Practical Data Analysis with JMP®

Two-by-Two: Bivariate Data

Chapter 3 was devoted to the concept of distributions and presented techniques for displaying and summarizing the variation within a single variable: univariate distributions. In many statistical investigations, we're interested in how two variables vary together (or not). For example, nutritionists might ask questions about how consumption of carbohydrates affects weight loss; marketers might ask questions about whether a particular demographic group responds positively to an advertising strategy. In these cases, it's not sufficient to look at one univariate distribution or even to look at the variation in each of two key variables separately. We need methods to describe the covariation of bivariate data, which is ...