Scientific Paper Publishing

Type				Team Size
Me, Myself and I	Group Collaboration	The Academic System	Software	1	2-10	11-30	30+

In short: This entry provides guidance for anyone who's planning or already engaging in the publishing of scientific research articles and describes the process from an initial idea to publication.

Writing scientific papers is to this day the main currency for most scientific communities. No matter how much you try to diversify your career, this will be the one component that you will be evaluated by. What you write about is one component of the research that is most closely associated with your profile and how you are seen by others as a researcher. The craft of writing is an altogether different topic, which we shall deal with in another entry. Here, we focus on the mechanics of the process of paper publishing. How do you approach this process? What do you need to know to publish your paper? Which steps are involved to get your paper published?

Starting a paper

All texts start with one single idea. Researchers typically recognize a gap, something where they want to contribute, an idea that is nagging them, or an urge to overcome a gap in our knowledge. The question is then how to get from this lofty and rather unspecific idea towards the successfully published paper. The first step is to get a draft of your idea, which can be an outline, a bullet point draft, or other forms of crude overviews. Any larger idea is usually approached through a grant application, because grants claim funding, funding pays people, and people do research. Hence does any larger endeavor in research not start with writing a paper, but typically with writing a grant. Tenured researchers may start with an initial paper before channelling the larger idea through a grant, and smaller ideas can also be pursued by a bachelor or a master thesis. A first step beyond the initial idea is to know who needs to be involved, who covers which expertise, and how the different responsibilities are divided. While in some papers there may be only one author, multi-coauthored papers are more common these days. Some disciplines such as economics mark a deviation from this, as they often value single-author papers much higher, and demand from PhD students to follow this norm. Be that as it may, most branches of science rely on multi-coauthored papers, usually way beyond the two that make up a student and a supervisor. Yet it is this combination that is the most crucial one in the beginning of a scientific career, because publishing papers is a craft that is learned from experience. In addition to these two authors there is often a second supervisor, ideally with a slightly different focus and set of experiences that are best complementary to the first supervisor. In addition, there may be more co-authors who either did parts of the empirical work or amended the list of authors through other facets of expertise. At this early stage, it is ideal to clarify the different roles, demands and expectations. As a rule of thumb, it can be quite clear that most papers are usually more work than one could anticipate before, hence be conservative in your estimate. A kickoff meeting can be helpful, and discussing the outline is a precaution that is unavoidable at this point. It is vital to write a general account of the overall plan to generate the paper early on.

Doing the actual work is then usually more straightforward, with the caveat that typically most work rests with early career scientists, whereas more experienced scientists do less legwork - they contribute exactly through their experience. Here, students can score by working hard in order to make their work stand out in comparison to that of other students, thus making their career gain an edge that allows them to rise above the ranks of their fellow students. An ideal supervisor will also lead students in a way that maximizes their learning curve despite often deep frustration and a demand for stamina. Any experienced researcher knows this learning curve and will try to honour this in the education of future researchers.

Once the direction and general aim of the paper are clarified, it might be already wise to consider potential target journals. Within normal sciences this can often be anticipated early on, yet also widely depends on the impact of the data and results, and thus may be decided only after the empirical results are roughly clear. It can be helpful to have a shortlist of potential journals ready to communicate to the co-authors. In case of a rejection of your first choice journal, the second best fitting journal can targeted. This is again the point where a good supervisor can give valuable advice to young researchers, as supervisors know the publication landscape more coherently. Depending on the general direction, length and framing of the paper can be already anticipated. Usual introductions are no longer than 20 % of the text. Theoretical foundations are common in some papers but should not be overly wordy and rather concise. The methods section and results depend much on the empirical work in terms of their lengths, yet the discussion should not be too short, making a solid quarter of most papers. Policy recommendations, limitations and other specific sections are relevant for some journals, yet certainly not for all. There is a specific norm when it comes to figures and tables, not only how these are used, but also how they are compiled. Be inspired by articles most important to your work, and try to follow them closely but not dogmatically.

Ask your co-authors for feedback. While much of the work relies on the first author, the last author should give general advice, and it has proven valuable to have a sparring partner who can give more direct feedback to your specific questions. Make sure to be interactive, but well-prepared and structured. Whenever asking your co-authors questions make sure that you anticipate how they can answer. Especially questions with obvious or trivial answers do not let you appear in the limelight you may want to reflect. On the other extreme, you may be bold, but not too much. Science is, despite its general aim, quite conservative in its norms, and you should follow these norms. They have proven valuable in the past, and are shifting slowly. Hence do not attack something head on that you hardly understand yet and will anyway only be able to change in the long run.

If you know one or several journals to which you want to submit, make sure to read the instructions for authors from the journal. Following these closely is of vital importance. To this end, see that you have a reference manager, and use it. Make your manuscript appear in writing, citing, and figures/tables as closely resemble the articles you admire.

It is always better to have a full manuscript that is imperfect as compared to parts of a manuscript that are lacking the context of the remaining parts. Try to finish whatever you can before passing anything to your co-authors. There may be several rounds of revisions, yet submission should only be approached when everybody involved agrees to the version. I think that depending on the relation of the authors about 2-3 revisions seem to be agreeable. It may be more, yet hardly ever fewer revisions to get an agreeable version that can be passed to the hands of peer-review.

One last point before submission is the publication fee. Many beginners believe that scientists are paid for articles. This is not true. Scientists are ideally paid by an institution to publish, yet almost all journals demand a publication fee, at least to make the article available free of charge, which is part of a new movement called Open Access. This fee can be up to several thousand euros. Some institutions have that covered, yet it is not always clear and needs to be clarified beforehand, as no one should pay this amount of money privately. Early career researchers need to discuss this with their supervisors. Yet it is important that the final paper is accessible to the people out in the world.

