書籍、官方指引與政策來源

研究問題與假設 · Research Question & Hypothesis

• 📚 BookHulley SB, Cummings SR, Browner WS, et al. Designing Clinical Research, 5th edition. Wolters Kluwer, 2023. ISBN: 978-1975174408 · FINER 框架原始出處。Wolters Kluwer, 2023. ISBN: 978-1975174408 · Original source of the FINER framework.
• ⭐ GuideRichardson WS, Wilson MC, Nishikawa J, Hayward RS. The well-built clinical question: a key to evidence-based decisions. ACP J Club. 1995;123(3):A12-3. · PICO 經典定義。ACP J Club. 1995;123(3):A12-3. · Classic definition of the PICO framework.
• 📚 BookSackett DL, et al. Evidence-Based Medicine: How to Practice and Teach EBM. Churchill Livingstone, 5th ed., 2018.Churchill Livingstone, 5th ed., 2018.
• 🛠️ ToolPROSPERO — 系統性回顧 protocol 註冊平台。— Registration platform for systematic review protocols.

文獻回顧策略 · Literature Review Strategy

• 📄 PaperPage MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi:10.1136/bmj.n71BMJ 2021;372:n71. doi:10.1136/bmj.n71
• 📘 DocEQUATOR Network — PRISMA 2020 Checklist — 完整 27 項 + flow diagram template。— Full 27-item checklist plus flow diagram template.
• 📄 PaperTricco AC, Lillie E, Zarin W, et al. PRISMA Extension for Scoping Reviews (PRISMA-ScR). Ann Intern Med 2018;169(7):467-473.Ann Intern Med 2018;169(7):467-473.
• 📚 BookBooth A, Sutton A, Clowes M, Martyn-St James M. Systematic Approaches to a Successful Literature Review, 3rd ed. SAGE, 2022.SAGE, 2022.
• 🛠️ ToolConnected Papers — 視覺化文獻 citation network。— Visualizes citation networks across the literature.
• 🛠️ ToolResearch Rabbit — AI 助理式文獻探索。— AI-assistant-style literature discovery.
• 🛠️ ToolPubMed + MeSH Browser — 生醫文獻檢索核心。— The core search stack for biomedical literature.

IMRaD 與學位論文章節 · IMRaD & Thesis Structure

• 📄 PaperSollaci LB, Pereira MG. The introduction, methods, results, and discussion (IMRAD) structure: a fifty-year survey. J Med Libr Assoc. 2004;92(3):364-7. · IMRaD 歷史源流。J Med Libr Assoc. 2004;92(3):364-7. · Historical origins of the IMRaD structure.
• 📚 BookMurray R. How to Write a Thesis, 4th edition. Open University Press, 2017. ISBN: 978-0335264414 · 學位論文寫作經典。Open University Press, 2017. ISBN: 978-0335264414 · A classic on thesis writing.
• 📚 BookDunleavy P. Authoring a PhD: How to Plan, Draft, Write and Finish a Doctoral Thesis or Dissertation. Palgrave Macmillan, 2003.Palgrave Macmillan, 2003.
• ⭐ GuideVitae Researcher Development Framework — 跨國 PhD 訓練框架。— A cross-country PhD researcher development framework.

緒論寫作 · Introduction Writing

• 📚 BookSwales JM. Genre Analysis: English in Academic and Research Settings. Cambridge University Press, 1990. · CARS Model (Create A Research Space) 原始出處。Cambridge University Press, 1990. · Original source of the CARS (Create A Research Space) model.
• 📚 BookSwales JM, Feak CB. Academic Writing for Graduate Students, 3rd edition. University of Michigan Press, 2012. · 國際公認最常用研究生寫作教科書。University of Michigan Press, 2012. · The most widely adopted graduate-level academic writing textbook internationally.
• 📚 BookBelcher WL. Writing Your Journal Article in Twelve Weeks. University of Chicago Press, 2nd ed., 2019.University of Chicago Press, 2nd ed., 2019.

