PDA

View Full Version : New Polls disagree with jaz - will wonders ever cease!!!


Matt Helm
09-08-2004, 11:22 AM
Newsweek Sept 2-3, 2004

Bush 54%
Kerry 41%

Gee, jiz. Looks like your polls may be somewhat biased?

jAZ
09-08-2004, 11:24 AM
Newsweek Sept 2-3, 2004

Bush 54%
Kerry 41%

Gee, jiz. Looks like your polls may be somewhat biased?
ROFL

(You are right on top of things, aren't ya?)

ROFL

Braincase
09-08-2004, 11:25 AM
I never trust polls. You have no idea how the question was phrased, or the location. For all you know, the poll was given right outide the Long Island NRA-affilliated gun club. jAZ's poll might be from right outside a locale frequented by libs.

Matt Helm
09-08-2004, 11:27 AM
ROFL

(You are right on top of things, aren't ya?)

ROFL

One of us has to be, it is an assured that you are not!!

Why do you lie? Does it make you feel better?

Radar Chief
09-08-2004, 11:27 AM
Don’t let Anton see all this talk of poles. :shake:

Matt Helm
09-08-2004, 11:28 AM
I never trust polls. You have no idea how the question was phrased, or the location. For all you know, the poll was given right outide the Long Island NRA-affilliated gun club. jAZ's poll might be from right outside a locale frequented by libs.

Yep, Newsweek normally takes it's polls there.

Now jaz, on the other hand, asks the question to himself 2007 times and uses those numbers for his polls. The rest he makes up.

jAZ
09-08-2004, 11:36 AM
One of us has to be, it is an assured that you are not!!

Why do you lie? Does it make you feel better?
You are about a week late on the Newsweek poll.... We've talked it to death already. It was flawed as was the Time poll. They both sampled more Republicans than Dems even though there are more Dems than Republicans. That kinda skews the results (sorta like polling the DC forum).

But if you want to call it "new" even though it's a week old... and believe Bush has an 11 point lead when its a dead heat... have at it.

Matt Helm
09-08-2004, 11:38 AM
You are about a week late on the Newsweek poll.... We've talked it to death already. It was flawed as was the Time poll. They both sampled more Republicans than Dems even though there are more Dems than Republicans. That kinda skews the results (sorta like polling the DC forum).

But if you want to call it "new" even though it's a week old... and believe Bush has an 11 point lead when its a dead heat... have at it.

This story was on CNN not more than 45 minutes ago. So you are saying that they are now lying about the continued lead that GWB has?

RINGLEADER
09-08-2004, 11:39 AM
You are about a week late on the Newsweek poll.... We've talked it to death already. It was flawed as was the Time poll. They both sampled more Republicans than Dems even though there are more Dems than Republicans. That kinda skews the results (sorta like polling the DC forum).

But if you want to call it "new" even though it's a week old... and believe Bush has an 11 point lead when its a dead heat... have at it.


Did you find the final weighting that Time and Newsweek used Jaz? Because I still haven't found anything other than what that one blogger was claiming.

Not talking smack, just never saw/heard what the final weighting on those polls ended up being.

KCTitus
09-08-2004, 11:42 AM
Did you find the final weighting that Time and Newsweek used Jaz? Because I still haven't found anything other than what that one blogger was claiming.

Not talking smack, just never saw/heard what the final weighting on those polls ended up being.

I saw a thread from Zogby that had the breakdowns...the republican percentage was higher than traditional voting patterns.

jAZ
09-08-2004, 11:46 AM
This story was on CNN not more than 45 minutes ago. So you are saying that they are now lying about the continued lead that GWB has?
Well if they are reporting the Newsweek Poll its not technically lying. If they aren't pointing out the methodology flaw, they are just hyping a flawed poll.

Chief Henry
09-08-2004, 11:51 AM
54-41

Is this accurate? The one released earlier by Newsweek says 52-41 % Which poll are we
talking about here?

Matt Helm
09-08-2004, 11:53 AM
Well if they are reporting the Newsweek Poll its not technically lying. If they aren't pointing out the methodology flaw, they are just hyping a flawed poll.

How about this one?

http://www.gallup.com/content/?ci=12922
GALLUP NEWS SERVICE

PRINCETON, NJ -- The CNN/USA Today/Gallup post-Republican convention poll -- the first national poll conducted entirely after the completion of that convention -- shows George W. Bush getting a small increase in voter support. Bush's share of the vote among likely voters increased two percentage points, from 50% to 52%, while Democratic presidential candidate John Kerry's share dropped by an equal amount, in the two-way race. Bush now leads Kerry by 52% to 45% among likely voters, compared with a 50% to 47% lead for Bush prior to the convention. Bush maintains that same 52% to 45% margin when independent candidate Ralph Nader, who receives 1% support among likely voters, is included in the ballot question.

