http://www.cbsnews.com/stories/2004/04/02/opinion/polls/main609944.shtml


In what must be the most inherently self contradictory poll I have seen in a while...

Nationally, Kerry is up 5 points, but in the 18 Battleground states Bush is up 2 points...

Delaware is a battleground state.. Even though Gore won by 13+% in 2000...?
Did something really big happen in Delaware that I missed...? (Perhaps a merger with Utah?)

For those of you have mastered math up to say, the grade 3 level, I will leave the anaylsis of this one up to you...

The "battleground states" that Bush leads by 2 points in  include Arizona, Arkansas, Delaware, Florida, Iowa, Maine, Michigan, Minnesota, Missouri, New Hampshire, Nevada, New Mexico, Ohio, Oregon, Pennsylvania, Washington, West Virginia, and Wisconsin..

These states make up almost exactly 40% of the nation's vote total, so in the other 60% of the nation Kerry must be up by just a hair under 10 points...

This poll was conducted among a nationwide random sample of 1,024 adults, interviewed by telephone March 30-April 1,2004. The sample included 834 registered voters. The error due to sampling could be plus or minus three percentage points for results based on the samples of registered voters and all adults. Results which may have been valid were surgically altered to resemble Saddam Huissen.  The use of Tarot cards was employed to weight data that had been deemed unreliable by the use of dice and a roulette wheel.  Next, as an additional confirmation, a double check was performed against "Big Fred's" famous 100% all beef foot long hotdog poll.  Lastly, a random number generator, hooked to a rube/goldberg device was used to normalize all data.

Vorlon,

Perhaps you should not make a new thread for every day's Rasmussen.

It does happen every day.

Perhaps you should just start one thread for Rasmussed every week?

Just a suggestion.

I second that

Done!!!! - I will just change the heading of the thread - (ie change the date) and update the first post...

Take this one with a grain of salt.  The result is not "nuts" but three points to keep in mind...

1)  It's a University poll...
2)  It's a University poll...
3)  It's a University poll...

you forgot the fourth point

4)It's a University poll

Thanks...

oh, ya, and it was done by a University....

Take this one with a grain of salt.  The result is not "nuts" but three points to keep in mind...

1)  It's a University poll...
2)  It's a University poll...
3)  It's a University poll...

[The Top Tier... Pretty darn good, most of the time...]

Terrance Group is a Republican firm, right?

Have they released results from any other states this year to compare?

Yes, Ed Goaes of Terrance group is the "semi" official chief GOP pollster.  But he is DAMN good...

The professional association of "public opinion researchers" (that is what "pollsters" call themselves) has awarded Goaes their "Pollster of the Year" award three times...

Goaes is as good as it gets.

He DOES make a bunch of weighting assumptions, but unlike most, he clearly spells out what they are, and fully explains why he made them...

Did Goaes do the Battleground poll in 2000?  I seem to remember that poll weighted Dems and Republicans equally in party ID and consistently ended up with pro-GOP results.  Greenberg's Democracy Corp weighted Dems three or four points higher in Party ID, and they were pretty much the only ones who got it right on election day.  There are still more Democrats than Republicans in this country!  (although some of those Dems in the South never actually vote for Democrats in federal elections)

Yes, Democracy Corps is a good firm - If you recall, I posted a list a while back of the "top tier" polling firms and they were on it...

I attached James Carville's name to the firm (he is a partner) just to tweak the Republicans by saying something nice about Carville...

The exact post was..

The Top Tier... Pretty darn good, most of the time...

<<list of other firms deleted to save space>>

Democracy Corps (Yes Republicans, James Carville's firm knows what they are doing.. marginal Dem bias)

........................

Actually, the 5 best results in 2000, in terms of accuracy were...

Actual = Gore + 0.51%

Harris Interactive = Predicted Gore + 0.7 => Error => .19%
CBS/NYTimes - Predicted Gore + 1 => Error = .49%
>>>>>>Historic footnote - Only time this poll has been within 8 points in 20 years!
FoxNews - Predicted Tie => Error = .51%
DDC/Yerxa - Predicted Bush +0.7 => Error = 1.21%
>>>>>Called 49/50 states correctly in the electoral college - they missed Wisconsin
Teeter/Hart - Predicted Bush +1 => Error = 1.51%

Battleground polll was HALF Goaes, the other half being a Dem firm (Snell/Lake/Perry) - Both good firms, but (apparently, if rumours are correct) a marriage made in hell...

