Causal Inference: Asynchronous

Our goal is to provide a clear and accessible understanding of how quasi-experimental methodology approximates the randomized control trials that have been used traditionally to reveal causality. A strong understanding of the assumptions that are required to make the inferences will be made explicit. Additionally, we will provide a thorough initial introduction into understanding the causal graphical models that are becoming more popular in the health and behavioral sciences. This workshop is designed to provide participants with the knowledge to determine which approach is most useful for revealing causal information for their research question.

Our workshop is designed for graduate students, post-doctoral fellows, faculty, and research scientists from the behavioral, social, and health sciences. Although it is recommended that participants have a working knowledge of the general regression model, these core concepts will be reviewed at the beginning of the workshop. Participants who seek a more in depth review prior to attending may also wish to watch our Linear Regression Playlist on YouTube.

Chapter 1: General Introduction
1.1 Introduction to the Causal Modeling Process
1.2 Randomization and Experiments
1.3 Potential Outcomes Framework
1.4 Counterfactuals

Chapter 2: Differences-in-Differences Designs
2.1 Review of Linear Regression
2.2 Relationship to Analysis of Covariance
2.3 Extension to Differences-in-Differences-in-Differences
2.4 Synthetic Controls

Chapter 3: Regression Discontinuity Designs
3.1 Introduction to the Standard Regression Discontinuity
3.2 Fuzzy Regression Discontinuity
3.3 Inference in Regression Discontinuity

Chapter 4: Longitudinal Causality
4.1 Piecewise Regression
4.2 Interrupted Time Series
4.2 Granger Causality

Chapter 5: Propensity Scores
5.1 Estimation via Logistic Regression
5.2 Matching Methods
5.3 Propensity Scores as Weights
5.4 Propensity Score Subclassification

Chapter 6: Graphical Models
6.1 Mediation Models
6.2 Directed Acyclic Graphs
6.3 Identification of Causal Effects
6.4 Search for Causal Graphical Models

Demonstrations are provided in R. Note that R can be downloaded for free. While it is helpful to have some familiarity with R, this is not necessary. The lectures for the workshop are software-independent.

Causal Inference is a five-day workshop originally taught live via Zoom by Doug Steinley. Daily lectures were held from 10:00 to 3:30 with morning, lunch, and afternoon breaks. Sessions consisted of comprehensive lectures, live software demonstrations of real-data examples in R, response to participant questions, and general discussion.

Self-paced participants receive lifetime access to all course materials, including complete course notes, lecture recordings, software demonstration notes, software demonstration recordings, and data and code for all examples. You can revisit these materials any time you like, without worrying about expiration dates. Pretty awesome, huh?