Random testing means selecting people for a test arbitrarily. Take, for example, a test meant to figure out how common COVID-19 exposure is in a town of 10,000 people. Testing everyone who works at the town's hospital would not be representative of the whole town, since hospital workers are at higher risk of being exposed to COVID-19. An example of how to generate a random sample of 200 people would be to use a list of all the residents in the town, assign them each a unique number between 1 and 10,000, and then run a random number generator that produces 200 unique numbers between 1 and 10,000, and then use those 200 people for the test.

Drafted 22 April 2020