Apropos of same discussions over and over, I want to remind you all that we have three works showing why results of replication studies do not necessarily mean reality or discovery of truth. Here’s the most recent one and you’ll find the others linked in the thread as well. https://twitter.com/zerdeve/status/1255168642938880001

We need many conditions being satisfied for meaningful replications and at the minimum these conditions require a good theoretical understanding of the phenomena we study. https://twitter.com/zerdeve/status/1255168670831005696

We cannot use reproducibility as a demarcation criterion. We should value replications not indiscriminately but when we can properly interpret and learn from them. What the original research question was and what it was based on matter a lot.

I’ll also link to this awesome thread by @IrisVanRooij on a talk I gave a while back. The link to the talk is in the thread as well but see this result first https://twitter.com/irisvanrooij/status/1111779141756563457?s=21

I’d really like to see replication champions someday meaningfully engage with our work and reevaluate some of their talking points in this light. There’s a lot that can be learned from the theoretical work we do which can be practically useful. It’ll only make for better science!

I’ll also add this tweet and thread here. https://twitter.com/zerdeve/status/978836297618870273?s=21