October 19, 2010, 19:35
KallehSimpson's Paradox
Though perhaps I should have heard of Simpson's Paradox before, I haven't. I was introduced to it today when reviewing some research findings. I read some online sites about it (including the Wikipedia article), but here's my question: Wikipedia said that "Simpson's Paradox disappears when any causal relations are derived systematically – i.e. through formal analysis." Is anyone here familiar with the concept? If so, is that statement about it disappearing with formal analysis true?
October 20, 2010, 00:57
arnieI hadn't heard of this by name. but I was aware of the problems involved in interpretation when combining two sets of data without making allowance for group sizes, etc.
October 20, 2010, 05:35
GeoffI thought it had to do with the unlikely success of Homer, Bart, Marge, et. al.