If you did, for instance case-cohort sampling and estimated an effect size, you can report that as a RR or an exposure OR (odds of exposure among cases vs non-cases). But that number CANNOT be called an outcome OR which is is how everyone would interpret it if you reported a just “an OR”.

The “case control OR” being referred to in this article is not a thing you should care about at all. When the author finds, eg, that the “cc OR” is collapsible it’s because its only valid interpretation is as an RR which is collapsible. There’s no mystery here. academic.oup.com/ije/article/...

If I learned anything from Groundhog Day it’s that counterfactuals are deterministic, not stochastic.

When you think about it, it’s super weird when article titles are just two variables. Like this one: “adiposity and the risk of aortic aneurysm” Is this a causal study? Descriptive? Predictive? It’s impossible to tell. link.springer.com/article/10.1...

I will never ceased to be amazed when people get an effect size like a risk ratio of 1.5 and then say "there was no effect" because it wasn't significant. Cults should be taking notes on how we can get people to say things like 1.5=1!

If you want to give $$ to support prenatal & infant health, what would be the best time in pregnancy? @dkarasek.bsky.social examines this question using natural experiment related to timing of anti-poverty benefits: authors.elsevier.com/c/1n60~2gOwG... @mdrc-news.bsky.social @ohsunews.bsky.social 1/

kagi really is that good y’all. big stress reliever to have a proper search engine again. (i don’t take compensation for posts as a rule, this is just me wanting them to have a solid subscriber base so they can offer the service long-term)

Happy #ClinicalTrials Day Trial design and conduct is no place for novices with best intentions. Patient participants make sacrifices. Medical progress holds its breath. Large sums of public and private money is invested. So get it right, from the start. statsepi.substack.com/p/dealing-wi...

Some day in the not too distant future journals will just publish titles of papers, no details, everything in supplement.

Why we should rethink causal mediation, and what to do instead? Come to hear the answer from Vanessa Didelez at the next CIIG seminar! The seminar will be hybrid. If you are in London, come join us in person at UCL! Otherwise, you can join on Zoom as usual. Registration links in comment below.

Eurovision Song Contest starting tonight! 🇪🇺🎤🤩

So I wrote part of a peer review in the wrong textbox (for a different article) and now I can't find it at all. Hope I didn't send it to the wrong review.

Proof the laws of physics have no arrow of time. From just this image I'd bet a lot of money that Greenway is sliding toward the bottom of the image and yet...

Lol

The NHL playoffs finish so late in the year that Europe has time to finish its own playoffs AND play the world championship. A bit silly really. #gohabsgo

A slide from this talk I'm giving on Monday. The conclusion will be a set of questions that one must be able to answer to ask associational research questions. The ball is in your court association people!

Well that's the last time I give a bunch of useful feedback in peer review. The authors keep requesting me as a reviewer! UGH

Done with my last talk before leave! This one was fun and I'm glad I added Columbo last minute. Slides here, in case you're curious: juliarohrer.com/wp-content/u...

I think one of the healthiest realizations I had in my life regarding engagement with sociopolitical discourse is that you can, actually, just pursue the good as you see it

This is the first of these memes that get.

the real danger is infraprocessed food

📣 The difference between replicable and not replicable is not itself scientifically replicable. 📣 New work with Erkan Buzbas, showing that verdicts such as "X% of results replicated" are based on an inferential machinery that doesn't work. arxiv.org/abs/2604.26268

My forthcoming book has a chapter on the role of luck in science that looks at forms of luck rarely discussed in history of science. Epistemic luck is all over science.

Can accommodate some more people at this Introduction to INSPECT-SR workshop tomorrow (April 30th). Book here or email/ message me directly: www.trybooking.com/uk/GATM It says (N America) because the time was chosen for that region, but all welcome

I've given "softer" versions of this talk in the past. This time I'm going to give the "all out" version which includes multiple questions that you must be able to answer (including some that aren't in my recent BMJ paper) in order to use associational language in your research question. 1/3

Cole Caufield told the Players Tribune to publish this story right before the Canadiens’ first home playoff game. You won’t want to skip reading this.

