Jump to content

Why the polls and the bounce mean nothing

Rick Munroe
This topic is 6654 days old and is no longer open for new replies.  Replies are automatically disabled after two years of inactivity.  Please create a new topic instead of posting here.  

Recommended Posts

1. The Time and Newsweek polls heavily oversampled Republicans (e.g., more people say they voted for Bush than Gore in 2000).


2. Less than 2 weeks before the Iowa caucuses in January 2004, Howard Dean led Kerry by 11 points. Of course, we know that Kerry then clobbered Dean, 38% to 18%.


3. In October 2000, less than 2 weeks before election day and Gore won the popular vote by 540,000, the Gallup poll had Bush leading Gore by 13%. Read this: http://www.dailykos.com/story/2004/9/4/17459/95741


4. According to Zogby and as reported in the Washington Times and U.S. News:


96% of Gore 2000 voters plan to vote for Kerry 2004

89% of Bush 2000 voters plan to vote for Bush 2004


Read this: http://www.dailykos.com/story/2004/9/4/222040/1943


So...let's get to work. It's time for a major get-out-the-vote. Have you made sure everyone you know is registered to vote and will vote?

Link to comment
Share on other sites



There is an interesting article in today's Boston Globe about the validity of polling. The article states that trditional telephone polling is not a valid as it once was, because folks don't rust pollsteras much anymore. Respondents are likely to hang up or lie to the pollsters. The traditional polling also completely misses folks who only use cell phones. Traditional polling gets folks who have listed land lines. So the question is what about the folks who only have cell phones: How big is that group?; How d you get to them?; Where do they fit into the demographic mix?; etc. etc.


This article went on to show that internet polling is now becoming more accurate than traditional telephone polling.


Why am I blathering on about the article, here is the link: http://www.boston.com/news/politics/polls/articles/2004/09/05/poll/

Link to comment
Share on other sites

Coming from another angle, a right wing political commentary called polipundit.com has been trashing the accuracy of polls for weeks. It`s an interesting, if occasionally infuriating, read. This site has compared polls reported in the media just before elections and then the final results. With a conservative spin, of course, but there are so many obviously inaccurate polls that it makes one wonder why the American political scene is so driven by them. If you link to it, keep scrolling and you will eventually reach those sections.



Link to comment
Share on other sites

>The traditional polling also

>completely misses folks who only use cell phones. Traditional

>polling gets folks who have listed land lines. So the

>question is what about the folks who only have cell phones:

>How big is that group?; How d you get to them?; Where do they

>fit into the demographic mix?; etc. etc.


Thanks for the link -- I think it's a preview of stories we will be reading November 4, headlined "Why The Polls Were Wrong."


A couple points:


--You don't have to have a listed number to be polled, because of random number generators.


--The pollster's automatic dialers are programmed to exclude cell phones, but with "number portability," that distinction is difficult to make. Numbers with land-line prefixes, may now be cell phones.


--People who rely just on cell phones are likely to be at the top, and at the bottom of the educational scale; in other words, more likely to be Democrats.

Link to comment
Share on other sites

Gallup poll came out today:


Among Registered Voters:


Bush 49

Kerry 48


Among Likely Voters:


Bush 52

Kerry 45


Gallup: "Bush's two-point convention bounce is one of the smallest registered in Gallup polling history, along with Hubert Humphrey's two-point bounce following the 1968 Democratic convention, George McGovern's zero-point bounce following the 1972 Democratic convention, and Kerry's 'negative bounce' of one point among registered voters earlier this year. Bush's bounce is the smallest an incumbent president has received."


I still say polls are useless, but it's interesting that the numbers are so different in the polls.

Link to comment
Share on other sites

That's what I already posted above.


But anyway, if anyone cares about polls (I hate 'em), today's Zogby numbers are really good for Kerry. Zogby is tracking swing states:



Bush 50 (47)

Kerry 45 (50)

3.9 MOE



Bush 48 (46)

Kerry 46 (48)




Bush 46 (46)

Kerry 46 (49)




Bush 49 (49)

Kerry 49 (50)




Bush 47 (45)

Kerry 50 (52)




Bush 45 (45)

Kerry 52 (51)




Bush 43 (45)

Kerry 51 (50)




Bush 49 (49)

Kerry 49 (49)




Bush 47 (46)

Kerry 48 (48)



New Hampshire:

Bush 45 (43)

Kerry 49 (51)



New Mexico:

Bush 44 (44)

Kerry 54 (50)



North Carolina:

Bush 50 (49)

Kerry 47 (50)




Bush 54 (51)

Kerry 43 (46)




Bush 43 (43)

Kerry 53 (54)




Bush 47 (44)

Kerry 50 (52)




Bush 53 (48)

Kerry 44 (50)




Bush 51 (49)

Kerry 46 (48)




Bush 44 (45)

Kerry 53 (53)



West Virginia:

Bush 49 (49)

Kerry 40 (42)




Bush 48 (46)

Kerry 50 (51)


Link to comment
Share on other sites

Numbers Game


>that it makes one wonder why the American

>political scene is so driven by them.


I had an interesting conversation earlier tonight with a client over dinner. He is a very intelligent man, in his mid 30s, he is in a car a great deal on calls and also reads the New York Times daily, rather than the Los Angeles papers. He listens to a lot of talk radio. The conversation turned to politics and I mentioned any number of things I found interesting (how, for example, largely religous voters in states such as Oklahoma, even thought it has high unemployment, a large number of displaced workers, and other low economic indicators, are likely to drive a double digit vote for Mr. Bush in this state simply because these voters perceive him as a religous man. Neither Bush nor Mr. Kerry are intending to campaign in the state); another interesting example is how the public perceives the "war on terror" with the "war in Iraq" and how the two are coupled, to Bush's advantage.


The polls are meaningless but it is what the media, by and large, print, electronic, web and otherwise, report on because it is quick, simple and easy. The harder task: to provide actual analysis or to hold any candidate to account, requires more work on the part of the media, facing deadlines and what the media perceives as a lack of interest on the public at large.


Largely, the public does not care. More people will not vote in this election than at any other time in the history of this country, in spite of significant get out the vote efforts by both parties and in spite of the fact that both parties anticipate that more actual voters will vote. Most people, including interested, intelligent and well informed individuals, are not paying attention to much of the campaign. Large numbers of the population are not interested nor could they be called well informed. These are people who believe that Iraq was involved in 9/11 and who consistently believe that WMD's were found, when they are asked.


In this environment, the horse race aspect, with the pretty graphics and colors, almost always leads the pundits talk shows, front pages and elsewhere, because it gives a simple, albeit simplistic thing to wrap your head around.

Link to comment
Share on other sites


This topic is now archived and is closed to further replies.

In order to post in the Political Issues forum, all members are required to acknowledge that their post is in compliance with our Community Guidelines.  In addition, you acknowledge that it meets the following requirements: 

  • No personal attacks: Attack the issue not the person
  • No hijacking: Stay on the subject of the thread 

  • No bullying, hate speech or offensive terms/expressions

In addition, if the moderators feel someone is reporting content simply because if it’s political stance (such as but not limited to reporting it as off topic but not other off topic replies by those that agree with your stance), the reporting person may receive a warning as well.

Content that does not comply with the above requirements will be removed.  Multiple violations may result in a loss of access to this forum.

  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Create New...