Analysis: Polling criticism unfounded
September 28th, 2012
05:32 PM ET
2 years ago

Analysis: Polling criticism unfounded

Washington (CNN) – 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.

– Follow the Ticker on Twitter: @PoliticalTicker

– Check out the CNN Electoral Map and Calculator and game out your own strategy for November.

Tune into conservative talk radio or search the Internet and there's a good chance you will see or hear the charges. Even from Romney surrogates.

"So there's a number of things going on with the polls, folks," Rush Limbaugh said Thursday on his radio program. "Let's review just quickly some things said yesterday. One of the reasons that you tout a big lead, at the same time early voting begins, is to suppress Romney early voters. Who wants to vote for a loser? Oh, gosh, here we go again. We really don't have a chance. They know that so many people on our side, all you gotta do is listen to the average conservative pundit and realize that they're mired in negativity and pessimism. So you build it up, put out a poll that shows a massive Obama lead and early voting goes on, and so you stoke Obama voters, you depress Romney voters."

A question you might ask yourself: Would Limbaugh be attacking these media polls if the results showed Romney leading?

Republican strategist Karl Rove stoked the same conspiracy flames earlier this week on Fox News.

"You've got to be careful about these polls," Rove said. "We endow them with a false scientific precision they simply don't have."

The criticism has ramped up over the past two weeks after a series of polls in key battleground states showed the president gaining an advantage over Romney. Heading into the homestretch of the long campaign, the results are a troubling development that Romney's top aides are being forced to address. And part of the strategy is criticizing the polling results.

"They have a Democratic voter participation that is higher than the participation in the electorate in 2008," Ed Gillespie, a senior Romney adviser said on Fox News this week. "I don't know anyone on the ground in any of these swing states who believes that there will be a higher Democratic percentage of the electorate in 2012 than there was in 2008."

And Romney surrogate Bay Buchanan, without an understanding of the facts, charged that CNN "deliberately" over-polled Democrats in a recent survey.

"There's a CNN poll last week," she said in an appearance on CNN. "We won in the poll the independents by 14 percent and we won Republicans by 98 percent. And somehow the CNN poll had us losing by six. Now, you and I both know if we win 14 percent of independents we're going to have a mighty good day. So what is that about?"

A top Romney campaign strategist said it uses its own internal polling when making decisions.

"The public polls are what the public polls are," Rich Beeson, Romney's political director, told reporters earlier this week. "I kind of hope the Obama campaign is basing their campaign on what the public polls say. We don't. We have confidence in our data and our metrics. I feel confident where we are in each one of our states. I have great faith in our data."

Steve Doocy of Fox News has expressed skepticism about polling results for this election.

"Could there possibly be some skewing going on by the media, the left-based mainstream media?" Doocy said.

Doocy's own network, though, has published polls that were very much in line with the topline results from other media surveys.

A Fox News poll of Virginia voters indicated Obama had a seven point (49%-42%) advantage over Romney. A Quinnipiac University/CBS News/New York Times survey in Virginia showed Obama up only four-points (50%-46%) over Romney, while a Washington Post poll had Obama with an eight point (52%-44%) lead over Romney in the Commonwealth.

In Ohio, a Fox News poll said Obama had a seven point (49%-42%) advantage over Romney, while an Ohio survey by The Washington Post put Obama up eight points (52%-44%) over Romney and a Quinnipiac University/CBS News/New York Times poll showed the president with a 10 point (53%-43%) lead over Romney.

A Fox News Poll of Florida voters showed Obama with a five point (49%-44%) lead over Romney, while The Washington Post survey of Floridians said Obama was up four points (51%-47%) on Romney and a Quinnipiac/CBS News/New York Times Florida poll indicated that Obama had a nine point (53%-44%) lead on Romney.

Hardly a concerted effort to skew poll results in favor of Obama - unless you agree that Fox News is part of the conspiracy.
Yet, not all conservatives are accusing the media of manipulating the data.

"I do not believe the polls are all wrong," Erick Erickson, editor of the influential conservative web site RedState.com, wrote on Wednesday. "I do not believe there is some intentional, orchestrated campaign to suppress the GOP vote by showing Mitt Romney losing. I actually believe that Mitt Romney trails Barack Obama.

