“How To Write and Publish A Scientific Paper”


October 12, 2014

“A scientific experiment, no matter how spectacular the results, is not completed until the results are published. In fact, the cornerstone of the philosophy of science is based on the fundamental assumption that original research must be published; only thus can new scientific knowledge be authenticated and then added to the existing database that we call scientific knowledge.”

That’s how Robert A. Day emphasizes the importance of writing about research results and publishing these in scientific papers in the preface of his book “How To Write and Publish A Scientific Paper”. Publishing is fundamentally important for science and research, but is almost equally important for scholars’ research careers since future collaborations, jobs and funding depend more and more on publishing records.

Day’s book covers both the overall aspects of publishing – what is scientific publishing? Why is it needed? – and the specific steps involved in writing and publishing a scientific paper as well as tips for each of these steps. The book is an interesting read, and I recommend it to anyone involved in research and scientific publishing.

For an overview of selected parts of the book – and, hopefully, as a teaser for reading the entire book – read the rest of this blog post. I’ll provide my own views on the process of the scientific paper, based on reading others’ papers and writing a handful of them myself, and combine this with quotes from the book.

The points I’ll cover are the following:

Authors of the paper

Deciding who should appear as authors of a scientific paper is important since authorships de facto are the “currency”, in which a scientist’s contributions are measured. However, authorship could and should only be claimed by those

“…who actively contributed to the overall design and execution of the experiments.”

Similarly:

“Colleagues or supervisors should neither ask to have their names on manuscripts nor allow their names to be put on manuscripts reporting research with which they themselves have not been intimately involved.”

Let me give an example of how I think this should work: In collaboration with my three Ph.D. supervisors, I have been discussing ongoing work at our joint supervisor meetings, while all details of this work and some additional discussions have been carried out together with only two of these supervisors. Earlier this year, the work was finished and written up in a journal manuscript, and the third supervisor – who had participated in the overall discussions, but not in any of the specific work leading to the manuscript – chose not be an author of the paper, exactly because he hadn’t been “intimately involved” in this work and didn’t take

“…intellectual responsibility for the research results being reported.”

On the other hand, long author lists occasionally appear with some members having made only minor contributions, which, however, shouldn’t happen for good scientists:

“…good scientists do not allow dilution of their own work by adding other people’s names for their minuscule contributions, nor do they want their own names sullied by addition of the names of a whole herd of lightweights.”


“Piled Higher and Deeper” by Jorge Cham www.phdcomics.com

How to Prepare the Title

As pointed out by Robert A. Day, the title of a scientific paper is guaranteed to be read by thousands of people since it will appear in journal tables of contents as well as in numerous online literature databases – where potential readers will spend a brief moment to read the title and decide if they want to know more. Therefore, each word in the title should be chosen with great care!

Firstly, titles shouldn’t be too long, which can be avoided by avoiding “waste” words:

“Without question, most excessively long titles contain “waste” words. Often, these waste words appear right at the start of the title, words such as “Studies on,” “Investigations on,” and “Observations on.” An opening An, or The is also a “waste” word. Certainly, such words are useless for indexing purposes.”

However, titles can also be too short:

“…most titles that are too short are too short because they include general rather than specific terms.”

Instead of general terms, use specific words that describe precisely, for example, which structure or type of material or species that is analyzed in the paper, and that specify as accurately as possible, which methodologies are used for this analysis.

In my own work, I expect to know from the title whether results are obtained from experiments, theory, computations or a combination of these. If the work is purely theoretical or computational, a specification of the theoretical approach or computational method(s) in the paper also adds value to the title; “Theoretical approach to…” or “Computational approach to…” are too general, while “Three-dimensional integral equation approach to…” accurately specifies the methodology.


“Piled Higher and Deeper” by Jorge Cham www.phdcomics.com

How to Prepare the Abstract

If a potential reader finds your title interesting, she/he will likely read your abstract next before deciding whether to read the entire paper (or parts of it) or not. Therefore, obviously, the abstract should, as clearly and concisely as possible, tell what is done and shown in the paper and what its results are.

However, one should avoid including too many details in the abstract:

“Occasionally, a scientist omits something important from the Abstract. By far the most common fault, however, is the inclusion of extraneous detail.”

Similarly, repeating half of the introduction – where the present work in detail is motivated and put into a context of other results and papers – will make the abstract long and not straight to the point on what the reader can expect from the paper.

A lot of details are important for the work presented in the paper, but not all of them are central in the overall picture, and most details should therefore be left out of the abstract. Choose two, three or four main points on methodology, structures and results and present them in four to six sentences – that is your abstract.


“Piled Higher and Deeper” by Jorge Cham www.phdcomics.com

How to Write the Introduction

Robert A. Day gives a cookbook recipe for writing the introduction of a scientific paper:

“Suggested rules for a good Introduction are as follows: (1) The Introduction should present first, with all possible clarity, the nature and scope of the problem investigated. (2) It should review the pertinent literature to orient the reader. (3) It should state the method of the investigation. If deemed necessary, the reasons for the choice of a particular method should be stated. (4) It should state the principal results of the investigation. (5) It should state the principal conclusion(s) suggested by the results.”

Including the above points will, in principle, lead to an introduction that has all the important components. However, in practice it may be hard to actually write, especially (2) for which you need to know the literature in great detail to position your own work and results – and to succeed in convincing the editor and peer-reviewers that your paper is worth publishing!

Personally, I have found that tagging my literature with suitable tags and categories has made writing this part a lot easier. The last time I had to write the introduction, for a manuscript on “quasi-normal modes”, I looked under the tag “quasinormal” in my literature database and started building the introduction around the papers in there – which naturally lead me to explain in what ways our work was new and different from previous papers.

