Assess graduate-level understanding of core concepts and assumptions in causal inference for observational studies, including propensity score methods (unconfoundedness, overlap, estimation and diagnostics), instrumental variable assumptions (relevance, exclusion, monotonicity/local average treatment effects, threats to validity), and directed acyclic graphs (DAGs) for representing causal structure, identifying confounders/colliders, selecting adjustment sets, and evaluating identification strategies and common sensitivity analyses.