Eric Polley

kottke.org/25/11/004787...

Ok, I’ll admit, I still reference Wikipedia in some of my lectures

Hi, It’s Me, Wikipedia, and I Am Ready for Your Apology “Wikipedia, the constantly changing knowledge base created by a global free-for-all of anonymous users, now stands as the leading force for the dum...

www.mcsweeneys.net/articles/hi-...

Great lecture on Stepped Wedge designs:

Had this debate at a faculty meeting recently. Some memorable quotes.

A shift in the operational definition of Research & Development

🗺️ Upcoming Workshop: “Intro to Spatial Data in R”

🗓️ Thursday, October 23, 2025
🕤 9:30–11:30 AM CDT
💻 Virtual; Register here: go.illinois.edu/GCCP-OCT23

This workshop is presented as part of the JCOIN Geospatial Consortium & Community of Practice (GCCP).
👉 gccp.healthyregions.org

Job Opportunities Alert

Associate Professor of Data Science: apply.interfolio.com/174768

Assistant Professor of Data Science: apply.interfolio.com/174766

EOE/Vet/Disability

Causal Bandits Podcast Causal Bandits Podcast: Causal AI, causality, causal inference, discovery, DAGs, machine learning and more!

Nice discussion with Mark van der Laan, covering many topics including the history of TMLE, and future efforts: causalbanditspodcast.com

SDOH&Place annual symposium next week in Chicago: symposium2025.sdohplace.org

Screenshot of the CRAN Task View "Clinical Trial Design, Monitoring, and Analysis" at https://CRAN.R-project.org/view=ClinicalTrials

🚨 Relaunch of CRAN Task View: Clinical Trial Design, Monitoring, and Analysis
#rstats

Complete rewrite by an int'l team of experts in #clinicaltrials

Ya Wang, Thomas Jaki @thomasjaki.bsky.social, Laura Pascasio Harris, Orla Doyle, Elias Laurin Meyer, Wilmar Igl

cran.r-project.org/view=Clinica...

Twenty-Five Things You Didn't Know You Could Do with R Explore 25 novel ways to utilize R, including pulling data from various sources, data visualization, mapping, using Quarto for reporting, automating tasks,.

Great set of R tips and helpful packages: rfortherestofus.com/2025/06/twen...

The Risks of Risk Assessment: Causal Blind Spots When Using Prediction Models for Treatment Decisions | Annals of Internal Medicine Clinicians increasingly rely on prediction models to guide treatment choices. Most prediction models, however, are developed using observational data that include some patients who have already receiv...

New paper in @annalsofim.bsky.social

"50 ways to misinterpret clinical prediction models for treatment decisions”

--> Published version: www.acpjournals.org/doi/10.7326/...

--> Open access version: arxiv.org/pdf/2402.17366

More like 48 minutes/week of dread before being able to start work on a redesign

Preparing for health care careers: UChicago master’s programs in biomedical sciences and public health Flexible programs equip students with real-world skills across variety of fields

news.uchicago.edu/story/prepar...

The style guide states this should be statistically significant, not just significant.

And then introduce notation to reinvent them?

No Kings stencil and sidewalk chalk

Just got a No Kings stencil and chalk in the mail, waiting for the rain to stop and head out to the sidewalk:

Doug Lowry had a helpful diagram explaining this fully funding impact at the beginning of the July CTAC meeting: www.cancer.gov/about-nci/ad...

YouTube video by The Late Show with Stephen Colbert David Byrne Performs 'Everybody's Coming To My House'

www.youtube.com/watch?v=07G4...

Heterogeneity of Cardiovascular Effects of Second-Line Glucose-Lowering Therapies in Adults With Type 2 Diabetes Across the Range of Moderate Baseline Cardiovascular Risk - PubMed Cardiovascular benefits of SGLT2is and GLP-1RAs exist across all levels of moderate cardiovascular risk, reinforcing the importance of choosing glucose-lowering therapies that can prevent MACE in all ...

New paper out today examining heterogeneity of cardiovascular effects between 4 classes of glucose-lowering therapies: pubmed.ncbi.nlm.nih.gov/40673553/

Bonus points if a tornado siren is going off in earshot?

Some of this may be trying to emulate what I recall from David Brillinger’s course in applied statistics. Check the data, then think about theory, but difficult to convene in a 9 week quarter

Real data does not look like most classroom examples, how do we help with this transition?

Still trying to find the best way to teach skepticism in data. When I say check the data, people find the issues, but without that statement there is a “Simon says” reaction. I have many examples of hiding messages (e.g. create a PCA plot it shows “nope” in the plot) but not everyone finds these.

After each round are you going to ask them to estimate an increasingly complex propensity score model?

👏👏

My matrices are all out at the jazz club tonight transposing in 5/4 time

#statsmeme

If only there was some process to help differentiate…