Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

A look at Iowa polls since the Pew data was collected starting Nov. 7:

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 11:04 AM
Original message
A look at Iowa polls since the Pew data was collected starting Nov. 7:
Edited on Mon Dec-03-07 11:09 AM by Stop Cornyn
Hillary (oldest date for beginning of data collection at top to newest at bottom):

31% (11/07) - AP/Pew
30% (11/14) - ABC/WaPo
29% (11/23) - SV
25% (11/25) - DMR
25% (11/26) - ARG
24% (11/27) - R


Obama:

26% (11/07) - AP/Pew
26% (11/14) - ABC/WaPo
29% (11/23) - SV
28% (11/25) - DMR
27% (11/26) - ARG
25% (11/27) - R


Edwards:

19% (11/07) - AP/Pew
22% (11/14) - ABC/WaPo
23% (11/23) - SV
23% (11/25) - DMR
23% (11/26) - ARG
24% (11/27) - R

Printer Friendly | Permalink |  | Top
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 11:20 AM
Response to Original message
1. So assuming (without foundation) equal polling rates
the average would be

HRC 27.3

BO 26.8

JE 22.3


So the poll still shows a shift from this averaged data.

It's all MOE stuff though. There is no discernible trend of any kind favoring either of them.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 11:31 AM
Response to Reply #1
2. My point was that the Pew data in the report released today is older than data in at least 5 polls
which were released earlier than the Pew poll. I'm questioning whether it is appropriate to suggest that Hillary's free fall has stopped based on data that was gathered nearer the beginning of her decline in the Iowa polls.
Printer Friendly | Permalink |  | Top
 
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 11:39 AM
Response to Reply #2
3. Well in all honesty you can't say that
without knowing the rate of gathering the poll responses. It's a BIT fairer to do it the way I tried to, but again not definitive either way. Only if the poll were massively front loaded would it say anything at all about any putative slide.

MY point is there is no valid reason to assume a slide ever existed.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 11:50 AM
Response to Reply #3
4. Not really. Only the AP/Pew poll is an outlier with data gathered over 18 days. ABC/WaPo gathered
data over 4 days, SV gathered over 2 days, DMR for 3 days, ARG for 3 days, and R over a day.

Other than the Pew poll, the window for data collection was appropriately narrow.
Printer Friendly | Permalink |  | Top
 
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:04 PM
Response to Reply #4
5. Not the point however
Since none of these "trends" showed movement outside MOE, and since a poll taken over the same time period - with an unknown rate of responses - shows results that are more in line with earlier data and in fact somehwat further away from the suggested "trends", there really is absolutely nothing we can honestly conclude about any movement in IA other than that it si close and fungible.

When you have a 3-5pt MOE, one candidate going up 2 or 3 and the other down 2 or 3 is not a trend. It's not even definitively a change.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:30 PM
Response to Reply #5
6. Are you sure Hillary's 7% drop from the Pew poll to the Rasmussen poll is within the MoE?
Edited on Mon Dec-03-07 12:30 PM by Stop Cornyn
Printer Friendly | Permalink |  | Top
 
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:53 PM
Response to Reply #6
11. Well the issue is a bit iffy between different polls but yes really
AP/Pew has a 5.5 MOE so Clinton could be anywhere down to 25.5 on that one 95% of the time and still have it a valid poll.

I can't find a Rasmussen poll which has a 7pt difference (she has 27% in the latest I can see) but it's very possible I missed one so if she had 24% on any, their tracking polls have a 3.5% MOE IIRC so she could be anywhere up to 27.5. With the 27% I see, she could be anywhere up to 30.5 of course and still be a valid poll

The Des Moines Register poll which shows Obama leading has a 4.4 MOE so Clinton could be anywhere up to 29.4 on that one 95% of the time and still have that a valid poll.

MOEs of course can go both ways. It's also very possible Obama has a somehwat comfortable lead, and has been possible for quite a while before he "took the lead". in released polls.

No matter which way you slice it, there is neither a clear front runner nor a clear trend in IA. I say that not to support, or denigrate, any candidate.


Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 01:02 PM
Response to Reply #11
12. MoE is based on sample size so when you attempt to aggregate multiple polls that enhances certainty.
Before we get onto a more detailed discussion about how the cross-checking of one poll with another poll taken contemporaneously affects the confidence intervals, let's not lose sign of the fact that these are polls which are a fair forecast of a primary result but which are less accurate at forecasting the results of a caucus.

My point was that some people are offering "new" polls as evidence of trends and those "new" polls aren't new for the porpose of spotting or refuting trends if they are based on data which is older than the "old" poll.

