What are drawbacks of correlation coefficient in psychology?

Question

What are drawbacks of correlation coefficient in psychology?

Answer 1

Well, there are limitations to the correlation coefficient in terms of drawing conclusions about causality.  For example, there might be a high correlation between cancer and smoking rates, but one does not mean that smoking causes cancer.  

Answer 2

If you test enough people, you will find a correlation between any two factors. This phenomenon has a name but i dont remember what it is. So theoretically if you keep getting a loq correlation, just test more people and eventually you can prove something like people with blonde hair are more likely to like cheese (or something totally random like that!)

Also, if two things are correlated it doesnt necessarily mean it is useful (look up the difference between statistical significance and clinical significance). Sometimes a low correlation may be very useful in the clinical setting, but it is often disregarded because of the low number.