Profile avatar
malcolmbarrett.malco.io
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.
475 posts 4,984 followers 686 following
Regular Contributor
Active Commenter
Posts 29 Comments 21

@malcolmbarrett.malco.io & I will be doing another Causal Inference workshop on Tuesday (and it will be my birthday!!) join us!

submitted 12 hours ago • 0 comments

New post about qualifications with qualifications: kucharski.substack.com/p/incredible...

submitted 1 day ago • 7 comments

We’re excited to announce that the session catalog for posit::conf(2025) has launched! View 100+ talks, sort by day and type, and add favorites to your schedule: reg.rainfocus.com/flow/posit/p... Our Early Bird discount is still available. Get your ticket today! #RStats #Python #PositConf2025

submitted 1 day ago • 0 comments

My dear health equity peers, the Robert Wood Johnson Foundation is doubling down on their DEI efforts. 💪🏽 Here’s a grant opportunity tinyurl.com/3pnxhy2d with a rapid response option those whose “health equity research projects have lost federal funding.” ⭐️ Please share ⭐️

submitted 1 day ago • 2 comments

Someone who understands how the US government handles its money, explain to me something: DOGE can cut contracts or whatever, but the money has already been allocated by Congress. Where does it go? It can't go back to taxpayers by just not spending it. Is it just sitting in an account somewhere?

submitted 1 day ago • 1 comment

How much of the US federal budget is allocated to AmeriCorps? 0.02% of the 13-digit budget goes to AmeriCorps. Every dollar invested turns into over $17 in economic and community benefits. But we guess DOGE considers that fraud, waste, and abuse.

submitted 2 days ago • 4 comments

Happy to share that {recipes} has a new release with many new features and all known bugs exterminated! www.tidyverse.org/blog/2025/04... #rstats #tidymodels

submitted 1 day ago • 0 comments

Breaking News: A federal prosecutor interrogated the The New England Journal of Medicine — which is considered the world’s most prestigious medical journal — in what its editor described as a “vaguely threatening” letter.

submitted 4 days ago • 217 comments

air is so, so good. Feels like the workflow joy of adopting usethis all over again—so many 2-second bits of tedium that I don't have to endure anymore. posit-dev.github.io/air/

submitted 4 days ago • 1 comment

PhD Timeline xkcd.com/3081

submitted 4 days ago • 607 comments

I was on a NASEM committee that received a stop work order earlier this week. It happened as the meeting was about to begin—and after flying everybody out. It was bizarre. You've got a dozen excited experts who are literally not allowed to take money for the work. This isn't a cost-saving action.

submitted 4 days ago • 1 comment

We’ve reached the arresting judges phase. www.nytimes.com/2025/04/25/u...

submitted 4 days ago • 0 comments

A lot has changed in the world, so my colleagues and I at the @fordschool are opening up a special emergency admissions cycle for folks who want to start a Masters of Public Policy this fall. Deets are here: fordschool.umich.edu/masters-prog... I hope to see you in class in September!

submitted 5 days ago • 6 comments

New #QuartoPub extension! Have you ever wanted to include epigraphs in your documents without needing to adjust CSS or LaTeX or Typst settings? Now you can the fancy epigraph shortcode: github.com/andrewheiss/... It works with HTML and PDF only—in other formats it appears as a regular blockquote

submitted 5 days ago • 2 comments

I'm coming out as an LLM moderate

submitted 5 days ago • 0 comments

Piles of linear algebra don't think or feel or reason or

submitted 5 days ago • 4 comments

priorsense 1.1.1, for prior and likelihood sensitivity analysis, is now in CRAN (n-kall.github.io/priorsense/) There is now an easier way to select which priors are power-scaled using the new prior tags feature in brms (2.22.11+), which allows focusing the analysis on specific priors. 1/3

submitted 5 days ago • 1 comment

My amazing independent study student wrote me a thank you card and drew this laptop with #rstats code on it 🥹

submitted 6 days ago • 0 comments

Also, don't worry so much about replacing every package. You can just install it when you need it.

submitted 5 days ago • 1 comment

#rstats folks: If you have a habit of updating your library after an R upgrade by copying the old one and run update.packages(). --> DON'T DO THAT <-- Especially not with 4.5.0. You will overwrite base packages with ones from a previous version and as these are not on CRAN, things go bad.

submitted 5 days ago • 4 comments

ICYMI: `install.packages()` is so much faster in R 4.5.0 #rstats

submitted 5 days ago • 1 comment

Check out the last Methods Series talk now on YouTube: youtu.be/QuQnZ4don54?...

submitted 7 days ago • 0 comments

Say what you will, these grifters are going to make a lot of fraudulent science together

submitted 6 days ago • 0 comments

UV 0.6.15 has preliminary pylock.toml support! That means there is either merged code or a release with pylock.toml support in: 1. Pip 2. Pip-audit 3. PDM 4. uv github.com/astral-sh/uv...

submitted 7 days ago • 2 comments

Me: Don't R CMD Check: ... Me: .... R CMD Check: ... R CMD Check: ❯ checking for future file timestamps ... NOTE unable to verify current time

submitted 8 days ago • 3 comments

The White House has turned COVID.gov from a public health resource into a politically charged platform promoting the lab-leak theory and attacking specific scientists and the pandemic response in general. Let’s fact check its 5 headline claims… 🧵

submitted 11 days ago • 101 comments

it's hard to find a motivation for DOGE other than 'hatred of the good.' these are bad people fundamentally envious or un-understanding of the idea that you might work or give for others, and attempting to punish it.

submitted 13 days ago • 52 comments

America, 2026

submitted 13 days ago • 0 comments

The loss of AmeriCorps funding is particularly painful for me. I owe my career and much else to my two years of service. AmeriCorps members serve in everything from disaster relief to education to health for extremely low pay. The return on investment on both sides is big.

submitted 13 days ago • 1 comment