chapter  18
Modeling Change over Time: From Description to Explanation
Pages 8

In some circumstances, and developmental research is a good example, an investigator desires not merely to establish that a relationship exists between two or more variables, but also to specify or determine the actual form of that relationship. There is an obvious difference, for example, between the assertion that 19-year-olds are taller than 13-year-olds and the assertion that the relationship between height and age is a negatively accelerated function during the teens. An even more precise statement, which specifies an actual formula relating height to age, might be constructed for a given data set.