My latest books cartoon for @theguardian.com

My comments studies that trito assess the spread of fabricated references in @nature.com . These papers are important but the uncertainty is high and confounds many. Stating as other commentators said that these papers establish a lower bound is an oversimplification www.nature.com/articles/d41...

Hubby, a statistician, trying to help out a friend from education w analysis of her survey data: Friend: I dropped all incomplete observations from the data Hubby: Uh, maybe send me the raw data to see how we should handle missing data Friend: Here, raw data, but I dropped everyone who seemed fishy

Can't help thinking what a woman Martha Mills would have become with @meropemills.bsky.social and @paullaity.bsky.social for parents.

This pretty much says it all. Thanks to the brilliant @estheraddley.bsky.social www.theguardian.com/society/2026...

In this cohort study of 367 authors and 1330 retracted RCTS, 6 superretractors coauthored 1/5 of all retracted trials, while 18 top-cited scientists w >10 retractions coauthored 1/4 jamanetwork.com/journals/jam... @ioanaacristea.bsky.social @metricstanford.bsky.social cc: @retractionwatch.com

1. The paper with the implausibly large effects of Omega-3 fatty acids on mental health was now retracted. A little thread on the process where @ianhussey.mmmdata.io and I was involved. www.sciencedirect.com/science/arti...

I rarely do content warnings but this is very bad. Like with some movies, I had to take breaks. Why this is so bad is the core of all traumatic events: the impossibility to understand the why. Men, get treatment. Srsly

In the light of the recently announced restrictions on reapplication to the ERC competitions, I have sent the following letter to the @erc.europa.eu President @marialep.bsky.social I assume she received many, I can only hope she will find the time for mine or that others made similar suggestions.

Our article is online in @jamanetworkopen.com. First two authors (and responsible of the bulk of the work), my stellar PhD students Chunwei Lyu and Minoo Matbouriahi (funded by @sharectd.bsky.social) w/ @floriannaudet.bsky.social and John Ioannidis. jamanetwork.com/journals/jam...

"Every 1% increase in military spending drives a 0·62% reduction in public health spending." On the cover, a new letter examines the relationship between military spending and public health expenditure during conflicts. Find out more in our latest issue: spkl.io/63325ANG6W

The ERROR project recruits independent experts to recheck social sciences papers’ data, statistics, methodology, code; now the project plans to publish the reviews in a new peer-reviewed journal. science.org/content/arti... @dalmeet.bsky.social @science.org #reproducibility

📣The [faked] PREVENT-TAHA8 trial paper was #retracted from @bmj.com after 5 long months 😅 🤝 well done @deevybee.bsky.social @steamtraen.eu @sophieehill.bsky.social @wjst.de @ioanaacristea.bsky.social @gidmk.bsky.social P.T. Wilmhurst and anonymous @pubpeer.com users #ResearchIntegrity 🔍 #FoSci

Can I just propose a sensitivity analysis called ‘excluding studies where researchers did not due their due diligence in fabricating the data”? Obvious fakes excluded?

In life either you die or you live long enough to see any villain being out-villained by an even greater villain

Why does my 4-year old (2nd child) keep morbidly selecting book about the arrival of a new baby, which she then listens to with dismay and horror? And also both (7yo & 4yo) are very into the new Peppa Pig w/ baby Ivy. Both THEIR parents could not have been clearer on this front: NO more babies.

From my collection, unpublished/rejected letters to the editor, which I share as preprints, because I think the conversation is important LSD for generalized anxiety disorder trial in @jama.com jamanetwork.com/journals/jam... Our letter: osf.io/preprints/ps...

Nearly identical reviews (abstract, images, the text I checked except the rephrased conclusion), the same @elsevierconnect.bsky.social journal, 3 years apart (2010 vs 2013) by overlapping authors (one common, 1 distinct) www.sciencedirect.com/science/arti... www.sciencedirect.com/science/arti...

RETRACTED FOR THE SECOND TIME: phony study on "biofield energy" treatment neurocritic.blogspot.com/2026/03/retr... An update on the saga of a supernatural intervention published in peer-reviewed scientific journals cc: @retractionwatch.com

@dingdingpeng.the100.ci still a long way to go with causal inference though. Everyone: I know about the mistakes. Don’t @me. I already am struggling with 3 phrases about challenges of living abroad which is Hausaufgabe.

In short, don’t dichotomize. It just gives the illusion of clarity, while losing information.

