如何撰寫高品質之期刊論文 李孟智教授中山醫學大學醫學研究所. 研究者歷經產生問題 假說之形成 申請研究計畫 執行研究 資料處理與分析之外 最重要的就是將論文發表 (Publication) 於期刊雜誌, 除了可與同儕切磋分享及個人和單位之得分外, 並可轉化為實務應用, 達成研究三贏之最高效益

Transcription

1 如何撰寫高品質之期刊論文 李孟智教授中山醫學大學醫學研究所 前 言 研究者歷經產生問題 假說之形成 申請研究計畫 執行研究 資料處理與分析之外 最重要的就是將論文發表 (Publication) 於期刊雜誌, 除了可與同儕切磋分享及個人和單位之得分外, 並可轉化為實務應用, 達成研究三贏之最高效益

2 精進研究 靈性成長 戰略層面戰術層面戰鬥層面 科技來自人文情緒與動機決勝 平安喜樂苦集滅道 心安 有愛提得起放得下 突破罩門 逆風而行從善如流聞過則喜 發問 尋求幫助 組織型學習與成果分享 多聽 多講 多問 多看多元學習 終身不懈的學習 服膺真理攀登高峯預測漁場 縱向與橫向學習 拓展及深入與研究主題的相關知識 研究流程 美國霍普金斯大學公衛學院 發現及確認問題概念形成文獻搜尋及建立理論架構假說形成建立研究架構撰寫研究計畫執行計畫資料處理及分析撰寫報告及論文投稿及發表研究結果之應用

3 發現問題 來世為何 行善為何 批判性思考 文獻查證 確認問題 人死了後會怎樣 行善與好報有關嗎 既有的理論運用同儕互動 形成概念 今生與來世 利他與利己 專家諮詢參與式研究 / 質性研究 概念架構 ( 心靈的視窗 ) 現在種種譬如明日生 人 行為 結果 先驅研究 研究假設 好人上天堂 好人有好報 研究架構 略 略 統計驗證 無法驗證 結束 多次測量 / 多變項分析 修正研究架構 好人大部份都有好報 發展模式或理論 行善福雖未至禍已遠離 Decide to Which Journal you will Submit University of Illinois -- Beverly Henry, PhD Start early when your proposal is approved Write for a specific journal - Match editor s interest - Consider acceptance rate - Consider the time-line Follow author guidelines Read editorials and articles

4 Choosing a Journal --AAFP Select the journal before you write the paper or plan the project Considerations: - Topic - Format - Audience - Prestige - Time to print - Acceptance rate

5

6 如何選擇研究題目 如何選擇研究題目 : 1. 瀏覽該領域的相關期刊 ( 主題分類或學門分類 ), 找到對味的期刊 2. 將近三年該期刊全部搬出來瀏覽目錄, 找到看得懂或有感覺的主題 3. 選擇三四個主題瀏覽摘要, 看看研究目的 ( 假說 ) 與方法 ( 資料來源 ) 4. 影印兩三篇類似假說與資料來源 ( 譬如死因統計 or 個案對照研究 ) 的論文 5. 仔細研讀這幾篇論文並在 references 中挑出相類似的文現再仔細閱讀 6. 挑出一兩篇幾乎可以彷做的論文, 以 PubMed 的 single citation 找出 related articles 7. 題目要在反覆來回於研讀論文 找尋資料與分析資料過程中, 不斷修改精進 構思讓題目更具原創性及趣味性 基本上, 博士論文對研究的過程及結果均須要求 Essential for titles Concise and precise Informative and descriptive Not misleading or unrepresentative Specific for example, include type of study and numbers(if large) Words appropriate for classification Interesting not dull

7 Developing a title in four steps ( lilleyman,1998) An epidemiological geographically based study of the quantity and effects of ionizing radiation received by male employees of a nuclear reprocessing plant and male residents working elsewhere in children of nuclear workers only. An epidemiological study of the links between the radiation received by male employees of a nuclear reprocessing plant and other local residents and childhood leukemia. Relation between working at and living near a nuclear reprocessing plant and childhood leukemia. Nuclear reprocessing, radiation exposure, and childhood leukemia: an epidemiological study. Title of Manuscript New England Journal of Medicine Concise and descriptive (not declarative) Lancet Concise but informative Annals of internal Medicine As brief as possible while conveying essential features of he article s content British Medical Journal Keep them concise

