New Zogby Poll: Beebe 49 Hutchison 45, MOE +- 4.2
The poll was conducted for the Wall Street Journal from Oct. 10th to Oct. 16th. You may have to click on the "gubinatorial races" tab and find Arkansas for the results. This poll did include Lendall and Bryan. This poll is highly "gender gapped", with a 30 point Beebe lead among women and a 23 point Hutchinson lead among men.
Going into the home stretch it looks like Beebe has a slight edge but it could go either way. The difference is within the margin of error. I would like to know the balance of single women vs. married women polled. Oversampling single women would tilt it the dems way. Over sampling married women with children would do the opposite. Maybe they got it about right. All I know is that the GOP needs a top-notch swing voter targeting and base turnout operation to win this race. They think they have one. We will know after 3 weeks.
Going into the home stretch it looks like Beebe has a slight edge but it could go either way. The difference is within the margin of error. I would like to know the balance of single women vs. married women polled. Oversampling single women would tilt it the dems way. Over sampling married women with children would do the opposite. Maybe they got it about right. All I know is that the GOP needs a top-notch swing voter targeting and base turnout operation to win this race. They think they have one. We will know after 3 weeks.
posted by Mark Moore (Moderator) at Saturday, October 21, 2006
3 Comments:
You neglect to mention this is a "Zogby Interactive" poll -- an internet poll where the respondents are self-selected. It is simply not reliable, just as it wasn't reliable in early November 2004 when the Zogby Interactive polls predicted a John Kerry victory.
Go to www.mysterypollster.com and compare the polls on the Arkansas governor's race. All polls generally track the others, except the Zogby Interactive polls. This follows a trend -- look at some of the other polls for congressional races. Zogby's internet polls consistently show the races closer, with a Republican edge, than the telephone interview polls.
Of course, it is possible that Zogby's internet polls are right, and all the others are wrong. Based on history, in both the 2002 and 2004 races when Zogby's predictions were wrong, I doubt it.
That is correct. All the others are wrong. This board has had in-depth articles explaining why that is. If the media floats 10 polls with bad methodology then all ten are wrong. If they double it and do 20 with flawed methodology then all 20 are wrong. They are just numbers. When you look at how they got those numbers you will see they are only good for two things
1) detecting trends
2) propaganda to discourage Republicans.
Your "in-depth articles" depend on the idea that more than 30% of the votes in this election will come out of the 3rd congressional district -- they won't. Wait and see. 27% at most, and even with that number Beebe still is up by 10 points. Also, you haven't explained why you believe the Zogby Interactive polls are correct, especially given Zogby's questionable track record with these self-selected polls.
BTW - new poll from Talk Business this morning showing Halter pulling away, and cracking the vital 50 percen mark. My prediction is Holt gets 47% of the vote, 3 points better than he did against Blanche, and just about the same percentage John Kerry got in '04.
Post a Comment
<< Home