Newsweek Poll: Tied

[California Dreamer]

Head to Head w/o Nader (registered voters)
Bush: 48
Kerry: 48

Head to Head w/Nader (registered voters)
Bush: 45
Kerry: 43
Nader: 5

M/E: 4%

Favorable /Unfavorable Approval Ratings:
Bush: 51/36
Kerry: 52/42

M/E: 3%

I am surprised after the bad week that Kerry has had and with terrorism in Spain that Bush didnt pull ahead...but there might be a bit of a delay factor.

I kinda think that between now and the conventions that the lead will swing back and forth.

Newsweek Article

[California Dreamer]

Head to Head w/o Nader (registered voters)
Bush: 48
Kerry: 48

Head to Head w/Nader (registered voters)
Bush: 45
Kerry: 43
Nader: 5

M/E: 4%

Favorable /Unfavorable Approval Ratings:
Bush: 51/36
Kerry: 52/42

M/E: 3%

I am surprised after the bad week that Kerry has had and with terrorism in Spain that Bush didnt pull ahead...but there might be a bit of a delay factor.

I kinda think that between now and the conventions that the lead will swing back and forth.

Newsweek Article

A few technical comments...

This poll was of 1006 ADULTS voters (as opposed to registered voters).  Typically a GOP candidate does 2 or 3 points better with "Registered" voters, and another 2 or 3better with "likely" voters (at least close to the actual election day)

I don't particularly quibble about not sampling down to"likely" voters 8 months out, since the meaning of "likely" here in March is a bit unclear.. at least to me..,

but not even limiting you sample to "registered" voters is damn near polling malpractice...

Doesn't ANYBODY know how to do a %^&^#$!!@#ing poll any more? (he mutters under his breath)

Newsweek did not poll close to the 2000 election, their last poll (about 7 weeks out) showed results very much at odds with the final outcome and other polls taken concurrently..

9/17/00
Newsweek Poll done nationally
Al Gore
54%
George W. Bush
41%

(actual outcome.. Gore + .51%)

dont you know how to read?

I posted seperate MofE because the Head to Head was registered voters (hence the higher MofE), but the approval ratings were all adults.

Also I am not sure that the 'likely voters' rules apply in this election because as the NPR poll the dems seem especially motivated this season.

[California Dreamer]

you need to read the fine print

You must be logged in to read this quote.

[California Dreamer]

dude...you need to relax. Do you think I am making this up?

try this link

if that doesnt work then take the online poll, after that it shows you the raw data of the NW poll (the fine print is under that)

[California Dreamer]

as I quoted from the poll, NW used all adults for general questions (approval ratings, stuff about spain, etc) and registered voters for voting questions ("if the election were held today...".

it's quite simple,

[California Dreamer]

I agree entirely..the MOE was 4% (due to the fact that that some questions were based on registered voters)...as stated in the raw data in the link I posted.

It is reasonable to assume that the questions that used the subset of registered voters were the questions relating to voting.

This was all in the link I provided, but if you cant be bothered to click the link Here is the raw data...look closely at the bottom

You must be logged in to read this quote.

Since I work in marketing research I know a thing or two about polling as well, but in this case the proof that I am right is there in black and white, no room for interpretation

The 'tie' between Bush and Kerry is Registered voters and not 'all adults'

All Adults were for the other questions.

[California Dreamer]

you just dont know when to give up

the raw data clearly states how it uses registered voters for some questions (obviously the voting ones) and that those questions have 4%

the 'overall' poll and those questions where everyone is intereviews is 3%

Read the data...go to the link, etc etc

If you look this poll has many questions not related to voting (such as the issues with Spain). In essense its two polls, but in the case of voting they used registered voters as the data says.

you got a problem with how they footnote things...contact Newsweek.