Estimated read time: 6-7 minutes
This archived news story is available only for your personal, non-commercial use. Information in the story may be outdated or superseded by additional information. Reading or replaying the story in its archived form does not constitute a republication of the story.
SALT LAKE CITY — With 40 days to go before Americans go to the ballot box to elect a president, voters will be inundated with polling data suggesting the election is moving in one direction or another. But does polling data really have an impact on the presidential race, especially with so many firms coming up with different numbers?
As of Thursday morning, at least 10 of the top polling firms have President Barack Obama leading Republican nominee Mitt Romney anywhere from zero to eight percentage points. Rasmussen Reports has President Obama and Romney tied, suggesting a close race. However, Pew Research has President Obama with an eight point lead, which could mean a significant victory in November. So why the difference?
"The difficulty is that these polls are making a prediction about a population that doesn't exist yet," said Quin Monson, director of BYU's Center for the Study of Elections and Democracy. "Normally you poll from a target population that you want to make inferences about, and that population exists — the general population of Utah adults or something like that. We know who those people are and we can identify them.
"For an election poll, you're trying to identify voters. But if you haven't had the election yet, that's not a population that you can easily sample from because, of course, the election hasn't happened yet," Monson continued. "That's the difficulty. There's a fair amount of uncertainty that gets thrown into it because you don't know who's going to vote. And we don't have full turnout in this country by any stretch."
The difficulty is that these polls are making a prediction about a population that doesn't exist yet.
–Quin Monson
And while polling firms attempt to approximate the general mood of the voters likely to cast a ballot this election, it is often the trend that is most likely the predictor of the race.
"The thing you look for in polls are trends, and then you try to figure out what those trends mean," said Kirk Jowers, director of the Hinckley Institute of Politics. "Some of the polls that talk about whether the country is on the right track or wrong track, whether they believe the economy is getting better, with those type of right track questions, it sometimes gets to the heart of what people feel without them trying to protect or harm their candidate of choice."
One such trend was released by Gallup on Tuesday, showing the increase in support of Americans trusting in government to handle international problems. The new polling numbers help the Obama administration, with more Americans trusting in his administration's ability to handle difficult situations.
Two-thirds of Americans (66%) said they have a "great deal" or "fair amount" of trust in the government, which is a steady increase since its low point in 2008. Even trust in domestic issues increased to 51 percent, following a decline in trust since 2001.
Related:
Additionally, the job creation index in the country has steadily increased since President Obama took office in 2009. Any trend moving in the positive direction will undoubtedly give the Obama campaign confidence, even though national polling numbers have the race relatively close.
However, trends change, even in a short period of time. In the world of Facebook and Twitter, public sentiment about a wide range of subjects can easily move, altering the state of the race. Additionally, the presidential debates will give Americans an opportunity to contrast the party platforms and candidates, which could alter polling and public sentiment about various subjects.
"What it will take to see significant movement is a major victory in a debate by one of the candidates. And that usually means a serious mistake on somebody's part — some show of temperament or lack of temperament," Monson said. "I think both candidates have substantial experience at debates, so the odds are they wrestle each other to a more or less standoff in the debates. Neither of them are so skilled in debating or so bad at debating that you would expect an easy victory."
#ElectionPoll
Nevertheless, the first debate may provide a more accurate description of American voters' attitude this election cycle.
"I think for this presidential race, the first poll that takes into account the first debate will be incredibly important," Jowers said. "Obviously things can change in a month, but that will really give you the best indication of where we really are."
"The important thing to note is that the portion of undecided (voters) is very low and has been for some time, which indicates there has been a fair amount of stability in the race," Monson said. "So those numbers haven't moved dramatically over the last few weeks, and there's not a lot of room for movement."
Early voting, however, is the wildcard in the presidential race that could alter the results of the polling data. On Monday, several states around the country opened up early voting, allowing voters more than a month to cast their ballot. Once cast, early votes are final, which means the upcoming debates or an "October surprise" will have little impact on early voters.
