MyDD :: Newsweek Poll

Newsweek Poll

by Chris Bowers, Sat Oct 02, 2004 at 06:27:13 PM EST

Some positive news from Newsweek (9/30-10/2, MoE 4 for the entire sample, all 9/30 interviews conducted post-debate):

Two Way, Registered Voter Overall Trend
      Bush  Kerry  Und.
10/2  46     49     5
9/10  50     45     5

Trend Among Republicans
      Bush  Kerry  O/U
10/2  89     6	  5
9/10  93     4	  3

Trend Among Democrats
      Bush  Kerry  O/U
10/2  12     86     2
9/10   7     87     6

Trend Among Independents
      Bush  Kerry  O/U
10/2  37     42    21
9/10  39     45    16

There was virtually no movement among the Party ID subsets, especially considering that the subsets have margins of error of plus or minus 6, 6, and 7 respectively. The difference is that the overall sample is composed 36D, 34R and 30 I/O now, compared to 35R, 32D and 33 I/O on September 10th. My opinion on Party ID has changed slightly, to the point where I now think that the three-way question is such an inaccurate measurement of Party ID that it should be discarded from polling (read this post for more). Still, this will clearly dent the "Kerry won, but he didn't move the needle" spin the GOP is trying to push.

Kerry's favorable ratio is +12, while Bush's is +3 (Cheney is at -3 and Edwards is at +19). Right Track / Wrong Direction is at -21. Bush's job approval is at 46 approve and 48 disapprove. All of these numbers have favorable trends for Kerry / Edwards.

Kerry only trails Bush 39-33 among Republicans in the "who won the debate" question. The holding Rove's water push poll award goes to question number 24:

24. Over his 19-year career in the U.S. Senate, John Kerry has changed his position on a number of issues. From what you know about Kerry, do you think this is MORE because ... (ROTATE ANSWER CATEGORIES 1-2)

WTF?!

Tags: General 2008 (all tags)

Permalink :: 8 Comments :: Post a Comment

The meme changes... (none / 0)

with this poll and it shifts the pressure to Bush. Now we need to hope that a "bandwagon effect" begins to build and the strong performance by Kerry on Thursday is repeated by Edwards then Kerry on Friday. At that point I think we will have solidified the perceptions and the rest will be clean-up and turnout.

by sandiegosteve on Sat Oct 02, 2004 at 07:04:18 PM EST

WTF!!! (none / 0)

How the hell though did Bush Gain Dems support? Kerry dropped a point and Bush gained several Dem undecided. How the hell did that happen?

by Michael on Sat Oct 02, 2004 at 09:25:34 PM EST

Unfortunately (none / 0)

What this poll actually shows is that Kerry has LOST GROUND, not gained ground.

I'm using Chris's estimated Party IDs of 40%-D, 37%-R, 23%-I (outlined in this post of Chris's from a couple of days ago).

If we weight by those Party ID's, the old (9/10) Newsweek poll had Kerry winning by 46.6% - 46.2%. The new (10/2) Newsweek poll has Kerry winning by only 46.3% to 46.2%.

Yes, I realize that Chris's opinion on Party ID "has changed slightly". But I'm still going by what Chris said a couple of days ago: Party ID will, on election day, end up to be about 40/37/23. I think that's as good a guess as any, and I intend to continue to use it in weighting future polls, even if it means that polls that normally would seem to look good for us (like this Newsweek poll) turn out to be not so good after all. I think it's better to have as accurate an idea as possible as to what's going on, and not to continually change my mind about Party ID depending on whether the result helps us or hurts us.

by BigModerate on Sat Oct 02, 2004 at 11:41:45 PM EST

That doesn't make sense (none / 0)

Kerry went from -5 to +3, a net gain of 8. The representation of Dems, relative to Reps, went from -3 to +2, a gain of 5. So the change in party ID can explain part of not all of the shift to Kerry.

Further reweighting the data based on support as of 9/10 and 10/2 among Dems, Reps and Indys gives a reasonable assessment of current support, but shouldn't be used to infer trends. Candidate support by party ID is based on a smaller, less reliable subset of interviews, and I believe that the 9/10 Newsweek poll shows lower Bush support among Dems than other September polls.

by EvanstonDem on Sun Oct 03, 2004 at 01:09:28 AM EST
[ Parent ]

Re: That doesn't make sense (none / 0)

The representation of Dems, relative to Reps, went from -3 to +2, a gain of 5.

That's incorrect. The representation of Dems, relative to Reps, went from -10 to +2, a gain of 12. (Note that Chris was in error when he he wrote that the Party IDs on 9/10 were "35R, 32D and 33 I/O"; in fact, the Party IDs were 41R, 31D, 28I.)

by BigModerate on Sun Oct 03, 2004 at 10:02:57 PM EST
[ Parent ]

despite the weighted question (none / 0)

It's a positive result--less than a majority believe that Kerry switches positions because he's a politician and it'll help him get elected?! Hell, even I believe that.

by asf6 on Sun Oct 03, 2004 at 12:42:46 AM EST

Questionable Kerry question (none / 0)

It's a leading question and should not have been asked. Or, out of fairness, they should also have asked the following:

"As President, George Bush has made a number of factually incorrect statements. From what you know about Bush, do you think this is MORE because he was lying in order to get re-elected, or becuase he was uninformed?"

by EvanstonDem on Sun Oct 03, 2004 at 01:13:39 AM EST

Newsweek's push poll (none / 0)

Where did they get question 24 from? That does sound like it comes directly from Karl Rove. Did Newsweek lose their mind?!?!

by mishiem on Sun Oct 03, 2004 at 09:02:11 AM EST

You are not logged in.

In order to post a comment, you must be logged in. If you have a member account, please log in to comment.

If not, you can make an account right here. It's quick and free.