This document discusses using simple linear regression to describe relationships between variables in data. It explains that regression finds the linear equation that best describes how a dependent variable (y) changes with an independent variable (x). The equation is the line that minimizes the sum of the squared residuals (deviations from the observed data points). Examples are given of regression analyses conducted to estimate the cost of computer networks based on number of computers, estimate real estate values based on house size, and forecast housing starts based on mortgage rates.