Tips on Writing a Scientific Paper

A well-written paper has a better chance of being accepted in any journal if it is written simply and concisely. Be certain your words express your ideas and your message.  In short, adhere to the following rules: (1) Have something to say; (2) Say it; (3) Stop as soon as you have said it.  Otherwise, the scientific worth of your manuscript may be obscured. 

General Suggestions

  • Adhere strictly to the journal format as described in the Guidelines for Authors. An incorrect format suggests that the paper was previously rejected by another journal and not changed for the resubmission.
  • Edit your paper carefully and eliminate errors of spelling, punctuation, and grammar.  After you type the final draft (and especially if someone else types it for you), force yourself to edit it once more.
  • Always check the accuracy of your references with the original, and not with secondary sources. Incorrect citations are a burden to the publisher and a disservice to the reader.
  • Do not expect the editor’s office to rewrite poorly-written manuscripts - that responsibility rests entirely with the authors.  Those who have difficulty writing scientific English should obtain assistance from colleagues or use a language editing service.
  • Organise your paper to answer four main questions; the reviewers and readers want them answered:  
    1. What did you set out to do and why? (Introduction)
    2. How did you do it? (Methods)
    3. What did you learn? (Results)
    4. What does it mean and how does it relate to what else is known? (Discussion)                     
  • It is easy to mix fact and opinion; keep the Results and Discussion separate.  Keep the discussion clearly reasoned, tightly written, and focused on the implications of the Results.
  • Keep the abstract brief and in the active voice. For original articles the abstracts should be structured. Avoid sentences such as “The implications are summarised.”  Instead, describe the implications.
  • Avoid repetition:
    1. Do not repeat the Abstract in the Introduction or Discussion.
    2. Do not disclose your Results in the Introduction.
    3. Do not repeat the Introduction in the Discussion.
    4. In the text, do not repeat figure legends, table titles, or the contents of the tables. 

  • Do not overuse tables. Presentation of a few facts will take up less space in the text than in one table. In particular, do not use a table for presenting simple word lists.

  • Bar graphs and pie charts should only be used where absolutely indicated and should be provided in colour, where possible the information should be presented in table format.

  • Use the active voice in the Abstract, Introduction, and Discussion; it is shorter, clearer, and more emphatic. The passive voice requires more words, suggests lack of conviction, extends reading time, and may be ambiguous. The passive voice, by convention, is appropriate in the Methods and Results.

  • Discussion. Emphasise the new and important aspects of the study and the conclusions that follow from them. Do not repeat in detail data or other material given in the introduction or the result sections. Include in the discussion section the implications of the findings and their limitations, including implications for future research. Relate the observations to other relevant studies.

  • Link the conclusions with the goals of the study but avoid unqualified statements and conclusions not completely supported by the data. In particular, authors should avoid making statements on economic benefits and costs unless their manuscript includes economic data and analyses. Avoid claiming priority and alluding to work that has not been completed. State new hypotheses when warranted, but clearly label them as such. Recommendations, when appropriate, may be included.

  • References should be numbered consecutively in the order in which they are first mentioned in the text, not alphabetically. List all authors when there are six or fewer; when there are seven or more, list only the first six and add “et al”.  All references must be cited in the text or tables. Unpublished data and personal communications are not accepted as references.

  • Avoid constructions that force the reader to stop and re-read the sentence, such as “respectively”. Example:“The mean values for men and women were x and y, respectively.” Substitute:“The mean value for men was x, and for women, y.” This version is direct and permits the reader to proceed.

  • The “cause(s) of bad writing are many”; this popular construction also stops the reader abruptly for the sake of supposed precision.  Use either the singular or plural, but not both.  Do not use “and/or”.  Your meaning is usually conveyed by “or” alone.  If necessary, add “or both” at the end of the phrase (“Subarachnoid haemorrhage can cause headache, stiff neck, or both”.)

  • Restrict the word “parameter” to its original mathematical definition; use the more specific “range”, “measurement,” or “variable” instead.  “Practice parameters” (clinical practice guidelines) is an allowable exception.  MRI or radiographic measurement factors (constants) are “parameters” and can be described as such.

  • “Incidence” and “prevalence” should have population denominators; otherwise, the correct terms, all synonymous, are “relative frequency,” “frequency,” or “percentage”.  A “mortality rate” also requires a population denominator and a time interval; deaths among a series of patients would provide a “case fatality ratio” and not a “mortality rate”.

  • Words and phrases that should be deleted on sight: 

    1. arguably (confusing)

    2. needless to say (unnecessary; just say it)

    3. recent (does it mean last week, month, year, or decade?)

    4. significant (except if it implies a statistical difference)

    5. “it…that” constructions, such as “it is a fact that”,  “it is apparent that” (use “apparently”), “it is believed that”,  “it is clear that (use “clearly”), “it is emphasised that”,  “it is generally believed that” (use “many think”), “it is known that”, “it is of interest that”,  “it is often the case” that (use “often”), “it is possible that (use “may”), “it is recognised that”, “it is shown that”, “it may be noted that”, “it should be noted that (use “note that”)  

  • Avoid using these words and phrases and substitute them with the alternatives shown:  

Instead of:

Use

Case

patient

Male or female

man or woman

Male or female children

boys or girls

Paediatric population

children

a total of 100 patients

100 patients

Consensus of opinion

consensus

Control groups

controls

CT of the brain

brain CT

Disease process

disease

end result

result

in my personal opinion

in my opinion

Reported in the literature

reported

Done

carried out

Commonest

the most common

 

Why articles may be rejected 

Professor David Phillips (University of Oxford), former Editor of the Oxford Review of Education, has offered the following explanations of why articles are rejected:  

  • The article is not ready; it is only a draft.

  • The article is too parochial (it will not appeal to a wider, international audience).

  • The article is written in poor English (if English is not your first language, seek help).

  • The manuscript is poorly prepared.

  • The article is too short or too long (check the article length specified in author guidelines).

  • The article has been submitted to the wrong journal (the material will not be relevant to the readers - check the aims and scope of a journal before submitting to it).

  • Nothing new is stated or found.

  • The article is under-theorised.

  • The article is under-contextualised.

  • It is not properly a journal article and would be better suited to another form of publication. 

 

A number of articles and websites provide detailed guidelines and advice about writing and submitting scientific papers. Some suggested sources are:  

  • The classic book - Elements of Style - by William J. Strunk, Jr (Humphrey, New York, 1918) is now published by Bartleby.com (New York, 1999) and is freely available on the web in a searchable format.