This document presents a comprehensive overview of the issues surrounding reproducibility and replicability in scientific research, as addressed by a committee of experts led by David B. Allison. The committee was tasked with defining these concepts across diverse scientific fields, investigating the prevalence of non-reproducibility and non-replicability, reviewing current improvement efforts, and considering their impact on the credibility of science within both the academic community and the general public. The committee emphasizes that improvements are needed, particularly in achieving reproducibility, which is technically challenging due to factors such as inadequate record keeping, non-transparent reporting, data obsolescence, and cultural barriers. The terms themselves—reproducibility (achieving consistent results using the same data and methods) and replicability (achieving consistent results across different studies addressing the same question)—are often confused, and clarity is crucial.
While problems exist, the committee cautions against declaring a crisis or becoming complacent. It stresses that neither reproducibility nor replicability alone guarantees scientific reliability; rather, the robustness of scientific knowledge emerges from diverse evidence, thorough synthesis, and transparent reporting of methods, uncertainty, and variability. Recommendations are provided for all stakeholders, including educational institutions, researchers, funders, journals, journalists, and policy makers. These include enhanced training in computational and statistical methods, improved archiving of data and code, clearer communication of uncertainty, and the need for multiple lines of convergent evidence in policy and media decisions. The document concludes that while challenges persist—particularly with the growing complexity and computational nature of science—a strong culture of transparency, methodological rigor, and continuous education can strengthen the reliability and public trust in scientific findings.
