3. Research question

One of the most complex parts of scientific writing is coming up with a topic. More specifically, this concerns a research question or questions you want to study. A question might come from a practical problem, or even better, you might be able to find an open problem in related literature. In either case, it is essential to document the motivation behind your work, as this will be needed as a justification for your paper or thesis.

3.1 Research question scopes your research#

Choosing your research question well is essential because it scopes your research. Too broad or vague research questions might not yield helpful enough answers. To give you a concrete example of a research question from one my papers (Vepsäläinen et al., 2023), consider the following: What are the technical opportunities and challenges of edge computing for static website hosting? To dissect the research question of my paper further, I have split it up below into separate components to consider:

• What are: - The question defines the type of the study. For example, here we are looking for a group of things.
• the technical opportunities and challenges of: - More specifically we are looking for technical opportunities and challenges of something.
• edge computing for static website hosting: - The final part of the question scopes the study to edge computing for static website hosting.

Although you might not be familiar with some of the technical wording within my research question, you can see how I composed the question for my paper. More importantly the question covers the keywords of my papers and also gives some idea of what should be covered in the paper. Especially those technical terms I had in my question should be visible in the background sections of my paper and likely they will be repeated across the paper given their importance. It is also likely some of the keywords may make it to the title of the paper itself.

3.2 Fix your research question early#

As you can see based on the brief example above, there are clear benefits to thinking about your research question as it can help to position and scope your work. When you are researching a topic you do not understand well yet, it is normal to start with a slightly vague question and then scope and adjust it over time. Typically you constraint your topic somehow by adding more specifiers to your questions although occasionally you might even go the other way and make your questions broader to match your research. It is more likely you will reduce the scope of your topic instead of expanding, though, especially if you are working under tight page constraints. It is also valuable to consider how many research questions to cover and what kind of angles each of them covers.

How many research questions to choose for a paper

It is possible to develop a paper around a single research question and this is what I recommend especially for Bachelor level work early on. For larger works, one option is to have a central question and several smaller ones that support responding to the central one as these questions may help you to craft your research agenda and choose the right research methods for your inquiry.

If you have to work against a tight page constraint, it is entirely possible you may have to leave some questions out of your paper and leave them for later. Even then, it is a good idea to document open problems at the end of the Conclusion section of the work so it is clear to the reader that you have considered the broader context beyond your paper.

How to construct a good research question

I tend to avoid questions with boolean answers as in "Is vantablack the darkest black color available?" as they tend to polarize research while often being trivial. Instead, I would go for questions that can produce variety of answers that may even depend on the method and researcher. Ideally someone else should be able to repeat your research and validate your findings although finding negative results, or refutations, is possible too.

I think the key to discovering good research questions is to do your own research well and map what others have done before you as this gives a good idea of what kind of questions people in your field or niche tend to use in their research. Likely you will notice some patterns and the same may go for the type of research that is done against the questions. Sometimes taking the same question but changing the research method somehow may be an option. It is typical that at Bachelor level students focus on surveys and literature reviews while more involved empirical work is performed for their Masters. A good researcher is able to use multiple methods and even combine them within the same study to gain unique insights. Further, good researchers know niches beyond their own and are able to pull ideas from other branches of science to improve the quality of their work.

You know the least about the topic when you should know the most

The fundamental problem related to choosing your research question is that you know the least at the beginning of a research process and the most at the end. For this reason, it is essential to spend time with literature related to your research domain and take notes. The fact that you know the least in the beginning has impact on which order you should write and we cover this problem in the Chapter 4. Elements of a paper.

As you gain experience of your niche, the nature of this problem will change as you have a better awareness of what has been done before and what is to be done. That said, inexperienced researchers may approach a topic from an unexpected angle or consider a problem that experiences scientists might overlook. Some niches are simply more interesting than others and it may even be a good strategy to study a topic that is not so crowded although that has benefits of its own at least in terms of availability of source material.

Time spent understanding the topic saves time in the long run

Spending time with the topic is valuable as it avoids you to arrive on a relevant research question that has not been explored in detail already. The worst thing that could happen is that after performing your research you find out that someone else has published the same, or similar, results already. You can largely avoid this risk by doing a diligent literature review before fixing on your research question. For this reason, I recommend writing a personal literature review early on especially when you are new to your topic. You should write your review in a scientific style with accurate references as this goes for training especially if you have not done a lot of scientific writing before. It can take a while to get used to scientific style and the more you work in the style, the easier and the more natural it gets. If you work towards a doctorate, you should consider publishing this early review in a form that you can use as a part of your thesis in one way or another to ground your research as the work is contributing towards your degree in a strong manner if you do this.

3.3 Deciding how to address the research question#

Your choice of a research question impacts the ways you can study the problem. In my example above, I was looking for specific qualities to characterize technical opportunities and challenges and focused specifically on performance where I did a small quantitative study to understand a specific factor in detail. Qualitative and quantitative research form the basic approaches and it is possible to combine them in a way like this.

Literature reviews and surveys are the easiest starting point

The most straightforward type of study is a literature review or a survey where you address your research question based on existing material and then try to synthesize adequate answers. Surveys are valuable because they can point out new, more specific research questions that can be addressed in future research efforts. Surveying also has the benefit that it strengthens your understanding of what has been done already and helps you connect the dots between different papers as you see how they address your research question. In short, surveys provide a more clear view of a topic while not necessarily producing a significant amount of new scientific information. Due to their nature, surveys are a natural fit for Bachelor level as they allow you to learn much about scientific writing, especially the first time.

Empirical studies are more complex by nature

Surveys can provide a good stepping stone towards empirical studies, where you might look into specific questions concretely, for example, by capturing user sentiments or benchmarking a particular property. These studies are expected at Masters level and industry as they provide specific insights into otherwise difficult-to-understand topics.

Preferred research methods depend on your field

It is difficult to give specific advice on researching your question as the exact methods and approaches tend to differ based on the field. By getting familiar with papers within your field, you can quickly see common types of documents and research approaches and better understand what approach you can apply in your paper.

3.4 Find good collaborators early on#

When you work in an academic context, collaboration is expected at some level although you might be doing most of the work alone. For thesis work, you most likely work with a supervisor and an advisor, or multiple advisors, that communicate the expectations from the academic side and help you iterate on the content. Given the impact of your collaborators, it is worth it to invest time into finding people that make sense given the scope and context of your work. A good supervisor is able to help you on a high level of the process while good advisors can give concrete ideas on your topic and perhaps how to present it.

Although you could come up with your own research question, often researchers working in the academia have a group of questions of their own that they are happy to share with students. Instead of understanding research space and its open problems, one way to get ahead is to work on a problem proposed by a researcher in your department. Using a topic available from your department gives you a natural access to supporting people as you are working on something interesting to them automatically.

3.5 Summary#

Choosing your research question well can have a significant impact on your work and it is important to invest time early on to discover a worthwhile problem as it is the starting point for your study. Especially early on as you might not understand your specific field well, it can be worthwhile to write a rough literature review without any specific publication target as doing this will get you into the habit of capturing information in a scientific format while developing your understanding of what kind of research has been done already.