方法寫作與可重現報告 · Methods & Reproducible Reporting

• 📘 DocEQUATOR Network — 500+ reporting guidelines 的權威清單。— The authoritative directory of 500+ reporting guidelines.
• 📄 Papervon Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. PLoS Med 2007;4(10):e296.PLoS Med 2007;4(10):e296.
• 📄 PaperHopewell S, Chan AW, Collins GS, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ 2025;389:e081123.BMJ 2025;389:e081123.
• 📄 PaperPercie du Sert N, et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol 2020;18(7):e3000410.PLoS Biol 2020;18(7):e3000410.
• 📄 PaperCollins GS, Moons KGM, Dhiman P, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ 2024;385:e078378.BMJ 2024;385:e078378.
• 🛠️ ToolZenodo — CERN 維護的 data/code repository，免費 DOI。— CERN-maintained data/code repository offering free DOIs.

結果與圖表設計 · Results & Figures

• 📚 BookTufte ER. The Visual Display of Quantitative Information, 2nd edition. Graphics Press, 2001. · data-ink ratio 經典。Graphics Press, 2001. · The classic on the data-ink ratio.
• 📚 BookWilke CO. Fundamentals of Data Visualization. O'Reilly, 2019. 免費線上版本O'Reilly, 2019. Free online version
• 📄 PaperWasserstein RL, Lazar NA. The ASA Statement on p-Values: Context, Process, and Purpose. The American Statistician 2016;70(2):129-133. · p 值單獨報告不充分的權威依據。The American Statistician 2016;70(2):129-133. · Authoritative basis for why reporting p-values alone is insufficient.
• 📄 PaperWeissgerber TL, Milic NM, Winham SJ, Garovic VD. Beyond Bar and Line Graphs: Time for a New Data Presentation Paradigm. PLoS Biol 2015;13(4):e1002128. · 為什麼 bar chart + SEM 是壞圖。PLoS Biol 2015;13(4):e1002128. · Why bar chart + SEM is a bad way to display data.
• 🛠️ ToolColorBrewer 2.0 — 色盲友善配色 picker。— Colorblind-friendly palette picker.

討論與限制 · Discussion & Limitations

• 📄 PaperHess DR. How to write an effective discussion. Respir Care 2004;49(10):1238-41. · Discussion 倒漏斗結構經典指引。Respir Care 2004;49(10):1238-41. · Classic guide to the inverted-funnel structure for the Discussion section.
• 📄 PaperIoannidis JPA. Limitations are not properly acknowledged in the scientific literature. J Clin Epidemiol 2007;60(4):324-329. · Limitations 寫法的批判分析。J Clin Epidemiol 2007;60(4):324-329. · Critical analysis of how Limitations sections are written.
• 📚 BookStrunk W, White EB. The Elements of Style, 4th edition. Pearson, 2000. · 校準語氣與簡潔表達的入門經典。Pearson, 2000. · An entry-level classic for calibrating tone and concise expression.

摘要、標題與關鍵字 · Abstract, Title & Keywords

• 📄 PaperHartley J. Current findings from research on structured abstracts: an update. J Med Libr Assoc. 2014;102(3):146-148.J Med Libr Assoc. 2014;102(3):146-148.
• ⭐ GuideBahadoran Z, Mirmiran P, Kashfi K, Ghasemi A. The Principles of Biomedical Scientific Writing: Title. Int J Endocrinol Metab. 2019;17(4):e98326.Int J Endocrinol Metab. 2019;17(4):e98326.
• 🛠️ ToolNCBI MeSH Browser — MeSH 關鍵字檢索。— Search MeSH keywords.

學術寫作語言與風格 · Academic Style & Language

• 📚 BookWilliams JM, Bizup J. Style: Lessons in Clarity and Grace, 12th edition. Pearson, 2017. · 主動句、強動詞、cohesion 黃金準則。Pearson, 2017. · Golden rules for active sentences, strong verbs, and cohesion.
• 📚 BookPinker S. The Sense of Style. Penguin Books, 2014. · 認知科學家視角的寫作指南。Penguin Books, 2014. · A writing guide from the perspective of a cognitive scientist.
• 📚 BookSword H. Stylish Academic Writing. Harvard University Press, 2012.Harvard University Press, 2012.
• 📚 BookSchimel J. Writing Science: How to Write Papers That Get Cited and Proposals That Get Funded. Oxford University Press, 2012.Oxford University Press, 2012.
• 🛠️ ToolHemingway Editor — 句子複雜度檢查器。— Sentence-complexity checker.

