PDA

View Full Version : AP Poll: Bush, Kerry in Dead Heat


Hel'n
10-21-2004, 03:40 PM
Oct 21, 4:22 PM (ET)

By RON FOURNIER


WASHINGTON (AP) - President Bush and Sen. John Kerry are locked in a tie for the popular vote, according to an Associated Press poll. Voters seem open to change in the White House - most disapprove of the president's performance at home and in Iraq - but still harbor doubts about making the switch.

Bush's strength continues to be in a perception by voters that he is the most qualified to protect the country, though his advantage has dwindled in recent weeks. Some 56 percent say the country is on the wrong track.

In the AP-Ipsos Public Affairs poll, the Democratic ticket of Kerry and Sen. John Edwards got support from 49 percent of those who said they were likely to vote, and the Republican team of Bush and Vice President Dick Cheney got 46 percent, within the poll's margin of error of plus or minus 3 percentage points. The Oct. 18-20 survey, released Thursday, included 976 likely voters.

A spate of other polls show the race tied or give Bush a slight lead nationwide. The presidency will go to whoever gets a majority of the 538 Electoral College votes, a state-by-state chase that is just as close as national surveys.

Likely voters are divided on many levels:

- They are just as likely to back Democrats for Congress as Republicans, with a 47-46 split favoring Democrats. That is essentially a tie.

- Twenty-four percent say they have already voted or will cast ballots before Election Day. Those who voted early were just as likely to back Kerry as Bush.

- A third of likely voters have been contacted by a candidate, campaign or outside group seeking support. Twenty-three percent said they were urged to back Kerry and 21 percent said they were asked to support Bush, a sign that two massive get-out-the-vote campaigns have had equal success contacting voters.

Less than half, 47 percent, approve of Bush's job performance. A rating below 50 percent spells trouble for any incumbent, and Bush falls below that threshold on the economy, domestic affairs and handling Iraq.

In each case, Bush's approval numbers have held steady since the AP-Ipsos poll taken after the first presidential debate.

That Bush performance, roundly criticized on style and substance, helped lower the president's standing against Kerry from early September, when the incumbent led in the head-to-head matchup and had higher approval scores.

A majority, 51 percent, support the president's handling of foreign policy and the war on terror. By 7 percentage points, they think he would protect the country better than Kerry. That is similar to the AP-Ipsos poll earlier in the month, but down from a 23-point advantage in March.

Voters are evenly split on who would do the best job on Iraq. They find the candidates equally honest and likable, but Bush is viewed as much more decisive.

By an 18-point margin, Kerry is seen as best suited to create jobs for workers.

http://apnews.myway.com//article/20041021/D85S1L700.html

the Talking Can
10-21-2004, 03:42 PM
being inside the MOE is not the same as being even or "tied", I believe...I wish reporters spent more time studying statistics so I wouldn't have too....can any savy pollsters explain that?

Hel'n
10-21-2004, 03:50 PM
being inside the MOE is not the same as being even or "tied", I believe...I wish reporters spent more time studying statistics so I wouldn't have too....can any savy pollsters explain that?

I'd like to understand it better too... Where's RL?

KCN
10-21-2004, 03:53 PM
What it means is that if you perform this poll 100 times, 95 of those polls will give you results within 3 points of those results.

the Talking Can
10-21-2004, 03:54 PM
What it means is that if you perform this poll 100 times, 95 of those polls will give you results within 3 points of those results.

for each candidates number, correct?

KCN
10-21-2004, 03:56 PM
for each candidates number, correct?

I'm not quite sure how it works with 4 possible questions (Bush/Kerry/Nader/Undecided)

the Talking Can
10-21-2004, 04:01 PM
the internet is truly amazing...here's an article w/links called "20 Questions A Journalist Should Ask About Poll Results"


faq about polls (http://www.ncpp.org/qajsa.htm)

Pollsters express the degree of the certainty of results based on a sample as a "confidence level." This means a sample is likely to be within so many points of the results one would have gotten if an interview were attempted with the entire target population. Most polls are usually reported using the 95% confidence level.

