📖 《把 AI 当助手——经济实证研究的诚信、方法与工具》双语网页版:每章中文部在前,English Edition 在后,内容与最新合订 PDF 一致。

附录 C · 产出物清单与提交格式

C.1 全手册产出物汇总表

先看演进链，再看汇总表。12 章的产出物不是并列的作业，而是一条逐章生长的研究档案链——后一环以前一环为输入：

第2章 research_question.md（问题卡：种子）
   v
第3章 one_page_proposal.md + 30papers.bib（选题落地 + 文献库）
   v
第4章 research_design.md 五栏方案 + theory_framework.png（设计定型）
   v
第5章 measurement_plan.md + sampling_plan.md（测量与抽样）
   v
第6章 data_source_matrix.xlsx（数据源定型）
   v
第7章 my_questionnaire_v2.docx + codebook.xlsx（问卷与编码）
   v
第8章 descriptive_stats.rtf + 体检log（数据变结论的第一步）
   v
第9章 .do 文件 + 全部 output（实证主体：基准/内生/稳健/机制/异质性）
   v
第10章 分轨产出（教学案例双件套 / 研究案例设计书——二选一）
   v
第11章 my_paper_final.docx + AI声明三版（终稿合流）

下表按章汇总全部产出物、格式、提交节点与归位目录：

表 C.1 全手册产出物汇总表。

L6|L6|L6|L6|L6|L6

章 & 产出物 & 格式 & 提交节点 & 归位目录 & 备注
第 0 章 & 读前自测 10 题 & 附录 D 对照自查 & 第 1 周 & — & 不提交，自测用
第 1 章 & + 标准文件夹截图 & .docx + .png & 第 1 周 & + & Stata putdocx 输出
第 2 章 & （研究问题卡：What/Why/How + 分析单元 + 构念清单） & Markdown & 第 2 周 & & 第 4 章的种子
第 3 章 & + + & .bib + .xlsx + .md & 第 3 周 & + & ≥ 30 篇经 Zotero 核实，零幻觉
第 4 章 & （五栏方案）+ （理论框架图） & .md + .png & 第 3 周 & + & AI 辅助 mermaid + 人工逐箭头把关
第 5 章 & （构念操作化表）+ （抽样方案一页纸） & Markdown & 第 4 周 & & 量表借用须注明出处（S5.1）
第 6 章 & （四维质量评估矩阵） & Excel & 第 4 周 & 同级的 & 出处+版本+访问日期留痕（S6.1）
第 7 章 & （预调研迭代版）+ 信效度检验输出 + & .docx + .rtf + .xlsx & 第 5 周 & + & 知情同意书随附（S7.1）
第 8 章 & （三线表）+ 数据体检 log & .rtf + .log & 第 5 周 & + & 完整日志可复现（S8.1）
第 9 章 & 实证分析 do 文件 + 全部 output（基准/内生性/稳健性/机制/异质性） & .do + .rtf/.docx + .log & 第 6-7 周 & + & 可复现三件套（S9.3）+ 规格曲线（S9.1）
第 10 章 & 教学案例轨：（正文 + 使用说明）；研究案例轨： + & 分轨二选一 & 第 7 周 & & 教学案例正文 ≥ 6000 字；研究案例须有理论对话对象
第 11 章 & （投稿版含 AI 声明）+ （风格定位一页纸）+ AI 声明三版（开题/投稿/答辩） & .docx + .md & 第 8 周 & & 投稿前过 S11.1 + S11.2\

C.2 提交格式要求（5 条硬规）

文件命名：学号_姓名_章节_产出物名.扩展名
- 示例：2024001_张三_Ch03_one_page_proposal.md
- 示例：2024001_张三_Ch09_baseline.do
- 章节统一用 Ch00-Ch11 两位数（便于按字典序排序）

压缩打包：每章产出物打包为 zip，命名 学号_姓名_章节_产出物.zip
- 示例：2024001_张三_Ch03_产出物.zip
- 包内目录结构与附录 A.3 标准研究文件夹一致

Markdown 文件：UTF-8 编码，无 BOM；行尾 LF（macOS/Linux）或 CRLF（Windows）一致即可
- 验证：file your_file.md 应返回 UTF-8 Unicode text，不应含 with BOM

Stata 产出：必须附 log 文件（证明可复现）
- log 命名：YYYYMMDD_HHMM_<dofile名>.log（带时间戳）
- log 内容必须包含 version / set more off / 完整命令记录
- 缺失 log 视为 S9.3 不合格，本章产出物退回

