Introducing Correlation and Regression
In This Chapter
Getting a handle on correlation analysis
Understanding the many kinds of regression analysis
Correlation, regression, curve-fitting, model-building — these terms all describe a set of general statistical techniques dealing with the relationships among variables. This chapter provides an overview of the concepts and terminology that I use throughout Parts IV and V.
Introductory statistics courses usually present only the simplest form of correlation and regression, equivalent to fitting a straight line to a set of data. But in the real world, things are seldom that simple — more than two variables may be involved, and the relationship among them can be quite complicated. You can study correlation and regression for many years and not master all of it. In this chapter, I cover the kinds of correlation and regression most often encountered in biological research and explain the differences between them. I also explain some terminology — predictors and outcomes; independent and dependent variables; parameters; linear and nonlinear relationships; and univariate, bivariate, multivariate, and multivariable analysis.