Technical Discussion of Robo-Rasmussen Polls
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
March 29, 2024, 10:10:52 AM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  Election Archive
  Election Archive
  2004 U.S. Presidential Election
  2004 U.S. Presidential Election Polls
  Technical Discussion of Robo-Rasmussen Polls
« previous next »
Pages: [1]
Author Topic: Technical Discussion of Robo-Rasmussen Polls  (Read 4559 times)
The Vorlon
Vorlon
YaBB God
*****
Posts: 4,657


Political Matrix
E: 8.00, S: -4.21

Show only this user's posts in this thread
« on: June 06, 2004, 12:42:50 PM »
« edited: July 07, 2004, 02:46:17 PM by The Vorlon »



When discussing the May 1-31 small sample Rasmussen Robo polls I floated the theory that because of the way his national survey was done, the following was likely to occur in the State level polls.

1)  States in the Eastern Time Zone would show a strong bias towards Kerry

2)  The Central Zones would be "ok" (maybe a bit pro-Bush in bias)

3)  The Mountain time + Pacific zones would swing hard to Bush in the polling to balance the EST bias toward Kerry.

Based on what we have seen, I am "semi" sure I am right.
Once all the states are in, we will revisit this thread.

I guess we will see when we get Arizona and a few more little ones.
Logged
millwx
Jr. Member
***
Posts: 402


Show only this user's posts in this thread
« Reply #1 on: June 06, 2004, 01:36:23 PM »

[img]Based on what we have seen, I a "fairly" sure I am right.
Once all the states are in, we will revisit this thread.
Couple of questions, Vorlon...

1) What about Pacific?  I presume, from your comments, that it would have the same bias as Mountain??
2) What, so far, makes you think your assessment is correct (I don't mean that to sound critical, I'm just not seeing it... the Kerry leads in ME and NJ look about right, and, on the other coast, OR *might* be slightly pro-Bush, but looks about right too... the only state which seem clearly to follow the suggested pattern is VA)?
Logged
The Vorlon
Vorlon
YaBB God
*****
Posts: 4,657


Political Matrix
E: 8.00, S: -4.21

Show only this user's posts in this thread
« Reply #2 on: June 06, 2004, 03:08:12 PM »
« Edited: June 06, 2004, 03:15:06 PM by The Vorlon »

[img]Based on what we have seen, I a "fairly" sure I am right.
Once all the states are in, we will revisit this thread.
Couple of questions, Vorlon...

1) What about Pacific?  I presume, from your comments, that it would have the same bias as Mountain??
2) What, so far, makes you think your assessment is correct (I don't mean that to sound critical, I'm just not seeing it... the Kerry leads in ME and NJ look about right, and, on the other coast, OR *might* be slightly pro-Bush, but looks about right too... the only state which seem clearly to follow the suggested pattern is VA)?

yes, will apply to pacific time zone as well.

BTW I think Maine is quite high at 19% - the other polls I have seen say 11% or so...

It's just a theory

http://www.boston.com/news/local/maine/articles/2004/06/05/kerry_outpolls_bush_in_maine/
Logged
CARLHAYDEN
Atlas Icon
*****
Posts: 10,638


Political Matrix
E: 1.38, S: -0.51

Show only this user's posts in this thread
« Reply #3 on: June 07, 2004, 09:59:01 AM »

From what I hear, Scott Rasmussen was mortified by the 2000 results and has recalibrated his polls to include a larger Democrat component than used in 2000.

I believe that this is a mistake.

In 2000 Gore had an unusually effective GOTV drive and Bush had essentialy none.

This year Bush appears to have an unsually effective GOTV drive in gear, and Kerry hasn't really got his act together on this one (its still five months till the election, plenty of time).

The bottom line is that if Kerry has a GOTV drive as effective as Gore had, due to the Bush efforts in this area this year, the 1-2 point advantage Gore accrued in 2000 will be eliminated, and if Kerry doesn't have such an effective drive then Bush may actually acrue a 2-4 point advantage (something Scott has not considered).

