Notes on Blacceleration - Journal #87

i read the post on Gelman’s blog about the Protzko et al. “high replicability” paper, and then i skimmed the published paper, and then i read the review i wrote for an older version of the paper when it was submitted to Nature back in 2020, and after that i was left with 2 questions:

@bmj.com Please look at PubPeer comments on an article you published last week. pubpeer.com/publications/C08779C45DB6E407… I think your research integrity dept shld act swiftly on this one, given clinical significance. I'm aware of even more evidence of problems so let me know if this is not sufficient.

When should a paper be corrected? Last December, @elisabethbik.bsky.social and other sleuths began flagging papers by a prominent bioengineer, Ali Khademhosseini. They found over 80 with image issues. Khademhosseini and his colleagues have issued over 40 corrections, but avoided retractions.

so I just commented on that paper in Cell Host and Microbe that is cited as evidence that the autism/microbiome link is solid: pubpeer.com/publications/9A4B5695950D3396… (not sure what the word “precision” is supposed to mean in the title) tldr; looks p-hacked 🫤 and I couldn’t find the ethics approval 🤔

In our new paper of how Bluesky users discuss retracted papers, we found: ✅ 90% of Bluesky posts show "good practices" (flagging issues/retraction status) ❌ Only 10% show "bad practices" This highlights Bluesky's vital role in responsible science communication! arxiv.org/abs/2605.04334