At the McNamara Center on the U of M campus last night, five U political scientists provided a cornucopia of research-based political insights into what their profession has figured out about this election heading into the final two weeks.
The presentations ranged over several specialties and research interests, so instead of trying to jam into some artificial frame, I’ll just spill some of the most interesting things they said. All of the panelists, whose thoughts are paraphrased below, are U of M professors, emeritus, regents or otherwise:
Larry Jacobs on Colin Powell’s endorsement of Obama:
Political scientists tend to separate the personal qualities that voters perceive in candidates into two categories: interpersonal qualities (is this person like me, does he understand people like me) and performance-based qualities (is this person experienced, steady, tough, etc). Voters generally have liked Barack Obama better on the interpersonal side but appreciated John McCain’s greater level of experience, especially on foreign and military matters. But “in one quick sound bite [on Sunday’s “Meet the Press” interview] Colin Powell put a dagger into the McCain advantage in experience.”
Joanne Miller on the impact of cell-phone-only users on the accuracy of polling:
There’s a lot of talk about Obama support among younger voters being undercounted by pollsters because so many of them don’t have a landline and most polls don’t call cell phones. Apparently, Hispanics are also over-represented among cell-phone-only households and are another group in which Obama does well. Pew Research Center studied this by conducting a special poll that included a sample of cell-phone-only users. They found that cell-phone-onlys make up about 14.5 percent of likely voters and 30.6 percent of 18-29-year-olds. In this particular poll, taken a while back when the race was closer, Pew found that Obama and McCain were tied among landline users while Obama led by 19 points among cell-phone-onlys. But when Pew took into account their relatively small portion of likely voters, among the full sample including cell-onlys and landliners, Obama led by 2 percentage points. So it’s possible that the undercount of Obama supporters because of the cell phone factor might be about 2 percentage points (which is within the margin for error).
Bill Flanigan on the Bradley effect:
The “Bradley effect” refers to the belief that black candidates may underperform their late poll numbers because some voters lie to the pollsters to conceal the fact that they won’t vote for an African-American. Flanigan said the verdict of current political science is one-sided: that there is no such effect. The chief evidence: While Obama significantly underperformed his late poll numbers in the New Hampshire (leading to Hillary Clinton’s stunning comeback into the race), Obama also outperformed his late poll results in many other states.
Kathryn Pearson on late-deciding swing voters:
If you think of swing voters as people who are reading a lot of information, paying attention and mulling their vote until the last minute, you are almost 180-degrees wrong. Typically late-deciding swing voters are people who haven’t paid much attention and that’s why they’re telling pollsters they’re undecided. They simply haven’t been paying attention and won’t until the last minute.
John Sullivan on negative ads:
Negative ads tend to be more policy-oriented than positive ads. Voters retain the information in negative ads better and longer. Negative ads tend to raise the negative ratings of both the target and the candidate sponsoring the ad (so they tend to make more sense for the candidate who is trailing in the race). Negative ads will boomerang and do more damage to the person sponsoring the ad if the facts and arguments in the ad are judged to be unfair. People who are already on one side or the other are usually not influenced by the ads because they can think of counterarguments or because they derogate the source of the ad.
Jacobs on the U.S. system of requiring voters to register and on the spotty nature of the registration surges that have been reported in some states:
In most European democracies, registration to vote is universal and automatic. That’s one of the reasons voter participation is higher in those systems. The U.S. system that requires people to register in advance (except in states like Minnesota that allow same-day registration) doesn’t encourage universal participation in voting. Jacobs implied that this was by design. While voter participation in the United States is low by the standards of other democracies, if you express turnout on election as a percentage of registered voters, it reliably comes in around 80 percent. Quite simply, if more Americans were registered, more would vote.
He also said that the Electoral College system has an impact on registration and turnout. In competitive swing states, the parties and the campaigns are motivated to promote new registrations, which has led to double-digit increases this year in many states. But in about half of the states that are not competitive in the Electoral College, registration levels have stagnated or even declined this year (another seldom mentioned downside of the Electoral College system).