引用與文獻管理 · Citation & Reference Management

• 🛠️ ToolZotero + Better BibTeX — 開源文獻管理 + LaTeX 同步。— Open-source reference manager with LaTeX sync.
• 🛠️ ToolEndNote 21 — 機構廣泛採用的商用文獻管理。— Commercial reference manager widely adopted by institutions.
• 🛠️ ToolPaperpile — Google Docs 友善的雲端文獻管理。— Cloud reference manager with strong Google Docs integration.
• 🛠️ ToolZotero Style Repository — 10000+ 期刊 CSL style files。— 10,000+ journal CSL style files.
• 🛠️ ToolRetraction Watch + Zotero 整合 — 自動警示引用 retracted papers。+ Zotero integration — automatically flags citations to retracted papers.
• 📘 DocCitation Style Language (CSL) project — 業界標準 XML 格式。— The industry-standard XML format for citation styles.

期刊選擇與投稿策略 · Journal Selection

• 🛠️ ToolThink.Check.Submit. — 免費期刊檢核 checklist。— Free journal-vetting checklist.
• 🛠️ ToolCabells Predatory Reports — 付費的掠奪性期刊精準名單。— Paid, curated list of predatory journals.
• 🛠️ ToolDOAJ — Directory of Open Access Journals — OA 白名單。— Open access journal allowlist.
• 🛠️ ToolScimago Journal Rank (SJR) + Scopus CiteScore — IF 之外的期刊評估指標。— Journal evaluation metrics beyond the impact factor.
• 🛠️ ToolSciRev.org — 真實作者投稿經驗 (peer review 時間、決定流程透明度)。— Real author submission experiences (peer review duration, decision-process transparency).
• ⚖️ PolicycOAlition S / Plan S — 強制 OA 的歐盟 funder 聯盟。— EU-led funder coalition mandating open access.

同儕審查與回覆信 · Peer Review & Response Letter

• 📘 DocCOPE — Committee on Publication Ethics — 同儕審查倫理權威。— Authoritative body on peer review ethics.
• ⭐ GuideWeb of Science Reviewer Academy (formerly Publons) — 免費 reviewer 訓練。— Free reviewer training.
• ⭐ GuideASAPbio Reviewer Trainee Programme
• 📄 PaperNoble WS. Ten simple rules for writing a response to reviewers. PLoS Comput Biol 2017;13(10):e1005730.PLoS Comput Biol 2017;13(10):e1005730.
• 📄 PaperBourne PE. Ten simple rules for reviewers. PLoS Comput Biol 2005;1(2):e22.PLoS Comput Biol 2005;1(2):e22.

倫理、作者貢獻與利益衝突 · Ethics & Authorship

• 📘 DocICMJE Recommendations (Jan 2026) — 國際醫學期刊編輯委員會權威文件。— Authoritative document of the International Committee of Medical Journal Editors.
• 📘 DocCRediT — Contributor Roles Taxonomy (NISO Z39.104) — 14 角色官方定義。— Official definitions of the 14 contributor roles.
• 🛠️ ToolORCID — 免費作者唯一識別 ID。— Free unique author identifier.
• 🛠️ ToolROR — Research Organization Registry — 機構唯一 ID。— Unique institutional identifier registry.
• ⚖️ PolicyCOPE Guidance — 抄襲、duplicate publication、salami slicing 處理流程。— Workflows for handling plagiarism, duplicate publication, and salami slicing.
• 🛠️ ToolICMJE Disclosure of Interest form — COI 申報通用範本。— Universal template for conflict-of-interest disclosure.
• 📚 BookResnik DB. The Ethics of Science: An Introduction. Routledge, 2nd ed., 2018.Routledge, 2nd ed., 2018.