Submitting a journal article

Peer-review is one of the pillars of modern science, and for better or worse is essential for most researchers to master the art of publishing papers. This includes learning how to submit articles to a journal, which is a small art in itself. Preparing an article first and then going through the often cumbersome act of submitting your manuscript to a scientific journal involves many diverse steps, which I will try to outline in this text.

The first step ist to decide which journal to submit your article to. Most journals have rather similar demands in terms of formatting, length and other stylistic details. While the journals Science and Nature demand an altogether different structure and writing of your article, most journals build on a structure that is more or less the same. Aim for a journal that maximizes your readership. A good rule of thumb are the journals that are most cited in your manuscript. More experienced co-authors can usually say which potential journals are out there.

Whatever your choice, you will have to save your text in a Word file, and have to make sure that all co-authors agree with this last version that you want to submit. Ideally communicate your final timeline to all co-authors weeks in advance, and make sure that everybody had the possibility to give their green light to the article being submitted. Some journals seem to demand that you slice your article into many different files. Do not do that. Instead save everything in one word file, and make sure that it contains the standard formatting demands in terms of font size, line spacing etc. I would say that most journals will not reject your manuscript outright if it does not match the cumbersome format they often demand as written on their homepage. Otherwise they will write to you and ask for an updated version. Just make sure that your formatting is as close as possible to their demands without the hassle to go through every little detail. I for once try to match their citation style, yet I also know that an AI will technically sort this out later.

In addition to the manuscript as a docx file, you also need to write a submission letter. This is ideally a one-pager that contains a very brief summary of your main findings, and why it would be a good match with the journal you submit to. Be humble, but also firm in how you sell your contribution. Ideally try to directly address the editor regarding the specific focus of your results that may make the article appealing to their readership, yet make sure to not ramble on for pages. After all, this is only a teaser to your article, and helps you to get a foot in the door.

Article type selection

First, you need to know which category your article falls into in the respective journal. These categories got more and more cluttered over the last few years, and often it is a mystery which category works best. Generally, there are empirical papers, review papers, conceptual papers and comments. Yet it seems that out of a thriving towards diversity more and more categories emerged over the last few years, and it is best to check which article types are published in the journal, and read some of these to find your best fit. Remember that if you are in the wrong category, then you can still change this later on.

Co-authors

At one point or the other, you need to list all co-authors. This can be a really cumbersome task if it is any, also because you may have to list at least their e-mail addresses, but often also their postal address and information such as their ORCID, which is a digital number identifying individual researchers in the publication system. What is more, you may need to submit the contributions of all co-authors. Make sure to be coherent and discuss this with co-authors, if need be. Otherwise tensions may arise. Therefore, it can be important to already log into the submission page weeks before the actual submission in order to discuss these things early on. Remember that the article is only submitted at the very end, so why not look around already a bit earlier?

Attach files

During the submission you will have to upload your manuscript and the submission letter, and in some cases other individual files such as larger figures of the online appendix. Make sure to upload them one by one, and best name them according to the journal’s standards, which is not hugely important, but helpful for the journal in case of doubt. Sometimes this uploading is a bit buggy, which is why Firefox may be the better choice as compared to Safari as a browser.

General information

There is a broad array of questions that journals may ask you, for instance regarding funding or ethical concerns, which fall into a category of general or additional information. All this is testimony that at some point somebody had a great idea, and now everybody needs to think about this. Hence, be transparent, but also be pragmatic.

Reviewers

Often one has to propose reviewers that may be potentially reviewing your manuscript. I always considered this a bit fishy, because it violates independence of the peer- review process. Of course you cannot suggest reviewers that are biased, i.e. reviewers that you have published with or that are in the same institute as you. Check out the ethical guidelines of your university or funding bodies in case of doubt, yet best propose people that are critical towards your research, yet are also constructive. Maybe opt for people that inspired you.

Data availability

Most journals include a data availability option, which indicates whether you want to upload your data. While this can be a huge work, it is of course usually worth the effort. Some journals require the publication of all data part of your research. I am skeptical when research data contains context information outside of the form provided by the journal. Special information and diverse data formats should be made available. A precondition for those datasets to be published is a corresponding contextualization and yet the misuse of those datasets cannot be ruled out.

Keywords

Sometimes you have to provide keywords, which is a nice old relic out of the time of punchcards. These keywords are therefore less important today, and can be evaluated as of minor importance.

Generating your PDF

Most journals generate a pdf out of all your files and information that you have to approve. Glance through the PDF and see that everything is in its right place and formatting. Do not read the whole document word by word. Otherwise you will find mistakes and have to upload it again and again. This will possibly never end. Hence inspect the PDF, but leave something for the reviewers. You probably read it already quite some times, hence why bother now once more. The last step is then to approve the PDF and make sure that the article is finally submitted to the journal. Celebrate your submission and make a note in the calendar. Depending on the journal you may start to wonder in three months what happened to the article. If you want to drive yourself semi-nuts you can also follow the status updates on the journal page, yet this gives usually only little information away. Hence test yourself in patience. Enjoy.

Revisions

After a few months the moment arrives that you usually receive feedback from the journal for your article. If it was not a desk reject, and these are not rare, you may have at least some helpful comments from competent reviewers, and you may even get the chance for a resubmission. Remember that only a small portion of articles is accepted in the journal where they were submitted, and this depends very much on your ambition. If you aim high, you are rejected more often. Yet once you have the chance to resubmit, you need to write a point by point reply. Every suggestion made by reviewers and editors merits a reply on your end, and ideally a change within the manuscript for the better.

---

The author of this entry is Henrik von Wehrden.