This Yale lecture, part of the course Understanding Medical Research: Your Facebook Friend is Wrong, tackles the multiple comparisons problem in statistics. The instructor explains why running many statistical tests on the same dataset inflates the chance of finding a false positive, using examples drawn from medical studies where researchers test dozens of outcomes or subgroups looking for significant results. The lecture walks through why a single p-value threshold like 0.05 becomes misleading once dozens of comparisons are made, and introduces corrections such as the Bonferroni adjustment that researchers use to control the overall error rate. The point lands on how to read published studies skeptically, especially ones that report a surprising result buried among many tested hypotheses. At seventeen minutes, it is a compact, classroom-style explanation aimed at viewers with some grounding in basic statistics who want to evaluate medical claims more critically.