-
How to write a scientific paperSorin Barac, MD PhDDivision of
MicrosurgeryVictor Babes University of Medicine and Pharmacy,
Timisoara
-
Scientific papers = important though poorly understood method of
publication. important: you cannot get money from the govt/univ
poorly understood: they are not written very well1.
Introduction
-
Example: the introduction is supposed to introduce the reader to
the subject, so that the paper will be comprehensible even if the
reader has not done any work in the field.1. Introduction
-
The real purpose of introductions, of course, is to cite - your
own work - the work of your advisor - the work of your spouse - the
work of a friend from college- the work of someone you've never
met, IF your name is on the paperNote! citations should not be
limited to refereed journal articles, but should also include
conference proceedings, and other published or unpublished work.1.
Introduction
-
At the end of the introduction: you must summarize the paper by
reciting the section headings.
1. IntroductionIn this paper, we discuss: scientific research
(R) (section 2), scientific writing (section 3), scientific
publication (section 4), and draw some conclusions (section 5).
-
The purpose of science = to get paid for doing fun* stuff2.
Scientific research*young and bright people are curios as how
things work, they are attracted to exploration
-
science = discovering something new about the Universe2.
Scientific researchbut this isn't really necessarywhat is really
necessary is a grant
-
Rules for obtaining a grant- your R will discover something
incredibly fundamental- you are the best person to do this
particular Rso you should cite yourself both early and often cite
other papers as well, if you are on the author list2. Scientific
research
-
Once you get the grant, your university, company, or government
agency will immediately take 30 to 70% of it so that they can: -
heat the building- pay for Internet connections, and - purchase
large yachts2. Scientific research
-
Now it's time for the actual R. You will quickly find out that
(a) your project is not as simple as you thought it would be and
(b) you can't actually solve the problem. 2. Scientific
researchHowever -- and this is very important you must publish
anyway!!!
-
You've spent years on a project and have finally discovered you
can't solve the problem you set out to solve. Nonetheless, you have
a responsibility to present your R to the scientific community. Be
aware! - () results can be just as important as (+) results- if you
don't publish enough you'll never be able to stay in science.3.
Scientific writing
-
Prepare to write a publishable paper1. Organize your
materialCover letterCopyright transfer pageTitle pageAbstract and
key wordsText - IntroductionMaterials and methodsResultsDiscussion,
with conclusionsAcknowledgementReferencesTables Figure legendsCopy
of each figure
-
Writing1. TitleGive a snappy title: - interesting, easy to
understand, simple, concise - short- indicate study design- begin
with a key wordProvide the species of animals used in the study
-
Writing2. AbstractTake time to polish the abstract - make a good
impression with a well written abstract- most people read only your
abstract! - structure it: background; methods, results, conlusions-
clearly describe the problem in the first sentence, begin with a
key word- indicate study design- avoid using the same sentences in
abstract/body of the paper Keep it shortProvide the species of
animals used in the study
-
Writing3. IntroductionBegin with thunder - write an introductory
reader grabber: a provocative question, a new perspective, a good
quotation- explain why your study is necessary Provide adequate
background information- start with a general, yet concise
description of the problem - reference previous work that supports
your ideas - define all potentially questionable terms
-
Writing3. IntroductionArticulate the purpose of your study- the
aim and hypothesis should b easy to find and understand- provide a
map showing the direction of your studyRemember! A good paper has
one main problem to solve, not 2 or 3
-
Writing4. MethodsReplicability of results is the heart of
science- present clear, but detailed description of research design
Organize the methods- study design - describe it carefully-
eligibility- randomisation, blinding- intervention and compliance-
statistical analysisDescribe your data collection method- who,
what, when, where, how, why
-
Writing4. MethodsEligibility- describe the - source of the study
subjects - inclusion/exclusion criteria- provide the beginning /
ending dates of the study- describe the informed consent
process
-
Writing4. MethodsIntervention and compliance- describe medical
devices, dosage, route of administration, - include the trade,
name, and the manufacturers name, location in parentheses- provide
evaluation of the tests- diagnostic tests - determine the presence
of disease- screening tests - detect individuals that need more
testing- prognostic tests - predict the outcome of disease
-
Writing4. MethodsSample size- description of the sample size
calculations is CRUCIAL Statistical analysis- describe the
statistical analysis carefully- provide reproducible details of
your statistical methods- explain why did you choose specific
tests
-
Writing5. ResultsOrganize the results- present your results
enthusiastically - remember Archimedes reaction- summarize the
similar types of numbers in tables, refer the reader to them and
shorten the text the ResultsPresent the results in a natural
orderStart your Results section with the major positive findings,
report negative associations at the end
-
Writing5. Results: Use statistical terms skillfully Significant
= probably caused by something other that mere chance -
statistically significant - probability of the finding to occur by
chance is less than 5% (P < 0.05).Random = each element in a set
has an equal probability of occurrence (not unplanned, incidental)
A sample = a finite part of a statistical population whose
properties are studied to gain information about the wholeA random
sample = a chance selection in which all members of the base
population presumably have the same chance of being selected
-
Writing5. Results: Describe people sensitively and
diplomatically Subject, individual, caseMan, woman, patient (pt),
participant45 males45 male patients, 45 men67 females67 female
patients, 67 womenHad surgeryUnderwent surgeryPatients with
complicationsPt who experience complicationsEpileptics People with
epilepsyMental disorderImpaired cognitive functionIn 43 pt used as
controlsFor 43 pt who served as controlsMan and wife Husband and
wife, man and womanOrientalsAsian people
-
Writing5. Results: Tables Checklist for creating a high quality
tableSimple and self-explanatoryNot a repetition of the textDouble
spacedUnits provided for each variable values are either SD or SE
of the meanExact P values included No vertical lines
-
Writing5. Results: Figures 1. Use graphs to illustrate only the
major points, particularly those that cannot be easily expressed in
the text or tables2. Obtain written permission for any figures or
tables that you borrow from published work
-
Writing5. Results: Figures Checklist for creating a high quality
figureThick linesLarge textExact P values included Clear, detailed
legendInformation that is not included in the textSelf
explanatory
-
Writing5. Discussion Start the discussion with your most
important pointYou should start with one sentence that clearly
shows that your paper contains new information: We found that Then
describe your findings and explain their importance The Discussion
is the place to discuss the implications of your findings - not
simply repeat them Do not discuss any data that you did not present
in the Results (that your study did not generated)
-
Writing5. Discussion Describe the NEW INFORMATION your paper
providesCompare your study with previous studies Discuss how your
results compare with landmark papersWhen your conclusions differ -
explain whySay what you mean, mean what you say, but dont say it
mean.
