How To Read and Summarize Technical Papers and Journal Articles
Updated: Apr 10, 2019
Reading technical papers is something no student can escape. This article presents tips on reading journal articles and technical papers quickly and efficiently.
April 8, 2019 By Destenie Nock
In graduate school reading technical papers (aka journal articles) is something that you cannot escape. A common mistake people make (particularly when coming from undergrad) is trying to read each research paper like a text book. If you can read and summarize a paper efficiently then this will help you save a ton of time. Also when you are writing your thesis/dissertation you will have to include some sort of literature review to justify the motivation for your project, so it will be good to catalog all the knowledge you are reading to save time down the road.
Below are my suggestions on reading and summarizing technical papers. What I have written below is a combination of everything I heard and learned from many mentors and my own experiences throughout graduate school.
Suggestions for Reading a Technical Paper
First read the Abstract - find out if it's interesting and if it relates to research. How is their topic similar/different to yours?
Skim introduction and lit review to get a sense of what else has been done, and motivation. Find the problem they want to answer, if you didn't already find it in the abstract. Think about how that problem is connected to your research.
Read over the methods and approaches (this will give you the base for understanding some of their results). If the Methods section starts to get too technical go to the results section.
Results and Discussion Section: What did they find? What do the results tell us? - Sometimes the answer to the second question is found in conclusions
Summarizing Papers (Making a Lit Review)
Put the author's name(s) and paper year at the top. This is important for tracking and not losing papers. Would also be good to put the web link.
Write 2-4 sentences on the problem the paper is trying to solve. What is the reason for doing the research in the first place?
Write 2 - 4 sentences on the solution the paper proposes. What is the solution they are proposing to solve the problem? This could be the methodology, metric, model, framework, algorithm, etc. How they will answer their overarching question?
State 2 strengths of the paper. What was good about the approach the author(s) took? Why should we listen to these authors? Did they collect good data? Has their approach been tested in other areas/articles? Did their approach use consistent data collection methods across different variables (ex: they found hourly data for wind and solar, as opposed to daily data for solar and monthly for wind)
State 2 weaknesses/unanswered questions of the paper. Were there holes in the data? Did they approximate a majority of variables? Are their assumptions inconsistent? Do their results change drastically under different assumptions? What are opportunities for future research directions? Sometimes weaknesses/unanswered questions can be found in the future research section.
Example Summary Template (this does not have to be followed for every paper)
Johnson et al. (2056)
Johnson et al (2056) focuses on the issue of ______________. This problem includes/encompasses/involves/etc _____________. As a solution the authors propose/create/etc ___________. The researchers used _______ to solve the problem above. A strength of this paper was that the data/model/metric/etc _________________. One weakness in this study is that a ___________. In addition the authors could have done _______________ to make their analysis stronger.
Hopefully these tips can help you move through those papers faster, and lead to less re-reading throughout your academic career. If you like this post and want to get more tips feel free to subscribe below (I promise not to spam you).
Note about the author: Destenie Nock holds a PhD in Industrial Engineering and Operations Research from UMass Amherst and two BS degrees from North Carolina A&T State University. She finished her PhD in 4 years thanks to the help of 7 undergraduate research aides and a wonderful adviser. Throughout grad school she spent a lot of time trying to figure out how to reach her optimal level of success...a task she still has on her to-do list.