Google Ads


Site Stats

HI-01: Dead Heat on a Merry-Go-Round

by: James L.

Tue Oct 05, 2010 at 2:40 PM EDT


Public Policy Polling for Daily Kos (10/2-3, likely voters, no trend lines):

Colleen Hanabusa (D): 48
Charles Djou (R-inc): 47
Undecided: 5
(MoE: ±3.9%)

It looks like Charles Djou and Colleen Hanabusa are in a game of jump ball here, which is a better result than I had been expecting for Team Blue here. If turnout were at 2008 levels, as Jed L. says, Hanabusa would be leading by 54-40.

Bonus finding: In the gubernatorial race portion of this poll, Duke Aiona and Neil Abercrombie are tied at 48-48. Considering this is Abercrombie's old district, this could mean that the Governor's race is a lot closer than previous polls indicated.
James L. :: HI-01: Dead Heat on a Merry-Go-Round
Tags: , , , , , , (All Tags)
Print Friendly View Send As Email

HI-01: Dead Heat on a Merry-Go-Round | 63 comments
Could there be the
same polling issues like there were in the special election?

19, Self Appointed Chair of the SSP Gay Caucus (I claimed it first :p), male, Dem, IN-09 (College IN-09) (Raised IL-03, IL-09)
Probably
The 2004 polling for president was also poor.
[ Parent ]
Yup
Remember Cheney's last minute trip to Hawaii in 2004?
[ Parent ]
Didn't think about this,
But I would not be surprised if that was the case.  I'd probably peg this as a Hanabusa up by 4 with that factored in.

20, Male, Democrat, CA-44 (home) CA-12 (college)
[ Parent ]
Very good question. And on top of that this is robpolling, so...
...it's possible hang-ups are a bigger problem?

Very hard to say.

43, male, Indian-American, Democrat, VA-10

[ Parent ]
The obvious question
How much did Hanabusa overperform the polls that were taken before the three-way race against Djou and Case? I doubt that Case was winning very many of these hidden voters, so her overperformance would be likely to be similar, no?

Kansan by birth, Californian by choice, and Gay by the grace of God.
[ Parent ]
Result was 39.4% 30.8% 27.6%
Final poll had the Dems tied so three points.
[ Parent ]
good question
I think the issue is the change in Hanabusa's total vote share from the last poll to the election, not the change in her margin v. Case.

41, Ind, CA-05
[ Parent ]
If Hanabusa leads in the polls
It's over.  There's a major hidden vote, mostly from native hawaiians/asians that the pollsters simply don't pick up on.  Polling had Hanabusa in third the whole way in the special and then she came back strong on election day and, without Case to screw it up, would've crushed Djou to pieces.  

I think Hanabusa wins this one much easier than anybody anticipates, 8-12 points probably.

23, Male, Democrat, OH-13

^This
I posted a similar comment in the digest. I believe that the hard-to-reach group is Japanese-American women, who do not respond to polls for cultural reasons. They are also, presumably, a group that favors Hanabusa, who is Japanese-American herself.

20, CD MA-03/NH-01/MA-08
[ Parent ]
Yup, and as I noted, on top of all that this is robopolling, so...
...I would think that would prompt a much quicker hang-up on the call.

43, male, Indian-American, Democrat, VA-10
[ Parent ]
It should be a good test for the credibility of Nate's model
He currently has HI-01 as a 59% chance of takeover.

If Nate's models just count the raw numbers, that probability should remain stable.

If Nate's models include appropriate intangibles (i.e. underpolling of issei, nissei, etc.), that number should go up, given the data from this poll.

[ Parent ]
Win yes
I doubt by that much though. So much for Djou's +8 internal.
[ Parent ]
Sounds about right

[ Parent ]
I'm more concerned about Abercrombie
I suppose the other district is more Democratic but not by much. We need a statewide poll here sharpish.
[ Parent ]
HI-02 is about 3 points more Democratic
It went 56% for Kerry and 73% for Obama, while HI-01 went 53% for Kerry and 70% for Obama. Looking at data from both elections, it appears that the best part of the state for Republicans is the part of Oahu (Honolulu County) that is in HI-02, and the best part of the state for Democrats are the outer islands of Hawaii, Maui, and Kauai (all in HI-02 as well.)

20, CD MA-03/NH-01/MA-08
[ Parent ]
HI-01 and HI-02
are basically drawn to more or less evenly reflect the state's Democratic lean.  There are only two spots in the state that Republicans show considerable strength in, wealthy areas in northeast Oahu (Kailua) and southeast Oahu (Hawaii Kai and other surrounding areas).  In fact, out of the 25 members of the Hawaii state Senate, the only two Republicans come from these two areas.