- By mail: postmarked by Oct. 9
- In-person & online: Oct. 22
Absentee voting:
- Miltary and overseas ballots: Sept. 22
- Mail-in ballots: Oct. 9
- All absentee ballots: November 2
Early voting:
- Begins: Oct. 23
- Ends: Nov. 2
- Click HERE to see your early voting location
General Election Day:
- November 6, 2012
- Polls open from 7 a.m. to 8 p.m.
In battleground states like Ohio, early voting is expected to surpass 2008 numbers, which could put polling data in question.
"I don't know that most of the polls factor in people that have already voted very effectively into their models," Monson said. "So, I don't know if we know too much about that frankly because the rise in early and absentee voting has been a relatively recent phenomena — just the past couple of elections especially."
And while polling data has a significant impact on the presidential race, it's the things beyond the candidates' control that can sometimes shape the race. Television programs like Saturday Night Live tend to give voters a lasting impression of the race, whether true or not.
"The bigger problem sometimes come from what Saturday Night Live and other more popular media — Colbert Report or the Daily Show — do with comments because that can really brand people," Jowers said. "Gerald Ford was probably one of the most athletic presidents we've had, but because of Chevy Chase we all think he was a huge klutz. Sarah Palin never actually said, ‘I can see Russia from my house,' but nine out of 10 people will swear on the Bible that she did because Tina Fey was so good at it."
Although polls give a snapshot in time, the results can quickly change. Four years ago at this point in the election cycle, Republican nominee Sen. John McCain had a three point lead, but in a matter of a week's time, the polls shifted in favor of then-Sen. Barack Obama by six points, giving him a three point lead.
"I think it will be relatively close," Monson said. "Even if it stays the way it is it will be close — it's close now."
"People seem to like Obama but they don't think he's capable," Jowers said. "And with Romney, they think he's capable, but he hasn't been as likable as people would like. It's incredibly close and if it keeps going the way it is, it will just end the way it is; it will just end up close."
General Election: Obama vs. Romney
| Poll | Date | Margin of Error | Obama (D) | Romney (R) | Spread |
|---|---|---|---|---|---|
| RCP Average | 9/25 | -- | 48.9 | 44.9 | Obama +4.0 |
| Rasmussen Tracking | 9/25 | 3.0 | 46 | 46 | Tie |
| Bloomberg | 9/24 | 3.5 | 49 | 43 | Obama +6 |
| Gallup Tracking | 9/25 | 2.0 | 50 | 44 | Obama +6 |
| Politico/GWU/Battleground | 9/20 | 3.1 | 50 | 47 | Obama +3 |
| National Journal | 9/19 | 3.0 | 50 | 43 | Obama +7 |
| Associated Press/GfK | 9/17 | 4.3 | 47 | 46 | Obama +1 |
| NBC News/Wall St. Jrnl | 9/16 | 3.6 | 50 | 45 | Obama +5 |
Battleground States: Obama vs. Romney
| State | Poll | Results | Spread |
|---|---|---|---|
| Colorado | Gravis Marketing | Obama 50, Romney 46 | Obama +4 |
| Florida | CBS/NYT/Quinnipiac | Obama 53, Romney 44 | Obama +9 |
| Iowa | PPP (D) | Obama 51, Romney 44 | Obama +7 |
| Massachusetts | Rasmussen Reports | Obama 55, Romney 40 | Obama +15 |
| Michigan | Rasmussen Reports | Obama 54, Romney 42 | Obama +12 |
| Nevada | PPP (D) | Obama 52, Romney 43 | Obama +9 |
| North Carolina | Civitas (R) | Obama 49, Romney 45 | Obama +4 |
| Ohio | CBS/NYT/Quinnipiac | Obama 53, Romney 43 | Obama +10 |
| Pennsylvania | CBS/NYT/Quinnipiac | Obama 54, Romney 42 | Obama +12 |
| Wisconsin | WeAskAmerica | Obama 53, Romney 41 | Obama +12 |