Also, Scott is deliberately omitting Nader from his results, which results in a slight skewing of the vote in Kerry's favor.

Finally, on a personal note, when I got home last night and played my messages on my answering machine, one was for a poll sponsored by a television station.  Wonder if I'll get a 'call-back.'  (Yes, I think Bush voters are being undersampled)
Logged
The Vorlon
Vorlon
YaBB God
*****
Posts: 4,657


Political Matrix
E: 8.00, S: -4.21

Show only this user's posts in this thread
« Reply #4 on: June 07, 2004, 10:03:16 AM »

From what I hear, Scott Rasmussen was mortified by the 2000 results and has recalibrated his polls to include a larger Democrat component than used in 2000.


I have emailed back and for with Scott a fair bit and while he has never (despite repeated attempts) told me the EXACT weighting he is using, reading between the lines and using the data he had published I am 99.9999% certain he is weithing at +3 to the Democratic side.

My best guess is 37/34/29 Dem/Rep/Ind, or in might be 38/35/27, but I am pretty sure it is the first one.
Logged
Fmr. Gov. NickG
NickG
Junior Chimp
*****
Posts: 8,150


Political Matrix
E: -8.00, S: -3.49

Show only this user's posts in this thread
« Reply #5 on: June 07, 2004, 10:08:24 AM »

From what I hear, Scott Rasmussen was mortified by the 2000 results and has recalibrated his polls to include a larger Democrat component than used in 2000.


I have emailed back and for with Scott a fair bit and while he has never (despite repeated attempts) told me the EXACT weighting he is using, reading between the lines and using the data he had published I am 99.9999% certain he is weithing at +3 to the Democratic side.

My best guess is 37/34/29 Dem/Rep/Ind, or in might be 38/35/27, but I am pretty sure it is the first one.

+3 Dem seems about right to me.  Do we know yet if he is weighting these state samples independently by Party ID?  Considering that Democrats in the South are much more likley to vote for Bush than Dems anywhere else, this could make a difference.  
Logged
CARLHAYDEN
Atlas Icon
*****
Posts: 10,638


Political Matrix
E: 1.38, S: -0.51

Show only this user's posts in this thread
« Reply #6 on: June 08, 2004, 09:31:21 AM »

Actually, there is a reason Scott doesn't provided the weighting formula.

Its not merely partisan, but racial weighting.

In 1996 (the first experiments). 1998 (rather sucessful in the Congressional results), and 2000 (spectacularly sucessful results) the Democrats herded large numbers of blacks to the polls.

While Blacks are generally Democrats, they are more inflexible than other Democrat racial groups.

For example, Bush is typically taking a fifth of the white self-described Democrats, but only about a twentieth of the black self-described Democrats.
Logged
MODU
Atlas Star
*****
Posts: 22,024
United States


Show only this user's posts in this thread
« Reply #7 on: June 08, 2004, 09:42:53 AM »


Vorlon (and everyone else),

A tangent off of your original post:  

One "the other" website (aka Zogby) I had floated my concerns regarding the media covering exit polls, and how it unfairly influences the Western states with their voting patterns.  While living in Hawaii, this was very obvious as people would stay at home watching the results coming in across the mainland before deciding to go to the polls to vote.

Especially after the fiasco last year with different networks calling Florida for both candidates, I would hope that they learned their lesson for this year.  While I love watching the unfolding election results come in, I would rather have the networks keep their mouths shut about poll figures until all the states had concluded and official results start flowing in.  This would lead to less tainting of the Western votes as well as possibly encourage more people to go to the polls and vote, since more doubt would be introduced on whether the candidate they'd like to see win is doing well or not.

Anyway folks, just a little issue I have with the media.  Debate it if you like.
Logged
CARLHAYDEN
Atlas Icon
*****
Posts: 10,638


Political Matrix
E: 1.38, S: -0.51

Show only this user's posts in this thread
« Reply #8 on: June 08, 2004, 09:52:20 AM »

Frankly there is no hard evidence that early reporting of exit polls influences the results in the West.

