Simple considerations when writing an article can make your work more accessible to a broader readership, finds Matthew Edmonds.
A recent Universities UK report found that 37% of UK-authored articles are freely available immediately, compared with 25% globally, and most UK articles can be accessed 12 months after publication through open access green and gold schemes. Around half of research articles searched for online are open access, and many funders now require research to be published in an open access journal.
It is difficult to measure how many people actually read a paper. However, a recent analysis did suggest that the proportion of articles that are never cited is gradually falling. Although there are likely other factors at play for citation rates, the bottom line is that primary research is becoming ever more available. Now, your research is not just picked up by others in your immediate field and your close colleagues, but by anyone who wants to read it. Changing your papers to serve new audiences is essential.
A change of perspective
When I worked in academia, I tended to only think about other researchers in the field when writing up my research. When I read an article, I would focus on the overall message – what was new – and technical details if they were relevant. I would be principally interested in the impact on my own work. This directly fed into the type of information I presented and the style in which I wrote.
Recently, I changed careers to medical writing. One of many new challenges I’ve faced is preparing systematic reviews. These use a pre-defined search and assessment protocol to focus on specific questions. An article may be identified for inclusion by these searches on the merit of only a small piece of information that isn’t necessarily its main focus.
Let’s say, for example, that your latest paper starts by comparing the effects of a panel of drugs on the growth of tumours. Naturally, the drugs that slowed or even reversed growth were the focus of the rest of your paper. But if my systematic review is addressing the efficacy of one of the “uninteresting” drugs, your paper would be cited only for a small part of its first experiment. Data that aren’t relevant to your story are really relevant to mine.
This approach has completely changed how I look at research papers. I check the methods section only for rigour of statistical analysis, appropriate controls, and ethics. In the discussion, only conclusions from the data are relevant; any conclusions or theories from the authors will be ignored.
When it comes to data, I consider only specific points, rather than the dataset as a whole. This can be frustrating: even in this age of open data and supplementary files I find many articles still present data in summary figures without actual numerical values.
I can understand why researchers don’t include specific data points: data has to be standardised for analysis in a way it wasn’t necessarily recorded in the lab. It can be fiddly work, and publishers are more interested in the message rather than reams and reams of collected datasets.
Besides, from an author’s perspective, the wider conclusions are what are important for stimulating debate in the field and finding new avenues of research. For me, though, science is the collection of data, and this should also be at the heart of published science. It surprises me that so many articles do not present the actual data gathered, but hide it behind overall trends and conclusions.
It can’t just be me who needs that data to get their own research done — how much more science is being held back because not all of the data is being published?
How can you write up your research so it is useful to a wider audience?
Here are a few things I feel would help published research go further. I wish I had this list when I was writing research articles:
- Write for an academic audience, but also be aware other non-specialists may be reading. Make sure conclusions from the data are clear, rather than just discussing research questions and theories.
- Summary charts are useful, but also include the actual numbers somewhere so others can also analyse them.
- Make sure your statistical methods are clearly outlined, even if they are standard methods in the field and seem obvious to you. This only needs to be a sentence or two in the methods section or figure legend.
- Perform a statistical analysis on all your data, where reasonable. Many papers only statistically compare groups of direct interest to the authors, despite having many other related groups in the same dataset. This doesn’t help someone reviewing evidence for a slightly different question.
- Define all abbreviations, even in conference abstracts. It may be that everyone in your field knows that AP sites are a type of damage in DNA, but that doesn’t mean anyone else does.
Why should you care?
With so many pressures on your time, you may wonder if it’s really worth the effort to change the way you write up your research. The short answer is “yes” — with more people than ever before able to access your work, you never know who will be reading it. The next job, grant or opportunity could come your way because you present your research with clarity and transparency.
But the best reason to make your results fully accessible is that your complete data will be available for researchers conducting meta-analyses and reviews of the literature. Then it is fully contributing to the body of scientific knowledge — as all published research should.
I’m sure those who read research articles for other purposes have their own wish list. Please feel free to share any tips of your own.
Matthew Edmonds is a Medical Writer at Insight Medical Writing in Oxford, UK. With a background in cellular and molecular biology and cancer research, he now writes about drugs and devices from all areas of medicine. He enjoys communicating and discussing science and the issues surrounding it. You can follow him at @benchmatt.