Supplementary materials for: The Status Quo of FAIR Data Sharing in Psychology

Data sharing is paramount for reproducible science, yet few studies have examined whether shared data meet FAIR (findable, accessible, interoperable, reusable) principles, often relying on small journal samples. Hence, we conducted a comprehensive analysis of 11,384 psychology datasets (2013–2024) indexed in PsycInfo and PSYNDEX, assessing FAIRness with the F-UJI tool. While data sharing has increased, FAIRness remained limited: datasets were generally findable and interoperable but less accessible and reusable. FAIRness varied mainly by repository, but not substantially by year, subfield, journal, or author characteristics. Higher reusability was linked to increased citation rates. Compared with F-UJI, humans rated datasets higher for accessibility but lower for interoperability and reusability. To improve data FAIRness, we provide a brief checklist for editors and reviewers that adds practical guidance beyond F-UJI. Finally, we highlight the value of coordinated efforts by repositories, authors, and journals.

Data sharing is essential for reproducible science, but quality varies. FAIR standards (Findable, Accessible, Interoperable, Reusable) aim to improve this, though data sharing differences remain across psychology subfields, journal policies, and authors’ practices. Automated tools for FAIR assessment are promising, though validation for psychology datasets is needed. A total of six research questions will be addressed, covering three main objectives: a) providing a general overview on data sharing in psychology, b) assessing the eligibility of an automated FAIR assessment tool, and c) inspecting relations of data FAIRness with repositories and psychology subfields as well as journal and article characteristics. Publication references with links to datasets will be retrieved from the psychology literature databases PsycInfo and PSYNDEX. Open publication metadata will be retrieved from OpenAlex. The FAIRness of the data will be automatically assessed using the F-UJI tool. To determine the validity of the tool’s results, a sample of 80 datasets will be manually coded by two human raters.