If you line the polls up according to the dates of the data collection, there is no reason to conclude that Hillary has "turned it around" or "Obama has peaked." It shows a decline for Hillary and a rise for Edwards and a Thanksgiving bump for Obama.
Printer Friendly | Permalink |  | Top
 
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 01:24 PM
Response to Reply #12
13. Only if all the polls are equal and valid but mostly agree
I don't think it DOES show a decline for Hillary OR that she has turned it around. I think it shows a race with no clear leader and lots of churn.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 02:38 PM
Response to Reply #13
15. Maybe you don't see a decline for Hillary, but apparently her campaign does:
Romney, Clinton shake up tactics:

CEDAR RAPIDS, IOWA -- -- Facing fresh polls showing their leads in Iowa disappearing, Democratic presidential candidate Hillary Rodham Clinton and Republican Mitt Romney rolled out new campaign tactics Sunday in an aggressive push to regain lost momentum.

Sen. Clinton of New York, who until recently would not even mention her rivals by name, used a news conference to question the ethics, character and "courage" of Illinois Sen. Barack Obama, employing the most pointed language she has used in the campaign.... The steps reflect the rapidly shifting dynamics in Iowa, whose caucus is set for Jan. 3. For months, both Clinton and Romney held strong leads in Iowa. But a Des Moines Register poll published Sunday showed that their advantage had collapsed. ... Clinton was in second place behind Obama, 25% to 28%.

Both Romney and Clinton would be shaken by a loss in this crucial state. Clinton has cast herself as the inevitable nominee, and a defeat here would shatter perceptions that she can't be stopped.... That Romney and Clinton would shake up the playbook with the caucus just a month away underscores the worry in both camps.

"It's really without precedent," said Gordon Fischer, a former chairman of the Iowa Democratic Party. Fischer, who is backing Obama, said: "They're both very, very concerned. They were the front-runners who are no longer front-runners."
Printer Friendly | Permalink |  | Top
 
Auntie Bush Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:02 PM
Response to Reply #1
19. Edwards is going up the fastest.
Hillary is going down the fastest.
Obama is the most steady.
Not going up or down much.
Printer Friendly | Permalink |  | Top
 
Tejanocrat Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 04:30 PM
Response to Reply #19
26. That's how it looks to me.
Printer Friendly | Permalink |  | Top
 
antiimperialist Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:32 PM
Response to Original message
7. The AP/Pew poll useless because it's very, very old
Several debates and events have taken place since Nov. 7 that render this poll useless. Even the voters who told the pollster their opinion between say, Nov.7-Nov. 15- could have changed their minds by the time the poll was over on the 25th.

This poll is garbage due to its useless time-frame.
Printer Friendly | Permalink |  | Top
 
Kurt_and_Hunter Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:34 PM
Response to Reply #7
8. You are correct. The AP poll was taken during the worst of Hillary's bad press.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:35 PM
Response to Reply #8
9. But the later polls show her support erroding further so I guess she's still in free fall mode.
Printer Friendly | Permalink |  | Top
 
antiimperialist Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 12:39 PM
Response to Reply #8
10. I disagree with you
I'd say Hillary's bad press stopped the day of the debate on Nov. 15th, when the media decided that gutsy Hillary kiced ass after "fighting back" against the rest of the pack.

Therefore more than half of this poll was conducted after the "bad press" against Hillary, who deserved it by the way.
Printer Friendly | Permalink |  | Top
 
sundancekid Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 01:32 PM
Response to Reply #10
14. couldn't disagree with you more - IMHO, your several posts above
in the aggregate show your bias interest rather than simple factual, head-on analysis ... what part of the Pew Polls are NOT, NOT, NOT, NOT, NOT, NOT the "newest" polls? Period. End of Question. And all other red-herrings are just that.
Printer Friendly | Permalink |  | Top
 
antiimperialist Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:39 PM
Response to Reply #14
23. you are in denial. hillary has lost ground
and the ap poll is so old that we cant even be sure that those surveyed on the first days had not switched candidates by Nov. 25th. the time frame is just useless.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 05:13 PM
Response to Reply #23
28. It's hard to spin the data any other way.
Printer Friendly | Permalink |  | Top
 
antiimperialist Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 09:11 PM
Response to Reply #28
29. and it's hard to make a case in favor of an ancient poll
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Dec-04-07 04:15 PM
Response to Reply #29
35. I agree
Printer Friendly | Permalink |  | Top
 
Quixote1818 Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 02:43 PM
Response to Original message
16. Edwards looks to have the most momentum. nt
Printer Friendly | Permalink |  | Top
 