Thus, for example, a "3 percentage point margin of error" in a national poll means that if the attempt were made to interview every adult in the nation with the same questions in the same way at the same time as the poll was taken, the poll's answers would fall within plus or minus 3 percentage points of the complete count’s results 95% of the time....

Remember that the sampling error margin applies to each figure in the results – it is at least 3 percentage points plus or minus for each one in our example. Thus, in a poll question matching two candidates for President, both figures are subject to sampling error.
----------


Certainly, if the gap between the two candidates is less than the sampling error margin, you should not say that one candidate is ahead of the other. You can say the race is "close," the race is "roughly even," or there is "little difference between the candidates." But it should not be called a "dead heat" unless the candidates are tied with the same percentages. And it certainly is not a “statistical tie” unless both candidates have the same exact percentages.

And just as certainly, when the gap between the two candidates is equal to or more than twice the error margin – 6 percentage points in our example – and if there are only two candidates and no undecided voters, you can say with confidence that the poll says Candidate A is clearly leading Candidate B.

When the gap between the two candidates is more than the error margin but less than twice the error margin, you should say that Candidate A "is ahead," "has an advantage" or "holds an edge." The story should mention that there is a small possibility that Candidate B is ahead of Candidate A.

When there are more than two choices or undecided voters – virtually in every poll in the real world – the question gets much more complicated.

Amnorix
10-21-2004, 04:14 PM
I took statistics for non-math types in college, which is a long time ago now. What I remember is:

1. Anything with a margin of error over 5% is junk. Just ignore it.

2. 5% margin of error is minimal acceptability. 3% is much better.

3. The margin of error essentially means that the number could be higher or lower by that amount. So if there is a 49-47% preference for A versus B, but a 3% margin of error, you cannot be truly certain that A is in fact preferred over B. All you can be really certain about is that (1) it's very close as to whether A or B is preferred, and (2) A is probably preferred over B. The numbers aren't useless, but if it's within the margin of error, it really means we don't know for sure, but it SEEMS like it's leaning this way or that way.

4. if the results are outside the margin of error, you are certain that A is preferred over B. Your actual percentages might not be precisely accurate, but if it's outside the margin of error, then you ARE certain that one is more popular (or whatever) than the other.

5. Nothing really changes where you have multiple options. If you had a 3 way breakdown of A at 37%, B at 35% and C at 28%, and a margin of error of 3%, then you effectively know (1) A appears to be slightly favored over B, but it's very close and we can't be certain that's correct, and (2) both A and B are definitely preferred over C.

Maybe this hekps a bit, maybe not. Hopefully my memory is accurate at least and nobody is going to tell me I screwed this all up. :eek:

RINGLEADER
10-21-2004, 04:19 PM
The AP/Ipsos poll is the only one that Kerry has led in for the last two weeks. There have been 29 other polls since the first debate and Bush has led in 21 of them, Kerry has led in 3 of them (2 of those 3 were AP/Ipsos polls) and 5 of them were tied.

There have been nearly 100 tracking polls since the RNC convention and Kerry has led in four of them.

ABC News poll for today has Bush up +7. Washington Post has Bush up +6. Opinion Dynamics has Bush up +7. Gallup has Bush up +8. Newsweek has Bush up +6. Even Harris' 2000 model, the one that was the most accurate of all the polls last time, has Bush up +8.

Zogby, Marist, Time, NBC/WSJ, CBS/NY Times, Pew all have the race tied or Bush up only +1.

Kerry leads in AP/Ipsos +3.

Which one looks like the outlier?

RINGLEADER
10-21-2004, 04:21 PM
I'd like to understand it better too... Where's RL?

