Network News

X My Profile
View More Activity
About Chris Cillizza  |  On Twitter: The Fix and The Hyper Fix  |  On Facebook  |  On YouTube  |  RSS Feeds RSS Feed

The Pros and Cons of Auto-Dialed Surveys

It's no secret to regular readers of this blog that The Fix's interest in political polling borders on obsession. I love the idea of combing through data to find the messages hidden within the numbers.

Since I spend so much time parsing the polls, I have formed opinions about what makes a poll worthy of attention and what doesn't. Ever since my time at Roll Call newspaper, I have generally viewed telephone surveys that employ live interviewers as the most reliable. I'm much more skeptical about polling that uses automated interviewers -- Survey USA and Rasmussen Reports being the two firms best known for using this approach -- so I largely avoid citing these surveys on The Fix.

But polling -- like everything else in politics -- is constantly evolving, so it makes sense to take a closer look at the arguments for and against auto-dialed polls to see how much weight they should be afforded when handicapping a political race. For those with questions about the Zogby/Wall Street Journal online polling, we'll get to that in a future "Parsing the Polls."

In order to best tackle this immense subject, I am going to break it down into three more manageable pieces -- methodology (or how Survey USA and Rasmussen conduct their polls), advantages and disadvantages.

First, The Fix needs to acknowledge the yeoman's work that Mark Blumenthal (a.k.a. the "Mystery Pollster") has done on this subject. Mark's site is an invaluable resource to anyone with a deep interest in the science and art involved in polling.

Method Makes a Difference

How a polling firm goes about reaching respondents is often the most critical element of a successful survey. Survey USA and Rasmussen make their methodology available to the public -- albeit in varying forms of candor.

The only fundamental difference between standard telephone surveys and those conducted by Rasmussen and Survey USA is that instead of a live interviewer the voice walking the respondent through the questionnaire is automated. Survey USA has partnered with media outlets (typically television stations and/or newspapers) in all 50 states and uses a professional announcer familiar to most locals. Rasmussen employs a "single, digitally recorded, voice," according to its Web site.

Much like a traditional telephone poll, the raw results of the Survey USA and Rasmussen surveys are then weighted by the two firms to ensure the sample is reflective of the demographics of a particular congressional district or state.

The Pluses

The biggest benefit of these "robo" polls is cost. "Survey USA is the first research company to appreciate that opinion research can be made more affordable, more consistent and in some ways more accurate by eliminating the single largest cost of conducting research, and a possible source of bias: the human interviewer," reads a statement on the organization's Web site.

Most polling companies -- partisan and independent -- use phone banks to make the thousands of calls necessary to produce a scientific sample. The process is both time-consuming and extremely expensive. Using a recorded voice makes it easier to survey larger samples, poll more frequently and put surveys into the field closer to elections. That last element of flexibility could well explain why these sorts of surveys have done as well if not better than traditional polling in predicting the final outcome in races. Most live interview, telephone polls end on the Thursday before the election, meaning that they might well miss the late breaks in a campaign.

The second major argument for auto-dialed polls is that by eliminating the middle man (in the person of the human interviewer), the margin for so-called "measurement error" is reduced. Survey USA's Web site lists a variety of ways in which a live interviewer can compromise an interview, ranging from mispronouncing names to tiring toward the end of a call list to simply reading the questionnaire too fast or too slow.

One intriguing but more difficult to quantify element of this measurement error is that people tend to exaggerate their likelihood to vote when interviewed by a live person, while they answer more honestly when prompted by a recorded voice. The psychology behind this trend? Voting remains a private act for many people and they are not comfortable sharing their past voting history with a stranger over the phone. In fact, the anonymity of the recorded voice, the thinking goes, more accurately simulates the conditions of a voting booth.

The Minuses

Establishment pollsters take issue with several elements of auto-dialed polling.

The first deals with the response rates. A traditional live interview telephone poll has a response rate of roughly 30 percent -- meaning that three out of every ten households contacted participate in the survey. The polling establishment has long held that people are less likely to respond to an automated survey than a call from a real person, meaning that auto-dialed poll have even lower response rates and therefore a higher possibility of bias in the sample. Neither Rasmussen nor Survey USA makes their response rates public, although, in fairness, neither do most media outlets or major partisan pollsters.

The second -- and potentially more troublesome -- issue revolves around the randomness of auto-dialed polls. In a standard telephone poll, the interviewer may seek to add another layer of randomness by asking to speak for a specific person in a household, such as whoever most recently celebrated a birthday. Automated polls do not attempt to do that. Establishment pollsters argue that by stripping a level of randomness from the polling process, auto-dial pollsters must more heavily weight their samples to achieve demographic diversity --- rendering the results almost meaningless.

Drawing Conclusions

What to make of these varying viewpoints on the strengths and weaknesses of auto-dialed polls? Smart people will disagree. But we defer to two of the smartest we know to provide perspective on the debate.

Fred Yang, a well regarded Democratic pollster, says that he counsels his clients to avoid drawing broad conclusions from any single survey conducted by Survey USA or Rasmussen and rather to "look at [their polls] as a package in order to gauge the overall trends in the race."

Mystery Pollster Blumenthal echoed that sentiment. "Surveys are by their nature prone to error," he said. "The more you look at, the more confident you can be."

Want to read more about the debate over polling methods? Start with Blumenthal's discourse on the future of survey research in Public Opinion Quarterly.

You can also read University of Oregon professor Joel David Bloom's 2003 paper on the topic.

Then there's an article yours truly wrote for Roll Call that takes a deeper look at Survey USA.

As always, the comments section below is open for you to continue the discussion.

By Chris Cillizza  |  July 12, 2006; 7:50 AM ET
Categories:  Parsing the Polls  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Landing a "Whale" for 2008
Next: Dems' Ad Moves Signal House Battle Is Heating Up


blackjack casino .

Posted by: blackjack casino | September 29, 2006 10:36 AM | Report abuse

casino games .

Posted by: casino games | September 28, 2006 8:35 AM | Report abuse

online casino play .

Posted by: online casino play | September 28, 2006 8:16 AM | Report abuse

generic viagra .

Posted by: generic viagra | September 13, 2006 7:58 AM | Report abuse

Hey, I am no FEMINAZI. Celebrating the right for women to vote, running for office and getting elected to help our nation create better laws for a civilized society in the USA is all good. I celebrate that.

THE FEMINAZI agenda spouts off about gender balance, as if Congress needs over 200 women in those seats before there is parity. That viewpoint is baloney. There are 67 or 68 women in Congress, 14 in the Senate, and many serving as governors. They worked to get elected based on skills, elected experience, and if voters believed they could do the job. THE FEMINAZI agenda is more like a quota system, and mostly based for the success of women who are Democrats not so much for women who are Republicans. The way you find a FEMINAZI is to ask a woman about politics and then ask her opinion of Eleanor Roosevelt. If she can spout off about all the wonderful things about Eleanor and her impact for Democrats, yap yap yap, she's a FEMINAZI. At least that is my experience. AND I AM NOT ONE OF THOSE THINGS.

Geena Davis played a difficult role as the female president on TV. But overall, she did a good job. Too bad the writers kept so much on the kids, unrelated to her job as president. Most of her day was consumed by her kids and her husband, and the TV show suffered from it. Geena is up for an EMMY as dramatic actress, so I hope she wins. Overall, the show was successful.

Posted by: Geena Davis is Wonderful | July 15, 2006 10:01 AM | Report abuse

Geena: You are very artful. The last sentence is what got you, and most of your co-horts. Being a fem is good, in my way of thinking. These are the ones that rear our and their kids, while holding down a full time and productive job.

Posted by: lylepink | July 14, 2006 4:57 AM | Report abuse

Hey guys, there is a huge story about Hillary in the July 13 issue of Wash Post.

Beyond the Poll Numbers, Voter Doubts About Clinton By Lois Romano

Anna Shelley, a mother of three from Utah, says she is ready for a female president, and she is sure that Sen. Hillary Rodham Clinton has what it takes.

But Shelley, a Democrat, is not sure she could ever pull a lever for Clinton. Her reservations are vague but unmistakable: Something about Clinton leaves her cold.

