It's a funny thing. The economy isn't so great, and people are not thrilled with America's standing compared to four years ago. Yet, whether it's the Obama got Osama effect, or the incredibly shrinking GOP field (Pawlenty is officially in and Daniels is officially out, though neither was true during the poll) Obama's job approval is at 55% and most people think Obama will be re-elected.
Despite having a common opinion that the United States is in worse shape than it was four years ago, 57 percent of Americans widely believe that President Barack Obama will be re-elected in 2012.
According to a new Fox News poll , this number is about double of the 29 percent people who thought so in December.
And the 36 percent of voters who don't expect him to be re-elected is a major shift from the 64 percent who felt that way earlier in December 2010.
We saw something similar in the Politico/GWU Battleground poll last week: economic jitters but stronger than expected numbers for Obama.
The president’s strong approval ratings are buttressed by the 59 percent who said they will either “definitely” vote for the president or “consider” reelecting him. Thirty-eight percent “definitely will not” vote for the president’s reelection — giving Obama a higher ceiling of support than his Republican rivals would hope to see.
Last week's Battleground poll, however, didn't ask the public who they thought would win.
So how good are voter predictions, anyway? On the whole, decent enough. When you look at the general voter (as distinguished from the highly interested and informed who might play at an online market like Intrade, which as of this writing has Obama's reelection chances at 62.2%), voters have a pretty good idea of what's going on by taking it all in, even if they lack information on specifics. The 57% in this poll is based on zeitgeist, not specifics.
While researching the question, I came across a 1999 academic paper entitled Voters as forecasters: a micromodel of election prediction by Michael S. Lewis-Beck and Charles Tien:
Much presidential election forecasting research employs macromodels based on national economic and political fluctuations. Micromodels based on surveys of individuals exist, but they are almost entirely pre-election explorations of vote intention. What has been neglected are micromodels derived from vote expectations. We show, by analysis of the American National Election Surveys 1956–1996, that voters themselves can forecast who will win the presidential election. We go on to explain some of sources of this forecasting ability, and to evaluate its precision. Voter forecasting models emerge as a useful alternative to current approaches.
How useful this far out, though? That's another question altogether. I wrote the co-authors asking that question, and Professor Tien responded with: "People are pretty good predictors of election outcomes, but they are better the closer they get to the election date. We found a significant improvement in prediction when it was made after September 15 of the election year. Having said all that, my reaction to a citizen forecast at this point in time would be “voters are usually very good at predicting election winners.” Professor Lewis-Beck agrees, and notes the voter prediction model can be "quite potent".
Mark Blumenthal (pollster.com) adds: "We need to consider the "potency" of voter predictions in the context of the potential of the early primaries and the nomination process generally to change voter perceptions about candidate viability. Iowa, New Hampshire and the nominating conventions have a way of changing our perceptions of who can win."
Still and all, political junkies stay glued to the ups and downs of the current job approval numbers, and the polling on who people see themselves voting for, even though we are still a year and a half away (aka "a lifetime in politics". Political scientist-blogger John Sides from The Monkey Cage talks about early polling at CJR and cites Election Year's July as an accuracy "take-off point".)
So, the bottom line is that it's too early for any of the polling parameters to be definitive, but while following voter preference going forward, don't neglect voter predictions. Take a peek at polls you come across and see if the prediction is included. They add to our overall Big Picture of what's going on, and will be especially useful in the home stretch.
In the meantime, just note that there's been a big swing in that department from December to now, and while no guarantee of anything, or even that voter predictions will stay the same, it's likely more than statistical noise. Voter prediction rates as another interesting parameter to follow along with voter preference. The long list of Republicans who have decided not to run this year must be basing their decision on something, and that zeitgeist is felt by ordinary voters just as much as the political pros and party smart guys.
h/t pollster.com for the John Sides reference.
