Richard Dawkins Proposes a Solution to the “Reproducibility Crisis"

A university student with a focus on physics, computer science, and math, Matt Kafker is an avid reader who enjoys works of scientific inquiry. One of Matt Kafker’s favorite science writers is Richard Dawkins, who was interviewed by Scientific American in August 2017. 

One topic that Dawkins addressed was the “reproducibility crisis,” which has created serious concerns about the reliability of scientific research, particularly in the medical field. At the root of this is a tendency for the results of studies to be summed up in neat ways that are designed to gain wide dissemination in the media but don’t stand up to attempts at replication by other scientists.

In addition, a “file drawer effect” exists, whereby those papers that do not disprove the null hypothesis, or accepted status quo, do not achieve publication, as they are considered too boring. As Dawkins views it, this can lead to the spread of falsehoods. This is because enough studies on a subject will eventually generate some outliers that “yield statistical significance,” even in cases where the null hypothesis is true. 

One way of resolving the file drawer effect proposed by Dawkins is for scientists to publish on the Internet their intention to undertake experiments beforehand. The results are then shared with the scientific community, regardless of whether the findings are positive or negative.