New Survey USA Polls just out

[CARLHAYDEN]

Mo.

Bush  +2

This surprised me.  Bush leading nationally by 5 should have been up by more.  Heck, he won by 3 last time with the national popular vote tied.


Penn

Kerry +2

This doesn't surprise me.


Ind, Kentucky, Kansas all Bush as expected.  Like I said, Mo. really is surprising.

SUSA leans Democrat in their polling.

Look at their record.

They're pretty consistently at the outer edges of credibility in their polls, consistently in the same direction (more Democrat than the other credible polls).

[CARLHAYDEN]

Mo.

Bush  +2

This surprised me.  Bush leading nationally by 5 should have been up by more.  Heck, he won by 3 last time with the national popular vote tied.


Penn

Kerry +2

This doesn't surprise me.


Ind, Kentucky, Kansas all Bush as expected.  Like I said, Mo. really is surprising.

SUSA leans Democrat in their polling.

Look at their record.

They're pretty consistently at the outer edges of credibility in their polls, consistently in the same direction (more Democrat than the other credible polls).

It really is tiring to here from you that the polls that may be 'surprising' or do not favor Bush the way you want them fit into your world are somehow.....Democratic leaning. If that were the case, why is not PA going Kerry by +5%???

The poll results are a bit closer than I expected as well. However, without delving too far into SUSA's methodology, they did have a pretty accurate record during the Democratic Primaries which, I'm sure most would agree, are tougher to poll for than a general election.

Perhaps you did not read my post very carefully or do no comprehyend the difference between polling for primaries and polling for general elections.

I repeat again, look at their record, they are NOT very accurate in general elections as they overload their polls with Democrats.

I have previously given examples of this (remember Nevada)?

[CARLHAYDEN]

Mo.

Bush  +2

This surprised me.  Bush leading nationally by 5 should have been up by more.  Heck, he won by 3 last time with the national popular vote tied.


Penn

Kerry +2

This doesn't surprise me.


Ind, Kentucky, Kansas all Bush as expected.  Like I said, Mo. really is surprising.

SUSA leans Democrat in their polling.

Look at their record.

They're pretty consistently at the outer edges of credibility in their polls, consistently in the same direction (more Democrat than the other credible polls).

It really is tiring to here from you that the polls that may be 'surprising' or do not favor Bush the way you want them fit into your world are somehow.....Democratic leaning. If that were the case, why is not PA going Kerry by +5%???

The poll results are a bit closer than I expected as well. However, without delving too far into SUSA's methodology, they did have a pretty accurate record during the Democratic Primaries which, I'm sure most would agree, are tougher to poll for than a general election.

First, you seem to be mixing up my post with the statements of the poster to whom I replied.

I was not suprised by SUSA's results.

Oh, and as to historical results, I suggest you check the record of the Literary Digest poll.

[CARLHAYDEN]

I'm sorry it's "tiring" for you.  

In 2000, Bush won Mo by 3, tied nationally.
In 2004, Bush up 2 there, yet up nationally by 5.

That is a bit surprising.  So what?

Easy agcat.....my fatigue was not directed at you  
Besides, all those numbers you cite ARE within the MOE

But back to SUSA - I'm not specifically defending them ...but.....those numbers for MO are about where they were pre RNC. As for accuracy of SUSA........

NH               SUSA              Actual
Kerry             33                   38
Dean             28                   26
Edwards       14                   12
Clark             12                   12
Lieberman      7                     8

Oklahoma
Clark             29                   30
Edwards       27                   29
Kerry            26                    27
Dean              7                     4

MO
Kerry            44                   50
Edwards       20                   24
Dean            15                    8
Clark              6                    4

TN
Kerry             35                   41
Edwards        25                  26
Clark             24                   23
Dean               9                    4

AZ
Kerry            34                   42
Clark            28                   26
Dean            18                   14
Lieberman      7                    7
Edwards        7                     7

At first glance at Kerry's numbers one might say not real accurate. Yet much of Kerry's higher numbers are the result of the 'piling on' effect once he had won New Hampshire and people were voting for who was obviously going to win. What struck me most about their polling was how accurate they were with the other candidates results. Not bad for primary polls from a firm that 'churns' out the polls.......

First, with the exception of the Oklahoma example, in all instances SUSA was outside the MOE for the winner.  Not a real good recommendation.

Second, SUSA likes polling Democrats, and in the general election polls loads their polls with extra Democrats (see their poll of Nevada of about eight weeks ago).

[CARLHAYDEN]

S-USA showing no bounce is odd.  They appear to have had a bit of an off round of polling.

Oh, mddem, your history of S-USA is not so good a way to refute Carl's claim that they have a pull toward the democrats.  There are no republicans to be pulling away from in those.

The problem is that Carl Hayden will cast doubt on 'any' poll that shows some advantage for Kerry or not bad news for Kerry. He has been doing that since long ago. The fact that he doesn't have a state or any avatar and he posts all the time attacks on Kerry disguised in some supposed polling expertise, makes him in my opinion the owner of the: 'stealth troll' award.

