Econometrics for Health Policy

Interactive modules — run R in your browser, no installation required

Econometrics is usually taught in the abstract, with health examples bolted on. These modules work the other way around: every method is introduced through a problem someone in public health actually needs to solve — birth weights and gestational age, hemoglobin trajectories in children, community health worker programs, bednets and malaria.

Each module mixes short theory sections with live R exercises that run directly in your browser (via webR). Fill in the blanks, run the code, get graded feedback, check the hints when stuck. Nothing to install.

These grew out of my years as a Teaching Fellow for GHP 525 (Econometrics for Health Policy) at the Harvard T.H. Chan School of Public Health. — Austin Tucker

The series

Foundations

Module	What you’ll learn
1 · OLS Foundations	What a regression line means, how OLS finds it, and how to read coefficients without fooling yourself
2 · Standard Errors & Clustering	Where uncertainty estimates come from, and why field studies that ignore clustering reject true nulls far too often
3 · Inference & Asymptotics	What p-values and confidence intervals actually mean — shown by simulation, not asserted

When things break

Module	What you’ll learn
5 · When OLS Fails	Omitted variable bias and measurement error: the two quiet killers of observational health research

Panel data

Module	What you’ll learn
6 · Panel Data	Fixed and random effects: using repeated observation to control for what you can’t measure

Module 4 (binary outcomes) and the causal inference sequence — randomization, difference-in-differences, regression discontinuity, instrumental variables — are in the works.

How to use these

Work through them in order if you’re new; jump straight to the topic you need if you’re not. Each module takes roughly 45–90 minutes. The exercises are the point — reading the theory without running the code is like learning to swim from a book.