Interrupted Time Series for Policy Evaluation and Analysis

When randomized controlled trials are impractical, how do we prove that a new policy actually worked? Interrupted Time Series (ITS) analysis is one of the most powerful quasi-experimental methods available for evaluating the real-world impact of public health, social, and economic interventions. This course guides you through the entire lifecycle of a policy evaluation, teaching you how to turn raw historical data into clear, statistically sound insights. You will learn to formulate evaluation questions, structure your data, and build regression models that isolate the true effect of an intervention from underlying trends. What you'll learn: 1. Understand the core principles of quasi-experimental design and interrupted time series frameworks. 2. Structure and clean longitudinal datasets using modern data analysis practices. 3. Build segmented regression models to estimate immediate level shifts and long-term trend changes. 4. Control for confounding factors including seasonality, autocorrelation, and historical events. 5. Interpret statistical outputs and translate complex regression coefficients into actionable policy insights. 6. Evaluate model assumptions to ensure your findings are robust and scientifically valid. The course starts with foundational definitions and the logic of causal inference before moving step-by-step through data preparation, regression modeling, and diagnostic testing. You will work with clear, written explanations and practical code examples designed to build your confidence. This program is designed for aspiring policy analysts, social science researchers, public health professionals, and data analysts who want to add a powerful evaluation tool to their skillset. No advanced prior knowledge of time series forecasting is required. Start reading today and learn how to measure the true impact of policy changes with statistical confidence.