8 摘要 (Abstract) 誘發讀者閱讀本文的興趣在短時間內能讓讀者明白本文之主要結果格式化的摘要撰寫為當今的趨勢 1. 目的 5. 介入方法 / 主要暴露 2. 研究設計 6. 主要結果測量 3. 研究單位 7. 結果 4. 研究對象 8. 結論 前言 (Introduction) 1. 精簡 ( 不超過一頁半 ): 含背景 問題及 尚未解決部分 2. 一篇論文一個目的 3. 可提出該問題 / 解決方法之重要性

9 方法 (Methods) 1. 方法可重覆性 2. 自變項及應變項明確定義 3. 測量之信度及效度 4. 採樣方法有無偏差 5. 樣本大小 : 足夠的檢力 (Power to detect differences) 1. 統計分析方法之適當性 2. 干擾 (confounders) 之控制 What to include in the methods section-1 How the study was designed Keep the description brief Say how randomization was done Use names to identify parts of study sequence

10 What to include in the methods section-2 How the study was carried out Describe how the participants were recruited and chosen Give reasons for excluding participants Consider mentioning ethical features Give accurate details of materials used Give exact drug dosages Give the exact form of treatment and accessible details of unusual apparatus What to include in the methods section-3 How the data were analysed Use a p-value to disprove the null hypothesis Give an estimate of the power of the study (the likelihood of a false negative the β error) Give the exact tests used for statistical analysis(chosen a priori)

11 A good methods section can answer these questions Does the text describe what question was being asked, what was being tested, and how trustworthy the measurements are? Were the measurements recorded, analysed, and interpreted correctly? Would a suitably qualified reader be able to repeat the experiment in the same way? 1. 回覆率有多少 結果 (Results) 2. 未回覆者 ( 或流失者 ) 如何處理 3. 結果說明層次分明, 平均值要有 S.D. 及勝算比要有 95 % C.I. 4. 結果有無反應研究目的 5. 圖表 : 通常是必要的 應能呈現主要的結果 應一目了然 與文字部份能相應合

12 Discussion: overall format Statement of principal finding(s) Appraisal of methods Comparison with previous work Clinical and scientific implications (if any) Further work Conclusion Acknowledgements 討論 / 結論 (Discussion/Conclusion) 1. 討論主要結果 2. 與以往研究之異同與辨證 3. 推論限制 研究限制有無說明 4. 研究方法之探討及對結果之影響 5. 結論應當明確 6. 建議應當可行

13 文 獻 1. 文獻是否必要, 是否參考最新文獻 2. 有無遺漏主要文獻 3. 有無錯誤地引用 4. 有無引用該雜誌之論文 英文寫作的基本功 多閱讀相同研究之論文多用主動式及簡單式表達 敘事之邏輯較用字遣辭來的重要通常使用三 四句話講清楚一個概念 : 1. 主題句 2. 支持句 3. 轉折句英文寫作的要點在於能將您的論點說清楚並具說服性

14

15 From Polished Product to Printed Page Responsibility: Authorship Reaching your audience: choosing a journal, submitting the paper Review process Revision Realizing your goal: publishing your paper

16

17 Guidelines for author submission Access the website via the URL by using a unique user identifier and password. Enter the author submission area. Choose the Submit a new manuscript link Guidelines for author submission Enter the following basic information: number of authors, type of article, title, manuscript keywords, abstract, cover letter to the editors, author details, suggested reviewers names, and word count. Upload the files of main text, tables and figures

18 Guidelines for author submission The article is converted automatically into a pdf The pdf file size is smaller than standard word processing and image files, and the software required to view I (Acrobat Reader) is a standard piece of software easily obtained free of charge from the web( Guidelines for author submission You then have the opportunity to view your submission before it is submitted formally to the journal. This allow you to make sure that what you are submitting is correct and of peer review standard. Once approved, the article is then considered to be a formal submission.

