While research mentors try to help on a lot of aspects of academic research, little attention, if at all, is given to how to read a research paper. Reading is an essential chore of the overall gamut of activities a researcher has to perform. A researcher has to spend a considerable amount of time reading other peoples’ work before he/she (consider “he” as sufficing for “she”, and “him” for “her”, and so on for the rest of this article) can propose solutions about what he intends to do. This article tries to help a beginning researcher master one of the essential spheres of academic research. That is reading.
In today’s multidisciplinary research environments, a new researcher can simply be overwhelmed by the sheer amount of academic literature present online almost on any conceivable topic. As one person put it, gathering information nowadays is tantamount to drinking water from a high-pressure water hose. Add to it the fact that almost every conceivable discipline is equally saturated with a lot of citations of works that comprehensively address and solve given problems. The truth is that even after reading a brilliant research article, a person leaves with a void in the mind with a big question mark about what should be done now?
Finding credible sources to read from is another problem that a person confronts. But that is not the scope of this article. This article merely tries to elucidate a nice reading strategy once an article has been screened and approved for reading by the reader. So where should we begin?
Searching Articles: Select an article first that you want to read. How are you going to do that? Normally people use Google scholar to scavenge for relevant articles. If you are satisfied with the credibility of the article, just give a quick eye-ball to the abstract and the conclusions. If they sound appealing and mostly relevant, you should decide that you are going to read the whole article. As a matter of fact, if you have enrolled in some place in a formal research position, then in the beginning, you would have to read almost everything that is relevant. This is to say that your reading choices do not merely depend on whether or not you like something to read. To this end, your reading choices also do not depend on whether or not you find something appealing. You have to read everything that is relevant to your work. And relevance at this stage is also quite relative.
Checking for Relevance: Relevance of what you are going to read can depend on many things. In today’s multidisciplinary world, a lot of literature that is published outside your domain can also be relevant to what you are doing. Consider that possibility when you are working in communication networks and literature published in the domain of fluid mechanics can be relevant to your problem. Similarly, while working on a problem in water distribution networks, a research paper from communication networks can have useful ideas for you.
To cut the long story short, skim through the abstract and the conclusions about the relevance of a paper’s contents to your work. Just read what the authors have proposed to do. If you find it relevant and appealing, just archive the paper for reading.
Read the Hard Copy: Once you have decided that you are going to read the paper, just print it out. Read the hard copy. It is a very good practice to read the paper from its hard copy and to take notes on it as you read it. Possibly try to keep a text highlighter and a pen with you to highlight important excerpts, ideas, terms, formulae, and findings and to take small notes on the paper. Reading the hard copy will engage you with the paper thoroughly. You simply cannot develop this type of engagement with a soft copy of the paper.
As you print hard copies of research papers to read, maintain an effective archival mechanism as well. Putting dates and your name on the paper could be a very good thing as it may remind you in future about what you read at a particular point in time in the past and why you read that. Having a place such as a cupboard, chest drawer or a file cabinet can really be helpful for archival.
Maintain a Reading Schedule: As a researcher, reading is going to be one large part of the larger set of activities that you would be doing during your tenure. Like all great things, becoming a great reader will also require immense focus and practice on your part if you want to become a better researcher. Other activities, along with immense reading, you would have to engage with during your research experience would involve designing and conducting experiments, contemplating on your ideas and writing up your work for various conferences, journals and eventually, your very own dissertation.
In order to become a better reader, you would have to do a number of things. The chief among them is to maintain a healthy and vibrant reading schedule. Having said that, you should plan ahead of time to maintain an effective reading schedule that fits in well with the other sets of activities that you would be doing as a researcher. As you experiment various ideas in the laboratory, and as some of them fail while others yield nice results, you will find ample free time in which you can fit in your reading schedule. You should simply learn to dedicate the free time slots to maintain a nice reading schedule.
Enjoy the Read: Once you have decided that you are going to read a particular paper thoroughly and downloaded it, plan a reading schedule. If you are slightly seasoned at reading this particular type of papers, you might find the read enjoyable. You may as well take your paper along with you on a long walk and read it while relaxing along the bank of a lake. Spend time reading it and highlight the important ideas.
It is quite important to be relaxed and calm as you read through your selections. This is important to fully grasp the ideas and to reflect on them. It would not hurt to maintain a pleasant ambiance in the place where you usually read. Keeping a table lamp reflecting a pale luminance on your readings can help you create a serene ambiance. Solitude also helps a lot in becoming a better reader. Find spots in your workplace and library that are normally quiet. You may also like to develop a culture of solitude in your home or dormitory. Remember that no great work is possible without great solitude.
