Many labs have semi-regular “journal clubs” where they read and discuss scientific papers relevant to their research. Journal clubs provide an opportunity to discuss, in detail, research going on in other labs. A large part of a scientist’s job is to think critically (but fairly) about research in their field. Journal clubs also provide an opportunity for students to learn and hone those skills. My experience has been that assigning people to “present” papers in shortened, verbal form usually isn’t that helpful or interesting. Instead, it’s best for participants to approach papers as an interested but skeptical outsider. As a reviewer, basically.
Here are some questions I ask myself when reading a paper:
- Why should anyone read this paper? What is the paper’s main point, the thing that everybody should take away from it? Is this point important?
- Is the paper right? How do the results support the main point? What are some alternative hypotheses for the results? Did the authors do the right controls? Did they do the right statistical tests?
- How is this paper wrong? All papers are wrong somehow. If you don’t find something wrong with a paper, you haven’t read closely enough. Some papers are more wrong than others. How do the errors in this one affect its main point? Are they fatal flaws or only minor quibbles?
- Is the paper well written? Is the data presented clearly? How could the writing be improved?
- What does the paper do right? What about this paper is worth emulating in your own work?
- If you were a reviewer, what would you recommend to an editor: accept as is, accept with minor revisions, send back for major revisions and re-review, or reject outright?