%0 Unpublished work %T Comparison of Preregistration Platforms %+ Analysis and Visualization (AVIZ) %A Haroz, Steve %8 2022-02-24 %D 2022 %R 10.31222/osf.io/zry2u %Z Cognitive science/PsychologyPreprints, Working Papers, ... %X Preregistration can force researchers to front-load a lot of decision-making to an early stage of a project. Choosing which preregistration platform to use must be therefore be one of those early decisions, and because a preregistration cannot be moved, that choice is permanent. This article aims to help researchers who are already interested in preregistration choose a platform by clarifying differences between them. Preregistration criteria and features are explained and analyzed for sites that cater to a broad range of research fields, including: GitHub, AsPredicted, Zenodo, the Open Science Framework (OSF), and an “open-ended” variant of OSF. While a private prespecification document can help mitigate self-deception, this guide considers publicly shared preregistrations that aim to improve credibility. It therefore defines three of the criteria (a timestamp, a registry, and persistence) as a bare minimum for a valid and reliable preregistration. GitHub and AsPredicted fail to meet all three. Zenodo and OSF meet the basic criteria and vary in which additional features they offer. %G English %L hal-03588379 %U https://inria.hal.science/hal-03588379 %~ CNRS %~ INRIA %~ INRIA-SACLAY %~ INRIA_TEST %~ TESTALAIN1 %~ CENTRALESUPELEC %~ INRIA2 %~ UNIV-PARIS-SACLAY %~ UNIVERSITE-PARIS-SACLAY %~ LISN %~ GS-ENGINEERING %~ GS-COMPUTER-SCIENCE %~ LISN-AVIZ