Straight from the horse’s mouth: How to read (and hopefully understand) a scientific research paper

Also available as an audio podcast here.

I’ll let you in on a little secret. Sometimes I don’t understand science. I’ve been either studying or working on infectious diseases since I was a bright-eyed undergraduate in 2007 (that’s more than 10 years, it’s more baggy-eyed now). Yet even now, I can settle down to read a research paper and be no more sure of what was said when I’m finished than when I started. If you’re not a scientist (and sometimes even if you are!) scientific papers can seem deliberately confusing! And often they are. Scientists are often pressured into writing in an overly complex style, by journals and also by the fact that ‘this is how it’s always been done’. We don’t wish to appear like we don’t know what we’re talking about, so sometimes we use lots of complicated words when really, fewer simple words would have worked just as well.

It’s starting to dawn on us (and in academia in general) that we need to start writing about our research in a way that non-scientists can understand. The explosion in fake news surrounding the COVID-19 pandemic just highlights that many people feel more comfortable watching and believing a sketchy YouTube video or Twitter thread uploaded by a pop star (Nikki Minaj, I’m looking at you) than the multiple, peer reviewed scientific papers from scientists who have decades of experience in that specific field.

We need to fix this and the responsibility (primarily) lies with science. We need to stop pretending we’re too clever to explain our science to the public and write plain language versions of our research. It’s starting to happen, certain journals now require scientists to publish a plain language abstract. But until Gareth who works in the corner shop, or Femi the primary school teacher can pick up a scientific paper, understand it and make their own informed decision, we’ve got work to do.

In this article, I’m going to try and explain how scientific papers work, so that everyone’s got a chance at understanding them. Some papers will be harder to read than others, but give it a go, the more you try, the better you’ll get. For a really in-depth explanation of how to read scientific papers, this is an excellent article.

Often scientific papers use a lot of abbreviations and acronyms. Often they have a table at the top of the article explaining what they all mean. They should also explain the full meaning or name the first time it is used, with the acronym in brackets after it. For example “this paper is going to talk about antimicrobial resistance (AMR). The main concern with AMR is…”

Some words you might find helpful when reading scientific papers

  • Aim – a statement that describes the purpose of a piece of research. It does not provide a prediction (e.g. we aim to identify how many mosquitoes carry dengue fever in Laos)
  • Citation/reference – a citation or a reference is another piece of research that the author has quoted in their paper, usually to help back up their argument, or to provide background information to the problem they are addressing. Reference names (Stevens et. al. 2005) or numbers(1) are usually found at the end of the sentence that is quoting it, with all of the full information listed at the end of the paper.
  • DOI number – stands for Digital Object Identifier. It’s a string of numbers, letters and symbols, which can be used to uniquely identify an article or document.
  • Hypothesis – a prediction of what the research might find, which will be tested via the methods used (e.g. we hypothesise/predict that patients will recover more quickly when given this drug). The difference between aims and hypotheses can be found here.
  • Non-significant – if a result is not significant, it means that when they tested their results using a statistical test, the results were just as likely to have occurred at random.
  • Open access – if a paper or journal is open access, it means that you don’t have to pay to read it (yay!). Often symbolised with a little open padlock.
  • Peer-reviewed – before a scientific paper can be published in a journal, it has to be read by other scientists, usually in a related field. Think of it as quality control, to make sure that the paper makes sense, the science is rigorous, and it is written well.
  • Significance – the word significant means something slightly different to scientists. Whilst generally it means something noteworthy, or having particular meaning, in science, if a result is significant, it means that the likelihood of that result occurring would not be due to chance. This website explains statistical significance. The way that scientists identify whether their results are true, or just random, is using statistical tests (an video of the types of statistical tests can be found here).

Types of scientific paper

  • Primary research: this is the most common type of paper and outlines a piece of research that a team have done themselves. It will often include the sections highlighted below (including a background, methods, results and discussion). An example of a primary research article can be found here (sorry, you know I love plague). In this example, the authors have described the problem (background), how they went about answering it (methods), what they found (results) and what these findings mean (discussion). These are often the hardest to read types of papers, because the authors will be discussing ideas and methods that, as a non-scientist, you may not have heard of before.
  • Review: as the name suggests, the author has read and analysed multiple papers in a specific area, then summarised what was found into more of a story. An example of a review can be found here. In this example, the authors have identified a question (how well prepared is Africa for COVID-19) and searched for relevant papers that answer the question. Reviews are easier for non-scientists to read, because they don’t go into such great detail, and provide more of an overview of what the primary papers have said.
  • Systematic review: a more rigorous version of the review (hint: the clue is often in the title!), a systematic review has a very specific process to include or exclude lots of papers, and that process will be outlined in the methods section. This paper is an example of a systematic review.
  • Editorial: These are opinion pieces, usually on a vaguely controversial topic. For instance, if a recent paper has come out and it contradicts previous literature, or a scientist believes that a certain area should be focussed on more. This editorial in the journal Nature is arguing that to improve the world’s laboratory capacity, the patents for CRISPR gene editing technology should be made free. This editorial explains what makes a good editorial (that’s pleasing).  

