Adam Graham
The meaningless CPAC straw poll
By Adam Graham
At the CPAC straw poll, Mitt Romney won for the third straight time with 20% of the vote, Governor Bobby Jindal finished second with 14%, Sarah Palin and Ron Paul were tied for third with 13%, Newt Gingrich got 10%, and Mike Huckabee got 7%. So, big Romney victory?
The big problem here is that the CPAC straw poll is pretty well worthless as a predictor of electoral fortunes or even top candidates.
The poll makes no sense when compared to public polls of Republican voters. Where's the disparity? You have to follow the numbers. The Internals of the poll. First thing that jumps out right away is that 52% of the people who voted in this poll are students, 59% are under the age of 25. At least 59% are male. 12% refused to say. So, if we assumed the "refuse to say" are half male and half female, which would mean this polling sample is 59% under the age of 25, 52% college students, and 65% male. Is there anywhere this demographic makes sense?
13% of the voters were from sponsoring organizations (mostly inside the beltway folks.) While 48 states were represented in the survey, can we say that they were represented proportionally? The beltway, Northern Virginia, and Southern Maryland will be far better represented than other states by nature of geography and having an easier time getting to where CPAC is held.
The CPAC straw poll is a notoriously bad predictor. In 1999, it awarded the straw poll victory to Steve Forbes. In 2005 and 2006, it went to George Allen, and of course 2007 and 2008 went to Romney.
What states are most likely under-represented? The ones that will decide the nomination. Iowa, New Hampshire, and South Carolina.
The bottom line: if you want to hold a conference so that big name conservatives can easily get there, hold it in Washington. If you want to hold a conference that will tell you anything about 2012, hold a simultaneous 3-day conference in Concord, Des Moines, and Charleston.
© Adam Graham
March 1, 2009
At the CPAC straw poll, Mitt Romney won for the third straight time with 20% of the vote, Governor Bobby Jindal finished second with 14%, Sarah Palin and Ron Paul were tied for third with 13%, Newt Gingrich got 10%, and Mike Huckabee got 7%. So, big Romney victory?
The big problem here is that the CPAC straw poll is pretty well worthless as a predictor of electoral fortunes or even top candidates.
The poll makes no sense when compared to public polls of Republican voters. Where's the disparity? You have to follow the numbers. The Internals of the poll. First thing that jumps out right away is that 52% of the people who voted in this poll are students, 59% are under the age of 25. At least 59% are male. 12% refused to say. So, if we assumed the "refuse to say" are half male and half female, which would mean this polling sample is 59% under the age of 25, 52% college students, and 65% male. Is there anywhere this demographic makes sense?
13% of the voters were from sponsoring organizations (mostly inside the beltway folks.) While 48 states were represented in the survey, can we say that they were represented proportionally? The beltway, Northern Virginia, and Southern Maryland will be far better represented than other states by nature of geography and having an easier time getting to where CPAC is held.
The CPAC straw poll is a notoriously bad predictor. In 1999, it awarded the straw poll victory to Steve Forbes. In 2005 and 2006, it went to George Allen, and of course 2007 and 2008 went to Romney.
What states are most likely under-represented? The ones that will decide the nomination. Iowa, New Hampshire, and South Carolina.
The bottom line: if you want to hold a conference so that big name conservatives can easily get there, hold it in Washington. If you want to hold a conference that will tell you anything about 2012, hold a simultaneous 3-day conference in Concord, Des Moines, and Charleston.
© Adam Graham
The views expressed by RenewAmerica columnists are their own and do not necessarily reflect the position of RenewAmerica or its affiliates.
(See RenewAmerica's publishing standards.)
