My thinking on how to read a scientific paper.

How to Read Scientific Papers

As beginning researchers, finding our own way to read scientific papers is the first problem we need to solve, which can help us get a better understanding of papers in less time. Appropriate reading habits also make us think clearly about how to continue our future work.

I’ve read lots of posts and suggestions about methods to read papers. Among them, the following three articles are my most recommended.

How to Read Paper

How to Read a Scientific Paper

How to Read a Paper? Ten Personal Tips

The first article introduces a universal way to read a paper, from which we can have preliminary knowledge on what need to do during reading and establish a basic framework for our own way. The second one also provides some steps, besides, it pays attention to what to do after reading a paper and the ways to sharpen our critical analysis skills. As for the third article, there are several personal tips that help me a lot. In addition, it is the only Chinese post and its author studies a similar major to us.

Maybe different researchers have different ways but effective ways must have something in common. Here I present some same points below, which are mentioned in most articles and may be the key points when reading papers:

• Categorize papers before reading.

  Before we begin, we must categorize papers according to their keywords and abstract to find out which papers to read carefully.

• Read a paper multiple times instead of trying to handle it just in the first pass.

  For beginners, there are vast quantities of information in a scientific paper, including unfamiliar terms, abstract concepts, new approaches, and so on. It is impossible for us to understand and memorize all of them in the first reading. Therefore, reading a paper time after time is necessary. Of course, each time we need to focus on different parts and achieve different goals. The details can be found in the first link above. In practice, I can get a better understanding of each paper by reading in this way, and also I am less nervous each time reading for not eager to solve all my questions in one pass.

• Pay attention to the title and sub-section headings.

  It is helpful to list the title and sub-section headings of a paper, from which we can have an overview of the content in each section and get to know how the author organizes the points.

• Focus on the figures.

  According to my observation, figures in a good scientific paper can imply all important information, consisting of motivation, methods, models, results of experiments, comparison, and analysis.

• Take notes.

  The importance of taking notes is obvious to everyone, but the key lies in what to write down. Besides the main content of a paper, I take notes of my questions and understanding of some nonintuitive points.

• Read critically and creatively.

  Whether to read a paper critically and creatively is the main difference between amateurs and researchers. When reading a research article, we can’t assume that the authors are correct. Maybe it’s difficult for beginners to ask valuable questions but it’s important to keep in mind that try to ask questions.

Besides the six points above, I have some personal thoughts on how to read papers after a period of reading. Here are some points I benefit from a lot:

• For the first or second pass, don’t be entangled with some small questions. Instead, we can write them down and solve them together after understanding the main content of the paper. Because dealing with these problems along with reading may interrupt our thoughts and waste more time.
• If possible, we’d better reserve a long and continuous time for reading a paper to avoid forgetting details in context.

​