NH: Granite State Poll: Sununu narrows gap to 4

[Small Business Owner of Any Repute]

Shaheen (D) 46%
Sununu (R)  42%
Other  1%
Undecided  10%

http://www.unh.edu/survey-center/news/pdf/gsp2008_summer_cong72308.pdf


In a rematch of their 2002 race, Republican Senator John Sununu and former Democratic Governor Jeanne Shaheen are in what is shaping up to be a close race for U.S. Senate. Sununu has trailed Shaheen significantly over the past year but he has narrowed the gap in recent months. Shaheen currently receives 46% of the vote with 42% going to Sununu, 1% for some other candidate and 10% are undecided. Sununu had Shaheen trailed by 17% in February and 12% in April.


These findings are based on the latest Granite State Poll,∗ conducted by the University of New Hampshire Survey Center. Five hundred (519) randomly selected New Hampshire adults were interviewed by telephone between July 11 and July 20, 2008. The margin of sampling error for the survey is +/-4.3 percent. Included was a subsample of 475 likely November voters (margin of sampling error, +/- 4.5%).

[Alcon]

This is an exceptionally weird-polling race.

[Sam Spade]

Another day, another summer poll.  uni vs. ARG.  Ugh.

Well, at least here the uni historically has a decent record, whereas ARG, well...

My gut says it's closer to the UNH, we'll see...

[Dan the Roman]

See my comments in the other thread. Party ID is waaaaaay off.

44% R 38% D 18% I

Do I have to  mention that their likely voter screen includes individuals not registered to vote?

[Sam Spade]

See my comments in the other thread. Party ID is waaaaaay off.

44% R 38% D 18% I

Do I have to  mention that their likely voter screen includes individuals not registered to vote?

The second thing is a problem.

I don't see the party ID having a six-point gap, unless I'm misreading something.

[TheresNoMoney]

See my comments in the other thread. Party ID is waaaaaay off.

44% R 38% D 18% I

Do I have to  mention that their likely voter screen includes individuals not registered to vote?

That is WAY off. Indies are 34% or 35% of the NH population and Reps and Dems are almost the same percentage (31%).

[Dan the Roman]

Sorry I confused it with the first district.

The party id in the poll is 44% R 44% D 12% I.

The problem with is poll is it overestimates Republicans and Democrats at the expense of independents. Independents in this poll however, seem to behave a lot more like Democrats than Republicans, so it ends up hurting Shaheen who leads by 20.

Shaheen leads Democrats 82-9, Independents 42-26, and loses Republicans 82-11. No way would that margin produce only a 4 point lead in NH.

Re weighted to 35% I 31% D 31%R which is probably still friendly to Sununu we get, a 44% - 37% Shaheen lead.

[Sam Spade]

Sorry I confused it with the first district.

The party id in the poll is 44% R 44% D 12% I.

The problem with is poll is it overestimates Republicans and Democrats at the expense of independents. Independents in this poll however, seem to behave a lot more like Democrats than Republicans, so it ends up hurting Shaheen who leads by 20.

Shaheen leads Democrats 82-9, Independents 42-26, and loses Republicans 82-11. No way would that margin produce only a 4 point lead in NH.

How many Independents in NH would truly identify as Independent?  Just a thought.

Actually, after looking at the poll, it appears like they ask party registration first and then push independents to party ID.  The registration numbers look accurate to me (maybe a point too Republican - the GOP has an >1% advantage now that I remember).

Of course, this still doesn't eliminate your point in the other thread, or this one, although I'm unsure as to whether the non-registereds are utilized when they push the Indys.

Or the fact that it's summer.  But that affects all polls.