In this SACC lecture, we begin with a concise introduction to causal inference, focusing on how researchers define causal effects and why estimating them is challenging in non-experimental settings.

We then turn to Difference-in-Differences —a foundational design for evaluating policies and interventions when randomization is not feasible. The lecture develops the intuition behind Difference-in-Differences as a comparison of pre-post changes in a treated group to contemporaneous changes in a comparison group, highlighting key identifying assumptions (especially parallel trends).