Pearson on how Hillary Clinton’s strategy backfired:
Clinton was trying hard not to reinforce gender stereotypes that have undermined previous female candidate. For example, her refusal to say that her vote in favor of the Iraq war had been a mistake was an attempt to show strength. But it backfired and helped unite the more anti-war elements of the Democratic Party behind Obama.
Flanigan on Hillary Clinton supporters’ attitude toward Obama:
The concern expressed at the end of the nomination contest that Democratic women who supported Clinton would not close ranks behind Obama is over. They have. But many men who supported Clinton have not.
Jacobs on the impact of presidential approval/disapproval:
Eighty to 90 percent of those who approve of the job the incumbent president is doing will vote to reelect that president or vote for the nominee of the president’s party. Eighty to 90 percent of those who disapprove of the job the president is doing will vote for his opponent or for the nominee of the opposition party. This is a short course on why President Bush’s historically low approval ratings are such a drag on McCain’s candidacy.
Party identification: Single biggest predictor of vote choice
In recent years, more and more voters are identifying themselves as Democrats, giving Democratic candidates an important advantage.
Pearson’s projection of the congressional elections:
Because the Dems had a big surge in the 2006 mid-terms — a gain of 30 seats in the House and six in the Senate — to take control of both houses, history would have suggested a pullback on this round as some of the more vulnerable Dem freshmen in red-leaning districts get defeated. But instead, the blue tide is still very strong. She projects a pickup of 15-20 House seats for the Dems and maybe the nine Senate pickups that Dems need to form a 60-member filibuster-proof majority.
Flanigan of the “embarrassing” inability of political science to understand why and how late-deciding swing voters finally do make up their minds:
“All these factors we’ve been talking about — partisanship, ideology, demographic characteristics, the candidates’ characteristics, the issues — are extremely predictive of vote choice for the electorate as a whole. But for people who make up their minds in the last two weeks, they’re almost totally unrelated to vote choice. That’s an embarrassment. Not just for us as analysts. We ought to be able to do better than that. But it’s also somewhat embarrassing to American democracy. You wonder what the hell is causing people to vote the way they do when they decide in the last two weeks if it’s not all of those factors. The simple answer is we don’t know… Maybe this year we’ll get lucky and figure it out. But I wouldn’t bet on it.”
Jacobs on the possibility that Gov. Tim Pawlenty’s veto power may lose its impact:
One of the less-noticed stories of this election year in Minnesota is that the blue trend is so strong that DFLers very well may achieve veto-proof majorities in both houses of the Legislature, which would make Pawlenty’s veto power almost irrelevant.
Miller on how the two presidential campaigns will use the last two weeks:
For Obama it will be identity messages: I’m like you. I’ll be good for people like you. For McCain it will be threat messages: It’s a dangerous world. Obama is a risky choice.
Sullivan on the endgame:
Sixty percent of Americans believe that McCain has run a negative campaign. He can’t change that now. Will the reintroduction of the Rev. Jeremiah Wright be the last arrow that McCain pulls out of his quiver? (McCain has until now forbid his campaign to exploit the controversial remarks of Wright, who was Obama’s long-time pastor.) It doesn’t matter if he does try to use Wright “because Americans have made up their minds.”
Flanigan’s mordant follow-up to the Rev. Wright idea:
If McCain emphasizes the controversies surrounding Obama’s former minister it runs the risk of informing the roughly 10 percent of Americans who think Obama is a Muslim that Obama is a Christian.
Jacobs on the audacity of Obama:
Obama is the “most audacious presidential candidate we’ve seen in a long time.” He is camping in the red states, trying to increase the number of states he can turn blue in 2008. “This guy is not playing to win. He is playing for a mandate and coattails. 2008 presents the most ideal set of circumstances we’ve seen in a long time for a partisan realignment election.”
Eric, great stuff, well worth reading, as always. Love the “mordant” comment about Rev. Wright.
I’d love to read more from these folks over the next two weeks, and longer. Get them back, give us more.