Notably, this is the first time Bush has had a lead over Kerry beyond the poll's margin of error since Kerry's surprise victory in the Iowa caucuses propelled him to his party's nomination.

Matt Helm
09-08-2004, 11:54 AM
54-41

Is this accurate? The one released earlier by Newsweek says 52-41 % Which poll are we
talking about here?

It was a typo, it is 52.

Saulbadguy
09-08-2004, 11:55 AM
This is funny

jAZ
09-08-2004, 11:55 AM
Did you find the final weighting that Time and Newsweek used Jaz? Because I still haven't found anything other than what that one blogger was claiming.
Zogby is pointing it out.

http://zogby.com/news/ReadNews859.html

8/7/2004
2004: It Is Not An 11 Point Race - by John Zogby
The Republican National Convention is over and score it a huge success for President George W. Bush. For one solid week he was on message and got Americans who watched to listen to the message he intends to carry in the fall campaign: leadership, decisiveness and success battling the war on terrorism. The convention actually followed another big week for Mr. Bush and equally dismal one for his opponent, Democratic Senator John Kerry.

Now the first polls are out. I have Mr. Bush leading by 2 points in the simple head-to-head match up - 46% to 44%. Add in the other minor candidates and it becomes a 3 point advantage for the President - 46% to 43%. This is no small achievement. The President was behind 50% to 43% in my mid-August poll and he essentially turned the race around by jumping 3 points as Mr. Kerry lost 7 points. Impressive by any standards.

For the first time in my polling this year, Mr. Bush lined up his Republican ducks in a row by receiving 90% support of his own party, went ahead among Independents, and now leads by double-digits among key groups like investors. Also for the first time the President now leads among Catholics. Mr. Kerry is on the ropes.

Two new polls came out immediately after mine (as of this writing) by the nation's leading weekly news magazines. Both Time's 52% to 41% lead among likely voters and Newsweek's 54% to 43% lead among registered voters give the President a healthy 11 point lead. I have not yet been able to get the details of Time's methodology but I have checked out Newsweek's poll. Their sample of registered voters includes 38% Republican, 31% Democrat and 31% Independent voters. If we look at the three last Presidential elections, the spread was 34% Democrats, 34% Republicans and 33% Independents (in 1992 with Ross Perot in the race); 39% Democrats, 34% Republicans, and 27% Independents in 1996; and 39% Democrats, 35% Republicans and 26% Independents in 2000. While party identification can indeed change within the electorate, there is no evidence anywhere to suggest that Democrats will only represent 31% of the total vote this year. In fact, other competitors have gone in the opposite direction. The Los Angeles Times released a poll in June of this year with 38% Democrats and only 25% Republicans. And Gallup's party identification figures have been all over the place.

This is no small consideration. Given the fact that each candidate receives anywhere between eight in ten and nine in ten support from voters in his own party, any change in party identification trades point for point in the candidate's total support. My polls use a party weight of 39% Democrat, 35% Republican and 26% Independent. Thus in examining the Newsweek poll, add three points for Mr. Bush because of the percentage of Republicans in their poll, then add another 8% for Mr. Bush for the reduction in Democrats. It is not hard to see how we move from my two-point lead to their eleven-point lead for the President.

I will save the detailed methodological discussion for another time. But I will remind readers that my polling has come closest to the final results in both 1996 and 2000.

None of this takes away from the President's achievement. He got out of his party's convention everything he needed to launch his campaign in earnest in the closing two months. But my poll still reveals lurking shadows for him. He still has a net negative job performance rating, a negative re-elect (i.e. more voters think it is time for someone new than feel he deserves re-election) and a net negative wrong direction for the country.

The poll also suggests that Mr. Kerry is behind and has a lot of work to do to refocus the campaign on the issues that must work for him: the economy, health care, and the execution of the war in Iraq. We also see now that at least in the short run, the advertising campaign against the Senator about his military service in Vietnam has raised questions about his integrity and has caused his personal unfavorable numbers to jump.

But with all that said, it simply is not an 11 point race. It just isn't.

John Zogby is the President and CEO of Zogby International- an independent polling firm, and writes this column for the Financial Times where it first appeared..

