Causal inference with Python and an unreasonable number of memes

What it does

This is an open-source textbook—delivered as Jupyter notebooks—that teaches causal inference and impact estimation in Python. It adopts the rigor of Angrist and company’s Mostly Harmless Econometrics lineage alongside Miguel Hernan and Jamie Robins’ causal inference framework, but trades the traditional academic tone for memes and an explicitly light-hearted voice.

The interesting bit

The author treats this as a pedagogical experiment: proving that intellectual accessibility and entertainment aren’t opposites of scientific rigor. The project has spawned community translations in at least six languages and a commercial follow-up for industry application—suggesting the informal approach resonated widely.

Key highlights

• Free, Python-only resource covering causal inference and sensitivity analysis
• Explicitly based on Angrist, Abadie, and Walters’ AEA econometrics courses and texts
• Community translations available in Portuguese, French, Chinese, Spanish, Korean, and Vietnamese
• DOI-backed via Zenodo for academic citation
• Companion commercial book focuses on industry application

Verdict

Worth bookmarking if you need to learn econometrics-driven causal inference but find standard textbooks impenetrable. Skip it if you’re already comfortable with Mostly Harmless Econometrics and just want a reference manual—this is a course, not a cheat sheet.