AI 工具與可重現實踐 · AI Tools & Reproducibility

• ⚖️ PolicyElsevier — Generative AI Policy for Journals (2025/09 update)
• ⚖️ PolicySpringer Nature — AI Policy
• ⚖️ PolicyCell Press — Author Guide (AI section)
• ⚖️ PolicyAPA Journals — Generative AI Policy
• ⚖️ PolicyTaylor & Francis — AI Policy
• 🛠️ ToolbioRxiv + medRxiv — CSHL 維護的生命科學/臨床 preprint server。— CSHL-maintained preprint servers for life sciences and clinical research.
• 🛠️ ToolQuarto — 2026 推薦的 reproducible publishing 框架 (R + Python + Julia)。— Recommended 2026 reproducible publishing framework (R + Python + Julia).
• 🛠️ ToolSnakemake + Nextflow — 標準 pipeline 管理框架。— Standard pipeline management frameworks.
• 📄 PaperWilkinson MD, Dumontier M, et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data 2016;3:160018. — Findable, Accessible, Interoperable, Reusable 原則出處。Sci Data 2016;3:160018. — Original source of the FAIR (Findable, Accessible, Interoperable, Reusable) principles.

📌 教學註記與細節

以下列出本教學中需要進一步澄清的論點、適用範圍限制，以及社群普遍存在的誤解。所有條目附原始來源。

The following points clarify boundary conditions, scope limitations, and widespread misconceptions in the tutorial. All entries cite original sources.

• ⚠ Errata 1

  IMRaD 結構並非通用標準 · IMRaD Structure Is Not Universal

  IMRaD (Introduction–Methods–Results–Discussion) 主要在生物醫學與實驗科學佔主導，但人文、社會科學、純數學、理論物理、工程設計、case study 與 review article 多採其他結構（如 thematic、chronological、 problem-solution 或 essay-style）。Sollaci & Pereira 2004 指出 IMRaD 至 1980s 才在醫學期刊達到 80% 採用率。投稿前應先閱讀目標期刊的 Author Guidelines。

  IMRaD (Introduction–Methods–Results–Discussion) dominates biomedicine and experimental sciences but is not universal. Humanities, social sciences, pure mathematics, theoretical physics, engineering design papers, case studies, and review articles often use alternative structures (thematic, chronological, problem–solution, or essay-style). Sollaci & Pereira (2004) note that IMRaD only reached ~80% adoption in medical journals by the 1980s. Always read the target journal's Author Guidelines before drafting.

  來源：Sollaci LB, Pereira MG. J Med Libr Assoc. 2004;92(3):364-7.；ICMJE Recommendations 2024 §IV.A.Sources: Sollaci LB, Pereira MG. J Med Libr Assoc. 2004;92(3):364-7.; ICMJE Recommendations 2024 §IV.A.
• ⚠ Errata 2

  期刊影響因子 (JIF) 的限制 · Journal Impact Factor Caveats

  JIF 是期刊層級的平均引用指標，不能用來評估單篇論文或個別研究者的品質。引用分布極度偏態 (skewed)，前 20% 論文往往貢獻 80% 引用。DORA (Declaration on Research Assessment, 2012) 已有超過 2,500 個機構簽署，呼籲停止用 JIF 評估研究人員。Leiden Manifesto (Hicks et al. 2015 Nature) 提出 10 原則為替代評估標準。

  JIF is a journal-level average citation metric and cannot evaluate the quality of individual papers or researchers. Citation distributions are heavily skewed — the top 20% of papers typically generate 80% of citations. DORA (Declaration on Research Assessment, 2012), signed by 2,500+ institutions, calls for abandoning JIF in researcher evaluation. The Leiden Manifesto (Hicks et al. 2015 Nature) proposes 10 alternative assessment principles.

  來源：DORA Declaration 2012；Hicks D, et al. Nature 2015;520:429-431 doi:10.1038/520429a；Seglen PO. BMJ 1997;314:498-502.Sources: DORA Declaration 2012; Hicks D, et al. Nature 2015;520:429-431 doi:10.1038/520429a; Seglen PO. BMJ 1997;314:498-502.
• ⚠ Errata 3

