by Andrew Walden
A June 21 Gubernatorial race poll showing Duke Aiona leading Mufi Hannemann by four points apparently set off a panicked frenzy of activity in the Star-Advertiser newsroom—but readers of the print edition would have no way of knowing.
The poll, conducted for the Aiona campaign by the Tarrance Group shows Aiona leading Hannemann and gaining sharply against both of his likely Democrat rivals as compared to a previous HNN/Advertiser poll. So the Star-Advertiser buried a single mention of the results in its little-read Political Radar online blog which is still hosted on the old Star-Bulletin site. The only print publication mention of Aiona’s polling lead came at the end of a West Hawaii Today article June 26. Hawai`i Free Press posted the poll results in full June 21.
Within three days of the Aiona campaign releasing its polling results, a new poll of both the Gubernatorial and Senatorial races from Rasmussen was rushed into production, making robo-calls to 500 “likely voters” on June 24.
Six days later, the Star-Advertiser finally had the kind of news it wanted to report: “New poll shows Abercrombie and Hannemann ahead of Aiona.” The article makes no mention of the Tarrance Group poll saying only: "The Aiona campaign's own polling, however, shows the race closer." An updated version of the same article also omits the results of the Aiona poll.
HNN also did not run a stand-alone report on the Tarrance Group poll. But unlike the Star-Advertiser, when the Rasmussen poll became available, HNN finally ran a story comparing the results from both polls: “Polls paint different pictures of Hawaii governor's race”.
Proof that the Rasmussen poll is stacked comes in the utterly ridiculous conclusion of the poll:
Abercrombie 58% vs Aiona 32% (26% spread)
Abercrombie 59% vs Carroll 30% (29% spread)
It is simply not credible to suggest that a sitting Lt Governor who has raised over $2.2 million for his campaign and has nearly universal name recognition is only 3% closer to Abercrombie than his little known Republican Primary opponent who has been out of office for three decades and has raised just over $9000. This conclusion alone shows that the poll is unreliable. Even Rasmussen’s announcement of the poll results points away from reliance on these polling numbers, writing:
“Rasmussen Reports considers the very favorable and very unfavorable figures more significant than the overall favorability totals.”
And what do those figures show?
Abercrombie 32% Very Favorable, 22% Very Unfavorably
Hannemann, 25% Very Favorable, 16% Very Unfavorable
Aiona, 22% Very Favorable, 21% Very Unfavorable
Carroll, 4% Very Favorable, 16% Very Unfavorable
Rasmussen notes that among the 500 likely voters sampled, Carroll “is by far the least well-known of the candidates.”
How did Rasmussen do it?
Part of the answer comes in the polling questions. By pairing Gubernatorial and Senatorial polling questions in the same 500 calls, the pollster was able to line up questions about Abercrombie and Hannemann with questions about support for popular Senator Dan Inouye. Questions about support for Aiona were paired with questions about unknown Republican US Senate candidate John Roco and Aiona’s little-known GOP Primary opponent, John Carroll. This is a standard poll-slanting techniques designed to create a “yes pattern” in favor of Abercrombie and Hannemann and a “no pattern” against Aiona.
Another technique utilized in the poll was the insertion of a question about the Arizona immigration law. This is designed to agitate progressives. As a result Abercrombie scored 90% with those who favor repeal of the Arizona law. Hannemann scored only 70% with opponents of the Arizona law and Aiona scored only 60% with supporters of the law.
The Gubernatorial polling questions were paired with questions about the US Senate race in which Sen. Dan Inouye, the Senate’s longest-serving member is facing only token GOP opposition. Rasmussen’s June 29 release of the Senate poll results starts out with a complete falsehood:
Democratic Senator Daniel Inouye looks like one incumbent without much to worry about this year. He holds a better than three-to-one lead over his only announced Republican opponent in Hawaii’s race for the U.S. Senate.
The latest Rasmussen Reports telephone survey of Likely Voters in the Aloha State finds Inouye, a member of the Senate since 1963, with 68% support. Conservative activist John Roco has 20% of the vote.
But the Hawaii Election Commission printout of candidates who have pulled papers as of June 24 shows not one but three Republicans competing in the Primary to run against Inouye. And of the three, only former State Rep Cam Cavasso has actually held elective office before. Cavasso was Inouye’s Republican opponent in 2004. Cavasso has a professional website and an active campaign. And yet he is not included by a pollster who claims Roco is Inouye’s “only announced Republican opponent.” To predicate an entire Senatorial election poll on this type of error reflects very poorly on Rasmussen.
Without mentioning the falsehood underlying Rasmussen's poll, BJ Reyes June 29 reports on the Senate results and notes: “To date, Roco is among six who have pulled papers to challenge Inouye, the second-longest serving senator in U.S. history.” Predicating an entire poll on false information is news, but the Star-Advertiser is not likely to report it—even on its blogs.
Ed Case is backing Neil Abercrombie in the Democratic Primary. Ed Case’s uncle, Grove Farm corporate lawyer Dan H. Case, is a part-owner of the Star-Advertiser. In addition to its bias in favor of gay marriage and against Aiona, the Star-Advertiser’s polls have consistently shown Hannemann trailing behind Abercrombie—just as they consistently showed Hanabusa trailing Case in the Democrats’ race for second place against Charles Djou.
NOTE: This article has been corrected. The Star-Advertiser did not commission the latest Rasmussen Poll of the Senate and Gubernatorial races.
Here is the recent Tarrance Group Poll which the Advertiser commissioned this to respond to: Poll: Aiona beats Hannemann in one-on-one matchup
REALITY: More Case family media manipulation