World-Class Causal Inference Curriculum

Understand Why Things Happen. Not Just How.

From Pearl's DAGs to the potential outcomes framework — master the science of causation through 88 rigorously designed lessons, live interactive simulations, and an AI tutor that knows causal inference inside out.

Start Learning Free → Try the DAG Builder
88+
Lessons
15+
Domains
50+
Methods
5
Curriculum Tracks

Curriculum

Five tracks. One complete education.

A structured path from first principles to frontier methods — or jump to the track that matches your level.

📐
Foundations
🔗
Basics
🎯
Identification
📊
Estimation
🚀
Advanced
Track 01
📐
Mathematical Foundations
Probability, statistical inference, linear algebra, and the mathematical language of causal models.
15 lessons · Beginner
Track 02
🔗
Causal Basics
DAG structures, confounding, colliders, selection bias, mediation, and instrumental variables.
20 lessons · Beginner–Intermediate
Track 03
🎯
Causal Identification
Backdoor criterion, frontdoor criterion, do-calculus, d-separation, and identifiability theory.
15 lessons · Intermediate
Track 04
📊
Estimation Methods
IPW, G-formula, AIPW, matching, DiD, synthetic controls, double ML, and causal forests.
22 lessons · Intermediate–Advanced
Track 05
🚀
Advanced Topics
Longitudinal methods, time-varying treatments, causal ML, mediation, and proximal inference.
19 lessons · Advanced

Domain Pathways

Learn in your field.

Every method is taught through real examples from your domain — the same rigorous causal questions, in the context that matters to you.

🏥Healthcare & Epidemiology 💰Economics & Development 💻Tech & Product 🧬Bioinformatics & Genomics 💊Pharmaceutical & Clinical 📈Finance & Risk 🌱Agriculture & Food 🎓Education & Learning 🌍Environment & Climate 🧠Psychology & Behavioral 📣Business & Marketing 🏙️Urban Planning & Policy ⚙️Industrial Analytics 🕸️Network Science 🤖Causal ML

Pedagogy

Designed for deep understanding,
not surface familiarity.

Every lesson follows an instructional design grounded in cognitive science and Bloom's taxonomy — from concrete examples to formal derivations.

01 — Activate
Concrete Before Abstract
Every lesson opens with a real-world scenario — a clinical trial, an A/B test, a policy intervention — before any formalism is introduced. Context first.
02 — Build
Incremental Complexity
Concepts layer precisely on the one before. No cognitive overload. No skipped steps. Each slide advances understanding by exactly one logical move.
03 — Visualise
Live DAG Interaction
Don't just read about d-separation — see it. Every structural concept is paired with an interactive causal graph you manipulate directly in the browser.
04 — Practice
Embedded Assessment
Knowledge checks, identification challenges, and estimation exercises are woven into each lesson — not bolted on at the end. Retrieval practice, not passive reading.
05 — Transfer
Cross-Domain Application
The same method is shown across healthcare, economics, and tech. Structural understanding transfers — you learn the causal reasoning, not just the recipe.
06 — Tutor
Slide-Aware AI Copilot
An AI tutor that knows exactly which slide you're on and answers questions in context — not generic help, but specific, grounded, Socratic guidance.

Platform

Everything you need to go from zero to research-ready.

🧠
AI Causal Copilot
A slide-aware AI tutor that answers questions in context. Ask "why does conditioning on a collider open a path?" and get an explanation grounded in what you just read.
🔬
Interactive DAG Playground
Build any causal graph, identify adjustment sets, simulate data, and run estimation — all in the browser. No code required. Professional-grade tools accessible to everyone.
📐
Formal Notation Support
Every lesson uses consistent notation — Pearl's do-calculus, potential outcomes, SCMs. When you finish, you can read the primary literature with confidence.
🎯
Scenario-Based Learning
Lessons are organised by causal question, not just topic. "How do I handle time-varying confounding?" — find it, learn it end-to-end, apply it immediately.
📊
Complete Methods Library
50+ estimation methods, each with intuition, formal derivation, assumptions, and worked examples. The most comprehensive causal inference reference anywhere on the web.
📱
Offline PWA
Works fully offline after first load. Install on any device. Full-screen presentation mode for teaching. Designed for researchers on the go.

Methods Library

Every method. Rigorous. Practical.

50+ identification and estimation methods — each with formal derivation, assumptions, and domain-specific worked examples.

Backdoor Criterion Frontdoor Criterion do-Calculus IPW / IPTW G-Formula AIPW / Doubly-Robust Propensity Score Matching Difference-in-Differences Instrumental Variables Regression Discontinuity Synthetic Controls Double Machine Learning Causal Forests Marginal Structural Models Structural Nested Models Parametric G-Formula Target Trial Emulation Survival Analysis Proximal Causal Inference D-Separation Principal Stratification Bayesian Causal Inference Sensitivity Analysis Mediation Analysis Panel Data Methods Effect Modification Adaptive Experiments Policy Learning Semiparametric Theory LATE / IV-LATE Graphoids + 20 more →

Get Started

The most rigorous causal
inference education. Free.

No sign-up. No paywall. Just the clearest path from correlation to causation — built for researchers, practitioners, and curious minds.

Begin Your Curriculum → Explore the Playground