Stata preconference short course

Causal inference with observational data

Instructor: Joerg Luedicke, Senior Social Scientist and Software Developer

Date: Wednesday, 2 September 2026

Time: 9:00 a.m.–1:00 p.m. ET

Fee: $25

Estimating treatment effects when the data are observational rather than experimental presents a challenge in many research areas. This course provides an introduction into the problems of causal inference that arise in the context of observational data and presents an overview of statistical methods and techniques to overcome these challenges. Commonly used techniques for estimating treatment effects such as regression adjustment, inverse-probability weighting, and propensity-score matching will be discussed. In addition, we will also present techniques for performing model checking and model selection, and we will introduce methods for dealing with more complex causal models such as causal mediation models. Finally, we will present several hands-on examples that demonstrate how to perform causal inference and treatment effect estimation. The examples are shown using the statistical software package Stata, but no prior knowledge of Stata is required. Attendees should be familiar with basic statistical methods and techniques such as regression modeling. [Read more ...]

Joerg Luedicke

Senior Social Scientist and Software Developer

Joerg Luedicke is a Senior Social Scientist and Software Developer at StataCorp LLC. He has a master's degree in sociology from Free University Berlin, Germany, and a PhD in sociology from Bielefeld University, Germany. Prior to joining StataCorp, he held positions as a research scientist and statistician at Free University Berlin, Bielefeld University, and Yale University. His interests lie in nonparametric statistics, survival analysis, discrete choice models, and causal inference.