  掠奪性期刊辨識的多重來源 · Identifying Predatory Journals Requires Multiple Sources

  不可僅憑單一名單判定期刊性質。Beall's List 自 2017 年起停止官方維護 (Jeffrey Beall 個人因壓力下線)，目前由志願者維護 archived 版本 (beallslist.net)；Cabells Predatory Reports 為付費精選名單；Think.Check.Submit 提供免費可重複使用的檢核 checklist (期刊網站透明度、editorial board 真實性、indexed in DOAJ/PubMed/Scopus、APC 結構)。Grudniewicz et al. 2019 Nature 提出共識定義：掠奪性期刊以自利優先、提供誤導訊息、偏離編輯與出版最佳實踐、缺乏透明度、使用激進招攬。

  Never rely on a single list to judge a journal. Beall's List has not been officially maintained since 2017 (Jeffrey Beall took it down under pressure; volunteer-maintained archives exist at beallslist.net). Cabells Predatory Reports is a paid curated list. Think.Check.Submit offers a free reusable checklist (website transparency, editorial-board verification, indexing in DOAJ/PubMed/Scopus, APC structure). Grudniewicz et al. (2019 Nature) provide the consensus definition: predatory journals prioritize self-interest, provide false information, deviate from best editorial and publication practices, lack transparency, and use aggressive solicitation.

  來源：Grudniewicz A, et al. Nature 2019;576:210-212 doi:10.1038/d41586-019-03759-y；Think.Check.Submit.；Beall's List (archived)。Sources: Grudniewicz A, et al. Nature 2019;576:210-212 doi:10.1038/d41586-019-03759-y; Think.Check.Submit.; Beall's List (archived).
• ⚠ Errata 4

  開放近用 (OA) vs APC 取捨與 Plan S · Open Access vs APC Tradeoffs & Plan S

  Gold OA 提供立即開放近用但需作者支付 APC (Article Processing Charges)，部分 hybrid journals APC 高達 $11,690 (Nature 2021)。Green OA (institutional repository self-archiving) 通常免費但有 embargo period (6–12 個月)。Diamond OA 對讀者與作者皆免費，由機構或學會補貼 (Bosman et al. 2021 OPERAS 報告約 29,000 種期刊)。Plan S (cOAlition S, 2018 起) 要求受資助研究自 2021 起立即 OA；2024-2026 過渡至 Rights Retention Strategy 與消除 hybrid model。APC 對 LMIC (中低收入國家) 作者形成嚴重結構障礙 (Smith et al. 2021 BMJ Global Health)。

  Gold OA provides immediate open access but requires authors to pay APCs (Article Processing Charges); some hybrid journal APCs reach $11,690 (Nature 2021). Green OA (institutional-repository self-archiving) is typically free but subject to embargo periods (6–12 months). Diamond OA is free for both readers and authors, subsidized by institutions or societies (Bosman et al. 2021 OPERAS report: ~29,000 journals). Plan S (cOAlition S, since 2018) mandates immediate OA for funded research from 2021; the 2024–2026 transition emphasizes the Rights Retention Strategy and elimination of hybrid models. APCs pose a serious structural barrier for authors in LMICs (Smith et al. 2021 BMJ Global Health).

  來源：cOAlition S / Plan S；Bosman J, et al. OPERAS Diamond OA Study 2021 doi:10.5281/zenodo.4558704；Smith E, et al. BMJ Glob Health 2021;6:e004088。Sources: cOAlition S / Plan S; Bosman J, et al. OPERAS Diamond OA Study 2021 doi:10.5281/zenodo.4558704; Smith E, et al. BMJ Glob Health 2021;6:e004088.
• ⚠ Errata 5

  預印本 vs 同儕審查：bioRxiv/medRxiv 統計 · Preprint vs Peer Review: bioRxiv/medRxiv Stats

  預印本 (preprint) 是未經同儕審查的初稿，提供 priority claim 與快速擴散，但讀者需自行批判性評估。bioRxiv (2013 創立)、medRxiv (2019 創立) 累計上傳已超過 30 萬篇 (Abdill & Blekhman 2019 eLife)，COVID-19 期間 medRxiv 單月上傳超 2,000 篇 (Fraser et al. 2021 PLoS Biol)。研究顯示約 70% 預印本最終被期刊接受並發表 (Abdill & Blekhman 2019)；預印本與最終發表版本差異多屬細節層面，主要結論變更比例低 (Carneiro et al. 2020 BMJ Open)。引用預印本應明確標示 "preprint" 並提供 DOI 與版本號 (v1, v2)。ICMJE 與大多生醫期刊接受預印本作為非重複發表。

