This lesson from Yale's Introduction to Medical Software course tackles a term that gets thrown around loosely in medicine: statistical significance. The instructor separates it from real-world clinical importance, then works through the machinery behind it, null hypothesis testing, how p-values are calculated, and the mechanics of an unpaired t-test. Sample size gets particular attention, with a walkthrough of how it shifts a trial's statistical power and its odds of detecting a true effect. The lecture also traces where the p less than 0.05 threshold actually came from historically, rather than treating it as an arbitrary rule to memorize. Aimed at students building software for clinical trial design, the lesson stays close to vocabulary and calculation rather than abstract theory, so viewers come away able to read a trial's statistics section and know what the numbers are actually claiming.