Tips for Publishing in Scientific Journals

Scientists in a lab at the University of La Rioja (Wikipedia)

by Katrina Kelner, Science Careers: http://sciencecareers.sciencemag.org/career_magazine/previous_issues/articles/2007_04_06/caredit.a0700046

A string of impressive publications can propel a young scientist to the next academic stage, whereas an insufficient publication record can derail a career.

Publications are the main way scientists publicize their work, and ultimately, it is by their papers that they will be judged.

So what makes a good paper? The most fundamental ingredient is excellent research. Work with the best scientists you can, in the best lab you can find. You will absorb the most about doing excellent science if you are surrounded by it during your training. Then make sure that the questions you investigate are important and of interest to others in the field.

As an editor at Science , I see that the most successful papers are those that present innovative research. But the best papers also present their story in a clear and logical way. The thinking behind the paper is clear, so the writing is clear. Writing research papers with all these qualities can require a bit of strategic thinking, practice, and know-how.
 
Choose a good environment for publishing

One of the signs you should look at when choosing a lab for your thesis or postdoc is the group's publication record. Look for consistent output of good publications, because this will tell you that the lab is run well and that the lab head manages research projects successfully.

Different members of the lab should also be listed as first authors, because this will show that projects and credit are distributed. Make sure that the papers are in journals in which you would want to publish. Then read the papers to find out about the writing skills of the lab's scientists.

Are the papers clearly written? Did they convince you of the importance of doing the experiments? Can you easily tell what the important conclusions are?

The best way for you to learn to write first-class papers is by getting as much practice as possible. Before deciding what lab to join, as you examine the facilities and find out what it is like to be part of the team, also make sure to ask about the writing process.

Do postdocs or graduate students get to write the first draft? Do they get valuable input from the head of the lab and other colleagues? Or does the head of the lab just write the paper and show it to the student or postdoc, which will not be so useful to them?
 
Think in figures

Once you are working in your new lab and producing data at full speed, you have to judge when you have enough data to write a paper. Write too soon, and you may be wasting your time. Wait too long, and you risk getting scooped. Stop and write when the data are sufficient to tell a story that is complete and makes sense.

The key is to constantly keep the paper in mind while you are performing the experiments. Think about the figures that can already go into the paper and the information they will contain. The reader must come to the same conclusions you have solely on the basis of your results.

So ask yourself whether, after grasping the results presented in your figures, the reader will be led to the correct overall conclusion. What convincing experiment might be missing? Are there alternative explanations? If so, what data will you need to collect to eliminate that other possibility? Before performing a new experiment, always ask yourself how it will contribute to the logic of the publication.

As you are immersed in the details of your work, it may be difficult to remain objective and see the holes. Test your reasoning on colleagues by asking them whether you told a logical and convincing story after giving a talk from your assembled figures, for example.
 
Choose an appropriate journal

Aiming your paper at the most appropriate journal can save much effort and reveal your results to the world sooner. The so-called top journals value novelty and unexpected findings, but other journals may be more interested in careful, extensive analyses of critical ( e.g., biological) processes.

Survey the various journals and see where your work would fit best. Get advice from colleagues and others in the field who have experience as authors, reviewers, and journal editors.

It may be tempting to send your paper to a top journal even if your results are not of the highest novelty or broadest interest. But you can save time and reduce your frustration if you send it to the appropriate journal first instead of waiting until it's rejected by a top journal.
 
Submit a high-quality paper

In the eyes of your readers - editors and reviewers included - the quality of the paper you send in directly reflects the quality of the science behind it. A careless approach to writing can undermine the most meticulous experiment. It is thus critical that the paper be free of careless errors, especially in the data.

Check and recheck that all information is consistent, that the images and graphs represent what you say they represent. Again, figures are your best ally to convey your story, so make them easy to understand. Each figure should make only one or a few related points, and together they should make all the paper's important points in an easy-to-grasp manner.

Put as much information about the data and the conditions of the experiment directly on the figure as you can. The figure legend is important, but the less the reader has to refer back and forth to it, the better.

Do not neglect the form. It is critical that the paper is written clearly and that it contains no spelling or grammatical errors, and that the logic is crisp and clean. Show your paper to your most critical friends and colleagues and take their advice seriously. Also make sure that all authors have seen and approved the submission!
 
Help ensure that the review process goes smoothly

Journals can be run by professional editorial staff (such as Science, which receives about 12,000 submissions per year) or by academics who take on the role of editor for a defined period of time. Both types of editors send papers out to peer reviewers - working scientists who evaluate your paper for accuracy, logic, and scientific interest.

