What the Polls Mean (and What They Don’t)
On Wednesday afternoon, the Vermont Democratic Party decided that, on second thought, it would allow reporters to cover its Sunday evening fundraiser starring Rep. Barney Frank of Massachusetts.
A wise move. When a powerful House committee chairman speaks in a quasi-public setting, he might commit what some reporters call a “new” (or perhaps a gnu?). That is, he might make news. News ought to be covered by folks whose business it is to cover the news, not because they have such a wonderful time doing so (it’s often a drag) but because it’s better to have an informed public that an uninformed public.
Such, at least, is the theory behind democracy, or, as it’s sometimes known, a republic, meaning both Democrats and Republicans should facilitate rather than encumber the process. For the nonce, the Democrats seem to have figured this out. Some of us will be watching all political parties for future compliance.
Speaking of politics, let’s deal with that poll about the governor’s race that came out the other day, and with the reaction to said poll.
According to a Rasmussen Reports survey of 500 randomly selected Vermonters, Lt. Gov. Brian Dubie, unopposed for the Republican nomination, is leading all five Democrats running in the Aug. 24 primary to oppose him.
Dubie’s lead over Secretary of State Deb Markowitz is a modest one, 47 percent to 40 percent, within the poll’s plus-or-minus 4.5 percentage point margin of error.
Against the other four, Dubie gets more than 50 percent, though just barely (51) against Sen. Doug Racine. But the lieutenant governor has a 15-point margin over Racine, and bigger leads against Sens. Peter Shumlin and Susan Bartlett and former Sen. Matt Dunne.
Bartlett fared worst, trailing Dubie by a 55-to-29 percent margin, so perhaps it was not surprising that her campaign manager, John Bauer, issued a statement questioning the validity of the poll. Bauer quoted the recent comment of Middlebury emeritus political science professor Eric Davis that ”Rasmussen polls should be interpreted cautiously(because) Rasmussen uses automated prompts with touch-tone responses rather than live operators to collect its data.”
That’s true, and Rasmussen’s polls are controversial, partly because Scott Rasmussen makes no effort to hide his own very pro-Republican sympathies.
But that doesn’t mean the polls are inaccurate. Even Nate Silver of 538.com, a critic of Rasmussen’s methods, concedes that the company’s polls have a higher than average record for calling races right.
Some of the poll’s findings at least raise the possibility that the survey over-sampled Republicans. For instance, it found Vermonters almost evenly split on supporting off-shore oil drilling.
Maybe they are, but it would be risky to run for office here on a pro-drilling platform.
Still, there’s not much reason to doubt that the Rasmussen poll is an accurate snapshot of the campaign right now.
But that’s all. The dynamic of the race will change on the night of August 24, when the primary results come in. (Rasmussen, confused, used the old date of September 14, but that doesn’t have any impact on the poll’s accuracy).
The more pertinent question right now, then, is not whether Dubie is ahead of the Democrats, but which Democrat is ahead of the others. That question can’t really be answered until next month, when Research 2000 releases its poll on the Democratic race. But it can sort of be answered now, as long as everyone understands that the answer is tentative and even a touch shaky, because two of the campaigns, Markowitz’s and Shumlin’s have hired respected pollsters to check out the race.
The results have not been made public. They haven’t even been quasi-officially leaked. But some people who have seen the results have…well, let’s just say they’ve chatted. In pubs, on the phone, in meetings. Hints have been dropped. With that shakiness caveat firmly in mind, the race seems to be shaping up this way:
Markowitz is ahead. Racine is second, though just how distant a second is hard to determine. Shumlin is an apparently not-very-close third. Dunne and Bartlett remain in single-digit percentage figures.
These results do not justify a prediction that Markowitz will win, or even justify an assessment concluding that she’s the front-runner. Nor do they mean that Dunne and Bartlett have no hope. To begin with, this is very hard race to poll. It all depends on turnout, which is likely to be quite low, and while the pollsters are probably “screening” for likely voters, that’s a most imprecise exercise under the circumstances.
All the pollsters can do is ask respondents whether they plan to vote in the primary. Ok, they can push a bit: “Are you sure you’re going to vote?” Or, “are you enthusiastic about the race?” Or some such. But for a couple of reasons, many folks may say they really, really plan to vote, but end up staying home.
Or, in this case, being away. August 24 is in the middle of vacation season. Some of those respondents who said they were going to vote may have forgotten that they have reservations at a resort or plans to visit their Aunt Louise in California on that date.
Yes, there’s early voting. It starts July 12, and all the campaigns have early-vote operations to urge their supporters to fill out their ballots before taking off for the beach or Aunt Louise’s. But one of the peculiarities of this campaign raises the likelihood that many voters may not bother.
The peculiarity is that all five of these Democrats are just fine with the vast majority of Democratic voters. This is not an ideological battle. There are some policy differences, but all the contenders are mainstream, center-left Democrats. Nor is it one of those contest between “regulars” and “reformers,” whatever that means. So far, at least, nothing resembling a “character problem” has arisen about any of candidates.
The result is that while few rank-and-file Democrats are all that enthusiastic about any of the five, neither do they harbor any hostility toward one or two of them. Harboring hostility is great turnout builder, absent in this primary.
But there’s one more thing to keep in mind. The dynamic of the primary is likely to shift next month, when that Research 2000 poll comes out and when the candidates report their fund-raising results on July 15. The candidates who bring up the rear in both measurements (and whoever trails in one is likely to be trailing in the other) are going to be in trouble. Both press and public will conclude that they are out of it.
That may not be fair. But it’s what almost always happens. At that point, the race will take on a new shape. Don’t expect anyone to drop out officially. But at least two of those Democrats will effectively spend the last month or so going through the motions, with most attention centered on the two or three at the top.
That’s when it’ll get interesting.
And Don’t Forget: The News Guy is on Vermont This Week tonight (repeated Sunday).
