---
title: "Session 14 — Adjust for GDP"
author: "Prof. Dae-Jin Lee (daelee@faculty.ie.edu)"
date: "Fall 2026"
output:
  html_document:
    toc: true
    theme: flatly
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE, message = FALSE, warning = FALSE)
library(tidyverse)
```

```{r}
world <- read_csv("data/country_indicators.csv")

m_naive <- lm(life_expectancy ~ democracy_score, data = world)
m_adjusted <- lm(
  life_expectancy ~ democracy_score + log(gdp_per_capita),
  data = world
)

summary(m_naive)$coefficients
summary(m_adjusted)$coefficients
```

Compare the democracy coefficient across models. Movement suggests GDP is entangled with the association — **still not proof** that the adjusted coefficient is the causal effect of democracy (omitted variables, reverse causality, measurement, etc. can remain).