This document discusses levels of measurement and choosing the correct statistical test. It begins by distinguishing between continuous and discrete variables, noting that continuous variables have many possible values while discrete variables have few categories. It then explains how this distinction relates to two general classes of statistical tests - those based on the normal distribution for continuous variables, and those based on the binomial distribution for discrete variables. The document provides examples of common statistical tests that fall into each class. It concludes by discussing debates around levels of measurement and notes researchers should be careful about artificially categorizing continuous variables.