Designing a research question


There are a number of steps to take in designing a research question:

  1. Finding a topic
  2. Researching the topic and finding a theme
  3. Narrowing down
  4. Turning your narrow sub-topic into a question
  5. Testing and refining your question

1. Finding a topic

The first step before arriving at a good research question is to think about a general topic area. It should be an area that is interesting to you obviously, and interesting to anyone likely to assess your research proposal.  If they are unlikely to find your topic interesting, they probably won’t respond well to the research question.

To do this, it’s good to think in terms of “hot”, “luke-warm” and “cold” topics (Davies, 2022).

  • Hot topics are at the cutting edge of research. Hundreds, if not thousands of publications are produced daily around the world in these areas—media articles too. Research grants are widely available. Examples include the COVID-19 pandemic and vaccine development, cancer research, the impact of climate change on weather patterns, ‘green’ energy, driverless cars, carbon capture and storage, etc.
  • Luke-warm topics: These are topical but less “hot”. They might be regularly discussed in the scholarly literature, but not on a daily basis. The impact of deforestation on the habitat of fruit bats is an example. It’s important (especially to ecologists) but hardly cutting edge.
  • Cold topics have had their day. They might have been important decades ago, but they hardly rate a mention in scholarship these days, and certainly don’t feature in the media. The Marxist revolution in Cuba or St Anselm’s third version of the ontological argument for God’s existence are examples. There’s nothing wrong with cold topics, but they appeal to a very narrow sectional interest groups.

“Hot topics” are great as they will ensure your research has currency and will potentially lead to spin-off research projects and possible future funding down the track. However, topics don’t remain “hot”—what it “hot” now might be cold or luke-warm by the time you have finished your research project. Another disadvantage is the vast quantity of publications you will have to keep abreast of on a daily basis. Finally, a hot topic also means difficulties in finding a good research question as other researchers may have ‘beaten you to it’, or it might be a question that has been ‘done to death’. As they are “hot” topics—with many scholars around the world working on them—there is also the problem that by the time your have finished your thesis, others may have ‘pipped you at the post’, i.e., made the same original contribution that you intended to make.

“Luke-warm” topics have a reduced range of scholarly papers being published regularly so it is easier to become familiar, and possibly to cultivate expertise, in the area. Something “warm” can later become “hot” too, and if you choose wisely you might be at the vanguard of a promising and developing new area later on. This is obviously a good thing.

“Cold topics” have been debated and discussed for decades (if not centuries). Interest in these area has lapsed for all except for specialists in the field. There is little recent research on these areas, and in pursuing them your work will lack currency. It is not unknown for a “cold” topic to re-emerge as a warm topic however (e.g., the 18th century Gallian idea of faculty psychology re-emerged into modularity of brain functions in late-20th century cognitive science) but you’d need to be pretty confident you were onto something that had the potential to be of contemporary interest to other scholars. 

In general, luke-warm topics are a better bet, but hot topics are good too—with reservations. Find one, then move to step 2.

2. Researching the topic area and finding a theme

Spend several weeks reading in your topic area. Make sure you read periodical articles as well as scholarly books, and book chapters.  You will start to notice regular themes emerging. These are sub-topics within the main topic area. You may like to do a mind map of these themes. Here is one on intelligence in the discipline of psychology:

A mind map on the area of intelligence (Cougle, 2012)

Now try to choose a sub-topic or theme within your chosen research area. It’s a good idea to let your emotional reaction to these themes guide you here. Which arouses your intellectual passions? Which theme has been a topic of discussion with other colleagues? Which one do you read most about—because it sparks something in you?

Alternatively, your choice might be guided by a problem area. This is an area of scholarship where the literature exposes a persistent research problem. An example might be the increasing rates of autism in children and the causes of this. Statistics show that this condition is on the rise, and there is no satisfactory explanation for it.

Whether you rely on your emotional response to a topic, or are guided by a research problem, remember that, if you are doing a PhD you will be working on the the topic for many years. It has to be something that fascinates you. You might decide, for example that ‘savant syndrome’ is your narrow field of interest. From this, you move to the third stage.