  The problem with Carl is he is a pure partisan, and I take his posts with a grain of salt.

Perhaps you are so new you do not remember the number of polls showing Bush farther ahead than I believe was the case.

To cite a few examples:

In June Harris produced a survey having Bush ahead by ten points.  Although I respect Lou and his organization, I stated at that time the poll was way out of line.

A few days ago when both Time and Newsweek showed Bush with a lead of 10 points or more, I criticized both of those polls as excessive.

Another survey research firm suggested that Bush had a lead of 16 points in Arizona (about a week ago), I stated that I though that was excessive.

A few days ago another survey research firm showed Bush with a large lead in Missouri, I criticized that.

Also, another survey research firm had a poll showing Bush ahead by 19 points in North Carolina.  Again I stated that was excessive.

Enough examples for you?

[CARLHAYDEN]

SUSA leans Democrat in their polling.

Look at their record.

They're pretty consistently at the outer edges of credibility in their polls, consistently in the same direction (more Democrat than the other credible polls).

They nailed 25/28 races in 2002.

Please be so kind as to post the SUSA polls you are claiming they "nailed."

[CARLHAYDEN]

Thank you for your posting.

I suggest you check out the Texas Senate race where Cornyn actually won!  I think you transposed the figures.

So, in 17 of the races, the error was 4 per cent or more.  In 16 of the 17 races with this error rate, the Democrat received a higher SUSA percentage than actual election results.

Of the 19 elections where SUSA had projections of 3 per cent of less, they over projected the Democrat candidate vote in 12, the Republican candidate in 3, and 'nailed it' in four.

Taking all thirty six races together, SUSA, in the aggregate tended to favor the Democrats candidates by 2.8%

As such, they are not out of the ballpark, but the clearly tend to favor Democrat candidates.

Enough said?

[CARLHAYDEN]

SUSA leans Democrat in their polling.

Look at their record.

They're pretty consistently at the outer edges of credibility in their polls, consistently in the same direction (more Democrat than the other credible polls).

They nailed 25/28 races in 2002.

Come on Tweed.

I want to see you again boast about the accuracy of SUSA.

[CARLHAYDEN]

So you don't dispute the MOE in the SUSA polls is rather large, or that SUSA pretty consistently favors Democrats relative to election results?

BTW, their average (mean) error with 4.7%.

[CARLHAYDEN]

Thank you.

I understated their mean error rate by one one hundredth of a per cent.

The point of my original posting on this thread was that the error rate was NOT equally distributed (i.e. they were not as equally likely to overstate the Republican vote as they were to overstate the Democrat vote).  With your data, SUSA leans Democrat on the average by 2.6%.

Now, every pollster to my mind is entitled to have an occasional error, so lets throw out the one race where SUSA most overstated the Republican vote, and one race where they most overstated the Democrat vote.

The result is the SUSA now leans Democrat by 3.7%.

I think I made my point.

Do you concur?

[CARLHAYDEN]

Mo.

Bush  +2

This surprised me.  Bush leading nationally by 5 should have been up by more.  Heck, he won by 3 last time with the national popular vote tied.


Penn

Kerry +2

This doesn't surprise me.


Ind, Kentucky, Kansas all Bush as expected.  Like I said, Mo. really is surprising.

SUSA leans Democrat in their polling.

Look at their record.

They're pretty consistently at the outer edges of credibility in their polls, consistently in the same direction (more Democrat than the other credible polls).

It really is tiring to here from you that the polls that may be 'surprising' or do not favor Bush the way you want them fit into your world are somehow.....Democratic leaning. If that were the case, why is not PA going Kerry by +5%???

The poll results are a bit closer than I expected as well. However, without delving too far into SUSA's methodology, they did have a pretty accurate record during the Democratic Primaries which, I'm sure most would agree, are tougher to poll for than a general election.

Also suggest you check out my post (no. 1080) on page 2 of this thread.

[CARLHAYDEN]

Facts are useful when discussing a pollster

Vorlon,

BTW, got my letter for charitable contributions from one charity the other day.

How much is the cut off for your site, and to where should I send a copy?

[CARLHAYDEN]

S-USA showing no bounce is odd.  They appear to have had a bit of an off round of polling.

Oh, mddem, your history of S-USA is not so good a way to refute Carl's claim that they have a pull toward the democrats.  There are no republicans to be pulling away from in those.

The problem is that Carl Hayden will cast doubt on 'any' poll that shows some advantage for Kerry or not bad news for Kerry. He has been doing that since long ago. The fact that he doesn't have a state or any avatar and he posts all the time attacks on Kerry disguised in some supposed polling expertise, makes him in my opinion the owner of the: 'stealth troll' award.

Oh, I posted the analysis of the SUSA numbers from 2002 (Vorlon was kind enought to post the numbers).

I suggest you read and and think before you post.