“How to write a scientific paper”код для вставки
Scuola di Dottorato in Scienze Ingegneria Medicina Seminario metodologico (Seminar on Professional Skills) led by Marina Bentivoglio and Luca ViganГІ 30 June 2011 вЂњHow to write a scientific paperвЂќ For those who have never published yet a scientific paper: вЂў Did you start writing one? If you did: which difficulties did you encounter? вЂў Do you regularly read scientific articles in your field? If so: what do you think is especially difficult to replicate in such writing? Key facts вЂў вЂњverba volant, scripta manentвЂќ вЂў A scientific paper is the way to communicate/deliver permanently your data to the scientific community вЂў Your only richness, in the scientific community, is your name and your reputation: use it properly вЂў The community to which you address your work did not discuss with you (data/concepts) previously, did not participate in the study planning and data collection and analysis (did not participate in your failures, torment, discussions with your supervisor) вЂў Scientific papers should be written in English (вЂњscientific EnglishвЂќ, which is some kind of вЂњjargonвЂќ and has, however very precise rules) вЂў This is a disadvantage if English is not your mother tongue вЂў Is the scientific community вЂњindulgentвЂќ concerning this? Absolutely not! вЂўTherefore one should get organized. What to do? п‚§Write your text in your mother tongue and find a translator? No! п‚§What else? Learn how to write your paper in English: it is feasible and mandatory! вЂў Selection of the journal: вЂњtop journalвЂќ, peer-reviewed indexed journal, non-indexed journals (no IF), on-line publications Format (in biomedical literature/life sciences): Presenting your вЂњprimaryвЂќ (unpublished) data (highest вЂњfilterвЂќof reviewers): вЂўShort communication вЂўOriginal research article Reviewing a topic (including your own data): вЂўReview article вЂўComments, opinions, вЂњlettersвЂќ вЂўChapters in volumes, conference/symposium proceedings вЂў Computer science: different targets Classical publication sequence in computer science: вЂў Workshop вЂў Conference вЂў Journal Paper categories вЂў Original work вЂў Survey (only/mainly for journal) вЂњFORMAL вЂќ interesting data, but is presented вЂў An article that contains in a вЂњsloppyвЂќ form (typos, mispelled words, bad or excessive use of abbreviations, etc) gives to reviewers (who represent the first filter) a bad impression: the editing should be professional! Editing tips: - Read carefully the Instructions for Authors of the journal - Select carefully the key words (words in the title are automatically indexed) - Use of acronyms: too many abbreviations may make the text difficult to read; a term can be abbreviated only if used more than 3 times and should be abbreviated when first mentioned (and therein should no longer be used in extenso) Editing tips: - Figures should be numbered consecutively and according to citation in text; figures have legends - Tables have a title and their own numbering (no legend) -Obviously, all refs quoted in text should be listed in the reference list and vice versa. Scientists have in general little time, many of them are under big pressure (grants, career, etc.) The article should be an вЂњappealingвЂќ and вЂњuser-friendlyвЂќ product: вЂў very clear вЂўnot verbose or repetitious, should be вЂњeffectiveвЂќ вЂўnicely illustrated with well readable figures (graphics, formulae, etc.: avoid too small labels and lettering; organize properly space avoiding blank space; evaluate the image reduction for paper formatting/ printing, etc.) What is вЂњappealingвЂќ? вЂў вЂњNoveltyвЂќ: вЂў if вЂњthese observations are here reported for the first timeвЂќ: specify it clearly (and, of course, make sure that it is true!) вЂў if the methodological approach is novel: present your approach properly and explain its advantages (as well as its limitations) вЂў вЂњImpactвЂќ on knoweldge: вЂў Explain the impact of your findings on current knowledge/interpretation: explain it clearly (no