数据文件：data/raw/ 中的原始数据不提交（文件太大；版权与隐私风险）
- 仅提交 do 文件 + data/clean/ 的描述（变量名 + 类型 + 缺失值规则）
- 涉人调查数据须脱敏后提交（S7.1 知情同意 + 数据脱敏）
- 教师机原始数据保留路径写在 README.md 中（导师审阅时可现场核查）

C.3 提交节点甘特图（8 周节奏）

周次       | 1   2   3   4   5   6   7   8 |
----------+-------------------------------+
第 0 章   | ###                             |（自测）
第 1 章   | ###                             |
第 2 章   |     ###                         |
第 3 章   |         ###                     |
第 4 章   |         ###                     |
第 5 章   |             ###                 |
第 6 章   |             ###                 |
第 7 章   |                 ###             |
第 8 章   |                 ###             |
第 9 章   |                     ### ###     |（分两周）
第 10 章  |                         ###     |
第 11 章  |                             ### |
----------+-------------------------------+

总节奏：8 周。三类学生（学硕 / 农村金融专硕 / 金融专硕）共用同一节奏，第 10 章分轨。

C.4 提交前最终自查清单（5 项）

提交任何一章产出物前，请逐项打钩：

命名规范：文件名含学号_姓名_章节，扩展名正确
格式合规：Markdown 无 BOM；Stata 有 log；docx 可正常打开（officecli validate 通过）
诚信检查：本章对应 S.X.Y 检查点全部通过（见附录 B）
可复现：data/raw/ 路径写清；do 文件可从零跑通
AI 披露：用到的 AI 工具 / 提示词 / 修改痕迹已在产出物末尾或 README 中声明

△ 任何一项不通过 → 不要提交，先解决再交。学术诚信是底线（S0.0）。

English Edition

Appendix C · Deliverables and Submission Format

C.1 Master list of deliverables across the handbook

Look at the evolution chain first, then the summary table. The deliverables of the 12 chapters are not parallel assignments but a research archive that grows chapter by chapter—each link takes the previous one as its input:

Ch.2  research_question.md (question card: the seed)
   v
Ch.3  one_page_proposal.md + 30papers.bib (topic settled + literature base)
   v
Ch.4  research_design.md five-column plan + theory_framework.png (design fixed)
   v
Ch.5  measurement_plan.md + sampling_plan.md (measurement and sampling)
   v
Ch.6  data_source_matrix.xlsx (data sources fixed)
   v
Ch.7  my_questionnaire_v2.docx + codebook.xlsx (questionnaire and coding)
   v
Ch.8  descriptive_stats.rtf + examination log (first step from data to conclusions)
   v
Ch.9  .do file + all output (empirical core: baseline/endogeneity/robustness/mechanism/heterogeneity)
   v
Ch.10 track-specific output (teaching-case pair / research-case design--choose one)
   v
Ch.11 my_paper_final.docx + three AI declarations (final manuscript converges)

The table below summarises every deliverable by chapter, together with format, submission milestone, and filing directory:

Table C.1 Master list of deliverables across the handbook.

L6|L6|L6|L6|L6|L6

Ch. & Deliverable & Format & Milestone & Filing directory & Note
Ch. 0 & Ten pre-reading self-test questions & Checked against Appendix D & Week 1 & — & Not submitted; for self-testing
Ch. 1 & + screenshot of the standard folder & .docx + .png & Week 1 & + & Stata putdocx output
Ch. 2 & (question card: What/Why/How + unit of analysis + construct list) & Markdown & Week 2 & & The seed of the Ch. 4
Ch. 3 & + + & .bib + .xlsx + .md & Week 3 & + & $\geq$ 30 entries verified in Zotero, zero hallucinations
Ch. 4 & (five-column plan) + & .md + .png & Week 3 & + & AI-assisted Mermaid + manual arrow-by-arrow review
Ch. 5 & (construct operationalisation table) + (one-page sampling plan) & Markdown & Week 4 & & Borrowed scales must cite their source (S5.1)
Ch. 6 & (four-dimension quality matrix) & Excel & Week 4 & , alongside & Source + version + access date recorded (S6.1)
Ch. 7 & (pilot-iterated) + reliability/validity output + & .docx + .rtf + .xlsx & Week 5 & + & Informed consent form attached (S7.1)
Ch. 8 & (three-line table) + data examination log & .rtf + .log & Week 5 & + & Complete log, reproducible (S8.1)
Ch. 9 & Empirical do-file + all output (baseline/endogeneity/robustness/mechanism/heterogeneity) & .do + .rtf/.docx + .log & Weeks 6–7 & + & Reproducibility triad (S9.3) + specification curve (S9.1)
Ch. 10 & Teaching-case track: (text + teaching notes); Research-case track: + & Choose one track & Week 7 & & Teaching case text $\geq$ 6,000 characters; research case must engage a theoretical interlocutor
Ch. 11 & (submission version with AI disclosure) + (one-page style positioning) + three AI declarations (proposal/submission/defence) & .docx + .md & Week 8 & & Pass S11.1 + S11.2 before submission\

