Articles

Tips and tricks: Scientific Papers

Writing articles is a core task in doing a PhD: enjoyable for some, but a hellish experience for others. Find below tips and tricks useful to streamline the process.

Key messages

A very useful first exercise is to check what the key messages of your contribution are, and if they will be enough to make (a high impact) article from. Knowledge of the existing literature is important here. For inexperienced students: a vivid discussion with your supervisor may be beneficial...

Planning

Next, a realistic planning is required. Be advised: in many fields, gathering the data for the paper is only a minor task. Making the figures and writing the article are considered a substantially more hefty task, involving dedicated intellectual input. Getting data is one thing, interpreting them well, and placing them carefully in the right scope within the existing literature is quite another...

Figures First

One thing that is recommend in the planning is to prepare (publication-ready) figures first. This exercise will make you also intimately familiar with your data, something which concomittantly facilitates further writing. Tips and tricks on making scientific figures can be found below.

Scientific figures

Learning along the way

During the writing you may learn that certain hypotheses/interpretations you held at the beginning of the process are no longer correct. This is nothing bad: in fact these represent valuable lessons that make your paper stronger. Make sure to communicate these results clearly to your coworkers and act accordingly. For example, it may be desirable to perform extra experiments to strenghten your paper.

Captions

Once you have made the figures, you can provide them with captions. As many people will only briefly read your paper once published, it is important that the figures can be interpreted without reading the article text. Explaining abbrevations again in the caption may help doing so.

Self-criticism

Be advised that as the first author you hold the core responsibility for what is written down and what is published. Do not write for your supervisor, do not write to please referees/editors, instead write to make a high-quality publication from your results. In whatever you do: be self-critical. Double check things. Do not forget: published is published. Referees are not going to stop you from making a fool of yourself!

Skeleton

With the storyline clear(er), you can formulate a skeleton in which you organize the different sections. You usually start by copying the structure provided by the journal template and further specifying the sections headings (mostly for the results and discussion). Next you order the paragraphs within the sections, by summarizing them in one sentence. Here you can also refer to the figures or tables you will use to prove your points/hypotheses.

Draft Abstract

With your key messages established, you may try to write an abstract. This exercise helps to formulate a storyline from your data/messages. This drafted abstract should at a later stage be further refined.

Journal (template)

With the key data identified, it useful to choose a good journal. Each journal publishes different types of articles and it is worth the effort to check which one suits your data best. In addition, most journals provide online instructions as well as templates from which your can work. Choosing the proper journal is a delicate task requiring some experience. So ask your superisor(s) for feedback. After you submit to a journal, it often takes several weeks before you hear something. Therefore, choosing the right journal from the start may save you loads of time...

Introduction: general to specific

Alright, you got your figures and your skeleton made in the right journal template: time to start writing! A part that can be started early in the process is the introduction. It is important that the introduction is used to briefly introduce the topic to the reader. Accordingly, it should go from a broad attention to the topic, to an interesting/promising part of that topic, to things that are desired/problematic in this particular field, to the super awesome stuff you have done that solve all these issues. Mind you; not everything needs to be explained. This is not a tutorial review. For common articles, keep it to ca. 1000 words or less. Usually: the more impact your work has: the smaller the introduction!

Introduction: the 'herein' paragraph

This last paragraph describes what you actually present in your work and should be sexy. It should seduce the reader to keep on going. Accordingly, it may be preferred to not simply and unambigiously diclose all your results, but to say what you did in a bit of a carsalesman's kinda way: "how certain overlooked aspects help to overcome existing limitations, achieving unforeseen performance by utilising the interplay between xxx and xxx." It should be a bit like the label of Nutella jar: you get all excited to put some on your knife! Do you get my draft?

Title

A title should be carefully thought of. Often during writing you can use a running title. The title should catch general attention, while remaining specific enough. Here experience helps... check other papers and ask your supervisors for help!