"I want to see her as a human being -- I can read a newspaper and see her agenda," said Shelley, 27,......."I think she's a little hard. She may be strong, but at the same time, if you're driven sometimes you're perceived as not having sympathy. And perception is reality for most of us."

It is a reality that Clinton's advisers are confronting as they seek to position the former first lady for a possible 2008 presidential run. They expect that any campaign would begin after this fall's election, in which Clinton, a Democrat, is running for a second Senate term from New York.

Never has a politician stepped onto a presidential stage before an audience of voters who already have so many strong and personal opinions about her, or amid arguments that revolve around the intangibles of personality and the ways people react to it.

Clinton's assets are formidable: an unrivaled ability to generate publicity and money, and approval ratings that are notably strong, given her polarizing reputation and the controversies she has weathered over 15 years in the national eye. In recent public opinion polls, she handily leads potential Democratic rivals.

Beneath these positives, however, there is evidence of unease -- about her personal history, demeanor and motives -- among the very Democratic and independent voters she would need to win the presidency.

A recent Washington Post-ABC News poll highlighted the paradox. Fifty-four percent of those responding view her favorably, and a significant majority give her high marks for leadership (68 percent), strong family values (65 percent), and being open and friendly (58 percent). At the same time, only 37 percent of Democrats in the poll say they would definitely vote for her for president.

A Gallup poll from last summer also highlighted a perception that she is too divisive, with 53 percent of respondents saying they do not view her as someone who would "unite the country and not divide it."

Follow-up interviews with skeptical Democrats and independents who participated in the Post-ABC News poll suggest that many view her as an inscrutable public figure who gets high marks for her ability and intellect but who nonetheless gives them pause because they find it difficult to relate to her on a personal level.

"The reason I am not able to say I am strongly supportive of her is because -- and this is just vibes -- she does not project a sense of what is inside of her like her husband did," said Sam Hack, 59, a self-described liberal Democrat from St. Louis.

Others said they see a persona too calibrated. "There's no question she's competent and very intelligent, but people want to see authentic human beings, and she has overly managed herself," said Peter Brooks, 68, a professor of English at the University of Virginia and a liberal Democrat who has an unfavorable view of Clinton.

Some Clinton advisers acknowledge these doubts and say they can be diffused if she runs by reintroducing her values and biography to a national electorate. They maintain that negative opinions often reflect misperceptions about her. Paid media and free media would give her opportunities to reacquaint herself with voters (she hasn't appeared on a Sunday talk show in more than a year). And the enormous popularity of her husband, former president Bill Clinton, is also part of the equation, some said.

"Many know of her but don't know her -- so the more they get to know her directly, the more they learn that what she does and what she stands for is what they are looking for," said Mark Penn, a longtime adviser and pollster for both Hillary Clinton and Bill Clinton.

Still, supporters say the powerful scrutiny she faces means that, far more than the typical politician, she has little room for public error or spontaneity, since even casual comments often draw national headlines. In addition, some political analysts believe that politicians who are women must work harder to be perceived as strong and serious.

Finally, those who have worked with her say that, unlike her husband, who easily conveys empathy and familiarity, Clinton is instinctively more reserved and harder to get to know.

The result is a public portrait of Clinton as highly self-contained. In an era when images of politicians biking or jogging are used to give them another dimension, she is rarely seen doing anything personally revealing.

Rhodes Cook, an independent political consultant who studies voting trends, maintains that, for the most part, "voters do have to find some kind of connection with candidates personally. All things being equal, style trumps substance in many ways.

The hope among her advisers is that she can do nationally what she did in New York in 2000. Then, she faced questions about why she was running in a state where she had never previously lived, and whether she was interested in the job solely as a springboard to the presidency. Clinton also faced more piercing questions -- including from many women -- about the state of her marriage in the wake of the Monica S. Lewinsky scandal. Was she staying with her husband simply for politics?

She quieted the doubts through relentless public appearances, old-fashioned retail politics on main streets and at local fairs, and paid advertising.

She eventually won with 55 percent of the vote, including what exit polls showed was 60 percent of female voters. These days, according to the Post-ABC News poll, 59 percent of women nationally have a favorable view of her, and among 18-to-35-year-old women, a notable 73 percent view her favorably. The flip side is that men are significantly more likely to say they would not vote for her.