"But I also believe the polls are reflecting a bigger Democratic strength than is really there," added Erickson, who is also a political contributor on CNN.

Criticizing public polling is hardly a new phenomenon. In fact, it's a regular campaign tactic.

What is new in this election cycle is that several polling organizations have started releasing their party identification numbers. And critics are seizing upon this information to formulate a flawed argument attacking the results, comparing party identification from these telephone polls to previous exit polls. It's not a valid comparison.

Interviews conducted by telephone prior to an election are much different than talking to a person who has just voted for a candidate. In a telephone interview several months or several weeks before an election, a person might provide a different answer than the one they give after emerging from a voting booth.

People's minds can change … which is exactly why there is such an effort by the campaigns to convince the undecided and soft-leaning voters to support their candidate.

Keep in mind, party identification is not a characteristic that is set in stone such as someone's race or sex. It can evolve over time and change much like education and income levels. A study conducted a few years ago interviewed the same people six different times during a presidential election and it found that 25 percent of the participants changed their answer on party identification at least three times during the course of the study.

Republicans made similar arguments in 2008 when they compared polls taken in that cycle to the 2004 exit polls. And Democrats also voiced the same argument in 2004 by comparing surveys in that election cycle to the exit polls from the 2000 election. In 2008, Democrats won the White House and only four years earlier President George W. Bush won a second term.

Frank Newport, editor-in-chief of Gallup, addresses the party identification argument in a September 27, 2012 column. If you are interested in this argument, we think it is worth a read. As is this September 19, 2012 column by Chris Cillizza of The Washington Post.


Filed under: 2012
soundoff (270 Responses)
  1. shruth

    Hey, let Republicans give themselves a 'pep' talk by denying facts and conjuring up conspiracy theories! May be they would then stop pouring Koch and Rove money into putting on rdio and TV ads full of lies and distortions! They have never had any use for facts or stats except when they favour them! Like the polls that indicate that the Affordable Care Act is not popular! Now Repubs would swear by those polls! One of the commenters here claims that polls are possibly inaccurate due to assumptions about turn out, noting that there are not many yard signs or Obama T-shirts like in 2008. Well, I see similar trends with Romney/Ryan as well, compared to McCain/Pailn signs in 2008. So I do not see that observation providing much support to the commenter's arguement!

    September 29, 2012 11:00 am at 11:00 am |
  2. mike

    Rush is a big fat hypocrite doper. Who would possibly listen to a word he says? He is a cheater, he demeans women and he lies. . What a vomit this creep is. I got two words for him, diet and exercise. All you sheep keep buying that garbage he hawks. I cant wait for the Rush toilet seat. Now that is American!!

    September 29, 2012 11:05 am at 11:05 am |
  3. kyphi

    True, I have not seen as many Obama yard signs; neither have I seen Romney yard signs. Vindictive people of both parties are stealing them. It's "blame the messenger" this year according to talk radio. People, change the station and listen to music instead. Your blood pressure, heart, and state of mind will be better off for it. May you have a good restful day.

    September 29, 2012 11:07 am at 11:07 am |
  4. shuman

    I thought this article was going to explain why oversampling Democrats in these polls is the way to improve predictive accuracy, but no. Just another factless innuendo piece.

    September 29, 2012 11:11 am at 11:11 am |
  5. McRCN

    Polling organizations just need to provide more facts about how the poll was conducted and let the reader make his or her own interpretation. I am getting tired of being told what to think by so called news organizations. Nothing worse than to hear about a poll showing a candidate way up and the results blasted all over the airwaves, but come to find out a day or two later it was a bias poll. No one goes back to retracts their remarks afterwards after advertising a bias poll and not everyone finds out about how the poll was conducted.

    September 29, 2012 11:11 am at 11:11 am |
  6. TruthHurts

    So multiple un-affiliated independant polls are all involved in an elaborate conspiracy to reelect Obama? Even Fox news' polls show that Obama has an advantage. Is Fox News involved in this leftist conspiracy? Romney... You reek of desperation. Quit now before you embarrass yourself further.

