Correlation: The Rise and Fall of Relationships
IN THIS CHAPTER
Understanding what correlation is all about
Connecting correlation and regression
Drawing conclusions from correlations
In Chapter 14, I show you the ins and outs of regression, a tool for summarizing relationships between (and among) variables. In this chapter, I introduce you to the ups and downs of correlation, another tool for looking at relationships.
I use the example of SAT scores and GPA from Chapter 14, and show how to think about the data in a slightly different way. The new concepts connect to what I show you in the preceding chapter, and you’ll see how that works. I also show you how to test hypotheses about relationships and how to use Excel functions and data analysis tools for correlation.
A scatterplot is a graphical way of showing a relationship between two variables. Figure 15-1 is a scatterplot that represents the GPAs and SAT scores of 20 students at the fictional Sahutsket University. The GPAs are on a 4.0 scale and the SATs are total SAT scores.
Each point represents one student. A point’s location in the horizontal direction represents the student’s SAT. That same point’s location in the vertical direction represents the student’s GPA.