Profile avatar
tcarpenter.bsky.social
🧪 Data science, survey science, social science 💻 Director of Data Science @ Microsoft Garage [Posts do not represent my employer] 🧮 Stats, R, python 📝 Science, Research: measurement, social biases, emotion. Ex-academic but scientist at heart
467 posts 3,421 followers 3,847 following
Prolific Poster
Conversation Starter
Posts 31 Comments 19

Pleased to share our ICML Spotlight with @eberleoliver.bsky.social, Thomas McGee, Hamza Giaffar, @taylorwwebb.bsky.social. Position: We Need An Algorithmic Understanding of Generative AI What algorithms do LLMs actually learn and use to solve problems?🧵1/n openreview.net/forum?id=eax...

submitted 1 day ago • 2 comments

Someone please explain why linear algebra isn’t taught more in high school in the United States? Seems like maybe if you’re lucky you get a few lectures on matrices and that’s it.

submitted 3 days ago • 2 comments

Is it 2015 or 2025?

submitted 4 days ago • 0 comments

“Data available upon request”

submitted 4 days ago • 1 comment

Under the Trump agenda, energy will cost more. And when energy costs more, everything costs more. www.nytimes.com/2025/06/04/c...

submitted 13 days ago • 1 comment

Many academics are taught “do more” instead of “prioritize”. They are taught the academic superhero myth, that “truly great/smart” scholars can handle it. So they sabotage their own success in service to the cult of personality

submitted 16 days ago • 0 comments

I love this technique because it gives a cool way to isolate change and stability components of within-person measurement using latent variables. The stable portion of IATs is far more predictive of other individual-difference measures than one would think given traditional scoring/analyses

submitted 17 days ago • 2 comments

I love this analysis technique because it gives a cool way to isolate change and stability components of within-person measurement using latent variables. The stable portion of IATs is far more predictive of other individual-difference measures than one would think given traditional scoring/anakysus

submitted 17 days ago • 0 comments

That moment when you had completely forgotten about a cool data app you’d made, only to find others posting about it ;) Thanks @calvinklai.bsky.social!

submitted 17 days ago • 0 comments

Paper w/ @tcarpenter.bsky.social & @alexgoedderz.bsky.social at PSPB!🚨 The standard IAT is only 5 min long. We found that making the IAT longer by taking it multiple times greatly improves predictive validity. 🧵below, with practical advice about how to run IATs! LINK: osf.io/preprints/ps...

submitted 17 days ago • 4 comments

What should universities look like 10 to 15 years from now? Lots of intersecting headwinds… dramatic change is likely. Biggest IMO is that teaching universities become cheap again by reverting to “room with chairs and a prof”. Think community college, but with amazing faculty. 

submitted 20 days ago • 0 comments

Skip the stress and the store – Copilot with Instacart can now handle your grocery runs for you! Recipe ✅ Shopping list ✅ Groceries at your door ✅ All seamlessly integrated in the Copilot app.

submitted 23 days ago • 0 comments

Fitting Bayesian models and finding it tedious to iteratively run prior predictive checks and tweak my priors. I’m thinking of a way to automate it. I know the plausible distribution of the data + plausible values for the parameters. The hard part is reconciling them.

submitted 24 days ago • 2 comments

Has anyone fit bayesian models with brms in Spark's implementation of R (sparkr)? What's your experience?

submitted 24 days ago • 0 comments

"In extremely rare move, Harvard revokes tenure and cuts ties with star business professor" Francesca Gino

submitted 27 days ago • 5 comments

The more I use AI to teach myself new skills, the more convinced I am that this is one of THE best ways to learn (especially if you have to learn selectively and quickly). However, strategy matters here. I threw some thoughts on using AI for skilling here. www.linkedin.com/posts/tompca...

submitted 32 days ago • 1 comment

Survey programmers be like

submitted 33 days ago • 0 comments

Folks, get in here. We’re admitting all the dumb wrong shit we believed until we were way too old.

submitted 34 days ago • 612 comments

Yes. www.billboard.com/music/music-...

submitted 38 days ago • 674 comments

In loving memory of Moshu 🍎

submitted 38 days ago • 358 comments

This is an excellent illustration of interaction effects with a (far too) resonant example.

submitted 41 days ago • 1 comment

The lesson in this story is that doing a good job doesn’t matter if no one knows you’re doing a good job so some self promotion is necessary. I think there’s a different lesson in the title — a promotion isn’t from doing your job well, eventually it’s showing you can also do your boss’s job well.

submitted 41 days ago • 2 comments

Seagulls, witnesses to history.

submitted 44 days ago • 25 comments

is the conclave vote ranked-choice

submitted 45 days ago • 10 comments

the precursor to direct sunlight is passive aggressive sunlight

submitted 46 days ago • 5 comments

!!! Look at what I found.

submitted 47 days ago • 12 comments

The Star Wars prequels are a trilogy about how an unfair and severe galactic tariff policy causes instability in the democratic political order that ultimately enables an authoritarian to seize power under the guise of restoring peace and prosperity to the republic. Why does this sound familiar….

submitted 47 days ago • 59 comments

Shot // chaser

submitted 49 days ago • 86 comments

Bottled water is a big ole bottle of microplastic. Don't buy it, support utilities so everyone has access to clean, less plastic-filled water. My university has a project to vastly improve how we monitor tap water but until NSF and its workers are restored, I fear that is on hold. 🧪 🚰

submitted 53 days ago • 2 comments

I wish this were a joke

submitted 57 days ago • 0 comments

'At current state-level vaccination rates, measles may become endemic again; the simulation model predicts measles may reestablish (>80% of simulations) with an estimated 851,300 cases over 25 years. Increasing vaccine coverage would prevent this' jamanetwork.com/journals/jam...

submitted 58 days ago • 2 comments