HE-R06 — Open Science and Collaboration

Course Code: HE-R06  |  School: School of Higher Education  |  Cluster: Research Management

Level: Intermediate  |  Estimated duration: 6 weeks (approximately 40 hours)  |  Language: English  |  Certificate: Professional Certificate (non-degree)  |  Format: Self-paced with AI support under human supervision

Overview

Open science is a reform movement with an unusually specific origin: the discovery, across psychology, medicine, economics and other fields, that a substantial proportion of published findings do not replicate. The Open Science Collaboration’s 2015 attempt to reproduce one hundred psychology studies succeeded in roughly a third of cases, and comparable results elsewhere made it impossible to treat the problem as confined to one discipline.

The movement’s proposed remedies — preregistration, registered reports, open data, open materials, open peer review and open access publication — are examined here individually rather than as a package, because they address different failure modes and carry different costs. Learners are expected to reach reasoned positions on which practices are appropriate to their own field rather than adopting the whole programme uncritically.

The course also treats open science as a question of equity. Article processing charges have shifted the cost of access from readers to authors, which advantages well-funded institutions in wealthy countries and disadvantages precisely the researchers open access was meant to serve. The UNESCO Recommendation on Open Science is used as the framework for thinking about openness as an infrastructure and justice question rather than only a methodological one.

Learning outcomes

On completion, a successful learner will be able to:

1. Explain the empirical basis of the replication crisis and the specific research practices that produced it.
2. Evaluate open science practices individually for their fit, cost and benefit within a named discipline.
3. Preregister a study or write a registered report protocol with genuine analytical specificity.
4. Prepare and share research materials, code and data at a standard that permits independent reproduction.
5. Compare open access publishing routes and assess their cost, equity and sustainability implications.
6. Design and manage an open, multi-institution research collaboration with clear governance.
7. Construct a realistic open research plan for a project, including the practices deliberately not adopted and why.

Who this course is for

Researchers across disciplines, doctoral candidates designing studies, journal editors and reviewers, librarians supporting scholarly communication, and research policy staff developing institutional open science positions.

Prerequisites

Familiarity with the research process in at least one discipline. HE-R02 Research Ethics and Integrity and HE-R04 Research Data Management are complementary but not required.

Syllabus

Module 1 — The Replication Crisis and Its Causes

The reform agenda only makes sense against the evidence that produced it, and that evidence is stronger and stranger than most summaries suggest. This module examines the primary studies.

Lessons. 1.1 The Open Science Collaboration reproducibility project and its critics  ·  1.2 Publication bias, the file drawer and the incentive structure  ·  1.3 Questionable research practices: p-hacking, HARKing, optional stopping  ·  1.4 Statistical power, the winner’s curse and inflated effect sizes

Deliverable. A diagnostic note assessing the replication vulnerability of one published study or one common design in your field.

Module 2 — Preregistration and Registered Reports

Preregistration constrains researcher degrees of freedom, and its value depends entirely on the specificity of the constraint. The module writes one properly and considers the legitimate exceptions.

Lessons. 2.1 Preregistration: what it fixes and what it cannot  ·  2.2 Registered reports and the shift of review before results  ·  2.3 Specifying analyses with enough precision to bind  ·  2.4 Exploratory research, deviation and honest reporting of both

Deliverable. An analytical brief comprising a complete preregistration for a real or planned study, with analyses specified to a binding level of detail.

Module 3 — Open Materials, Code and Reproducibility

Reproducibility requires that someone else can run the analysis and obtain the result, which is a higher standard than sharing a spreadsheet. This module builds the package to that standard.

Lessons. 3.1 Computational reproducibility and dependency management  ·  3.2 Sharing code, materials, instruments and protocols  ·  3.3 Repository choice, versioning and persistent identifiers  ·  3.4 Testing reproducibility with an independent reader

Deliverable. A design artefact: a reproducibility package for one analysis, tested by a peer who reports whether they obtained the same result.

Module 4 — Open Access and the Equity Problem

Open access has succeeded in changing who can read and failed to solve who can afford to publish. The module examines the routes, their economics and their distributional effects.

Lessons. 4.1 Green, gold, diamond and hybrid open access routes  ·  4.2 Article processing charges and their effect on author geography  ·  4.3 Preprints, overlay journals and scholar-led publishing  ·  4.4 Institutional and national open access policy and mandates

Deliverable. An evaluation report comparing open access routes for one real article, with a full cost and equity analysis.

Module 5 — Open Collaboration and Governance

Large open collaborations fail on governance rather than on technology: authorship, decision rights and data access must be settled before the work begins. The module settles them.

Lessons. 5.1 Many-analyst studies, consortia and multi-site replication  ·  5.2 Authorship and contributorship at scale  ·  5.3 Collaboration agreements, data access and dispute resolution  ·  5.4 Equity between well-resourced and under-resourced partners

Deliverable. An implementation plan comprising a collaboration governance agreement covering authorship, data access, decision rights and dispute resolution.

Module 6 — Capstone: An Open Research Plan

The capstone commits the learner to a specific set of open practices for a real project and requires justification of every practice rejected. Uncritical adoption of everything is penalised as heavily as adoption of nothing.

Lessons. 6.1 Selecting practices to fit discipline and constraint  ·  6.2 Building the reproducibility and sharing architecture  ·  6.3 Choosing a publication route  ·  6.4 Justifying exclusions

Deliverable. A 2,500-word open research plan for one project specifying preregistration, data and code sharing, publication route and collaboration governance, with a reasoned account of every open practice deliberately not adopted.

Assessment

Pass mark 70 per cent. All assessed components must be attempted. Every mark in this course is issued by a human assessor; no assessment outcome is generated automatically.

Rubric criteria

Each assessed artefact is marked against four criteria at four levels (distinction, pass with merit, pass, fail).

1. Evidential grasp. The replication literature is engaged with accurately, including its criticisms.
2. Specificity. Preregistrations and reproducibility packages are precise enough to constrain and to be executed by others.
3. Critical selectivity. Practices are adopted or rejected with disciplinary reasoning rather than by default.
4. Equity awareness. Cost, access and distributional consequences of openness are addressed substantively.

Reading list

Core. Open Science Collaboration (2015) ‘Estimating the Reproducibility of Psychological Science’, Science, 349(6251), aac4716.

• Munafò, M. R. et al. (2017) ‘A Manifesto for Reproducible Science’, Nature Human Behaviour, 1, 0021.
• Nosek, B. A. et al. (2015) ‘Promoting an Open Research Culture’, Science, 348(6242), pp. 1422–1425.
• UNESCO (2021) UNESCO Recommendation on Open Science. Paris: UNESCO.
• Suber, P. (2012) Open Access. Cambridge, MA: MIT Press.
• Wilkinson, M. D. et al. (2016) ‘The FAIR Guiding Principles for Scientific Data Management and Stewardship’, Scientific Data, 3, 160018.

Academic integrity and use of AI

Generative AI may be used in this course as a drafting, translation and critique aid, and its use must be disclosed. Every submission carries a short use-of-AI statement naming the tools used, the tasks they performed and the checks the learner applied to the output. Using AI to fabricate data, invent sources, impersonate an interview participant, or produce an artefact the learner cannot explain in a live viva is prohibited.

All citations are verified before submission; a reference that cannot be located by the assessor is treated as fabricated. Fabrication, plagiarism and undisclosed ghost-authorship result in failure of the component and referral to the academic integrity panel. Marks in this course are issued by human assessors, and any learner may be asked to defend a submission orally before a mark is confirmed.