The drawing of the congressional districts splits these two areas into the two different districts so that Democrats are easily dominant in both districts.  However, it's not a perfect split as you noticed.  The other islands besides Oahu are so insanely Democratic that even with some conservative areas of Oahu thrown in, HI-02 is a few points bluer than HI-01.

[ Parent ]
It would be subject to the same problems
other polls have in Hawaii, which are discussed above.

Polling in Hawaii consistently undercounts democratic support. In Alaska this is reversed, and republicans almost always do better than the polls indicate (while very conservative republicans outperform moderate repubs--hence the result of the Miller-Murkowski primary).

[ Parent ]
I have little doubt that Abercrombie will lose
The only way a republican could win the Gov's seat would be if Linda Lingle ran again, but she is term limited.

20, Male, Democrat, CA-44 (home) CA-12 (college)
[ Parent ]
You mean no doubt he wins?

[ Parent ]
Yeah...
This always happens...

20, Male, Democrat, CA-44 (home) CA-12 (college)
[ Parent ]
I think the same

The close result for Abercrombie is close to be an outlier. This can be another positive sign for Hanabusa from this poll.
[ Parent ]
Whatever it takes, I just want Hanabusa to win it (n/t)



Indepedent/Lean D. Dude.
All 5s (now TX-5; frmly VA-5 and CA-5)  
It would be nice
to have another woman, especially a woman of color in Congress.

19, Self Appointed Chair of the SSP Gay Caucus (I claimed it first :p), male, Dem, IN-09 (College IN-09) (Raised IL-03, IL-09)
[ Parent ]
Here's some census data for HI-01
Non-Hispanic white: 18.6% (PPP has it at 32%)
Asian: 51.8% (PPP has it at 45%)
*Japanese: 19.1%
*Filipino: 15.8%
*Chinese: 7.1%
*everyone else: 9.8%

I'm a big fan of PPP, they're extremely transparent and (usually) accurate, but as others have said, I'm not ready to buy this poll (particularly for the gubernatorial race).

Politics and Other Random Topics

24, Male, Democrat, NM-01, Chairman of the Atheist Caucus, and Majority Leader of the "Going to Hell" caucus!

Race
Race might matter here, yes?  Abercrombie is white.  Aiona is Asian.  That could be part of what is going on.  In fact, while HI-02 is a tad more Dem than HI-01, HI-01 is Abercrombie's own district.  That he's tied here makes me wonder if he's not slightly down in HI-02 and the state overall.  I had HI-Gov as Safe Dem, but this makes me think it should be Likely Dem at best and perhaps even Lean Dem.

43 - Male - GOP/Libertarian - FL 22
[ Parent ]
Likely until we see a proper statewide poll
I had it there already since the last poll was only a 12 point margin.
[ Parent ]
Duke Aiona is Chinese and native Hawaiian
Chinese only make up about 4% of Hawaii's population (or about 10% of the Asian population). Most Asians on the Island are Japanese (14.2% or about 37% of of the Asian population) or Filipino (13.7% or about 35% of the Asian population). I doubt Aiona is doing all that well among the Asian vote based on that.

Politics and Other Random Topics

24, Male, Democrat, NM-01, Chairman of the Atheist Caucus, and Majority Leader of the "Going to Hell" caucus!

[ Parent ]
Good your point about the census

I agree.
[ Parent ]
a question.
In past elections where there has been a "Shy Asian" effect, wasn't it always an Asian candidate vs. a Non-Asian candidate?  Is there any indication (more than a gut feeling) that it's specifically underestimating Democrats?   Would a Japanese woman lean more to "Duke" over Neil because of race?  

I looked for a close election for some indication: The 2002 Gov Race.  It was Lingle (R) against Hirono.  The polls I found were R+1, Tie, R+4.  Lingle won by 5.  In 2006, she was running as an incumbent against Iwase.  I can't find any polls, but Lingle won 62-35 (27 points).  It's tough to look at the Senate blowouts for any proof.  There just haven't been a ton of competitive races in HI.

Since I lived in AK, we had an affinity for HI and I'm fascinated by this theory and am going to continue to compile evidence to see how valid it is.  I'm not discounting it at all, I just want to see if the numbers are showing what we think and hope they are...If anyone can point to a definitive study, I'd love to read it.  Thanks!

[ Parent ]
You live in AK?
What are your observations on the Senate race?  Are those McAdams commercials getting good airtime?

