METRICS Conference: improving biomedical research 2015: Challenges and solutions
Last modified: June 28, 2017
by Jelte Wicherts
When I left Amsterdam for my trip to the 2015 METRICS conference at Stanford University, an autumn storm had hit my home country. Dark clouds and gusty winds resembled feelings I sometimes have when considering the common lack of transparency, rigor, and replicability of findings in the scientific literature. But when I arrived at the meeting organized by Steven Goodman and John Ioannidis and their colleagues at METRICS, my mood rapidly improved.
In a way, the METRICS conference restored my optimism as if the Californian sun dissipated the clouds of low reproducibility. The meeting featured fascinating talks by top researchers in meta-research, policy makers, educators, journal editors, science writers, funders, and other stakeholders. Speakers and attendees were well aware of the problems faced by contemporary science and considered ways to improve rigor, transparency, and reproducibility in science.
The organizers had asked presenters and the many actively involved attendees to focus on solutions rather than problems. Discussions were sharp and engaging and the breakout sessions were aimed at formulating concrete solutions at the different levels of the scientific system (data transparency & open science, evaluation of science, peer review, research policy & incentives, overall methods, education and engagement).
There were plenty of good proposals and things would certainly improve if we would implement these today. But just like the weather, the scientific system is complex. A few days of sun does not mean it is summer. The METRICS conference highlighted some of the core problems in the (biomedical) science and highlighted that much work and continued concerted efforts are needed to further improve it.