如何選擇一個好的研究課題

Literature review 3: How to generate a good research question. Three rhetorical patterns

Hey, it’s Dr. Liu here with Better LIFE Research TIPS. Today I want to talk about the three rhetorical patterns for generating a good research question, and what makes a good research question.

A good research question provides guidance for your work, and is the heart of your research. As Aristotle said “Asking the right question is half the answer.” However, developing a good research question can be hard. Why is it hard? Because in order to develop a good research question, first you need a good understanding of your field. You need to read a lot. Someone said, 90% of a research job is done after you have developed a good research question. There is some truth in it. I would say, a good student knows the right answer, but a great student knows the right question to ask. If you know the right answer, you may be able to get high grades, or write a good paper. But if you know the right question, you will be able to start writing grants, and to apply funding for your research.

There are three common rhetorical patterns for generating a good research question. The first pattern is to fill a research gap. In this pattern, we present a picture of current knowledge, and identify gaps or holes in the field. And then we can argue that our proposed research question, if getting answered, will plug up one of the holes. The existing research gaps are usually old problems. We already know their significance. But solving of old problems are often technically challenging. That is why they are still a problem. In this pattern, the key is to find the innovation part of our research. It could be a new solution with new technologies, or some new ideas borrowed from other areas. Also, since it is a old problem, the area may be crowded, we need to find out how others have addressed the problem. How is our study different from other studies. And how does our study fit in with what others have done.

The second pattern is to extend a previous study. In this pattern, we trace the history of research in this field, one achievement after another. Each one study is built on the work of the previous one, and makes the foundation for the next one. Extend a previous study may involve new problems, or even new areas, which often have more opportunities, comparing with old problems. But it is also easy to get lost for unexperienced researcher. In this pattern, the key is to learn how to focus.  In order to focus, we should ask ourselves, why does our work need to be conducted. We need to always keep in mind who is our target audience. And what’s the benefit of our research.

The third pattern is to replicate in a different setting. In this pattern, we identify an existing successful research, and we propose some kind of replication in a different setting, such as a different range of variables, or a different sample population, or even a different field. This is a relatively safe, and a very frequent pattern. In fact, it is a natural way for new knowledge to propagate. In this pattern, the key is to avoid copying errors. Something can be copied, but something can not. When we try to replicate, we should always pay attention to all the premises, assumptions of the work we try to replicate, and the local conditions of our own research, and address the differences they could make.

In order to generate a good research question, these three patterns can be employed alone, or in combinations. There are two types of the best research questions. In the first pattern, after your work, nobody can further improve it, you actually plugged up the hole, and complete the current knowledge framework in the field. In the second pattern, after your work, more people will follow your path and conduct more research, you opened the door for more opportunities. You breakthrough the current knowledge framework.

But what makes a good viable research question? First, a good research question needs to be specific and focused. All the terms or variables in the research question should be well-defined. A good research question provides direction and set clear boundaries in the same time. It should be right for the field based on what is needed, and it should also be right for you based on your resources. The proposed research should be doable. Which also means, in the research question, there should be no preconceptions, or unchecked assumptions about causal relationships. The research question should be defined in a way that it could actually be answered by the data you propose to collect, without any unchecked assumptions. Developing a well-focused research question usually involves a narrowing process, which I will cover in details in a future video. In a typical quantitative research, the research question is often narrowed down to study a specific variable, or the relationships between specific variables. Examples of typical quantitative research questions include: How much is variable A at the given settings? Is there a difference between variable A and variable B? Are variable A and variable B related? Or one step further, is variable A affected by variable B? In the last case, in addition to a significant correlations, we need more evidences, or a valid experiment design, to support a claim of causality.

Next, a viable research question should be arguable. The potential answer to the research question should be open to debate, rather than accepted facts. Karl Popper, a great scientist and philosopher, he claims falsifiability as a criterion for distinguishing the scientific from the unscientific, which is widely accepted by the academic world. If you answer to the research question is unfalsifiable, then it is unscientific. Actually, the answer to a real research question is usually just a statement of what is most probable given the currently known data. The answer should always be open for revision when there are new data available. Research questions are often based on theories. And a common goal of our research work is to make contributions that break or add to the current theories in order to get us one step closer to the truth. A specific type of research question is a testable hypothesis.

Last, a good research question should inspire innovation. Usually, what we want to do is depending on what is needed and what is possible with the current technology. This area is often crowded and have limited opportunities. But technology is advancing all the time. It will create new opportunities for innovation. The new technology may come from your own area or it could also come from other area. That is why sometimes you need to read outside of your area. That is why interdisciplinary, or multidisciplinary efforts are encouraged in the research world. In many cases, that is where the innovations come from. You want to follow technology news to know where the world is going. One shortcut is to identify the leading research groups in your field, and monitoring their activities. Also, as you read papers, especially recent papers, you want to pay attention to their comments about directions for future research. A good paper usually include future directions in their conclusion section. Another tip is, devoting some time to get out of your daily routine, and think about the big picture, which may help you to generate new ideas, or even original ideas.

Thanks for watching, I am Dr. Liu with research tips for the underdogs. In the next video, I will talk about the two principles to set good research objectives. Until next time, keep studying for a better life. 

To Subscribe Dr. Liu ResearchTIPS:
https://www.youtube.com/channel/UCXMJUogKw0lXduef7nu3I7A/?sub_confirmation=1