I agree. Being inside the MoE isn't tied. That's why I think the polls since the debates are so revealing...for a candidate that supposedly won all the debates and has the momentum Kerry is doing rather poorly.

KCN
10-21-2004, 04:22 PM
I took statistics for non-math types in college, which is a long time ago now. What I remember is:

1. Anything with a margin of error over 5% is junk. Just ignore it.

2. 5% margin of error is minimal acceptability. 3% is much better.

3. The margin of error essentially means that the number could be higher or lower by that amount. So if there is a 49-47% preference for A versus B, but a 3% margin of error, you cannot be truly certain that A is in fact preferred over B. All you can be really certain about is that (1) it's very close as to whether A or B is preferred, and (2) A is probably preferred over B. The numbers aren't useless, but if it's within the margin of error, it really means we don't know for sure, but it SEEMS like it's leaning this way or that way.

4. if the results are outside the margin of error, you are certain that A is preferred over B. Your actual percentages might not be precisely accurate, but if it's outside the margin of error, then you ARE certain that one is more popular (or whatever) than the other.

5. Nothing really changes where you have multiple options. If you had a 3 way breakdown of A at 37%, B at 35% and C at 28%, and a margin of error of 3%, then you effectively know (1) A appears to be slightly favored over B, but it's very close and we can't be certain that's correct, and (2) both A and B are definitely preferred over C.

Maybe this hekps a bit, maybe not. Hopefully my memory is accurate at least and nobody is going to tell me I screwed this all up. :eek:

I took grad-level stats for non majors last semester and already forgot much of it. But they all appear correct, except for #4. I believe the margin of error is done to within a 95% confidence interval, so you aren't 100% certain.

EDIT: It was three semesters ago. I meant last year. Ay, my brain is failing me.

the Talking Can
10-21-2004, 04:22 PM
here's another article that explains what the margin means..basically, this poll is saying it is 85% likely that Kerry does lead by 3 points..I think:

link (http://www.washingtonmonthly.com/archives/individual/2004_08/004536.php)

"In fact, what we're really interested in is the probability that the difference is greater than zero — in other words, that one candidate is genuinely ahead of the other. But this probability isn't a cutoff, it's a continuum: the bigger the lead, the more likely that someone is ahead and that the result isn't just a polling fluke. So instead of lazily reporting any result within the MOE as a "tie," which is statistically wrong anyway, it would be more informative to just go ahead and tell us how probable it is that a candidate is really ahead. As a service to humanity, here's a table that tells you:"
http://www.washingtonmonthly.com/blogphotos/Blog_MOE.gif

Calcountry
10-21-2004, 04:26 PM
Bush has the lead and the ball, and Kerry has maybe one time out left. One more first down, and Bush runs out the clock.

RINGLEADER
10-21-2004, 04:26 PM
I took grad-level stats for non majors last semester and already forgot much of it. But they all appear correct, except for #4. I believe the margin of error is done to within a 95% confidence interval, so you aren't 100% certain.

Yep.

The general rule of thumb, as Amnorix rightly points out, is that if the MoE is higher than 5.0% it really is a dubious poll. And the 95% rule, as KCN points out, is also true - 1 out of every 20 polls is just going to be way-off wrong.

Like I pointed out above, there are six current polls that show Bush leading 6%-8% and two current polls showing Bush and Kerry tied, three showing Bush leading 1%, and one poll showing Bush up 2%.

One poll shows Kerry ahead by 3%.

That's not one out of 20, but it's close.

the Talking Can
10-21-2004, 04:27 PM
Yep.

The general rule of thumb, as Amnorix rightly points out, is that if the MoE is higher than 5.0% it really is a dubious poll. And the 95% rule, as KCN points out, is also true - 1 out of every 20 polls is just going to be way-off wrong.

Like I pointed out above, there are six current polls that show Bush leading 6%-8% and two current polls showing Bush and Kerry tied, three showing Bush leading 1%, and one poll showing Bush up 2%.