over-statements, no arrogance!) Start raising these issues in the Introduction, but do properly highlight them in the Discussion (and in the Abstract) вЂњSUBSTANTIALвЂќ (LIFE SCIENCES) вЂў Title вЂў Authors and affiliation/s вЂў Key words Essential for indexing: think about it (words in title are automatically indexed) вЂў Abstract Body text: вЂў Introduction вЂў Material and Methods вЂў Results вЂў Discussion (including Conclusions) вЂў Acknowledgements вЂў References вЂў Figure legends вЂў For an вЂњoriginal research articleвЂќ (different, of course, for review articles); вЂў Format/section subdivision may vary, but all these parts are anyhow included вЂњSUBSTANTIALвЂќ (COMPUTER SCIENCE) вЂў Title вЂў Authors and affiliation/s вЂў Abstract вЂўKey words Body text: вЂў Introduction вЂўBasic definitions (background, pre-existing defs) вЂў Theoretical results (new defs, theorems,вЂ¦) вЂў Experimental results (data and their analysis) вЂў Discussion вЂў Acknowledgements вЂў References вЂў [Appendix] Title: вЂўProvides the first impression to the reader and influences whether a reader is interested in reading the manuscript (while doing an e-search, the title determines whether who does the search will go ahead and read the abstractвЂ¦. and then the paper) вЂўAvoid long titles вЂўAvoid titles which begin with redundant words such as вЂњA study ofвЂ¦вЂќ Who is an Author of an abstract/scientific publication? Authorship should be based on a substantial intellectual contribution! (can be reflected in the order of names) International Committee of Medical Journal Editors: definition of authorship as being based on: 1) Substantial contribution to conception and design, acquisition of data, or analysis and interpretation of data 2) Drafting the article or revising it critically for important intellectual content 3) Final approval of the version to be published (NB all authors are equally responsible and should approve the version to be submitted) Acquisition of funding or general supervision of the research group alone does not constitute authorship... Many journals now request specification of the contribution of each Author For example: Author Contributions Conceived and designed the experiments: M.B., K.K., A.K.N. Performed the experiments: A.K.N., A.A., J.Y.F., D.M. N., J-J.M. Analyzed the data: A.K.N., S.P., P.F.S.E. Wrote the paper: M.B., A.K.N., K.K., P.F.S.E.вЂќ ABSTRACT вЂў Limited length (every word is precious!) вЂў Very important: openly accessible вЂў It is the first text part encountered by the reviewer (and by the reader); it could be the last part to write, when the main вЂњtake homeвЂќ message and its вЂњnovelty and impactвЂќ are fully clear to you Sometimes reviewers are picked based on the abstract вЂў Should contain, very synthetically, all the elements of the study: rationale (or motivation), methods, main results, conclusions and вЂњtake-home messageвЂќ ABSTRACT вЂў One or two sentences (at the beginning) of background information вЂў The question/s asked вЂў What was done to answer the question/s вЂў Findings that answer the question/s (in a logical order) вЂў The answer to the question/s It must make sense when read in isolation for those who only read the abstract (NB e-searchers may only retrieve the abstract!) It must also provide a clear and accurate recapitulation of the manuscript for readers who read the entire manuscript (must not contain data which are not included in the Results) (From: Sue Jenkins, 1994 вЂ“ and still valid!) INTRODUCTION вЂў вЂњSets the stageвЂќ вЂўPresent very clearly the rationale for the study (essential), in the framework of current knowledge (NB but should not be a вЂњdiscussionвЂќ, should be an вЂњintroductionвЂќ) вЂўIntroduce the methods you have adopted to solve your problem and start emphasizing the original contribution/s of the study INTRODUCTION Background to the topic (past verbe tense) вЂў What is known or believed about the topic вЂў What is still unknown or problematic вЂў Findings of relevant studies (past verbe tense) вЂў Importance