Less Likely
Statistical science in medicine & nutrition
Independent essays · Established 2018New York
The featured essay / statistics
When Can We Say That Something Doesn’t Work?
When are we allowed to conclude that an intervention doesn’t work? Is it when a study or several studies fail to find significant differences?
No matching items
Climbing Down the Ladder: Abstraction, Dashboards, and Forking Paths
statistics
visualization
data analysis
NYC comparator methods lab: demo dashboard
statistics
causal inference
interactive
How much can six months tell us? NYC comparator methods lab
statistics
causal inference
interactive
The P-value Is Not a Verdict: An Interactive Consonance Curve
statistics
interactive
No matching items
End of selection
All articles →
Recurring themes
N° 01
Sensitivity analysis
How conclusions move when assumptions wobble — tipping points, multiverse runs, robustness.
N° 02Causal inference
Drawing causal claims from observational data, and the price of doing it carelessly.
N° 03Bayesian thinking
Posteriors, priors, and the uncomfortable but useful business of saying what you actually believe.
N° 04The bootstrap
Resampling for uncertainty quantification — when it works, when it lies, and why.
N° 05Surrogates & trials
Endpoints that aren't outcomes, and trials that almost answer the question you asked.
N° 06Nutritional epidemiology
Why food-and-disease studies are hard, and what to read when someone reports otherwise.
Comments