Finding ideas for your thesis


Finding ideas for your thesis

Initial exploration

Do a brief literature review 

Explore relevant research agendas, meta-reviews, introductions to the current research landscape and recent publications of impact oriented researchers and institutes.

Then go deeper into individual studies and explore a few of these (you could use this tool to find relevant papers). Keep a log of interesting papers and ideas as you go.

Aim to identify what is currently unknown and where there are any ‘gaps’ in existing knowledge. This will enable you to determine how your own research will contribute to and build on what is already known.

Talk to other researchers

We think most people tend to be too hesitant to reach out to others for advice. We recommend connecting with researchers working on your research direction to ask what they think some of the most important questions are.

You could talk to:

  • researchers working on the research direction you’re interested in, for example because you’ve read papers they’ve written.
  • your supervisor or potential supervisors.
  • researchers in your research group.

Consider connecting with organisations

You could also explore writing a thesis tailored to the needs of specific stakeholders as a way of having a direct impact with your research.

You could contact some of the organisations listed on a research direction profile you’re interested in and ask if there are research projects you could undertake that would be decision-relevant for them.

Read our advice on sharing your research beyond academia to learn more about this possibility.


While you’re gaining an overview of the field, you could start a brainstorming document to actually put your ideas down on paper (both to help you keep track of and develop your ideas and to make it easier for others to give feedback later). Try writing down the general research questions you’re interested in first, then splitting the original question into more concrete sub questions.

Rank your questions and start building theories of change

While writing your brainstorming document, consider what the theory of change would be for the questions you’re interested in pursuing, and whether some of your questions seem more promising than others.

Here’s a possible framework for ranking research questions and beginning to sketch out the steps by which answered them might have a positive impact, based on Jaime Sevilla’s post on “How to generate research proposals” and a methodology used by some FHI Summer Research Fellows:

  • Importance – How large in scale and/or severity is the problem your question would address? 
  • Tractability – how realistic is it that you would make progress? Is your research question concrete and manageable, and do you have a clear strategy to tackle it?
  • Neglectedness – How likely does it seem that others won’t work independently on this question if you don’t? 
  • Learning value – Will you learn useful things from working on it?
    • Will this project help you build useful research skills?
    • Will this project help you decide what to do next?
    • Will this project help with building your model of how something important works (as opposed to relying on the views of others)?
    • Will this project help you refine a nebulous sense of confusion/uncertainty/curiosity into a crisp, important question?
  • Actionability – The research has a clear audience and can inform action 
    • Will this project generate genuinely new findings/data that are useful from some relevant point of view?
    • Will this project help to translate/communicate important ideas that need a signal boost?
  • Respectability and credentials – Will the output signal your research competence?
    • Will the project reflect well on you, and is it shareable with others (or could it be developed into something shareable)?
    • Will the project allow you to interface/build relationships with people whom it will be helpful to know going forward? 
  • Personal fit & situational fit  – Will you provide a differential value for the question, am are you in or can you find a good environment to work on this question; for example, can you find a good supervisor?

What weight these criteria carry in your decision depends on their importance to you in this specific research project. For an undergraduate thesis, learning value might be of greater importance than later in your career, while signalling competence becomes relevant when you want to be hired as a researcher – read more about our ideas on how to find a research question that fits your degree level here.

Reach out to people for feedback

At this point, identify some experts who might be interested in talking about these questions, and then reach out for feedback. Getting feedback might allow you to prioritise between the questions, develop your methodology further or discard projects before investing too much effort in them. 

You could seek feedback via two strategies: sending the document to people asking for general comments and seeking out people you know work on some particular ideas related to yours and questioning them / asking for their feedback on specific ideas.

Finalise your research question and develop a proposal and methodology

To develop a really good methodology:

  • Check what are standard methodologies in your field. 
  • Identify how research on the topic has previously been conducted in terms of, for example: approach, methods, analysis of data. 
  • Ask your supervisor whether there are relevant differences in different methodologies you could use.

What you might want to consider when choosing a methodology:

  • If you want to learn a specific skill, e.g. data analysis or conducting interviews, it can make sense to pre-commit to using a specific methodology and then develop your questions with this in mind.
  • Usually however, you would first develop the question and the appropriate methodology to answer that question depends on the conclusions you want to draw.
We use cookies to enhance your experience. By continuing to visit this site you agree to our use of cookies. More info