What is correlation? What are the different types of correlation? Why is it important to determine correlation

Question

What is correlation? What are the different types of correlation? Why is it important to determine correlation

Answer 1

Correlation in statistics is the probability that a change in Y will occur in parallel to a change in X.  Correlation can be positive (i.e. Y increases with X) or negative (Y decreases as X increases).  Correlation can be linear (Y changes in direct proportion to changes in X,  Y = k x X) or nonlinear (Y changes according to a more complex curvilinear process, an infinite number of such possiblilities exist).  Science is very much a process of relating phenomena in nature to understand what causes what.  For instance, one could ask does increasing the temperature of a gas change the pressure in a fixed volume.  Measurements would show that increases in pressure are correlated to increases in temperature.  This is how we develop principles to explain our universe.  But, it is very important to point out that correlation per se does NOT imply causation.  Strong correlation between two parameters (A and B) could occur if each is responding independently to a third stimulus.  In this case changes in B would not be CAUSED by changes in A.