Here’s an interesting example of why you should never just look at an *r* value and a regression model without looking at the scatterplot first. Anscombe’s Quartet consists of four data sets. Each has the same sample size and nearly the same mean and standard deviation for the *x* and *y* variables, *r* values, and regression coefficients. But in only one of the data sets is the correlation and regression analysis correct.

http://en.wikipedia.org/wiki/Anscombe’s_quartet

Advertisements