Battleground weighted at 38/36/26 Dem/GOP/Ind in 2000.  They had Bush at +5 in 2000 (Waaaaay out on their turnout model - projected at 49%)

(There is a VERY interesting story about how turnout level projections  drove some very "strange" campaign behavior in 2000.... note to self to remember to post this at a later date...)

Believe me, Battleground is NOT The Terrance Group...

Ironically Both Terrance Group (Goaes) and Celinda Lake (Snell/Lake/Perry) released polls independently in 2000 that contradicted their "joint" Battleground poll.

I note that you've gone over to the FOX news channel coloring scheme, oddly.  You don't like the more traditional scheme used by Liep.  i.e., RED reminds me of communism, therefore red=democrats,  BLUE reminds me of patriotism, therefore blue=GOP.  Why FOX, of all groups, would have tried to popularize the reverse is a little bizarre.  Why you would choose to help them popularize it is a total mystery.

(It's my own April Fools gag... it flips back tomorrow... Hence the Quantum Reality tagline...)

Terrance Group is a Republican firm, right?

Have they released results from any other states this year to compare?

Yes, Ed Goaes of Terrance group is the "semi" official chief GOP pollster.  But he is DAMN good...

The professional association of "public opinion researchers" (that is what "pollsters" call themselves) has awarded Goaes their "Pollster of the Year" award three times...

Goaes is as good as it gets.

He DOES make a bunch of weighting assumptions, but unlike most, he clearly spells out what they are, and fully explains why he made them...

The good folks at Terrance Group AND Teeter/Hart are currently polling Ohio.  Terrance group should be releasing on Friday.

These are two really EXCELLENT firms... likely both in the top 3 or 4 firms anywhere.

The results should.. within the limits of polling, be about as good a snap shot as we're gonna get of this really key state...

The two firms use very similar methodologies, which should likely skew things maybe a point or two towards the GOP this far out, so I will personally in my own mind be taking a point or two off the Bush total (They use very tightly constained turnout models) but that is very minor nit-picking...

Stay tuned... will post when I get details..

Edwards is a heck of a campaigner, and I like his optimist and generally positiver campaign style.

He would make a lousy VP addition to the Kerry ticket for the following reasons..

1)  He adds no regions or states to the ticket.  Unless the Democrats suddenly nominate Robert E. Lee as the VP, the south is long gone...

2) The role of the VP is to attack...  People like Edwards because he was NOT a pitbull, turn him into a pitbull and folks don't like him any more.

I would expect that Edwards will pick one of the following:

A) A Rust Belt VP (help out in Ohio/Michigan/Pennsylvania/Michigan)

B) A Hispanic to take a shot at Nevada/New Mexico/Arizona

C) A midwesterner VP candidate to take a shot a Iowa/MIssouri/Minnesota/Michigan area.

I've made this point time and time again, so what is once more?  

Many of the seniors who voted for Gore in 2000 are dead!!!!

Many of the seniors who voted for Gore in 2000 were dead already.

LOL!

The Voter Fraud Enablement & Crooked Politician Protection Act Moter voter act is, on paper, a nice expression of the democratic sentiment that voting is good.

As a practical matter, I have seen voter turn out in certain areas that..... seems to defy historic turnout patterns....

Here is a stab at bi-partisanship...

Do any of the Democrats on this board think it would be an intrusive violation of somebody's voter rights to require voters registered via the Moter Voter act to produce the very drivers licence that got them registered..?

Or, (while we are on role...) have everybody who shows up to vote present maybe 1 piece of picture ID...?

And if the person has no ID, let any other voter with ID "vouch" for them by signing a declaration that they know the person trying to vote and certify they are who they say they are...?

As a historic note, the Democrats in the Senate have fillibustered these measures again, and again, and again...

Why?

I am generally fairly non-partisan here on these boards, but voter fraud IS an issue, and I don't think any of the stuff above is draconian in any way...

Comments...?

(Yes Dems, you have a legitimate issue on "voter disenfranchinesment" - but that's a DIFFERENT legitimate issue than the voter fraud I am talking about here)

The friday night sample was essentially a tie - Kerry won 44.9 to 44.4

Bush is up 1.4% on the 3 day roll, down 0.1 on the 7 day.

None of these numbers has any statistical meaning.

Rasmussen has had some strange samples past couple days, so this one is going to bounce around a bit the next few days... It's noise... don't pay any attention...

                I Day         3 day        7 day  
              Kerry/Bush  Kerry/Bush   Kerry/Bush
28-Mar   43.4   45.3   45.0   45.5   46.2   44.7
29-Mar   47.6   43.1   46.2   44.5   46.4   44.8
30-Mar   49.1   45.1   46.7   44.5   46.7   44.5
31-Mar   41.9   45.3   46.2   44.5   46.5   44.5
1-Apr   41.6   43.7   44.2   44.7   46.0   44.7
2-Apr   47.9   50.5   43.8   46.5   45.5   45.0
3-Apr   44.9   44.4   44.8   46.2   45.3   45.2

The most likely pickup for Bush is probably Pennsylvania...

Ah Gustaf (spelled correctly !) - You read my thread on PA, and like magic a few days later a poll with Bush +6 magically appears..

And you are convinced that PA is in play......

You seem to believe Florida will change hands..

You are mistaken young Jedi...

I will soon enlighten you to the ways of the force as it fortells Florida...

(but not now... have work to do.....)

Md went dem with southern candidate (Gore, Clinton, Carter) but went rep with guys like Dukakis or Mondale (so Kerry can loose)

Mason Dixon had Kerry up 11 points.

You'll make a lot more money betting on Mason-Dixon than you will  on "Gonzales Marketing & Research"

Yeah, the name of that firm who did the poll was a little odd to me. I had never heard of them before.

Gonzales does a lot of polling in Maryland.  I've usually seen them refered to as "Gonzales-Arscott"; I'm not sure if they have changed their name.

Actually, the principle of the firm used to be the President of Mason-Dixon...

An inside joke.. that nobody got....

As comedians say... if you need to explain the joke, get another joke....

You never did respond to repeated requests for elaboration about your feelings on uni polls.

Why UNI polls have... issues.... would require  the writing of a book...

I will start the book here...

Lets follow the path of a poll to get it right...

Step 1; - Your list of Phone Numbers...

You cannot simply "randomly" pick phone numbers...

Rich people have more phone lines than poor people..
Families with teenagers have more phone lines
Young people have more cell phones than old people
Young people have more land lines than the old
Many people have a cell and an land line..

etc... etc...

To get a list of phone numbers that actually allows you to CONTACT a truly RANDOM sample of the population you need to do a huge amount of cross correlation of census tracks, phone directory information, etc...

This is far too complex a matter to get into here, but I'll make one point..

The generation of a truly random contact list is such a huge and complex matter that not even Gallup does it in house - Gallup buys their phone lists (as do most top-tier firms) from a company called Survey Sample Inc, who does nothing but sell these lists..

If Gallup (which has as many ph'ds as a lot of Universities) has concluded that not even they can do this VITAL step in house, what do you think the chances that one marketing professor and two grad students are going to get it right?

Likely voters?

Most of these Universities have never come within 50 miles of an actual election poll under "combat" conditions.

People lie when they are polled.  Gallup uses 13 questyions to sort the "likely" from the "unlikely" most Universities use 2 or 3...

To pick on Keystone (since it is the topic of this thread) They are projecting that 35% of the entire sample of 565 registered voters meets their criteria as a likely GOP Primary voter..  this means that about +/- 70% of all registered GOP voters will vote in the primary.. are they on Drugs...?

in 2000, in a hotly contest Presidential race, plus a hotly contested Senate race on top of it, only 63% of the registered voters turned up to vote..

70% GOP turnout in a lukewarm Senate Primary...?

Give your head a shake...

Topics to still be covered....

Who are your operators?
Callbacks?
Is the ph'd in Math or Marketing?

TO BE CONTINUED.....

Md went dem with southern candidate (Gore, Clinton, Carter) but went rep with guys like Dukakis or Mondale (so Kerry can loose)

Mason Dixon had Kerry up 11 points.

You'll make a lot more money betting on Mason-Dixon than you will  on "Gonzales Marketing & Research"

I like a man that doesn't need a partner for a debate.  