Chew it Well: Chewing a paper well simply means to read it extremely thoroughly, no matter how long it takes. It is to read the whole paper verbatim while also grasping even the minutest ideas presented in the paper. A well-chewed paper should result in a thorough, vivid and profound understanding of most of the ideas presented in it on your part. Remember, however, that you have to chew a paper really well if you find it extremely relevant to your work or if it has extremely ideas presented in it that you can use for your work. For other types of papers, you will either skim through them casually, read them slightly thoroughly, or have a tendency to litter them as soon as you reached the middle of the paper and found them rather uninspiring.
Grasp the Central Ideas: An important thing to be aware of while reading a paper is to know what you are looking for while reading it. Many novice researchers simply do not have a clue about why exactly they are reading a paper, and they keep on reading paper after a paper unless they eventually realize that what their motives should have been while reading all those papers. So the things you should be trying to understand as you read so many papers can be summarized as follows:
- The problems that the paper tries to address.
- Try to understand how they formulate the problem. This is a very important aspect.
- How they addressed the problem? What algorithms, methodologies, and techniques were applied to solve the problem.
- In various problems concerning empirical research, various simulators are normally employed. Read the paper to look for the specific simulators that have been employed. Employing those simulators can be quite beneficial for your work.
- In most research articles addressing a particular problem, already pre-processed datasets for that particular problem are employed. Look for how data was gathered and stored as it was used in research. Especially, read about how the same datasets can be obtained or reproduced for your work.
- Understand the results of the paper and what sorts of comparisons were made with other methods.
- Try to understand what methods were used for making comparisons and benchmarking the work. Normally researchers employ statistical methods and numerical techniques to benchmark their work. Try to understand these techniques as you will find them useful when you advance in your research.
Reading Recursively: Good research papers normally present only the crux of the most useful results of works of others and your own. This is absolutely true about great research papers that are published by journals like IEEE Transactions, ACM and the likes. You might consider the literature presented in a research paper as consisting of juice, the pulp of which is left in the various citations which the paper cites. This type of writing can haunt you, especially in the beginning. A nice article normally only cites important ideas that are written elsewhere. A citation is normally given. But understanding the whole article without understanding the central ideas on which the article is built is normally not possible unless you have become a domain expert. This simply implies that in order to fully grasp the ideas in a research paper, you will have to do something that I call to read by recursion. This means that you will have to download a substantial amount of literature the research paper cites, read it, understand it and then come back to the research paper and read it in the light of your understandings of what the paper cited. This might sound daunting in the beginning, and as you read this, but this is only how you can become a better researcher.
Reading Spans: Depending upon what you are reading, your reading spans for a single article can be as small as a few hours to as long as a few months. Actually when you have gained enough expertise in a certain domain, or a certain approach or a methodology that addresses a particular type of a problem in a certain domain, your reading can last a couple of hours. You can really be done and dusted with reading a paper in that much time. However, if you are a novice, or if you confront a certain type of a paper that you have never ever seen before, your reading span may as well last from a week to a few months for understanding the paper. Depending upon your domain, or possibly the mathematical rigor required to understand a research paper, your reading span may as well last for six months or even longer. In this case, you may as well have to read recursively all the relevant literature and ideas that the research paper has cited and come back to reading it when you have understood everything so well. But just relax, there is a way to do that effectively and to make the whole experience more enjoyable as well.
Reading Bit By Bit: Given the nature of reading involved for a typical research problem, it can be quite mundane at the same time it is interesting and exciting. Similarly, the practice of reading can be thwarting and thrilling at the same time. As a matter of fact, even for a problem that you find extremely interesting, there could be patches of reading involved in it that you may find extremely boring, monotonous, dry and sweating. There could be points of time in your life where you may totally regret your decision to work on a particular problem as a researcher. There can be workarounds too, however, and you may make your reading experiences more fun, enjoyable and thrilling. A lot may depend on your personal motivation and commitment to your problem and your professional choice.
Working to solve a research problem can be literally challenging. You could have spans of time where you might think that your life couldn’t have been more fun without it, and at the same time, you could have times where you totally regret your professional decisions. The key is to learn to be calm and cool. Perseverance with a relaxed mindset can take you a long way. At the same time, always remember that cracking a research question in most cases is going to be like eating an elephant. What you will have to learn to do is to eat it bit by bit.
There is an increasing interest in upgrading the E-Model, a parametric tool for speech quality estimation, to the wideband and super-wideband contexts. The
If you found an error, highlight it and press Shift + Enter or click here to inform us.