Some Tips for Writing Research Papers

Some Tips for Writing Research Papers

WHAT A SCIENTIFIC PAPER SHOULD COMMUNICATE

• Why did you perform the study? (Introduction)
• What is the research question? (Introduction)
• What did you do? (Methods)
• What did you find? (Results)
• What do your results mean? (Discussion)

TIPS FOR WRITING A BETTER PAPER

Avoid cutting a single study into a large number of small parts.

Discuss authorship early and candidly.

Invest time and energy in the planning stage

Consult with experts and a statistician

Know the relevant literature

Don’t attempt to make more of your results than they deserve.

Select a target journal before you write the paper.

Plan to submit a paper that is shorter than that journal’s average.

Write in the active voice, not passive:

The patients were treated… vs We treated the patients…

THE TITLE OF THE PAPER

An accurate promise of the paper’s content

Specific concerning the scope of the study

Indicates study design

States subject (not conclusion) of article

Avoids abbreviations and acronyms

Simple, short, and concise (10-12 words)

Interesting; eye-catching; “reader-grabber”

Easy to understand

ABSTRACT

Adhere to the Journal’s instructions.

Keep it short.

Write it last. Rewrite and polish it.

Keep conclusions specific and conservative.

Data and conclusions must agree with the body of the paper.

INTRODUCTION

Provides adequate background information

Defines any new, unusual, or vague terms

Points out gap in current scientific knowledge

and explains why this study fills that gap

Clearly states the purpose of the study

Should be short

METHODS

Simplest section to write

 (Could be written before data collected)

Must be complete and accurate

Reader must be able to replicate study

Statistics should be easy to understand

No results in this section!

RESULTS

Start with the major positive findings

Include a table describing the study patients

Present the results in a logical order

Do not repeat in detail information that is given in the tables and figures

Report the results in the target journal’s format

Describe people sensitively and diplomatically; avoid pejorative & outdated terms

Make a long results section easier to read by using subheadings

Include only results in the results section—no additional methods, no discussion

References should not be required in this section

Do not use more than the average number of tables and figures in similar articles

in the target journal

PRESENTING STATISTICAL INFORMATION

Report relative risk and 95% confidence intervals

Use statistical terms correctly (e.g. significance)

Provide exact p values; do not use “NS”

TABLES

Simple, self-explanatory

In format of target journal

Not a repetition of the text

Double-spaced

Units for every variable

Exact p values

Appropriate rounding

Format consistent with other tables

FIGURES

Use to illustrate the major points

Label axes and other elements clearly

Edit or redesign graphs from slide or poster presentations

Use style parallel with others in field

Thick lines; large text

Information that is not included in text

Axis labels clear, easy to understand

Units, symbols in target journal’s style

Clear, detailed legend

Should be self-explanatory

DISCUSSION

Start with your most important point

Present no new data in this section

Focus on the implications of your results

Stick to the subject; keep it focused

Compare your study with previous studies

Discuss its weaknesses and deficiencies

Discuss alternative explanations for the results

Write clearly and in plain English

Keep this section as short as possible

CONCLUSIONS

Provide cautious conclusions that are fully supported by your data.

Answer the question, “Who cares?”

Limit conclusions to the boundaries of the study.

Describe precisely what future research is needed.

REFERENCES

In target journal’s style

Double-spaced

Proof-read; accuracy confirmed

BEFORE SUBMITTING THE MANUSCRIPT

Have it peer reviewed internally.

Read the journal’s instructions carefully and follow them to the letter.

Shorten the Introduction and Discussion;

Lengthen the Methods and Results

Eliminate all jargon

HOW TO RESPOND TO PEER REVIEWERS

Write a detailed cover letter to the Editor with your revision.

Thank the reviewers, praise their insight, and don’t be nasty.

Deal point-by-point with every issue raised by the reviewers.

Don’t just respond with a point-by-point rebuttal.

D J Pierson 7/07

Editors’ and Reviewers’ Most Common Criticisms of Manuscripts*

Study designInterpretation of findings

Poor experimental designErroneous/unsupported conclusions

Vague/inadequate method descriptionConclusions disproportionate to results

Methods lacking sufficient rigorDesign does not support inferences

Failure to account for confoundersInadequate link of findings to practice

No control, or improper controlUncritical acceptance of statistical results

No hypothesisFailure to consider alternative explanations

Biased protocolUnexplained inconsistencies

Small sample sizeInadequate discussion

Inappropriate statistical methods, or Inflation of the importance of findings

statistics not applied properlyInterpretation not concordant with the data

Importance of the topicPresentation of the results

Rehash of established factsToo long, verbose

Insignificant research questionExcessively self-promotional

Irrelevant or unimportant topicPoor grammar, syntax, or spelling

Low reader interestPoor organization

Little clinical relevancePoorly written abstract

Not generalizableFailure to communicate clearly

*Paraphrased from Byrne DW. Publishing your medical research paper.

Baltimore, Williams & Wilkins, 1998:48-50.

Resources

Byrne DW. Publishing your medical research paper. What they don’t teach in medical school. Baltimore, Williams & Wilkins, 1998. Good practical resource; $26 paperback.

Special Issue: Research and Publication in Respiratory Care. October, 2004 issue of Respir Care (2004;49[10]:1145-152). 18 articles covering many aspects of writing articles and successfully navigating peer review; most are general rather than specifically about respiratory care. All articles are available free as PDF files at the Journal’s website:

or via PubMed.

Strunk W Jr, White EB. The elements of style. Longman, 4th edition, 2000. also, “The elements of style, illustrated,” 2005. [The classic; concise, accessible, and correct]

Style Manual Committee (Huth EJ, Chair), Council of Biology Editors. Scientific style and format: the CBE manual for authors, editors, and publishers. New York, Cambridge University Press, 6th edition, 1994. (new edition due soon)

American Medical Association manual of style. A guide for authors and editors. Hagerstown, MD, Lippincott Williams & Wilkins, 9th edition, 1998.

Chicago Manual of Style. Chicago, University of Chicago Press, 15th Ed., 2003.

D J Pierson 7/07

OVERCOMING WRITER’S BLOCK

The following guide is from William M Tierney MD, Indiana Univ School of Medicine, Editor of JGIM, via the World Association of Medical Editors (

TITLE PAGE
Write the paper’s title below:
List the study’s authors (one to a line, including middle initials and highest degrees):
List the paper’s authors’ institutions (in the same order as the authors):
INTRODUCTION
Provide a brief discussion of the general topic: Why is it important?
Provide a brief discussion of prior work by you and/or others:
Provide an explicit statement of your study question/hypothesis:
METHODS
Describe your study site:
Describe your study population (source, inclusion and exclusion criteria):
Describe your recruitment methods in detail:
Describe your intervention (if an interventional study):
Describe the data you collected and how you collected them:
Describe your data analysis in detail (dependent variables, independent variables, comparisons, primary and secondary analyses, statistical methods used, p-value accepted as significant, etc.):
RESULTS
Describe your subjects: numbers approached, enrolled, excluded, characteristics (do not repeat table data—describe the table data in qualitative terms, where possible)
Describe main analysis results (again, do not repeat table data):
Describe secondary analysis results (again, do not repeat table data):
DISCUSSION
Write down the most important take-home point you want the reader to remember. Do not merely repeat the results. Then provide commentary based on what prior relevant studies have found.
Write down the second most important take-home point and discuss it.
Write down the third most important take-home point and discuss it. (Some papers will not have 3 take-home points.)
List and discuss the study’s limitations:
Write down your conclusions (usually a repeat of take-home point #1):
Give future directions (often the next study you want to do following this one):

D J Pierson 7/07