Writing a scientific manuscript ─ getting started

As scientists we are not all gifted writers, nor is writing a manuscript a particularly enjoyable task compared to other aspects of research where the thrill of discovery drives your career. Very few researchers actually ‘enjoy’ writing a scientific manuscript, partly because of the time involved, and the possibility of rejection by a reviewer who doesn’t consider the months or years of hard work summarized in ~5000 words or less. Sometimes the ‘getting started’ stage can be the most difficult. This article is an overview of the process of writing a manuscript. I will expand on these points in upcoming articles.

In general, you are ready to write your manuscript when you have reached that point in your research that triggers the decision to stop your experiments and start writing. For some research questions that point is fairly self-evident if you have closely followed an a priori hypothesis. Some types of research can go on indefinitely (or as long as you have funding) if the answers to the research question generates more questions.

I have often heard senior researchers comment on the very clever researcher or collaborator who cannot write to save her/his life (or career). Getting into the necessary headspace to write can be the hardest part of research.

Here are some tips, gleaned from over three decades in research, on getting started with that manuscript that you’ve put on hold.

  1. Choose your journal. Create a list of journals that are suitable for your subject area and review their aims and scope. Use published articles that your research builds upon, if relevant, and see where they are published. For more on journal selection, see my earlier blog titled ‘Choosing the right journal – think before you submit’.
  1. Review your results. Thinking about the ‘design’ of your manuscript as you review your results will help you decide what other experiments you need, and what questions potential reviewers may ask. Plan your tables and figures before the write-up stage. These can be anything from flow-charts to graphs or illustrations that can be updated as more data comes in. You will know when the write-up stage is ready when the tables and figures tell a story. ‘A picture tells a thousand words’, so use them to your best advantage.
  1. Start with an outline. Create a blueprint that will guide your manuscript. This can be as detailed or as basic as you wish, and helps to ensure flow and logic. With your journal in mind, lay out the different sections in your document, i.e. Title page, Abstract, Introduction, Methods, Results, Discussion. Include any relevant formatting or word count requirements outlined by the journal. You may even fill in subheadings in the Results section. It is not necessary to list these chronologically, but I have found that it does help with outlining the logical thought processes that guided the work. The ultimate goal should be to tell the story as simply as possible without the day-by-day diary of events that led to the final result. It may in fact be better to exclude experiments that did not substantively contribute to the result. It is a judgment call whether a failed experiment (I don’t mean negative results – more on that in another blog) adds to the current manuscript, or is a distraction from the main finding.
  1. Order of writing. Choose the order according to what inspires you most. Like a jig-saw puzzle, you can start at the top or bottom, left or right. I personally find it helps to write the Methods first, as this should be the least difficult section. Where you have complicated experiments, keeping a good lab book (a requirement of most research institutes) will help avoid frustration and ‘memory loss’ about what you did months (or years) ago. In some situations it may be helpful to write up results bit by bit as you do your experiments or analyses, rather than waiting for the penultimate experiment.
  1. Write now, edit later. Very often my first draft will be full of ‘notes to self’ and ‘stream of consciousness’ text. There will necessarily be several rounds of editing before you are ready to share it with collaborators and co-authors. At the editing stage, aim for clarity and conciseness. Avoid overuse of transition phrases like ‘Next’ or ‘We then showed that’ etc. Any sentence that you need to do a double-take on needs to go! Fragment long sentences. Don’t get caught up with formatting requirements at this stage. That should be the absolute last thing you do prior to submission, along with ensuring you have met the required word count. Remind your co-authors that you need substantive comments back when you send them the first draft — not formatting! 
  1. Keep a close eye on potential plagiarism. Most journals use plagiarism software, as do most reputable research institutions. Invest in plagiarism checks with your institute. Automated software will almost always detect a level of plagiarism. I recently got a report back from a fairly reputable journal, and was somewhat amused that standard phrases in my manuscript, like “associated with an increased risk” and “as a risk factor for” were flagged as plagiarism! Reasonable journals will not ask for changes to every phrase that the software picks up on, but will highlight the ones that require your attention. So at all cost, avoid lifting entire paragraphs or sentences from published manuscripts or online resources. Also, ‘self-plaigarism’ needs to be avoided. If the topic of your paper is one you have published on before, it can be tempting to save time and recycle sections that you legitimately wrote in a previous publication. This should be avoided because regardless of who originally crafted the words, the software will pick this up as legitimate plaigarism. 
  1. The nitty-gritty.
    1.  The title can be written at any stage of the process, and crafting one is not a minor element of the paper. Avoid long drawn out titles. Some journals put character limits on this. As a general guide, titles that run into 2-3 lines are too long. Try to find the phrase that ‘sells’ your work and encourages the reader to want to know more. Also, shorter titles that are easy to understand and informative are more likely to reach a wider audience and shared on social media than those with scientific jargon (reported in Journal of Clinical Epidemiology 2017). More on crafting a title can be found in my blog titled “It matters how your write”.
    2. The Abstract may be one of the last sections you write. Word count limit forces you to again focus on the main message. Where there is a lot of information in the Results, prioritizing what to include in the Abstract can be daunting. Work your way back to the a priori hypothesis and the main result. Remember to include key words and search terms in your Abstract that may be used by other researchers for Pubmed or Google Scholar searches.
    3. The Results section is what reviewers will focus on. Take the time to properly format your tables and figures, and avoid sloppiness. As a reviewer, I have seen manuscripts with tables that look like they were just pasted into a Word document from Excel without any attempt to reformat them. The same applies to images. Be pedantic about data presentation. You run the risk of a reviewer expecting the rest of the paper to be sloppy if they cannot make sense of your data.

    Planning for a manuscript takes time and commitment. My best advice to doctoral students and early career researchers is try to start writing early on in your career, as there is no avoiding it. Pay attention to the writing styles of your mentors, and be open to their advice, edits and comments on your draft. Good mentors will also provide guidance in developing writing skills. Allocate time at the end of each day to actually write up your results as your research progresses. Writing skills take practice, so start early. 

    Finally, celebrate your achievements!! Acceptance of a manuscript for publication are major personal milestones worth celebrating with your colleagues. So even if it is one small step in the advancement of your career, make it memorable!

    At SugarApple Communications we are experienced in all stages of the publication process, and can advise on data presentation and analysis. If you are time-poor and need to get that manuscript out, get in touch today and let us talk.

    ← Back to All Articles

    Unfog the science…ensure quality, clarity and accuracy.

    CONTACT US