Matt Helm
09-08-2004, 11:56 AM
CCN
http://www.blogit.com/Blogs/Blog.aspx/ccnews6633/

Monday, September 6, 2004
Gallup, Time, Newsweek Show Bush Leading By Large Margin in Polls
Gallup: Bush 52, Kerry 45, Nader 1 Newsweek: Bush 52, Kerry 41, Nader 3 Time: Bush 52, Kerry 41, Nader 3 Source: RealClearPolitics.com Sign in to see full post.
posted by ccnews at 6:22 PM Comments (2) (permalink)

Sunday, August 29, 2004
Bush Pulling Ahead on Eve of Republican National Convention; Go Bush!
RealClearPolitics Poll Averages: 3-Way: Bush 46.2, Kerry 44.6, Nader 3.6 Head-to-Head: Bush 47.4, Kerry 45.8 Bush JA: 50.2 Approve/46.5 Disapprove National Polls: Time: Bush 46, Kerry 44, Nader 5 NBC/WSJ: Bush 47, Kerry 45, Nader 3 Gallup: Bush 48, Kerry 46, Nader 4 FOX News: Kerry 44, Bush 43,... Sign in to see full post.

Matt Helm
09-08-2004, 11:57 AM
http://www.chicagotribune.com/news/columnists/chi-0211140128nov14,0,6863919.column
The coming decline of political polls
Steve Chapman

November 14, 2002

When a reporter for The New York Times Magazine recently asked White House political adviser Karl Rove if President Bush is too closely identified with big business at a time of corporate scandal, Rove began reciting the latest poll findings.

"Forty-five percent of the people think Bush's proposals for reforming accounting go too far or are about right," he noted, "versus 39 percent who say they do not go far enough. Now that's compared to 39 percent who said they go too far or are about right a month ago, and 43 who said they do not go far enough."

Then Rove stopped, realizing he was making his boss look like a human windsock. "Not that we spend a lot of time on these," he assured his listener.

Bush is hardly the first president to keep a close eye on such data. Bill Clinton commissioned a poll to find out if he should come clean about the entire Monica Lewinsky scandal. He was told Americans could forgive adultery but not perjury and obstruction of justice, and--well, you know the rest.

Once upon a time, politicians weighing policy decisions had to rely on their own sense of what was right and what was appealing to voters. Today, officeholders and candidates are all hooked up to IVs that continuously drip fresh poll data directly into their veins. An entire industry has grown up to tell them what every demographic group thinks about every conceivable issue and how each segment of the electorate may be won over by tweaking the candidate's message.

But last week's election outcomes left some pollsters resembling contestants trying to catch a greased pig--with their quarry escaping and their faces splattered with mud. A late Zogby poll had Republican Jim Ryan a hair ahead of Rod Blagojevich in the Illinois governor's race, but the Democrat won by seven points. An Atlanta Journal-Constitution/WSB-TV survey a week before Election Day had Republican Sonny Perdue trailing incumbent Georgia Gov. Roy Barnes by 11 points. When the votes were counted, Perdue won by five points.

The last polls in Minnesota couldn't be wrong, because one of them had Democrat Walter Mondale with a five-point lead in the U.S. Senate race, while another had Republican Norm Coleman ahead by six. (Coleman won.)

Failures like these are not the product of bad luck or incompetence but of changes that pollsters have not been able to cope with. Factors beyond their control are making it harder and harder to measure and interpret what the public thinks.

One is that a lot of people simply refuse to pick up the phone and answer questions. About the only public sentiment that pollsters can vouch for is that cold calls from strangers are about as popular as West Nile virus.

The rise of cell phones, which generally don't get called, has added another hurdle. The problem has gotten so big that Karlyn Bowman, a polling expert at the American Enterprise Institute in Washington, says, "I expect that in 10 years, phone interviews will be a thing of the past, replaced by Internet polling."

Political polls also have to make adjustments to reflect how likely people are to vote, since Candidate A won't win if he has lots of supporters who stay home on Election Day. Such adjustments become harder as American society grows more diverse, because new ethnic groups may not follow the same patterns as older ones. Even if you can get people to tell you whom they plan to vote for, you may have no idea which candidate is ahead.

Surveys can still yield lots of useful information, if the pollster has the time and money to keep going back to non-responders to make sure the sample is representative. In the heat of a campaign, that option doesn't exist. So polls often mislead.

But politicians and campaign managers continue to use them for lack of anything better. At some point, though, they may decide it makes more sense to hire an astrologer. They may have to confront a new environment in which they put their finger to the wind and find there is no wind.

