An open letter to evolutionary and human sciences; Statistics has moved on and so should we. A proposal for more transparent research, and some notes on p < 0.003

Abstract

Statistical reasoning and inference have become integral to how we reach conclusions as scientists. Few papers in evolutionary, biological, and human sciences, are published at present without at least one report of a p-value, or the term “statistically significant”. As a product of this, discoveries and results often hinge on p-values below the infamous threshold of 0.05. This is due to how well ingrained these notions are in many of our higher education systems, and our practices as researchers. Nevertheless, the concept of a p-value, borrowed from statistics, and dating as far back as the late 20th and early 21st century, has undergone an evolution that we may not be completely aware of, if we do not follow or keep ourselves up to date with the most recent advances in statistical research. The present short communication can be framed as an opinion piece that simply aims to call our attention to the fact that statisticians, in recent years, have asked that we completely abandon the notion of “statistically significant (p < 0.05)”. It is important to point out that none of these observations are new, however certainly merit being re-addressed. Here I attempt to open a dialogue among archaeologists, palaeontologists, palaeoanthropologists, biological anthropologists, evolutionary biologists, and all researchers in related fields, about such statistical research. Finally, I propose a means of moving forward, suggesting we move from a world where p < 0.05 is considered a binary threshold for conclusive results, to a more nuanced practice where different gradients, down to p < 0.003, are interpreted as increasingly suggestive evidence – provided that findings are supported by additional, transparently reported corroborative data.