Most primary and systematic review scientific papers are split into the following sections:

Title

The first thing you’ll read, the title of a scientific paper says a lot about the contents. Sometimes it is posed as the question it’s trying to answer, like this paper ‘Antibiotic usage and stewardship in patients with COVID-19: too much antibiotic in uncharted waters?’, sometimes as a description of the research, such as this paper titled ‘Identification of Risk Factors Associated with Resistant Escherichia coli Isolates from Poultry Farms in the East Coast of Peninsular Malaysia: A Cross Sectional Study’ and sometimes it’s a statement of the results, like this one ‘Increase in Methicillin-Resistant Staphylococcus aureus Acquisition Rate and Change in Pathogen Pattern Associated with an Outbreak of Severe Acute Respiratory Syndrome’. Most of the time, they’re sensibly titled. We scientists do have a sense of humour though, and sometimes we can’t resist making puns. I’ve read these amusing titles to real scientific papers multiple times before and it still makes me snigger. You’re welcome.

Abstract

Usually the second thing you’ll read, abstracts are short overviews of what you’ll find within the paper, often with the key results highlighted. Great for getting an idea of what the paper is about, but doesn’t always tell the whole story, so don’t just read it and assume you’ve understood the paper!

Background/Introduction

This section of a paper explains what we know already about this particular area of science and cites other researchers’ articles, to highlight what we know and also what we don’t. It’s the gaps in our collective knowledge that the paper you are reading is trying to fill with their research. Backgrounds are good places to learn about a certain field, as they provide an up to date overview of what is already known. A good background should have lots of references to other people’s work, as well as key facts and numbers from them. If they don’t, then it’s hard for you as the reader to compare what that paper has found with what is known already (e.g. if the paper you are reading states that in Khartoum, Sudan, 90% of the E. coli that they found in ward patients were resistant to gentamicin, but there is no data in the background about what similar studies found, you (and the author) don’t know whether that’s a high rate of resistance or not. However, if the paper states in the background that a study in Nairobi, Kenya found only 10% resistance, you will know that the 90% found in this paper is probably quite high!

Aims and hypotheses

This section is sometimes found at the end of the background section. This is usually a paragraph or two where the author states what the problem is and what they believe their study might find. The definitions of aims and hypotheses can be found above. An example of this could be:

“our review of the current literature (e.g. what they’re written about in the background) shows that there is a lack of data on the amount of antimicrobial resistance seen in farm animal infections. In this paper, we aim to identify what percentage of infections seen in cows and goats on farms in the Copperbelt region of Zambia. We hypothesise that we will more resistant infections in cows than in goats”.

Methods

This section explains how the researchers did the research to get the results. In theory, the methods should be detailed enough for a scientist in a completely different laboratory to repeat that experiment exactly, including where they got their chemicals from, what volumes they used, and exactly how long they incubated their cells for. Sometimes, the paper might provide a reference to another paper if they followed the exact methods that they used. The authors should also explain what programmes and statistical tests they used to analyse the data.

Results

This section is where the authors display everything that they found out when they did their experiments. It should be pure facts (we found this, this and this), as the author should have left the interpretation of the results to the next section (this is why we think we found this). The data might be displayed in many different ways depending on what it is, it could be a table, or a graph, or maybe a map with different colours. Under each of these will be a description, which should be detailed enough to explain the entire image/table/map etc. and what it’s supposed to show.

Discussion

The discussion is a really important part of the paper as in this section, the author will try to interpret their results and explain why they think they saw what they did:

we found that more mosquitoes were breeding in fresh water than dirty, which is likely to be related to the species of mosquito found in this region

The fact that they know about the species of mosquitoes in the region will be from previous research too:

Arruda et. al. (2013) showed that it is mostly Aedes mosquito species that are found in this region

If their results fit with what previous literature has found, they should back this up with other references:

Our results show similar patterns in mosquito breeding habits to Tembo et. al. (2020)”.  

If their results are different to what was previously found, the author should suggest why this might be (did they collect the mosquitoes at a different time of year? Was it in an urban rather than rural setting?). Just because they’re different doesn’t necessarily mean they are wrong, although often papers will state any limitations that they had too:

these results were limited by the fact that, due to poor weather, samples could only be taken from non-flooded areas

The discussion is where the author tries to justify and explain their results. It’s this section where you will need to think hardest and try to decide whether you think their results are justified by their reasoning.

Conclusion

The conclusion to a scientific paper ties all the sections together in a summary statement and often finishes with a few sentences about how the author thinks their research will impact future research. They might suggest that their findings may help policy makers make better decisions, or that doctors might use this new drug alongside another to treat patients quicker. Often research raises more questions than it answers!

References

The list of references will enable you to find and read any of the previous research that the author has quoted. Often there are web addresses or DOI numbers (see definitions) so you can click on it and be taken straight to that page. Make sure you look a the age of references (when they were published). If all of the author’s references are from 20 years ago, then they’re definitely not up to date on their background information! Some older references are fine, but there should be more newer ones too. Some scientific fields move faster than others though!

Published by Linzy Elton

I'm Dr Linzy Elton and I'm a postdoctoral researcher at the Centre for Clinical Microbiology at UCL. I'm part of the PANDORA-ID-NET consortia I focus on One Health, laboratory capacity development, antimicrobial resistance, whole genome sequencing, tuberculosis and science communication. I'm also the principle investigator for a multi-site study identifying the effects of COVID-19 infection prevention and control measures on hospital acquired infections (Twitter: @AmrCovid). Whilst my background is in parasitology, focusing on helminths and specifically the prevalence and control of schistosomiasis (I completed my MSc on Medical Parasitology at the LSHTM), I completed my PhD on the role of biofilms in the transmission of Yersinia pestis (plague) in fleas and lice. I've worked on research projects in a number of countries, including Egypt, South Africa, Tanzania and Uganda. My personal Twitter account is @LinzyElton

One thought on “Straight from the horse’s mouth: How to read (and hopefully understand) a scientific research paper

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create your website with WordPress.com
Get started
%d bloggers like this: