Battleground Fall Track Poll (Great stuff!)

THE TARRANCE GROUP, INC. and LAKE SNELL PERRY ^

[narses]

Battleground Fall Track (Week 7)
FINAL
______________________________________________________________________________
STUDY #9936
THE TARRANCE GROUP, INC. and LAKE SNELL PERRY
N = 250 per day of registered “likely” voters
Field Dates: October 25 –28, 2004
Hello, I’m _______________ of The Tarrance Group, a national survey research firm. We’re talking to people long
distance today about public leaders and issues facing us all. May I please speak with the youngest male in the
household who is registered to vote in this state?
A. Are you registered to vote in this state?
IF “NO”, ASK: Is there someone else at home who is registered to vote in this state?
(IF “YES,” THEN ASK: MAY I SPEAK WITH HIM/HER?)
Yes (CONTINUE)
No (THANK AND TERMINATE)
Now, thinking ahead to the elections that will be held this November --
B. What is the likelihood of your voting in this upcoming election -- are you extremely likely, very likely,
somewhat likely, or not very likely at all to vote?
Extremely likely...........................................82%
(CONTINUE) Very likely ....................................................17%
Somewhat likely ...........................................1%
(THANK AND TERMINATE) Not very likely
UNSURE (DNR)
C. Are you, or is anyone in your household, employed with an advertising agency, newspaper, television or radio
station, or political campaign?
Yes (THANK AND TERMINATE)
No (CONTINUE)
* = Less than .5%

[sitetest]

Dear narses,

I googled, and this poll had Bush 50% and Gore 45% in the closing days of the 2000 election. These folks don't have the most sterling reputation for accuracy. I will note that they did much better in 1996.


sitetest

[mwl1]

FOX said this morning they will release new Opinion Dynamics poll at 2 pm today... any guesses?

[Greek]

Suspect, because moderate numbers are usually closer to 25%

[zeebee]

The poll was only directed at the youngest male in the houshold who is registered to vote. It is not a representative cross section.

[Tribune7]

That is a good poll.

[JustDoItAlways]

This poll is for "the youngest male in the household who is registered to vote." The demographics show the age was spread pretty evenly so it is really just a male poll.

I would expect Bush to be leading 46, 41 in this demographic so it is not spectacular.

[Tribune7]

It's 6 percent Latino (4 percent in the first question and 2 percent in the follow up)

[deport]

This poll is done by a joint effort between a major GOP firm and a major DEM firm. That makes it the most respected poll I know. Plus, they publish their questions and answers for all to see.

---

They blew 2000 when they had Bush up by 5% over Gore.

[Hootch]

Could you even begin to imagine the backlash if they did something like that?

lol

[narses]

Polls are not predictions, much as we'd like them to be. This poll discloses what was asked, who was asked, the methodology used in asking and the results. That allows people like us to analyze more than just the horse race. You can use aggregators - http://www.electoral-vote.com/ or http://www.realclearpolitics.com/bush_vs_kerry.html or http://synapse.princeton.edu/~sam/pollcalc.html but they tend to have their biases too. We will know soon enough. I think this could be a huge blowout but I'll settle for a repeat of 2000 if I must.

[pandemoniumreigns]

This is purely anecdotal, but at a ballgame in Neyland stadium last week I was taking note of Bush vs. Kerry bumber stickers and then badges worn by the fans.
I saw exactly 3 Kerry badges vs. hundreds of Bush/Cheney badges.
And these were not all just middle aged guys like me
Lots and lots of the kids were Bush, even the vendors.
As a matter of fact, all the Kerry badges were worn by the older crowd.

[sitetest]

Dear narses,

I'm not suggesting that this poll serves as a prediction. In fact, I'm suggesting that it may not serve well as a prediction at all. Nonetheless, we assign some predictive value to polls, especially to polls taken just prior to the actual event - the election, otherwise, we likely would ignore them.

We see "50 - 45," and we're heartened. To the degree that we take heart from this poll, we are considering it to be at least partly a prediction. Else, who cares about the "50 - 45"?

Some polls have better predictive value than others. This poll, in the last election, had poor predictive value.


sitetest

[deport]

I agree..... I think this will be a big win for the President..... But BG's prediction wasn't that good in 2000 when others got it dead on. So it becomes who's on this time around and who's off. Therefore, I tend to not follow a poll for specific predictions but rather look at it for trends......

[narses]

Sure and my thought is that the data presnted is much more interesting than just the BIG conclusion. I haven't seen any of the other polls release this much data. The truth will be known next week.

[Scott from the Left Coast]

It is rumored they will charge Bush with having sex with his Laura before they were married.

ROFL! That'd be a great one...Bush would GAIN votes!

[sitetest]

Dear narses,

To your credit, you have emphasized the "internals," although I don't think they quite support the conclusion of a "blowout." As well, I'd want to compare their internals from last year's final poll, as well. I don't have them handy, and don't feel like looking for them.

I'm hopeful of a modest, meaningful win. 51% - 47%? 52% - 46%? 300 EVs? 325 EVs? I'd be happy. I'm really doubtful of much more than that, and wouldn't be at all surprised by a closer race than that.


sitetest

[narses]

"300 EVs? 325 EVs? I'd be happy."

Me too. Very happy. Especially if Thune wins as well.

[Scott from the Left Coast]

I think that's a very accurate picture of what the population thinks of themselves. It's, obviously, not an accurate picture of reality -- but quite in line with the fantasy fog many people live in.

Let me give you a quick annecdotal example that's a bit of fun: My dear 87-year old mother considers herself an independent and would call herself conservative. She thinks this because once way back in 1972 she voted for Nixon. The only time in her life she EVER voted for a Republican! She thinks she is a conservative because she opposes abortion, though she has never voted for anyone who opposed abortion themselves!!! Economically, she's a radical unionist, somewhere right about on the mark with Vladimir Lenin. Yet, she would answer this question as being "somewhat conservative". She thinks Bill Clinton was "very conservative"!

[infohawk]

While we have no proof, the 2000 final battleground poll was probably taken before the DUI story broke on the final Friday before the election. Also, the MSM calling the state early for Gore probably dissuaded people from voting in the panhandle. These folks are strong for Bush. One more thing. Bush made a mistake by not campaigning through the final weekend. He became overconfident. Gore spent time in Florida over the final two or three days, which certainly helped to rally his base.

[sitetest]

Dear infohawk,

I don't disagree with you. There may be very legitimate reasons why the Battleground poll was off on election day.

But it was off. By a lot.

Frankly though, I don't thing the DUI story moved the polls more than a point our two, and I don't think the lamestream media early call had much effect outside of Florida.

The thing is, that small movement added up to being magnified in proportion.

I think without those two items, Mr. Bush would have won the popular vote by perhaps 1% or 2%, and taken Florida relatively easily, as well as New Mexico, Wisconsin, and one or two other states. Instead of the fiasco we had, we'd have had a modest but discernible win in the popular vote, as well as the electoral vote. And none of this "selected, not elected" crap.

That still leaves the final 2000 Battleground poll as a mediocre also-ran.


sitetest