Teaching resources for statistics and R: Monte Carlo simulation of t-test p-values

Why doesn't a small p-values automatically prove the alternative hypothesis? In this third article of my series on teaching resources for stats and R, I share a graphical Monte Carlo simulation I developed to teach my students about classical hypothesis testing and p-values.