Elizabeth Stuart

I'm looking forward to the conversations and learning at #JSM2025!

A Classroom Experiment

I appreciate the push for research at the end of the story about AI + education. But it doesn't clarify who will pay or do that needed research -- with the gutting of the federal education research agency (IES) it's not clear there will be unbiased groups to do it.

www.nytimes.com/2025/07/09/b...

I am grateful for this coverage. Fiona is a friend and sadly just one example of many devoted public servants (who I know personally to be deeply committed, smart, and caring) whose expertise we are currently losing from the government.

I was SO sad that my own session conflicted with this one!

@noahgreifer.bsky.social was my post-doc and I cannot recommend him more highly if you need a super smart statistical consultant / programmer. He is the force behind MatchIt, cobalt, and other packages, and is also just a fantastic team member.

Excited for what will be a flash trip to New England in mid-June for the AI + precision medicine conference in Portland, ME, followed by the Society for Epi Research in Boston! Join me at both!

My daughter was taking the same exam today! I kind of love the coordinated timing all around the country and the idea of a bunch of kids all sitting down for the tests in a coordinated timing kind of way.

Great way to learn the basics of policy trial emulation and the importance of careful study design for health policy!

“Scale is not a substitute for scrutiny” may be my new favorite quote. Thanks @adamjkucharski.bsky.social!

This is so important. And especially in non experimental studies where bias - not variance - is the first order concern.

Great way to learn the basics of policy trial emulation for free! Paper here: pubmed.ncbi.nlm.nih.gov/39374529/

This has been one of the most meaningful and rewarding (and sobering) collaborations I have been involved in, especially the integration of advanced stat methods and deep substantive expertise. Crucial empirical data and results that we hope can inform policy discussions.

Make sure it's done multiple times -- otherwise imputation will make estimates appear more precise than they really are.

One add'l insight is that in some contexts a "missing data indicator" approach is actually okay (that's what twang does in implementation), and even if you do MICE [with outcomes and treatment included too!] it might make sense to include missing data indicators, as they may carry info about people.

The WSJ editorial board out with a piece today opposing confirmation of RFK, Jr.

www.wsj.com/opinion/rfk-...

Sad to be missing the opportunity to cross country ski in DC but can’t really complain about being at #ICHPS. Already have run into many friends and ran on the beach!

a countdown clock with the number 10 in the center ALT: a countdown clock with the number 10 in the center

Let us start 2025 in a positive mood: here are 10 methods things researchers can worry *less* about in 2025

Thank you for this thread; information like this is sorely needed. One quibble -- I would clarify that "massive study" does not always equal good evidence -- quality matters, not just sample size. Agree on the substance though; no rigorous evidence for a link between vaccines and autism!

That was Union station in DC.

There are worse places to have to be very early in the morning. Excited to spend the day talking causality at Penn!

In my experience it isn’t even that they might want to become a professor some day but more that someone told them it would help them get into college. Would love to know what admissions officers think about that.

I’m excited by many speakers at the Bloomberg American Health Summit - needed inspiration for those of us in public health - and especially fun to see @annieandrewsmd.bsky.social live, talking about gun violence prevention and how to normalize discussions of gun safety along pediatricians.

You can watch the Summit here -- important conversations about public health in the US: www.youtube.com/live/sKg4Tcl...

ASA Community The ASA Community is an online gateway for member collaboration and connection.

A busy day today -- excited for the Bloomberg American Health Initiative summit, and sneaking out mid-day to do this (free) webinar on causality + interdisciplinarity! community.amstat.org/discussion/t...

Thank you for nothing and writing!

Thanks for the plug! And just to clarify to estimate the ATT data on the comparison group is used. The inference population is just the treated group.

Great group to work with, including me!

Glad to be in Cape Town for timely conversations about data and evidence.

Thank you!!

Roderick Little - Survey Sampling Inference My research in survey sampling focuses on model-based (Bayesian) methods for complex survey designs that are robust to misspecification, and comparing the resulting inferences to classical methods bas...

The survey world makes distinctions between model based and design based as well, outside of causal inference. Rod Little has nice papers on that, e.g., sites.google.com/a/umich.edu/...