C.2 Submission format requirements (five hard rules)

File naming: studentID_name_chapter_deliverable.ext
- Example: 2024001_ZhangSan_Ch03_one_page_proposal.md
- Example: 2024001_ZhangSan_Ch09_baseline.do
- Always use two digits for the chapter, Ch00–Ch11 (so that lexicographic sorting works)

Zipped packaging: pack each chapter’s deliverables into a zip named studentID_name_chapter_deliverables.zip
- Example: 2024001_ZhangSan_Ch03_deliverables.zip
- The internal directory structure must match the standard research folder described in Appendix A.3

Markdown files: UTF-8 encoding, no BOM; line endings may be LF (macOS/Linux) or CRLF (Windows) as long as they are consistent
- Verification: file your_file.md should return UTF-8 Unicode text and must not contain with BOM

Stata output: a log file is mandatory (proof of reproducibility)
- Log naming: YYYYMMDD_HHMM_<dofilename>.log (timestamped)
- The log must contain version / set more off / a complete command record
- A missing log means S9.3 is failed and the chapter’s deliverables are returned

Data files: the raw data in data/raw/ are not submitted (too large; copyright and privacy risk)
- Submit only the do-file plus a description of data/clean/ (variable names + types + missing-value rules)
- Human-subjects survey data must be de-identified before submission (S7.1 informed consent + de-identification)
- The path where raw data are kept on the teacher’s machine goes in README.md (so the supervisor can verify on site)

C.3 Submission milestone Gantt chart (eight-week rhythm)

Week      | 1   2   3   4   5   6   7   8 |
----------+-------------------------------+
Ch. 0     | ###                             |(self-test)
Ch. 1     | ###                             |
Ch. 2     |     ###                         |
Ch. 3     |         ###                     |
Ch. 4     |         ###                     |
Ch. 5     |             ###                 |
Ch. 6     |             ###                 |
Ch. 7     |                 ###             |
Ch. 8     |                 ###             |
Ch. 9     |                     ### ###     |(spans two weeks)
Ch. 10    |                         ###     |
Ch. 11    |                             ### |
----------+-------------------------------+

Code walkthrough (English note) — the Chinese version of this chart labels each row “Ch. N” and the right-hand annotations read “(self-test)” and “(spans two weeks)”; the bars are unchanged.

Overall rhythm: eight weeks. All three student types (academic master’s / rural-finance professional master’s / finance professional master’s) share the same rhythm, with the Ch. 10 track split.

C.4 Final pre-submission checklist (five items)

Before submitting any chapter’s deliverables, tick each item:

Naming: the filename contains student ID, name, and chapter, with the correct extension
Format compliance: Markdown has no BOM; Stata output has a log; docx opens normally (passes officecli validate)
Integrity check: all S.X.Y checkpoints for this chapter pass (see Appendix B)
Reproducibility: the data/raw/ path is stated; the do-file runs from scratch
AI disclosure: the AI tools, prompts, and edits used are declared at the end of the deliverable or in the README

△ If any item fails → do not submit; fix it first. Academic integrity is the bottom line (S0.0).

附录 C · 产出物清单与提交格式 ​

C.1 全手册产出物汇总表 ​

C.2 提交格式要求（5 条硬规） ​

C.3 提交节点甘特图（8 周节奏） ​

C.4 提交前最终自查清单（5 项） ​

English Edition ​

Appendix C · Deliverables and Submission Format ​

C.1 Master list of deliverables across the handbook ​

C.2 Submission format requirements (five hard rules) ​

C.3 Submission milestone Gantt chart (eight-week rhythm) ​

C.4 Final pre-submission checklist (five items) ​