One poll shows Kerry ahead by 3%.

That's not one out of 20, but it's close.

what are the 6 polls showing Bush with a 6-8% lead?

RINGLEADER
10-21-2004, 04:29 PM
here's another article that explains what the margin means..basically, this poll is saying it is 85% likely that Kerry does lead by 3 points..I think.

This poll, using these respondants and this methodology shows Kerry up 3% unless it's a screwed up poll (it's consistent with the last AP/Ipsos poll, however, that showed Kerry up by 4% though), so I'd imagine the results are representative of the electorate (despite what a dozen other polls say) according to the AP.

Doesn't mean they're right as a dozen other polls suggest.

RINGLEADER
10-21-2004, 04:38 PM
what are the 6 polls showing Bush with a 6-8% lead?


ABC News: Bush up 51-44
Washington Post: Bush up 51-45
Gallup: Bush up 52-44
Harris Poll: Bush up 51-43
Fox News/Opinion Dynamics: Bush up 49-42
Newsweek: Bush up 50-44

Kerry has a big problem and a big opportunity in these results: The big problem (and one that even extends to the polls that show the race tighter) is that Kerry can't seem to muster more than 45% support in ANY poll outside of AP/Ipsos. The average in the polls above for Kerry is 43.6%. He will not get elected if that trend continues. The big opportunity for Kerry is that the polls above use a likely voter screen that may not fully reflect the turnout this year (that's the Dems hope at least). If they can get something above 60% turn-out then I think the polls that are weighted by partisanship (that by their very nature build in a turn-out advantage for the Democrats) that show a closer result may turn out to be prescient. If not, the proven models of Gallup and Harris and Opinion Dynamics spell almost certain doom for the Kerry campaign.

Of course the GOP is the party that has been focusing on turning out the base so it will be interesting to see how the ground game evolves in the next two weeks. I suspect certain states like Iowa will benefit from it while others - such as Wisconsin and Ohio - will favor the president.

RINGLEADER
10-21-2004, 04:51 PM
Another bad trend for Kerry:

In the past week the lowest Bush has scored was 46% (twice), 47% (twice), 48% (three times), 49% (twice), and 50% or more (five times).

Kerry, on the other hand, has scored his highest response of 49% (once), 48% (twice), 47% (twice), 46% (none), and 45% or less (nine times).

Them's not good numbers for Kerry...

Chief Henry
10-21-2004, 05:03 PM
ABC News: Bush up 51-44
Washington Post: Bush up 51-45
Gallup: Bush up 52-44
Harris Poll: Bush up 51-43
Fox News/Opinion Dynamics: Bush up 49-42
Newsweek: Bush up 50-44

Kerry has a big problem and a big opportunity in these results: The big problem (and one that even extends to the polls that show the race tighter) is that Kerry can't seem to muster more than 45% support in ANY poll outside of AP/Ipsos. The average in the polls above for Kerry is 43.6%. He will not get elected if that trend continues. The big opportunity for Kerry is that the polls above use a likely voter screen that may not fully reflect the turnout this year (that's the Dems hope at least). If they can get something above 60% turn-out then I think the polls that are weighted by partisanship (that by their very nature build in a turn-out advantage for the Democrats) that show a closer result may turn out to be prescient. If not, the proven models of Gallup and Harris and Opinion Dynamics spell almost certain doom for the Kerry campaign.

Of course the GOP is the party that has been focusing on turning out the base so it will be interesting to see how the ground game evolves in the next two weeks. I suspect certain states like Iowa will benefit from it while others - such as Wisconsin and Ohio - will favor the president.

I'm wondering about all of those P. Diddy - Rock the Vote and other
new register'd voters...If i had to guess, i would say a rather LARGE majority of those new register'd by those groups mention'd above
will be voting for Kerry. IMO......