Kurt_and_Hunter Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 02:44 PM
Response to Original message
17. An intellectually honest person would use the mid-point or end-point.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:06 PM
Response to Reply #17
20. How is the mid-point or the end-point more "honest" than the beginning-point?
Printer Friendly | Permalink |  | Top
 
dmallind Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:16 PM
Response to Reply #20
21. because using the latter
creates the implication that ALL the poll respondents answered when Clinton was doing much better, therefore reducing the current applicability of the poll results. Since it's very unlikely that a poll conducted over 18 days was so significantly dependent on the first few days, that's not a fair extrapolation to make.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:28 PM
Response to Reply #21
22. First, using the mid-point doesn't change the order and using the end-point only flips the order
of the two oldest polls.

Second, I clearly stated that I was listing the polls in the order of "oldest date for beginning of data collection at top to newest at bottom." How can I create an "implication" contrary to what I specifically said?
Printer Friendly | Permalink |  | Top
 
Tejanocrat Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 02:55 PM
Response to Original message
18. From the Pew poll. Who has the "strongest" support, who would you never vote for:
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:40 PM
Response to Reply #18
24. Six (!) times as many Iowa Dems definitely won't caucus for Hillary as compared to Edwards or Obama!
Printer Friendly | Permalink |  | Top
 
ClarkUSA Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 09:51 PM
Response to Reply #24
31. Wow, that's bad news for Hillaryworld!
No wonder they're attacking like rabid dogs.
Printer Friendly | Permalink |  | Top
 
Tejanocrat Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 03:57 PM
Response to Original message
25. What's up with Hillary supporters floating old numbers and calling them new? Mother Jones is on it:
New Iowa Polls Put Clinton Up Big, but the Numbers Are Phony

There's been a minor splash because of two new polls that seemingly contradict the Obama surge in Iowa.

The first is an AP-Pew poll that has the three-way race looking like this: Clinton 31%, Obama 26%, Edwards 19%

The second is an Iowa State University poll that is even more startling: Clinton 31%, Edwards 24%, Obama 20%

Here's the catch. The AP-Pew poll was conducted November 7-25. Some of the results there are two to three weeks old. The Iowa State University poll was conducted November 6-18. All of the results there are two to three weeks old. They all predate the juvenation the Obama campaign has gotten going in Iowa these past few weeks.

More current numbers all show the race tied or with Obama leading slightly. An American Research Group poll conducted 11/26-11/29 has Obama 27%, Clinton 25%, Edwards 23%. A Des Moines Register poll conducted 11/25-11/28 has Obama 28%, Clinton 25%, Edwards 23%. A Rasmussen poll conducted 11/26-11/27 has Clinton 27%, Obama 25%, Edwards 24%.

The average, according to Real Clear Politics, is dead even: Obama 27.5%, Clinton 27.2%, Edwards 22.3%.
Printer Friendly | Permalink |  | Top
 
Stop Cornyn Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 04:55 PM
Response to Reply #25
27. I'm glad to see Mother Jones on top of this.
Printer Friendly | Permalink |  | Top
 
Czolgosz Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 09:43 PM
Response to Original message
30. k & r
Printer Friendly | Permalink |  | Top
 
Sancho Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Dec-03-07 10:00 PM
Response to Original message
32. Some dates are on Real Clear Politics:
Average 11/07 - 11/29 - 27.5 27.2 22.3 8.3 4.7 Obama +0.3
American Res. Group 11/26 - 11/29 600 LV 27 25 23 4 8 Obama +2.0
Des Moines Register 11/25 - 11/28 500 LV 28 25 23 9 6 Obama +3.0
Rasmussen 11/26 - 11/27 1156 LV 25 27 24 10 4 Clinton +2.0
Strategic Vision (R) 11/23 - 11/25 600 LV 29 29 23 6 4 Tie
AP-Pew 11/07 - 11/25 460 LV 26 31 19 10 2 Clinton +5.0
ABC/Wash Post 11/14 - 11/18 500 LV 30 26 22 11 4 Obama +4.0
Printer Friendly | Permalink |  | Top
 
waiting for hope Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Dec-04-07 01:50 PM
Response to Original message
33. I like those trends -
Thanks!
Printer Friendly | Permalink |  | Top
 
Zynx Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Dec-04-07 01:56 PM
Response to Original message
34. All of those polls are virtually identical in results.
Remember back in 2006 when people freaked out over little swings here and there from 46% to 51% or whatever the case might be? This was true in all races. In the end, it all mattered very little. Close races are uncallable and always have been. The best general rule is throw out the top and bottom outliers and average from there.
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Fri May 03rd 2024, 11:07 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC