A very common issue at the beginning of research projects among many students, PhD candidates, Early Career Researchers (ECRs) and even among senior academics is uncertainty about the question that the project should answer.

A good research question addresses an important problem, the answer to which contributes to both societal needs and to science. Your contribution to science and society should answer a question that is both unresolved and important. It should be a question that we do not yet have the answer to or an incorrect answer, and it should be a question that matters.

Your research question should be an interesting question. That is: the answer should be relevant for yourself, for society as well as for science.

It is important that you are personally intrigued by the question you are asking, because you will devote a substantial amount of time to your research. You better work on something you find interesting. One way to avoid getting bored is by asking a question that is relevant to you personally. This will keep you motivated when you get a lot of feedback, when you are waiting for your data to come in, or when have to engage in repetitious work to prepare the analyses.

How then do you find your question? Start by reading literature recommended to you by your supervisor. As you are reading these materials, see which of these spark your interest. If your supervisor has not yet provided a review article that presents the state of the art on the topic, try to find it. Section 3.4.2 provides suggestions that help you find such an article. Research reports and review articles often end with suggestions for future research. Which of these questions are most appealing to you? Which questions do you think are most relevant for science and society?

Next, draft your research question using the suggestions below. Write down the reasons why you find the question important for science and society. Don’t worry if you find it difficult to do this. Talk about your motivation with friends, not only to see whether they find the question important, but also to see whether you are able to make the case. If you can’t explain why it is an important question, it may not be an important question, or you have not mastered it yet. Through further discussions you discover which arguments are convincing. Revise the question using the feedback you get.

Finally, send your revised research question to your supervisor. Be prepared to revise the text. Don’t be disappointed by the comments and suggestions that you will undoubtedly get – they do not mean that you have not thought it through or that you are on the wrong track. It is simply your supervisor’s role to improve your research ideas.

How should I construct my research questions?

In addition to the substance of your research questions, the way you formulate them is an important feature that determines the success of your research. Broadly speaking, there are three types of research questions (Ultee, Arts & Flap, 2009):

  1. Descriptive questions;
  2. Explanatory questions;
  3. Policy questions.

A descriptive question asks for a description of a phenomenon, about its development over time, or how it varies between persons, contexts. Descriptive questions invite readers to go on a discovery mission: “let’s see what the world looks like!”

An explanatory question asks about causes of the phenomenon you are interested in. Explanatory questions invite readers to delve deeper into things they already know: “let’s find out why the world looks the way it does!”

Finally, a policy question asks about ways to change the world in a desired direction. Policy questions invite readers to solve problems or avoid them: “let’s fix this!”

Work in the right order

Before you pose a research question, make sure that it is a question to which we do not yet know the answer. You do not want to present a research question without having read previous research. In all likelihood, the question has been asked and answered by others before you. Make sure you have read previous research on your research question. See section 3.4 below for suggestions on how to select relevant research.

Again, also in the formulation of research questions keep in mind to work in the right order (Ultee, Arts & Flap, 2009). The three types of questions can be ordered in the form of a pyramid. Explanatory questions build on descriptive questions; policy questions build on explanatory questions. Descriptive questions are the most basic type of questions, that have to be answered before explanatory questions can be answered. Policy questions, in turn, presuppose that the answer to a set of explanatory questions is known. You can see that policy questions are the most difficult to answer because they require both valid answers to descriptive as well as explanatory questions.

An example serves to illustrate this cascade of questions. Suppose your research question is: “To what extent can differences in volunteering between religious groups be explained by differences in altruistic values and being asked?” (Bekkers & Schuyt, 2008).

This research question presupposes that you know the answer to the following descriptive research questions:

  1. “What is the proportion of the Dutch population engaged in volunteering in 2018?”
  2. “How does the proportion of volunteers differ between religious groups in the Netherlands in 2018?”

The explanatory research question in this example is: “How can differences between religious groups in the proportion of volunteers in the Netherlands in 2018 be explained?”

When you ask an explanatory research question, make sure that the assumptions are correct. The explanatory research question in the example assumes that there are differences between religious groups in the proportion of volunteers in the Netherlands in 2018. Check this. If there are no such differences, and all groups have about equal percentages of volunteers, your explanatory question is misguided. You cannot answer it in a meaningful way.

Reading previous research, you may get the impression that the relations that you are interested in have changed over time. In this case, it may be interesting to ask the following set of questions:

  1.  “How has the proportion of volunteers in religious groups changed since 1998?”
  2.  “How can changes in volunteering by religious groups be explained?”

This second set of questions follows the first set of questions. For a thesis, I would not recommend you to try to answer these four questions, but only two. I have numbered them consecutively, but you could answer questions 3 and 4 in a thesis without answering questions 1 and 2. In a dissertation, questions 1 and 2 would be good for one chapter, and 3 and 4 for a second chapter.

The above examples follow two rules for research questions:

  1. Work in the right order: first ask descriptive questions (which may concern a certain point in time or a change over time), then explanatory questions, and finally policy questions;
  2. Ask a question about mediation: ‘how can the influence of X on Y be explained?’ and/or moderation: ‘under which conditions does X influence Y?’.

Interested to learn how you can craft a good research question? Read more here.