Posted on 10/10/2004 3:00:59 PM PDT by timbuck2
Methodology
The vote preference questions in this poll were conducted by telephone Oct. 6-9, 2004 among a random national sample of 2,407 adults, including 2,030 registered voters and 1,589 likely voters. The results have a 2.5-point error margin among likely voters. ABC News and The Washington Post are sharing data collection for this tracking poll, then independently applying their own models to arrive at likely voter estimates. Sampling, data collection and tabulation by TNS of Horsham, Pa.
ABC News polls can be found at ABCNEWS.com at http://abcnews.com/Politics/PollVault/.
(Excerpt) Read more at abcnews.go.com ...
Can we assume the Wall Steet Journal might come up with a 52% advantage for Bush using "independent" models?
-T
We have no way of knowing which likely voter models are accurate and thus any criticism is pure speculation
well it's only a percent difference, that isn't an indication that the model they're using is outrageous.
Are you sure you are not looking at the old numbers for ABC? Maybe they haven't updated their website yet. Yesterday's ABC/Wash Post (both) had it at 50/46. Make sure you check everything before you start making accusations.
But .. it is an indication they are using a model which is weighted in favor of Kerry.
Cableguy,
I made that mistak this morning. It's the same data with two different results.
It's going to be fun to watch how ABC consistently sees better news for Kerry in the same data than the WashPost. I look forward to dueling interpretations for the next three weeks. Should be entertaining.
-T
It could also be rounding.
No, the ABC report you are referring to is using the data from yesterday, not today. The Wash Post poll just came out at 5pm. So when ABC updates its website, the two polls will match.
No ABC bias or screw-up here. You are just looking at ABC poll from yesterday, and Wash Post poll from today.
http://www.washingtonpost.com/wp-dyn/articles/A9363-2004Oct5.html
10/10: 1,438 likely voters; 1,912 self-identified registered voters Oct. 6-9.
http://abcnews.go.com/US/story?id=153043&page=2
The vote preference questions in this poll were conducted by telephone Oct. 6-9, 2004 among a random national sample of 2,407 adults, including 2,030 registered voters and 1,589 likely voters.
Sure looks like the same data to me with very different conclusions inferred.
-T
Looks different to me.
Same dates, same data.
-T
You are ABSOLUTELY right. I apologize.
Now you just opened the door to a whole new different question.
;-)
What were the results? I don't want to register...if you don't mind, thanks.
ping
51%-46% for WASPOST
50%-46% for ABCNEWS
And this is EXACTLY why I am thankful the Voter News Service system FAILED! It, and these polls, are purely for throwing elections.
Perhaps the Post is weighted toward Bush?
Who knows.
Not sensible to get worked up about a single point difference. There are all sorts of totally legitimate weightings that could account for it, since they are using a different model. Let this one go and save it for a true outrage, like mark halperin (sp?) telling ABC to go easy on Kerry.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.