Vera Wilde
LightNews LightNews
verawil.de
Vera Wilde
@verawil.de
220 followers 300 following 350 posts
Scientist (PhD), writer, risk literacy and citizen science tool builder. Nerd-of-all-trades (research methodologist). <3 FOIA, babies, clocking bias and error. Seeker of truth, especially wild.
Posts Replies Media Videos
verawil.de
57% of my pilot participants aced Bayesian stats problems. The lit's ceiling is 24%. Either I'm the greatest science educator alive, or everyone used LLMs.

Silver lining: my crappy data suggest a measure to estimate how contaminated your own crappy data are!

wildetruth.substack.com/p/lowered-ex...
February 9, 2026 at 10:04 AM
verawil.de
When you see "Post too long for email" in your current top Substack draft and think "ok, I've done my job."
February 8, 2026 at 12:18 PM
Reposted by Vera Wilde
dingdingpeng.the100.ci
New preprint! So, what's a multiverse analysis good for anyway?>

With @jessicahullman.bsky.social and @statmodeling.bsky.social

juliarohrer.com/wp-content/u...
What’s a multiverse good for anyway? Julia M. Rohrer, Jessica Hullman, and Andrew Gelman Multiverse analysis has become a fairly popular approach, as indicated by the present special issue on the matter. Here, we take one step back and ask why one would conduct a multiverse analysis in the first place. We discuss various ways in which a multiverse may be employed – as a tool for reflection and critique, as a persuasive tool, as a serious inferential tool – as well as potential problems that arise depending on the specific purpose. For example, it fails as a persuasive tool when researchers disagree about which variations should be included in the analysis, and it fails as a serious inferential tool when the included analyses do not target a coherent estimand. Then, we take yet another step back and ask what the multiverse discourse has been good for and whether any broader lessons can be drawn. Ultimately, we conclude that the multiverse does remain a valuable tool; however, we urge against taking it too seriously.
February 4, 2026 at 10:24 AM
verawil.de
DAGs describe discrete snapshots. Most causal mechanisms of interest in science on humans and other living things operate continuously. The two may not correspond. Is it time to panic?

wildetruth.substack.com/p/can-we-bel...
February 4, 2026 at 11:13 AM
verawil.de
Anyone else notice the hot water seems to be out when you're too tired to take a shower
February 3, 2026 at 6:50 AM
Reposted by Vera Wilde
sarahcrook.bsky.social
writing: this is fun
editing: this is not fun
checking proofs: this is terrible
reading my published work within c.5 years: this is horrific beyond measure
reading my work c.6-10 years later: who wrote this
February 2, 2026 at 4:52 PM
Reposted by Vera Wilde
f2harrell.bsky.social
Wow an important paper. And Sander Greenland has thought deeply about this and brings rigorous thinking to causal inference debates.
February 2, 2026 at 1:45 PM
verawil.de
Why is there not a saying "the cold takes 30 [IQ] points," like "the camera adds 10 pounds"?
February 2, 2026 at 2:37 PM
verawil.de
Greenland "states that 'causal inference from observational data using formal causal models remains a theoretical and largely speculative exercise (albeit often presented wo explicit acknowledgement of that fact).' "
pmc.ncbi.nlm.nih.gov/articles/PMC... - Aalen et al, "Can we believe the DAGs?" 2014
Can we believe the DAGs? A comment on the relationship between causal DAGs and mechanisms
Directed acyclic graphs (DAGs) play a large role in the modern approach to causal inference. DAGs describe the relationship between measurements taken at various discrete times including the effect of interventions. The causal mechanisms, on the ...
pmc.ncbi.nlm.nih.gov
February 2, 2026 at 12:02 PM
verawil.de
New blog ft. overdiagnosis, underdiagnosis, what risk literacy gets wrong, updates to Conceptual (preconception gender nudge app) and Rarity Roulette (Bayesian reasoning simulator for mass screenings), modeling equilibrium in screenings, & weird zebra jokes.

wildetruth.substack.com/p/two-screen...
Two Screening Failures
Underdiagnosis, overdiagnosis, and what risk literacy gets wrong. Plus: updates to Conceptual and Rarity Roulette.
wildetruth.substack.com
February 2, 2026 at 10:32 AM
verawil.de
Florence Nightingale invented the rose chart to show policy effects over time.

I made one for mass screenings for low-prevalence problems: verawilde.github.io/rarity-roulette/policy/

Watch false positives accumulate year after year while your "99% accurate" test remains worse than a coin flip!
January 30, 2026 at 1:12 PM
verawil.de
Waiting for the "Day Tripper" remake "Vibe Coder."

