36. Advanced Quasi-Experimental Designs: RD, DiD & Propensity Scores Explained

About this lesson

In this video, we dive deeper into quasi-experimental research designs used in healthcare and public health research when randomization is not possible. Learn how advanced methods like Regression Discontinuity, Difference-in-Differences (DiD), and Propensity Score Matching strengthen causal inference and reduce bias in observational data.

What You’ll Learn

• How Regression Discontinuity uses cutoff-based assignment to estimate treatment effects
• How Difference-in-Differences compares group changes over time to evaluate interventions
• How Propensity Score Methods statistically balance groups to reduce selection bias
• Applications of these methods in real-world healthcare and policy studies

Ideal for

Public health students, researchers, MPH candidates, clinicians, and analysts using observational data.

Core Concepts Covered

• Natural experiments & eligibility thresholds
• Matching techniques for causal inference
• Assumptions like parallel trends and balance diagnostics
• Strengths and limitations in real-world contexts

Lesson notes

Full written notes for this lesson are in production. Until they publish, watch the video above and use the summary on this page as your study guide.