Moreover, with the rapid recent rise in absentee voting, a high percentage of the vote is cast where it cann't be 'polled,' upon leaving (exiting) the 'polling place.'

Just ask "govenor" bradley (of California) about this.

My major complaint is that televison spends too much time with moronic anchors droning on with nonsensical hogwash (rather is the worst of this) and so-called expert analysts who frequently don't know what they're talking about.

Instead, I'd like to see more raw data, clearly labled as to source (absentee vs. precinct), by county, etc.
Logged
CARLHAYDEN
Atlas Icon
*****
Posts: 10,638


Political Matrix
E: 1.38, S: -0.51

Show only this user's posts in this thread
« Reply #9 on: July 07, 2004, 11:40:45 PM »

While you have established that the Democrat component of Scott's sample is slightly higher than you originally suspected, did you ever get the data on the black percentage of the sample?

If it is a high as I suspect, then Scott is overstating (compared to what my models suggest for the 2004 turnout) that Kerry is a point or two higher, and Bush a point or two higher.

If this is the case, then a Rasmussen tie, would actually mean a Bush lead of two to four points.

Also, as you have previously correctly noted, summer polls tend to favor Democrat candidates.  I have gotten home twice in the past couple of months and found messages on my answering machine for a robo poll from a local TV station.  

Logged
The Vorlon
Vorlon
YaBB God
*****
Posts: 4,657


Political Matrix
E: 8.00, S: -4.21

Show only this user's posts in this thread
« Reply #10 on: July 08, 2004, 12:14:06 AM »
« Edited: July 08, 2004, 12:15:25 AM by The Vorlon »

While you have established that the Democrat component of Scott's sample is slightly higher than you originally suspected, did you ever get the data on the black percentage of the sample?

If it is a high as I suspect, then Scott is overstating (compared to what my models suggest for the 2004 turnout) that Kerry is a point or two higher, and Bush a point or two higher.

If this is the case, then a Rasmussen tie, would actually mean a Bush lead of two to four points.

Also, as you have previously correctly noted, summer polls tend to favor Democrat candidates.  I have gotten home twice in the past couple of months and found messages on my answering machine for a robo poll from a local TV station.  



in his National sample Rasmussen stratifies for race at:

White 81%
Hispanic 7%
Black 12%

Actual 2000 turnout based on US Census bureau estimates was:

White + Asian 81.2%
Hispanic 5.4%
Black 11.6%

(does not = 100% due to "other")

The Hispanic number is a crapshoot - who knows in 2004?




Logged
CARLHAYDEN
Atlas Icon
*****
Posts: 10,638


Political Matrix
E: 1.38, S: -0.51

Show only this user's posts in this thread
« Reply #11 on: July 08, 2004, 10:45:01 AM »

Thank you for the info.

It pretty much confirms what I was saying.

Scott's percentage of respondents who are black is slightly higher than the actual turnout in 2000, in which election (due to extraordinary get out the vote drive by Democrats) the percentage of black voters was far higher than usual.

If black turnout returns to the normal percentage of the electorate, and if we preseme a normal party distribution of the vote by race, then Scott will have overstated Kerry's support by just approximately two per cent, and understated Bush's support by the same extent.
Logged
Fmr. Gov. NickG
NickG
Junior Chimp
*****
Posts: 8,150


Political Matrix
E: -8.00, S: -3.49

Show only this user's posts in this thread
« Reply #12 on: July 08, 2004, 11:00:30 AM »

While you have established that the Democrat component of Scott's sample is slightly higher than you originally suspected, did you ever get the data on the black percentage of the sample?

If it is a high as I suspect, then Scott is overstating (compared to what my models suggest for the 2004 turnout) that Kerry is a point or two higher, and Bush a point or two higher.

If this is the case, then a Rasmussen tie, would actually mean a Bush lead of two to four points.

