When your primary job is producing academic papers: some thoughts

A series of articles covering thoughts on: 1. Packaging your Research for Publication, 2. Learning on the job: The art of Academic Writing, 3. The art of a minimal draft 4. What are you really trying to say? Questions for authors 5. Levels of Editing of a Scientific Paper. 6 Growing your fluency in Scientific English

Double Helix column by Dawn Field
  1. Packaging your Research for Publication
If you want to be a scientist, you’ll have to write papers. The ‘paper chain’ you produce largely defines the trajectory of your academic career.

We all evolve as writers from our first efforts as children to this day. The more we write, the better we become, especially if we get and absorb feedback from readers of our writing. This holds true for scientific writing as well — and most scientists learn on the job.

Scientific writing, just like any type of specialized writing, has its own rules and practices. This is the first thing to become aware of if you plan to try it: it sounds and reads a certain way and you will need to emulate it to publish. This still means writing as clearly as possible, but formal scientific language is objective, concise and aimed at experts. Readers of scientific journals expect a high level of fluency in domain-specific language. If the authors logically and clearly convey their science with the right terminology, precision and conciseness, it is inferred the authors did equally well designing and executing their experiments and analyses — but it is up to the reviewers and reading public to agree.

The heart of a well-formed, memorable paper is great science. If your science is solid, paper writing is vastly simplified. Doing good science, with adequate controls and according to best practices, protects against your paper getting stuck in review or rejected for methodological reasons. Doing good science means designing good experiments. This is embedded in the grants you write: scientific careers are a cycle of thinking, designing, executing, and communicating.

Some researchers start writing while designing a study. This is wonderfully organized, but it can’t dim your objectivity or limit your interpretations of the results. Sometimes evidence from experiments takes you in a direction you could have never predicted. If you start too early you can get stuck in a loop of rewrites, outdated material, and scrapped analyses. Then writing can seem like a huge drag on your time — because it is.

Huge amounts of time can be wasted on papers with too little data to be a ‘publishable unit’. Often researchers feel pressure to publish as quickly as possible, especially when working in a competitive field, needing a paper as preliminary data for a grant, or for career progression. While these pressures are understandable, don’t let them cloud your judgement. Open questions, incomplete data, unnecessary sources of variation and less than optimal experimental design or outdated technology all raise red flags in reviewers’ minds. Take the time to be sure of your conclusions and wait until you can tell a complete story.

If your data are producing a clear conclusion, you can far more easily ‘package’ up your research for publication. Collate and process your data first, devise the ‘story’, build your narrative (why, what, how, etc) and put it into words for dissemination.

In forming a solid paper, weigh up how it might be cited. This is asking yourself what the value your work is to the wider scientific community. What further work might your work make possible? Viewing your paper through ‘the lens of time’ offers a way to refine your message, find your slant, and weed out the extraneous. What are the lasting messages and contributions? Think hard about how your work advances your field and forges new avenues of research.

Target the appropriate publication venue. Journals down the ranks will be more for specialists. Sometimes this is exactly what you need and don’t shy away from picking a journal whose readership can best use your work. At the same time, don’t undersell your work.

2. Learning on the job: The art of Academic Writing

This article is published as part of the Double Helix column at Oxford University Press Blog: Learning on the job: The art of Academic Writing

3. The art of a minimal draft

One of the quickest ways to get a good paper into swing is to focus on the 
“big picture”. If you start with a big picture ‘minimal’ draft it will vastly accelerate your process. Early on, you’ll have made sure all main points are agreed upon by co-authors.

It’s all about the data and the story it tells

Sit down and sketch the minimal draft when you have 1) a finished data set, 2) finalized figures or tables to support your primary and secondary results, and 3) a target journal.

Agreeing upon the data and the story it tells, saves endless hours of revision. The data define the venue, slant, introductory and discussion aspects of the paper, not to mention the heart of the results and materials and methods sections.

The best way to make sure your story is strong is to practice it verbally before you go the extra mile to put it into words for posterity. Presenting it informally, such as in a journal club, or giving a poster or talk at a conference are excellent ways to rehearse the content. Questions are invaluable as they ‘push on your ideas’ and reveal any holes. You see how your science fares before an audience. The more times you present, the easier the write up — you’ll know the story inside and out and it’ll be bullet proof.

