Causal Inference in R

Welcome! This workshop is taught by Lucy D’Agostino McGowan, Associate Professor of Statistical Sciences at Wake Forest University.

Over the course of this workshop, we’ll work through an introduction to causal inference, from specifying a causal question and drawing assumptions, to estimating and communicating causal effects (using R!).

Slides

• The Whole Game
• From Casual to Causal
• Estimating counterfactuals
• Causal Diagrams
• Causal Estimands
• Building Propensity Score Models
• Propensity Score Diagnostics
• Outcome Models
• Continuous exposures and g-computation
• Sensitivity Analyses
• Exercises in R

R Setup

Posit Cloud

To follow along, the preferred way to access the materials is through this Posit Cloud Workspace. After clicking this link, you will sign up for a (free) account (or sign in if you already have an account). Then make a copy of the project by clicking “Save a Permanent Copy”:

You can download these files locally to your machine after, if you would like to. The reason this method is preferred is I have pre-installed all of the needed packages here.

Local

If you want to work through the material locally, you’ll need R and RStudio installed. Then install the required packages by running the following in your R console:

install.packages(c(
  "broom",
  "dagitty",
  "ggdag",
  "ggokabeito",
  "gtsummary",
  "halfmoon",
  "MatchIt",
  "propensity",
  "rsample",
  "survey",
  "tidyverse",
  "remotes"
))

remotes::install_github("LucyMcGowan/touringplans")

You can then download a zip folder of all exercises from here: All Exercises

Due to time constraints during the workshop, I will not be able to troubleshoot local installation issues, but I am happy to discuss them afterwards.