Random samples also mean there's more noise in the data. When someone tells you a fact , like "average crime is low", or "very few people get struck by lightni

Random samples also mean there's more noise in the data. When someone tells you a fact , like "average crime is low", or "very few people get struck by lightning", that average is skewed by useless noise. If there's high crime in your neighborhood, who cares about the low crime neighborhoods? If you work in a corn field all day, your risk of getting struck by lightning is quite high. That's the great myth with statistics. They are deliberately chosen to tell a story. By segmenting the data, you will get signal that matters to you. If my friends and coworkers are all leaving, that is more meaningful to me and the decisions I'll make over the next year than if the majority of the entire population is staying. It also has an impact economically on the State if specific segments of the work population are leaving.