# Load packages
library(tidyverse)
library(foreign)
library(quantreg)
library(modelsummary)
library(kableExtra)
# Load the shared function library
source("R/ACA_Function_Library.r")
# Set seed for reproducibility
set.seed(42, kind = "L'Ecuyer-CMRG")Applied Causal Analysis
Covariate Adjustment & Quantile Regression
0.1 Welcome
This lecture covers two important topics in the analysis of randomized controlled trials (RCTs):
Covariate Adjustment — How adding pre-treatment covariates to a regression can improve precision without introducing bias, even in a randomized experiment.
Quantile Treatment Effects (QTE) — How treatment effects can vary across the outcome distribution, and how to estimate and visualize them using quantile regression.
0.2 Datasets
We use two datasets throughout this lecture:
| Dataset | Description |
|---|---|
| Butler & Broockman (AJPS, 2011) | A field experiment on political representation and race. Legislators received emails from constituents with racially distinct names. |
| Penn Job Training Experiment | An experiment testing the effect of job training on unemployment duration. |
0.3 Setup
The code chunk below loads all required packages and the shared function library used throughout the lecture. Run this first before proceeding to any other section.