Lots of impt discussion on preanalysis plans after @The_JOP announced they will be req'd for experimental studies

Short 🧵 on some of what we know about PAPs, their effects, what researchers think of them from scientist surveys & census of posted PAPs 👇 https://twitter.com/The_JOP/status/1350385896315617280
Census of every PAP in EGAP and AEA registries and survey of PAP authors by @gofosu @DanielNPosner finds big variation in what is in PAPs, but authors do not report many of the bad consequences (e.g., scooping) critics worry about.

https://www.georgeofosu.com/files/Ofosu-Posner-191212.pdf
Big survey of social scientists by @g_theonion @betsylevyp @polarwander @tedmiguel @reblitt et al finds fast adoption of open science & our support outpaces our guesses about what our colleagues think. We think < 1/3 support PAPs but actually > 1/2 do.

https://osf.io/preprints/metaarxiv/5rksu
Best estimate of fx of registration on publication bias I know of comes from @alfang @maqartan @grantmgordon analysis of medical registries. No effect. They note medical registries ask for minimal details about designs, so fx of PAPs may differ.

https://albertfang.com/research/ps_paper.pdf
On publication incentives, @gofosu & @DanielNPosner analyze NBER working papers and find that those with PAPs are less likely to be published than those without PAPs — but more likely to be accepted into a top-5 econ journal and have more citations

https://www.georgeofosu.com/files/main.pdf 
Lots more in them: what people are writing in their PAPs, how often they are followed in papers, how views of PAPs vary by age, what PAP writers and broader community think about objections to PAPs, & detailed stories about authoring PAPs and the review process...
Big missing piece: @prof_mirya, @jrpjrpjrp, @Timlongman & others raised important questions of equity & inclusion in requiring PAPs in a top journal. We need more thinking on this. I don't see this discussed in these papers; pls point out if I missed it. https://twitter.com/prof_mirya/status/1350453577265930241
https://twitter.com/dbroockman/status/1350668409600557058
You can follow @graemedblair.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.