In article <[EMAIL PROTECTED]>, Anders <[EMAIL PROTECTED]> wrote:
>I hope someone can help me. I have two continuous variables and I want >to find out if they correlate with each other. How do I do that? It's >data for an entiure population, not a sample. If you really have data on the entire poplulation of interest, then you just compute the correlation coefficient (see any elementary stats books for the formula, or just use SPSS), and see whether or not it is zero. I can tell you right now that the coefficient will almost certainly not be zero. So you really hardly need even bother. They're correlated. The fact that one can get the answer without looking at the data should be a sign to you that you aren't asking the right question. The problem is that you almost certainly DO NOT have data for the entire poplulation of interest, though you may well have data for everyone in a set that isn't the population of interest. You may even have data for an entire population that someone told you was the population of interest. But they were mistaken. Radford Neal ---------------------------------------------------------------------------- Radford M. Neal [EMAIL PROTECTED] Dept. of Statistics and Dept. of Computer Science [EMAIL PROTECTED] University of Toronto http://www.cs.utoronto.ca/~radford ---------------------------------------------------------------------------- . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
