NASEM Report Analyzes Factors for Reproducibility and Replicability in Scientific Research
In response to a Congressional directive to conduct a comprehensive study of issues related to reproducibility and replicability of scientific research, the National Academies of Sciences, Engineering, and Medicine has released a new report, Reproducibility and Replicability in Science. The consensus study report defines the terms “reproducibility” – or getting similar results using the same research methods – and “replicability” – getting similar results across multiple research methods – as they relate to research practices. The report sets forth several steps on how to improve the reproducibility and replicability of research, including identifying clear descriptions of how the reported result was reached, providing training for scientific institutions on proper statistical analysis, investing in tools and infrastructure that support reproducibility, encouraging journals to consider reproducibility factors in publications, and having the National Science Foundation (NSF) facilitate the sharing of data for NSF-funded research. The report also considers how the public’s confidence in scientific findings can be improved by two factors: first, that scientists avoid overstating the implications of their research to public-facing audiences, and second, that journalists report on scientific results with “as much context and nuance as the medium allows.” The full report can be found on the National Academies website.