How to read an empirical paper
| Type | Team Size | ||||||
|---|---|---|---|---|---|---|---|
| Collaborative Tools | Software | Personal Skills | Productivity Tools | 1 | 2-10 | 11-30 | 30+ |
In short: This entry provides tips on how to read an empirical, scientific paper.
Just as much else, reading scientific papers is a matter of practice. When I started reading scientific papers, it took me hours to get through one paper, and I can very safely assume from my perspective today that I understood next to nothing back then. In between 'beginning to engage with science' and the position I find myself in today, there is a long but continuous journey of learning and gaining experience. I cannot tell you exactly how to learn to read scientific papers, but here are some suggestions that may help you in the beginning. These tips focus on empirical papers - for conceptual papers however, much is similar, and you will realize the differences as you read more papers of all kinds.
1) Read the abstract first. The abstract is a systematic from to condense information, and what the authors feel is most important. Therfore, grasping the abstract should reveal what the paper is all about. Reading the abstract twice can be a good first step to learn what is interesting about the paper.
2) Cross-read the introduction. After all, it leads to the research questions or hypotheses. However, it often contains deep knowledge that is not necessary to understand the bigger picture. Also, do not go down the rabbit hole of following the referenced literature for now - you may come back to it when you've gotten the bigger picture. First, take it step by step.
3) Do not try to understand every word. In the beginning, much is overwhelming, because much is new. Try to hold on to what you understand. Each discipline has its own words, and sometimes different words even mean the same in different scientific cultures. Over time you will learn more of these big words, and what begins as a puzzle will form a picture.
4) Note down the main points that the paper is focussing on. Many papers are well structured, and you can easily trace down the structure throughout the text. Your notes will reveal the methodological and theoretical logic of the paper, which is a good thing to come back to when you feel lost in the details.
5) Do not preoccupy yourself with undressing the methods if you have no knowledge about these methods. Peer-review should have usually safeguarded that the applied methods are OK. So if you have no knowledge about the specifics of these methods, just accept that something was done, and try to understand what you can.
6) The results are the heart of every paper. Read these in detail, and try to write down the main points. Also, be on the lookout how the authors qualify the individual results. Is a result more generalisable, or rather specific?
7) In the discussion, look for the mentioning of limitations of the results. Then, try to see how the authors connect their specific results to the wider research picture. Lastly, check whether the authors give some recommendations for future research.
8) Learning to read papers takes time. In the beginning, read a lot, but be patient with yourself. You may understand very little. The first 100 papers are more about familiarising yourself with the general form, not the content as such. Reading papers needs to become a habit, not a pain.
9) Last piece of advice: Make your peace before reading. Instead of dreading towards the time where you feel you have to have to have to read the paper, just take one or two minutes, breathing in, and breathing out. Clear your mind, to prepare it for reading, just like you would stretch a bit before going running.
10) Exchange with others, and use peer pressure to your advantage. If you sit together and read, and then discuss what you got from the paper, then you can learn to understand how other people read papers, which can be quite a beneficial skill. Learn together, and nudge yourselves through the papers. You can do it.
The author of this entry is Henrik von Wehrden.
