This month I published my 60th journal article. It took me 9 years to get here. A lot of rejections and lessons. And, importantly, I finally developed a writing system that could spare you from desk rejections.
I will be sharing it with you bit by bit.
I use this writing system every time. And it helps ensure that my manuscript is clear and concise. Start with the Results I learned this from my PhD supervisor. He always said, while you're waiting on the next results, you should do data analysis and visualization for the results you have. In the end, by the time I am thinking of writing a manuscript I already have figures, tables, charts, and annotations. With my data visuals in place, I create a storyboard.
What do I mean? I pick out the visuals that show the key results, and then arrange them into a story. The results that are important but not key, I throw them into the supplementary materials. Each visual answers a research question. But my policy is a manuscript can only have 7 visuals. I always try to have at least three research questions. So, that means they can be two or three visuals answering the same question. Now that I have selected the principal results and arranged them properly to address specific research question, the next step is writing the Results section.
And this is how I do it.
1. Question - I write the question that I am interested in. This is normally a paraphrase of what I will later write in the Introduction. I use phrases like, "We investigated the effect of XYZ on the ABC." At the end of that sentence I cite the data visual that answers that Question.
2. Experimental approach - I then write the experimental approach or research method I used to answer that question. To save space, I always combine the question and experimental approach into one sentence, where possible e.g., "The effect of XYZ on ABC was determined using 123."
3. Principal results - always start with the most important results. This is the top tip I received from my postdoc supervisor. Don't repeat what's on your visuals and avoid data dumping. That's a tip I received from more than 200 peer reviewers, who have gone over my manuscripts over the years. How then do you write the principal results? Focus on trends, patterns, and sometimes outliers. But don't use superlatives to describe these - no was higher or lower etc use exact values. Don't use significant(ly) without corresponding statistical values.
4. Answer - this is the wrapping up part where you just show how your principal results answered the research question you stated in the introductory sentence. It's always smart to write your answer in a manner that connects with the question asked in the next set of results.
Repeat this for all the principal set of data visuals you have. I guess you can agree, it's possible to write the results section in one day even if you're a slow writer. More in future posts.
University of Northumbria, UK