A Research 2000 poll done for a Madison TV station and released yesterday shows Barack Obama winning Wisconsin over John McCain, 49%-43%. The poll, with its margin of error of +/-4%, might suggest to the ill-informed or hopeful McCain fans that the state remains in a "dead heat" or a "statistical tie" since the difference between the two--6%--is less than double that margin.
However, that's not how polling works, as Kevin Drum explains here:
In fact, what we're really interested in is the probability that the difference is greater than zero — in other words, that one candidate is genuinely ahead of the other. But this probability isn't a cutoff, it's a continuum: the bigger the lead, the more likely that someone is ahead and that the result isn't just a polling fluke. So instead of lazily reporting any result within the MOE as a "tie," which is statistically wrong anyway, it would be more informative to just go ahead and tell us how probable it is that a candidate is really ahead.By those standards, Barack Obama has a 93% likelihood of being ahead in the state. A mere 7% chance that McCain is really even or ahead of Obama is hardly enough to suggest that this is a dead heat or a statistical tie right now.
In fact, if you look at the Pollster.com composites, you can see that early May was the last time any polling firm found McCain to be ahead in Wisconsin, when Rasmussen Reports noted a 47%-43% split in McCain's favor. Every poll of the state since, even during McCain's convention bounce, shows Obama leading. The current average at the site is Obama 48.3%-McCain 43.9%. That's a 4.4-point difference, over what must be thousands of respondents. Using the same math as above, that means there is a greater than 99% chance that Obama is currently winning Wisconsin. We are not in a dead heat or a tie, and to pretend as much is ridiculous.
(Usual caveats apply--don't get complacent, don't forget to give to Obama so he can fund his GOTV here, and so on.)
No comments:
Post a Comment