技能 数据科学 开放科学与透明化研究指南

开放科学与透明化研究指南

v20260724
joc-open-science-and-transparency
本指南为撰写投稿至《传播学期刊》(JoC)的稿件提供详细指导。它详细阐述了开放科学和透明化的要求,涵盖数据可用性声明(DAS)、开放科学徽章(开放数据、开放材料、预注册)等关键要素。帮助研究人员理解如何确保定性、定量和计算研究的材料具有高度可重复性,实现“写作即准备”的流程。
获取技能
479 次下载
概览

Open Science & Transparency (joc-open-science-and-transparency)

JoC requires a Data Availability Statement on every article and offers Open Science Badges for open data, open materials, and preregistration. Transparency is not an afterthought — build the statement and supporting materials as you write so submission and any badge claim go smoothly.

When to trigger

  • Drafting the Data Availability Statement (a submission requirement)
  • Deciding whether to pursue Open Science Badges (open data / open materials / preregistration)
  • Preparing a preregistration for a prospective design (note it in the cover letter)
  • Data cannot be fully shared (privacy, ethics, platform/legal restrictions) and you need the path forward

What JoC expects (verify current wording on the policy page)

  1. Data Availability Statement (required). State where the data are (repository + identifier), under what conditions they can be accessed, and — if they cannot be shared — why, with instructions for how others might obtain them.
  2. Open Science Badges (optional, earned).
    • Open Data — data and a codebook deposited in a trusted repository with a persistent identifier.
    • Open Materials — stimuli, instruments, code, and protocols deposited so the study can be reproduced.
    • Preregistration — a time-stamped, registered design/analysis plan; note it in the cover letter.
  3. Quantitative / computational materials. Data, code, codebook, and documentation sufficient to regenerate every reported result; master script + README + pinned versions + seeds; for automated text measures, include the human-validation materials.
  4. Qualitative materials. Share the materials that support the claims (e.g., coding schemes, evidence tables, de-identified excerpts) with access controls where needed (QDR for controlled access).

When data cannot be shared (restricted-data path)

  • Explain why in the Data Availability Statement (ethical/privacy concerns, platform terms of service, or legal restrictions by the provider).
  • Provide instructions on how others can obtain the data (access process, application, provider contact).
  • Where possible, provide synthetic or de-identified data so the code can be run.

Build-as-you-go checklist

  • Data Availability Statement drafted (repository, identifier, access conditions, or exemption)
  • One master script regenerates every table and figure from raw/constructed data
  • README documents data provenance, construction, and how to reproduce each exhibit
  • Seeds set and reported for every stochastic step; software/package versions pinned
  • Open Science Badge materials staged (open data / open materials / preregistration) where claimed
  • Content analysis: codebook + intercoder-reliability report included
  • Materials anonymized (no author-identifying paths/links) for double-anonymous review

Transparency expectations by communication method (decision table)

Because JoC spans communication research broadly, "transparency" differs across traditions. Match the deposit to the method rather than forcing one template:

Method What a JoC referee wants deposited Badge most relevant
Survey / message experiment data + codebook + stimuli + analysis script open data + materials + preregistration
Content analysis codebook + coder instructions + reliability subsample + texts open materials (+ open data)
Computational / text-as-data corpus or query, model/version, seeds, human-validation set open materials + open data
Qualitative / critical coding scheme + evidence tables + de-identified excerpts open materials where ethics allow

For computational measures, the human-validation set is itself the evidence that the automated label means what the paper claims — depositing the classifier without it leaves the construct unverified. Confirm badge mechanics against the journal's current guidance.

Worked micro-example: a DAS for a copyrighted-news corpus (illustrative)

A computational content analysis of 40,000 news articles (illustrative) on climate coverage hits a familiar wall: the texts are copyrighted and the feed bars redistribution. The transparency path: (1) deposit the codebook, article IDs/URLs, query parameters, and analysis code so a same-license reader reproduces the pipeline; (2) deposit the human-validation sample, shareable derived data; (3) write a DAS naming the restriction, provider, and access route, and offering de-identified derived features (frame proportions per article) so modeling re-runs without raw text — earning open-materials credit and an honest, hedged DAS.

Anti-patterns

  • Submitting without a Data Availability Statement (it is required)
  • Claiming a badge whose materials are not actually deposited or do not reproduce the results
  • A personal URL instead of a trusted repository with a persistent identifier
  • Claiming data are restricted without giving an access path or synthetic substitute
  • De-anonymizing the manuscript via an open-materials link during review

Output format

【Data Availability Statement】drafted? repository + identifier or exemption? [Y/N]
【Reproduces tables/figures?】master script verified locally? [Y/N]
【Badges sought】open data / open materials / preregistration (materials staged?)
【Documentation】README + provenance + seeds + pinned versions? [Y/N]
【Restricted data?】exemption note + access path + synthetic data?
【Qualitative transparency】coding scheme / evidence documented? [Y/N/NA]
【Next】joc-review-process

Supplementary resources

信息
Category 数据科学
Name joc-open-science-and-transparency
版本 v20260724
大小 6.41KB
更新时间 2026-07-28
语言