New Florida poll by Quinnipiac

[Friar]

Democratic challenger John Kerry leads President George W. Bush 47 – 41 percent  among Florida voters, with 4 percent for independent candidate Ralph Nader, according to a Quinnipiac University poll released today.  With Nader out of the race, Sen. Kerry leads President Bush 49 – 42 percent.

From August 5 - 10, Quinnipiac University surveyed 1,094 Florida registered voters with a margin of error of +/- 3 percentage points.

http://www.quinnipiac.edu/x12887.xml

[A18]

Polls with registered voters always favor Democrats and don't mean anything. You need a likely voters poll.

[Friar]

Because you say so?

Remember that the democrats are very energized this time. So, I expect a higher torunout and in my oppinion the 'likely voters' polls are less reliable this year because they can't 'catch' the anti bush sentiment.

Even if you don't agree with me, a 47-41 with 1094 sample shuld make Bush very worried.

[MODU]

ARG - Bush 43%  Kerry 50%
Strategic Vision - Bush 48% Kerry 47%
CNN/USA/Gallop - Bush 50% Kerry 46%
Mason Dixon - Bush 48%  Kerry 46%
Insider Advantage - Bush 46%  Kerry 46%

Take your pick.  Don't start your celebrations yet.  Besides, as the poll summary reads, Kerry is still showing some signs from his convention bounce.  Bush has yet to have his.  

[CARLHAYDEN]

Polls with registered voters always favor Democrats and don't mean anything. You need a likely voters poll.

Philip,

First, if your statement were rephrased to state that Democrat candidates usually perform better in voter registration polls than they do among actual voters in a general election, then it would be correct.  Qualifiers are important in this area, and superlatives should usually be avoided.

Second, a mathematical deconstruction of the poll in question demonstrates that it is more pro-Democrat than actual Florida voter registration.

If you go the links section of this forum, and then go to the Florida link, you will see that the most recent available (June) voter registration by party statistics show that the partisan registration is broken down as follows:

Republicans     3,652,311     38.07%
Democrats       4,009,765     41.80
Others             1,931,732     20.13
Total                9,593,808    100.00

If you then go to Q.1 of the survey, and multiply the data from that question by the correct partisan allocation, you will see that Jeb Bush's approval would be 47%, NOT 45%, as indicated by the poll.  This is an indication that the poll contains a base more Democrat friendly than the actual voter registration base.

If the base is further modified to project actual voters we find that the approval numbers for Jeb Bush change to 48% approve, 42% disapprove, instead of the 45% approve, 44% disapprove, as suggested by the survey.

Finally, Quinnipiac (a Connecticut college), has a history of of their polls being more Democrat than actual results.

[The Vorlon]

I usually cut the student's at Quinnipiac a bit of slack, but on this poll, I am afraid they have strayed a little too far from home.

You will note that this sample has Kerry winning Hispanics about 2 to 1, which in Florida is just laughable, and also they specifically mention that hispanics are "less than 75" out of a sample of more than 1000.

In reality the Hispanic + Cuban community is about 12% in Florida.

I am just "eye-balling" it, but this poll also looks like it is +6 or +7 to the Dem side in terms of voter ID.

(According to Exit polls, in 2000 it was +2 to the Dem side among those who actually voted in Florida.)

This poll has the same problem the ARG poll structurally has in that they don't interview in Spanish, and hence miss the entire, and strongly pro-Bush "Little Havana" community.

Crank voter ID back to "normal" and actually talk to some Cuban folks and this poll suggests Florida is very, very close.

I have posted it before, and I will post it again....

It's a Uni-poll, it's summer, and Florida is an absolute bugger to poll.

Mason Dixon will be out with a Florida poll just before the GOP Convention, let's all wait for that one...

All together now... let's sing in Harmony...

"If a summer time poll (especially in Florida) does not say Mason-Dixon....."

For what it's worth, my model has Florida in the Tied-Kerry+2 "bracket" and in the dead of summer, I trust my models a lot more than just about any poll.

As always... "Hey, It's one poll"....

We will see what happens....

[Beefalow and the Consumer]

For what it's worth, my model has Florida in the Tied-Kerry+2 "bracket" and in the dead of summer, I trust my models a lot more than just about any poll.

As always... We will see...

It's interesting that you have Missou leaning Bush, but Arkansas leaning Kerry.  Do you really think Bush could win MO and lose AR?