of the topic Statement of the research question e.g.: вЂњTo determine whetherвЂ¦вЂќ вЂњThe purpose of this study was toвЂ¦вЂќ вЂњThis study tested the hypothesis thatвЂ¦вЂќ вЂњThis study was undertaken toвЂ¦вЂќ Approach taken to answer the question (past verbe tense) (From: Sue Jenkins, 1994 вЂ“ and still valid!) MATERIAL AND METHODS вЂў NB: your work should be вЂњreproducibleвЂќ (replication of the study is a requisite of a scientific work and its richness!) вЂў Specify all the sources of reagents: manufacturer/company, city, country; website for softwares, etc; the providing colleague/laboratory (kindly supplied byвЂ¦) вЂў Remember ethical statements related to animal experiments (mandatory) RESULTS (LIFE SCIENCES) вЂў Should be precise, complete, synthetic вЂў Pay attention (throughout the text) to use/selection of words: вЂњshow, indicate, suggest, demonstrateвЂќ вЂњobvious, evident, important*вЂќ, etc. (*what would really be вЂњnot importantвЂќ?) вЂў Guide the reader in identifying the data you consider key findings RESULTS (COMPUTER SCIENCE) вЂў Should be precise, complete, synthetic вЂў Definitions: define each notion before use вЂў Theorems: proof must be understandable/convincing (reader should be able to follow all the steps) вЂў Algorithms: readable, appropriate level of abstraction вЂў Pay attention (throughout the text) to use/selection of words: вЂњshow, indicate, suggest, demonstrateвЂќ вЂњobvious, evident, importantвЂќ, etc. вЂў Experimental data: HW, SW, cpu time, memory, program options/parameters, choice of algorithms/strategies, input data NB: data should be interpreted/analyzed (not just reported) DISCUSSION вЂў It could help to start with a вЂњsummaryвЂќ of results/contributions (in the order in which they are then discussed) вЂў Divide into subsections (possibly according to logical reasoning) вЂў DO NOT REPEAT THE RESULTS (or the Introduction) вЂў Discuss your data comparing them with previously published data Computer science: compare with related works вЂў AVOID SPECULATIONS, STICK TO FACTS вЂў End, possibly, with вЂњConclusionsвЂќ or вЂњConcluding remarksвЂќ in which the study impact and the вЂњtake-home messageвЂќ should be very clear. Computer science: also directions for future research, open problems Life sciences: the same, making it clear in which вЂњdirections for future researchвЂќ are your own findings inserted DISCUSSION To remember: вЂў Indication of the novelty/impact/relevance/implications of the study (originality/uniqueness of the work), i.e., the priority of the study вЂў Explanations of: - Any results that do not support the answer/s to your question/s - How the findings concur with those of others - Any discrepancies of the results with those of others - Eventual unexpected findings - The methodological limitations of the study which may affect the study validity and/or generalization of the study findings ACKNOWLEDGEMENTS вЂўGrant/Funding support (all details: grant number) вЂў All help should be acknowledged (be generous in thanking!): technical advice/assistance/support, help in the preparation of the manuscript, etc.) Tip When you start writing a paper, try to finish it within a relatively short period of time, not to loose concentration and contact with the pertinent literature (which keeps being publishedвЂ¦) But when you think that the paper is ready, let it вЂњrestвЂќ for at least a couple of days and read it again with вЂњnew eyesвЂќвЂ¦вЂ¦ Reviewers: Who are they? Friends, enemiesвЂ¦? Young, old? Lots of information / guidelines / manuals are available, e.g. вЂўhttp://lemire.me/blog/rules-to-write-a-good-researchpaper/ вЂўhttp://www.findaphd.com/student/study/study-6.asp вЂўWilliam Strunk, Jr. and E.B. White вЂ“ The elements of style вЂў http://en.wikipedia.org/wiki/The_Elements_of_S tyle вЂў http://www.bartleby.com/141/ вЂў http://www.crockford.com/wrrrld/style.html вЂўetc. For those who have already published a scientific paper: Which difficulties did you encounter in writing - and then getting accepted for publicationyour paper/s?