Really curious what tenant of modern parenting has made it so that your kid wants you and me to exchange phone numbers to play with my kids, but you are like ‘oh it’s not necessary, we’ll see each other around’. Privacy? Attachment? Sheer insanity?

The very same people (I have two big names of open science in mind, add their fans) who literally predicted 5-6 years ago that researchers who don’t code will be out of science (haha) are now predicting LLMs will be doing most of the science in 5 years. Cassandra, something’s gotta give.

New paper, on a worrying trend in meta-science: the practice of anonymising datasets on, e.g., published articles. We argue that this is at odds with norms established in research synthesis, explore arguments for anonymisation, provide counterpoints, and demonstrate implications and epistemic costs.

What is the best reporting guideline acronym? ROB-ME by @mjpages.bsky.social gets my preliminary vote. Any other good ones, #AIMOS2025?

Oh finally. We can like many things about Denmark, and still dislike their vaccination model. We have a more comprehensive schedule in Italy. Thank God.

Mesmerizing. You cannot watch this and remain unmoved.

My favorite genre of Epstein emails are powerful people emailing WITH JEFFREY EPSTEIN about how MeToo has gone too far and the woke mob must be stopped.

And modestly I am proud of myself that without any substantial causal inference training (on my list this year along with relearning German, learning French and Gen alpha…for now) I intuitively thought of many of these limitations. And some designs (RCTs) to me should only have robustness analyses

But data is available from the authors upon reasonable request. For a meta-research paper this is simply ridiculous Honestly, read our paper more pubmed.ncbi.nlm.nih.gov/35623542/

TFW you reply to all but it was the wrong all including the person who should not have read the mail. 😐

Academics vying for a spot in Epstein‘s world. There are so many. I feel the need to make a thread, so I don’t keep confusing them. 1/

It absolutely baffles me that in America high-profile professors at the top higher-education institutions (before cuts and anything) would secure funding by literally courting a millionaire sexual predator. How did you declare that funding on submissions again?

Who knew the real sociopaths were all studying morality and social behavior

Alison Luchs, who has worked at the National Gallery of Art for 47 years, agreed to learn Gen Z slang and make videos because she wanted to raise interest in the museum’s art. She never expected to slay. https://wapo.st/45BXc3S

📣 Our introduction to structural causal models in science studies is now published: doi.org/10.1162/QSS.... @tklebel.bsky.social and I tried to make our introduction as accessible as possible. We illustrate the theory by three case studies based on a simulated model of Open Science. 🧵(1/6)

Sometimes, rarely, I read a study which tackles a problem in such an elegant or just smart way, I experience the rare feeling of genuine admiration.

I just gifted two separate people "Work in Progress" (thanks @dingdingpeng.the100.ci) and now I know my new thing is reading pre-2022/2023 articles about topics everyone is panicking and saying not much new today worksinprogress.co/issue/the-sp... worksinprogress.co/issue/real-p...

My aerobic fitness instructor literally said to me "there are some things you do really well" (some jumps everyone else had abandoned) and it brought back the glowing pride of when the teacher read my essay in front of class in 1999/2000. (She probably meant it like "even you do something well" 😂)

Aaron Beck said this. About cognitive therapy. 55 years ago 😬

One of my concerns with AI-assisted research in biology is that we're already looking only in the lamplight and AI incentivizes us to narrow the beam. Seems like it's happening... www.nature.com/articles/s41...

New PubPeer comments: @ioanaacristea.bsky.social on the post-hoc study registry = violation of the Helsinki declaration) @ianhussey.mmmdata.io with a full table of effect sizes (many impossibly large, d > 7) me just about the large SMD of the main result and what this means (shit vs. chocolate)

How long should it take to retract a paper with incontrovertible signs of data fabrication? Sleuths think 2 months is too long, particularly when clinical risks are involved. deevybee.blogspot.com/2026/01/an-o... #retraction #stemcells #cardiology @erictopol.bsky.social

Which year is this? How old am I? Live Updates: U.S. Captures Venezuelan Leader, Trump Says www.nytimes.com/live/2026/01...

Don’t tell that to the humble brags who simultaneously communicated they are 1. Speakers of multiple languages; 2. Protected from dementia (haha); 3. Epidemiologically and methodologically illiterate

Colleagues describing my ERC project

Two victims of previous school shootings — Mia Tretta, who was shot in the stomach in 2019 at Saugus High School, and Zoe Weissman, who survived the 2018 shooting at Marjory Stoneman Douglas High School — are now at Brown. Gun violence is out of control in this country.