|Year : 1998 | Volume
| Issue : 4 | Page : 247-250
Practical suggestions in the writing of a research paper
Medical and Vision Research Foundation, Chennai, India
Medical and Vision Research Foundation, 18 College Road, Chennai - 600 006
Source of Support: None, Conflict of Interest: None
Writing a scientific article requires proper planning and a methodical approach. This article provides practical tips to organize the materials before writing, and discusses how to approach the writing of different parts of an article; that is, introduction, materials and methods, results, and discussion. It also provides guidelines on authorship, citing references, selecting photographs, tables and legends, and finally on style, grammar and syntax.
Keywords: Scientific article, introduction, materials and methods, results, discussion
|How to cite this article:|
Biswas J. Practical suggestions in the writing of a research paper. Indian J Ophthalmol 1998;46:247-50
The man of science appears to be the only person who has something to say just now, and the only man who does not know how to say it - Sir James Barrie.
A scientific work, no matter how spectacular the findings are, is not complete until the results are published. The scientist therefore, must not only "do" research but also must write up the research results to communicate with others. However, scientific writing is not easy. Barring a few fortunate exceptions, writers are made, not born. The writing of a research paper is often a tedious and painstaking process, and a very intense struggle between the writer and a blank sheet of paper. Presented here are a few practical pearls gathered over the years by scientific researchers (the author and others), who believe that a scientific experiment or observation is not complete until the results are conveyed to others, which can be achieved by publication. Before writing a paper it is important to answer the following five questions:
- 1. Assess whether the work is fit for publication. Are there enough well documented data and findings? (Remember, a negative result can also provide good scientific information and may be worth a paper.)
- 2. What is the potential audience for this research?
- 3. Which is the most suitable and prestigious (must be peer-reviewed and indexed) journal for publication of the article?
- 4. How would you prioritize this work among other papers for publication? (Do not mess with too many papers at a time!)
- 5. What is the time frame to be followed? (Remember, a paper is not complete until it is despatched and accepted for publication. Plan a realistic time line.)
| Getting Started: A Few Practical Tips|| |
1. Writing a paper requires concentration. You should select a place where you will not be disturbed. Block off your writing time also. Some people find the first hour of work in the morning to be the best time to put down their thoughts. Some find that the evening, after routine work is completed, is the best time to write uninterrupted. Find out which one suits you the best. You should also know how long it takes you to get going and how long it takes before you run out of steam. You should plan a time slot accordingly.
2. Keep everything you need ready before you start working on the paper: for example, data sheets, photographs, and photocopies of key articles in the field.
3. Think about how exactly you will write. Some people are comfortable with pencil and paper, while others prefer to type directly on the computer.
4. It would be of advantage if you could type and use the computer efficiently, particularly the word-processor. If you cannot, take a computer course. If you pursue a career in science you need to write. In the scientific world, the well known adage is, "Publish or perish". Learning to use a computer will make your life easier, and save a lot of time.
| To get going|| |
Once a piece of work is targeted for publication, a journal must be identified early on so that time is not wasted in formatting as journals differ in style, though in minor ways. Keep a list of instructions to authors of the preferred journals to make your job easy.
| Which journal should you send the article to?|| |
- Assess the relevance of your work to a particular journal.
- Check for the impact factor (the impact factor is the number of times a published article is cited in other papers).
| When and where should you start?|| |
It would be best to start after finishing all the critical experiments and completing the analysis. Keep all the bits and pieces of information together with a list of references. Keep track of references, new information coming into the field, and put all information and photographs in a separate file. Select the best photographs you have taken or plan your photographs. Make sure the site of interest is photographed in the proper magnification. Keep aside, separately, the results and records of the research. The basic format of papers is what is generally called the IMRAD format, which is now at least 100 years old.
Introduction Why was this work undertaken ?
Method How was it done ?
Results What did you find? and
Discussion and conclusion What does it imply?