19 Systems for web-based manuscript submission and peer review Common features The submitting author must begin by registering with the manuscript submission system. This Gives the author a username and password. This step will not need to be repeated for subsequent submissions to the same journal. Only the submitting author has access to the online system, unless he or she shares his username and password with co-authors (not recommended). Submissions are made by filling in web forms and uploading files to the journal s server. Only a standard web browser is required. File uploads journal s server. Only a standard web browser is required. File uploads can be slow, and the web forms sometimes require detailed information about co-authors, ethical approvals, suggested reviewers, and so on. The submission process can be started at on time and completed at another (the system will remember the tasks already completed.) Usually a wide variety of files (e.g. Word documents, jpeg images, tiff images, pdfs) can be uploaded, but there are limitations (e.g. not all systems will accept Excel files) and it is wise to check the instructions for authors before proceeding. Image files can be sent as relatively low-resolution jpeg files to reduce the file size and upload time, but high-resolution copies of images will be required eventually if the article is accepted for publication. The submitting author can track the manuscript s progress through editorial and peer review by logging in to the system. Submission of revisions is generally much simpler than the original submission process, as much of the information required is already on the system. If the article is accepted for publication, hard copies of some documents (such as copyright assignment or permissions) may be required. Systems for web-based manuscript submission and peer review Examples Editorial Manager (Aries Systems)< used by Elsevier Science, Australasian Medical Publishing Company, University of Chicago Press, and others. BioMed Central< : an open-access publisher of over 170 electronic journals. ejournalpress< > :used by Nature Publishing Group, Proceedings of the National Academy of Sciences of the United States of America, Palgrave Macmillan, and others. Bench>Press (Highwire Press) < > :used by BMJ Publishing Group, World Health Organization, and others. Manuscript Central (ScholarOne) <scholarone.com> :used by Blackwell Publishing, Taylor & Francis, Oxford Journals, and others.

20 審查流程 通常副主編為樞紐, 主編為決定者 主編 Editor 編審委員會 Editorial Board 審查人 Reviewers 副主編投稿 Associate Editor 接受 / 修正 / 退稿 Accept/Revise/Reject Thanks for your attention! Wish you a successful submitter.

21 敬請批評指教 The purpose of letter Usual Comment (positive or negative) in response to a previous publication Concise communication of clinical or investigative data Communication of case report(s)

22 The purpose of letter Less common General medical or political comment (e.g. guild issue ) Comment concerning the nature or format of the journal Advertisement of interest to collaborate or to gain access to patients or study material Guidelines for a letter in response to an article Be courteous and interested not rude or dismissive Make specific rather than general comments Give reasoned argument, not biased opinion Do not repeat aspects already covered in the original article Introduce different respective or additional data to the topic Attempt to make only one or a very few specific points Be concise

23 Presentation of a concise report as a letter Introduce the topic Brief explain rationale and objectives of study. Presentation of a concise report as a letter Present methods and results Reference methods as much as possible to reduce length of their description Include only essential data If possible present data in a table and / or figure.

24 Presentation of a concise report as a letter Present conclusions Emphasize only one or a few main conclusions Place findings in context of previous literature Highlight caveats and strengths of study Suggest future studies that are still required in this area. Presentation of a concise report as a letter Avoid extrapolating too far from data. Avoid repetition of data or conclusion. Be Concise

25 How to please an editor Adhere strictly to the Guidelines for Authors throughout the text Do exactly what the Guidelines dictate: no more. No less. Make the editor s life easier. Avoid basic errors such as incorrect numbering of figures or tables; forgetting to attach figures and using the wrong reference format. If invited to submit a revision, attend to every detail raised by the editor and assessors in a structured, unemotional manner. Contact the Editorial Office if you are concerned that your manuscript is not being dealt with efficiently. Communicate courteously and correctly with the Editorial Office, arguing your case coherently and professionally. Never submit a manuscript to more than one journal simultaneously; editors find out, and they loathe the practice. Make sure that all the authors have read and contributed to the manuscript. Would they be willing to stand up in public to defend their work? How to be a Vancouver Group positive author All persons designated as authors should qualify for authorship. Each author should have participated sufficiently in the work to take public responsibility for the content.