With ONE fact you can make an easy choice,..

With TEN facts you can make the hard choice,...

With ALL the facts you can make the right choice,...

Ayn Rand

Here is the link...

http://pdn.philly.com/2004/03/31/keyzpoll.pdf

Two points....

1:  My thread of a week ago asking if PA was "in play" looks a lot more rational than it did... (my March 27th prediction map reflects PA leaning Bush, as you will all note....)

2: Don't pay all that much attention to this poll (as he blatently contradicts point 1)

Firstly, this is a University/College poll.  I don't know all that much about the Keystone Poll, but in an act of blatent "pollster profiling" unless I have crushing evidence to the contrary, I generally ignore College/University polls.

Secondly, the poll found a 47/44 GOP voter identification/registration advantage.

Pennsylvania is actually basically even in GOP/Dem voter registration, so this mini-blip in GOP registration advantage is likely due to the semi-hot GOP Senate primary race which is likely juicing up GOP registrations a bit, so I'd trim Bush's 6 point "lead" back 3 points or so in my own mind.

(That being said the same poll found a 5% GOP registration advantage in February so the actual drop in Kerry support may have been a bit bigger than this poll suggests, however...)

Another question I have about this polls is that all the data tables are headed with the prefix "Weighted Data" but no where does the text of the poll explain their weightings, which renders it very hard to make any qualitative evaluation on this poll at all...

Lastly, and perhaps most importantly, this poll was designed to measure the GOP senate race, not the November Presidential race...

As my woodworking friends keep telling me.. "The right tool for the right job...."

In 2000 I followed the Battleground polling at their website. This year it seems to have stopped with their results from last September. Have they gone to private polling only?

Muon2...

Didn't I see you violating matter/antimatter symmetry (or was that your evil twin Kaon..?) in another thread...?

Battleground is a joint venture between Terrance Group  and Snell/Lake/Perry (two excellent firms)

Terrance Group is polling the $%^# out of Ohio right now, so they are a tad busy..

I have no idea what the set confidence value is for a political poll, 95%?

95% = two standard deviations from the norm, in an evenly divided bi-nomial expansion.

They picked it 'cause its a place where the math is easy..

[The Top Tier... Pretty darn good, most of the time...]

ALWAYS keeping in mind the margin of error, which firms can we trust...?

This is my personal rankings...

(Firms are NOT in any order within their categories)

The "Bias" I refer to is NOT ideological, but Methodological...

The Top Tier... Pretty darn good, most of the time...

Gallup (the real one, not the CNN/USAToday/Tracking version)
Battleground (Goaes/Lake/Perry) (mild GOP bias, maybe 1-2 points)
Tetter/Hart (Wall Street Journal/NBC News) (maybe a marginal GOP bias)
Mason Dixon - Damn good firm...
Fox News (Yes, liberals, check their historic record...)
Democracy Corps (Yes Republicans, James Carville's firm knows what they are doing.. marginal Dem bias)
TIPP/Infometrica (Investors Business Daily)
Snell/Perry/Lake
Yerxa/DDC
AP/Ipsos Reed
Harris

The Second Tier - Interesting, not worthless, but don't bet the farm....

Rasmussen (Misses the top tier, but not by much...still on "probation" after their 2000 presidential call)
ABC News (Strange weighting, but knocking on the top tier door)
Survey USA
ARG
Research 2000
PEW - very consistent, but has a structural 4 point Democratic bias.
Quinnipiac - Only University poll (I am familiar with) that's NOT totally $%$ing useless...

Third Rate...

CBS/NYTimes..

ok..they got LUCKY in 2000, but from 1980 to 1996 they AVERAGED missing the actually result by 2 and a half times the margin of error of their poll! - If they do ok this year, I might believe them.... maybe...

Newsweek..

(Historically waaaaay off the mark)

Time

(see Newsweek)

All off on his Own..

Zogby gets his own Category.  He is brilliant, and erratic.  

For the record, in 2002, he polled 17 races, he got 12 right, 5 wrong...  

Zogby is the "Drudge Report" of pollsters... he's often utterly wrong, but right just enough to keep you reading...

What do you think about Ohio?

The Terrance Group (Ed Goaes/Battleground) and Teeter/Hart are both polling Ohio right now.

