Malcolm 朝精 Barrett
LightNews LightNews
banner
malcolmbarrett.malco.io
Malcolm 朝精 Barrett
@malcolmbarrett.malco.io
5.2K followers 750 following 640 posts
Ph.D., epidemiology. research software engineer @ Stanford Health Policy. living in Ann Arbor. open-source data science. causal inference. doing poems on aircrafts. approximately Bayesian. formerly Posit, Apple, AmeriCorps. 心を燃やせ。sic semper tyrannis.
Posts Replies Media Videos
malcolmbarrett.malco.io
Oh shit I think the spam texters are pissed at me
Can I have a quick word?
November 2, 2025 at 6:29 PM
malcolmbarrett.malco.io
Kids in the street
A child with an emoji mouse face and a protest sign that says “no kings”
October 18, 2025 at 6:15 PM
malcolmbarrett.malco.io
A protest sign that says “sic semper tyrannis”
October 18, 2025 at 6:10 PM
malcolmbarrett.malco.io
There are three kinds of lies: lies, damned lies, and
a checkbox on a website that says "keep me logged in"
September 29, 2025 at 2:42 PM
malcolmbarrett.malco.io
Casual AI is the future
LinkedIn screen shot that says “GPT-5 is able to predict the outcome of experiments that we haven't even done. That saves us weeks and even months." Immunologist and professor Dr. Derya Unutmaz explains how he uses GPT-5 as a collaborator on medical research. It’s our most intelligent model yet, capable of generating novel hypotheses, accelerating discoveries, and pushing the frontier of science.”
August 15, 2025 at 3:31 PM
malcolmbarrett.malco.io
GPT-5 is like having a “team of Ph.D. level experts in your pocket.”

The Ph.D. level experts:
Marty Makary, RFK, and Jay Bhattacharya during a terrible press briefing
August 8, 2025 at 12:02 AM
malcolmbarrett.malco.io
And also
A meme of kronk with a devil on his shoulder saying "vectorize it" and an angel saying "you'll spend more time rewriting your code than you save in computation time"
July 15, 2025 at 8:19 PM
malcolmbarrett.malco.io
kronk with a devil and angel on his shoulder. The angel says "vectorize it" and the devil says "you'll spend more time rewriting your code than you save in computation time"
July 15, 2025 at 6:31 PM
malcolmbarrett.malco.io
New target population just dropped
A screen shot of a news article headlined “scientists have discovered third state between life and death”
June 17, 2025 at 2:32 PM
malcolmbarrett.malco.io
Zuck: The average American has three friends
A movie poster of the movie Clueless with three woman
May 2, 2025 at 12:08 AM
malcolmbarrett.malco.io
That’s not what’s happening. We’re not talking about intellectually nuanced conversations with people of differing views. We’re talking about a Nazi platform full of fucking nazis. This is not hyperbole. Elon is a Nazi.
Elon Musk, a Nazi, making a Nazi salute
March 1, 2025 at 2:54 PM
malcolmbarrett.malco.io
It’s Thanksgiving in Detroit. Go Loins!!!
A wooden decoration of the Detroit Lions logo with the words “Detroit Loins” on it
November 28, 2024 at 5:37 PM
malcolmbarrett.malco.io
A very whimsical Thanksgiving to you from Grinnin’ Barrett’s Bakery
Three loaves of homemade sourdough bread in a whisker basket
November 28, 2024 at 4:44 PM
malcolmbarrett.malco.io
Scooby do unmasking meme: the first panel says “research question”, the second panel, unmasked, says “whatever coefficient logistic regression gives me”
November 26, 2024 at 4:00 AM
malcolmbarrett.malco.io
WHAT’S YOUR ESTIMAND
A dog in a conference room surround by people with a tv of virtual attendees. The dog is barking and looks like he’s yelling at someone across the table
November 26, 2024 at 1:08 AM
malcolmbarrett.malco.io
Got an email about a bitchin’ new ATT estimator #causalsky #eleventoqueestabaesparando
A screenshot from an email with a clip art car that is labeled “ATT”. The caption says ¡llega el evento que estabas esperando!
