How to write a scientific paper in 7 easy steps

With the start of classes this last week, I’ve been meeting with lots of students at various levels about how to write a scientific paper.  Even after writing a number of scientific papers myself, I tremble at the thought of giving advice, but since I learned through trial and error, I thought I would put down some key principles of writing a scientific paper down in the hopes that others don’t have to go through the long convoluted path that I had to get a paper published.

Step 1)  What the heck is a scientific paper anyway?

Scientific papers are not an essay, book report, review, news story, executive summary or blog post. Scientific papers are the summaries of collected data accompanied by a detailed interpretation of that data. They are vetted through a rigorous review process and must be of broad interest, but provide new information. They are also known as peer-reviewed papers, since the published article is reviewed by a panel of experts. Most news stories only need to get past the editor; scientific papers have to get through a panel of experts and the editor of the journal.

A literature review of a particular topic is not a scientific paper, nor is a personal essay of your thoughts on a topic. As students we are programed to write literature reviews and summaries. My entire undergraduate college degree was writing long summaries of books and chapters. It is so easy to fall into this as you research your scientific paper. It is very hard to not suppress a desire to see all those words and citations to impossible to find articles in print, but scientific papers are not reviews.

Don’t fear putting all that hard work outside of the formal manuscript that you are working on. Think of it as the pre-paper, necessary to write, but all in preparation for the real scientific paper that will come later.

Step 2) Get data

Do the experiment! Collect the fossils! Make the observations! Go and get data. This is what the reviewers are going to look at in detail. No matter how fantastic your writing skills, without good hard evidence, the interpretations you bring will not be believed. This part of the progress can take years and years to collect the data to support a scientific paper.

Step 3) Interpret the data

Interpret the data you collected. Does it reveal something new? Do you need more data, repeat step 2. Keep collecting data until you feel confident.  Share your data with people you respect and get feedback from them.  Make charts, tables, graphs, and photographs of your data. Treat your data well, and give it the full treatment. At this point it is worth working with trusted colleagues.

Step 4) Choose the journal

Today there are thousands of journals that you can submit a manuscript to for consideration, and most of them are very specialized in what they are looking for.  Scientific journals were developed out of scientific societies, these initially were presented as lectures or letters read before a public audience. Charles Darwin and Alfred Russel Wallace’s early ideas about evolution were presented as letters to the Linnean Society of London. These lectures, letters and demonstrations were published by the society since they represented an endorsement of its scientific members of the society. In fact most scientific journals are edited and managed by a society of member scientists. Scientists don’t receive any money for a publication, and the copyright is often transferred over to the society, which manages the dissemination of the journal. Libraries would purchase copies, while members of the scientific society would pay dues and receive a copy with their membership.

Things have suddenly changed with the advent of the internet, and the growth in specialized scientific journals. Libraries can no longer afford to have so many subscriptions, while scientists following the literature could not pay for membership is so many new societies. Many journals were taken over by large publishing companies, and bundled for libraries to subscribe to in bulk. This gave a distinct advantage to scientists in academia, since they had access to libraries with subscriptions to these journals, while scientists outside of academia had limited access to these journals. If you have a fondness for dinosaurs to follow the latest research you would need to subscribe or be a member to over fifty organizations and societies to receive all the journals that might publish an article about dinosaurs. For scientific journalist this is extremely expensive if you are following a broad field of science.

Authors also wanted to have a larger platform to share their scientific discoveries, so open access journals were born. Open access scientific articles differ fundamentally from traditional science journals in the fact that the author pays for the article to be published. The journal makes is money off the author rather than the reader. This also may limit where you wish to publish your work. If you want to have a wide readership, you may come up with the funds to submit to an open access journal, or if you want to have the article read by experts in a particular field of science, a traditional journal might be better for you. Beware of the rapid growth of open access journals, many of which are rather dubious and prey on authors, since all you need to publish online is a website.