Also, as you have previously correctly noted, summer polls tend to favor Democrat candidates.  I have gotten home twice in the past couple of months and found messages on my answering machine for a robo poll from a local TV station.  



in his National sample Rasmussen stratifies for race at:

White 81%
Hispanic 7%
Black 12%

Actual 2000 turnout based on US Census bureau estimates was:

White + Asian 81.2%
Hispanic 5.4%
Black 11.6%

(does not = 100% due to "other")

The Hispanic number is a crapshoot - who knows in 2004?


Is Rasmussen not counting Asians?  They are one of the fastest growing minorities, and are now a large enough group that they should be counted on their own.  

I expect the % of black and especially Hispanic voters to be higher than in 2000, simpy because their % of the population has increased.  Not much higher, but Rasmussen's figures look about right, assuming the Asian thing is somehow explained.
Logged
The Vorlon
Vorlon
YaBB God
*****
Posts: 4,657


Political Matrix
E: 8.00, S: -4.21

Show only this user's posts in this thread
« Reply #13 on: July 08, 2004, 11:21:46 AM »

While you have established that the Democrat component of Scott's sample is slightly higher than you originally suspected, did you ever get the data on the black percentage of the sample?

If it is a high as I suspect, then Scott is overstating (compared to what my models suggest for the 2004 turnout) that Kerry is a point or two higher, and Bush a point or two higher.

If this is the case, then a Rasmussen tie, would actually mean a Bush lead of two to four points.

Also, as you have previously correctly noted, summer polls tend to favor Democrat candidates.  I have gotten home twice in the past couple of months and found messages on my answering machine for a robo poll from a local TV station.  



in his National sample Rasmussen stratifies for race at:

White 81%
Hispanic 7%
Black 12%

Actual 2000 turnout based on US Census bureau estimates was:

White + Asian 81.2%
Hispanic 5.4%
Black 11.6%

(does not = 100% due to "other")

The Hispanic number is a crapshoot - who knows in 2004?


Is Rasmussen not counting Asians?  They are one of the fastest growing minorities, and are now a large enough group that they should be counted on their own.  

I expect the % of black and especially Hispanic voters to be higher than in 2000, simpy because their % of the population has increased.  Not much higher, but Rasmussen's figures look about right, assuming the Asian thing is somehow explained.

The US census lumps them in as "Asian/Pacific" and they are about 2.5% or so.

The more or less vote like hispanics.

In 2000 Hispanics broke for Gore 61/38, Asians broke 62/37 for Gore.

Rasmussen does not take Asians into account directly, but is a bit high on both Blacks and Hispanics which should "about" compensate.

Asian is a very broad term however.

Japanese vote most Republican, while Koreans vote Democratic.

(Unrelated, but if you want to see real racial division, get the Japanese talking about the Koreans or the other way around) Lumping all asians together is just wrong, but at 2.5% they are still pretty small to be dividing up the pie as it were.
Logged
Fmr. Gov. NickG
NickG
Junior Chimp
*****
Posts: 8,150


Political Matrix
E: -8.00, S: -3.49

Show only this user's posts in this thread
« Reply #14 on: July 08, 2004, 11:25:33 AM »

While you have established that the Democrat component of Scott's sample is slightly higher than you originally suspected, did you ever get the data on the black percentage of the sample?

If it is a high as I suspect, then Scott is overstating (compared to what my models suggest for the 2004 turnout) that Kerry is a point or two higher, and Bush a point or two higher.

If this is the case, then a Rasmussen tie, would actually mean a Bush lead of two to four points.

Also, as you have previously correctly noted, summer polls tend to favor Democrat candidates.  I have gotten home twice in the past couple of months and found messages on my answering machine for a robo poll from a local TV station.  



in his National sample Rasmussen stratifies for race at:

White 81%
Hispanic 7%
Black 12%

Actual 2000 turnout based on US Census bureau estimates was:

White + Asian 81.2%
Hispanic 5.4%
Black 11.6%

(does not = 100% due to "other")

The Hispanic number is a crapshoot - who knows in 2004?


