In hypothesis testing there are two types of errors that can result in the experimenter drawing the wrong conclusion. These errors are expressed as probabilities (α and β risks), and are always present to some extent. The goal is to make sure these errors are reasonably small, given the actions that will or will not be taken as a result of the hypothesis test.

Type I Error – A conclusion that a true change has occurred in the underlying population, when in reality it has not. The probability of making a Type I error is the α risk.

Type II Error – A conclusion that the underlying population has not changed, when it reality it has. The probability of making a Type II error is the β risk.

Typical values for acceptable α and β risks are 5% and 10% respectively.