High Replicability of Newly-Discovered Social-behavioral Findings is Achievable

• DOI: 10.31234/osf.io/n2a9x
• Publication Date: 2020-09-10T11:09:29Z
• Authors: John Protzko; Jon Krosnick; Leif D. Nelson; Brian A. Nosek; Jordan Axt; Matthew Berent; Nick Buttrick; Matthew DeBell; Charles R. Ebersole; Sebastian Lundmark; Bo MacInnis; Michael O'Donnell; Hannah Perfecto; James E Pustejovsky; Scott S. Roeder; Jan Walleczek; Jonathan Schooler
• Keywords: Replicate; Replication; Sample (material)

Failures to replicate evidence of new discoveries have forced scientists ask whether this unreliability is due suboptimal implementation methods or presumptively optimal are not, in fact, optimal. This paper reports an investigation by four coordinated laboratories the prospective replicability 16 novel experimental findings using rigor-enhancing practices: confirmatory tests, large sample sizes, preregistration, and methodological transparency. In contrast past systematic replication efforts that reported rates averaging 50%, attempts here produced expected effects with significance testing (p<.05) 86% attempts, slightly exceeding maximum based on observed effect sizes sizes. When one lab attempted discovered another lab, size replications was 97% original study. high rate justifies confidence rigor enhancing increase discoveries.