Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: randog

Nah, they're simply not as far off as they usually are... LA Times would rather argue its a "dirty campaign" bounce than a convention bounce.

The decoder key to LA Times polls are their partisan sampling... This one shows a decent comparison of Republicans to Demonrats.


30 posted on 08/25/2004 8:18:20 PM PDT by dangus
[ Post Reply | Private Reply | To 2 | View Replies ]


To: dangus

Thanks for posting the sampling results. I refuse to register with them, so I was wondering if it was 14% Dim heavy like a few moths ago.


39 posted on 08/25/2004 8:20:08 PM PDT by RightthinkinAmerican (I oil, lube, service & remotely operate the Republican attack machine!)
[ Post Reply | Private Reply | To 30 | View Replies ]

To: dangus

Down Goes Kerry! Down Goes Kerry!!Down Goes Kerry!


78 posted on 08/25/2004 8:47:00 PM PDT by CWW (John Edwards -- Democrat and Whore Trial Lawyer)
[ Post Reply | Private Reply | To 30 | View Replies ]

To: dangus; Blood of Tyrants; randog; goldstategop; Reagan Man; over3Owithabrain; snooker; commish; ...
Normally one might expect that a MSM organization such as the LA Times would artificially push Bush's pre-convention poll numbers higher by over-sampling Republicans, first to minimize any post-convention bounce, second to make it appear that his numbers are going steadily south... However, it appears that the 'Swift Boat Vets for Truth' have had a large impact on Kerry's approval ratings in the national polls and it was suddenly necessary for the LA Times poll to shore up Kerry's poll numbers to keep him with in the margin of error of their August 21-24, 2004 poll.

Here are the LA Times August 21st-24th, 2004 published poll internal numbers for political affiliation for the three-way race between Bush, Kerry and Nader. These LA Times poll data are in a table format for easy reading.

This national three-way race poll which had 1,352 Registered voters, now shows Bush in the lead by 3% with Bush 47%, Kerry 44%, Nader 3%. The MoE is ±3.0% for this LA Times poll question number six.

Demographics Republican Democrat Independent   Poll  Numbers
Favor  Bush 92.00% 14.00% 41.00% Bush: 47.00%
Favor  Kerry 4.00% 78.00% 40.00% Kerry: 44.00%
Favor  Nader 2.00% 3.00% 7.00% Nader: 3.00%
Other/UnDec 0.00% 0.00% 2.00% Other/UnDec 1.00%
Do  not  know: 2.00% 5.00% 10.00% Do  not  know: 5.00%
  100% 100% 100%   100%

Source: LA Times Presidential poll - August 21-24, 2004 PDF file, released August 25th, 2004: Question #6, page 9 of 25.

Here below is the distribution of the 1,352 registered voters necessary to recreate the numbers show in the table directly above. Remember that the numbers below must be rounded to NO decimal places to match the above LA Times published data.

Registered 1352        
Demographics Republicans Democrats Independents Total  Raw  Votes Poll  Results
Bush 399 80 151 630 46.60%
Kerry 18 432 146 596 44.08%
Nader 8 14 24 46 3.40%
Other/UnDec 0 0 8 8 0.59%
Don’t  know 7 29 36 72 5.33%
Won’t  Vote 0 0 0 0 0.00%
Total  Raw  Votes 432 555 365 1352 100%
Percentage: 31.95% 41.05% 27.00%   100%
           
Bush 92.36% 14.41% 41.37%    
Kerry 4.17% 77.84% 40.00%    
Nader 1.85% 2.52% 6.58%    
Other/UnDec 0.00% 0.00% 2.19%    
Don’t  know 1.62% 5.23% 9.86%    
Won’t  Vote 0.00% 0.00% 0.00%    
Total: 100.00% 100.00% 100.00%   Demographics


So the partisan affiliation weightings that were used by the LA Times are indicated in the table below. Remember, because of LA Times rounding of their poll numbers, the below breakdowns might have a ±1.2% variance. That being said, it appears that this latest LA Times three-way race presidential preference poll has 9.1% more Democrats than Republicans in its sample population. A 2%-3% advantage for Democrats over Republicans might be a more realistic set of demographic splits.

    Results    
    (Weighting)    
`        
  Republican 31.95% Republican  
  Democrat 41.05% Democrat  
  Independent 27.00% Independent  
         
    100.0%    
         


So what would the LA Times poll look like with a more reasonable national sample of say 35.1%(R), 37.1%(D), 27.8%(I)? Look below and you would see that Bush would now have a 7.8% lead, which would be clearly outside the 'Margin of Error' of the LA Times poll. Kerry has been sinking swiftly...

Adjusted  Numbers
49.3%
41.5%
3.4%
5.9%
100.0%
 
 
New  Weighting
 
 
35.1%
37.1%
27.8%
 
100.0%
 


Hope this helps...

dvwjr

159 posted on 08/26/2004 4:46:38 AM PDT by dvwjr
[ Post Reply | Private Reply | To 30 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson