The Institute for Replication (I4R) works to improve the credibility of science by promoting and conducting reproductions and replications.

Pinned Tweet
🧵1/ Our first meta-science paper (with 350+ coauthors) is published today in Nature. It presents one of the largest-ever reproducibility projects in economics & political science. Here’s what we found 👇
9
230
585
108,654
Over the past two years, I4R partnered with Psychological Science on a large reproducibility project. 67 independent teams reproduced 64 articles published in 2024–2025, about 44% of eligible papers. Paper: econstor.eu/bitstream/10419/… Here's what we learned 🧵
1
15
39
6,413
See Appendix for hyperlinks to the 67 reports and authors' responses or our website: i4replication.org/reports
1
1
131
Reach out to instituteforreplication@gmail.com if you'd like to participate in Round 2! We are now checking articles that went through the STAR team and should have pre / post comparisons in a year from now.
1
135
1/ Episode 2 from our 1-day I4R workshop is out. Jan Marcus, Josephina Weinerova & René Bekkers ask: How can we link publications to replications, so anyone reading a paper can instantly see whether its findings have been retested? 🧵
1
1
4
1,624
3/ We discuss what better infrastructure could look like: Who should create these links—journals, publishers, metadata systems, replication databases, or researchers? And how should we distinguish reproductions, robustness checks, and different types of replications?
1
1
286
Our comment on “Raising Health Awareness in Rural Communities” is finally published in REStat (MIT Press). But the retraction we had been told was coming never happened. Can lawyers derail a retraction? This case raises that uncomfortable question. 🧵
2
25
56
13,125
Again, our criticism here is NOT directed at the REStat editors. We thank them for pursuing this and ultimately publishing the relevant record. The concern is at the publisher level (MIT Press). If legal pressure can stop or indefinitely derail a retraction that...
1
4
22
7,450
editors believe is warranted, that is a serious problem for scientific publishing. The scientific record should not depend on who is willing to lawyer up.
19
1,458