In a poll of 625 registered voters, PilotOnline is reporting that George Allen leads Jim Webb 47% to 43%. The poll has a margin of error of +/- 4%.
I don’t know what’s up with Mason-Dixon using registered voters as opposed to likely voters. In any event, I just got polled by M-D in the last hour on this race and the 2nd CD race. Will be interesting to see what the results of that poll are.
UPDATE: The Daily Press has its take on the poll and I’m wondering what’s going on. Jim Hodges reports that it was likely voters, not registered voters.
UPDATE2: The AP story says registered voters.
UPDATE3: My source says it was likely voters.
VIVIAN J. PAIGE | All Politics is Local 
Looking at the graphic from the poll, it does say registered voters, BUT at the bottom it says that all of the voters polled said that they are likely to be voting in the election. That might cause some reporters to label it as ‘likely voters’. That is not what likely voters usually means. “Likely voters” is not a self identified category, but based on voting patterns in previous elections. To me, the AP story is correct, the M-D poll is registered voters, not likely voters. That might explain why Allen’s numbers are higher in the other recent polls. Allen does better with likely voters than registered voters. But, the trend is interesting. Allen is up 4% over the last Mason Dixon poll on September 29. Webb stayed exactly the same. Undecided voters went down 4%, between the polls, the exact percentage that Allen picked up. It does not appear that the undecideds are breaking for the challenger as the democrat myth insists.
The graphic does show likely voters. I think the M-D poll is of 625 registered voters, all of whom were likely voters. Richmond War Room says that it was likely voters and explains how M-D ends up with registered voters = likely voters.
Actually, “likely voters” is self-identified and is not taken from the voting records. One of the first questions asked is how likely you are to vote. (I actually have experience with constructing poll questions and working with a pollster.)
I thought likely voters were those who had voted in the past elections. Not taken from voting rolls, but asking the voter ‘did you vote in 2004, did you vote in 2002’, etc. The question is not ‘are you likely to vote in this election’ but is a series of questions to determine voting patterns. For instance, a registered voter who says he’s likely to vote this time, but has missed voting in 3 of the last 4 elections, would not be considered a ‘likely voter’. Or someone who has never voted in an off year election. Or someone who can’t remember when he last voted. For most pollsters simply stating that you are likely to vote, or somewhat likely to vote, in this election, does not make you a ‘likely voter’. There must be some further proof of voting patterns. M-D seems only to ask the question ‘do you intend to vote in this election’ to determine ‘likely voters’. That’s not how it works in other polls. The pollster determines, based on a series of questions, if the voter is a ‘likely voter’ in this election.
Nope. There is no other proof other than what the voter says. All the polls use the same criteria: they ask the voter how likely they are to vote in the election. If a voter says they are not likely to vote, then they are not included in the figures.
The phone calls are made from the registered voter list (unless it is a random poll, like the one that was done after the last debate).
As I understand it, the voter files maintained by the registrars don’t keep much voter history. (They do need to know if you have voted in the last # of elections so that they can purge the files.) The files maintained by the respective parties is where the detail on voter history is kept.
Have you never been polled, Jane? I get polled regularly and have yet to have any questions asked to elicit my likelihood of voting other than the very first one (how likely are you to vote?).
Vivian, how is it that you get polled so often? I’ve never been polled in my entire life. Is it just happenstance? (P.S. Thanks for the sign 🙂 and I hope you are feeling better.)
I guess I get polled because I vote 😉 That’s the only reason I can come up with. I’ve been polled more this election season that I ever have, though. They must be doing a lot of polling this year.
I’m much better, thanks! And glad you got the sign. Let me know if you or anyone else you know wants one.
I have voted in every single election that has been held in the close-to-40 years I have been eligible to do so. I haven’t changed phone numbers, I’m in the phone book. That’s what I don’t get. No one wants to know what *I* think. 😦
I will admit to never having run for office, though. 😉
Well, then it’s not about who votes. I think it’s a fluke I’ve been polled so many times this year. I’ve never been polled this many times in one election cycle.