I promised a few weeks ago that I would blog about how I write a paper from start to finish. I was hoping to have screenshots of every stage of my paper writing, but obviously doing my own research, fieldwork and travelling to academic conferences to present papers (and writing those papers in haste!) didn’t allow me to do this in a much more planned manner. So here are 8 tips I use to write a research paper from start to finish.
1. Create an outline
This tip would be kind of obvious, but I am far from being the first one to suggest that writing an outline allows you to put complex ideas on paper in a sequential, articulate, cohererent form. If you’ve already started writing the paper, then Professor Rachael Cayley’s approach is the best – e.g. create a reverse outline. At any rate, you should have a skeleton of what your paper is going to look like. One way in which I do this is I break down my abstract into the sections that I need to fill out and/or the questions I need to answer to have my paper actually show my full argument. So, the outline comes directly from the paper abstract. What I have found is that often times, my outline doesn’t show the same thing that the paper does at the end of it. That’s fine. At least you answered the questions and/or filled the sections you needed to and refined your abstract and paper on the basis of these responses.
2. Write the abstract and introduction first
The one sure way in which I know I am going to make progress on a paper is writing the abstract and the introduction. Normally what I do is I expand the abstract and write the introduction from the abstract. I also make sure that I develop the structure of the paper as I write the introduction. Often times, this will change and I will have to come back and redraft this section, but at least I have a basic structure for the paper.
2. Break down the paper into separate documents.
I am someone who doesn’t react well to word counts. In fact, I loved a recent blog post by Tseen Khoo entitled “Your Word Count Means Nothing to Me“. I am disciplined about writing every day for two hours, but I don’t really like the idea of “I write 3,500 words every 1.5 hours”. Some days I write a lot, some days I write much less. And some days, I just simply can’t write (though I summarize papers and reflect on them during my #AcWri period those days to keep generating text that I might use at some point, particularly research and reading memoranda).
So what I do instead is, I break the paper down into sections for which I then create separate documents. For example, for my recent paper on environmental mobilizations against Nestlé in British Columbia and in California, I created a separate document for the story around Nestlé in British Columbia and another one for the story on Nestlé in California. To avoid getting frustrated, I just focus on writing on one of the sections at a time.
4. Begin drafting some conclusions as you complete the analysis
As I write my paper, I always make sure to include some early conclusions. For example, for my recent paper on the comparative analysis of 6 remunicipalization cases, as I completed each section and the history of each remunicipalization, I started integrating and summarizing my results in the analysis section and immediately after, I wrote a couple of sentences about the implications of my analysis for the conclusions section. By the time I finished the sixth case, I had 6 paragraphs in the conclusions section of my paper. This is particularly important as it helps me see the light at the end of the tunnel. As I was finishing the table that summarized my paper’s findings, I was able to also have a feeling of completion. By the time I had completed 3 case studies, my table looked quasi complete and I began feeling excited about finishing the paper.
5. Make sure you’ve told all the stories
As I was trying to finish my MPSA 2016 remunicipalizations paper (with a comparative table of 6 cases – Paris, Grenoble, Berlin, Atlanta, Hamilton and Buenos Aires), I got frustrated that I had assembled the paper too early for my liking and therefore I was not sure if I had completely told all the stories. For me, a story is fully told when there is at least 4-6 paragraphs that outline the overall issue and provide some analysis. That’s why at least 4-6 paragraphs would be necessary (history, the issue at hand, why is this issue relevant, what does my theoretical framework say about this particular issue) to fully outline and sketch the story. So, while I recognize that I had assembled the paper early, I used a summary table to ensure that I had already completely told all the stories. This table also helped me finish the paper because I could use the insights gained from this exercise for the analysis section and the conclusions section (see tip 4).
6. Leave text for the next day
This tip sounds counter-intuitive, but this is exactly how I finish papers: I leave myself some room to complete sections, paragraphs and sentences. For example, for my environmental mobilizations paper, I wrote the section on the history of the environmental protest against Nestlé in British Columbia, on the Tuesday, and even though I wasn’t exhausted, I decided to just start the first few sentences of the California case. This tip is particularly important to me because I write in the morning. I start at 4 or 4:30am, wake up, start a pot of coffee, and write from 4-6, 4:30-6:30 or 5-7 am, because those are the hours when I am most productive.