🎸 Vibe coder
Yeah he prompted it right
Vibe coder
Stack Overflow's his kryptonite
It took me sooooo long to find out
But I found out
He's got a repo full of bugs now
He's got a repo full of bugs now
He's got a repo full of bugs now 🎸
January 30, 2026 at 11:49 AM
Reposted by Vera Wilde
isager.bsky.social
New blog post introducing Causion - a web app for causal inference teaching and learning: pedermisager.org/blog/causion....
Introducing Causion: A web app for playing with DAGs | Peder M. Isager
Personal website of Dr. Peder M. Isager
pedermisager.org
January 28, 2026 at 9:23 AM
verawil.de
When you find someone's R package / Shiny app that does the same thing your app does, but with more visualizations!

Rarity Roulette verawilde.github.io/rarity-roule...

meet riskyr spds.uni-konstanz.de/en/node/1170
riskyr: A toolbox for rendering risk literacy more transparent | Social Psychology and Decision Sciences
spds.uni-konstanz.de
January 28, 2026 at 10:17 AM
verawil.de
wildetruth.substack.com/p/a-little-l...

New readings: Feufel on perspicacity vs computation, Hofmann on why consent may be impossible, Eklund on screening smarter not harder, & why my Rarity Roulette pilots may have asked the wrong questions.

Come for the mass screenings, stay for the uncertainty!
A Little Learning Is a Dangerous Thing
Recent readings on screening, heuristics, endpoints -- and why frequency trees may be the wrong tool for my risk literacy job
wildetruth.substack.com
January 26, 2026 at 12:03 PM
verawil.de
So what are your favorite recent causal diagram-based analyses of observational and/or mixed (obs + experimental) data? Bonus points for using both DAGs/do-calculus and DCGs/dealing with feedbacks. BONUS bonus points if scaffolding already relates to mass screenings for low-prevalence problems!
January 25, 2026 at 3:13 PM
verawil.de
This is really interesting. I find it easy to get stuck thinking about problems with existing endpoints. These people are solving some!
jrboehnke.bsky.social Jan R. Boehnke @jrboehnke.bsky.social · 18d
Core outcome sets are a strong tool to improve quality of and communication about research, but especially to align PRACTICE and research. A #COS for #depression has the potential to move the dial considerably:
www.jclinepi.com/article/S089...
#GlobalHealth #GlobalMentalHealth
January 22, 2026 at 1:51 PM
verawil.de
V: Someone should make a parody video of "All About That Bass" for Bayes.
Internet: Hold my beer.

www.youtube.com/watch?v=kNz8...

Five-year-old: Turn it off.
All About that Bayes - All about that bass parody
YouTube video by Ian Signer
www.youtube.com
January 18, 2026 at 6:05 PM
verawil.de
Pilot 2 results are in for Rarity Roulette, the Bayesian reasoning simulator you need.

Hey, what if the educational tool you worked on for years actively made people worse at the thing it teaches? Asking for a friend.

verawilde.github.io/rarity-roule...
wildetruth.substack.com/p/rarity-rou...
Rarity Roulette v 2.0 Pilot Study
Now can people see the Bayesian forest for the frequency trees?
wildetruth.substack.com
January 7, 2026 at 1:06 PM
verawil.de
We baked a chocolate chestnut holiday cake from an AI recipe. It was dry, dull, and used too many em dashes.
January 7, 2026 at 9:36 AM
verawil.de
Mass screenings for low-prevalence problems, bathroom edition: when you check for gray hairs and try to get the one, but it's so rare, uncertain, and you would get too many others, so you give up and post on social media about how you're getting old
January 7, 2026 at 6:55 AM
verawil.de
Earl Grey tastes like shoe. But, sometimes, you just need a hot cup of shoe.
January 6, 2026 at 6:27 AM
verawil.de
Performative prediction: The prophecy that causes itself. The screening that changes who it's screening. The dating app that manufactures your "type."

If you're into equilibrium effects and other things that break DAGs because life isn't hard enough, enjoy!

wildetruth.substack.com/p/all-the-wo...
All the World's a Performative Prediction
And all the people merely players
wildetruth.substack.com
January 5, 2026 at 11:01 AM
Reposted by Vera Wilde
j2bryson.bsky.social
"Mental health" really just seems like a proxy for "wellbeing" when you read things like this. The organism fails when it suffers abuse and stress, even from impersonal, undirected factors such as pollution or uncertainty of essential support.

www.science.org/doi/abs/10.1...
Poverty, depression, and anxiety: Causal evidence and mechanisms
www.science.org
January 2, 2026 at 11:23 AM
verawil.de
Even highly accurate mass screenings for rare problems generate overwhelmingly false positives. Frequency trees help people grok this better than tables. So I redesigned my risk literacy app (Rarity Roulette) using them. Next up, pilot 2. Science!

wildetruth.substack.com/p/rarity-rou...
Rarity Roulette v 2.0
Gearing up to pilot a redesign of the Bayes' Rule mass screenings app
wildetruth.substack.com
January 2, 2026 at 3:16 AM