When a poll is too good to be true
May. 21st, 2012 09:21 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
maxomaiCheck out this poll from Vanderbilt:
Tennessee
Obama 41
Romney 42
MOE 4
N = 1002
Obama's within one point of Romney in deep red Tennessee! Wow!
Just one problem: this is a poll of Tennessee residents. The same poll shows Romney up seven with registered voters, which is a proper subset of residents. The likely voter model in Tennessee, when those polls start coming out, will be even worse for Obama. And in November, barring the absolute collapse of the Romney campaign and a consequent Obama landslide, Tennessee will go in Romney's column.
Tennessee
Obama 41
Romney 42
MOE 4
N = 1002
Obama's within one point of Romney in deep red Tennessee! Wow!
Just one problem: this is a poll of Tennessee residents. The same poll shows Romney up seven with registered voters, which is a proper subset of residents. The likely voter model in Tennessee, when those polls start coming out, will be even worse for Obama. And in November, barring the absolute collapse of the Romney campaign and a consequent Obama landslide, Tennessee will go in Romney's column.