Brian Tripplett, 47, a Democrat and a United Parcel Service manager from Kentucky, says he has a strongly unfavorable view of Clinton based on impressions 15 years old. "It seems that her public image is different from her private image. It bothered me when I read she was verbally abusive to employees," he said.

Valerie Herzig, 42, an independent from California who leans Democratic, said in the survey that she has an unfavorable view of Clinton, largely because she doesn't have a feel for her. "You hear a lot about her, but you don't hear from her," Herzig said in an interview. "My impression when she was 'Mrs. Clinton' was that she was the driver in the family. . . . But I have no idea what she's been doing for the past five years."

The intense curiosity about Clinton -- as well as the challenge she faces in supplying politically salable answers -- is not new. In her Arkansas and White House years, she at times eschewed the traditional images of the political wife, initially not taking the Clinton name, pursuing her legal career and, on the campaign trail in 1992, offending some with remarks that they interpreted as disparaging to women who stay at home and bake cookies.
Here is some more clips from the article........But some voters wonder what is behind this controlled persona. "I was just talking to my friends about this," said Jeny Guy, 55, a registered independent from Falls Church, who expressed a "favorable" view of Clinton but said she would not vote for her. "I find her too stiff and packaged."

"I guess she would do a good job, but I just don't think she can get the votes," said Julie Troy of Michigan, who describes herself as an independent and a liberal but says she definitely would not vote for Clinton. "I find that men don't like her and that's a problem. . . . I don't think we're ready for her."

THE REASON I BRING THIS TO THE ROOM IS THAT IF HILLARY IS RUNNING IN 2008, the voters of New York need to consider it before they vote for her in 2006. Some people are very happy with Hillary as their senator and are disgusted with her agenda to get back into the White House. Plus, the NY Times presented a front page article about the Clinton marriage, and the 2 for the price of 1 slogan they had in 1992.

Geena Davis was great in Commander in Chief, but she was NO HILLARY. Geena played a warm and caring wife, loving mothers, and hard ass defender. But that is not what makes up Hillary.

The buzz now seems to be who will be the UN-Hillary in 2008. Time for some state polling to show which states would be supporting Hillary for president or Mark Warner or Biden or Evan Bayh or John Kerry or John Edwards or Bill Richardson or even Gore.
And the Washington Post comment about quite a few men who don't like Hillary? She reminds of them of the dominating wife who tries to wear the pants in the family.
Geena Davis was a full partner, kind and courteous, not a Bit**. The feminist agenda is too militant and that is going to drag down Hillary and her army of feminists. That is my opinion.

Posted by: Geena Davis is wonderful | July 13, 2006 1:53 PM | Report abuse

Chris --

In our state, auto-dialed, recorded message surveys are illegal.

Mont. Code Anno., § 45-8-216 (2005)provides:

"45-8-216 Unlawful automated telephone solicitation -- exceptions -- penalties.

(1) A person may not use an automated telephone system, device, or facsimile machine for the selection and dialing of telephone numbers and playing of recorded messages if a message is completed to the dialed number for the purpose of:
(a) offering goods or services for sale;
(b) conveying information on goods or services in soliciting sales or purchases;
(c) soliciting information;
(d) gathering data or statistics; or
(e) promoting a political campaign or any use related to a political campaign.

(2) This section does not prohibit the use of an automated telephone system, device, or facsimile machine described under subsection (1) for purposes of informing purchasers of the receipt, availability for delivery, delay in delivery, or other pertinent information on the status of any purchased goods or services, of responding to an inquiry initiated by any person, or of providing any other pertinent information when there is a preexisting business relationship. This section does not prohibit the use of an automated telephone system or device if the permission of the called party is obtained by a live operator before the recorded message is delivered.

(3) A person violating subsection (1) is subject to a fine of not more than $ 2,500."

Rasmussen Reports that Tester leads Burns. Absent orther information, I would conclude that Rasmussen Reports has violated Montana law.

