Another new blog article, elaborating on ordinal regression within the cumulative probability model family. One general model contains an amazing array of special cases used all the time. Do we need to keep teaching special cases&assuming so much? www.fharrell.com/post/cpm/ #Statistics #StatsSky

New blog article proposing a general statement of the fundamental question a standard RCT should answer and providing robust and powerful statistical methods to answer it: www.fharrell.com/post/rctmod/ #Statistics #StatsSky #EpiSky #clinicaltrial #rct #pharma @stephensenn.bsky.social

Statistical Models Answer the Fundamental Clinical Question and Provide Clinical Trial Estimands – Statistical Thinking

Specific goals and estimation targets for randomized clinical trials have still not been well defined for general outcome variables. Proponents of causal inference calculus have claimed to define goal...

fharrell.com

Weekend reads: Young girl’s death after gene therapy never made public; FDA retracts lettuce test result; ADA delays editorial publication after controversy

Weekend reads: Young girl’s death after gene therapy never made public; FDA retracts lettuce test result; ADA delays editorial publication after controversy

If your week flew by — we know ours did — catch up here with what you might have missed. The week at Retraction Watch featured: Exclusive: A couple’s daughter died after they paid more than $…

retractionwatch.com

Here's a useful page with a **ton** of common tasks in #RStats, done with parallel syntax for tidyverse vs. data.table vs. base. arelbundock.com/posts/dt_tb_... #rdatatable #tidyverse

Weekend reads: Private info in ‘most’ arXiv preprints; why did George Church decline authorship; 28 misconduct cases in China

Weekend reads: Private info in ‘most’ arXiv preprints; why did George Church decline authorship; 28 misconduct cases in China

If your week flew by — we know ours did — catch up here with what you might have missed. The week at Retraction Watch featured: ‘Disappointed’: Cochrane journal asked researchers to publish article…

retractionwatch.com

The Middle TN Chapter of the American Statistical Association @amstatnews.bsky.social is hosting a half-day applied improv course in Nashville, Friday 8/14 8:30am-12:30pm w/lunch and parking included. No experience required; sign up by July 22 for 10% discount. www.eventbrite.com/e/fundamenta...

Fundamentals of Applied Improvisation for (Bio)statisticians

This half-day course introduces applied improvisation as a novel way to pursue interpersonal skills as a quantitative scientist.

eventbrite.com

the most surprising thing about this is that Twitter never tried this first

Aaron Rupar@atrupar.com · 3w ago

This is nuts: "Trump Media has unveiled a paid-for, licensed data feed that will give banks ​& trading firms 'the fastest' access to posts from influential Truth Social ‌accounts Trump’s posts on the platform regularly move markets. High speed access to these could be very valuable to traders"

What an incredible resource!! I've added a link to it from hbiostat.org/data and hbiostat.org/bbr/ancova. How did you find this Darren? Having ready-to-analyze #RStats RDS files is wonderful. #rct #Statistics #StatsSky

Darren Dahly@statsepi.bsky.social · 3w ago

🤩 bingkaiwang.github.io/RCT_Bench-ma...

🚨 Protocol -> preregistration -> preprint -> publication 🚨 125 clinical prediction model articles developed on unreliable data on stroke and diabetes. Since preprinting our results 7 articles have been retracted. doi.org/10.1186/s129...

It is so sad to lose Norm. I learned so much from him-one of the few researchers/teachers credentialed in both comp sci and #Statistics; he used that dual expertise to great effect. He contributed greatly to the community&engaged in vigorous and thought-provoking discussions to the end. #RStats

Dirk Eddelbuettel@eddelbuettel.com · 4w ago

Norm Matloff (1948-2026) @matloff.bsky.social I was made aware of the post below made by his daughter. We lost another friend and valued long-time contributor the #Rstats community, gone far too soon, and sure to be missed. His memory will be a blessing.

I finished translating my CV from a complex mess of LaTeX constructions to a clean and much simpler @typst.app document. Typst handles multiple bibliographies out of the box, among other things. It also instantly processes BibTeX files without creating auxiliary files sub as .bbl.

Diagnostic models fail because we optimise sensitivity instead of probability calibration. @f2harrell.bsky.social explains why dichotomising outcomes destroys information and what doctors actually need 📊

UM Division of Biostatistics Seminar Series (2013-14) - Frank Harrell Jr, PhD

For more info and seminars, visit:http://biostat.med.miami.edu

youtube.com

Brown University's response to widespread student laziness has been underwhelming

Jacob T. Levy@jacobtlevy.bsky.social · 4w ago

www.insidehighered.com/news/faculty...

Calling FIFA

Mayo Clinic’s former director of research operations alleges she was sidelined, then terminated, for raising objections over lax standards of studies involving AI.