Experimental: Been there

The experimental part is often the easiest. Simply write done what you did and how did it. Ask yourself the question: "would people be able to reproduce exactly what I did? if the answer is no: specify further. If you used techniques that you have used before: simply see how you wrote it down before. There is no need to reinvent the wheel ;)

Experimental: sample coding

Sample coding that is easily interpreted will make everybody happy! So do yourself and all others a favor and think about this just a bit longer! As mentioned above, people hardly ever read your paper in full detail, so having a intuitive sample coding will ensure your paper gets the attention it deserves!

'3. Results and discussion'
versus
'3. Results, 4. Discussion'

The old school scientists (see work of F. Mumpton above) often prefer to split the results from the discussion. In this case the results are the rather dry part, whereas the discussion is the more lively part. More recently, the parts are often taken together. The choice may depend on the type of data and the preferences of the journal or your supervisor. Clear direction by the journals on this aspect is often missing.

Present versus past tense

During my studies, a strict use of past and present tense was imposed. However, in publications these rules are surpisingly rarely enforced. In general, try to write the abstract in the present tense. In addition, try to write general truth in the present form, and the experiments you performed in the past tense. An example: "Elemental analysis of Nutella revealed that the sugar content was 58 wt.%. This implies that Nutella is of similar sweetness compared to other commercial chocolate pastes"

Double data

In general, papers should be as concise as possible. As such, data depicted in a figure or included in a table does not need to be explicity repeated in the text. Rather, describe the trend the data reveals.

Word inflation

It is recommended to be economic with words. If you are self-critical enough, you may be able to recognize some rather 'useless' words in your writing. For example: "In fact, it seems that". These 5 words do not add anything and are best left out. The work by R. Day (see link above) contains a nice list of examples.

Overcautious writing

Especially new students often write too carefully. For example, if a certain figure shows an increasing trend, it should not be described as 'seems to be increasing' but with 'is increasing'. If something is obviously clear, then write it like this. Only when you are speculating on possible implications you should apply caution with words as 'may, might, should'.

Order

Prior to writing: think! Make sure you have the order and flow of you arguments crystal clear before starting writing. You may even use a flow chart to plan your writing. This may save you valuable time rewriting and reordening afterwards.

Example paper.pdf
File Size:	4902 kb
File Type:	pdf

Download File

Figure and Table references

When to refer to a figure or table? This is a tough one to answer in a universally-applicable fashion, especially if the data if repeatedly referred too. Refer to them the first time you mention the relevant data. Try to limit the amount of referring to a certain figure to 1 per paragraph. If it is hard to avoid repetitive refering to the same figure, you can make a little general introduction of where the figures can be found. An example of this approach can be observed in the start of the R&D of the paper on the left.

Information versus technique

In structuring your discussion. Try to focus on the information you have gathered instead of on the technique used. For example, instead of dedicating a section on nuclear magnetic resonance (NMR), you can focus it on coordination. After all, we do not want to do NMR, we want to learn about the coordination of a certain element within the sample (using NMR). This is clearly demonstrated in the above Example paper.

Focus

Multi-tasking, consisting out of doing many things at the same time, usually leads to a lower quality. Try to do things one at a time, and do them properly. You can easily spent a few days making a great abstract: it will be worth the effort if the quality is top. Simply light a candle, play some Marvin Gaye, and write that sexy abstract!

Flow: Sentence to sentence

Yes: every sentence in a paper should flow into the next. That means that every part of your paper deserves your glorious and dedicated attention. It is very common that especially parts in the end are poorly written simply due to a lack of love! An useful trick is to, instead of reading your manuscript from the first part (abstract), start reading from the back...

Say it as it is

During writing you may sometimes find it hard to describe things accurately. In such a case, it may be good to take a step back, take a deep breath and to try to think of what's going on (Marving Gaye reference). Then, when you are ready, 'say it as it is' using initially simpel words. Afterwards, when you are happy with your argumentation, you substitute those simple words for more accurate ones. Another valuable tool is to shorten sentences to tackle the topic part by part.