All journals should start with title, author(s), affiliation, abstract, and key words. It might be very difficult to begin the first few lines. Therefore it would be better to start with the materials and methods. This is the part that the author has worked on and should be easy to write. It is wise to write this part while the work is in progress. One simple reason is that it will also facilitate consultation with co-authors.
| Some Important Considerations|| |
| Authorship|| |
Authorship of the paper is an important issue. Unfortunately there is no generalized rule to decide who can be the authors and what should be the order. Most feel that the main author should be the one who takes the intellectual responsibility for the experiments/works done for the paper. Sequence of authors should be decided unanimously before the research work or the work of the paper is started. This definitely avoids confusion or misunderstanding later on. A thorough discussion on "Guidelines on authorship of medical papers" has been published by E.J. Huth.
| Title|| |
First impressions are strong impressions; therefore a title ought to be well studied, to give so far as its limits permit, a definite and concise indication of what is to come - T.C. All.
The title is one of the most vital parts of the paper. It should be short, interesting and should describe the significant results. It may work better to write a few titles and choose the one that will best convey the contents of the paper.
| Abstract|| |
A great deal of importance should be placed on writing the abstract. Do not try to write the abstract before you finish writing the paper. A good paper often makes the writing of a good abstract easy. The abstract should provide the gist of the work done. Currently many journals want a structured abstract with sub-headings such as purpose, methods, results, and conclusion. The abstract should be written in the past tense. Make it clear and simple. Avoid extra details. Provide the main conclusions. Be economical with words. Abstracts are usually limited to 250 words. The abstract gives a very short summary of what the hypothesis is, what has been found and why it is important. With most journals being linked to the web, computer searches often include only the abstract. Therefore your abstract will be read and used more than the article.
| Key Words|| |
The key words should be well chosen so that they are common to the database. Well-chosen key words greatly help in scanning the literature.
| Introduction|| |
A bad beginning makes a bad ending - Euripides.
The introduction basically conveys the rationale of the study. It describes the hypothesis, the background and scope of the study. Important references relevant to the field of study should be chosen. The following steps can guide the writing of the introduction.
- Define the problem.
- Provide a brief review of the literature done in the field.
- Summarize what is currently known on the question.
- State the principal conclusion in the paper.
Do not be tempted to keep things in suspense. Spell it out clearly here. As Ratnoff (1981) very truly says, "Reading a scientific article is not the same as reading a detective story".
The introduction is written in the present tense. This is also a good place to introduce abbreviations to save time and space.
| Materials and Methods|| |
There is a certain method in this madness - Horace.
This describes what has already been done and will be the easiest part to write. The purpose of this section is basically to provide enough details so that a competent worker can repeat the experiments. It is not necessary to give details for every part of the experiment, but if the technicalities of a major experiment cannot be detailed due to lack of space, they should be outlined and references provided at the end. In a clinical study, start with why you selected the topic, what examinations were done, investigations carried out, and statistical methods employed. This is the part that is most carefully reviewed. If your work involves other investigators who did part of the work, the paper should be written in collaboration with or after discussion with them. In experimental studies provide the technical specifications, the quantity, source and method of preparation. Follow the chronology of events.
| Results|| |
This is the core part of the paper. Present only representative data and briefly describe the results. Some explanations can be provided using illustrations. Choose a judicious combination of tables to present similar results.
| Should data be put in the table or the text?|| |
A tabular presentation of data is often the heart or, better, the brain, of a scientific paper - Peter Morgan.
- If there are only a few (one to two) factors to be discussed (such as sex distribution), incorporate them in the text.
- Present the data in tables/graphs, rather than displaying an array of numbers in the text that might confuse the reader.
- Try to present the results as clearly and succinctly as possible. Avoid verbosity, as far as possible.
- Select your tables and figures carefully.
- The table has horizontal rows and vertical columns. Like elements should be read top to bottom and not across.
- The table should have a title and footnotes, when there is a need to amplify. Abbreviations are often needed in the table.
| What about figures?|| |
A picture may instantly present what a book could set forth only in a hundred pages - I.S. Turgenev.
Only figures of prime importance should be presented. They should highlight and illustrate the most important points of your paper. Every finding need not be accompanied by a photograph. Use clear arrows or letters to indicate the area of interest, especially in microphotographs, CT scans, etc. The patient's name should not be seen in any of the figures. If an external photograph (face) is shown, the area of identification can be covered by black tape. For microphotographs crop the unnecessary part, use an inset for highlighting a certain area. The legend must clearly describe what it shows without reference to the text. Remember, some readers will begin your paper by scanning the figures first.
| Discussion|| |
Scientific enquiry requires investigators to challenge the validity, and interpretation of evidence; hence the name research - W.G. Watson.