Posted by: Concerned Montanan | July 13, 2006 1:23 PM | Report abuse

Updates for Preview of Friday Line- Governors

July 6
Angelides (D) 46%
Schwartzenegger (R)* 44%

Dem Primary
Strategic Vision
July 7-9
Cox 42%
Taylor 48%

July 5
Blagojevich (D)* 45%
Topinka (R) 34%

Dem Primary
Survey USA
July 12
Gabrielli 27%
Patrick 36%
Reilly 26%

Statehouse News
June 28-30
Gabrielli (D) 38.6%
Healey (R)30.3%
Mihos (I) 9.2%

Patrick (D) 40.1%
Healey (R) 30.5%
Mihos (I) 9.5%

Reilly (D) 36.3%
Healey (R) 31.3%
Mihos (I) 12.6%

Univ. of Wisconsin
July 2
Doyle (D)* 49%
Green (R) 36%

Posted by: RMill | July 13, 2006 12:51 PM | Report abuse

Also, Gore did not make a splash vs. Dukakis in 1972.

Gore and Dukakis ran for President in 1988. The only thing Gore made a splash in 1972 was as a journalist for the Tennessean, just back from covering the Vietnam War as a military journalist and getting Tipper pregnant with their first child Karenna

Posted by: RMill | July 13, 2006 11:56 AM | Report abuse

There is truly nothing funnier than a proud member of today's Southern-dominated Republican Party calling the GOP the "party of Lincoln."

Posted by: Loudoun Voter | July 13, 2006 11:13 AM | Report abuse

Ok, the topic is polls, right? So if polls are a test of the market of candidates, then it makes sense to see who the people think will be able to do the job as president. But right now, with Rudy Giuliani and John McCain raising millions and Hillary raising millions, I wonder if the political race is for millionaires only?

Case one: Look at John Kerry. He took out a loan for $6 million on his Boston mansion to compete against Howard Dean with $50 million in his pocket. So either being self-financed or able to raise many millions seems to be what the voters want.

Now I sure would like to see a poll from Iowa and New Hampshire. Any out there?

Posted by: Helen | July 13, 2006 12:04 AM | Report abuse

In the book, THE VULCANS, Condi does say that she voted for Carter, but it was his lack of leadership in handling the Iran crisis, with US hostages held for over 400 days that led to her leaving the Democrats.

She is strong on foreign policy like FDR and Truman. So that is why she became a Republican in 1980 and proudly voted for Reagan.

Now if the Democrats want to keep beating up on Joe Lieberman, a fellow on Hardball tonight, Ed Rogers, said Joe should come to the BIG TENT of the Republicans. Hey, if Joe became Republican, I would welcome him as well for making a stand on strong defense of our nation and strong foreign policy as well.

In fact, many Democrats who are strong on foreign policy left the party in 1980. Some of them were in Congress and the Senate, and left the Democrats. So that is a lame argument about Condi being a flip flopper. She is where she belongs and I only wonder what took her so long to come to the party of Abraham Lincoln and the Equal Rights Amendments for African Americans.

Posted by: Slim Girl in Pearls | July 12, 2006 11:39 PM | Report abuse

LBJ's VP was Hubert Humphrey. What total ignorance. Shameful

Posted by: Larry | July 12, 2006 7:04 PM | Report abuse

LBJ chose Herbert Hoover for VP? Wasnt Hoover pres during the great Republican depression in 1929? Condi used to be a democrat anyway another flip flopper. Jeez neocon fantasy land gets stranger all the time. Neocons show their stupidity and lack of basic facts again.

Posted by: Larry | July 12, 2006 7:02 PM | Report abuse

First of all, the BIG DIG is a federal taxdollar boondoggle thanks to Teddy Kennedy, John Kerry and the corrupt/ignorant fellow who Mitt Romney campaigned he would get rid of if elected, (which the unions blocked the Governor from, to be so blunt). Now the people want to dump all of the blame on Romney when he was in charge. Sounds like the Democrats need to do some research about the Big Dig.

Next. Bhoomes requested some insider info, and I will try to help share strategy.

Bhoomes believes Condi means it when she says she is not interested in running for president. I do believe she could win the republican nomination if she ran but it will not be given to her.

Debbie responds that a draft effort is underway, to get support for Condi in states like Virginia where a petition must gather over 100,000 names of registered Republicans to get Condi on the primary ballot. That will cost money to hire people to collect the names of registered people, making sure they are valid and legal.

