entity Updated 2026-07-24 Tags: Project, Research-Methods, Social-Science, Reproducibility

Replication Games

Replication Games are the hackathon-style research-auditing events described in Don’t hate the replicator, hate the game. Abel Brodeur first tested the format in Oslo in October 2022 after an open workshop invitation drew roughly 70 to 80 registrants.

The format is cooperative rather than prize-driven. Teams spend a day using [[ReplicationPackage|replication packages]] to reproduce published results, then run selected Robustness Checks to see whether conclusions depend on fragile choices. The Montreal event in the episode had 71 replicators and showed why Crowdsourced Academic Auditing can change research norms even when most checked papers survive the day.

Connections