Is Rasmussen not counting Asians?  They are one of the fastest growing minorities, and are now a large enough group that they should be counted on their own.  

I expect the % of black and especially Hispanic voters to be higher than in 2000, simpy because their % of the population has increased.  Not much higher, but Rasmussen's figures look about right, assuming the Asian thing is somehow explained.

The US census lumps them in as "Asian/Pacific" and they are about 2.5% or so.

The more or less vote like hispanics.

In 2000 Hispanics broke for Gore 61/38, Asians broke 62/37 for Gore.

Rasmussen does not take Asians into account directly, but is a bit high on both Blacks and Hispanics which should "about" compensate.

Asian is a very broad term however.

Japanese vote most Republican, while Koreans vote Democratic.

(Unrelated, but if you want to see real racial division, get the Japanese talking about the Koreans or the other way around) Lumping all asians together is just wrong, but at 2.5% they are still pretty small to be dividing up the pie as it were.

Does this mean he is not polling any Asians or people who refuse to disclose their race?  His White+Black+Hispanic %'s add up to 100%.

Also, I don't see how lumping Asians together is any worse than lumping all Hispanics together....Cubans and Puerto Ricans, for example.

Logged
The Vorlon
Vorlon
YaBB God
*****
Posts: 4,657


Political Matrix
E: 8.00, S: -4.21

Show only this user's posts in this thread
« Reply #15 on: July 08, 2004, 11:31:00 AM »
« Edited: July 08, 2004, 11:37:17 AM by The Vorlon »


Does this mean he is not polling any Asians or people who refuse to disclose their race?  His White+Black+Hispanic %'s add up to 100%.

Also, I don't see how lumping Asians together is any worse than lumping all Hispanics together....Cubans and Puerto Ricans, for example.


Sorry if I was unclear.. my "understanding" from Rasmussen is that he sets self identified Blacks to 12%, Hispanics to 7%, and "Other" (mostly whites) to 81% - so "other/Asian" gets lumped into the 81%

In don't know if Scott is deliberately vague or just a bad communicator, or I just am bad at reading, but all his answers are always a tad unclear to me.

Re Asians - yes lumping the together for plling purposes is just fine - similar as you point out to Hispanics.

I was just saying that in terms of thinking about them as "voting blocks" - The are not like Blacks which vote massively Democrat, but rather various asian nationalities vote in very different ways.

Logged
CARLHAYDEN
Atlas Icon
*****
Posts: 10,638


Political Matrix
E: 1.38, S: -0.51

Show only this user's posts in this thread
« Reply #16 on: July 08, 2004, 11:40:42 AM »

First, I stongly suggest that you don't get the Japanese talking about the Koreans, or the Koreans talking about the Japanese.  They do get a little bit heated (personally, I sympathize with the Koreans).

Second, the voting patterns of Asians in 2000 were heavily influenced by two factors.  In Hawaii, Asians vote Democrat since the Republican party in that state has not treated them well historically (@#$% country club Republicans).  In California, a lot of Koreans vote Republican in Presidential elections because the Democrat party seems more interested in other racial groups.  Further, the 2000 election had a higher than normal Democrat vote for President as Gore's campaign worked Asian groups better than Bush's campaign did.

Third, Bush is picking up support (relative to 2000) among hispanics.  While Kerry will probably carry this group nationally (largely due to the overwhelming advantage among Puerto Ricans), I expect the overall advantage to be a historically low 60-40, and possibly as good as 55-45.  

Fourth, while Native Americans (indians) only account for about one per cent of the electorate, Gore did unusually well among this group in 2000 because (again), his campaign worked this group, and Bush did not.  Expect Bush will take a larger percentage of this group this year than he did in 2000.

Logged
Pages: [1]  
« previous next »
Jump to:  


Login with username, password and session length

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.237 seconds with 14 queries.