Worked example: narrowing urban trees and summer heat

This is a fictional study scope to show the steps. It reports no findings.

  • Broad topic. "Urban trees and summer temperatures." Too wide: it covers every city, every kind of tree and several meanings of temperature.
  • Known and disputed (starting points). Suppose a review article and two city case studies suggest that tree cover is linked with cooler surfaces, while the size of the effect may depend on city and season. Assume this has been checked against the actual sources.
  • The gap, in one sentence. "Tree cover has been studied mostly in large cities, but we know less about mid-sized cities such as the one in this study, which matters because their planners lack local evidence for where to plant."
  • Candidate A. "Do trees reduce heat?" Fails the scope and definable-terms checks: which trees, which heat, where?
  • Candidate B. "Do street trees cool air temperature in neighborhood X?" Fails feasibility if no air temperature sensors exist and there is no budget for them.
  • Candidate C. "In residential neighborhoods of one mid-sized city, is higher tree canopy cover associated with lower afternoon land surface temperature on summer days, after accounting for building density?" The scope is named, the terms are measurable, and the answer is not a yes or no, because it asks about direction and size.
  • Choose C and write its note. Why: it uses data that the study could plausibly obtain, such as public satellite imagery and a municipal canopy map, assuming they are available (check this first). Excludes: air temperature, health effects, causes, and other cities. Rests on: the assumption that land surface temperature is a reasonable stand-in for what residents experience. It is not the same as air temperature, so state this as a limit.

Notice the wording: "associated with" matches what an observational comparison can support. Claiming that trees "cause" cooling would need a different design. Putting such a distinction in the note keeps the question honest.

From question to proposal outline

Proposal sections vary by program and by field, so follow your department's template and use this branch as a drafting scaffold. A common set of children under Proposal outline is:

  • Working title. Provisional. Note the question it answers.
  • Background. From the known and disputed branches. Keep the citations in the notes.
  • Problem. The one-sentence gap, plus why it matters.
  • Aims and question(s). The chosen question, and any sub-questions that break it into parts.
  • Methods. Design, data, sample and analysis as four children. For each, note what it would let you say and what it would not.
  • Feasibility and timeline. What you need (data, access, software, time), and a rough order of work. Check against the feasibility test above.
  • Ethics. Whether people, personal data or sensitive places are involved, and what review your institution requires. The example above uses no human participants, but a study of residents' experiences would need a different answer.
  • Expected contribution. What a reader could do or know afterward, stated without promising results.
  • References. Link to your source notes, for instance those from the research paper notes template.

For each child, the note should answer: why is this here, what does it leave out, and what does it assume? Those are the places a reviewer is most likely to push.

Get feedback and then read widely

Bring the map, not only the draft, to your supervisor. Use Cmd+Shift+F to focus on the candidate questions and the chosen one, and export the map as a PDF to share; Mappy exports maps as PDF. Ask specific things: Is the gap real? Is the scope realistic for the time available? Is the key assumption acceptable? The thesis meeting notes guide shows how to record the answers so they land in the right node.

When the question is settled, move to the literature review mind map, where themes become branches under your question. Expect the question to change as you read. Revise the note with the reason for each change, so the map records how your thinking developed. Mappy does not suggest topics, write proposals or check a design, so those judgments stay with you and your supervisor.

Building and saving your own maps needs Mappy's one-time Full Unlock. The free download shows a guided tour and an editor preview, but it does not create or save maps, so use the template below on paper or in any notes app if you are only trying the method.

Research question worksheet

Use this outline in any notes editor, or recreate its branches in Mappy. It is a plain-text worksheet, not an automatically importable .mindmap file. No signup is needed for the template.

Download the text template (.txt)

RESEARCH QUESTION WORKSHEET
Broad topic:

What is known (2 or 3 starting sources, one line each):
What is disputed or unclear:

The problem or gap, in one sentence:
  Why it matters, and to whom:

Candidate questions (write at least three):
  1.
  2.
  3.

Test each candidate (yes / no / unsure, with a reason):
  Feasible: can I answer it with the data, time and skills I have?
  Interesting: will it hold my interest, and my field's?
  Novel: does it confirm, refute or extend what is known?
  Ethical: can it be done without undue risk or burden?
  Relevant: does it matter for knowledge, practice or policy?
  Specific scope: population, place, period, exclusions named?
  Not trivial: is the answer more than yes or no?
  Key terms definable: can I say how each will be measured or identified?

Chosen question:
What it excludes on purpose:
Assumption it rests on:
What would make me revise it:

Proposal outline (sections vary by program; follow your template):
  Working title:
  Background:
  Problem:
  Aims and question(s):
  Methods (design, data, sample, analysis):
  Feasibility and timeline:
  Ethics:
  Expected contribution:
  References:

Questions for my supervisor:

Researched and drafted by an AI agent that works for Mappy. The FINER criteria are attributed to Hulley and colleagues, the example reports no results, and proposal sections are described as varying by program. How these guides are written.