7. Don’t write beyond your physical limits
Recently, I finished a book chapter by inserting 3,500 words that I wrote in the first 1.5 hours of the day into a draft that had 3,400 words. So I finished an 8,000 word paper in about 2 or 3 days. Obviously this only works if you’ve already simmered and thought about the paper for a very long time. I had been spinning my wheels for the past few days when I knew that I had made no progress on this paper in the past 4.75 months. This week, I just decided that I needed sleep and I stopped trying to write (yes, I too try to push my limits and do some “spree-writing”) so I went to sleep early. I woke up on Wednesday at 5 am, and by 6:30pm, I had finished the book chapter.
The reality is that academia has this toxic culture of overworking as though it were a badge of honor. But I can’t do that anymore. I used to work 24 hours in a row, sometimes even 36. Right now I can’t push my physical limits and I will not endorse overwork. So I know for a fact that I improved my writing since I started sleeping at a decent hour and at least 6 hours a day. And that’s exactly why I never write beyond my physical limits even if I am not done with the paper and I have a deadline. I prefer to ask for an extension or simply say “No, I can’t write your book chapter/paper/article” because I will no longer push myself beyond my physical limits.
8. Assemble the paper 80%-90% into the process
When I assemble a paper too early into the process, I end up seeing all the gaps in the paper and this demoralizes me. So now what I do, is I assemble the paper about 80-90% into the process. I assemble the introduction, conclusion, body of the paper and I collect my handwritten notes of what needs to be improved and corrected. And then I go over the paper and figure out if I am missing something. That way, whenever I sit down and work on this paper again, I feel that I am about to be done.
Applying this process helped me complete 3 draft papers (2 for MPSA, 1 book chapter, and two I’m working on) in about 5 weeks, all the while travelling every week and teaching one class every week. This is not to brag, but it’s just to show that if I follow a systematic process, I can move forward even under conditions of relative duress (e.g. when I am travelling). So, every single day I was able to work on research and write for a few hours because I was working every day on a different, single component of my paper and research project. As I have often said, I follow Aunty Acid’s advice: I take life one panic attack at a time.
Posted in academia.
Tagged with academic writing, AcWri, research paper, writing.
ByRaul Pacheco-Vega–April 16, 2016
How to undertake a research project and write a scientific paper
Author information ►Article notes ►Copyright and License information ►
Accepted 2012 Feb 14.
Copyright © 2013 Royal College of Surgeons
This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
This article has been cited by other articles in PMC.
Research and publishing are essential aspects of lifelong learning in a surgical career. Many surgeons, especially those in training, ask for guidance on how they might start a simple project that may lead to a publication. This short paper offers some practical guidelines on the subject.
Keywords: Research Techniques, Research Activities, Publications, Journal Article
How to get started with a project
How to get started varies depending on whether the project is suggested by a trainer or educational supervisor. Projects suggested by a senior are always offered as an encouragement to a trainee, who should be careful not to respond in a negative way by ignoring the suggestion, coming up with a string of excuses or doing the project badly! Here are some simple steps that may contribute to an organised start on the project. You need a protocol but first you must be clear about what the project will involve.
Undertake a literature search on the suggested topic.
Read all the papers from the last ten years and summarise them on a single page of A4.
Make a note of how many similar series have been produced, their size, the length of follow-up and any special aspects of the subject that have already been addressed.
List aspects of the topic that have not been well covered, perhaps morbidity or surgery for rare indications, or long-term outcomes.
Discuss your thoughts on the subject with your colleagues.
With the strengths and weaknesses of the current literature clear in your own mind, summarise your thoughts in bullet points on a single side of A4 and arrange ten minutes to discuss them with the senior who suggested the topic.