    September 29, 2012 11:19 am at 11:19 am |
  7. joe

    Poor Obozo! He and his media campaign numbskulls, they just dont get it. They will get the dumb vote, i.e the 19 percent Liberals in the country the media is making sure that they keep dumb, Dumber! So goodbye B.O and take your stupid, racist VP. With you!

    September 29, 2012 11:21 am at 11:21 am |
  8. jonat

    The lib media is skirting the issue as usual. The claim is that liberal media polls are based on skewed data based on the 2008 election. The polls sample more Democrat voters based on the percentage of registered voters in 2008. The problem with that is that Democrat registered voters have declined and Republican registrations have increased (thanks to voters fed up with obama). So the real issue is accuracy. Of the top twenty national polls, Rasmussen was the only one that got the number right in 2008. Their current poll shows a tie between Romney and Obama

    September 29, 2012 11:21 am at 11:21 am |
  9. Jim

    There's only one thing missing from this lengthy report: a denial that poll results are adjusted to reflect the pollsters opinion about how many Dems and how many Reps will vote in November. I wonder why they aren't able to poll 1000 or 2000 people and just publish the raw results. Will they adjust the actual vote numbers on election day as well?

    September 29, 2012 11:28 am at 11:28 am |
  10. Terri

    Keep dreaming, Carol that your mittens could possibly ever win!

    September 29, 2012 11:31 am at 11:31 am |
  11. Alex

    Jah, Romney is right: 47% bums, all democrats.

    September 29, 2012 11:32 am at 11:32 am |
  12. B.

    Rush Limbaugh, National Disgrace and a symbol to the World of why progress takes a long time in this country!

    September 29, 2012 11:32 am at 11:32 am |
  13. 1608

    What say you about a fox polls? This must be a Rush thing.

    September 29, 2012 11:33 am at 11:33 am |
  14. AB

    @Carol,

    You don't see a slew of Obama regalia because it's not new, it's not in, but most of all it's not needed. Obama was dealt a crappy hand and has made some tough decisions with very little help from Congress. No one is saying otherwise, but to think Romney will do better is ludicrous.

    No one knows who Romney is. He's secretive about everything. He's made a lot of money, but won't say how. He won't talk about his religion. He won't stand on an issue and explain it. He won't take a side, other than retreating from his own rhetoric. He only want's the job as POTUS to fulfill some inner need/bucket list thinking he will glide through his duties. He cares nothing for the common American or their plights. With all he's said and done, it's easy to see where his real loyalties lie.

    Romney is a man of no substance.

    September 29, 2012 11:38 am at 11:38 am |
  15. Dizzyd

    And if the polls showed Romney leading, these same guys would be saying 'the polls have spoken! They're completely true!' lol

    September 29, 2012 11:46 am at 11:46 am |
  16. Suck

    It is time to reconsider freedom of media. NoBama!

    September 29, 2012 11:50 am at 11:50 am |
  17. Jennifer

    Is it just me, or is there something very entertaining about watching GOP's/ tea baggers/ birthers sweat and squirm nervously?

    September 29, 2012 11:50 am at 11:50 am |
  18. Suck

    Let's have media select a president!

    September 29, 2012 11:52 am at 11:52 am |
  19. Jadie

    Republicans are being criticized for challenging polls that they feel are skewed in favor of the Democratic party. Yet, in 2004 Democrats were the ones criticizing polls because they were skewed in favor of Repbulicans. Why is it when they are go against the Democratic party it is okay to question their authenticity, but if it seems skewed against Republicans they are conspiracy theorists. Polling is only a guage and what really matters is who actually shows up to the polls and votes. Finally, I am also amazed that Carol had an intelligent, well-worded, fact based thought with no malice toward Democrats, yet the response was a negative rant at Republicans and not one with data and any statistical backing.

    September 29, 2012 11:53 am at 11:53 am |
  20. Robertmiii

    As long as Obama can hold that 8 point lead until Nov 6th I don't care what anyone says or does.. my remaining fear is that Romney may do well in the debates-that could change the race. Romney tends to flub and stutter when thinking on stage though.. If Obama can doge the post Benghazi bullets he should win the election with close to 300 electoral votes..

    September 29, 2012 12:06 pm at 12:06 pm |
1 2 3 4 5 6 7 8 9 10 11