Significant differences are indicated with 🐦‍⬛

Guess who booked a vacation to Spain in June 💀

As is typical, I thought I could solve my R data formatting problem with a regular expression, so I flailed around for like an hour, despaired at my ignorance, then solved the problem a different way much faster once I gave up.

When I told an AI that the resource it linked me to was written by me, and thus I knew it did not contain the information it claimed it did, it said that was "a real 'I am the Senate' moment". It's not! It was an 'I am Pagliacci' moment, at best. Damn thing couldn't even get that right.

A great learning moment in mentoring a student. She wrote a part in the paper, and just left a blank intending to find a citation to support the point. Then after hours of reading, discovered that the point she was trying to make was actually wrong! Lesson is: Always read first, then write!

Ok #stats people, I want to know your most frivolous use of your data skills. Can only be for the love of the game, can't be for work. I'll start: I record all of my oracle's actions in my tabletop rpg I play (pathfinder 2e), and use ggplot to create data viz of the results like in the attached.

Hey, I wrote a thing about AI in astrophysics ergosphere.blog/posts/the-ma...

This is like asking me where I work, and then me submitting freedom of information requests at every business in town, then doing string match til I find my name in all the replies.

The Orange Cat Brain Atlas is here. 🧠🐈 Today, we published the first comprehensive cellular map of the orange cat brain. The new atlas reveals a single, specialized neuron responsible for behaviors like staring at walls, knocking objects off tables, and the 3am "zoomies."

SCORE, a collaboration of 865 researchers, is now released as three papers in Nature, six preprints, and a lot of data (cos.io/score/). SCORE examined repeatability of findings from the social-behavioral sciences and tested whether human and automated methods could predict replicability.

I got awesome answers to my question from real experts here. Btw, I tried asking Copilot before this to see if I was "missing out" and it: a) Incorrectly proposed that it had no scholarly basis b) Misrepresented citations c) Incorrectly attributed 1st appearance to teaching materials from 2005

Hey #stats people, I'm trying to find the source of something that seems like bad advice, but my students seem to find online. Cooks distance is a measure of influence for outliers. Some people out there recommend 4/N as the cutoff for an influential value. Any idea where this cutoff comes from?

Bayesian SaaS companies and OnlyFans creators are very similar if you think about it🤔 Both make money by showing off their expertly crafted posteriors.

Someone was trying to take a portrait of their two fave Chickens, and this happened.

I am now officially promoted to University Teaching Fellow (like full professor, but for teaching-stream faculty)! 🎉 The true reward is now never needing to collate a file for promotion ever again

Ok this is getting embarrassing for me, so #stats people you know the Akaike information criterion (AIC) right? Uh ... how do you pronounce Akaike? I gotta verbally say it sometimes in teaching, and am tired of making a fool of myself lol

I almost never do any kind of forensic data auditing, but got do spend a bit of time on it in a consulting gig and its ... kinda fun. No maleficence in this case, but folks: Remember to always read the primary sources (someone used a secondary citation of certain numbers, which was wrong).

This reminds me of when Dalhousie changed our tagline from the respectable "Inspiring minds" to the supervillan-esque "Where infinite ambition meets global impact.” I joke now that since 2022, Dalhousie no longer inspires minds

I just heard "postplagairism era" as a term to describe what education will be like post-AI (e.g., everyone will start using brain-computer interfaces and use AI to assist in their writing). It's so frustrating that these kinds of pieces assume nobody will resist (or that resistance is futile).

