Mark Blumenthal | September 4, 2008
Topics: Bounce , Bump , Daytime interviewing , John McCain , Mike Murphy , Sarah Palin
From a pollster's perspective, we are in something of the eye of the storm. As Gallup's Frank Newport puts it, "a huge influx of election-related news [is] swirling around these days." The convention speeches garner giant audiences; 38.4 million for Barack Obama's acceptance speech last week, 37.2 million for Sarah Palin last night. And history tells us that the convention period can often alter vote preferences in ways that persist through election day.
So something about public perceptions of the candidates is changing, but the real-world limitations of telephone survey interviewing leaves us uncertain about exactly what. At this hour, if you're willing to play what Mark Mellman calls "pick-a-poll," you can make whatever case you want. Three surveys conducted over the last three nights -- from Gallup, Rasmussen Reports and the Democratic Party affiliated Democracy Corps -- all show "no dent" in the Obama advantage gained after the Democratic convention. However, one new survey conducted over the same three-day period from CBS News shows McCain closing an eight point gap since the weekend.
Of course, virtually all of the interviews for those surveys were conducted before Sarah Palin spoke last night. Given the huge audience, and the precedent for convention bounces, future surveys will likely yield a different result.
Some of you may have seen a survey conducted by SurveyUSA today that shows 60% of the voters surveyed giving Palin's speech an "A." But please note the fine print: The interviews were conducted earlier today and released by SurveyUSA at 4:55 Eastern time (according to the email release we received). That means they used their automated methodology to interview whomever happened to be at home during the day today. Do you think that might be a source of some bias in the results, even after weighting by age, gender and race? I'd say that is a very good possibility.
The bigger problem with this sort of survey is that it asks voters to react like pundits. How did Palin perform? Is she an "asset or a liability" to the Republican ticket? Does the selection "reflect well" or poorly on McCain? Only the last question begins to get at what matters, which is whether Palin's selection or speech has changed voter preferences or their underlying judgements about McCain or Obama.
The focus groups conducted last night by Democratic pollster Anna Greenberg for the Women's Voices Women Vote Action fund found results that are essentially consistent with grades that SurveyUSA's respondents gave the speech: Palin's speech "connected with these voters in a way that made her seem authentic, independent and strong," and her post speech favorability scores were improved. On the other hand, the groups showed no net gain for McCain in terms of vote preference. And as always, these were non-projective focus groups of just 22 voters.
Thus, it's still premature to conclude much about the electoral impact of last night's speech by Sarah Palin, and obviously we have not yet heard (as of this writing) from John McCain himself.
My own hunch -- and for the moment it is nothing more than that -- is that Republican consultant Mike Murphy is right. He likes Palin and had no complaints about her speech, but takes issue with the "politics of the choice:"
What I don't like is the effect I think Palin will ultimately have on the ticket. With all her charm, she is still a pick aimed squarely at the Republican base. In a high turnout Presidential year, I am not worried about turning out the base. I'm worried about everybody else we need to win and I fear that among those voters, Sarah Palin will be a dud.
I know, I know, she's a "hockey mom" and through the magic of identity politics she is going to make female voters swarm across party lines in numbers that Gerry Ferraro never dreamed of since this identity politics hokum is only a good idea that is certain to work when, um, we Republicans try to do it.
Instead, I think she'll ultimately be a polarizer. After last night's smash, Republicans are in deep love. Nothing thrills 'em like a good "us vs. them" speech. But I'd guess that most Democrats had the opposite reaction. In a year where the Democrat generic numbers are 10+ points better than the Republican, I don't like the math of a strategy that just polarized the election along party base lines. Among the vital sliver of voters in the middle, I think Palin's rock solid social conservatism will be a turn off. And while voters may value vision over experience, Palin's inexperience is a weakness, denying McCain an argument that has been helping him against Obama.
The survey data that will interest me most, once the dust settles next week, is that which attempts to test Murphy's argument. Does net effect of this Republican convention help move voters in the middle?