技能 数据科学 学术数据透明与复制标准指南

学术数据透明与复制标准指南

v20260724
poq-transparency-and-data-policy
本指南详细阐述了学术期刊提交稿件的严格数据透明和可复现性要求。它指导作者完成两大核心任务:一是系统性地提供方法学披露要素(Appendix A),二是构建完整的复制包。作者必须确保所有发表的表格和图表都能通过在永久数据仓库中提交的数据和代码进行准确重现,这必须在排版前完成。
获取技能
475 次下载
概览

Transparency & Data Policy (poq-transparency-and-data-policy)

POQ has two transparency demands that distinguish it from generic social-science venues: AAPOR-standard methodological disclosure of every survey, and a replication package that reproduces exactly all published tables and figures, archived in POQ's Dataverse before the paper advances to typesetting. Build both as you go so acceptance does not stall.

When to trigger

  • Assembling Appendix A: Disclosure Elements for every dataset reported
  • Building the replication/reproduction package for POQ's Dataverse
  • A manuscript was accepted and the office requested materials before typesetting
  • Data cannot be fully shared (privacy, ethics, provider restrictions) and you need the exemption path

AAPOR disclosure — Appendix A: Disclosure Elements

POQ requires manuscripts to disclose these elements for all data reported (if applicable) or link to publicly available / published documentation, gathered in an "Appendix A: Disclosure Elements":

  1. Funding — who sponsored the data collection.
  2. Exact question wording of all survey questions used.
  3. Population under study — a clear description.
  4. Sample design — frame, selection, stratification, clustering.
  5. Method and dates of data collection — mode(s) and field period.
  6. Response rate and how it was calculated — per AAPOR Standard Definitions (state RR1–RR6 and show the disposition-code computation).
  7. Sample sizes and a discussion of the precision of the findings.
  8. Design effect due to clustering and weighting — discuss the adjustment.

For secondary data, link to the public documentation rather than reproducing it.

Replication package — POQ Dataverse (Harvard)

  1. Deposit to POQ's Dataverse on Harvard Dataverse (the journal's permanent archive with persistent identifiers) — not a personal website or generic cloud link.
  2. Reproduce exactly all published tables and figures: data + code + documentation sufficient to regenerate every reported result. Master script + README + pinned versions + seeds.
  3. Archived before typesetting — accepted manuscripts do not advance until materials are in place.
  4. Data Availability Statement (DAS) in the article's endmatter describing where/how to access the data.

When data cannot be shared (exemption path)

  • Explain why the data are restricted (privacy, ethics, provider/legal limits).
  • Provide README instructions on exactly how others can obtain the data (access process, contact).
  • Where possible, provide synthetic data so the code can be run.
  • A twelve-month embargo from the article's appearance may be requested (confirm current terms — 待核实).

Build-as-you-go checklist

  • Appendix A drafted with all eight disclosure elements (or links) for every dataset
  • Response rate stated with AAPOR definition + calculation
  • One master script regenerates every table and figure
  • README + provenance + seeds + pinned versions
  • Exhibit numbers in the manuscript match package output exactly
  • Data Availability Statement drafted for the endmatter
  • Restricted data: exemption note + access path + synthetic data + embargo request if needed

Anti-patterns

  • A response rate with no AAPOR definition or calculation (an Appendix A failure)
  • Treating the deposit as a post-publication afterthought (it gates typesetting)
  • Code that does not actually reproduce the printed tables/figures
  • A personal URL instead of POQ's Dataverse
  • Claiming data are restricted without an access path or synthetic substitute

Output format

【Appendix A】all 8 disclosure elements present (or linked)? [Y/N]
【Response rate】AAPOR definition + calculation shown? [Y/N]
【Repository】POQ Dataverse (Harvard) — package staged? [Y/N]
【Reproduces tables/figures?】master script verified locally? [Y/N]
【Documentation】README + provenance + seeds + pinned versions? [Y/N]
【DAS】drafted for endmatter? [Y/N]
【Restricted data?】exemption + access path + synthetic + embargo?
【Next】poq-review-process

Supplementary resources

信息
Category 数据科学
Name poq-transparency-and-data-policy
版本 v20260724
大小 4.89KB
更新时间 2026-07-29
语言