Open Science Training Handbook

Open Science Training Handbook
Name	Open Science Training Handbook
Subject	Open science, research skills, reproducibility
First published	2018
Contributors	rOpenSci; Mozilla; Australian Research Data Commons; Carpentries
Language	English

Open Science Training Handbook The Open Science Training Handbook is a community-authored guide for teaching research data management and open access skills to researchers, librarians, and educators. It synthesizes practices from rOpenSci, The Carpentries, Mozilla Foundation, European Open Science Cloud, and Australian Research Data Commons to support reproducible workflows, ethical sharing, and collaborative scholarship. The handbook informs training programs at institutions such as University of Cambridge, Stanford University, Max Planck Society, and National Institutes of Health.

Introduction

The handbook emerged from collaborations among networks like Force11, CODATA, Research Data Alliance, OpenAIRE, and SPARC to address gaps identified by reports from European Commission, National Science Foundation, Wellcome Trust, Horizon 2020, and Howard Hughes Medical Institute. It frames open practices in the context of initiatives such as Plan S, FAIR principles, Budapest Open Access Initiative, Panton Principles, and Declaration on Research Assessment (DORA).

Principles of Open Science

Core principles include transparency exemplified by Open Science Framework, accessibility reinforced by Directory of Open Access Journals, reusability aligned with Creative Commons licensing, and reproducibility advanced by GitHub, Zenodo, and Figshare. Ethical sharing acknowledges standards from Declaration of Helsinki, data protection laws like General Data Protection Regulation, and guidance from Committee on Publication Ethics and International Committee of Medical Journal Editors. Community norms reference practices from arXiv, bioRxiv, medRxiv, and publishing models used by PLOS, eLife, Nature Communications, and PeerJ.

Curriculum and Learning Objectives

Learning objectives map to competencies promoted by European Research Council, Wellcome Trust, National Institutes of Health training programs, and syllabi developed by The Carpentries and rOpenSci. Topics include version control workflows using Git, collaborative coding on GitHub, data stewardship guided by DataCite metadata standards, and open licensing with Creative Commons. Students gain skills in preprint submission to bioRxiv and arXiv, protocol sharing via Protocols.io, systematic review registration at PROSPERO, and citation practices informed by CrossRef and ORCID.

Training Methods and Materials

Pedagogy draws on active learning approaches from Carnegie Mellon University, University of Oxford, Massachusetts Institute of Technology, and Harvard University. Materials include modular lesson plans, interactive notebooks using Jupyter Notebook, containerized environments with Docker, continuous integration via Travis CI or GitHub Actions, and datasets deposited in Dryad or Figshare. Trainers adapt resources from The Carpentries Lesson Development Handbook, rOpenSci package development guides, and community curricula hosted on GitHub and archived on Zenodo.

Implementation and Institutional Adoption

Adoption strategies feature policy alignment with funders such as Wellcome Trust, Bill & Melinda Gates Foundation, National Institutes of Health, and European Commission Horizon programs; institutional incentives mirror changes at University of California, Imperial College London, ETH Zurich, and University of Toronto. Roles include research support librarians from Association of College and Research Libraries, research data officers influenced by CODATA frameworks, and centers like Digital Science and OpenAIRE that coordinate campus-wide initiatives. Budgeting and sustainability reference models used by Jisc, Microsoft Research, Google Research, and Allen Institute for AI.

Assessment, Evaluation, and Certification

Assessment tools adapt methods from European Framework for Research Careers, accreditation procedures at Council for Higher Education Accreditation, and badging systems pioneered by Mozilla Open Badges. Evaluation metrics draw on reproducibility audits from Center for Open Science, open citation indicators developed by OpenCitations, and compliance checks used by Plan S signatories. Certification pathways leverage platforms such as Coursera, edX, and institutional continuing professional development offered by University of Edinburgh.

Case Studies and Best Practices

Case studies document projects at University of Melbourne, University of São Paulo, University of Cape Town, University of Tokyo, and National University of Singapore. Best practices highlight community-run initiatives like rOpenSci onboarding, Mozilla Science Lab, Data Carpentry workshops, and policy shifts triggered by Plan S and DORA. Exemplars include reproducible analyses published alongside code in repositories maintained by PLOS Computational Biology, GigaScience, and F1000Research.

Category:Open science