Posted on 09/05/2004 4:01:55 AM PDT by dvwjr
Analysis of recent Newsweek/PSRAI post-RNC convention polling data, with derived and revised Kerry and Bush post-convention 'bounces' and poll leads. Comments welcome...
Keep me on! :^D
Thanks for the ping. Keep me in your list.
Fascinating analysis. I've long thought that every new poll should have to prominently discuss the "weighting" formulae used in calculating results.
Which leads to a question. How do they arrive at the "weighting" formula (party affiliation). Is it based on an average of their last few polls? What if the electorate is swinging away from one party and towards another?
I would think an interesting timeframe to look at would be the months immediately before and after 9/11. Did more voters begin identifying themselves as Republicans? For discussion's sake, let's say Republican I.D. went from 38% to 47% in a 3 week period. How would that change in party affiliation be reflected in polls? If the firm's policy is to use averages that occurred BEFORE the defining event, wouldn't the new polls underrepresent Republicans?
I'm asking this because I think there's a possibility that Kerry's miserable August has resulted in a slightly higher Republican self-identification. I just don't know how that would be reflected in the weighting. Can you help me here?
BTTT and a bookmark for me ;^)
Thank you for the clear and cogent analysis! Please keep me on your ping list!!
The entire exercise, from Newsweek's attempt to set up a Kerry comeback to your ability to net out the results, is fascinating.
This is quit an extensive write-up, but for what? I’m no statistician, but this reminds me of the phrase garbage in garbage out.
Please tell me how in a country about evenly divided between Republicans and Democrats, a poll that included 374 Republicans and 300 Democrats is valid.
This is a 20% over-sampling! If this were reversed, and Democrats were over-sampled, would you take this seriously?
Let's give this some legs and get it out everyway we can, telling people that the current polls appear to look skewed on purpose by the leftists who take these polls, only to come back to show Kerry gaining big on Bush.
Please remove me from the ping list that you use.
Just because I’m paranoid doesn’t mean they’re not all out to get me.
I will have to bookmark and print, for further study, your poll analysis. Your talent to analyze, imho, is incredible, dv!
Now, for us liberal arts majors (English/Psychology) lol, please tell me, the numbers you posted at the very bottom chart, is this the REAL numbers? (Bush got a 8.93% increase in those who would vote for him? and Kerry lost 6.79%, for those who would vote 9/2?) And then the convention bounce for Bush is a shock and awe of 18.73%, with Kerry bombing to -2.17%? (sorry for sounding stupid, numbers is not my forte.)
Okay, okay… now that I actually read your analysis, I see that you’re just as critical as I of the over-sampling and are adjusting the results.
The nom-paranoid explanation of the party distribution is that some right-leaning Independents declared themselves as Republicans after deciding to vote for Bush.
please add to ping list
I'm guessing that Newsweek is gambling that more people decided on the spot that they were Republicans after the speech, which seems reasonable to me (although not 37%). If their bounce showed Bush leading by 12% and dvwjr's showed Bush leading by 8% I'd guess the truth is somewhere in between.
Yeah, DU suffers from numerous problems, the lack of any intellectual honesty leads the list. Their hatred and vulgarity also diminishes their site, although that is the part that makes their site fun to visit as they implode. In their eyes, their spin is the gospel truth and our spin are outright lies. Democratic spin is usually more factually-challanged.
That’s not only non-paranoid, it’s unnatural . 10-20 percent may change candidates overnight, but not party affiliations.
Looks like you did a really thorough analysis here. Well done. It's one thing when one poll comes out with an unusual result, but it's quite another when 2 polls come out showing the same thing. Yes, Zogby and Rasmussen are showing different stuff at the moment, but the full effect of President Bush's speech is still unknown.
The bottome line is: President Bush is AHEAD, not behind like his dad was in 1992.
Can everybody agree with me on one thing: 2004 is NOT A REPEAT OF 1992???????
thanks for the breakdown, please keep me on the list
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.