4 ms·
It seems like a sensible check is for this collaboration to include original studies, like the one mentioned in the article lede, that have already been replica
by jessriedel 11y ago
It seems like a sensible check is for this collaboration to include original studies, like the one mentioned in the article lede, that have already been replicated elsewhere. (Ideally they would keep blind the relevant members of the collaboration to this fact.) Then when you say "we failed to replicate X% of the studies" you also say "of the subgroup that had already be replicated, we failed to replicate Y%". If Y isn't much smaller than X, you know the replication collaboration is probably botching this.