Sunday, October 21, 2012

Gallup's Likely Voter Screen

I wrote the other day about Gallup's tracking poll and how it is out of step with every other pollster. This morning, Gallup's tracking poll release has Romney ahead by 7 (not a typo) among likely voters and ahead by 3 among registered voters.

Whatever is going on with Gallup, this is not just a problem with their likely voter screen. They've got Romney up by 3 among registered voters! Compare that with the latest NBC/WSJ poll just out this morning. They've got Obama and Romney tied at 47 among likely voters but Obama is up by 5 among registered voters (I will have more to say about this poll later in the day).

So, what is going on with Gallup? Given that they have such a different result among registered voters, it means there is something different in their sampling. They are simply reaching a different mix of respondents.

Alan Abramowitz (whose work is really great) has a good theory. Gallup does not release information on the racial composition of its samples but it appears they are over-representing white voters:
Although Gallup does not report the racial composition of its likely voter sample (or any of its other samples), based on the results presented in their October 16 report on the standing of the presidential candidates among whites and non-whites, one can use interpolation to estimate the racial composition of the likely voter sample. The results show that about 80 percent of Gallup's likely voter sample consisted of non-Hispanic whites while about 20 percent consisted of non-whites.

Gallup's estimate that only 20 percent of this year's likely voters are non-white is far lower than the 26 percent non-white share of voters found in the 2008 exit poll or even the 23 percent share found in the 2004 exit poll. It is actually very close to the 19 percent share found in the 2000 exit poll. So according to the Gallup tracking poll, the racial composition of the 2012 electorate will be similar to that of the 2000 electorate despite the dramatic increase in the nonwhite share of the voting age population that has occurred in the past 12 years.
It is probably the case that the non-white portion of the electorate is likely to be much closer to the 2008 number than the 2000 number. Indeed, there is every possibility there will be fewer whites in the electorate in 2012 when one thinks about where growth in the electorate is occurring. For now, it is probably best to simply ignore Gallup's numbers as they are radically out of step with every other pollster's numbers.

UPDATE: By the way, IBD/TIPP has their new tracker out this afternoon and they have Obama ahead by 6 points among likely voters. This seems equally odd and I don't know exactly what's up there. It is a bit less relevant because they don't have the reputation that Gallup does and therefore, they affect the conversation far less. The one thing I do see in their poll that is very different is their numbers on male voters (Obama leading by 1). That's not right, I can assure you but how they got there is anybody's guess.

No comments: