How to Read a Research Paper
Key Guidelines:
Be patient!
-Especially if the field is new. The work might be totally incomprehensible at first, but be persistent!
-If it really is impossible, you might want to do some background research and reading from textbooks, slides, reviews, lecture notes, before jumping into primary articles.
-Even for someone who experienced skimming a paper could take half an hour. Reading it in totality could take many hours. Papers are dense. Read it like a paper, not a novel.
Do multiple passes
-If you’re not just skimming the paper, it’s often worthwhile to read the paper multiple times
Quick skim
Read to get a deeper understanding of the mechanics of the work. (Can ignore some details)
Scrutinize details (methods, proofs), step by step
How to Find a Research Paper:
Google Scholar/PubMed
Library Database search-EBSCO Discover Search
From a secondary article
Browse a journal or conference proceedings
Click around in the references of other papers
Ask a labmate, mentor, friend, for a reading list
Questions to ask:
What am I trying to learn from this paper?
Look at the structure of the paper: What sections are there? Why? What is the style of the paper? What is the field of study?
Screening questions:
Who wrote the paper? What lab? Is it part of a collaboration? Is it a review paper?
-Is it from a reputable source?
Where is the article published?
-Be careful of questionable journals and questionable scholarship. Always be skeptical!
Look at what other researcher’s comments on the work. Who’s citing the paper? What’s the paper’s impact? In general? Specifically in your subfield? Is it part of a collection of papers?
1st and 2nd Pass:
Get a gist of things:
-First read the abstract thoroughly to get a bearing on what the article is about
-Look for the two or three most impactful figures and tables, and try to understand what’s going on.
-Read the introduction to understand the broad survey of things
-Read the conclusion to understand what the main points of the paper are.
-Skim the references to see how the article relates in the wider picture.
If the article is still relevant and interesting:
-Sometimes if you needed to understand a result, you can stop here
-If you need more detail, go into the content:
If you need, come up with a list of key terms and their definitions. Write out any words you don’t don’t understand.
-It might help you with the jargon
Identify the following:
-What’s the big idea?
-What are the key results?
-What motivates the paper?
-What are the questions that the paper seeks to answer?
-What are the methods the authors uses to achieve their results
Annotations
-Can highlight key points or write in the margins
-Good to take some brief notes
-Draw pictures and diagrams to help your understanding.
Going into the content (Pass 3):
Look through the references at the same time as the review of the literature
Go step by step through the methods and results sections
-What methods were used. Sketch out the flow of the experiment
Look at the data tables and the results sections, do they match up?
-Make sure that you understand the statistical results, vs. the interpretation of those results.
If there are any proofs, go step by step
Look at the supplemental information section if there is one and go over methods, results, or figures.
Further Reading:
How to read a scientific paper | Science | AAAS (sciencemag.org)
How to (seriously) read a scientific paper | Science | AAAS (sciencemag.org)