Understanding Political Polls
http://thedianerehmshow.org/shows/2012-10-22/understanding-political-polls
Voters look to political polls to understand who is winning the presidential race. But interpreting these numbers can be difficult. Diane and guests explore the art and science of political polls.
Guests
Andrew Kohut
director of the Pew Research Center for the People and the Press
Whit Ayres
president and founder of North Star Opinion Research
Mark Blumenthal
senior polling editor of the Huffington Post
Comments
Please familiarize yourself with our Code of Conduct and Terms of Use before posting your comments.
Speaking of biased polls, have your guests never heard of what's called "push polls?" They're used frequently in local races in North Carolina and while I haven't gotten one this year were also being used during the 2000, 2004, 2006 and 2008 campaigns.
There are many reasons for having a landline. My husband and I both have a cell phone for work but if you have children (who are too young for cell phones) you NEED a landline. We are as liberal as they come so there are many families who will lose their landline when their kids are old enough.
But she has a (R) pollster as a guest. C'mon! Cut down on the caffeine.
KadeKo wrote:
"Oh, and while we're on the subject, please, more on how crap polls (hello, Rasmussen!) "
Crap why? Because they've been the most accurate in recent years?! Gallup, also has a good track record.
Compare their accuracy to NBC/WSJ or NYTs. Then you will see what is crap.
We can't trust the main stream media media on any issue being covered especially on politics, so maybe it's better they stay busy with their bogus poles. The last presidential election Rasmussen and Pew were the most accurate so I use these as my public barometer. Poles are fun if you like the horse race, I do. The only problem I have is bad polling and the news organizations behind them are forgotten in the next election cycle, we need to be constantly reminded who had the most inaccurate pole numbers last time so we know which ones should be taken seriously this time.
which do you think are more reliable, exit polls or machine counts?
KadeKo wrote:
"I'll remember to call you as a character reference if I ever want to go AWOL from military service."
You are channeling that disgraced talking head, Dan Rather.
Third party candidates are often not included in the polls. The reason given for this is that they don't poll high enough to be included, but isn't that a catch 22? If polls say Obama, Romney or Other...How can someone like Gary Johnson get a rating in the poll? If most polling information doesn't include a third party candidate, how can people hear the name and research it? The voters are not informed and the polls are perpetuating that. Gary Johnson has recently polled high enough to be included in the Presidential Debate, but he is still being blocked. This needs to be discussed. A law suit is in the courts today to determine if he can participate in tonights debate.
In Ohio & Wisconsin pollsters are calling low income neighborhoods, asking if people knew that "Voter Fraud is a felony, punishable by up to 3 ½ years and a $10,000 fine"
How do polls take into account people like me who are simply fed up with intrusive phone calls and who have given up answering the phone until after the election. I would never think of responding to a political poll. It's nobody's business.
edwards wrote:
In Ohio & Wisconsin pollsters are calling low income neighborhoods, asking if people knew that "Voter Fraud is a felony, punishable by up to 3 ½ years and a $10,000 fine"
Good point, and how to pollsters take
this into account?
@ecgbehrt,
Yes, Rasmussen isn't "accurate" now.
From Labor Day to Halloween, they have their thumb on the scale for Republicans. Nobody knows their methodology. And when Election Day rolls around, they have "closing speed", and start to agree with other pollsters.
When it comes to making headlines and influencing narratives, Rasmussens' job is to make everything "Good news for Republicans". And that is what they do.
They've got a track history of outliers in Septembers and Octobers. This is a feature, not a bug.
If they're supposed to be a compass, they're as reliable as a compass next to a magnet.
I'd like you to explore how many polls exclude other viable candidates in their polling. Candidates like Gary Johnson of the Libertarian Party hardly ever get a mention. Pollsters call and when confronted with a Gary Johnson supporter, list them as 'undecided'. This is a deceitful travesty. Particularly considering that the Commission on Presidential Debates bases their 'criteria' on polling results. How can a candidate reach all of the criteria if they aren't included in the polls?
KadeKo wrote:
"they have their thumb on the scale for Republicans. Nobody knows their methodology."
Side by side, those are two fascinating statements!
Here's the problem with your analysis, KadeKo:
First, this from wiki
2004
In the 2004 presidential election, "Rasmussen...beat most of their human competitors in the battleground states, often by large margins," according to Slate magazine. Rasmussen projected the 2004 presidential results within one percentage point of the actual vote totals earned by both George W. Bush and John Kerry.
