Read this about Gallup's Ohio Poll, Guys...

[sonicx]

I love poll watching. :) but maybe not with gallup

http://www.emergingdemocraticmajorityweblog.com/donkeyrising/archives/000650.php

Gallup's new poll in Ohio appears to have the same problem as their latest national poll--there is again a huge discrepancy between their results for registered voters (Bush 48, Kerry 47) and their results for likely voters (Bush 52, Kerry 44). Based on the numbers of registered and likely voters in the Gallup sample, this means that they are projecting that 90 percent of Bush supporters will vote but only 78 percent of Kerry supporters will vote. Again, this seems way out of line with evidence from previous elections and with other polls. In contrast, in Pennsylvania and Washington, Gallup's results for registered and likely voters are much more similar, and in line with other recent polls.


take that however you like it.

[Cha]

1. I'm taking that all these polls with bush

in the lead and doom and gloom for Kerry are just going to make Dems bust their butts that much harder!:D

[Union Thug]

2. why oh why are all these polls suddenly slanted towards republicans?

[Not Me]

6. It'll support the BBV fix they have in.

n/t

[snippy]

7. It is because of the methodology used to identify likely voters.

Every polling company uses a different methodology to identify likely voters. Some companies ask two or three questions relating to the respondents' intentions and voting history and use the responses to those questions to chose which respondents will be included as likely voters. Other companies ask a higher number of questions.

For example, when Gallup includes "likely voters" as a subset of registered voters it bases the determination of likely voters on responses to a series of seven questions. The answers to those seven questions by each respondent result in a numerical score being assigned to that respondent. All respondents then are ranked by this score from most likely to vote to least likely to vote.

Gallop then estimates the probable turnout percentage for the population being sampled. That estimated turnout percentage then is applied to the ranking of registered voters and used to determine a cutoff point for likely voters. So if turnout was estimated to be 53% then the 53% of respondents with the highest likely voter score are included in the results of the poll of likely voters and the 47% of respondents with lower scores are not.

Because nearly all polling companies, including Gallup, use questions about voting intentions, the different groups of likely voters will include a disproportionally high number of voters from the party that has the strongest support from its base at the time the poll was done and a disproportionally low number of voters from the party that has weaker support. The extent of the disproportionality will depend on the extent of the difference between the levels of support.

Most of the polls showing significant gains for Bush were taken during or shortly after the republican convention when worship of the LordGodBush was at a fevered pitch among the republican base and many democrats and independent Kerry supporters were somewhat demoralized. Consequently, the likely voter groups in these polls included a disproportionally high number of republicans. This should change as democrats and independents supporting Kerry become more optimistic.

One of the polls which does not show that significant of a gain for Bush is the Fox News poll. The reason Fox shows a smaller bounce for Bush is that it always has included more republicans than democrats in its definition of likely voters. Consequently, the increased intensity of Bush supporters following the convention did not result in a similar increase in the number of republican likely voters included in the Fox poll.

The Gallup methodology illustrates the importance of turnout and the fact that no pollster knows how to identify a likely voter. If turnout is high and includes many first time voters any poll that is accurate will owe much of its accuracy to luck.

[RevolutionStartsNow]

11. Thanks, Snippy! Good explanation of polling methodology

It confounds me, really, and I hate that I watch the polls, because I know how wrong they can be especially in a race like this.

Turnout, turnout, turnout. Can't say it enough. Wish I could wave a magical "turnout" wand and get people to the polls! As is it, next weekend I will be registering voters in California. Wish I could do this in a Swing State, where it seems that much more important.

[sonicx]

12. yes, if 60%+ of voting age pop. votes, we win. turnout baby!! nt

[sonicx]

3. kick

[cthrumatrix]

4. listen...nobody in Kerry camp is leaving and the debates will solve

the leadership issue.

[tritsofme]

5. Gallup's likely voter model looks good on paper

Gallup tries to avoid the lack-of-interest problem by specifying that 55% of its sample (which corresponds with a 55% projected turnout in the election) will be considered likely voters. Gallup asks each respondent seven LV screening questions, and gives each person an LV score of 0 to 7. The top 55% are classified as likely voters. This classification works even if interest is low early in the year, because low-interest respondents are considered likely voters as long as they are in the top 55% of the sample on the overall LV scale.


But its been giving out some weird number the past few weeks.

[snippy]

8. The composition by party of Gallup's likely voters varies with the

level of intensity of the support in each party when the particular poll was done. So if republicans are having multiple wargasms and democrats are demoralized at the time the poll is done there will be a higher percentage of republicans in the group identified as lkely voters.

[DarthDem]

9. Then That Would Mean

That the poll was totally unreliable during, say, a party convention, with one side all fired up and fascist-like and the other just sitting back and watching in horror.

[snippy]

10. Every poll is totally unreliable as a predictor of the outcome of an

election that is still two months away. But many polls can be reliable as showing the state of the race at the time the poll was done and several polls done by the same polling company over time can be reliable to show a trend. Bounces are called bounces because they show support going up BUT then coming back down. We will have to wait and see whether Bush got a bounce or made a permanent move. I think it is a bounce.