Articles:
Power failure: why small sample size undermines the reliability of neuroscience
Power-up: A Reanalysis of ‘Power Failure’ in Neuroscience Using Mixture Modeling
UK funders demand strong statistics for animal studies
Science Isn’t Broken - It’s just a hell of a lot harder than we give it credit for.
Reproducibility: Seek out stronger science
Book:
Rigor Mortis: How Sloppy Science Creates Worthless Cures, Crushes Hope, and Wastes Billions
Website:
Every researcher has a story about a compelling result that didn’t reproduce. Sometimes this is due to technical challenges or the experimental approach. However, there are also biological mechanisms that contribute to irreproducible results, and their importance is increasingly recognized by researchers, drug developers, funding agencies, and the general public.
In this discussion, we consider the issue of data reproducibility in preclinical research through a biological lens, focusing on often-neglected variables such as sex, age, microbiome, and circadian rhythms. The roundtable brings together stakeholders from across the biomedical enterprise to explore practical ways of controlling for these variables to maximize the translational potential of preclinical research. The participants include investigators Michael Holtzman, Dana Philpott, and Jeffrey Haspel, who encounter these variables in their research; Patricia Valdez, representing the NIH Extramural Research Integrity Office; and NPR Science Correspondent Richard Harris, who moderates the discussion and brings a lay perspective.
The goal of the discussion is to explore both the tensions and opportunities for exciting science that irreproducible results create in preclinical research. It also considers economical ways for researchers to address biological drivers of irreproducible results, to comply with new standards set out by funding agencies, and to enhance the potential of their work to lead to novel therapies.