Statistics and polls are not science. The latter can be worded in such a way that it can be geared towards receiving the desired answer. The former can be manipulated with language to mean anything you want them to. For instance, smokers and lung cancer. The actual percentage of smokers who die from lung cancer is surprisingly low given the cancer hammer with which we're continually bludgeoned. I forget the exact number, but it's over 90%, so you can say, more than 9 in 10 smokers will not die from lung cancer, more than 9 in 10 non-smokers will not die from lung cancer. So do you have the same chances of dying from lung cancer or not? Well, in reality, no. While incidence difference is somewhat significant, as given above, it's not as great as we're led to believe.