Understanding Scientific Research Papers: Before Reading, Do These Things
Here’s seven important things to consider before you start to read a research paper.
I’ve read a lot of papers in the process of doing my PhD and I’ve got a lot better at being able to read them, digest the information and use it in my writing.
I’m hoping to demonstrate, by using a research paper which is almost definitely going to be outside your field, that it is possible to understand and grasp research papers without a degree in science.
If you want to follow along, the paper I have chosen, is by Zeppenfeld et al., and was published in 2017 in PLoS One (an open access journal). It’s called “Rhizosphere hydrophobicity: A positive trait in the competition for water.” If this doesn’t interest you, then skip the bits in italics!
The paper I’ve chosen, is a good paper to consider, because there are bits in it I don’t understand (modelling 🤦♀️), but it doesn’t mean we can’t read and analyse our thoughts on their work.
The post is the first part in my series on how to read and understand research papers and scientific articles. This post concerns all the things we want to do before we dive in and read.
Before we start
There’s loads of things we should do before we even start reading the paper. Time spent here can save you time in the longer run plus identify potential new avenues. I’ve pulled together seven key things to do before starting to read a scientific research article.
1️⃣Who are the authors and where are they from?
Don’t worry if you haven’t heard of the authors in early stages, but use it as a guide to identify people who come up and time and time again in your literature searches; it may be worth checking out their work directly, searching out seminars and other work, like a personal blog. These can provide valuable additional insights.
What is also very important to be aware of, is something called ‘helicopter research.’
Helicopter research is when wealthy countries, like the US and UK send teams of scientists into less wealthy countries, do science and take any science home with them, leaving scientists and people in said country high and dry as to any benefits which come with said science. In other words, if a study takes place in Ethiopia, make sure there is a decent team of scientists from Ethiopia on the author list.
Zeppenfeld et al. (2017): In this paper, the authors are a mixture of spatial ecologists and modellers at least one of whom I have heard of before. They are mostly based in Switzerland and have the expertise to carry out the work.
2️⃣Check the journal.
We’re not interested in whether it is a so-called ‘prestigious journal’ (e.g. Science/Nature), we are interested in whether it is an honest journal. There are many smaller publishers out there who produce excellent, but very specialised science.
What we want to be sure about is that it is not a predatory journal. There are some borderline journals with questionable methods like Frontiers, but I know some editors and reviewers on these journals that are highly respected, so it can be hard to gauge.
Although a little controversial at times, you can use Beall’s list to determine whether a journal has predatory tendencies if you are batting in the dark.
Zeppenfeld et al. (2017): PLoS ONE are an interesting journal. They were the first completely open access journal, and with this regard they set the scene for others to follow. They are also unusual because their subject remit is incredibly broad. If authors are unsure about their intended audience, this is a good one to submit to.
3️⃣When was it published?
Although academia is notoriously slow moving, science moves on fast. The 2010s may not seem like that long ago, but there may have been follow on work since, sometimes which changes the interpretation of preceding work.
If you are looking at medical journals, then 1 year, is old.
Conversely, don’t discount old papers. Due to the limitations of distribution methods in the actual ye olden days, many papers didn’t get the recognition they deserve. There is much foundational, excellent quality research to be found in the 20th century and even the 19th. They also make for easier understanding because the methods, techniques and statistics were much simpler.
If you’re looking to start somewhere with your research topic, the history of your field is a great place to start.
My supervisor loves to see the use of old papers. In my field of work the work of Hiltner from 1904 is credited with the original conception of the ‘rhizosphere’ as “Einflusssphäre der Wurzel” or ‘soil influenced by roots’. Because the paper was in German and not available digitally, many scientists rely on translational pieces like Hartmann, Rothballer and Schmid (2008) to understand the work; which of course isn’t open access, and is another problem plaguing older papers.
Zeppenfeld et al. (2017): This paper was published 6 years ago. It is likely there has been follow on work since then, so it is worth bearing in mind. As this is a modelling paper, often the next step is to carry the work out experimentally. As far as I’m aware this has not been done yet.