Ad hoc, ad loc and quid pro quo!
So little time, so much to know!
[ Parent ]
USED to
I did live in AK, which is what drew me to this board in the first place after the Miller win.  (It's funny, I put a post up about how none of my friends there had ever heard of McAdams and his answers to the first media inquiry were vague and I was accused of trolling...).  Since this is not an AK post, I don't want to derail, but very quickly the people that I still know up there seem to like McAdams now, but I don't know if there are enough of them.  I really liked the "tie" ad and I think it struck the perfect tone for the typical AK voter based on my experience.  The usual partisan playbook is worthless there, and if Miller is going to run that kind of campaign, he could be in trouble.  

Quick aside...it's hilarious to see Palin on the national stage now as such a polarizing dimwit.  Some Dems actually kind of liked her in up here for taking on the Republican establishment.  She was way more independent in the beginning, which is why she had 70-80% approval ratings.  The second she became the face of anything national, they took a nose dive.  I apologize for the brief derail...I hope the strong HI/AK connection will make it acceptable.

[ Parent ]
Duke is Chinese, I doubt Japanese women would have any particular affinity towards Duke
The communities are quite different (and actually, not even particularly friendly towards one another either, if memory serves). This isn't really based on a study, just on the idea that "Asian" is more of an umbrella term that groups together a lot of very different groups (much like "Hispanic")

Politics and Other Random Topics

24, Male, Democrat, NM-01, Chairman of the Atheist Caucus, and Majority Leader of the "Going to Hell" caucus!

[ Parent ]
yeah
like I've said before, "Asians" are such a massive and diverse group in Hawaii that it's hardly even useful to try to conflate all the data together under a blanket term like that in Hawaii.
[ Parent ]
Hopefully
The Census will release CD or county-level ethnicity data. I noticed that they asked specifically what part of Asia you traced your roots to if you identified as Asian, and there was a similar ethnicity question if you identified as Hispanic.

It would be interesting to see Hispanic broken down into Mexican, Puerto Rican, Brazilian and "other," and Asian broken down into Arab, Chinese, Indian, and "other" when the Census releases its results. That would better help redistricters keep ethnic communities intact.  

20, CD MA-03/NH-01/MA-08

[ Parent ]
pretty sure they've done that for a long time
Asian is split between Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, and Other Asian. Hispanic is split between Mexican, Cuban, Puerto Rican, and Other Hispanic. I don't think that Arabic is mentioned and a lot of them will answer White or Other.

And i'm pretty sure they asked that stuff again and will mention it as well.

Seems sort of hard to keep ethnic communities together in most situations (especially in Hawaii).

[ Parent ]
Do they release it to the public on CD level, though?
If so, do you have a link? It would be cool to see.

20, CD MA-03/NH-01/MA-08
[ Parent ]
Here ya go:

Demographics for HI-01

They do it for ever congressional district and county (though some counties, because of size, are only for 2000).



Politics and Other Random Topics

24, Male, Democrat, NM-01, Chairman of the Atheist Caucus, and Majority Leader of the "Going to Hell" caucus!

[ Parent ]
The HI polling for Pres-2004 had Bush in dead heat or even leading, but...
...Kerry won 54-45.

So there are some more data points for you that Democratic strength is understated in polling.  And that was a white man vs. a white man, neither having any personal connection to the state or running an aggressive campaign there.

It's not a LOT of evidence, to be sure, but no doubt there's enough evidence to conclude "Hawaii is difficult to poll accurately."

43, male, Indian-American, Democrat, VA-10

[ Parent ]
Doubtful as to Duke
Asians are hardly monolithic.  Granted I've never been to Hawaii, but I live in a noticeably Asian area of the second-most Asian state and there are more differences than commonalities between the political alignment of, say, Japanese-Americans and Vietnamese-Americans.  Each might support one of their own across party lines, but in my experience "Asianness" isn't a particularly cohesive group identity.
[ Parent ]
There are good surveys out there that show the distinctions between "Asians"......
Vietnamese are indeed among the most politically conservative and Republican of Asians.  Although they are too small in numbers to have been polled or otherwise tracked politically, I suspect other Southeast Asians are in the same situation here, as refugees from communism, so it might be possible to lump together Vietnamese, Laotians, and Cambodians as having a shared political orientation.

This linked page has links under the heading "voting rights" to exit poll data.

http://aaldef.org/cgi-bin/mt/m...

43, male, Indian-American, Democrat, VA-10

[ Parent ]
Reminds me of
the split between Cubans and other Latinos in Florida.
[ Parent ]
Cambodia was under a fascist, brutal regime, not a Communist regime.


Ad hoc, ad loc and quid pro quo!
So little time, so much to know!
[ Parent ]
Hasn't
Hasn't it been under both?