[MODU]

That's why I included the last M-D poll showing Bush up 2.  

[ThePrezMex]

I just want to make some points about the Cuban community and Hispanics in Florida.
The bulk of the Cuban community in Florida came to the United States since the beginning of the Cuban revolution in 1959. Most of them are now second and third generation Cubans that are fluent in English. Even those that came originally, were either kind of fluent in English or have become now after 40 years. So, I don't think that language is a huge barrier to identify results among the Cubans.
Moreover, even if they have traditionally supported strongly the Republican candidates in the past, we would like to know what effects are causing the recent policy changes that the Bush administration decided about restricting traveling to Cuba; have those changes undermined Bush's support among Cubans? if so, by how much? or has a negligible effect?
About other Hispanics if Florida. Well, basically Mexicans don't amount to much there (they traditionally vote more Democratic). You have a huge presence of people from Central America, Puerto Rico and South America, especially Venezuela, Brazil and Argentina. They all could have some of the language problems that Vorlon is talking about. But also many of them have come recently to the United States - don't know how many of them can vote, or have voted in the past (are there exit polls that show how Non Cubans Hispanics have voted in Florida?)

[MODU]

I read an article not too long ago discussing how incumbants have used Cuba as a means to attract the Cuban-American votes.  Each time they've come up for re-election, they've tighten the screws on Cuba, either by decreasing aid, imports/exports, travel, diplomatic relations, etc.  Now, the article didn't say how much of an impact (good or bad) these policies had on the election, but it was a re-occuring pattern.

[The Vorlon]

I just want to make some points about the Cuban community and Hispanics in Florida.
The bulk of the Cuban community in Florida came to the United States since the beginning of the Cuban revolution in 1959. Most of them are now second and third generation Cubans that are fluent in English. Even those that came originally, were either kind of fluent in English or have become now after 40 years. So, I don't think that language is a huge barrier to identify results among the Cubans.
Moreover, even if they have traditionally supported strongly the Republican candidates in the past, we would like to know what effects are causing the recent policy changes that the Bush administration decided about restricting traveling to Cuba; have those changes undermined Bush's support among Cubans? if so, by how much? or has a negligible effect?
About other Hispanics if Florida. Well, basically Mexicans don't amount to much there (they traditionally vote more Democratic). You have a huge presence of people from Central America, Puerto Rico and South America, especially Venezuela, Brazil and Argentina. They all could have some of the language problems that Vorlon is talking about. But also many of them have come recently to the United States - don't know how many of them can vote, or have voted in the past (are there exit polls that show how Non Cubans Hispanics have voted in Florida?)

You make a good point re 2nd Generation+ Cubans and language..

I should rephrase..

When a really good firm does an area with a high hispanic population, they typically use bi-lingual operators, and "about" 40% of the interviews you do of Hispanics will be in Spanish.

Many of this 40% might be able to handle the interview in English, but (not surprisingly) the validity of your results goes waaaaay down (too many undecides, lots of incomplete interviews, etc)

Exit polls are sketchy at best regarding cuban-hispanics voting for Bush.

The best "guess" I have seen based on an anyalsis at the precinct level was that Cubans went "about" 78/20 for Bush in 2000, while non-cuban hispanics went "about" 53/43 for Gore. (Note the use of "about".. these are educated guesses)

[ThePrezMex]

Thank you Volron. I think it was more clear this way.
In any case, because of what we just pointed out, the 'winter birds' as you call them and the military people, Florida is really really hard to forecast unless somebody has a clear lead.

Can we say that as of today, Kerry is like 1 or 2% above?

[The Vorlon]

Can we say that as of today, Kerry is like 1 or 2% above?

The light pink on my signature "map" indicates I "think" Kerry is up in the 0-2% range, but franky in the middle of August I don't trust state polls in general, and I especially don't trust a Summer time-Uni poll coming out of Florida.

The old refrain that "The polls don't count till labor day" is good advice anywhere, and especially good advice in Florida.

[The Vorlon]

Florida is really really hard to forecast unless somebody has a clear lead.

You are right on with that comment

For what it is worth, in 2000 just before election day, the Florida polls (at least the ones I still have on record) ranged from Bush +2 to Gore +13, with about 6 or 7 stops in between.

The 2000 Florida polls were a F$%king mess.

[TommyC1776]

hey at least it looks like Nader really isn't taking alot of votes from Kerry b/c either way Kerry's in first.