Making Sense of Reproducibility in Published Science

• Activity Number: 358
• Type: Roundtables
• Date/Time: Tuesday, August 5, 2014 : 12:30 PM to 1:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Victoria Stodden*+
• Companies: Columbia University
• Keywords: reproducible research ; open data ; open code ; publication standards ; replication ; science policy

Abstract

In the last year, there has been a near firestorm of press articles, journal commentary, and publications on the topic of reproducibility of published scientific results. How much of what we publish is reproducible? How much is capable of being reproduced? How does computational work measure up, compared to bench science? What do we make of the new publication standards by high-impact journals such as "Nature" and "Science"? How much of the issue is poor statistical procedures versus outdated reporting standards? Are open data policies improving reproducibility? Do we need open code standards as well? What should be the role of the scholarly society, funding agency, journal, institution, and individual researcher in ensuring reproducible research? This roundtable discussion will untangle these questions with an eye to policymaking and standards to address issues in reproducible research.