26 How to be a Vancouver Group positive author Authorship credit should be based only on substantial contributions to: (1) conception and design or analysis and interpretation of data; (2) drafting the article or revising it critical for important intellectual content ; and (3) final approval of the version to be published. Conditions 1-3 must all be met. Participation solely in the acquisition of an article of funding or the collection of data does not justify authorship. General supervision of the research group is not sufficient for authorship. Any part of an article critical to its main conclusions must be the responsibility of at least one author. How to be a Vancouver Group positive author Editors may ask authors to describe what each contributed; this information may be published. Increasingly, multicentre trials are attributed to a corporate author. All members of the group who are named as authors, either in the authorship position below the title or in a footnote, should fully meet the above criteria for authorship. Group members who do not meet: these criteria should be listed, with their permission, in the acknowledgements or in an appendix.

27 How to be a Vancouver Group positive author The order of authorship should be a joint decision of the co-authors. Because the order is assigned in different ways, its meaning cannot be inferred accurately unless it is stated by the authors. Authors may wish to explain the order of authorship in footnote. In deciding on the order, authors should be aware that many journals limit the number of authors listed in the table of contents and that the US National Library of Medicine(NLM) lists in Medline only the first 24 plus the last author when there are more than 25 authors. An example of contributorship Byline: A,B, C, D, E, F, G, H Contributors: A carried out the trial, helped in data analysis, and wrote the paper. B was involved in design, implementation, and data analysis, and contributed to the writing of the paper. C was involved in execution of the trial, data management and analysis, and quality assurance of the turnip assay. D was involved in trial execution and data entry, management analysis, and data entry, management analysis, and quality assurance, E was involved in trial execution and data management with emphasis on analysis. F and G were involved in the design and contributed to the writing of the paper. H was involved in the design, implementation, analysis, and biochemical interpretation, and contributed to the writing of the paper. Guarantors: A and H

28 Acknowledgements according to Vancouver At an appropriate place in the article (the title page footnote or an appendix to the text; see the journal s requirements), one or more statements should specify: (1) contributions that need acknowledging but do not justify authorship, such as general support by a departmental chair; (2) acknowledgements of technical help; (3) acknowledgements of financial and material support, which should specify the nature of the support; and (4) relationships that may pose a conflict of interest. Acknowledgements according to Vancouver Persons who have contributed intellectually to the paper but whose contributions do not justify authorship may be named and their function or contribution described for example, scientific adviser, critical review of must have given their permission to be named. Authors are responsible for obtaining written permission from persons acknowledged by name, because readers may infer their endorsement of the data and conclusions. Technical help should be acknowledged in a paragraph separate from that acknowledging other contributions.

29 Style makes a difference Good Style assists effective communication Style should aid, not hinder, comprehension Clear writing helps articles be accepted for publication Well-presented papers make editors jobs easier House style gives publications consistency and identity Authorship (from Ref.[13]) Authorship credit should be based only on substantial contributions to: Conception and design or analysis and interpretation of data. Drafts of the article of critical revisions for important intellectual content. Final approval of the version to be published.

30 Research misconduct Errors of judgement Inadequate study design Bias Self-delusion Inappropriate statistical analysis Research misconduct Misdemeanours ( trimming and cooking ) Data manipulation Data exclusion Suppression of in convenient facts

31 Research misconduct Fraud Fabrication Falsification Plagiarism Prevention of research and publication misconduct Education Research training Research ethics Publication ethics

32 Prevention of research and publication misconduct The research Protocol driven Establish contributors and collaborators -Define roles -Agree protocol -Agree presentation of results Define methodology for data analysis -Statistical advice Ethical approval Project and personal licence (Home Office) Supervision -Identify guarantor -Good communication -Ensure good clinical practice Meticulous record keeping Prevention of research and publication misconduct The publication Disclose conflict of interest Disclose previous publications Approval by all contributors Submit to one journal at a time Assume research data audit

33 Thanks for your attention! Wish you a successful submitter.