Reposted by Ben MoranThe Onion @theonion.com · 16/09/2026Gas Station Price Sign Using Scientific Notation 283659748
Reposted by Ben MoranMatt Kerlogue @matt.kerlogue.co.uk · 09/09/2026It’s only causal inference if it comes from the Inférènce region of France, otherwise it’s just a sparkling correlation. 031
Reposted by Ben MoranIván Díaz @idiaz.bsky.social · 10/09/2026(1/22) Over the last few days I have been discussing the role of formal causal inference in RCTs, and in general the role of formalism and rigor in statistics in twitter with @f2harrell.bsky.social. Thread with an example for why both are fundamental for biomedical research: 3135
Reposted by Ben MoranRaj LB @rajlb.bsky.social · 06/09/2026Nobody knows how paracetamol works www.tandfonline.com/doi/full/10....tandfonline.comParacetamol (acetaminophen): A familiar drug with an unexplained mechanism of actionParacetamol (acetaminophen) is undoubtedly one of the most widely used drugs worldwide. As an over-the-counter medication, paracetamol is the standard and first-line treatment for fever and acute p... 2125
Reposted by Ben MoranVincent Arel-Bundock @vincentab.bsky.social · 04/09/2026Big news! 🎉 𝚖𝚊𝚛𝚐𝚒𝚗𝚊𝚕𝚎𝚏𝚏𝚎𝚌𝚝𝚜 1.0.0 for #Rstats is out. It’s a big number and it feels like a big step. I wrote a blog on the challenges of interpreting statistical models, SPEED, cool new features, the future, and a 5 year package development and writing odyssey. arelbundock.com/posts/margin... 1642299
Reposted by Ben MoranPierre-Simon Laplace @learnbayesstats.bsky.social · 31/08/2026In this special episode of Learning Bayesian Statistics, @alex-andorra.bsky.social is joined by Andrew Gelman, Aki Vehtari, and Richard McElreath to discuss Bayesian Workflow. From simulation and hierarchical pooling to causal inference, there’s a lot to unpack 🎧 lnkd.in/geX2QkxV #Bayesianlearnbayesstats.comBayesian Workflow - Gelman, Vehtari & McElreathAndrew Gelman, Aki Vehtari, and Richard McElreath discuss their new book on Bayesian workflow, reverse Bayes and hierarchical pooling. 13710
Reposted by Ben MoranAndrew Gelman et al. @statmodeling.bsky.social · 25/08/2026Bayesian Workflow free pdf! statmodeling.stat.columbia.edu/2026/08/25/b...statmodeling.stat.columbia.edu Bayesian Workflow free pdf! | Statistical Modeling, Causal Inference, and Social Science 17429
Reposted by Ben MoranLars Mølgaard Saxhaug @load-dependent.eurosky.social · 22/08/2026Editors should not accept for publication papers using the phrase «real world evidence». It’s a marketing term, to embellish non-causal studies. #metasci #stats #cardiosky www.ahajournals.org/doi/10.1161/...ahajournals.orgStopping Oral Anticoagulation After Catheter Ablation for Atrial Fibrillation in Patients With High Risk of Stroke: STOP‐OAC Study | Journal of the American Heart Association 263
Ben Moran @benmoran.bsky.social · 14/08/2026I feel like you’ve violated my eyes!!static.klipy.comSimpsons: My Eyes, The Goggles Do Nothing!ALT: Simpsons: My Eyes, The Goggles Do Nothing! 010
Reposted by Ben MoranAki Vehtari @avehtari.bsky.social · 13/08/2026All four books I've co-authored are freely available online for non-commercial use: - Bayesian Workflow at avehtari.github.io/Bayesian-Wor... Links to other three books are in the quoted post 👇 (too many books to fit in one post!)avehtari.github.ioBayesian Workflow book: Website – Bayesian Workflow bookWebsite for the Bayesian Workflow book by Gelman, Vehtari, McElreath, et al. — case studies, code, and exercises in R and Stan. 8285134
Reposted by Ben MoranBen Harrap @bharrap.bsky.social · 12/08/2026The Statistical Society of Australia has a new President... No, it's not Randy the greyhound (although he would've done a fine job) It's me! If you want to know a little more about me, I have the blog post just for you :) benharrap.com/post/2026-08... #statssky #datasciencebenharrap.comGet to know the new President of the Statistical Society of Australia (it’s me) – Ben Harrap 13768
Reposted by Ben MoranCameron Patrick @cameronpat.bsky.social · 12/08/2026I, for one, welcome my new overlord 1131
Reposted by Ben MoranRafe Meager (they/them) @economeager.bsky.social · 11/08/2026we have got to put that into a causal inference lecture 743271
Reposted by Ben MoranJulia M. Rohrer @dingdingpeng.the100.ci · 08/08/2026New blog post in which we discover the secret to immortality: accounting errors. www.the100.ci/2026/08/08/m...the100.ciMy Immortal Time BiasPreamble: This shortie is my attempt to come up with an example that makes the problem underlying immortal time bias as obvious as possible. In Japan, people who reach the age of 100 are handed a co... 128031
Ben Moran @benmoran.bsky.social · 07/08/2026I feel the same about the wakefield package! Genius! github.com/trinker/wake...github.comGitHub - trinker/wakefield: Generate random data setsGenerate random data sets. Contribute to trinker/wakefield development by creating an account on GitHub. 000
Reposted by Ben MoranRobin Blythe @rbly.bsky.social · 01/08/2026Me, once a month #statsky #rstats 3374
Reposted by Ben MoranJulia M. Rohrer @dingdingpeng.the100.ci · 30/07/2026New blog post! Odds ratios are...a bit weird. Here I explain two problems (interpretation, non-collapsibility) and provide some simple recommendations for improved reporting. www.the100.ci/2026/07/30/r...the100.ciReviewer notes: Odds ratios are really oddWhen psychological researchers investigate binary outcomes, they routinely report odds ratios to quantify the strength of an effect or an association. This is quite understandable given that psycholog... 2428879
