Approximately two thirds of published scientific articles are estimated to be non reproducible, likely due to questionable research practices (QRPs). This is the opposite of science. ‘Science’ is not just a result, it is a method.
The solution/ deterrent for many QRPs is pre-registration of study protocols and the associated hypotheses being tested using publically available platforms. This infrastructure already exists: osf dot io, aspredicted dot com, clinicaltrials dot gov, zenodo dot org, etc.
Proposed policy: Following implementation on a given date, publically funded entities shall priviledge and give preference to scientific results that have a clear documentation of registration. Evidence bases driving public policy should be based on pre registered studies only. The registered versus non-registered evidence bases should be compared to point out areas and directions of further investigation.