Prior to the first Obama-Romney debate, some critics blasted polls as Democratic-leaning.
The more accurate statement: Polls reflect the people more willing to answer pollsters at the time of the poll. And now, it seems Republicans are the more motivated interviewees.
Before the debate
Before the debate, Democratic voters were motivated. After the conventions, President Obama’s lead had been growing slowly but steadily. A September poll from the Pew Research Center for the People and the Press showed Obama with a 51-43 lead over Romney.
A look at the poll’s sample shows a greater number of Democratic voters responding:
Soon after, Republican candidate Mitt Romney committed several gaffes (e.g. his reaction to the attack on the U.S. consulate in Libya and the “47 percent” comment), and it seemed Obama’s lead was growing. Some Republicans even came out publicly with their frustrations with Romney’s campaign.
After the debate
But then came the first debate, which, by most accounts, Romney won handily. Republicans and conservatives were energized by his feisty, focused performance.
And, it seems, they became more willing to answer pollsters’ questions.
The latest Pew poll released Oct. 8 showed Romney above Obama among likely voters, with a 49-45 advantage.
Unlike earlier polls, Pew captured marginally more Republican voters than Democrats:
Does that mean the poll is “biased”? Pollsters say no because you are grabbing a random sample of the entire U.S. population, which gives everyone an equal shot at participating. Polls are a moving target, something that the sage Nate Silver reminds of consistently in his must-read FiveThirtyEight blog; it’s best to look at a collection of polls and make sense of the aggregated data.
One concrete conclusion we can draw: The race is extremely tight, and at this stage, it’s impossible to know how many of those who say they support Obama or Romney today actually come out to vote for those candidates on Election Day.