12. A correlation of r = 0.80 is stronger than a correlation of r = -0.80. 13. The intercept in a regression equation can never be negative. 14. The sign of the slope in the regression equation always matches the sign of r. 15. To computer, the form of the relationship must be non-linear. 16. If you correlate two quantitative variables, called "x" and "y," changing the units of measurement for "x" and "y" will lead the correlation to change. 17. A very strong relationship between variables can result in a value of r that is larger than 1 . 18. If you correlate two quantitative variables, called "x" and "y," the correlation will not change if you switch which variable is "x" and which is "y." 19. To computer, both variables must be quantitative (or numerical). 20. The correlation coefficient is not affected by outliers