If you’re curious about how to do great research and how using different tools (physical and digital) can help you do this, I’ve just started a new free newsletter called Brain STREAM. I’d love to have you on board!
4️⃣How did you find it?
Was it recommended to you? Was it through a literature search engine?
Whilst this may seem a weird point to have here, how a paper was found can tell you about how it might link into your work. If a paper is easy to find, it’s more likely to be popular and well-received or conversely, controversial. If a paper is hard to find, it’s likely more obscure and specialist, but may hold some valuable insights or conclusions which are different from the ‘popular’ articles.
Often we find papers relevant to something else we are working on when we don’t intend to. It’s quite exciting to find a paper which explains the result in another paper unconsidered by the authors.
Zeppenfeld et al. (2017): I found this one when looking for one of the authors, Carminati, specifically and it came up again when I was creating a literature overview of ‘rhizosphere repellency’.
5️⃣How many times has it been cited by subsequent authors? Who cited it?
Sometimes excellent papers get lost with time and don’t get the recognition they should have, so few citations may not indicate a poor paper, though it may do.
On the other hand, if a paper is well cited, it’s going to be a must read for the topic you are considering.
Do be aware of self-citations. You’ll likely get to know the authors that do this. I don’t have a problem with it per se. If you are in a niche field and you are the expert, and you need to build on previous work, then OK. But I think it needs to be clear the author(s) did the work, rather than mentioning it in third person as if someone else did it.
Zeppenfeld et al. (2017): This paper has been cited 18 times. This is an expected number for a paper in this kind of cross-disciplinary area (soil physics, modelling and plant ecology). It would be worth me checking who cites it, as the work is highly relevant to my current experiment. In doing this I can see a couple of papers published in the last couple of years which have not come up on my radar yet.
6️⃣Why are you reading the paper?
Consider why you chose the paper to read. What do you hope to get from it? Did they promise exciting results in the abstract? Consider whether it is new or controversial.
We can save time when reading a lot of papers by honing in on what we actually want to get from reading it. We might be looking for a result which supports our own. Other times we might be reading about a new technique. The most exciting ones use new techniques and have results contrasting our own.
Consider how you feel before and after you end up reading it. Do you end up disappointed or are you enthused with ideas?
Zeppenfeld et al. (2017): In this paper, I hope to find out how making the rhizosphere (soil around plant roots) repellent to water (hydrophobic), can help plants compete better for water. Making something repellent to water to get more water is counter-intuitive, but I’ll hopefully find out in this paper how it could be used to good effect by a plant species, to prevent other plant species from getting water. Will I be disappointed? I have to bear in mind this is a modelling paper, so it’s about testing the potential of a theory rather than actually testing it experimentally.
7️⃣Where does this paper fit in the context of other papers?
This is similar to many of the last points combined, but here we want to consider the wider context of the paper. We want to consider where the paper fits in the research landscape. Is it central to your work? If it is, we need to prepare a long focus session to read it. If it’s on the periphery and might be interesting, then reading as quickly as possible might be a better bet, just to see if anything sparks your interest.
A technique I learnt from Jeremy Nguyen is that of creating a birds eye view with loglines of the literature. Basically, fire a search term into Google Scholar or Web of Science, and using just the abstract of papers, write a few summary sentences. If you aim your search term to throw up 80–100 papers, we can very quickly (<1 hour) get an idea of how multiple papers around a research theme fit together.
Although this is the last step in this list, it is also one of the most important.
Zeppenfeld et al. (2017): This paper is a really nice link between a soil property (repellency) and plant community dynamics. There aren’t many papers around which consider how interactions between plants can influence soil physical properties (beyond bulk density, which is the ‘go-to-doesn't-tell-you-much’ soil physical measure).
Now that we have scrutinised the paper in its wider context, now it’s time to set your physical and digital environment up for reading. That’s Part 2!
(I’ll add the link in here once I’ve written it!)
If you want to skip ahead and see how I use Obsidian throughout my research and writing process, here’s a fast forward to an overview of the entire thing:
If you’re curious about how to do great research and how using different tools (physical and digital) can help you do this, I’ve just started a new free newsletter called Brain STREAM. I’d love to have you on board!