Nature Human Behaviour published A manifesto for reproducible science last Tuesday. The manifesto was written by an international group of prominent academics, led by John Ioannidis of Stanford University. The authors propose a number of measures to increase the reliability, credibility and efficiency of scientific research. Too often, they write, scientific discovery is actually nothing more than self-deception.
Publish or perish
One of the authors of the manifesto is Eric-Jan Wagenmakers, professor in psychological methodology at the University of Amsterdam. “The aim is to make science healthier”, he told Dutch newspaper NRC on Wednesday. Wagenmakers said that we are currently living in ‘a culture of publish or perish’. “For the individual scientist, achieving results has become more important than achieving the truth.”
Because scientists are under heavy pressure to see novel and unexpected patterns in data, there is a risk of ‘being misled by our tendency to see structure in randomness’, the manifesto reads. Simply put, scientists are encouraged to see animals in the clouds or a face in the moon – and so they do. The manifesto emphasizes that it’s not only up to researchers to improve science, but that stakeholders like universities, scientific journals, research institutions and industries have a responsibility as well.
The authors propose a number of ways to improve four major aspects of research: methods, reporting and dissemination, reproducability, and evaluation and incentives. The goal is to move towards better, more reliable and more useful science. “When we are doing science, we are trying to arrive at the truth. In many disciplines, we want that truth to translate into something that works”, John Ioannidis said. “But if it’s not true, it’s not going to speed up computer software, it’s not going to save lives and it’s not going to improve quality of life.”