Here is my suggestion for finding the right science journal to submit too. Read articles in many different journals in your field of study. Which journals feature the best articles? Which journals feature articles most similar to the research you are conducting? Don’t worry about impact factors or other metrics. Most cited papers are often in the most obscure journals, while papers in higher impact journals may not get as many citations. Citations are related to how active the field of research is, so instead select a journal that you personally respect.

Once you have settled on which journal you are going to submit to, visit their website, and learn their formatting for manuscripts and how you go about submitting a journal article. This will guide you how you are going to write your manuscript.

Note that many journals have different lengths of turn-around times for publishing.  Some journals have hundreds of manuscripts backlogged, others are actively looking for submissions, some might rely on volunteer editorial staff that conducts summer field research, while others are eager to see your manuscript in print as soon as possible.  Be aware that the editor of the journal will be the first to review your manuscript, as well as a copy writer. So follow the rules of the journal closely, and feel free to contact the editor to see if your research would be of interest to the journal beforehand.

Step 5) Writing

At this point you should have an outline of what you are going to write, with the journals headings and sections, all your raw data, as well as figures, graphs, tables, and charts that you would like to include in the article. On top of this you have conducted a thorough review of the literature and previous research.

As you write, pay attention to the flow of the paper and try to minimize scientific jargon. Don’t use an abundance of acronyms, and focus on the structure of your sentences. Sentences should move your ideas from one point to the next point. Avoid spelling errors, bad grammar and inconsistent formatting. Once you have written your article show it to those people you trust will give it a good reading. When I was starting out I had a number of professors who would read a paper, and offer a handful of comments. You want someone to read it so thoroughly that they notice the difference between a comma, and a period.

Next step is optional, but it is a great exercise to improve your paper. Save your manuscript, and open a new blank document. Now write and try to reconstruct the article from memory.

What? This is a wonderful exercise. It really helps to improve the flow of the writing. When you are first writing your paper you are doing it slowly and deliberately, without taking in the major points. You are focusing on detail after detail after detail after detail. When you start writing from scratch you remember the major points of the paper, these ideas float to the top and get repeated, while the details that you don’t easily recall get left out of the re-written version. This also helps you generate two different versions. If you don’t like this re-written version, you can just submit the original. However, I find that I like the flow of the second version because I wrote it from memory, and much more rapidly than the original paper. I can also pick and change various sections of each manuscript to make one that is best.

Students spend so much time writing the first manuscript that they can’t let it go. I remember hating having some section of my writing edited or removed, because I spent so much time working on it. But often the best improvements in a manuscript are made through brevity and clarity, rather than a pontification of ideas that may not relate to the major point you are trying to make. For science this is hard to do, because we tend to focus on the details within the data.

Step 6) References

Now it is time to go back and make sure you include all references you cited. I find that it is really important to do a literature search for recent articles just before you submit your paper. If a new paper just came out as you were writing, it will be the first paper that your manuscript will be compare too when it goes to review. Be sure to cite papers of scientists that are likely to review your paper, also cite papers that are authoritative in the field. Cite many papers, especially if they are related to your article. Make sure that all your papers are cited both in the text of your manuscript, as well as listed at the end of your article in the format of the journal. I will create a folder with pdf documents of all the papers I cite in a manuscript, so I can quickly refer to an article if a question comes up.

Step 7)

Now it is time to submit. Make sure your manuscript is formatted correctly, your figures are saved as separate documents, and that you have followed all the rules of the journal. Many journals ask for possible reviewers for your article or a subject editor to review the submission. I highly recommend that you list people to review your manuscript. These are people you know will do a good job reviewing the article, and provide timely feedback. If you don’t know anyone, ask your advisor or professors for suggestions.

Once you have submitted your manuscript, then it is just waiting to see what the reviews think. Don’t be blinded by what reviewers say. Great writers know how to take comments and use them to improve a manuscript. My early manuscripts underwent major revisions, and each one improved the paper. I’m very happy that my early drafts were never published. So stay the course until it is published!