Reading research papers without drowning

In the first months of a PhD, the reading list grows faster than you can read it. What helped me most was accepting that not every paper deserves the same depth.

A well-known approach, described by S. Keshav in a short note called How to Read a Paper, is to read in passes. I use a loose version of it.

First pass: decide. Title, abstract, introduction, section headings, conclusion, and a glance at the figures and tables. Five to ten minutes. The only question is: is this relevant to what I am working on, and is it worth more time?

Second pass: understand. Read the whole paper, but without getting stuck on every equation or robustness check. Look closely at the research question, the data, the method and the main results. Note the references that look important.

Third pass: rebuild. Only for the few papers that are central to my own work. Here I try to reconstruct the argument and the model, check the assumptions, and ask how I would have done it differently. For DEA papers, this usually means looking carefully at which inputs and outputs were chosen and why.

A few other habits:

• Write one or two sentences after each paper, in my own words. If I cannot summarise it, I have not understood it yet.
• Read reviews and meta-analyses early. They map a field faster than single studies.
• Follow the citations in both directions. Older papers show where an idea came from; newer ones citing it show where it went.
• Keep a reference manager from day one. Rebuilding a bibliography later costs far more time.

Reading efficiently is not about reading less. It is about choosing where to go slowly.

---

Photos: sam651030 on Pixabay, used under the Pixabay Content License. These are not my photographs.