Author: Sparhawk Mulder
Interview with Sarah Charles
Sarah Charles is currently working on a PhD on the psychobiology of social bonding. She has an undergraduate psychology degree from the University of Sheffield, and a graduate degree from the University of Goldfield. Her work has appeared in the The Times. Alongside her research, she also lectures, both on psychobiology and the movement for Open Science.
What is Open Science?
Open Science is a number of policies which push the contemporary practice of science better fit the Mertonian norms of good science, those norms being:
Communism/Communalism: research produced is collectively owned. Progress in any field is good for all fields. Scientists should be “on the same team” as each other.
Universalism: Scientific validity should be independent of the socio-political status of the person conducting the research. Researchers in different socio-political settings should get the same results (unless the research is about the socio-political environment).
Disinterestedness: Institutions should act for the benefits of all of science, not just researchers and universities, so as to not to create research bias towards profitable fields and conclusions.
Organized Skepticism: upgraded peer review, allow the methodology and institution to be reviewed too.
These goals can be accomplished in a number of ways. The first is the promotion of Public Access. Most cutting-edge research is behind a paywall, and it doesn’t need to be; information can be hosted online very cheaply, and the money from this paywall doesn’t go to the researchers. Also part of the open access plan is the promotion of public educational resources, so that people can better understand those scientific articles. Luckily, this is already partially started: Khan Academy and many Youtube channels (Sarah Charles recommended Philosophy Tube, 60 Symbols, and the Organic Chemistry Tutor) offer simple free online resources, and open source coding softwares allow practice in coding for free.
Most open science policies are geared more toward academia, however. Open Data policies require the submission of not just the results of studies, but submission of all of the data (anonymized if need be), the initial hypothesis, research background, and the team’s thoughts throughout the process. These requirements function as protections against P-hacking (measuring multiple factors and only choosing to look at the data for the most interesting one) and harking (creating the hypothesis after data collection). Those can be tempting shortcuts in an academic climate where getting published is almost the only thing that matters, but they obscure and skew data trends for other researchers. Open Data policies have started to be implemented in many fields, but are not yet universal.
Preregistration and Registered Report are the other biggest Open Science policies. Preregistration is when a study must submit its procedure plan and methodology to a journal before beginning. Registered Report is a step up from that: if the journal approves and the study follows their submitted method, then the journal must publish the study (even with a boring null result). This serves to stop publication bias towards only positive results, since researchers don’t need to worry about getting a result “interesting enough to publish” (and aren’t tempted to cheat), since “not discovering something is itself a discovery” according to Charles.
As for implementation, all of this depends upon the cooperation of funders and the journals. Luckily, many are already on board! Some journals already use the registered report system, and still others give benefits/commendations to studies which follow extra rules like Open Data. Some even award studies badges for extra credibility. As for funders, some large funding groups like Soalition S have agreed to only fund studies which comply with open science guidelines, and some national groups like certain sub-boards of the EU’s science funding department are catching on too!
To many people, this doubtless seems like a far-away postgraduate academic movement, but it’s not! Open Science laws not only make more complex and specific information available to more people (who might need it, if it’s medical information), they also benefit undergraduates and smaller/newer research groups too! Preregistration and registered reporting gives promising and creative studies a better chance to be published, even if they originate from smaller or newer labs. Open data policies help ensure wack results from trimmed datasets don’t get published as easily, and open access facilitates the spread of new discoveries. So most undergraduates should be in favor of it!
The only problem is how to get started. All the extra requirements and policies can be a little confusing, especially to newer researchers. Luckily, the One Science Foundation has step-by-step guides and resources for any prospective researcher, on their website. Additionally, as more and more funders and journals adopt these policies, requirements will become more clear and uniform. But until then, Sarah has one suggestion: spread the word! Tell other people, and help your undergrads!
Thank you to Sarah Charles for giving such an informative lecture on this important topic! Check out (some of) her work here!