A minimal draft can be shared in the body of email — the point is to be brief. You are defining the absolute core of the paper. As simple as a list of bullet points, it’s content meant to be glanced at but still substantive enough to provoke deliberation, opinion and consensus. A skeleton draft is easy to restructure: iteration early on that produces a final structure is a hole in one.

Defining the Scope of your paper

Putting down the high-level features of your paper is about making sure you have defined the scope of your paper.

1. Go to your selected journal and either get a paper template with the required sections or create one yourself from Author’s Instructions

2. Familiarize yourself with the Author’s Instructions for the journal if you aren’t already

3. Select a working title, if not the final one

4. List your Authors, and think about how they may contribute to the writing of the paper

5. Put a bullet points of your three primary findings in the place of the Abstract

6. Format the matching Figures or Tables to support these three primary findings into your doc.

7. Make a list of the key references that are essential for the paper — the research upon which your work builds (might only a handful to start, or even 1 key study)

Adding the Details

Once you have the approval for the ‘skeleton’ of your paper, fill it in with details that continue to build your balanced structure.

1. Add a justification for this work to your Introduction.

2. Add a list of secondary findings and relevant Figures and Tables (and decide which belong in Supplementary Information)

3. Add a list of headers for your Materials and Methods

4. Add in the key points you want to make in your Discussion

5. Add in bullet points for ‘future work’ that might be undertaken

6. Add in bullet points for your main conclusions

Circulate that more detailed draft for comments and once approved by co-authors you are in the homestretch:

1. Fill in the sections of the paper based on your agreed bullet points

2. Add all necessary references

3. Mark any sections which need input from specific authors

The last step is polish. Make sure that:

1. The title sums up the paper

2. The message in the Abstract is reinforced in each of your sections

3. The logical flow of the paper is solid, the content is balanced, any repetitions or extraneous materials are removed, all content is complete and in its proper place

4. The grammar and spelling is correct, each paragraph addresses one topic

5. Add the Acknowledgements

6. Make sure submission of all data to public repositories is complete, all software is public and all relevant web materials updated

From minimal draft to submission, the process of paper drafting are straightforward if you start with the data and the story it tells.

4. What are you really trying to say? Questions for authors

You aim is to make the structure of your paper rock solid. Having a place for everything and everything in its place makes it as easier to consume often complex scientific contents. Without a logical structure, data, background information and interpretation may sit out of order or worse conflict — in early drafts.

It’s not worth doing detailed editing until the structure is in place. This means getting the story right.

When the lead author is experienced and the paper is well-structured it’s a joy to read. Then a read by a colleague is just about whether it’s sufficiently polished for submission or could be improved in any way.

Great structure means the text has a good flow, nothing is duplicated or left out, and all reader questions are addressed. By the time all the details are added, the paper is balanced and complete. Information is organized and contextualized so that readers aren’t left searching back and forth for key details or methodology. The entire package in ‘unified’: the whole is more than the sum of the parts and all parts are justified and required.

More specifically, a well-written paper has the following features:

1. The title sums it all up.

2. The abstract reflects best what is in each of the other sections

3. The ‘story’ of the paper is easy to grasp, and all the parts serve to reinforce the story.

It’s surprising how often these basic principles are ‘ignored’, but it is usually because the author is still struggling with getting the story straight or even worse is still interpreting the underlying evidence (data). It’s essential to get the data, interpretation and conclusions down pat before structuring a final draft.

Sometimes the lead author has a great structure in mind, it just isn’t yet on the page. Sometimes part of the structure is there, and the rest of the paper looks more like ‘notes’ or ‘what if’s’. Getting the structure can involve shuffling sections, deleting or adding text or doing further analyses.

Questions for authors in support of finding the ‘Big Picture’

Often, authors are so focused on the details that the ‘Big Picture’ goes missing. Sometimes, it’s helpful to sit back and take the bird’s eye view. The ‘big messages’ can be hidden in the sea of words. Speaking freely and informally is a great way to find them beforehand. Researchers often describe the core of their work more easily in conversation but hide it under dry and convoluted in formal form.

Before you start a paper, or hand it over for reading, answer these general questions. If you are doing a read for someone, getting these answers up front can be a great compass:

1. What is the ‘primary’ idea behind this paper as you see it in a few words? This is like the ‘elevator’ pitch and it is the center of your paper.