Most people go into politics with some clear ideas of what they want to accomplish, but most end up parroting poll-tested slogans that some consultant says will charm (or fool) voters. They'll change their approach only if they learn that the polls are unreliable.

That development might put a lot of political consultants out of business. But it wouldn't be a bad thing if our leaders spent less time trying to figure out what the citizenry believes and more figuring out what they believe.

----------

E-mail: schapman@tribune.com

jAZ
09-08-2004, 11:57 AM
How about this one?

http://www.gallup.com/content/?ci=12922
GALLUP NEWS SERVICE

PRINCETON, NJ -- The CNN/USA Today/Gallup post-Republican convention poll -- the first national poll conducted entirely after the completion of that convention -- shows George W. Bush getting a small increase in voter support. Bush's share of the vote among likely voters increased two percentage points, from 50% to 52%, while Democratic presidential candidate John Kerry's share dropped by an equal amount, in the two-way race. Bush now leads Kerry by 52% to 45% among likely voters, compared with a 50% to 47% lead for Bush prior to the convention. Bush maintains that same 52% to 45% margin when independent candidate Ralph Nader, who receives 1% support among likely voters, is included in the ballot question.

Notably, this is the first time Bush has had a lead over Kerry beyond the poll's margin of error since Kerry's surprise victory in the Iowa caucuses propelled him to his party's nomination.
Gallup doesn't seem to have the same flaw as Time and Newsweek. It's certainly a more accurate poll.

Matt Helm
09-08-2004, 11:58 AM
Ask our Experts > Can political polls and statistics ever be trusted?
Posted Aug. 10/04

Carleton University faculty and staff experts answer life’s perplexing questions. Scott Bennett, Associate Professor of Political Science, answers this month’s question:

In the recent federal election, many political pundits said the predictions of the polls were pretty far off. Can political polls and statistics ever be trusted?

The track record for pollsters predicting elections has been good since the 1960s. Widespread improvements in methodology have made misleading projections rare. Our recent national election may be one of the problematic cases. However, if one examines all the polls and puts them in context, the 2004 electoral projections don’t look too bad. To appreciate this, consider some of the things that can go wrong in projecting future electoral behaviour from polls:


Random Sampling Error—In media releases on polls, this is often reflected in the small percentage of error one might expect 19 times out of 20. This is usually not a major problem, but, by definition, this sort of error can infrequently exceed desired bounds even with soundly derived samples. Further, even estimates within the range of acceptable error will seldom be a perfect reflection of a population.

The Problematic Link Between Survey Responses and Actual Behaviour—Poll respondents can give intentionally or unintentionally misleading responses. There are methods for dealing with this, but it can still be problematic. This is particularly true when opinion and preference are changing rapidly and there is a time gap between polls and balloting.

Increasing Problems in Getting Good Population Coverage Through Telephone Interviewing—This is something the polling industry must watch closely and is generally aware of.
In our recent election, the primary problem in projecting popular vote arose from the second point mentioned above. It was not so much that people gave intentionally misleading responses. Rather, they were legitimately changing their vote preferences in significant ways between the last publicly released polls and election day. One company (COMPAS Research) led by Carleton Professor Conrad Winn) polled on election day and found many people were making final choices at the last minute. The highly accurate COMPAS election-day survey showed that same-day and previous day voting decisions strongly favoured the Liberals over the Conservatives, thereby suggesting that the previous polls of all the various firms were relatively accurate at the time they were released.

To the credit of other companies, some of them did pick up subtle changes in the second half of the campaign and were suggesting caution in interpreting popular vote projections from their final polls.

---

Do you have one of life’s perplexing questions? Need an answer? Email us, and we’ll track it down. Email: editor@carleton.ca

Chief Henry
09-08-2004, 11:59 AM
It was a typo, it is 52.


THank you for clearing that up.

Matt Helm
09-08-2004, 11:59 AM
How Reliable are Political Polls?

http://www.stanford.edu/group/gradethenews/dreamhost%20files/pagesfolder/Pollstory1.htm

Pollsters are reluctant to call even their final polls before an election a “prediction” of voting results. Most of these surveys concluded their interviews 7-10 days before election day. Pollsters quite properly point out that a disturbing revelation about a candidate during the last week of a campaign could render even the most careful survey moot. They also argue that a last-minute gust of advertising sometimes blows the numbers around. And trends from previous polls need to be taken into account -- if one side appears to be rising or falling.



But as the venerable Mervin Field Poll asserts in a self-assessment of its accuracy, “no polling organization can stay in operation long if there are frequent or wide variations between poll findings and election results.”



