Measuring the "Howard Stern Effect" - FrankLog Somehow I doubt that this works. The summary is that Howard Stern actually caused voters to go to Bush due to his comments against the administration. In a shocking surprise, this is a great example of people not understanding statistics. Because there was no control group and because no one held all other variables stationary, you can't actually do this kind of analysis. Especially since Bush's numbers were up EVERYWHERE which means you could draw the exact same conclusions about states where he was not on the radio. Interesting numbers about voter turn out though... 