Introduction

Open science is not a single technique but a way of organizing research around transparency, reproducibility, accessibility, and collaboration. This intensive PhD summer school introduces both the rationale behind open science and the practical tools researchers can use to integrate these principles into their own work.

The course begins with the replication crisis and the research practices that contributed to it. It then moves through preregistration and pre-analysis plans, reproducibility and replicability, replication packages, open data, and transparent dissemination. Throughout, the emphasis is on translating principles into practice: participants critically evaluate published research, draft preregistrations, work with reproducible research tools, and consider how open-science practices can be adapted to quantitative, qualitative, primary-data, and secondary-data research.

Open science is treated neither as an all-or-nothing checklist nor as a demand to make everything public. Particular attention is paid to the trade-offs researchers face when working with sensitive data and to the broader question of how to be as open as possible while remaining ethically and methodologically responsible.

Learning goals

By the end of the course, participants should be able to:

  • Explain the replication crisis and how it motivated more transparent research practices.
  • Understand the core principles and practices associated with open science.
  • Apply open-science principles to their own research projects.
  • Design and write preregistrations and pre-analysis plans appropriate to different research designs.
  • Create and evaluate replication packages and reproducible research workflows.
  • Communicate analytical decisions and research results transparently.
  • Make informed decisions about open data sharing, open-access publishing, and other forms of research dissemination.
Course format
  • Four intensive days combining lectures, demonstrations, discussion, critical reading, and hands-on exercises.
  • Practical work with preregistration platforms and reproducible research tools.
  • Exercises designed around participants’ own research whenever possible.
  • A participant-driven final day in which the group selects additional open-science topics to explore.
Assessment

Successful completion requires attendance at all sessions. Participants can also submit either a preregistration or a replication package applying open-science principles to their own research and receive detailed feedback. Assessment is pass/fail.

Course material

Open science and the replication crisis

The first day asks why open science became necessary in the first place. It introduces the major principles of open research and examines the replication crisis in the social sciences through concrete cases and empirical evidence.

The session focuses on the incentives and research practices that can undermine the credibility of published findings, including selective reporting, flexibility in analytical decisions, publication bias, low statistical power, and questionable research practices. At the same time, it introduces open research practices as institutional and methodological responses to these problems.

Rather than treating failed replication as a problem confined to one discipline, the readings draw on communication science, psychology, economics, and political science to examine how reproducibility and replicability have become central concerns across the social sciences.

Lecture material
Main readings
Additional readings
Practice material

Preregistration, pre-analysis, pre-what?!

The second day turns to one of the most visible open-science practices: preregistration. The goal is not simply to learn how to complete a template, but to understand what preregistration can—and cannot—solve.

The session distinguishes preregistrations, pre-analysis plans, and registered reports and compares platforms such as the Open Science Framework and AsPredicted. Particular attention is paid to the decisions that should be specified before analysis and to the inevitable trade-off between precision and flexibility.

Preregistration is also treated as a flexible practice rather than a one-size-fits-all protocol. The readings consider its application to experimental research, secondary-data analysis, qualitative research, and more complex analytical workflows. Standard operating procedures provide a further way of making recurring analytical decisions explicit before researchers encounter the data.

Lecture material
Main readings
Additional readings
Practice material

Reproducibility and replicability

The third day moves from planning research to making the completed research process reproducible and replicable. Participants examine large-scale replication initiatives such as Many Labs and consider what replication projects can teach us about the reliability, generalizability, and cumulative nature of social-science evidence.

A major practical focus is the replication package: how to organize data, code, documentation, and output so that another researcher can understand and reproduce an analysis. The session introduces the DA-RT and FAIR principles and uses reproducible reporting tools such as R Markdown to connect data processing, analysis, and reporting.

The module also considers several less obvious threats to reproducibility, including poorly documented data, inconsistencies in reported statistics, and low statistical power. The broader objective is to move from reproducibility as an afterthought to reproducibility as a feature of the research workflow from the beginning.

Lecture material
Main readings
Additional readings
Practice material

Everything else you always wanted to know

The final day puts the principles of open science and open education into practice by allowing participants to determine the agenda. Rather than fixing the complete program in advance, participants identify the open-science questions that matter most for their own research and vote on the topics to explore.

Possible topics include open-access publishing and preprints, open peer review, open education, multiverse analyses, inclusivity and diversity in open science, citizen science, the statistical foundations of open science, and reproducible research software.

For the 2025 edition, participants selected four topics: multiverse analysis, R Markdown, open data sharing versus privacy concerns, and open peer review. Together they highlight an important theme running through the entire course: openness is rarely a mechanical rule. Researchers must make transparent and defensible choices about analytical uncertainty, documentation, privacy, and the dissemination of knowledge.

Lecture material
Topics covered
  • Multiverse analysis — making consequential analytical choices visible by examining the set of reasonable analyses rather than reporting only one analytical path.
  • R Markdown — integrating code, results, and prose in a reproducible document.
  • Open data and privacy — balancing the benefits of data sharing against confidentiality, consent, disclosure risks, and other ethical constraints.
  • Open peer review — examining alternative approaches to transparency in the evaluation and publication process.
Practice material

Open-science toolbox

The course concludes with a set of resources participants can continue using after the summer school. These platforms, communities, principles, and podcasts provide starting points for implementing open science and for keeping up with debates about reproducibility and research transparency.

Platforms and communities
Podcasts
Additional reading