How the analysis was done


Grade the News conducted a Lexis-Nexis search through the courtesy of Santa Clara University’s Markkula Center for Applied Ethics. We looked at press reports in Bay Area newspapers contained in the Lexis-Nexis online library. We compared the results of the final poll before the San Francisco city elections in November and December’s run-off and the final poll before the spring California primary with actual voting results.



We focused only on races the Bay Area press reported on in papers indexed by Lexis-Nexis. These races presumably were most important to the local public. Note that Lexis-Nexis does not catalogue the Contra Costa Times or other smaller newspapers, nor does it contain all San Jose Mercury News stories. Likewise, local TV news transcripts were not available.



To examine the predictive power of these polls, we relied primarily on a method proposed by Warren Mitofsky, one of the nation’s most respected and veteran pollsters. Director of elections and surveys for CBS News for 27 years, Mitofsky now runs his own international polling organization.



We chose his method because of his reputation both among commercial and academic pollsters, because the method is the same for propositions and candidates, and because he was not involved in any of the surveys we analyzed.



Mitofsky suggested we assess “accuracy” by comparing the margin between the two leading candidates or two sides in a proposition in the final poll with the actual vote margin. If the difference between the winning and losing sides that is predicted in the poll is within two times the margin of error of the sample, then the poll is accurate.



For example, the KTVU (Channel 2)/San Francisco Examiner Poll showed the Yes’ ahead of the No’s on Prop 22 (limiting legal marriage to heterosexual couples) by a margin of 55 to 38 with 7% undecided. The poll reached 634 likely California voters, so the margin of error was calculated at +/- 4%.



Mitofsky argues that we can’t know what the undecided voters will do, so we can only work with those who express an opinion. The margin or gap between the Yes’ and No’s is 17 (55 minus 38). Given the margin of error, however, the gap really could be as wide as 59 Yes’ (55+4) and 34 (38-4) No’s -- a range of 25 points. On election day, Prop 22 won 61.4 to 38.6. The actual margin, 22.8 percentage points, lay within the range predicted. So the poll predicted this race accurately.



The results


In the California open primary, held March 7, the side leading in the final poll before the election won about 8 times in 10. However, the actual vote fell within the margins of error only about 4 times in 10, even though all but one race the press focused on was a landslide. (Prop 26 was decided by a 2.4 point margin; the next closest was decided by 23 points.) The more public opinion falls on one side of a contest, the more likely even a haphazard survey is to predict the winner.

jAZ
09-08-2004, 11:59 AM
Monday, September 6, 2004
Gallup, Time, Newsweek Show Bush Leading By Large Margin in Polls
Two of those 3 are flawed. The 3rd (Gallup) is about normal (showing a 2 point bounce from the convention).

jAZ
09-08-2004, 12:00 PM
This is funny
Isn't it?

Matt Helm
09-08-2004, 12:01 PM
Gallup doesn't seem to have the same flaw as Time and Newsweek. It's certainly a more accurate poll.

What you are doing is claiming that one is flawed because skerry is shown with -4%points?

All polls are flawed. Anyone and everyone can find a poll to support their claim, it's just that you have never attempted to be fair with the numbers you throw up. There is even suspicion that you manufacture the numbers yourself to justify your cause.

The only "Poll" numbers that will count will be the ones cast on Nov. 2.

Matt Helm
09-08-2004, 12:02 PM
Two of those 3 are flawed. The 3rd (Gallup) is about normal (showing a 2 point bounce from the convention).

As well as showing a negative bounce for skerry, or did you forget that?

Saulbadguy
09-08-2004, 12:03 PM
The only "Poll" numbers that will count will be the ones cast on Nov. 2.

So you start a thread about polls...ok.

Matt Helm
09-08-2004, 12:03 PM
So you start a thread about polls...ok.

In your usual fashion you have missed the point completely.

jaz lied people died.

Saulbadguy
09-08-2004, 12:04 PM
jaz lied people died.
ROFL

StcChief
09-08-2004, 12:05 PM
I saw a thread from Zogby that had the breakdowns...the republican percentage was higher than traditional voting patterns.

Zogby polls over time have been very accurate.

jAZ
09-08-2004, 12:06 PM
As well as showing a negative bounce for skerry, or did you forget that?
And if Kerry gains 2 points, its back to where it was before. It's fun with numbers!

Matt Helm
09-08-2004, 12:07 PM
And if Kerry gains 2 points, its back to where it was before. It's fun with numbers!

But, you forgot one point, he hasn't.