Some journals (such as Science) have an initial screening step in which papers unlikely to make it through the review process are rejected. Science editors make these initial screening decisions with advice from the Board of Reviewing Editors, a group of more than 100 working scientists.

Reviewers are chosen by the editor on the basis of their expertise in the field, often utilizing extensive databases assembled by the journal and the editor’s knowledge of the area. Some scientists are better reviewers than others--they are more critical and thorough, a fact that quickly becomes known to editors. The review process can take anywhere from a few days to several weeks.

After review, the editor makes a decision about publication, taking into account all of the feedback he or she has received. The editorial goals of the journal--sometimes journals decide that certain areas are of particular upcoming or lessening interest--factor into the decision, as does knowledge about the reviewers themselves and the background behind their opinions.

You can help the review process go smoothly by providing a cover letter that includes, in very clear language, a concise version of the whole logic of the paper that makes clear its importance and context.

If there are any special considerations that the editor and reviewers should take into account, include these in the cover letter. These might include information about your own availability, related work being reviewed at other journals (from your lab or other labs), or the names of other scientists who are working on the same problem and so would have a conflict of interest in reviewing your paper.

Keep the list short; otherwise, the editor will be forced to ignore your list or get an uninformed review. If it is necessary to ask that a few individuals be excluded from review, explain why.

All of the related data not included in the main body of the paper should be clearly accessible to the reviewers, either as an appendix or through a publicly available database.
 
Respond to reviewers' comments positively and constructively

Good news: The journal wants to publish your paper. Still, only on rare occasions will reviewers recommend that your paper be accepted without revision. New experiments - usually ones that can be done within a few weeks - are often among their requests for revisions. Science editors also often give authors extensive advice on how to revise their papers.

Remember that the editor and reviewers are interested in your paper. They want to see it improved and published. You increase the chances of your paper being accepted if you make the assumption that the reviewers are offering their suggestions as constructive criticism.

Make all possible attempts to comply with their requests, including performing extra experiments, even if you think they are unnecessary.

Of course, sometimes the reviewers' requests are misguided or based on faulty reasoning. In these cases, especially if you have agreed to address the rest of the reviewers' comments, the editor may be willing to consider a reasonably worded argument that the request does not need to be fulfilled for acceptance of your paper.

When you send your revised paper back to the journal, you should include a detailed, point-by-point explanation of how you have addressed each of the reviewers' and editor's comments. Remember that the editor may send your responses to the reviewers, so if you are refusing to address one of the referees' comments, you should word your argument carefully to be clear but not offensive.

Always treat the reviewers' comments and motives with respect. It is never a good idea to engage in personal attacks or observations about reviewers or reviews. Also be polite to your editor. The editor will be most disposed to work with you when it is not unpleasant to do so.
 
How to deal with rejection

In spite of your best efforts, you have received a rejection letter from the journal of your choice. This does not mean that your paper is not good. At Science, we have to reject more than 90% of the papers submitted to us.

A rejection can be upsetting, and it is often sensible to let at least 24 hours pass before thinking about your next steps. It is not a good idea to fire off an angry e-mail to the editor explaining why the journal's process was unfair and biased.

If, after careful consideration, you think there has been a misunderstanding or error, some journals will entertain a request for reconsideration, usually in the form of a clear letter or message explaining your point of view. Some editors might be willing to have a phone conversation.

In most cases, the best and most time-efficient course is to reassess quickly your choice of journal, fix any weaknesses that may have been pointed out in the review process, reformat the paper for your second-choice journal, and send it off.

About 70% of papers rejected by Science are eventually published elsewhere. Even a submission that ends in rejection is an opportunity to hone your writing and editing skills.

Katrina Kelner is Deputy Editor, Life Sciences, at Science magazine.

What’s Your Back-Up Plan?

by Thesis Whisperer: http://thesiswhisperer.com/2014/12/10/whats-your-back-up-plan/

Let me tell you a story about my PhD.

I studied how architects and students gestured when they talked to each other.

I was trying to find out what role gesture played in classroom communication and what implications this might have architectural education.

Gesture is a visual medium, so I decided to use participant observation and video ethnography to collect my data.

I sketched out a research design and picked two universities as sites for case studies. I did an ethics application and then approached people I knew at each university. I asked if I could film them inside their classrooms and was grateful when they accepted.

I spent many days with these teachers and their students and encountered many unanticipated problems along the way. Cloud computing and large capacity smart phones were not yet available so data handling was … awkward. As it turns out, analysing film data is extremely time consuming; every mintute took me hours and hours.

