Causal Inference: The Mixtape is a book for practitioners. The purpose of the book is to allow researchers to understand causal inference and work with their data to answer relevant questions in the area. It is the emphasis on the use of statistical software that sets Cunningham’s book apart. In each chapter, theoretical details are clearly presented, followed by how to apply the theory to answer causal inference problems using statistical software. The examples are accompanied by readily available data and replication code.
The book starts with some basic concepts and then moves into the most commonly used causal inference models and estimators. Topics include directed acyclical graphs, potential outcome models, matching and subclassification, regression discontinuity, instrumental variables, difference in differences, and synthetic control. The inclusion and discussion of synthetic control and directed acyclical graphs differentiates this book from others in the literature, which do not cover these topics or do so tangentially.
The structure of the book lends itself to teaching a course on causal inference, but at the same time it is a useful reference for any researcher delving into causal inference.
Do Not Confuse Correlation with Causality
Optimization Makes Everything Endogenous
Example: Identifying Price Elasticity of Demand
Estimation Using an RDD
Challenges to Identification
Replicating a Popular Design: The Close Election
Regression Kink Design
Intuition of Instrumental Variables
Homogeneous Treatment Effects
Parental Methamphetamine Abuse and Foster Care
The Problem of Weak Instruments
Heterogeneous Treatment Effects
Popular IV Designs
Data Exercise: Survey of Adult Service Providers
Providing Evidence for Parallel Trends Through Event Studies and Parallel Leads
The Importance of Placebos in DD
Twoway Fixed Effects with Differential Timing