Another important part of the introduction is to tell what you do and what you find – don’t only, as Robert A. Day points out, save this for the end of the manuscript:

“Reading a scientific article isn’t the same as reading a detective story. We want to know from the start that the butler did it.”

How to Write the Materials and Methods Section

Describing your methods precisely and in all details is important:

“Be precise. Methods are similar to cookbook recipes. If a reaction mixture was heated, give the temperature. Questions such as “how” and “how much” should be precisely answered by the author and not left for the reviewer or the reader to puzzle over.”

This is obvious, but nevertheless I regularly read papers that are not precise and don’t give all details – for example, quantities are introduced in the text, equations or figures without proper definition. This leaves the reader in a state of “how” and “what”, which moves the attention away from the important points of the paper. Therefore: Be precise and give all details.

Likewise, to save space, refer to other papers where previously published methods are used. In some cases, it may be useful to reproduce central parts of previously published methods, but in that case state this very clearly to avoid confusion on which parts of your manuscript are new and which aren’t.

How to Write the Results

In the results section, representative and well-chosen results should be presented:

“Most importantly, in the manuscript you should present representative data rather than endlessly repetitive data…”

or put more bluntly:

“The fool collects facts; the wise man selects them.”

In my most recent paper, I had in an initial draft included four figures that in many ways conveyed the same information – and it was therefore natural to remove two of these to avoid redundancy. The four figures together constituted a collection of facts; the two that remained a selection of facts.

On top of choosing a selection of figures, make sure these figures are informative and illustrative; include axis labels, legends for data in the figure and an inset with the structure or measurement setup (if it makes sense and is possible). Likewise, write an elaborate figure caption that clearly states what the figure shows, values of parameters etc. When I skim through a paper, I look a lot at the figures, and the more the figures speak for the themselves, the more likely I am to keep and read the paper.

As for the methods, tell exactly what values of parameters and settings were used since otherwise the results may not be reproducible:

“…the potential for producing the same or similar results must exist, or your paper does not represent good science.”

Again, this should be obvious, but from time to time I read papers where central parameters are not stated – which makes it hard and frustrating to try to reproduce and validate the results!

How to Write the Discussion

The discussion is the part where the significance of the methods and results presented in the previous sections is elucidated. As Robert A. Day explains, this is often not done sufficiently:

“Too often, the significance of the results is not discussed or not discussed adequately. If the reader of a paper finds himself or herself asking “So what?” after reading the Discussion, the chances are that the author became so engrossed with the trees (the data) that he or she didn’t really notice how much sunshine had appeared in the forest.”

That is to say, make sure to explain what your method and results mean, how they improve exisiting methodologies or shed light on new phenomena, since that may not be apparent from the bare data presented.

Also, avoid overselling your results and write the discussion and interpretation as simply as possible:

“When you describe the meaning of your little bit of truth, do it simply. The simplest statements evoke the most wisdom; verbose language and fancy technical words are used to convey shallow thought.”

Keep it Simple and Avoid Jargon

In short: Use the simplest words and expressions possible in the paper. Most likely, your work is concerned with something complex, but that doesn’t mean you cannot utilize (ehmm, use) simple words:

“The favorite type of verbosity that afflicts authors is jargon. This syndrome is characterized, in extreme cases, by the total omission of one-syllable words. Writers with this affliction never use anything – they utilize. They never do – they perform. They never start – they initiate. They never end – they finalize (or terminate). They never make – they fabricate. They use initial for first, ultimate for last, prior to for before, subsequent to for after, militate against for prohibit, sufficient for enough, and plethora for too much (…) Who would use the three-letter word now when they can use the elegant expression at this point in time?”

Also, keep in mind that scientific publishing isn’t literature:

“Some of my old-fashioned colleagues think that scientific papers should be literature, that the style and flair of an author should be clearly evident, and that variations in style encourage the interest of the reader. I disagree. I think scientists should indeed be interested in reading literature, and perhaps even in writing literature, but the communication of research results is a more prosaic procedure. As Booth (1981) put it, ‘Grandiloquence has no place in scientific writing’.”

The Review Process

The peer-review process may be long and frustrating, but keep in mind that in most cases its outcome is an improved manuscript:

“All editors, and most authors, will affirm that there is hardly a paper published that has not been improved, often substantially, by the revisions suggested by referees.”

Also, don’t be afraid to communicate with the editor, who in many ways is on your side:

“With rare exceptions, editors are awfully nice people. Never consider them adversaries. They are on your side. Their only goal as editors is to publish good science in understandable language. If that is not your goal also, you will indeed be dealing with a deadly adversary; however, if you share the same goal, you will find the editor to be a resolute ally. You are likely to receive advice and guidance that you could not possibly buy.”

And finally, if you feel that comments and critique from a reviewer is wrong, then simply explain and argue for this:

“…if you dispassionately point out to the editor exactly why you are right and the reviewer is wrong (never say that the editor is wrong), the editor is very likely to accept your manuscript at that point or, at least, send it out to one or more additional reviewers for further consideration.”


“Piled Higher and Deeper” by Jorge Cham www.phdcomics.com

2 thoughts on ““How To Write and Publish A Scientific Paper”

  1. Jens Parbo

    Jakob det er et fremragende blogindlæg. Det kunne jeg godt have brugt til at forstå processen da jeg begyndte som PhD-studerende. Jeg vil anbefale vores phd-skole-leder at bruge det.

    Reply

Leave a Reply to Jens Parbo Cancel reply

Your email address will not be published. Required fields are marked *