UPDATE: This is a good poll.
I just received the cross-tabs from Gravis Marketing and the sample was 39% Democrats, 36% Republicans and 23% Other (Independents.) My only criticism is that there were too many Republicans and Democrats and too few Independents. But since the sample polled 3% more Democrats than Republicans and the relatively small number of Independents (23%) broke for Allen by a double digit margin – which is consistant with other polls, this survey actually understates the vote for Allen. Essentially, the Allen lead should be a bit more than 5 points.
I am comfortable saying that this Gravis Poll passes the Virginia Right! test for trust.
There is a new poll by Gravis Marketing that shows both Virginia Senate Candidate George Allen up over his Democratic opponent Tim Kaine by 5 points – 48% – 43% and the poll also shows Republican Mitt Romney leading Barack Obama 49% to 44%.
Now before my Conservative friends get all giddy and start hiring workers again thinking the worst is over, or my Liberal friends begin looking for work again thinking that the free ride is ending, just cool your jets.
While the results are definitely looking good for the right, there is not a shred of demographic data included in this poll. So even though the results are pretty close to the “normalized” results I calculated in the Presidential race here, I am unable to declare this poll credible without the numbers.
I have written to the proprietors of Gravis Marketing requesting the internals in this poll. Until I get those (and I am not sure I will) I will rank this poll as suspect.
My feeling is that a poll should proudly proclaim their figures behind the results or they are hiding something.
I am not saying Gravis is conducting bogus polls, but I simply have no information on this outfit, and this is a new entry into the polling arena.
As soon as I get more information on their methods I will update this post. If they fail to provide their numbers, I can’t give this poll any authoritative status.