-
Why Causal Mechanisms Are Harder Than You Think
The hidden costs of explaining how something works: cross-world counterfactuals and heroic assumptions.
-
Why Better Models Can Create Stranger Counterfactuals
When AI explanations respect real-world constraints, the "what-ifs" become fewer but more meaningful.
-
What Directed Acyclic Graphs (DAGs) Teach Us About Choosing Covariates
Why adding more controls can backfire, and how causal graphs help you pick the right ones.
-
Three Fundamental Conceptual Shifts in Causal Inference
From missing data to survivor bias: three ideas that will reshape how you think about cause and effect.
-
Significance Stars: A Cautionary Tale
P-values, power, and the three types of significance: a guide to interpreting quantitative evidence.