Statement on Statistical Significance and P-values. The world’s largest professional association of statisticians, the American Statistical Association, has made a public statement regarding the use of P-values in research. This statement, presented by Wasserstein and Lazar, aims to provide some clarification on the proper use and interpretation of the P-value which ‘can be a useful statistical measure but it is commonly misused and misinterpreted’. Understanding the meaning of P-values is clearly important and essential to build a strong basis for good statistical practice. It is worth noting that this statement, no matter how official, is not novel: dozens of previous publications have already emphasized the importance of understanding the concepts of null hypothesis testing and P-values. However, this message has not been recognized and it seems unclear what needs to be done so that some action is taken. The answer is very simple: we, the biomedical researchers, can only change the way we currently perform data analysis if we know and understand what the proper alternatives are. This is something that is not being communicated clearly at the moment and that is well worth an effort. (
Junior biomedical scientists and preprints. This article in F1000 Research by Gary McDowell is focused on preprints, the purpose of the ASAPBio meeting, potential issues and benefits, and ‘whether submitting preprints are a worthwhile endeavor’. ASAPbio was created as a focal point for engaging the biology community in a discussion about the role that preprints could play in communicating results in life sciences. A preprint is a manuscript in a finished form that has not yet been published in a traditional journal but is available on preprint servers like bioRxiv or the Self-Journal of Science (SJS). The practice of publishing preprints is already quite often used in some fields of physics and mathematics since the creation of the ArXiv preprint server in 1991 but still not very well accepted in other life science areas. As the author states: ‘Preprints can provide evidence of the work and mind of the scientist in a more up-to-date fashion and, divorced from the impact factor of a particular journal, the work itself must be considered‘.
Team up with industry. Industry-academia partnerships produce highly reproducible work because their incentive-structures are different, argues Aled Edwards, chief executive of the Structural Genomics Consortium, in a Comment piece in Nature. He distils eight principles that make for reliable research: 1. Require full commitment, and reward efficiency; 2. Define objectives that cannot be achieved with current technology; 3. Establish clear quality criteria and make them public; 4. Mandate data sharing; 5. Subject work to independent oversight before public release; 6. Enshrine public ownership for all research outputs; 7. Ensure that industry and academic scientists collaborate and 8. Create an active governing body. Edwards concludes that ‘adherence to these principles builds an ecosystem that supports reproducible, innovative research. Scientific publications are no longer the sole units of achievement; reaching predefined milestones and making useful tools are also key to continued funding.’ (