Marc Fisher has an interesting article up on his blog this week about a new way of polling, and how it may impact things in future campaigns. I concur.
But one aspect of polling that bothers me is when results are intentionally skewed to favor certain position. That's what happened in the recent Virginia polling conducted by Moveon.org.
MoveOn conducted a poll in Virginia's 2nd Congressional District showing the race tied between Thelma Drake and Phil Kellam. To see more information on this poll, check out 750 volts which put the results into nice looking graphics.
While I concur that this will be the closest race in Virginia this year, the sleaziest way of conducting polling is to do what MoveOn did here. They put on a large TV ad buy into a vacuum, where nothing else political was going on, and then claimed to push the race into a dead heat.
With a small budget any competent ad agency can make the general public respond to a certain question the way they want them too, but it only lasts for a short period of time before it slowly wears off. In a political campaign, once the real ads begin, it wears off much more quickly and this type of stunt in the spring has almost zero impact by election day.
In short, MoveOn has probably hurt the Kellam campaign more than they helped them by announcing that he trailed by nine points prior to their ad campaign. If Phil began this tracking period trailing by nine points, that almost certainly means he trailed outside the margin of error in Virginia Beach City, which makes up about 3/4 of the Congressional District. In other words, all the spin of Kellam's popularity in the Beach and longtime ties there is not yet swinging his home city to his campaign. That one fact makes this a more uphill battle then I anticipated before.
Old Rating: Slight-Lean Republican
New Rating: Leans Republican
