Statistics draws a careful conclusion about a whole group from a random sample of it. The mean and standard deviation of a bell-shaped data set fit a normal model, and the 68-95-99.7 rule turns the model into percentages. A sample survey describes a group, an experiment assigns treatments at random and can speak of cause, and an observational study only watches. A simulation shows what chance alone produces, so a result can be called surprising or not, a margin of error can be built, and two treatments can be compared. Every report is read by asking who was measured, how they were chosen and what was assigned.