Don't be silly, Pluto isn't manipulable.

Good news everyone 🥳 Our (w @vincentab.bsky.social) primer on models as prediction machines (with the marginaleffects package) is finally officially published!> journals.sagepub.com/doi/10.1177/...

Ugh. Here comes the language police saying I can't use the word association! (You definitely can use the word association but to mean what it's actually supposed to me!)

Me to students: you should talk about consistency and positivity in your papers Published causal studies: pubmed.ncbi.nlm.nih.gov/40037557/

Causal inference conferences: welcome to my talk. First, assume all causal assumptions hold…

Causal inference courses: here's how you estimate causal effects and these are the assumptions you rely on and by the way the assumptions are always violated kthxbye

Graceland. When it came out, in 1986, it was quite a surprise, even considered "risky" and "experimental" by some people. Simon flirted with African-influenced music but the result was extremely successful; joyous and uplifting. Still a great listen after all these years.

A problem in science is we spend a disproportionate amount of time on problems that more tractable rather than on more difficult problems.

And not only do you get those benefits from paying taxes, another huge positive is you get to live in a society where everyone gets those benefits.

The fact that Mamdani just seems like a cheerful normal guy who's policies are things like "we're going to tax second homes to fund pothole repairs" or whatever and yet this generates panicked responses like "This is basically Stalin's great purge turned up to eleven" is... telling. It tells things.

Claims and ideas which happen to favour the ruling class don't dominate public life by accident. They're paid for by oligarchs and corporations then promoted by media outlets other oligarchs and corporations own. Once they've been drummed into our ears enough, they're internalised as "common sense".

"The rise of extreme wealth is one of the clearest signs of this imbalance. In 1987, billionaires held wealth equal to 3% of global GDP. Today this tiny elite, just 0.0001% of the world population, owns the equivalent of *16%* of world GDP in wealth."

One upside of this is that it means I get to re-use this joke I made in the past

Absolutely stunned that the Society for Epidemiologic Research chose John Ioannidis to deliver the conference keynote in the Year of Our Lord 2026.

apparently @societyforepi.bsky.social selected one of these doctors (ioannidis) to deliver their conference keynote this year. it's disturbing that legitimate public health organizations are STILL platforming people who spread deadly misinformation. hopefully they will read this and rethink

Recording now available if you've missed it:

I do not regret to inform you that we are going to win

You can follow the results for Hungary here. Looks very much like Tisza has won. Tight for a supermajority. And of course we don't know how Orban will react. europeelects.eu/2026/04/04/n...

Statisticians are trained in how to compute statistics, but not why to compute statistics. At least 10% of their training should be in philosophy of statistics. They should also spend 10 credits on their own empirical research project to understand how data is acually collected.

You will never convince me that we won't see more incredibly creative art like this with universal basic income. I want UBI for many reasons, but especially this reason. I want more people exploring new ideas and pushing boundaries and getting really good at stuff they love. I 100% adore AdP. 🏆

One year after our original Guardian essay, Naomi Klein and I are excited to reveal the North American cover of our forthcoming book END TIMES FASCISM - and the Fight for the Living World. It will be on bookshelves on September 15th and is available for pre-order. More details below. bit.ly/4cbLfEr

I mean obviously as a certified non posable action figurine appreciator I'm a biased source here, but I do basically agree with this. Being unironically enthusiastic about weird niche things is one of the true joys of living in a society and if you can gain access to it you should.

Poverty is a key driver of health; our latest article reviews the growing literature on economic policies as an upstream solution. However, several policies have insufficient evidence, and data barriers remain: www.annualreviews.org/content/jour... @hsph.harvard.edu @irpwisc.bsky.social

This is so, so well-articulated.