Then there were the people problems. Being in the field was confusing and difficult for everyone. Observing in the classroom was full of basic, but tricky problems. Where should I sit? How long should I stay? Who should I film? I’m grateful, to this day, that my teacher informants, Pia, Anna, Simon and Peter stuck with me and the students put up with me.

After a year of struggling with these problems I finally started to see patterns in what I was observing. I talked about what I was seeing with whomever would listen. People seemed really interested (or, at least, were polite enough to pretend).

I developed some theories about what was going on and wrote a small conference paper. I started to plan the last semester of data gathering and map out a structure for my final thesis.

More importantly, after being consumed with self doubt during the early stages I started to feel really good. I had this research thing totally sorted!

Then, one day, I got a call from a member of department at one of my research sites. This person had become aware of my research through casual conversation with his colleagues. He demanded a meeting where he asked me long and searching questions about my ethics approval approval and my process of data collection, in, I might add, a quite aggressive tone.

I was mystified at the sudden attack, but tried to stay calm. I used all my recently acquired participant observation skills to try to work out what was really going on in the conversation.

It became apparent that this person had formed the opinion that my research was deliberately designed to bring his institution into disrepute. This was bewildering until I remembered an important detail: the two research sites were in different universities who had long considered themselves rivals.

I had studied and worked at one for a long time before moving to the other to do my PhD. I had thought nothing of this rivalry at the time but now it seemed I had a tribal warfare problem. Perhaps this person considered me a ‘defector’ out to do my old institution harm?

I talked fast, trying to convince him that my data could not be used to find out if one university was ‘better’ than the other - even if I had been interested in this question. But he just looked at me skeptically.

My heart sank when I realised another important detail: this person did not have a PhD and had no experience of doing research. Talking about epistomology was probably pointless. He was never going to believe me.

The meeting ended on a sour note, with him promising to refer the matter to the ethics committee, who, I knew, had the power to shut the project down and make me destroy the data.

I went home in a state of extreme agitation. This was a problem I literally could not solve. Had I made a fatal mistake? I had followed all the rules set up by the ethics committee to the letter - but would that be enough to protect me? If I lost half my data, the case study design would no longer work. Could I even finish my PhD?

I waited on tenterhooks for the axe to fall, all the time berating myself for not anticipating the problem. A week went by. Then another. Then a month. I asked at my university, but no complaint had been received. I didn’t know whether to relax or not. I was having trouble sleeping.

Two months or so later I ran into the person at a social event and asked him if he had sent the complaint yet. He chugged his beer and eyed me for a moment, in silence. Then he said he had decided not to bother as no one but the examiners would read my thesis anyway.

I politely thanked him and walked away, stifling the strong urge to throw my drink in his face. I wonder to this day if his concern was genuine, and I had allayed his fears through my mad skills of research ‘splaining, or if he was just screwing with me.

This story has a happy ending: I finished my PhD, but I think it illustrates an important point: you need a back up plan. Preferably a plan that involves losing ALL your data. This sounds drastic, but it’s really a sensible risk management strategy and is time well spent. I suggest you sit down and ask yourself a couple of difficult questions.

If you lose your data, can you easily source an alternative, equivalent set?

Researchers are being encouraged to share their results so in some fields there’s lots of open data around. Spend a bit of time searching and see what’s out there already (ask a librarian to help you if you are confused).

Or have a look at data in the public domain such as government data or archival data. Find a set that looks promising and write a quick research plan around it. Maybe it’s so good you don’t need to collect your own data after all! (hey - we can all dream).

If you lose your data, can you alter your research question slightly so you still have a thesis without any data at all?

This is what Stephen Johnson calls ‘the adjacent possible’. Could you look at the historical dimensions of a problem, rather than the contemporary issues? Can you do a completely theoretical project? Decide on a few tacks and write a quick research plan based each of these new approaches and evaluate them.

Of course, some disciplines are more forgiving than others in this regard: most scientists will have no project without data. In that case you need to ask yourself the really hard problem:

If you lose all your data how long, realistically, will it take you to either design and conduct a new study in the time you have left, or support yourself long enough to do it all over again?

If you need to delay your submission, where will you live? How will you support yourself? Are there visa problems? What are the rules around extending your candidature?

I hope I’ve convinced you to spend a day or so and make yourself a back up plan now - hopefully you will never have to use it!

How about you? Have you encountered a big problem which put your whole project at risk? What did you do? What did you learn?