Question

A Six Sigma Black Belt runs two regression models. In one, she regresses yield (y) versus temperature (x) measured in degrees Farenheit. In the second she regresses yield versus dwell time measured in seconds.

She states in a meeting that dwell time is more highly correlated with yield since the slope of that model is larger than the slope of the yield versus temperature model. How do you respond to her claim?

Answer 1

We can't predict correlation between any two variables by using the slope of the regression line. The correlation is determined by how close the points are to the regression line, more closer implies that those variables are more correlated. For determining correlation we have to use R-square term which takes into account how far each point is from the regression line. The value which is more closer to 1 implies that those set of variables are more correlated.