The six steps listed above can be easily completed within a couple of weeks. Once you have discussed and agreed the aims of the project as well as how they can be achieved, you can write your protocol. It is also possible that having studied the literature you decide the suggested project is unlikely to add to our current knowledge and that another topic might be better studied.
A protocol and approval from your trust’s research and development (R&D) department as well as from the research ethics committee (REC) are needed before you begin a research project. If you are planning a service evaluation, REC approval may not be needed. When you have secured the approvals, the process of collecting the data begins.
Examining a case series, there may be hundreds of medical records that need to be studied and it is crucial to draw up a ‘proforma’ on which to record patient data. This should ideally fill no more than one or two sides of A4 and needs to include all the data that you have decided to collect for your particular study. It is crucial not to leave out a dataset you might later wish to look at but on the other hand it is also important not to collect too many data. Because of this fine balance, it is important to draw up a proforma and agree its composition with your supervisor and any co-workers on the project before starting to collect data from the medical records.
Data collection can be time consuming and it may be that several colleagues can work on this to speed the project along. Once all the data proformas are filled in, the data need to be entered into the database, spreadsheet or statistical package of your choice. It is best to use the software favoured by the department or colleagues in medical statistics.
Having looked at the data, discipline yourself to produce a succinct summary on one side of A4. Again, arrange a meeting with your supervisor and any other co-workers to discuss the findings, and give everyone the opportunity to comment and correct the summary. Once the findings are agreed, you are ready to write up the project.
Sometimes you will want to develop an idea of your own. It is even more important with a self-generated projectto do a thorough literature search to make sure that your ideas will contribute to our knowledge. The discussion of a more ambitious project like a randomised trial should be with as many colleagues as possible, both for advice and also to garner support for your idea. Having produced a single side of A4 summarising your idea, identify a senior colleague who can advise you and proceed as described above.
As noted previously, REC approval is needed for any clinical research involving patients or their data. You will need to prepare an application on the Integrated Research Application System website (https://www.myresearchproject.org.uk/). If you have never done this before, seek advice from your trust’s R&D department. REC approval is time consuming; the following comments may help:
Much of your initial work producing a summary of your idea will be helpful in completing the ethics committee form. It is crucial that submission to your local ethics committee is checked by all your co-workers.
Colleagues from medical statistics and any other parallel disciplines such as radiology or medical chemistry need to be involved right at the start of this formal submission so that all aspects of the study are academically correct. It is especially important to have expert statistical input because it is very demoralising to finish a trial only to be told that your study is woefully underpowered and cannot answer the question that it set out to address!
It is wise to present your idea to the committee in person as this can save time and iron out minor misunderstandings. These ‘glitches’ in an ethics submission can soak up months of precious time and a personal meeting with the REC can help to avoid them.
Many institutions also have research governance or internal review boards that must also pass a project after it has gained ethical approval. Their role is often to assess the financial and organisational impact of a study.
This process seldom takes less than 3 months and may take nearly 12 months. Do not be disheartened by this. If your study is worth doing, then it is worth persevering.
The recording of data using a concise proforma, entry into appropriate computer software and production of a summary of your findings are all conducted in the same way as in the first section of these guidelines.
Writing up a study
One of the most challenging aspects of surgical research is writing a paper. Putting together a manuscript for submission to a journal can be broken down into several simple and relatively self-contained steps:
Journal guidelines: All journals have a set of instructions for potential authors. The suggestions below are an overall guide to writing a paper but should be viewed in the context of the specific guidelines on submission to the journal you have chosen for your work.
Title: Keep this simple and concise.
Authorship: This topic may be a source of some problems. My own observation about authorship is that if you leave somebody out who feels they have contributed to your project, you can make an enemy for life! It is easy to forget colleagues, especially when a project has run for several years. Try, within the internationally agreed authorship guidelines, to include all colleagues who have contributed significantly to your study.
The order of authorship may also cause problems. It is generally agreed that the main researcher who also produced the first draft of the paper is the first author. The second author has usually been the second main contributor to the project. The last author is the senior person supervising the work. Between these positions come all other authors who fulfil the guidelines for authorship. If in any doubt about who should or should not be in the authorship, discuss it with your senior author.
