---
title: "Session 3 — Histograms and boxplots"
subtitle: "Probability and Statistics (MIR)"
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}
ci <- read_csv("data/country_indicators.csv")
```

# Histogram

```{r}
ggplot(ci, aes(life_expectancy)) +
  geom_histogram(bins = 10, colour = "white") +
  labs(title = "Life expectancy", x = "Years", y = "Count")

ggplot(ci, aes(gdp_per_capita)) +
  geom_histogram(bins = 12, colour = "white") +
  labs(title = "GDP per capita", x = "USD", y = "Count")
```

# Boxplots

```{r}
ggplot(ci, aes(x = "", y = life_expectancy)) +
  geom_boxplot() +
  labs(x = NULL, y = "Life expectancy")

ggplot(ci, aes(x = factor(recent_internal_conflict),
               y = life_expectancy)) +
  geom_boxplot() +
  labs(x = "Recent internal conflict (0/1)", y = "Life expectancy")
```

**AI paradox:** If a chatbot “summarizes” the plot without showing it, demand the figure — shape is the point.