  A preprint is an unrefereed manuscript that establishes priority and accelerates dissemination but demands critical reader appraisal. bioRxiv (founded 2013) and medRxiv (founded 2019) have together accumulated 300,000+ submissions (Abdill & Blekhman 2019 eLife); during COVID-19 medRxiv exceeded 2,000 uploads per month (Fraser et al. 2021 PLoS Biol). Roughly 70% of preprints are eventually accepted by journals (Abdill & Blekhman 2019); differences between preprint and final published versions are mostly minor, with low rates of major conclusion changes (Carneiro et al. 2020 BMJ Open). When citing a preprint, explicitly label "preprint" and include DOI plus version number (v1, v2). ICMJE and most biomedical journals accept preprints as not constituting duplicate publication.

  來源：Abdill RJ, Blekhman R. eLife 2019;8:e45133 doi:10.7554/eLife.45133；Fraser N, et al. PLoS Biol 2021;19:e3000959；Carneiro CFD, et al. BMJ Open 2020;10:e035458。Sources: Abdill RJ, Blekhman R. eLife 2019;8:e45133 doi:10.7554/eLife.45133; Fraser N, et al. PLoS Biol 2021;19:e3000959; Carneiro CFD, et al. BMJ Open 2020;10:e035458.
• ⚠ Errata 6

  AI 寫作工具：ICMJE 2024 揭露要求 · AI Writing Tools: ICMJE 2024 Disclosure Requirement

  ICMJE (International Committee of Medical Journal Editors) 2024 年 1 月更新 Recommendations 明確規定：(1) AI / chatbots / LLMs 不能列為作者 (cannot be authors)，因為無法承擔 accountability；(2) 作者必須在 cover letter 與 manuscript 中揭露 AI 使用 (在 Acknowledgements 或 Methods 段落)，包含工具名稱與版本、使用範圍、生成段落定位；(3) 作者對 AI 生成內容的正確性、原創性與無偏誤負完全責任，包含潛在的捏造引用 (hallucinated citations) 與抄襲。WAME (World Association of Medical Editors) 2023 與 COPE 2023 立場相同。Nature 與 Science 系列亦在 2023 同步更新政策禁止 AI 共同作者。

  In January 2024, ICMJE (International Committee of Medical Journal Editors) updated its Recommendations to specify: (1) AI / chatbots / LLMs cannot be listed as authors because they cannot take accountability; (2) authors must disclose AI use in both the cover letter and the manuscript (in Acknowledgements or Methods), including tool name and version, scope of use, and location of generated passages; (3) authors bear full responsibility for the correctness, originality, and bias-freeness of AI-generated content, including hallucinated citations and plagiarism. WAME (World Association of Medical Editors) 2023 and COPE 2023 take the same position. Nature and Science families updated their policies in 2023 to prohibit AI co-authorship.

  來源：ICMJE Recommendations 2024 §II.A.4；WAME Chatbots and ChatGPT Statement 2023；COPE Position Statement on AI 2023；Thorp HH. Science 2023;379(6630):313 doi:10.1126/science.adg7879。Sources: ICMJE Recommendations 2024 §II.A.4; WAME Chatbots and ChatGPT Statement 2023; COPE Position Statement on AI 2023; Thorp HH. Science 2023;379(6630):313 doi:10.1126/science.adg7879.
• ⚠ Errata 7

  審稿人回覆信架構：Noble 2017 ten simple rules · Reviewer Response Framework: Noble 2017 Ten Simple Rules

  Noble (2017 PLOS Comp Biol) 提出 reviewer response 的 ten simple rules：(1) 提供 reviewer 概覽 (overview)；(2) 逐點 (point-by-point) 對應，不可遺漏；(3) 用顏色或編號清楚標示原文修改位置；(4) 即使不同意亦保持禮貌；(5) 解釋為何同意或不同意；(6) 修改正文以反映回覆；(7) 在文本中加 (line X, page Y) 指引；(8) 控制長度，必要時附 supplementary；(9) 對 editor 提供獨立摘要；(10) 不要要求 reviewer 進一步審稿時間。Annesley (2011 Clin Chem) 與 Williams (2004 Chest) 亦提供互補建議。常見錯誤：對審稿人態度防衛、忽略次要 comment、修改但未於回覆信中說明。

