Anton Musil retweeted
pokud vás téma #metascience a #openScience zajímá, v zimním semestru se v pátek ráno můžeme setkávat v novém kurzu Opens Science in Practice na @FSV_UK
is.cuni.cz/studium/predmety/…
- mix @ReproducibiliT
a praktického semináře, otevřen pro všechny studenty na @CharlesUniPRG
příští úterý (29.9.) na FSV UK pořádáme dva otevřené metascience workshopy: @giladfeldman ukáže, jak využít AI ve výzkumu a jeho kontrole, a @BartosFra makes meta-analysis easy with JASP. registrace na forms.gle/8srinjHrqB26J44S6, více info ve 🧵👇
1
2
230
Save the date — Perspectives on Scientific Error 2027 is heading to Copenhagen, 27th to 29th of January 🇩🇰
Join us alongside keynotes Stephan Guettinger, Evangelia Christodoulou, Adeyemi Adetula and James Heathers
Sign up for newsletter at errorsin.science
#Metascience
2
3
332
Preprint authors say "highly cited papers show more significant reductions in citation after retraction and garner more attention to their retractions as they occur."
cambridge.org/core/journals/…
4
339
#statstab #555 To Be FAIR: Theory Specification Needs an Update
Thoughts: what's the use of better stats if our theories are nonsense?
#theory #metascience #fair #epistemology #research
journals.sagepub.com/doi/10.…
1
4
182
To me all this suggests that (a) there should be a lot more focus on adversarial collaborations; and (b) academia needs to work out ways of running many @I4Replication type projects full time.
Pre-publication peer review, by itself, has severe limits. #MetaScience
Asking for robustness in ref reports wastes everyone's time and DOES NOT WORK.
Authors selectively respond to requests, interpret things differently, and still have many degrees of freedom. This is a stupid dance.
Robustness checks are only useful if done by 3rd parties. 2/
3
265
📢 New publication in Nature! #MetaScience
A major international collaboration investigating the robustness of social & behavioural science results includes DSE research fellow @MatteoMMarini1 among its authors 🌍🔬
➡️ rdcu.be/fbewp
🧵Here's what they found 👇
1
2
5
511
Great piece by @samturner330849 on my approach to fixing the reproducibility problem in computational science. Worth a read 👇
happeningscommunity.substack…
#OpenScience #Metascience #Holochain #ResearchIntegrity
1
3
3
225
Why does reporting quality matter? 🧩 A meta-epidemiological review buff.ly/KI1s5DE of 239 SMT trials shows significant variability across studies, raising concerns for evidence interpretation.
Standardized reporting must improve.
#ClinicalResearch #MetaScience #SMT
3
14
1,121
"The replication crisis in Psychology has undermined both intra-scientific and public trust. This review shows–using Wilholt's notion of epistemic trust–that replicability is fundamental for epistemic reliability" #metascience
sciencedirect.com/science/ar…
1
11
28
3,454
"Ultimately, the idea is simple: if we better understand science, we can better improve science."
@LSEImpactBlog @LSEblogs
spr.ly/6016Acyir
#SciencePolicy #Metascience
2
3
124
"Ultimately, the idea is simple: if we better understand science, we can better improve science."
@LSEImpactBlog @LSEblogs
spr.ly/6010fnJ0n
#SciencePolicy #Metascience
2
4
122
On our way for a research planned with AI, evaluated with AI and done with AI… the question here is <with, by, or assisted by> AI #metaScience
1
3
280
NEW PAPER in @JClinEpi looking at the content of open peer-review reports of RCTs in the field of oncology
sciencedirect.com/science/ar…
Peer review reports [...] were short (median 276 words), superficial, and rarely constructive #metascience #peerreview #RCTs
1
14
40
2,393
Really interesting history piece accompanying the launch of the UK #Metascience Unit's first year report (by @RoRInstitute's James Wilsdon, @LudoWaltman, Andre Brasil + the Metascience Unit's own Ben Steyn)
assets.publishing.service.go…
I need to post one small correction though...
1
2
255
Very happy to be part of this inspiring community @Metascience2025. Two lovely thoughts: “sorry bloody Bibliometrics!!!” And “precision does not entails prescription” #metascience @OSLab_UC3M
2
244
On "navel-gazing" in science..
and why we should rethink how we use this term in research discourse.
🧵 1/12
#SciSci #MetaScience #MetaSci #ResearchOnResearch #RoR
1
1
5
226
⏰ for the coming online events #OpenTalk this month:
@klebel_t from the Know Center (know-center.at/en/) will share their latest work on "The Academic Impact of Open Science"👏👏👏
#openscience, #metascience, @hcp4715
3
4
565
#metascience What would increase the rate of discovery more now AI can catch junk grants (no hypotheses; hangers-on dipping their beak for 5%, no pre-reg, N, or power calcs, Bad stats, no replication, closed data, over pricing, salami slicing, ignoring existing art, etc.)
1. AI instant-feedback: Author fixes & reuploads.
2. AI auto-fail, saying why, allow reapply in a year.
3. Make the AI a lead-reviewer, producing a clear, on the record demolition of bad grants, forcing the humans to say they don't care about truth if they want to fund their buddy or fear retaliation.
40%
Feedback and re-upload
40%
Feedback and reject
20%
Honest AI at the table
5 votes • Final results
1
2
290