21 September 2018 Vol 361, Issue 6408
Special Issue; Metaresearch
A recipe for rigor
By Kai Kupferschmidt
Science21 Sep 2018 : 1192-1193 Full Access
A simple strategy to avoid bias—declaring in advance what you will study, and how—is rapidly catching on.
When a series of scandals hit the field of psychology in 2011 and eroded trust in a lot of research results, some scientists proposed a radical solution: preregistration, or describing the research they plan to do, and how, before they gather a single piece of data. Researchers hope this will lead to more negative results being published. But most importantly, it limits what scientists can do with their data, making practices like p-hacking or HARKing, which can lead to results that only seem to be significant, less likely. The practice has taken off since then, with the number of preregistrations doubling every year. Metaresearchers are now trying to suss out what positive or negative effects the new approach has. But that is proving difficult.