  Noble (2017 PLOS Comp Biol) presents ten simple rules for responding to reviewers: (1) provide an overview for each reviewer; (2) respond point-by-point with no omissions; (3) clearly mark manuscript revisions via color or numbering; (4) remain courteous even when disagreeing; (5) explain why you agree or disagree; (6) edit the manuscript to reflect responses; (7) cite (line X, page Y) pointers in-text; (8) keep length manageable; attach supplementary if needed; (9) write a separate summary for the editor; (10) do not demand additional review time from reviewers. Annesley (2011 Clin Chem) and Williams (2004 Chest) provide complementary advice. Common errors: defensive tone, ignoring minor comments, revising without explaining in the response letter.

  來源：Noble WS. PLoS Comput Biol 2017;13(10):e1005730 doi:10.1371/journal.pcbi.1005730；Annesley TM. Clin Chem 2011;57(4):551-554；Williams HC. Chest 2004;126(2):347-349。Sources: Noble WS. PLoS Comput Biol 2017;13(10):e1005730 doi:10.1371/journal.pcbi.1005730; Annesley TM. Clin Chem 2011;57(4):551-554; Williams HC. Chest 2004;126(2):347-349.
• ⚠ Errata 8

  ORCID 與 ROR：研究人員與機構持久識別碼 · ORCID & ROR: Persistent Identifiers for Researchers & Organizations

  姓名歧義 (name ambiguity) 是 bibliometrics 與資助追蹤的核心問題 (相同姓名作者、姓氏拼寫變體、結婚改姓、機構轉移)。ORCID (Open Researcher and Contributor ID, 2012 起) 提供 16 位數 ISO 27729 標準 ID (e.g., 0000-0002-1825-0097)，目前已有超過 1,800 萬研究人員註冊 (ORCID 2024 Annual Report)；多數主要期刊與資助機構 (NIH、NSF、Wellcome Trust、ERC) 強制要求。Khan & Daud (2021 Scientometrics) 系統性回顧顯示 ORCID 採用顯著提升作者識別準確度與 cross-database 連結效率。ROR (Research Organization Registry, 2019 起) 為機構提供類似 ID (e.g., https://ror.org/02jbv0t02 for Stanford)，補足 GRID、Ringgold、ISNI 等舊系統。投稿時應同時提供 ORCID 與 ROR，未來 funder reporting 與 metadata 互通將以此為基礎。

  Name ambiguity is a core problem in bibliometrics and funding tracking (homonyms, surname spelling variants, marriage-related changes, institutional mobility). ORCID (Open Researcher and Contributor ID, since 2012) provides a 16-digit ISO 27729 identifier (e.g., 0000-0002-1825-0097); over 18 million researchers are registered (ORCID 2024 Annual Report). Most major journals and funders (NIH, NSF, Wellcome Trust, ERC) now mandate it. Khan & Daud (2021 Scientometrics) provide a systematic review showing ORCID adoption substantially improves author-disambiguation accuracy and cross-database linkage efficiency. ROR (Research Organization Registry, since 2019) provides analogous identifiers for institutions (e.g., https://ror.org/02jbv0t02 for Stanford), complementing legacy systems such as GRID, Ringgold, and ISNI. Submit ORCID and ROR together — future funder reporting and metadata interoperability will rely on both.

  來源：Khan S, Daud A. Scientometrics 2021;126(8):7461-7479 doi:10.1007/s11192-021-04098-4；ORCID Initiative；ROR — Research Organization Registry；Haak LL, et al. Inf Serv Use 2012;32(1):51-55。Sources: Khan S, Daud A. Scientometrics 2021;126(8):7461-7479 doi:10.1007/s11192-021-04098-4; ORCID Initiative; ROR — Research Organization Registry; Haak LL, et al. Inf Serv Use 2012;32(1):51-55.