Reposted by Ben MoranJack Iwashyna @iwashyna.bsky.social · 31/07/2026Thank you for your kind words 2 responses: 1/ Source control is good for patients with sepsis, so if we can find the right patients, this could be more than just marginal for those patients 2/ ICU care in general is a game of inches; much of excellence is just cumulating marginal benefits, right? 171
Ben Moran @benmoran.bsky.social · 28/07/2026It’s not nice to kink-shame the ML-bros (TM)! I am assuming it’s the bad kind of paddling, of course! 020
Ben Moran @benmoran.bsky.social · 28/07/2026Doesn’t exist anymore. One of the major tenets of the medical profession is to interfere with natural selection. I’ve got T1DM- I should’ve died at age 7. I’ve got bad genes that shouldn’t be propagated, but here I am, procreating with reckless abandon!! H/T to Idiocracy for the insight! 010
Ben Moran @benmoran.bsky.social · 28/07/2026You know you’re getting old when the consultants look like interns! 210
Ben Moran @benmoran.bsky.social · 28/07/2026Yeah, but have you written a book about John Snow that when Nolan makes said movie, it will take you to a place of insurmountable money, sex and power!?! 101
Ben Moran @benmoran.bsky.social · 27/07/2026It’s a tough read, especially from someone who has sold over 1M copies of their book since it was announced Nolan was making the movie. 010
Ben Moran @benmoran.bsky.social · 26/07/2026I was being sarcastic. I was referring to ye olde times (TM) when estimands were niche statistics, rather than mainstream. I’ll show myself out… 110
Ben Moran @benmoran.bsky.social · 26/07/2026Ha! These young ones with their gaming! They can’t even spell estimate properly! 100
Reposted by Ben MoranCritical Care Reviews @criticalcarereviews.com · 25/07/2026Our latest blog is out: The Nothing-to-Lose Fallacy criticalcarereviews.com/2026/the-no… 0117
Ben Moran @benmoran.bsky.social · 23/07/2026I think we’ve reached the point that it is unethical to enrol critically ill patients into a vitamin C trial. It increases harm. Research funding is siphoned away from studies that may actually benefit pts. 030
Ben Moran @benmoran.bsky.social · 23/07/2026100%!!!!!! I’ve banned the word AI in my house. My youngest calls them LLMs around me, although I’m sure he rebels when he’s with his friends! I suspect this battle will join other worthy ones: - Independent risk factors - The term “association” - WIYEMF!?! 050
Reposted by Ben MoranCameron Patrick @cameronpat.bsky.social · 23/07/2026I fear this is a losing battle, but I'm trying very hard to avoid using the term "AI". a word that can refer to anything from LLMs/GenAI through to control systems and prediction models through to linear regression has no useful meaning at all 6344
Reposted by Ben MoranCameron Patrick @cameronpat.bsky.social · 22/07/2026happy pi approximation day (22/7) to all those who celebrate 0304
Ben Moran @benmoran.bsky.social · 21/07/2026I find that viewpoint fascinating! It explains some of our divergent views. Can you expand on your rationale? Also, is that a common belief in the US? 000
Ben Moran @benmoran.bsky.social · 21/07/2026I agree with you that learning how to employ research into practice is not a trivial process. But observational research deserves the same rigor that we place on RCTs if we use it in our clinical practice. 100
Ben Moran @benmoran.bsky.social · 21/07/2026I said they should understand the process of CI, just like they do for an RCT. There’s more obs research out there than RCTs, so I think that they should understand the process. Likewise I think that if you are actually doing obs research you should know the basics of CI (excl descriptive studies). 100
Ben Moran @benmoran.bsky.social · 21/07/2026Why? K isn’t that high (granted >5, but <10). The other thing I would also look at is the RE normality assumption and try different RE distributions (t, skew-normal, mixture). 010
Ben Moran @benmoran.bsky.social · 21/07/2026My point was that clinicians incorporate research into their daily practice. They should be aware of the process of CI in observational studies to help with critiquing it. I’m not asking them to go and do it, but be aware of it and discard the crap. 110
Ben Moran @benmoran.bsky.social · 21/07/2026Absolutely agree with the clinical component, but that knowledge comes from somewhere. In Australia we do a research project to complete training, and the ability to have an overview of the process enables that understanding. Otherwise bad research can become the basis for those guidelines. 010
Ben Moran @benmoran.bsky.social · 21/07/2026I agree with the sentiment in that I think they should know about the causal inference process: Causal estimand -> identification -> statistical estimand -> estimator -> estimate With quantitative bias analysis to test for identification assumption violations. This would stop a lot of pain. 210
Ben Moran @benmoran.bsky.social · 21/07/2026Duh, he’s a rapper from London. Dood, get with the times!! 020
Ben Moran @benmoran.bsky.social · 21/07/2026I’d be worried about RE identification with a small(ish) number of studies. Test: - Common-effect model - Arm-based model (one-stage) - RE model with informative priors on tau. At least with partial pooling you get shrinkage estimates. Then compare model fit with LOSO-CRPS. 130