In 2004, Slate said they “publicly doubted and privately derided Rasmussen” polls because of the methodology. However, after the election, they concluded that Rasmussen’s polls were the most accurate.
2008
According to Politico, "Rasmussen’s final poll of the 2008 general election — showing Obama defeating Arizona Sen. John McCain 52 percent to 46 percent — closely mirrored the election’s outcome." In reference to the 2008 presidential election, a Talking Points Memo article said, "Rasmussen's final polls had Obama ahead 52%-46%, which was nearly identical to Obama's final margin of 53%-46%, and made him one of the most accurate pollsters out there."
Now, here's what I want to know from you. The only poll that is truly verifiable is the one on election day. In those cases, Rasmussen has been the most accurate. On what basis do you then say, "Yes, Rasmussen isn't "accurate" now. "? Because their results differ from other polls? Since Rasmussen is most accurate on election day, how do you know they are anything other than the most accurate today?
You say, "When it comes to making headlines and influencing narratives, Rasmussens' job is to make everything "Good news for Republicans". And that is what they do. They've got a track history of outliers in Septembers and Octobers.".
So? I want to know what your objective evidence is that they are wrong and others are right. Do you see the problem?
It has been said that people lie to gain an advantage. As far as I'm concerned, that makes polls irrelevant. Not to mention the interpretation of the polls. How do we know that the poll takers (and poll interpreters) aren't unbiased?
We don't!
I live on the west coast in Washington state. Polls from the eastern states announce the winner before we are finished voting. This leaves us feeling like our votes do not count. Why do we not postpone announcement of exit polling until all results are in?
Fivethirtyeight
November 4, 2010, 10:41 pm
Rasmussen Polls Were Biased and Inaccurate; Quinnipiac, SurveyUSA Performed Strongly
By NATE SILVER
Every election cycle has its winners and losers: not just the among the candidates, but also the pollsters.
On Tuesday, polls conducted by the firm Rasmussen Reports — which released more than 100 surveys in the final three weeks of the campaign, including some commissioned under a subsidiary on behalf of Fox News — badly missed the margin in many states, and also exhibited a considerable bias toward Republican candidates
See the above dated article for the complete report on what can only be considered the charlatan of polling, Scott Rasmussen
@Etaoin Shrdlu: How about that Etaoin.... a Canadian Republican!!! And by the way...admitting that you attended Dalhousie explains a lot about your comments.
Jim Gamble wrote:
"See the above dated article for the complete report on what can only be considered the charlatan of polling, Scott Rasmussen"
I'm not positive, but I don't THINK that 2010 was a Presidential election year ... but I could be wrong.
I've been looking for "Nate Silver's" analysis of the "charlatan of polling, Scott Rasmussen"'s results in 2004 and 2008, which, like 2012 are Presidential election years, but I seem unable to find anything.
Fact: Obama won the last election by a 7.3 margin, 52.9 to 45.6 percent
Fact: CNN & Ipsos came closest with Obama winning by a 7-point margin 53/46
Fact: Rasmussen & Pew underestimated Obama with a 6-point margin, 52/46
Fact: Gallup overestimated Obama with an 11-point margin, 55/44
The 105 polls released in Senate and gubernatorial races by Rasmussen Reports and its subsidiary, Pulse Opinion Research, missed the final margin between the candidates by 5.8 points, a considerably higher figure than that achieved by most other pollsters. Some 13 of its polls missed by 10 or more points, including one in the Hawaii Senate race that missed the final margin between the candidates by 40 points, the largest error ever recorded in a general election in FiveThirtyEight’s database, which includes all polls conducted since 1998.
Well done, Numeros!
After the 1992 election there was a two panel political cartoon that ran in Newsweek (I think) which perfectly captured my feelings about polling. It consisted of a man talking to a woman.
Panel one:
Man: I bet you're happy now that the election is over.
Woman: I'll say!
Panel two:
Man (pulling out a clip board and pencil): Are you happier, less happy, or as happy as you were four years ago?
Woman: AAAAGGGGHHHHH!
Hey! Nice going Numeros! A whole point difference! (on only one side, mind you). Now, how did CNN or Ipsos do in 2004? Consistency is the name of this game, you know. (Hint, Ipsos back then was Ipsos-AP, not Ipsos-Reuters)
Moreover, Rasmussen’s polls were quite biased, overestimating the standing of the Republican candidate by almost 4 points on average. In just 12 cases, Rasmussen’s polls overestimated the margin for the Democrat by 3 or more points. But it did so for the Republican candidate in 55 cases — that is, in more than half of the polls that it issued.