November 21, 2024 at 3:07 PM
malcolmbarrett.malco.io
#causalsky hell of a new new instrumental variable identified
A sign outside a liquor store that says “super liquor IV” with pictures of whiskey
November 15, 2024 at 6:07 PM
malcolmbarrett.malco.io
Learning about all the cool new diff-in-diff estimators from #econsky the past few years, but just thinking about one thing the whole time #causalsky #causalinference
A photo of a Yankees fan interfering with the World Series game. He is labeled "parallel trends assumption". A baseball player has a ball in his glove. He is labeled "ATT". The fan is trying to take the ball from the glove.
October 30, 2024 at 8:38 PM
malcolmbarrett.malco.io
A meme of Bruce Banner saying "That's my secret, Cap. I'm always partially identifying"
October 22, 2024 at 3:35 PM
malcolmbarrett.malco.io
All numbers are nerds but not all nerds are numbers.
A screenshot of Fox News on twitter. It says Jason Kelce blasts statistics in a wild rant: stats are numbers, numbers are nerds, nerds are losers
October 4, 2024 at 1:01 PM
malcolmbarrett.malco.io
Oh good I found a likely real causal effect here. Scam people into improving their grips -> make money
An article headline with a picture of someone hanging from a bar with both hands. It says: If You Can Hang From a Bar For This Long, Scientists Say You're Likely to Live a Longer, Healthier Life And what simple tests like this one can mean for entrepreneurs hoping to serve on of the biggest addressable markets on the planet
August 15, 2024 at 12:39 AM
malcolmbarrett.malco.io
We are excited to be hosting Causal Inference in R Workshop at the 10th annual New York R Conference on May 15! Join us in NYC or virtually online! Get 20% off your ticket now with promo code NYR10 at rstats.ai/nyr #rstatsnyc | #rstats
A card about the workshop on May 15
April 4, 2024 at 5:41 PM
malcolmbarrett.malco.io
If you’re familiar with structural equation models (SEMs), a modeling technique commonly used in psychology and other social science settings, you may notice some similarities between SEMs and DAGs. DAGs are a form of non-parametric SEM. SEMs estimate entire graphs using parametric assumptions. Causal DAGs, on the other hand, don’t estimate anything; an arrow going from one variable to another says nothing about the strength or functional form of that relationship, only that we think it exists.
December 1, 2023 at 9:42 AM
malcolmbarrett.malco.io
We've made a major update to the causal diagrams chapter of Causal Inference in R! Check it out to learn about DAGs! I think there's a lot in this chapter that is undercovered in other sources. #rstats #causalinference @lucystats.bsky.social @travisgerke.bsky.social www.r-causal.org/chapters/05-...
A screenshot of Causal Inference in R: Causal Diagrams A screenshot of a DAG from Causal Inference in R showing two open pathways in an analysis a scientist must account for to get the correct answer A screenshot of a DAG from Causal Inference in R showing how to describe feedback loops in causal diagrams
November 29, 2023 at 9:30 PM
malcolmbarrett.malco.io
We just overhauled the first chapter of our book, Causal Inference in R! Learn about causal questions, their structure, and the relationship between description, prediction, and causal inference. www.r-causal.org/chapters/cha... @lucystats.bsky.social @travisgerke.bsky.social #rstats
A screenshot of the first chapter of Causal Inference in R: What is a causal question? A sentence diagram of a causal question showing it's essential components: exposure, outcome, eligibility criteria, time-zero, target population, and follow-up period A visualisation of a descriptive question: what were the number of deaths in 2020 vs the expected deaths based on historical data (by country)? The figure shows the shocking toll of the early pandemic, with actual 2020 deaths far above expected deaths. This is a useful graph for understanding risk without making causal claims
September 15, 2023 at 5:55 PM