Behind the Gallup Numbers (Likely Voters), I don't Get It

[louis c]

The latest Gallup tracking poll has Obama ahead by 8 points among registered voters. That's OK ,I think that's in line. Now, here's the part that seems curious. Gallup published its own survey of voter enthusiasm. Look at these numbers. The Gallup question on a poll conducted only days ago (Oct. 10-12) was this, "Compared to previous elections, are you more enthusiastic than usual about voting, or less?" The results were that 74% of Obama voters answered that they were more enthusiastic and only 15% replied that they were less. On the other hand, only 48% of McCain voters said that they were more enthusiastic and 40% said they were less. That's a plus 59% for Obama and just a plus 8% for McCain, for a whopping 51% difference in Obama's favor. If that's the case, how the hell can Obama's registered voters numbers decrease when it turns into likely voters? It deifies logic, considering that it's the exact same polling agency. With all the right wing news outlets (Drudge and Fox News) using this "traditional" likely voter number as though it were a tightening of the race, I'd like an explanation. It just doesn't make sense to me.

Link;
http://www.gallup.com/poll/111115/Democrats-Election-Enthusiasm-far-Outweighs-republicans.aspx

[mucifer]

1. They say younger voters might be motivated but they tend to not turn out on election day.

Maybe that's what they mean. Obama has a lot of new younger voters. We need to encourage them to early vote and then volunteer and prove these polls wrong.

[kikiek]

2. From what I understand Traditional is those who have voted before. It doesn't take into

account, therefore, any of the newly registered voters. Or voters already registered who historically haven't voted. In an election year like this when Dems are very motivated to go vote it misses a large portion of those who will.

[louis c]

3. But why only 4 points among the expanded likely voters?

[kikiek]

4. Expanded doesn't count them either. The most accurate one is the one where he has the largest lead.

[kikiek]

6. The other problem with that poll, according to huff post, is they polled equal dems and repubs.

11 pct more of the population id themselves as Dems. They said it isn't plausible that the Repubs made up that difference of voter regs in such a short time. And Dem voters are much more motivated this year.

[metapunditedgy]

7. If that's true, it's probably understimating the Obama vote,

because thinking back to 2000 and 2004, the Republican excitement for Bush was higher than for McCain, while the Democrat excitement for Kerry (and maybe Gore) was lower than currently for Obama. This would suggest to me that Democrats who weren't excited enough to vote previously (and therefore aren't counted in the poll) may show up in greater force this year. Hopefully.

[kikiek]

8. Yes I think they all are underestimating the vote, and enthusiasm.

[Tansy_Gold]

5. Because "likely" in polling-speak means. . . . .

. . . . whether or not they've voted in the past.


It's not a matter of them asking "how likely are you to vote?" because they don't base their "likely" on what the surveyed person says. They base it on what they've already done, meaning, voted in previous elections.

Therefore, a newly-registered voter who has never voted before is (almost never) counted as a "likely" voter in a poll.

Many of those newly-registered voters, however, may be among the most enthusiastic.


That's how it gets skewed. It's normal.


And the only poll that counts is the one that gets tallied on Nov. 4.



Tansy Gold

[louis c]

10. Here's what's obvious

Why would someone register for the first time, vote in a state primary, and then skip the most historic election in American history? Those newly registered voters will certainly come out in a very high number. Secondly, african-Americans have traditionally been one of the lowest demographics in elections. although they are reliably democratic by percentage, they turn out in low numbers. Logic should tell you that with Barack as the candidate, African American turnout should double, just as it did in the primaries. These numbers are not reflected in either model of Gallup's likely voters.

[Tansy_Gold]

11. Well, yes, some of it's obvious, but some isn't

And I don't say this to rain on anyone's parade but to remind everyone that it ain't over 'til it's over (thank you, Yogi!).

Because of all the voter registration drives, there MAY be a lot of newly registered voters who took advantage of a quick and easy and convenient way to register. At the DMV. At a table in front of the grocery store. Whatever. And I'm not saying that these people necessarily won't vote. A good portion of them, motivated by the current circumstances of candidates and conditions, will do it, as will some who have been registered for years but haven't been diligent voters.

But registering isn't voting, and the pollsters know this. Even voting in a primary, which usually has significantly lower turn-out (even this year) than a general election and therefore less stress, isn't the same.

There will be long lines this time. There will be frustrations. There may be bad weather, no place to park, etc., etc., etc. On the other side of the motivation coin, yes, there is the driving desperate need to rid this country of the festering sore that sits in the oval office.

All I'm saying therefore is that the pollsters can only base their "likely voter" model on verifiable patterns of behavior from the past, not on speculations for the future.

Do I personally think there will be significantly higher voter turnout? Yes. Do I personally think there will be significantly higher turnout in the African American voting bloc? You betcha! Do I personally think there will be significantly higher participation from newly-registered voters? Absofuckinglutely. But that's not how the pollsters determine "likely."

KMFX


Tansy Gold

[Suich]

9. What counts is the Electoral vote.

Check this site out: http://www.electoral-vote.com/

You can also see what was going on in 2004.