I have two variables for which I want to calculate the correlation coefficient. One variable has an interval level (the values are 200, 400, 600, 800) the other has a ratio level. Is there a special correlation coefficient which can be applied?
The Pearson (product-moment) correlation works with scales of interval or ratio strength. In most statistical software, it is the "default" option when you select a correlation analysis.
I would be hesitant to do anything before I had a 2D plot of the data. I assume the two variables are paired?. The reason I ask is, I once had a colleague who asked how do I regress y on x if n(y)=14 and n(x)=10 and sure enough they were not paired observations.. Best, D. Booth