Category Archives: Polls

State polls or national polls?

“There have been plenty of elections before when the outcome was highly uncertain down the stretch run or on Election Day itself. But I am not sure that there has been one where different types of polls pointed in opposite directions. Anyone in my business who is not a bit terrified by this set of facts is either lying to himself — or he doesn’t know what he’s doing.”

That’s Nate Silver, host of the influential FiveThirtyEight blog of the New York Times. He explains the issues surrounding national polls that say one thing and state polls that indicate another.

— John Robinson

Who won the VP debate?

The real results to that question won’t be known for awhile (if ever). But overnight results are in.

CBS: 50% of uncommitted voters gave it to Vice President Joe Biden and 31% said Rep. Paul Ryan won. (19% said it was a tie.) “The percentage of voters who say they believe they can relate to Biden spiked from 34 percent before the debate to 55 percent; 48 percent think Ryan is relatable, up from 31 percent before the debate.”

CNN: Among voters, it went 48% Ryan, 44% Biden, (within the margin of error). “Seven in ten said Biden was seen as spending more time attacking his opponent, and that may be a contributing factor in Ryan’s 53%-43% advantage on being more likable. Ryan also had a slight advantage on being more in touch with the problems of average Americans.”

What does it mean? Well, not a whole lot. But satirist Andy Borowitz of the New Yorker had one faux pollster’s take: “In a poll of Democratic voters taken immediately following Thursday night’s Vice-Presidential debate, a wide majority said they wanted Vice-President Joe Biden to appear in all remaining 2012 debates.”

— John Robinson

Stop polling in North Carolina? Never!

The president of Suffolk University Political Research Center made a bit of a splash earlier this week when he said that the center had stopped polling in North Carolina, Virginia and Florida — three swing states.

 “I think in places like North Carolina, Virginia and Florida, we’ve already painted those red,” David Paleologos said. “We’re focusing on the remaining states.”

He later told Huffington Post: Paleologos said that Obama has consistently polled at 47 percent of the vote in the three states, making a win for him “not impossible, but highly improbable,” especially in North Carolina. The state, which Obama carried in 2008, is currently trending red, and several pollsters agree that it’s one of the strongest battleground states for Romney.”

The good folks at the Huffington Post asked Elon University Poll Director Ken Fernandez his thoughts.

“We’ve seen the lead changes in North Carolina a dozen times and results have almost always been within the margin of error. I don’t think you will see many well-funded polling organizations pulling out of North Carolina, Florida, or Virginia,” he said.

Fernandez noted that pollsters with limited time and money might choose to focus on different states, but “it seems premature to make a methodological decision based on a single campaign event and poll change,” he said.

Fernandez pointed to 1948, when Gallup stopped polling two weeks before the election and predicted that Thomas Dewey would defeat Harry Truman. “When you make such a decision with incomplete information, you are increasing your chances of being wrong, very wrong.”

The Elon University Poll isn’t abandoning North Carolina. It goes into the field with another statewide poll later this month.

— John Robinson

Who won the debate?

Who won last night’s presidential debate?

Overnight polls say Gov. Romney in a landslide.

CNN— “…67% of debate watchers questioned said that the Republican nominee won the faceoff, with one in four saying that President Barack Obama was victorious….While nearly half of debate watchers said the showdown didn’t make them more likely to vote for either candidate, 35% said the debate made them more likely to vote for Romney while only 18% said the faceoff made them more likely to vote to re-elect the president.”

CBS“…46 percent of voters gave the economy-centric debate to Romney, 22 percent said they believed the president was the winner, and 32 percent called it a tie. More good news for the GOP nominee: 56 percent of those polled said they viewed Romney in a better light after watching the debate.”

And an analysis from the New York Times’ Nate Silver: “There is not a lot of empirical research on the relationship between instant reaction polls and their eventual effect on the head-to-head polls. However, these were strong numbers for Mr. Romney where comparisons to past post-debate polls are available”

His conclusion on the FiveThirtyEight blog: “My own instant reaction is that Mr. Romney may have done the equivalent of kick a field goal, perhaps not bringing the race to draw, but setting himself up in such a way that his comeback chances have improved by a material amount.”

— John Robinson

 

Are the polls purposely skewed?

There has been much discussion lately — mostly by Republicans — that polls showing President Obama leading the presidential race are skewed. The answer is no. Here’s a roundup.

CNN— It has the best description of the “discussion.”

“It’s a conspiracy theory of the highest level: media organizations allegedly manipulating data in public opinion polls to try and help President Barack Obama win a second term. Democracy crushed. The accusations are predicated on the idea that some media organizations are interviewing too many Democrats in their surveys, which skew the results in way to benefit Obama over Republican presidential nominee Mitt Romney.”

Washington Post — The best explanation of polling. “What all of the above points to is the reality that polling is equal parts art and science. The best of the best — like the folks at the Post — understand that putting together the sample for any poll involves weighing what we know the electorate looked like in the past with what it looks like today and what it will look like on Nov. 6.”

Gallup — It has the best defense of polling. “Interested observers often opine that when a given poll shows that Candidate X is ahead, it cannot be correct because there is a higher percentage of voters who identify with Candidate X’s party in the sample than there should be, based on comparison to some previous standard. There are several reasons why this is a faulty approach to evaluating a poll’s results.”

Slate — It has an interesting “unskewing” of polls. “Dean Chambers, who runs unskewedpolls.com, quicky worked his usual magic on the Fox data but this time his “unskewing” wasn’t enough to move the numbers in the GOP’s favor.”

— John Robinson

Obama-Romney in N.C.: It’s tight, still

While the Elon University Poll conducted last week reported that Gov. Romney has a 47%-to-43% lead over President Obama in North Carolina, other polls of the state reported slightly different results.

Public Policy Polling, which leans Democratic, has the race tied up, 48%-48%. A High Point University/Fox8 poll has Romney at 46% and Obama at 43%. Interestingly, a few days earlier, HPU reported poll results showing the race deadlocked at 43%. And a Civitas Poll, which leans Republican, has the race in a dead heat at 45% each.

But if you’re serious about comparing, make sure you read the methodology and the margin of error at each of the sites. They differ vastly.

— John Robinson

Stressed? At least it’s not caused by the political campaigns

As Isaac bears down on the Gulf Coast, the presidential race is deadlocked. Still projections continue.

Gallup looked at the results of the 15 presidential elections since 1952 and determined that, “All else being equal, the leader of the Gallup poll prior to the convention has an 80% probability of winning the election, according to past data.

“Of course, all else is not equal. When pre-convention polls show a tight race, as is the case this year, conventions have been more likely to create new leaders or galvanize support for a heretofore weak leader. Thus, both President Obama and Romney have the potential this year to gain an upper hand as a result of the convention process.”

So who is ahead? Gallup won’t post that information until later today.

Meanwhile, a CNN/ORC International poll gives President Barack Obama the nod on social issues, and Mitt Romney has the edge on the economy, which is what Americans say is their No. 1 concern. But the poll also makes clear that “there are a lot of issues where neither candidate has an appreciable edge, and many of them – including welfare, taxes, health care, and Medicare – have been fodder recently for both presidential campaigns.”

So, to put it all into perspective, Harris Interactive reports that 73% of us are stressed at work. But it’s not the political campaign that is causing it. Instead, “The top source of stress, felt by 11 percent of survey respondents, is low wages. For women, that rises to 14 percent.”