As far as voters have heard about this year’s governor’s race, no one’s in the lead. Or rather, “undecided” is — the top pick of participants who responded to an Idaho Politics Weekly poll about the race in November.
Candidate Raúl Labrador’s internal polling has him ahead in the primary. And what about polls run by all the other candidates?
Surely one of these must be “right”?
With nearly a dozen candidates — Republicans, Democrats and independents — for Idaho’s open governor seat this year so far, voters looking for insight as to who’s in the lead won’t find many choices.
Sign Up and Save
Get six months of free digital access to The Idaho Statesman
And not all polls are created equal. Deciding whom and what to poll can change the outcome.
This is why independent, unbiased polls have become an important part of elections — voters need sources of information not manipulated and managed by candidates or special interest groups.
But Idaho largely lacks independent election polling, primarily because it is so expensive. That means what few polls do come out are newsworthy — but also makes voters here more dependent on their accuracy and approach.
Two recent polls — the only two available to the Statesman on Idaho’s 2018 gubernatorial race — are markedly different in results and methodology.
The first was paid for by Idaho Politics Weekly, an online publication substantially funded by Utah-based Zions Bank. The website’s 2018 election polls — there have been a couple of installments now — are the only ones so far that are widely available to the public. Its most recent poll on the governor’s race was conducted Nov. 8-15 by Dan Jones and Associates from Utah.
Zions Bank, of course, is the namesake tenant in the Eighth & Main high-rise built in Downtown Boise by Salt Lake City-based Gardner Co. GOP candidate Tommy Ahlquist is Gardner’s chief operating officer. Ahlquist’s campaign and Idaho Politics Weekly both say the candidate has no influence over the site or the polling.
“We have actually published more Labrador stuff than Ahlquist stuff, just because Labrador’s people produce more regular opinion essays and press releases,” said LaVarr Webb, Idaho Politics Weekly publisher. “We welcome submissions from all the campaigns and incumbents.”
The site’s poll asked: “If the election for governor were today, for whom would you vote?” Its results: 36 percent of people were undecided, 21 percent supported Brad Little, 17 percent Labrador, 14 percent Ahlquist and 13 percent were split among other candidates.
The second poll, paid for by Labrador’s own campaign, was conducted on Oct. 11-12 by Magellan Strategies from Colorado. This poll has not been publicly released; Labrador’s campaign provided it to the Statesman.
It asked: “If the Republican primary election for governor was being held today, for whom would you vote if the candidates were Raúl Labrador, Brad Little and Tommy Ahlquist?” Labrador led in the results with 37 percent support, followed by Little with 23 percent, Ahlquist with 21 percent and an undecided 19 percent of respondents.
The same, but not the same
Both polls share a few things in common:
▪ Both polls had comparable margins of error: Labrador’s at 3.66 percent, Idaho Politics Weekly’s at 3.94 percent.
▪ Each poll had half male and half female respondents.
That is where the polls’ similarities end.
▪ Labrador’s poll was focused on the Republican primary in May 2018. It polled 714 registered Republican voters deemed likely to vote in the party’s closed primary.
Idaho Politics Weekly’s poll took a broader look. It polled registered and non-registered voters across political parties about the election in general. Of its 619 respondents, 43 percent said they were Republican, 17 percent Democratic, 31 percent unaffiliated and 8 percent identified with another party.
“Our poll was not structured to be a primary poll,” said Judd Nielsen, an analyst at Dan Jones. “This makes it difficult to compare to the Labrador poll … because if we polled for a primary, we would likely include a different methodology — starting with the decision of whether to poll registered Republicans, or to target not only Republicans, but Republicans who are most likely to participate in a primary.”
▪ Labrador’s poll was automated. Respondents had to choose from one of four choices — the three named candidates or “undecided” — limiting their options.
Idaho Politics Weekly used live polling, which means the pollsters and the people they called interacted directly about a list of possible candidates. That provided more flexibility and more options to respond. But that list was incomplete: It included Republican Attorney General Lawrence Wasden, who is not running for governor, and it did not include Democratic candidates like A.J. Balukoff, who filed his candidacy paperwork two weeks before the poll was conducted.
▪ Both polls talked to residents from both of Idaho’s congressional districts.
Labrador’s campaign would not say how many people they polled from each district, so it’s unclear how many were among his current constituents. Idaho Politics Weekly’s respondents were evenly split between the two districts.
▪ Labrador’s poll only surveyed people with landline telephones at home. That’s a dwindling group of people in Idaho, the state with one of the largest percentages of residents who only own a cellphone.
Idaho Politics Weekly called both landlines (59 percent of respondents) and cellphones (24 percent). Another 16 percent of responses came from an opt-in online panel.
Polling via telephone is not as simple as it used to be. Nationwide, about half of all households no longer have a landline, so pollsters can miss half of their audience by not reaching out to cellphone-only residents. But, there is a catch — federal regulations prohibit making automated calls to cellphones, which means pollsters must use the more costly live-calling method to reach cellphone users.
▪ The pollsters focused on different age groups. By age, 70 percent of Labrador’s respondents were 55 or older; 49 percent of Idaho Politics Weekly’s respondents were 60 or older.
Just 17 percent interviewed for Labrador were age 44 and younger, compared to about 32 percent for Idaho Politics Weekly that were age 49 and younger.
What’s all this mean?
Given all that, what should you, the voter, take away?
The biggest challenge with these two polls is that there aren’t any others to compare them to.
And so, we’re left with one pollster’s take on the Republican primary, and one pollster’s image of Idaho at large.
Is either skewed? Both firms believe they hit their target audiences.
The Idaho Politics Weekly poll sought a quick, moment-in-time snapshot to one question, of Idaho adults across age, gender and voting interest. It may be accurate about Idahoans’ feelings, but it’s unclear how well it will forecast this year’s elections.
Labrador’s approach was tailored to closely match past Republican primary voters. According to Idaho voting records, 73 percent of voters in the 2016 Republican primary were age 52 or older — fairly close to Labrador’s percentages. And while he only polled people with landlines, this demographic is more likely to still have one. According to a national study, just 23 percent of people age 65 and older have ditched their landline phones.
Ahlquist and Little, his GOP competitors, would not release their respective internal polls to the Statesman. But both campaigns confirmed they use live polling to reach both landlines and cellphones.
Half of Idaho’s 800,000 registered voters are age 51 or younger. But in the past, they simply haven’t shown up for primaries. About 20 percent participated in the 2016 Republican presidential primary and a paltry 12 percent participated in the closed Republican primary.
“Young voters are simply less engaged, less likely to vote, and less partisan,” said Corey Cook, dean of the Boise State School of Public Service.
Where they do engage is in the general election. In the November 2016 presidential election, they turned out in force — 84 percent of Idaho registered voters age 51 and younger voted. They still weren’t the majority — 53 percent of voter turnout still belonged to ages 52 and above — but the divide was close.
“Nationally, millennials comprise the single largest segment of eligible voters,” said Cook. “But due to their lower likelihood of voting, baby boomers are still the plurality of voters, even in presidential election years when turnout is highest.”