-
Writing5. Discussion Be skeptical of published workDo not fail
to consider the null hypothesisMost people - especially younger
doctors and students - assume that any published report is true and
unconsciously add their name to the author list It is safer to
assume that the author was wrong and see if he or she can overcome
your skepticismEx: are conclusions based on recent data?, a large
sample size?, are data drawn from a population appropriate for your
needs?
-
Writing5. ConclusionsConclude with a Bolt of lightningWhat do
you recommend?Provide conclusions that are fully supported by your
dataAre the conclusions correctly derived from the data
presented?Are the recommendations too general?
-
Writing5. ConclusionsAnswer the question: Who cares? What are
the clinical applications of your findings?Describe precisely what
further research is neededIf you recommend additional research
explain whyEx: a larger sample is needed to permit more
statistically significant conclusions - explain why you paper makes
a contribution despite this problem
-
Writing5. ReferencesUse full length articles from peer reviewed
journals Abstracts are not allowed Paper submitted but not accepted
are not allowed Limit references to key citations (20-40)Place
reference citations properly- Previous reports have shown an
incidence of 50%12,24,29
-
Writing5. Referencing systemsCitation-Order,
Citation-by-Reference, or Vancouver SystemAuthor-Date,
Name-and-Year, or Harvard SystemAlphabet Number SystemFormat
references correctly (see abbreviations)
-
While writing a scientific paper, the most important thing to
remember is that the word "which" should almost never be used. Be
sure to spend at least 50% of your time (i.e., 12 hours a day)
typesetting the paper so that all the tables look nice. 3.
Scientific writing
-
You've written the paper, and now it's time to submit it to a
scientific journal. The journal editor will pick the referee most
likely to be offended by your paper at least the referee will read
it and get a report back within the lifetime of the editor.
Referees who don't care about a paper have a tendency to leave
manuscripts under a growing pile of paper until the floor
collapses, killing the 27 students who share the office below.4.
Scientific publishing
-
Be aware! every scientific paper contains serious errors. If
your errors are not caught before publication, you'll eventually
have to write an erratum to the paper explaining (a) how and why
you messed up and (b) that even though your experimental results
are now totally different, your conclusions need not be changed. 4.
Scientific publishing
-
Errata can be good for your career. They are easy to write, and
the convention is to reference them as if they were real papers,
leading the casual reader--and perhaps also the Science Citation
Index--to think that you've published more papers than you really
have . 4. Scientific publishing
-
The conclusion section is very easy to write: - take your
abstract and change the tense from present to past. - mention at
least one relevant theory only in the abstract and conclusion. By
doing this, you don't have to say why your experiment does (or does
not) agree with the theory, you merely have to state that it does
(or does not).5. Conclusions
-
We (meaning I) presented observations on the scientific
publishing process which (meaning that) are important and timely in
that unless I have more published papers soon, I will never get
another job. These observations are consistent with the theory that
1. it is difficult to do good science2. write good scientific
papers, and 3. have enough publications to get future jobs.5.
Conclusions
-
Thank you
*******The grant agency must also believe that*The grant agency
must also believe that*The grant agency must also believe that*The
grant agency must also believe that*The body of a good paper is a
thunderbolt in reverse*Otherwise the reviewer might agree with you,
send the manuscript back and encourage you to enlarge your sample
*Otherwise the reviewer might agree with you, send the manuscript
back and encourage you to enlarge your sample *Otherwise the
reviewer might agree with you, send the manuscript back and
encourage you to enlarge your sample *The grant agency must also
believe that*The grant agency must also believe that*The grant
agency must also believe that*The grant agency must also believe
that*all you have to do is toIt's considered good form to*The grant
agency must also believe that