Yeah, I’m thinking something along those lines. Another example would be a test of differences in reports of adverse events between an active and placebo treatment

I'm more interested in auxiliary tests, which aren't the central outcomes but would partly (or fully) invalidate the focal outcome. So studies would either not be published at all (less likely) or the test result itself simply wouldn't be reported.

Yeah, that's also my experience with trying out some "reverse" selection models via simulation. Good to hear it's not just me!

Yeah, that’s the sort of thing I’m thinking. But how to evaluate this in a body of research? I have some ideas but want to find out what others have done

Like a scenario where someone does a computational reanalysis, finds no issue, then doesn't write anything up?

Nooooooo! This instance was nowhere near as consequential as your case

Perfect

Publication bias is the idea that non-significant results are less likely to be published. But what about the opposite case where a *significant* result is less likely to be published? Anyone have any papers or tools for detecting and/or correcting for “reverse” publication bias?

wait, what am I supposed to select here?

Fascinating insights into the early history of Zotero, including the fact that the name is the Albanian word for "to learn well or master". Hats off to everyone involved in this project. Also if you're not using Zotero, why not? Dump that closed rubbish you're using now and switch.

Great to work with @benjurek.bsky.social and Julia Winter on this one

Yeah, I get the feeling its an overzealous editorial assistant (human or AI) signing on behalf of the Editor

Thanks for this heads up, I was actually considering this journal for an upcoming submission of another manuscript. On paper, the journal seems like a great concept, so it's a shame that manuscript handling looks like a mess

I've had two recent instances where journals keep making vague requests to make changes to submitted manuscripts. I do my best to make the change, but they still tell me there's an issue. I ask for more specific info regarding the request, but the manuscript keeps getting returned to me with no info

Without actually sending your scripts/data to other people, it can be hard to ensure others can easily use your materials. This looks like a great solution. Another cool feature is that users can also tweak scripts to see how conclusions can change when different parameters are used.