Break the dogma

Often, especially as a novice Ph.D. student, you will run into things that everybody says or does, but that simply do not make sense. In such a case, instead of covering it up and avoiding the issue, name it, try to get advice from your supervisors, and -if possible- solve it. This may require extra experiments, but will strenghten your paper. You may even plan carefully and use the experiments for a second 'spin off' paper!

7 C's: Complete, Concise, Clear

Scientific sentences should be complete, concise, and clear. An approach to achieve this is to remove any possible ambiguity from the sentence. Ask yourself the question: 'can the sentence be interpreted in any non-intended way?' If the answer is yes, try to refine your sentence.

7 C's: Consistent, Correct, Considerate, Convincing

The remaining C's: These will help to enforce the above 3 C's, and need little further elaboration. Within the academic circle it is important to write also considerately: Eventhough you may not agree with other works, you should always describe other work in a fair and objective manner. In trying to be as convincing as possible, you could use examples.

The broad picture: dare to generalize

After you have finished your R&D, you may have gathered a unique perspective on the matter. Well done, that's the purpose of being a scientist! Ideally, you could use this perspective to generalize or review the existing works on the topic in your paper. To do this in the strongest possible fashion, you could compare your data with those from the existing literature. Generalizing can be nicely done using conceptional graphs. The good thing of such graphs is that they are extremely insightful but do not require extra experiments to make. Moreover, these figures will make your work more cited! Of course, be careful in your generalizations, and add disclaimers if needed.

Supporting information

If you find yourself in a surplus of figures, you may put some in the supporting information. Particularly those that are derived from routine analysis and bring no added value may be easily be placed there. Sadly, the supporting information is also often used to 'hide' unfavourable results...

Conclusions

At the end of the paper you may conclude. Here it is important to highlight why your results are so important. You could even predict some possible advantageous implications of your work. See also the useful tips from F. Mumpton (link above).

Acknowledgments and funding

Mention those that contributed but not enough to justify being an author. Funding should be always acknowledged. The latter may seem unnecessary, but do not forget that the people in charge of giving the grants (including the one you may be enjoying right now) are also keeping track!

References

The references should be done in the style of the journal. Although software can be used to facilitate this (such as Endnote), they often imply difficulties further down the process. Accordingly, especially if you have ca. 60 or less references: you could try to do it by hand! Old School those babies!

Graphical abstract

The graphical abstract should cover the most basic message of the paper. It is often depicted rather small. Therefore, try to keep it simple. People have little time to study it and it should bring your message across within several seconds. PowerPoint can be suitably used to make graphical abstracts (see tips in the Arts section). Bonus tip: if you are lacking inspiration: ask your colleagues and even friends for help or inspiration!

To the Arts section

Are you sure?

Published is published. Make sure that you are sure of what you write. You do not want to make an ass out of yourself. Also, think twice about introducing or 'coining' new terminology: before you know it, you may be forced to reuse your own poorly-chosen words time and time again...

Abstract

The abstract will be the most read part of your paper. Make sure it is absolutely spot on. Abstracts typically contain 1-2 introductionary sentences, followed by a somewhat dry summary of the results, and a final sentence regarding the implications.

Abstract vs herein vs conclusions

It can seem confusing to distinguish the abstract, the herein paragraph of the introduction, and the conclusions. An visual aid to highlight the differences may come in the form of the nutella jar: the abstract is formed by the dry nutrition facts. The tasty-looking label is the herein paragraph. Then, the conclusions, showing the implications of a tasty chocolate paste, is a content and happy child eating Nutella. (Finally, in the supporting information you can mention that exposing your kids to a high sugar diet might be bad for their health).

Key words

These key words are used online to link your work with other works and requests placed in the journal's search engine. Accordingly, to optimize your exposure, choose them wisely.

Bonus: patent literature

One thing that remains (sadly) often completely overlooked is the patent literature. This is surprising as these works reflect the 'real' world probably more accurate than scientific papers. If you really want to make a dent, make sure to also check patent literature. Particularly the background of the inventions may enlighten you!