Reproducibility of the Statistical Literature: Is Fooling Our Deans Enough?

• Activity Number: 46
• Type: Contributed
• Date/Time: Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Scott Emerson*+
• Companies: Univ of Washington

Abstract

Recently there has been much attention focused on the lack of reproducibility of many analyses presented in the applied scientific literature. However, it is not immediately clear to me that the statistical literature fares any better in this regard. I believe that my perceptions are due to a break down in the peer review process. Some main line statistical journals have gone on record as valuing speed of review over quality of review and novelty of topics over correctness of the publication. In this talk I present some recent examples of what I believe to be a failure of the peer review process in the statistical literature, and I present some suggestions that I believe can greatly improve the reproducibility of the statistical literature. This in turn may facilitate more reproducible applied science literature.