|
Edited on Tue Sep-09-03 09:38 PM by papau
Zogby's 9/4-5 poll of 1,013 likely voters (+/-3% error margin) had Bush's approval rating dropping to just 45 percent, but Cook is upset that this is lower than his 9/2-4 Ipsos Public Affairs/Cook Political Report poll of 766 registered voters (+/-3.6% error margin) showing a 52 percent approval rating, and the CNN/Time poll of 9/3-4 of 1,003 adults (+/-3% error margin) that also showed Bush's approval rating at 52 percent. He asserts the use of a "do you approve or disapprove of the job George W. Bush is doing as president" poll always polls 5 to 8 points higher than the Zogby method of asking folks to rate Bush's performance on a scale of "excellent, good, fair or poor." with the "excellent" and "good" categories together called "positive", while the "fair" and "poor" responses are deemed "negative." Cook also notes Zogby's use of likely voters, while most other media pollsters generally sample all adults this far from an election - BUT COOK does Not note that Bush does better with "likely" than he does with "all".
Finally he disputes the trend line, citing the 9/10/01 Gallup of 51%, ignoring the Aug 26 01 of 57% was where Gallup was on trend, and then compares to Gallups current 59%.
In previous emails Cook has stated that one should not compare between polling organizations - and just look for trends with in one groups poll. Well doing that we find Zogby has Bush Job Performance as (9/2003) 45 to 54, (8/2003) 52 to 46, (3/2003) 54 to 45, (9/2002) 64 to 36, (9/2001) 82 to 17, (8/2001) 50 to 49, and on 1/2001 a postive- but a bit uncertain - 42 to 36.
Looks like a trend to me!
|