43 - Male - GOP/Libertarian - FL 22
[ Parent ]
yes
It was a French colony until 1953, then a monarchy until a right-wing military coup in 1970, then a Maoist dictatorship under the Khmer Rouge in 1975 until they picked a fight with Vietnam in 1978, resulting in a less brutal Communist dictatorship after the Vietnamese crushed the Khmer Rouge. Then there was a civil war ending in 1993, and now it's a constitutional monarchy again.

41, Ind, CA-05
[ Parent ]
My hometown
and high school are about 6% Cambodian, and I have quite a few Cambodian friends, all pretty liberal and probably Democratic (although granted these are children of immigrants, not immigrants). Then again, my city went 83% for Obama. So that begs the question, are first-generation Vietnamese people more liberal than their parents? Also, are Vietnamese people in LA more liberal than Vietnamese people in Orange County?

FWIW, I am pretty sure the Hmong population in Wisconsin/Minnesota (including many political refugees) is solidly Democratic.

21, dude, RI-01 (registered) IL-01 (college)
please help Japan. click "donate funds" in upper right and then "Japan Earthquake and Pacific Tsunami." http://www.redcross.org/

[ Parent ]
Yeah, and
Aiona is a Native Hawaiian-sounding name, no Japanese people would mistake him for one of their own. For that matter I doubt most Chinese people would know he's Chinese unless he explicitly went around advertising it. FWIW, he is also part white (I think Portuguese).

21, dude, RI-01 (registered) IL-01 (college)
please help Japan. click "donate funds" in upper right and then "Japan Earthquake and Pacific Tsunami." http://www.redcross.org/
[ Parent ]
I might just believe that.
From what I know about CA Asian turn-out, that sounds just about right. Asians are 12.3% of the population in CA, and they were about 6% of the electorate in CNN's exit polling in 2008. You have similar numbers for Hispanics.

Of course, Asians in Hawaii are less of an immigrant community than Asians elsewhere, so it might be different.

http://mypolitikal.com/

[ Parent ]
I dunno, I have a hard time believing Aiona has a real shot here...
I'm hardly an expert on Hawaii politics, but my best guess on a voter turn-out model looks something like...

Democrat - 46%
Independent - 30%
GOP - 24%

In other words, one of the most Democrat-friendly in the nation, especially given that miniscule # for the Republicans. I think PPP is oversampling Independents.

Abercrombie - 86/44/6 = 54%
Aiona - 14/56/94 = 46%

I'd personally peg this at Lean D. Anything can happen this year, and I think the HI-01 race is a bonifide dead heat (though I think Hanabusa wins), but the electorate is almost as bad for the GOP here as Alaska is for the Democrats.

For daily political commentary, visit me at http://polibeast.blogspot.com/ and http://twitter.com/polibeast

so
any guesses on how turnout gets hindered by the matter of half the voting day taking place with polls closed in other parts of the country?

Unrelated fun note: to my knowledge, this is the first November election to be held under daylights savings time. At least anybody having to do last-minute GOTV will have some better luck at having the sun out

I can understand turnout
Being hurt by an early presidential call but I seriously doubt it would make a difference in these races.
[ Parent ]
turnout percentages
of registered voters (statewide / Honolulu)

2008: 66% / 66.1% (election essentially over before the polls closed at 6pm Hawaiian Standard Time)
2006: 52.7% / 53%
2004: 66.7% / 67.4%
2002: 57% / 57.4%
2000: 58.2% / 57.9%
1998: 68.6% / 68.7% (Lingle/Cayetano, Marriage amendment)
1996: 67.9% / 67.7%
1994: 77% statewide
1992: 82.4% statewide

I think there's a pretty real risk of a subpar Dem turnout for the last few hours with the likely news of results coming in involving Democrats losing races. Which is even more sad since this might be one of the races to determine who wins the House.

[ Parent ]
Not by those figures
Low in 2006 and high in 1994.
[ Parent ]
1994 also had a very competitive Gubernatorial race
but the best bet, get the votes in before the afternoon/evening or absentee.

Another note on the differences in time here. Indiana and Kentucky close at 6pm Eastern, which is noon in Honolulu.

(It'd be a bit unrelated to wish for some states to close their polls closer to 7 than 6, for the sake of turnout and all. Might be just as relevant voting reform as some of the stuff we hear about)

[ Parent ]
Well if this poll is true
There is a very competitive race this year too. Seriously, I wouldn't worry about this. Hawaiian Democrats have every reason to vote whatever happens in the Lower 48.
[ Parent ]
Agree, an open Governorship and hard-fought Congressional seat are plenty motivating, plus the Governorship is...
...not a blowout, even though I think this poll is inaccurately GOP-friendly in the result.

43, male, Indian-American, Democrat, VA-10
[ Parent ]
Bah
I have serious issues with this poll, as with many pollsters that try their hand at Hawaii (which continues to be an astoundingly difficult state to poll accurately).

Hanabusa should win easily.  Abercrombie will also win easily though despite what you may think, he will probably do a couple points better in the other district, HI-02.  Aiona will get destroyed on the other islands.

Also, that PPP poll finds Inouye leading Cavasso by 66-28.  This sounds reasonable until you remember that Cavasso is a little-known perennial candidate and six years ago Inouye demolished Cavasso 76-21%.

Combine that all with the fact that their sample only voted 56% for Obama and I'm inclined to view the poll as somewhat of a worst-case scenario for Dems.  Democratic turnout may be down elsewhere in the country but I'm expecting good performance here due to the excitement behind retaking the governor's mansion.

I posted this after the Hawaii Primary...
.... but it is worth repeating here.

Colleen Hanabusa received 85,732 votes in this years primary (The other Dem in the race received 22,874 votes). In 2006 there were 162,794 total votes cast in CD1 during the General Election. Hanabusa's vote total in the primary is 52.6% of the 2006 general vote total. The total Dem vote in the 2010 primary was 66.7% of the total vote in the 2006 general. Unless there is a spike in general election turnout or the voters who voted Dem in the primary sit out the general she is in good shape.

"Where free Unions and collective bargaining is forbidden, freedom is lost." - Ronald Reagan

well
2006 was fairly uncontested in Hawaii, up and down the board.  Akaka and Lingle cruised to re-election, and Hirono and Abercrombie easily defeated overmatched Republican opponents (despite Hirono running for an open seat, the race simply never became competitive).

This year, the Democratic primary got a lot of interest, and with Republican primaries boring as always, pretty much everyone votes in the Dem primary.  So I can see why Hanabusa's primary performance is so impressive compared to 2006's general election totals.

I guess I'm just used to seeing massive Democratic primary votes in Hawaii, given our open primaries, that I start to discount the usefulness of that data.

[ Parent ]
Total Votes in CD1 race in off year elections
2006- 162,794
2002- 180,733
1998- 189,579
1994- 176,706

You are right that 2006 was a low turnout year but the fact that over 108,000 votes were cast in the Dem Primary and off year election total votes have never surpassed 190,000 is a good sign for dems.

"Where free Unions and collective bargaining is forbidden, freedom is lost." - Ronald Reagan

[ Parent ]
122482 votes in the primary
counting the 13,000 blanks of people who showed up for the Gubernatorial primary.

119846 votes in the 2006 primary (82K Abercrombie, 21K challenger, 16K blank). Should be noted there were more blank Democratic votes in that primary than total Republican votes.

Interesting quirk for how bad a campaign can be. Randy Iwase won 119058 votes in the primary and 121717 votes in the general election. Especially in an open primary, winning more votes in the primary than the general and losing the general is the sign of an exceptionally bad campaign. But people who win more votes in a primary than in the general election and lose is a topic for another time.

As for primary v. the general election total, i've already posted on here how bad turnout was in Hawaii in 2006.

[ Parent ]
The big reason this poll
is bogus, is that the Democratic primary results for the General election discredit it; the total number of votes case in the Democratic primary amounted to 62% of the total number of votes cast in the 2006 GENERAL Election, Hanabusa's total of votes alone put her at 48% of the 2006 General Election. This sample is ridiculously favorable to Republicans.  
There must have been some late votes added...
... because the primary totals have risen on the official website since the day after the primary.

"Where free Unions and collective bargaining is forbidden, freedom is lost." - Ronald Reagan
[ Parent ]
HI-01: Dead Heat on a Merry-Go-Round | 63 comments

Copyright 2003-2010 Swing State Project LLC

Primary Sponsor

You're not running for second place. Is your website? See why Campaign Engine is ranked #1 in software and support among Progressive-only Internet firms. http://www.mediamezcla.com/

Menu

Make a New Account

Username:

Password:



Forget your username or password?

About the Site

Publisher: DavidNYC

Senior Editor: James L.

Senior Editor: Crisitunity

Contributing Editor: Jeffmd

Search the SSP archives:

Sign up to receive a daily e-mail whenever the SSP is updated:

SSP Resources

Presidential Vote by CD

SSP Document Collection

Race Tracker Wiki

Dave's Redistricting App

Blogroll

Daily Kos Elections
DemConWatch
Five Thirty Eight
Governing
Hotline On Call
Political Wire
Pollster.com
Prop*50
Public Policy Polling
The Thicket
Powered by: SoapBlox