If one focused solely on the final poll issued by Rasmussen Reports or Pulse Opinion Research in each state — rather than including all polls within the three-week interval — it would not have made much difference. Their average error would be 5.7 points rather than 5.8, and their average bias 3.8 points rather than 3.9
Moreover, in the last two Presidential races, Rasmussen has done the best job of predicting the final result. That is a fact. That is not an opinion. What everybody cares about the most is the Presidential race. Should Rasmussen nail this race too, it will be very difficult for Democrats to dismiss him - including Nate Silver. His blog says he supports Barrack Obama, which tells me that, though he might be really smart, he's not very sensible.
Silver did a very good job in 2008, but it was a landslide for Democrats - so, not that tough.
There is no evidence, even based on the 2010 mid-terms, however that Silver's bias is any less than that attributed to Rasmussen by the left, whether or not a Rasmussen bias exists. Silver's predictions favored Democrats in 2010 and he was wrong. He under-predicted the Republican landslide in the House by almost 15%.
The bottom line, however, before I bow out of this discussion is that opinion polls are not called "opinion" polls for nothing - and a lot of discussion about them is little more than opinion either.
The poll that matters is the one two weeks from Tuesday. That's the one I'm waiting for.
Speaking of opinion, here's the first impression of the debate from Political Wire:
Reaction to the Third Presidential Debate
The third and final presidential debate was President Obama's best moment in the campaign so far. He was prepared on every issue and knew Mitt Romney's record of past statements just as well.
Obama succeeded because he conveyed his unique view of the world from the Oval Office. For undecided voters watching, all they probably heard was that he's the commander-in-chief. And that's what Team Obama wanted.
For the most part, Romney made an effort to look presidential by not attacking. He was exceedingly careful and desperately tried not to make a mistake. In fact, despite his rhetoric for the last two years, he now apparently agrees with most of the Obama administration's foreign policy.
As a result, Romney's biggest opponent was not the president, it was his own words. Obama did a brilliant job of bringing up past Romney statements -- on Iraq, on the nation's biggest adversary, on Afghanistan, on Osama bin Laden -- to make him look unprepared for the presidency.
As the debate went on, Romney tried many times to move the international affairs discussion back to the economy where he was more comfortable. It was as if he had only 30 minutes of foreign policy talking points for a 90 minute debate. As a result he seemed to string together random thoughts which often made him sound incoherent.
Obama won the debate hands down.
Bob Schieffer started off doing a good job as moderator, framing questions but still letting the candidates engage each other. But he lost control of the debate as both Romney and Obama often preferred to talk about the economy. It's clear that both candidates know that most voters don't care much about foreign policy.
When The Church of Jesus Christ of Latter Day Saints (LDS) - The Mormons, drastically downsized The Deseret News (the Utah church owned newspaper) the top LDS church officials dispersed many of the writers (and other LDS faithful) to go worldwide to function in the capacity of contributing writers to the media (radio, tv, magazines, and newspapers. Let's not forget about the possibility of the LDS church historians working the internet and public opinion polls. Their contributions to the media and internet have been most apparent during this 2012 Presidential Election with alot of the earlier pro Romney and pro LDS/Mormon media coverage. What about the millions of dollars spent by the LDS church for their prime time "I'm a Mormon" television campaign under the guise of religious freedom. I cannot help but wonder if all the new polls are LDS church historians or LDS faithful professionals (Whit Ayers) again making contributions to the media to manipulate public perception of the presidential election in Romney's favor. Diane, did you invite Whit Ayers to be on your show or did he contact you to offer his perspective and presence as an "Ambassador"?
Fact: Obama won the last election with a 7.3 spread, 52.9 to 45.6 percent
CNN & Ipsos projected a 7-point spread.
Rasmussen projected a 6-point spread.
7 is closer to 7.3 than 6 is.
CNN & Ipsos projected Obama winning with 53%.
Rasmussen projected 52%.
53 is closer to 52.9 than 52 is.
Do the math. Rasmussen was not the most accurate in 2008.
Numeros said:
"Rasmussen was not the most accurate in 2008."
Congratulations. Never said he was. Here's what I said:
'Now, how did CNN or Ipsos do in 2004? Consistency is the name of this game, you know.'
'In the last two Presidential races, Rasmussen has done the best job of predicting the final result.'
Is that not clear enough for you? Rasmussen has been the most consistent and done the best job over the past two Presidential races. I repeat: that's a fact.