How is association in statistics identified and demonstrated?

I need a clear understanding of this. Most of the stuff that I have read has been very vague.

How is association in statistics identified and demonstrated?
Association and be quantified with a correlation coefficient. Two common coefficients are pearson's (parametric - i.e. can assume a distribution) or spearman's (non-parametric).





Both of these correlation coefficients range from -1 to 1.


1 -%26gt; perfect direct relationship


-1 -%26gt; perfect inverse relationship


0 -%26gt; no association whatso ever.





All values inbetween have sort of a vague interpretation...so with a corelation coefficient you pretty much have to speak very generally about how strong an association really is....





As another analysis of association, you can TEST for association between 2 variables. If the p-value is small enough, typically %26lt;.05, you can say that the 2 variables are associated %26amp; it is statistically significant.





Note that association is not the same as causation. Just because 2 variables are associated doesn't mean that 1 variable directly causes the other.





I agree that it can be difficult to find a direct answer to a statistics question... so to answer your question directly, to quantify association I would look at:





1) correlation coefficients (quantifies association)


2) tests for association (determines if association is statistically significant).





Check out this site for some good info on this:


http://www.med.umich.edu/csp/Course%20ma...





good luck~