The results are not significant at all. As I said in my previous post, I ran a simple simulation with pseudorandom numbers and obtained very similar results. I would be seriously surprised if the average was exactly zero, or even near zero.newp175 said:Like I said, you sometimes get the opposite results from what you are trying to achieve. The site also has some references (amongst other things) where in similar studies these observations were made. Anyway, as long as the results are significant, this works. Does anyone have any suggestions on why though?