- This is usually the most difficult section to write. Be specific, do not beat about the bush.
- It may help to go through the discussion of other papers related to the present work. Avoid the temptation to lift discussions from other papers (a common mistake!). The discussion should be as relevant as possible.
- Do not replicate the results section, but start with the major findings.
- Point out exceptions or lack of correlation.
- Show how the results and interpretations agree (or disagree) with previously published works. Then describe the theoretical and practical implications of the present study.
- End the discussion with a short summary or conclusion, if no separate conclusion exists in the prescribed journal format.
To quote Thistle - finally, good writing, like good music, has a fitting climax.
| References|| |
Manuscripts containing innumerable references are more likely a sign of insecurity than a mark of scholarship - W.C. Robert.
| Rule of thumb|| |
- List significant published references. You do not need to annotate all the references unless you are writing a major review.
- Unpublished references should be cited only if necessary.
- All the references should be complete, with author name(s), year, article title, journal name, page extent, and publisher, and should match the text citation.
- Place each reference at that point of the sentence to which it applies, and not necessarily at the end of the sentence.
- Become familiar with the style the journal uses for references.
- Avoid cross referencing from other articles; check it yourself, if available in your library.
- Choose references that strengthen your argument.
| Style, grammar, syntax|| |
- Keep medical and English dictionaries, and a thesaurus handy for quick referencing of technical terms.
| A few common tips|| |
- Do not use double negatives.
- Do not use the present tense for one's own work.
- Do a spell check if you have a computer.
- Avoid repetition of words as far as possible.
| Final Pointers|| |
- Review the paper critically once the writing is over.
- It would be wise to ask a colleague to go through your paper. Often one is too attached to his work to be very critical. Others may be able to see the mistakes more easily.
- Take a break from the work and review the review paper again after a week. Schedule enough time for several revisions.
- Do not send a manuscript unless you are totally satisfied. Once a paper gets published, you cannot change it.
- Do not discard previous drafts till your manuscript is finally sent for publication.
- If the manuscript comes back with the rejection letter from the editor's desk, do not get disheartened or brood over it. Go through the reviewers' comments carefully, make changes if necessary, and send it to some other journal.
Remember most of the esteemed journals have at least a 50% rejection rate.
- Check whether the reviewer has missed any points; if so rewrite for clarity. You may even think of an appeal to the editor in exceptional cases.
- Keep a copy of the revised manuscript as you may have to compare it with the pre-print form which the editor sends for correction.
- At the end you may add acknowledgments. If you have taken significant help regarding preparation of the manuscript use this opportunity to note it. This could be for reviewing your manuscript, help with the pathology slide or doing a part of the technical work. However, you should not include a host of names like who took photographs, typed the manuscript, etc.
- Once the article is published, keep a copy of the paper. It would be too embarrassing if you are not able to recollect your own published findings.
- Keep a back up of the manuscript, photographs and data for at least six months after publication since one may have to clarify the work at a later date if, for instance, there is a letter to the editor from some reader regarding your paper.
- Last but not the least, you should enjoy writing.
| Suggested Reading|| |
| References|| |
Day RA. How to Write and Publish a Scientific Paper
. 3rd ed. New York: Oxford Press; 1988.
Iverson C, Dan BB, Glitman P, King KS, Knoll E, Meyer HS, et al. American Medical Association Manual of Style.
8th ed. Baltimore: Williams and Wilkins; 1989.
Goodman NW, Edward MB. Medical Writing: A Prescription for Clarity.
Cambridge: Cambridge University Press; 1991.
Gopen ED, Swan JA. The science of scientific writing. Am Scientist
Edever F. Refereeing clinical research papers for statistical content. Am J Ophthalmol
Ratnoff OD. How to read a paper. In: Warren KS, editor. Coping with the Biomedical Literature.
New York: Praeger; 1981.p 95-101.
Huth EJ. Guidelines on authorship of medical papers. Ann Int Med