RINGLEADER
10-21-2004, 05:10 PM
I'm wondering about all of those P. Diddy - Rock the Vote and other
new register'd voters...If i had to guess, i would say a rather LARGE majority of those new register'd by those groups mention'd above
will be voting for Kerry. IMO......


Seriously, you would think so. But Bush has been polling better than Kerry among the 18-35 demo though. I get the sense that for every granny Kerry is scaring with his lies about Bush privatizing social security he's putting a younger voter in Bush's column.

But at the end of the day you could be right.

Ugly Duck
10-22-2004, 12:38 AM
Bush has been polling better than Kerry among the 18-35 demo though. I get the sense that for every granny Kerry is scaring with his lies about Bush privatizing social security he's putting a younger voter in Bush's column.

Poll: College students moving toward Kerry

By Michael Ludden CNN Thursday, October 21, 2004

ATLANTA, Georgia (CNN) -- College students say they are much more interested in politics this year, more likely to identify with a party and more likely to vote, a new Harvard University study shows.

And it appears a majority of those responding to the survey intend to vote for Sen. John Kerry.

The 10-point lead that Kerry held over President Bush in a March survey has grown to 13 percent.

While Bush's support among students has remained consistent, many undecided voters have moved into the Kerry camp.

In swing states, Kerry's lead is even higher, about 16 percent. Swing-state students also are saying they are more likely to vote than students in non-swing states.

http://www.cnn.com/2004/ALLPOLITICS/10/21/college.poll/index.html

RINGLEADER
10-22-2004, 01:01 AM
Poll: College students moving toward Kerry

By Michael Ludden CNN Thursday, October 21, 2004

ATLANTA, Georgia (CNN) -- College students say they are much more interested in politics this year, more likely to identify with a party and more likely to vote, a new Harvard University study shows.

And it appears a majority of those responding to the survey intend to vote for Sen. John Kerry.

The 10-point lead that Kerry held over President Bush in a March survey has grown to 13 percent.

While Bush's support among students has remained consistent, many undecided voters have moved into the Kerry camp.

In swing states, Kerry's lead is even higher, about 16 percent. Swing-state students also are saying they are more likely to vote than students in non-swing states.

http://www.cnn.com/2004/ALLPOLITICS/10/21/college.poll/index.html


I was relying on the real polling firms...not the Harvard University study.

Although the new ABC News poll shows Kerry a few points up among 18-30 year olds.

Taco John
10-22-2004, 01:08 AM
According to Newsweek, young voters are neck and neck, but are trending Kerry...

http://www.msnbc.msn.com/id/6273246/site/newsweek/

RINGLEADER
10-22-2004, 01:39 AM
According to Newsweek, young voters are neck and neck, but are trending Kerry...

http://www.msnbc.msn.com/id/6273246/site/newsweek/


Well the draft thing was a great political ploy even if it is a complete lie. If someone in the media would stand up and point out that a president can't enact a draft it probably would blunt that particular lie a bit. Even Charlie Rangel (the Dem author of the only draft bill to appear in congress in a generation) admits that someone saying there was a plan to bring back a draft - as Kerry does - would be lying.

Of course it would also help if Bush could explain this simple fact to people when it gets brought up. :rolleyes:

Taco John
10-22-2004, 03:43 AM
The 1190AM station here just cited a new poll from ABC giving Kerry a 19 point advantage in the youth vote...

Looking for it now...

DaKCMan AP
10-22-2004, 07:48 AM
Politics are very heavy around campus this semester. It is pretty close among students with a lean towards Kerry. Young voters generally aren't included in polls because we generally don't vote. I hope that changes this year. I truly believe young voter turnout will be higher than in '92 and could have an impact on the election.

On campus here they had a huge voter drive called "Chomp the Vote" and they registered 8000 student voters. They have a student-government funded limousine service that drives students to polls for early-voting. Everyone around campus is wearing Kerry-Edwards or Dubya stickers/buttons and politics is a major topic of discussion.

I hope its similar around the country..