---
title: "Session 12 — CI and test connection"
author: "Prof. Dae-Jin Lee (daelee@faculty.ie.edu)"
date: "Fall 2026"
output:
  html_document:
    theme: flatly
---

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

```{r}
survey <- read_csv("data/survey_attitudes.csv")
two <- survey |>
  filter(education %in% c("Bachelor", "Postgraduate"))

tt <- t.test(support_international_cooperation ~ education, data = two)
tt$p.value
tt$conf.int
```

If the 95% CI for the difference excludes 0, the two-sided test rejects at $\alpha=0.05$ (same assumptions). Prefer reporting the CI because it shows **magnitude**.