Both are absolutely first rate firms (Terrance group uses a turnout model that slightly favors the GOP, IMHO,  but thats  more a nit-pick than anything else)

By friday we should have a pretty darn good idea of what Ohio looks like.

The CNN/USAToday/Gallup poll of March 8 was all over the CNN site.

http://www.cnn.com/2004/ALLPOLITICS/03/08/elect04.prez.poll/index.html

For some reason CNN has not posted anywhere todays poll.

Why isn't CNN publishing their own poll..?

Does anybody have a link to the "guts" of today's CNN/USAToday/Gallup poll...?

How is a popular former governor a marginal candidate?

Ok... let's say Shaheen was a GOOD candidate.... the argument still stands...

Re West Virginia, ok a bit of a brain cramp there..

Was thinking Virginia, not WV geographically..., but the rest of the argument is valid...

[West Virginia]

Vorlon, why do you think the dems will win NH and WV? And why does FL lean GOP?

Keep up the good analyzing, it's greatly interesting to read it.

West Virginia

WV is a naturally Democratic state -Heck Dukakis even carried West Virginia.

Also, in 2000, Gore was, to put it mildly, a total and complete idiot when it came to West Virginia.  I think the Gore campaign book was entitled "Coal is bad, lets throw all the West Virginia coal miners out of work and then publically shame them for contributing to global warming" (ok.. that's a mild paraphrase, but I got the spirit right..

I still have West Virginia close, but I have to believe that in a real close race it goes for Kerry.

New Hampshire

Has been trending Democrat for a while.  Bush only won by 1.5% (or so) in 2000.

In 2002 a GOOD GOP senate canidate (Sununu) only won 51/47 against a marginal Democratic candidate on a strong GOP election night

New Hampshire is in Kerry's back yard.

New Hampshire has been hit quite hard by high tech out sourcing

Bush got pounded for 4 months during the Democratic Primary.

New hampshire goes for Kerry, write it down..

Florida

I just do not buy this "angry democrat" argument.

Firstly, Donna Brazille (Gore's 2000 campaign manager) did an AMAZING job at voter turnout in 2000 in Florida.  

Sometimes you just have to take your hat off and say "well done!"  Donna.."Well done!"

On the downside, I just don't think there is any more Democratic turnout to be had.  Put bluntly, Kerry's people an 2004 are just  not as good as Brazille is at get out the vote

Secondly, what about 2002..?

Were these angry democrats also not going to sweep Brother jeb from the Governors office?

Granted, a Governors race is not a Presidential race, and Jeb is a better Governor than George is a President, but Jeb won by a semi-landslide of 13%..

You just can't ignore 13%+ GOP margin in a governors race.

Finally, there is the Seniors issue.  Bush actually won Florida seniors 50/46 in 2000.

Older, government dependant seniors (say 70+) break more for the Dems than any other group - this group is also the group that is dying off the fastest.

Younger, affluent seniors (60-70) break modestly for the GOP, and their relative percentage in Florida is growing as it becomes more and more a retirement location for affluent, GOP leaning seniors.

The Kerry strategy seems to be targeting Ohio rather than Florida.  Gore decided to go toe to toe, dollar for dollar with Bush in 2000, and if essentially bankrupted his campaign in a bunch of other states - Arkansas, Tennesse, New Hampshire, Ohio, Missouri - I think Kerry realizes that even a few of the many millions spent in Florida might have won any one of a number of other states for the Democrats.

I would not be at all surprised if Bush ends up carrying Florida by maybe +5 or 6 actually...

Pardon my "french" here, but the notion that somehow we should ban foreigners from political discussion here is about as intollerant a ^&&^%%^$%$ing idea as I have ever heard.

Last time a checked the First Amendment, and a few other ideas I generally am rather fond of, held to the believe that an open discussion of all the fact led to a better democracy.

Ok.. call me a crazy Libertarian....

An argument stands or falls on the merits of the argument.

 Gustaff (an excellent member of this community! - Go Gustaff!!)  and Seige make their  points, sometimes I agree, sometimes I do not, but regardless of the argument coming from Sweden, Latvia, or Mars, the argument carries the weight it deserves.. origin is irrelavant.

I rather enjoy the European perspective actually, it serves to remind me how important Jefferson's warning about limiting the size of government were!