All papers have a corresponding author responsible for answering queries after submission of the manuscript. It is best if he or she is a permanent member of the department as queries may arrive several years after a paper is published.
Abstract: This is usually 200–250 words and should be written in the style of the journal. Generally, this includes sections on background, methods, results and conclusions.
Introduction: This should introduce the reader to the subject covered in the study and explain why this particular study has been undertaken. It should be kept to two or three paragraphs. The first paragraph sets the scene and summarises the current literature. The second paragraph should justify why this particular study or series of cases has been put together.
Patients and methods: The most frequent mistake in this section is to include results as well as patient details. It is important to stick to describing the study population, how they were collected and, crucially, how any analyses were undertaken. Always describe what statistical tests were used and justify why they were appropriate.
Results: These should be presented concisely with as few tables or figures as possible. Use a logical sequence and follow the same sequence in the methods and discussion sections.
- Discussion: Many surgeons don’t know where to start in this section! Over 25 years I have found the following outline helps to clarify one’s thoughts when discussing a study. Using these five headings can keep the discussion concise, relevant and, most importantly, just five paragraphs in length!
What are the main findings of your work? State clearly what you can conclude from your observations, taking care not to overestimate what you can conclude.
Why are these findings valid (sample size, methods etc)? Explain what leads you to conclude that your findings may be relied on. Also make sure you highlight any potential weaknesses in your data and consider other potential confounding variables that might invalidate your conclusions.
How do your observations compare with other work in the same area? Discuss how results from your work compare with other papers on the same subject, either explaining similarities or examining differences.
Any other business? Are there any unexpected side observations that merit separate discussion? This might include unexpected complications in a trial or a unique subset of patients in a clinical series.
Restate your main findings and suggest what further work might be helpful in providing more information on the topic of your project.
References: Make sure these are presented in the style of the journal you have selected.
Publication of the paper
This can be the biggest hurdle you have to clear! Some basic rules will help to make this easier. First, never submit a paper without all authors having read it and agreed to the content. Second, never submit a paper to more than one journal at a time. Finally, remember that submission is not the end of your paper but just the beginning.
Selection of the right journal is important. On the basis of their impact factor, journals may be divided into four divisions. Think of it like the football league! The premier division contains journals with impact factors greater than 10, the second division those with impact factors from 5 to 10, the third division with impact factors from 1 to 5 and, finally, the fourth division with impact factors less than 1. Just as with football, journals may be promoted or relegated so it is wise to check online for a journal’s current impact factor.
Discuss with your co-workers what your target journal should be. It is acceptable to aim just higher than you think your paper ranks but obviously pointless sending a small case series to one of the premiership journals. A second consideration is which articles have appeared in your target journal over the last 12 months. If there have been one or more papers on the same subject as your work, it may be better to select an equally ranked journal that has not had a paper on your topic for several years.
Peer review is the process used by journals to select papers for publication. Many papers are rejected immediately but those deemed of potential interest are sent out for peer review. This process usually takes 3–4 months (although some journals such as the Annals of The Royal College of Surgeons of England have a much quicker turnaround). There are four potential outcomes:
Accept without corrections – this is very rare!
Minor corrections needed followed by resubmission for publication
Major corrections needed and resubmission invited but without any promise to publish
Major criticisms and rejection (for most major journals this is the single largest category of outcomes)
When you receive the reviewer’s comments don’t take them personally! The best way to regard the reviewer’s criticisms is as helpful suggestions to improve your paper. It is crucial to deal with each of the reviewer’s comments carefully, systematically and politely. If possible, respond to the comments within a few days of receiving them.
If your paper has been rejected, then the reviewer’s comments are an excellent set of suggestions to improve the manuscript for submission to another journal. This should probably be in one division lower than your first submission. Again, there is no reason to delay resubmission to another journal more than a few days. Make sure that all possible advice on rewriting and correcting your paper is taken and your work will almost certainly get published eventually!
Articles from Annals of The Royal College of Surgeons of England are provided here courtesy of The Royal College of Surgeons of England