Former executive accuses Mayo of cutting corners on AI research

Mayo’s former director of research operations alleges she was sidelined, then terminated, for raising objections over lax standards of studies involving AI.

startribune.com

At a basic level, I think this depends on what you care about. If you want quick pubs for a grade or a promotion, just do t-tests. If you are interested in your data, learn how to fit and interpret ordinal models.

Dr Mircea Zloteanu 🌺🌞🍃@mzloteanu.bsky.social · last mo.

#statstab #574 When the Mean is Misleading: a Guide to Ordered Regression for Meaningfully Modeling Ordinal Outcomes Thoughts: The "it's fine to analyse likert/ordinal data with t-test" camp is not doing themselves any favours #ordinal #likert #probit #regression link.springer.com/article/10.1...

Nice article that calls for more replication studies but lacks sufficient emphasis on avoiding bad design and analysis in the first place. Initial non-replicating studies can often be shown to fail #Statistics standards: www.polytechnique-insights.com/en/columns/s... #StatsSky

The reproducibility crisis: is science deceiving us? - Polytechnique Insights

The reproducibility crisis: is science deceiving us? – Read the column on Polytechnique Insights

polytechnique-insights.com

Weekend reads: Taylor Swift teaches botany; NEJM retracts key study in Amgen drug; hidden prompts at conference ‘snare AI peer reviews’

Weekend reads: Taylor Swift teaches botany; NEJM retracts key study in Amgen drug; hidden prompts at conference ‘snare AI peer reviews’

If your week flew by — we know ours did — catch up here with what you might have missed. The week at Retraction Watch featured: Physicist in Iraq fired over publishing scam claims fake Columbia aff…

retractionwatch.com

Fantastic talk about making fantastic talks, plus a link to a great book about making R graphics. #RStats #StatsSky #Statistics

Kieran Healy@kjhealy.co · last mo.

Somewhat more constructively, and in lieu of me saying something rude.

I couldn't leave the USA without giving my DAG talk one last time! Come and join my 'farewell lecture' in New Haven (or online!) on 15th July at 4pm Eastern Time! Sign-up to attend in-person or online at: yale.zoom.us/webinar/regi...

I found this surprisingly thoughtful and balanced: r-statistics.co/R-vs-Julia.h... (although some of the labeling on Figure 2 is a little confusing)

R vs Julia for Statistics: Julia Is Faster, But Is That What Matters?

Julia benchmarks faster than R on tight loops, but R ships 21,000+ statistical packages and a huge community. Honest trade-offs with real benchmark code.

r-statistics.co

Excellent discussion Julia. @stephensenn.bsky.social would have a field day with this paper.

Julia M. Rohrer@dingdingpeng.the100.ci · last mo.

Had to read this one to figure out whether it's as click-bait-y as the title. As somebody who has written a lot about causal inference with observational data, I'm quite conflicted about it.> asteriskmag.substack.com/p/the-case-a...

Several great colleagues published an excellent paper on multiplicity adjustment. Wish it considered that "assertions are too easy" is miscast as multiplicity. See hbiostat.org/bayes/bet/mu... @sabinehoffmann11.bsky.social @gscollins.bsky.social @georgheinze.bsky.social

Nice discussion of an important upcoming presentation by @stephensenn.bsky.social to the @royalstatsoc.bsky.social on the importance of statistical modeling in randomized studies: discourse.datamethods.org/t/royal-stat... #Statistics #StatsSky #rct #clinicaltrial

Excellent new blog article by Steve Ruberg: analytixthinking.blog/2026/06/28/b... #Statistics #rct #pharma #clinicaltrial #StatsSky

No. 24: Estimands Part 1 – Just Do ITT?

Randomization and complete data on the randomized experimental units are needed for cause-and-effect inference. What happens when that logic is broken?

analytixthinking.blog

On the fourth anniversary of the overturning of Roe v Wade remember that they lied to get their seats on the Supreme Court

Carl Sagan hat das ja auch sehr eindrücklich erklärt...

📘 With the release of our textbook "Bayesian Workflow" (avehtari.github.io/Bayesian-Wor...), I figured I'd also share the content of my graduate course on the topic at UBC. 🌎 charlesm93.github.io/stat547/ The course contains overlapping and complementary material, homeworks and reading.

Bayesian Workflow book: Website – Bayesian Workflow book

Website for the Bayesian Workflow book by Gelman, Vehtari, McElreath, et al. — case studies, code, and exercises in R and Stan.

avehtari.github.io

The forensic report is available here: storage.courtlistener.com/recap/gov.us... Incredible stuff

Nick Brown@steamtraen.eu · 2mo ago

The testimony of Julian Ackert, the forensic expert in the Gino case, is fascinating for computer nerds. <raises hand> Whoever changed the metadata might have gotten away with (at least part of) it if they had remembered to include seconds in the timestamp.