Replies

I work this campaign twice a week and it seems that we are indeed poll takers for the most part. The phone system here in FL is impressive. The dialing is automatic, responses are keyed into the phone and two pushed buttons leave a message and free you up the instant you hear the first words from an answering machine. If you put your head down and just work, you can easily make 200 calls an hour.

Every call that is registered as a wrong number or a disconnect along with other unknown phone number addresses is fed into a different system that generates “door knocking” lists. We are sent out with scripts and bubble result pages that are scanned once we return to the office. Of course we go out with a bunch of literature for unanswered doors andhand outs and we say a few kind words to try to convince and undecideds we meet. But by and large I have been a poll taker giving Gov Romney his “internal poll”.

Given that the Romney office is 15 miles away the nearest obot office is 5 minutes, mischief has most certainly been on my mind. But I am not sure I could stand the smell in such close confines.

But by and large I have been a poll taker giving Gov Romney his “internal poll”.

I hope I don't come across as splitting hairs/dealing in semantic here (I don't intend to at all) but in campaign management what you're doing would be classified as being part of GOTV - Get Out The Vote. Extremely important, and we all really appreciate your effort, but unlike polling where the goal is to figure out both how your candidate and the other guy are doing, GOTV is focused on determining actual supporters and then motivating them to get to the polls on election day.

A solid GOTV effort, for instance is what gave Bush Ohio (and therefore reelection) in 2004.

"Internal polling" really refers to the same kinds of polling that the MSM are doing - random calling (or sometimes repeat calling) of numbers to achieve a snapshot of how the candidate/opposition are doing. The difference with the campaign polling is that the campaign pollsters use different weighting and turnout projections in their polls. For instance (probably harping a bit on this at this point), 2008 was a D+7 election. MSM polls have consistently been using that as the basis for their results, and in many cases higher numbers in the D+9 to D+13 range.

What an internal poll will do is start with the assumption that 2008 was an historic election with a very elevated D turnout. So they'll dial back the D numbers to what they think actual turnout will be (they can use other polls to narrow this down). So for instance they'll go with D+2 (as Rasmussen seems to be doing). With that, the whole picture will change - Obama may be up by 4 in a D+13 poll, but in a D+2 poll Romney will be ahead. Which one is more accurate? Commonsense right now says the D+2 one ... but we'll see on election day.