3. Narrowing down

Now that you have found a sub-area or theme within a topic of interest you need to narrow down further. It’s insufficient to construct a research question such as What is savant syndrome? That’s far too broad. To develop a more refined question, you need to read more about the specific area and narrow down the sub-topic ever further. It has to be as narrow and specialised as possible without losing your interest. For example:

Narrowing down your sub-topic down

This step is essential for developing a research question, which is the next step.

4. Turning your narrow sub-topic into a question

The task now is to create a question that will simultaneously be broad enough to allow for sustained scholarship in the area, and yet not so wide as to lead to spill-over into adjacent related areas. If there is spill-over it might make the focus of your question unclear. You don’t want to start researching other types of savant syndrome, or other brain disfunctions. This would result in a loss of focus. Avoiding this problem means refining and testing your question.

Open, not closed: The question should also be “open”, not “closed”, i.e., not lead to a superficial ‘yes/no’ response, but rather open-up opportunities for debate and reflection. For example:

  • Is there left hemisphere disfunction in middle-aged autistic females?

is a closed question. The more detailed question:

  • Is there statistically significant rates of left hemisphere disfunction in middle-aged autistic females and does this correlate with evidence of savant syndrome?

while a better question is still a closed question. The answers they foster are ‘yes/no’ answers. But things might not be so unsubtle. It might be a matter of balance: In some cases there might be a statistically significant rate, but in other circumstances there might not be. By asking a closed question you are effectively closing down the debate.

Getting the right question starter: Asking questions in such a way already limits the range of the study that you undertake and closes down opportunities for analysis. Instead, use an “open” question that uses question-starters like:

  • HowWhy … Is… and To what extent….

For example:

  • To what extent do statistically significant rates of left hemisphere disfunction occur in middle-aged autistic females and does this correlate with evidence of savant syndrome? 

This opens the issue up to permit discussion and debate. How it is answered becomes a matter of degree. To take another example:

  • Should the constitution be changed so that the President of the US can serve more than two terms?  

That’s a closed question. A subtler question might be:

  • To what extent is there merit in amending the US constitution to admit of changes to presidential term limits?   

Which question is likely to lead to fruitful discussion and debate? Clearly, it is the “open” question.

Here are some alternative question types and research question formulations you might like to consider.

Research aimsResearch question formulations
Describing and exploringWhat are the characteristics of X? How has X changed over time? What are the main factors in X? How does X experience Y? How has X dealt with Y?
Explaining and testingWhat is the relationship between X and Y? What is the role of X in Y? What is the impact of X on Y? How does X influence Y? What are the causes of X?
Evaluating and actingWhat are the advantages and disadvantages of X? How effective is X? How can X be achieved? What are the most effective strategies to improve X? How can X be used in Y?
Alternative question forms (McCombes, 2021)

5. Testing and refining your question

Testing and refining: You should not rest with your first iteration of a question however. It needs to be assessed by scholars in your field of study who know the area.

This is where your colleagues come in. Shop your question around with your colleagues in departmental seminars and get their opinion.

  • Is the question clear?
  • Is it interesting?
  • Is it sufficiently focussed?
  • Is it complex and subtle enough?
  • Is it researchable? (Can research actually be done on the topic?)
  • Is there likely to be sufficient, current literature in the area? (insufficient literature might mean a “cold” topic; too much literature might mean the field is overdone).

You are now in the position to begin your deep dive into the literature and to start on your Literature review.

References and further reading

  • Cougle, B. (2012). Psychology, Chapter 11. Intelligence. https://www.mindmeister.com/80682022/psychology-ch-11-intelligence
  • Davies, M. (2022). Study skills for international postgraduates. Bloomsbury.
  • McCombes, S. (2021). Developing strong research questions. Criteria and examples. https://www.scribbr.com/research-process/research-questions/

For a downloadable helpsheet, see Designing and Research Question. See also: