We all know that reading the literature is essential, but it's not enough to just sit and read a stack of articles. Reading journal articles is a skill and, like any other skill, how you approach it makes a big difference. So it's not just a question of how to read a journal article, but how to learn how to do it.
When you're an experienced academic, reading journal articles is relatively easy. This is because you already have knowledge of the field and enough experience to recognise the significance and quality of the work. In other words, you can see where an individual paper sits in the wider context.
But as a PhD student, initially at least, you don't have that experience. So, while plenty of academics have written guides on how to read journal articles, you can't necessarily follow the same process as them (yet).
While you can learn a lot from experts, the processes they follow now aren't always the best ways to train your own skills. This is because their processes rely on prior experience.
How to learn to read an academic journal article
Instead of focusing on how to read an individual article, let's think about what you're actually trying to achieve. Are you trying to build your knowledge of the field? Or trying to understand a specific technical point? Are you trying to find a gap in the literature? Depending on the answer, the papers you choose to read, and what parts of those papers you focus on, should be different.
Of these, the first aim should be to build some knowledge of the field. How do you do this without drowning in thousands of sources?
1. Identify the most influential sources
Journal articles can be divided into two main categories; there are groundbreaking, influential papers that change the way the field thinks or operates, then there are incremental papers that just add a little bit, but don't have a huge impact.If you understand the influential, groundbreaking work, this gives you a foundation for understanding the incremental work that followed. Fortunately, there are relatively few of these groundbreaking papers and they are easy to identify (because they are highly cited).
Try to find around 5 highly influential sources related to a specific topic. If you can understand;
- what they discovered/ invented/ proposed
- what problem this discovery/ invention/ proposal attempted to solve
Then you have a good start. However, there's a problem...
If you look just at those original sources, they might be very hard to understand. Simply re-reading them won't necessarily help, because academic articles usually assume a lot of pre-existing knowledge.
2. Focus on the concepts you identify, not the sources
If you've identified important developments in the field, but don't really understand them, you now have a new aim; to find sources that explain those concepts.
Primary research articles may not be the best things to focus on, because they aren't written to teach. Instead, look for textbooks, review articles, Wikipedia pages, YouTube videos, or people in your institution you can ask.Y
ou can go back to the original source later. For now it's enough to know where the idea came from, while looking elsewhere for an initial explanation.
3. How did these concepts influence the field?
A key part of understanding the literature is understanding trends in your field. So what effect did these influential papers have?
For example, did a particular theory spark a bitter debate in the field? Did a particular invention open up new possibilities for research? Or did a particular discovery reshape the fundamental understanding of some phenomenon?
This gives you an initial, broad context for understanding some of the finer detail contained within the literature. But equally importantly it gives a focus to your reading; a specific, achievable aim that isn't too overwhelming.
4. How to read journal articles for context
You can strengthen this contextual knowledge by reading the introductions of recent papers. Every research article starts with a brief overview of the background and current state of the art, so reading just the introductions of a few recent journal articles is a great way to get a quick summary of what's happening in the field.
You may find that a lot of them say more or less the same thing. In this case, you know what the field considers to be important. Or if you see they all say contradictory things, you know that there is no consensus in the field.
Again, the aim determines how you read the individual journal articles and what you focus on.
5. Get some practical experience
It isn't enough to know how to read a journal article. You will find that it gets much easier to read and understand once you have some practical research experience. It's by doing research yourself and making mistakes that you're able to spot problems in the published literature and to really appreciate the best work that's been done.
Reading helps with the practical work, but the practical work helps reading too.
How to read a journal article once you've got a bit of experience
Once you've built up some broader contextual knowledge, you can think about how to read a journal article in isolation.
Many people advise reading the abstract first, then the conclusion, then going back to the introduction. The exact order varies (and this one advises skipping the abstract altogether), but it's basically a way of systematically assessing whether the article is worth reading in depth.
Again, though, I'd say that how you read should depend on what you want to achieve and what led you to the article in the first place. Context is everything. You might read differently depending on whether you've just done a search that gave you 5000 results and need to filter through them, or you've done a search that resulted in 5. In the former case, filtering by title initially is the only way to go. In the latter case, slow down and read everything.If you found a paper because you've noticed a lot of relevant sources referring to it, it's probably best to treat it like one of the ground-breaking papers, but also noting what the authors you have already read are saying about it.
So you have to adapt, but as a general approach to reading journal articles...
Read the introduction first
I would usually advise reading the introduction first. The introduction should set out what the paper aims to do, and if you skip this then nothing else will make sense.
The first thing to look for is what problem they are working on. If it isn't clear, or if the problem isn't interesting to you, move on to the next article.
What you do next depends on the immediate relevance to your project
Different literature will be useful to you at different times, depending on what you are trying to achieve. You may re-visit some sources several times throughout your PhD, and what isn't relevant or useful now may become useful later.
If the work seems highly relevant to what you are doing, or helps you to solve a current problem in your research, either slow down and read carefully, or if you don't have time, put it to one side but make sure you have a way of remembering where to find it later.
You don't have to summarize everything
Some say that you should summarize everything you read, but I don't think this is the most important thing to do. It's possible to fill fifty notebooks with summaries, but that isn't the same as having knowledge.
What you need is a picture of the literature (the key discoveries, most relevant articles and the trends in the field) which you carry in your head at all times. Then you where to look to find relevant sources to fill in some details when you need to.
What I did in my own PhD was build up collections of literature around certain topics. I put printed copies into ring-binders by sub topic. Any notes were taken in the margins of the paper, so they stayed in context with the whole text.
I knew which were crucial to my work and I knew which were most influential. I also knew the kinds of problems that were being worked on in the field, and what techniques were being used, and I knew where to look to find details when relevant.