A Guru Appraises Systematic Reviews
Before any scientist starts writing a research article, he or she is responsible for interpreting results and figuring out how the data advances knowledge in the field. This list of key do’s and don’ts can shed some light on navigating the evolving publication life cycle — and help you learn from others’ mistakes.
Every scientific article tells two stories. One story is neatly divided into an Introduction, Materials and Methods, Results, Discussion, and Conclusion Sections. A research article ties together a proven/disproven hypothesis, illustrated with figures and tables. The other story is one of sweat equity involving long hours of frequent disappointments. Then there is the unspoken edict hanging over every budding scientific career: “publish or perish.”
Today, the life cycle of this article will involve a pre-publication phase, publication of the manuscript and accompanying visuals (usually 2 or 3 slides to highlight the takeaways), and a post-publication phase with social media and citing tools to promote the article. A scientist must consider the benefits of accelerating a research article through this life cycle against quality tradeoffs. Moreover, it may take years before one has a “eureka” moment and then the relevant experiments may be summarized in only two articles.
Walter Sutton (April 5, 1877 – November 10, 1916), widely viewed as the first proponent of an association between Mendelian genetics and chromosome behavior, made a lasting contribution to his field through the publication of only a few articles. Can you imagine what would happen to a scientist today if he/she published only one or two articles? What modern funding organization would take the time to wade through the twenty-plus years of labor that went into Darwin’s “The Origin of the Species?”
To earn the career-defining currency of a peer-reviewed publication, an aspiring scientist must mimic Sutton and Darwin by picking a good scientific problem. Ideally, this problem will be addressed within the review cycle of a grant funding the individual or the laboratory. At the very least, public expectation is that the article should hit the presses shortly after the outbreak of an epidemic.
Regardless of the challenges, no one will bother reading your amazing discoveries if you don’t follow basic academic writing rules.
Tips and Pet Peeves
Make your data readable
“The P-value was never intended to be a substitute for scientific reasoning.”
Ron Wasserstein – Executive Director (American Statistical Association)
Do assemble relevant essential data (additional information can go into a supplementary section published as a stand-alone document online), controls, and accompanying statistics into easy-to-interpret formats for your peers.
Don’t omit explanations for non-standard abbreviations/abbreviations, crowd figures/tables/images with too much information (they are not meant to be eye charts), forget to include the number of times each experiment was performed, or over-emphasize the P-value. For instance, a P-value does not tell you to what extent a treatment has worked. That would have to with the effect size.
It would be great if the data was directly in line with an original hypothesis. However, data often meanders in a different direction and may lead to a discovery that was not initially envisioned by the researcher. Once data is understood and interpreted within its proper context, assembly of the remaining parts of the article becomes an easier task.
Tell your Data Story Well
Do provide a discussion based on relevant references that corresponds to the results, discuss strengths and limitations without the sudden introduction of new ideas, consider if there is another way to interpret your data, discuss what else could be done to address questions arising from your research in future studies, be descriptive (without being repetitive) in compiling the methods section, tell the reader how your study will advance the field in a short conclusion, and “hook” the reader with a concise title/abstract/introduction that conveys why they should be reading your article in the first place.
Don’t bore the reader by discussing the “data you wish you had,” group discussions of many figures in run-on sentences, speculate on interpretations of your data that are not rooted in facts, treat your introduction like a systematic review of all the literature in your field, and use the conclusion section as an opportunity to repeat the discussion.
Do use reputable editing services if English is not your native language.
Don’t forget to read your galley proofs. Trust, but verify. Editors are fallible human beings and mistakes happen. The meaning of your great discovery will be lost in translation if figures/tables are jumbled, equations are missing, sources are not properly cited, or your peers fail to understand what you are trying to say.