I have a joke about multiple imputation after pooling, but it's only average

Meanwhile my public health colleagues

I have an exam scheduled today for an undergrad class. On St. Patrick's day. During a rainstorm. With a week long student strike happening. While I am still recovering from the flu. Feels cursed, ngl

This is Hazel. She did some minor redecorating while you were out. Hopes you like it. 12/10 (TT: kelly_smith02_)

As I try to reconcile a million takes on AI and academia, my current thinking about my own role is analogous to artisan manufacturing. Mass production was incredibly disruptive, but there are still tons of thriving small producers of goods using labour-intensive methods able to make a living. 1/2

Calling all Bayes #stats people, I had a student who wants to do exploratory factor analysis in a Bayesian framework. They are working with indigenous people and developing a scale, and want to incorporate prior knowledge about the factor structure & factor loadings. 1/2

Just learned that among the many savage budget cuts happening in #novascotia right now, they are cutting the Nova Scotia Graduate scholarships. Just straight up ending all the provincial graduate scholarships. Just horrible. It's a tough time to be a university student these days.

New post! "Valuing the Process vs. the Product in Research," in which I try to describe some of the tensions around using GenAI/LLMs in scientific research, and why it can be so difficult to have productive conversations on the topic. getsyeducated.substack.com/p/valuing-th...

A week ago, we had an open conversation amongst senior academics in leadership roles and their use of LLMs (in both their leadership capacities and individually). Small group, not representative, but it already taught me a great deal about the diversity of perspectives (& their consequences). 1/

Me? I'm just happy to be here #rstats

#EcoSpace dailygalaxy.com/2026/02/japa...

Are baby chicks more bouba or more kiki? 🐥🐣🐤

This week on the blog: Against the State! Another military theory primer, this time focused on groups that take on the state itself, both with violent and non-violent means. Given the enormous disparity in force, how can they win? How have they won? acoup.blog/2026/02/13/c...

A business-related #stats consulting question for those of you out there that do paid consulting. I was contemplating hiring a student assistant for my consulting business. I normally charge hourly at a set rate, but how would you do billing if you had an assistant do some (but not all) things?

nothinghuman.substack.com/p/the-tyrann... Being popular causes all kinds of bad context collapse phenomena

It's severely important how we talk to school kids about subjects that they're interested in or not interested in, and it's criminal that it still seems to be centered on "what labor can we extract from you once you reach adulthood"

an image flashed in my head of a discord exec seeing everyone get together in IRC in protest and rambling to himself like the grinch trying to figure out christmas: "they chat without exclusive unlock quests. they chat without memes. they chat without emojis, video chat, and streams."

dplyr 1.2.0 was released last week and since I use {dplyr} a lot in my work, I wanted to take some time to try some of the new functions. This post provides some supplemental examples of the new functions, beyond what is provided in the new Posit materials. #rstats cghlewis.com/blog/dplyr_u...

average person picks 3 on sliders factoid actualy just statistical error. average person picks 0 on sliders. Sliders Georg, who has a monitor resolutiob over 16,000 x 10,000 aspect ratio, is an outlier adn should not have been counted

dplyr 1.2.0 is out now and we are SO excited! - `filter_out()` for dropping rows - `recode_values()`, `replace_values()`, and `replace_when()` that join `case_when()` as a complete family of recoding/replacing tools These are huge quality of life wins for #rstats! tidyverse.org/blog/2026/02...

Carney + 40 in Quebec lmao

Whenever Microsoft Word or some other software tries to finish my sentences for me now, even if that WAS what I was gonna say, I rewrite it now partially out of spite, but also because it's a good way to know that I was going to be overly predictable and basic in my writing

Tired: Assigned to a committee Wired: Assigned to be chair of a task force Making this committee I got assigned to be a "task force" makes it sound so important lol

Teaching my personality students a little bit about factor analysis (to a very non-stats audience, in a non-stats class), and in a group assignment a humanities student described items loading on two factors as being "balkanized into two groups" which I kind of love linguistically

Andrew Gleman just casually mentioning on his blog that he charges $850 an hour for stats consulting (American dollars, no less) makes me realize I live in a totally different financial world to some academics.

In the age of students submitting AI slop as papers, it's striking how that the new heuristic for quality in papers tends to be the presence of typos and grammatical mistakes (i.e., having errors is correlated with the ideas being more solid), which is the inverse of what it used to be like...

I just thought everyone should see this