In many cases changes to a statistical model can improve things - the paper provides an example where the differences would be important in terms of policy or intervention.

Off the shelf statistical models don't in general do that - they're about modeling patterns in data and can be meaningful in the sense of worthwhile, useful and so on which is the sense of the term being used here (and in most contexts).

I'm not sure I understand the latter point unless its about the semantics of the term meaningful, there's as argument that all models are wrong and some useful. In this case I think the ordinal model is more useful than the standard linear one.

I don't disagree with the first part - the claim here is that a particular model of mean effect can be misleading.

His blog is very good too: reluctantcriminologists.com

I can't see Jonathan Brauer on Bluesky but he is on linked-in (where I first saw this).

They general principle won't surprise most statistically aware people (except maybe LPM proponents), but the point is to demonstrate just how misleading it can be in a fairly innocuous use case where the average effect is estimated pretty accurately by either model.

When the Mean is Misleading: a Guide to Ordered Regression for Meaningfully Modeling Ordinal Outcomes - Journal of Quantitative Criminology Journal of Quantitative Criminology - We demonstrate the utility of ordered regression models for analyzing ordinal outcomes frequently encountered in criminology, comparing their predictive...

Just saw this new paper by Brauer on how linear models of mean effect can be very misleading for ordinal data link.springer.com/article/10.1...

Introducing the EEG and ERP Methods Template: Q&A with Gisela Govaart and Antonio Schettino Interview with Gisela Govaart and Antonio Schettino, developers of the new ERP preregistration template on the Open Science Framework (OSF).

Now available on the OSF as part of a growing collection of preregistration resources, the new EEG & ERP Methods template guides researchers through every stage of ERP study planning. In our Q&A, two of its creators share how the template can help researchers at all stages:

We're thrilled to open registration for the Utrecht Replication Games. The event will be at the at the University of Utrecht on June 4th. Psych, public health, pol sci and econ studies will be reproduced!

Register here: www.surveymonkey.ca/r/Replicatio...

Graph of award probability of R35 and R01 from NIH factbook as a function of review rank percentile. As is apparent, 2025 is a significant departure, with lower award probabilities at all scores <40 and significant departures from norm, where even being in the top 10% is no longer a nearly certain indicator of success. Data source: https://report.nih.gov/nihdatabook/report/302

The data is in: the NIH goalposts have shifted.

What were once almost certain fundable scores have become coin flips and what used to be likely grants have become aspirational, leading to fewer awards.

Another manifestation of how HHS policies have led to fewer awards and less science.

Politics London discussing driverless taxis casually referring to “jaywalking pedestrians”.

Jaywalking is not a thing in UK law.

Transport for London and Government had better not be asleep at the wheel on this.

if you're an older person who finds words like mogg and maxxing annoying, just starting using them. people over the age of 30 have the superpower to end trends by simply adopting them

Exclusive: Key US infectious-diseases centre to drop pandemic preparation Staff members have been instructed to scrub this topic and ‘biodefense’ from the agency’s website.

“Staff members at the United States’s premier infectious-disease research institute have been instructed to remove the words “biodefense” and “pandemic preparedness” from the institute’s web pages, according to e-mails Nature has obtained.”

www.nature.com/articles/d41...

Meme. Photograph of the sculpture "Eel Walker" by Mark Newman, showing naked woman walking an large eel on a leash. The eel -- a massive moray-looking thing -- is floating along off the ground, so we can assume that they are walking through a stream, or on the bottom of the ocean, or through space. OR that it's a magical eel, and it just flies. The eel is fierce and terrible, and probably is expensive to feed. The woman is done in a golden bronze color. Her hair is floating backwards, suggesting that there is a breeze or an ocean current catching it. I've cleverly (or obnoxiously) written the meme text over her breasts. Sorry, folks; she saving those for the eel. She is muscled and powerful, and the way her left foot is places in front of makes her seem more like a dancer than a walker. She looks like the goddess of eels, taking one of her minions out to gnaw into the dreams of unbelievers and scientists. Meme text reads: "Let's talk eel sex, baby"

Tomorrow is Valentine's Day, so let's talk about sex.

Eel sex!

Well...we can't, really.

We don't know exactly where, or how, eels get busy. It's a 2000+ year old puzzle. It's not like we haven't tried. But the sex life of eels is a mystery smothered in secret sauce. 1/5
🗃️🧪

Children trapped in Texas immigration facility recount nightmares, inedible food, no school A photo of Liam Conejo Ramos, a scared 5-year-old, drew attention to a detention center in Dilley, Texas. Advocates say his experience reflects what hundreds of children have endured out of public vie...

Maria is one of hundreds of children, who — like Liam Ramos — have been held at a Texas detention facility where parents say children languish as they’re served contaminated food, receive little education and struggle to obtain basic medical care.

Here's our attempt at telling their stories. 2/

I wrote a blog for the Meta-Research Center expressing my infinite frustration about not getting data. What else is new, you might think? Well, I added an extra layer of annoyance directed at the journals who do NOTHING to enforce promised data sharing.

metaresearch.nl/blog/2026/2/...

All fine by me.

Firmly believe this would serve multiple needs.

If I don't know the area I'm likely to read the introduction carefully from the start.

Varies, but I usually skim intro and discussion. Then look at methods and results. Usually results in detail and back to methods for clarification. If it is important I'll go back and read the paper from start as one read-through is rarely enough.

Wouldn’t have made it out of the VEEP writers room

I'm in the UK but still getting a lot of coverage of this. I mostly follow friends in science and academia, but many are in the US.

Oh and Election Concerns too ...

It's still on my trending topics under the NRA Criticism label. I think the reason it is not so prominent is maybe that a lot of people are just emotionally exhausted and maybe switching off from news for the rest of Sunday.

I was a sensitive kid. I loved cats. I had OCD, undiagnosed.

When I was 9, a fraternity brother tortured a cat to death in my hometown. The details even now are beyond imagination. archive.is/xdmUL

I was traumatized for years from reading the news story.

The torturer? Now the head of the NRA.

This is the paper:

Felix B. Muniz & David P. MacKinnon (21 May 2025): Three
Approaches to Testing for Statistical Suppression, Multivariate Behavioral Research, DOI:
10.1080/00273171.2025.2483245

Yes. They are closely linked. Not really my area but I did read a nice paper on it recently. Suppression is a more general concept and whether they are identical depends on your definition of suppression.

I just wrote a short blog post on effect sizes for simple mediation. Not a topic usually write about but it was prompted by something I read that sent me to check the literature and down a bit of a rabbit hole.

psychologicalstatistics.blogspot.com

What can history teach us about what happens when a populist strongman with an idiosyncratic taste for low interest rates undermines central bank independence?