AK: Rasmussen: Miller (R) leads write-in Murkowski big; McAdams (D) in third

[Small Business Owner of Any Repute]

New Poll: Alaska Senator by Rasmussen on 2010-09-20

Summary: D: 25%, R: 42%, I: 27%, U: 5%

Poll Source URL: Full Poll Details

[Capitan Zapp Brannigan]

Obama still with a 42% approval rating here.

Also, I wonder how screwed up the write in campaign is going to make the polling look for Alaska.

[tpfkaw]

Once we get confirmation from PPP I'll be confident enough to de-sig him.

[King]

I don't think enough people will take the time to write in M - u - r - k - o - w - s - k - i on election day.

Senator Miller it is.  Alaskan Senators are usually odd balls anyway (see Mike Gravel, Ted Stevens, etc) so this is no big deal.

[Sam Spade]

Don't trust Alaska polling - it tends to underestimate Republicans, sometimes heavily.  The voters also lie a bit.

That being said, I think Rasmussen is polling the correct way (from what I read).  Don't inform, but take responses.

[Eraserhead]

Once we get confirmation from PPP I'll be confident enough to de-sig him.

... huh? lol.

[tpfkaw]

Once we get confirmation from PPP I'll be confident enough to de-sig him.

... huh? lol.

I only put people who I mostly agree with and who might be in trouble at the polls.

[minionofmidas]

Murkowski higher than I thought she'd be (now).

[The Vorlon]

Polling a well known "write in" candidate is a nightmare.

Off the top of my head, I would be inclined to use a so-called  "unaided" ballot where you don't read the person being interviewed the actual candidates, but rather basically just ask them "Who are you going to vote for?"

Polls that read off the names tend to over state third party/fringe candidates because they get a share of what is essentially a "don't know" vote.

If you ask people who they will vote for and they then say "Murkowski" chances are pretty good they will at least try to vote for her.

Does anybody know the rules re write in's in Alaska?

Does it have to be, for example, Exactly "Murkowski", or does "Murcowski" or "Murkowsky" or "Lisa M" count?

How close does the voter actually have to get?

[Eraserhead]

Murkowski will come in third.

[Lief 🗽]

Polling a well known "write in" candidate is a nightmare.

Off the top of my head, I would be inclined to use a so-called  "unaided" ballot where you don't read the person being interviewed the actual candidates, but rather basically just ask them "Who are you going to vote for?"

How would that work with automated polling, like Rasmussen's or PPP's?

[minionofmidas]

Does it have to be, for example, Exactly "Murkowski", or does "Murcowski" or "Murkowsky" or "Lisa M" count?

the latter.

You must be logged in to read this quote.

no clue.

[Fmr. Pres. Duke]

I expect her actual numbers will be much lower than this poll indicates. This may be the highest we see her. Miller will win simply because people will see the (R) by Miller's name.

[tmthforu94]

I expect her actual numbers will be much lower than this poll indicates. This may be the highest we see her. Miller will win simply because people will see the (R) by Miller's name.

This. If Murkowski had barely won the primary and Miller ran a write-in campaign, she'd also win by a comfortable margin due to that (R).

[The Vorlon]

Polling a well known "write in" candidate is a nightmare.

Off the top of my head, I would be inclined to use a so-called  "unaided" ballot where you don't read the person being interviewed the actual candidates, but rather basically just ask them "Who are you going to vote for?"

How would that work with automated polling, like Rasmussen's or PPP's?

Rasmussen has the ~~technical~~ capacity to record the voices and ask an open ended question, and then digitally record the reply. - He rarely uses this capacity.

It would dramatically add to the cost of the poll, as you would need a human to listen to the tapes and manually record the answers.  I also suspect the results would be skewed from a Rasmussen baseline.  The reason "the bot" gets such a low "no reply/undecided" is that people (so the theory goes) feel more comfortable pushing a button and not replying to a human being.  Does getting their voice recorded, so there is an actual mp3 file of them saying "I support X" alter that feeling of anonymity? - Hard to know for sure, but I would sure want to test it before I relied on the results.

Rasmussen does do "candidate research" polls using his voice recording capability, and a number of candidates have used his services. - Essentially what he does is ask an open ended question saying, "Describe in just a very few words what you think of Candidate X" and then the recording are turned over to the campaign.

The best campaign advertising reinforces a perception that is already there, so if a lot of people unprompted said "candidate X was untrustworthy" that might be an effective negative ad, or, on the flip side, if a lot of people unprompted said "Y" sounded like a smart person, they might want to use that in a positive ad.