2. What are the biggest findings of the paper — at least one and no more than three

3. Are these findings in the abstract and highlighted accordingly in each section?

4. Which figure/table backs each one of these findings?

5. What secondary findings are you including and why?

6. When and why do you want this paper to be cited in the future?

7. How does it fit into your career’s ‘paper chain’?

8. What main papers does it build on? (Key references)

9. What descriptor best describes this paper and why: novel findings, method paper, confirmatory evidence, theory, contentious findings (field is a debate, data goes against general thinking, etc), other

Logistics of submission

You can also ask helpful ‘logistical’ questions that might help you give the most useful feedback:

  1. What journal is it for? (is there a history to this paper? Has it been previously submitted/reviewed? details?)

2. How likely you think it is it will get accepted first time? (1 to 10 with 1 being uphill battle and 10 being sure thing)

3. What stage in the writing process you are at (1 to 10–1 being just starting, 10 is ready to go out the door)

4. Any incomplete parts?

5. Any known weaknesses (good to know if you are still working on them!)

6. References — complete or still in progress?

7. Tables and Figures — complete or still in progress?

8. Data: is it finalized or still in the process of deciding what to include/exclude/put in supplemental?

5. Levels of Editing of a Scientific Paper

This article is published as part of the Double Helix column at Oxford University Press Blog: Levels of Editing of a Scientific Paper

6. Growing your fluency in Scientific English

If English is not your mother tongue, paper writing might take you longer or prove more stressful. There are some obvious things you can do to help overcome any remaining language barriers when it comes to publishing.

1. Write your paper in your native language first if it helps make sure you prioritize the logic and flow and not the details of the language. Anything that helps you focus on content and structure is helpful, you can always fix the grammar and word choices later.

2. Co-author with native English speakers. You can’t always pick your collaborators, but one of the easiest ways to improve your English science writing it to work with one or more co-authors who are native speakers. Barring this, getting a read-through from a native speaker as early as possible can be highly beneficial.

3. Avail yourself of spell-checkers and grammar checkers before showing your work to anyone. Online tools like Grammarly also offer tips on grammatical issues and word choice.

4. Enlist professional editing services.

Stages of Fluency

Here are the stages through which every author progresses on the way towards fluency in Scientific English — whether native or not:

1. Searching for words, not knowing how to express a concept, unable to describe certain parts of your work

2. Able to find words and describe everything but sounding ‘awkward’

3. Able to write fluently but lacking specialist scientific language required for scientific fluency

4. Able to express ideas and concepts in Scientific English accepted as ‘publication quality’

5. Able to write compelling, easily understood and widely read scientific papers

Leveling the playing field

The need to level the playing field is gaining increasing recognition. For example, in response to requests to help improve publication rates, the Indonesian government put together this extensive document on “Academic Writing for Publication”, which primarily overviews available courses.

A good general starting point is: “Science Research Writing for Non-native Speakers of English” (2009), by Hilary Glasman-Deal of Imperial College.

Many have gone before and experienced specific issues expected translating from specific languages to English. Sometimes this is an issue of syntax, or emphasis. Other times it is a more fundamental non-overlap of concepts: things that have words in one language just don’t in English, or vice versus.

For Korean scientists there is: “BioEditor: a writing guide for Korean scientists” (also available on Kindle). This experienced editor highlights that the most common translation quirks of Korean-speakers are due to 1. The lack of articles in Korean compared to English 2. Forgetting to add s/es to English nouns to make them plural and 3. Issues with verb tenses.

Often there are significant differences in communication styles across cultures and learning to write ‘good English’ is about more than good grammar. This author encourages Koreans to adopt a far more direct method of communication than they might be used to. Her advice is to make sure readers don’t have to search for the meaning of what you are really trying to say. Write directly. This is good advice for all writers of science. Best of all, the author offers a checklist for getting your paper ready for an editor.

Many English-speaking scientists have done editing for colleagues, usually during a stay abroad. Some have graciously written up their cumulative experiences. The wonderful text “Scientific English as a Foreign Language” summarizes a range of issues that crop up especially when German and French scientists write in English.

This web article, “5 Ways To Improve Scientific Writing For Non-native English Speakers”, has tips on writing, related links and specific advice for Spanish-speaking scientists communicating science in English, along with links to similar resources to help you do science in English.

Some scientists have even formally published on their experiences. The paper “Writing scientific articles like a native English speaker: top ten tips for Portuguese speakers” reviews one scientist’s experiences in Brazil.

For more detailed help, look for materials written in your native language. If you know of additional resources, please add them in the comments section.