During my postgraduate studies I went to a journal club every week organised by my PhD supervisor. It was here that I learned how to read a scientific paper, and gained confidence in critiquing published studies. Now I run a journal club for my MSc students, and sometimes (but not often enough!) for my PhD students.
Here are some tips I give my students before they lead a journal club meeting:
- Pick a data paper: it is hard to lead a fun discussion on a review paper.
- Don’t pick one of your own papers: this is egotistical and stifles discussion from non-authors (or could set you up for a bruising!).
- Send everyone a PDF of the paper a few days before the journal club, so they have time to read it.
- Read the paper thoroughly yourself several times. If you have time, see if it has been discussed in subsequent review papers, or in blogs
- Don’t use powerpoint – you will end up lecturing instead of leading a discussion; instead, make sure that everyone has a copy of the paper in front of them
- Start out with a one or two sentence statement of the major headlines of the paper
- Give some context to the paper, especially if the authors have failed to acknowledge that what they are saying is controversial.
- Give a good summary of the methods. Think carefully about any shortcomings. The author may not have mentioned these in the introduction or discussion
- Talk your audience through each figure and table. Most of your time should be spent on these
- If there is something you have not been able to get your head around, tell the audience this and ask if any of them have understood it. This will either help you to learn something, or will uncover a flaw in the paper
- Summarise the author’s conclusions
- Say whether or not you find the paper convincing
- If the audience has not chipped in by now, highlight points that your audience might want to discuss, or ask them questions about what they thought about particular aspects of the paper
- Reserve about half of your allocated time for discussion
- Encourage as many people in you audience to contribute to the discussion as you can
- Try to keep the discussion on the topic of the paper, and especially its methods and the interpretation of its results.
- Don’t let people go off on big tangents unless the whole group is interested
- Beware of the loud person who hasn’t read the paper, but uses it as a springboard to talk at length about one of their hobby-horses
- Use colloquial language
Common mistakes in leading a journal club:
- Spending too long on background context – you are here to discuss a paper, not give a lecture
- Plod through the paper in a linear fashion, reading out sentences from it, and never giving a broad overview or saying anything that explains in colloquial language what is going on. This is a sure sign of not having understood the paper
- Just focussing on the introduction and conclusions. This might seem an easy way to go, but you will have little to discuss and will not be able to critique the paper.
- Defending a paper to the hilt even though it has obvious flaws
- Being too negative about a paper because you can see a few weaknesses in it. All papers have weaknesses
How to be a good participant in a journal club discussion:
- Read the paper
- Make sure you understand the figures
- Contribute questions and comments about the papers methodology and its interpretation of results
- Don’t make long speeches, especially not using the paper are a springboard to talk about a tangential issue that you have thought a lot about before, or feel strongly about
- Be open about what you don't understand
If you have more ideas about how to make journal club meetings work well for everyone, please add them as comments below!