NSF Reproducibility Framework 1 A Framework for Ongoing and Future National Science Foundation Activities to Improve Reproducibility, Replicability, and Robustness in Funded Research Prepared for the Office of Management and Budget

The confidence in and reliability of science and engineering research is truly invaluable and especially so at the National Science Foundation (NSF). Although outright fraud or falsification in science is rare, a number of factors can compromise the reproducibility of published studies in science and serving the public interest requires guarding against them. “Reproducibility”, “replicability” and “robustness” are broad terms that encompass research aspects that relate to confidence in published findings. Across NSF’s directorates and disciplinary domains, various science and engineering fields and subfields require different practices to build confidence in their results. Nevertheless, all directorates within their domains of practice value reproducible, replicable, and robust science . This may involve proposal of a new hypothesis, testing of a hypothesis under new conditions, confirmation, if not exact replication, of an original result, and reliable extensions of findings. Thus issues of reproducibility, replicability, and robustness in research are closely related to the nature and practice of scientific research, which differ across domains.