Also, in New Hampshire, it might be a statewide WRITE-IN campaign to get the Republicans to vote for a CONDI win of all the delegates in the state.

Next, by following the state rules for getting Condi on the primary ballots, it will either require more petitions signed and/or registration fees paid. And of course, that would all be handled from the funds raised by the independent group(s) willing to do this.

Bhoomes says, "our best bet is after the dust settles, talk her in into accepting the VP slot for the good of the party."

Most vice presidents have run for the presidential nomination before they got the VP nod. In 1960, JFK selected Johnson to win the South region, it made a balanced ticket. Johnson choose Herbert Hoover ( a Senator from Minnesota) for his leadership and also Hoover had run for president, showing he would also represent a region of support (the upper Northwest states) again to balance the ticket.

Sadly, Bobby Kennedy was killed in 1968, and Senator Eugene McCarthy did not have enough support to win the nomination, so it went to VP Hoover (Johnson bowed out in March 1968)

Carter and Mondale ran for president, if I remember correctly, in 1976. Carter of the South choose Mondale, again bringing the Democrats of the North region to balance the ticket.

In 1980, Reagan had a good idea about naming Gerald Ford as his VP. (Ford would bring his White House experience to the ticket) but the insider power brokers wanted George H W Bush since he represented more support in the delegate numbers.

Now why Bush chose Dan Quayle out of the Senate might have been to avoid being overshadowed by a more experienced Republican. Yes, No? As I said, MOST VP's get selected for the balance of the region or a major factor which would balance the ticket.

Mondale selected Geraldine Ferraro from the Congress of NY to give him gender balance and some charisma. A Bit of Pizzazz to energize his low poll numbers. But Mondale only won Minnesota and the DC area.

Then we go to Clinton and Gore. Gore had made a big splash against Dukakis in 1972, and with his years in the Senate representing the South, I think Gore brought youth and a WONKISH balance. Fresh ideas against the Old Boy in the White House so to speak.

Bob Dole in 1996 selected Jack Kemp for his boyish flair, his football athletic charm, and his Northern region support. (Also, Kemp ran for president)

Bhoomes also said "That is unless you know something I don't." Debbie say "I think I have laid out a good plan showing that Condi needs to show which states she would win and bring to the 2008 ticket. Based on her being the top choice at the Republican convention in California recently, most newspapers amazingly stated that if she was on the Ticket in 2008, that she might make a BLUE state into a RED state. Again, she is highly favored in her birthstate and early schools state of Alabama. She is a Southern bell in the Republican party, more based on her education and experience and her temperment.

Dear bhoomes, thank you for supporting Condi, and if she becomes VP, she brings talent, strength and diplomatic experience to the White House. But I still think she will win as president. If not, as I said, she still has to run for president to show that she has delegates. But then again, if Bush wants her as VP, he will be able to get her on the ticket at the convention just be saying that she is his legacy. Oh boy, wouldn't the Democrats hate it if Condi was going to the Bush Legacy. Man oh Man, can you hear them screaming, crying, and whining? Yep, that would be a remarkable Republican convention and the media would eat it up. If there is no VP selected before the convention, then you can bet it will be CONDI, just in case she won't run for president.

Posted by: Debbie Watson | July 12, 2006 6:37 PM | Report abuse

thats it Bhoomes, Massachusetts is the most corrupt state in the Union. LMAO, its all thanks to that Republican Governor of ours.

Posted by: Rob Millette | July 12, 2006 5:30 PM | Report abuse


If all the polls showed your candidate was hopelessly behind, you might not bother to vote or contibute.

Or, conversely, if you thought your candidate was a lead-pipe cinch, you still might not.

Either way, it can influence.

Posted by: Drindl | July 12, 2006 4:18 PM | Report abuse

See my above response to Slim Girl in Pearls.

What information a candidate releases to the press or public and what they actually read can be very different things.

Posted by: RMill | July 12, 2006 4:12 PM | Report abuse

Regarding poll bias,

When it comes to predicting political races, what is the purpose of trying to skew the results? All that gets you is less information. Why would a candidate puruse biased polling? (other than of course the push polling discussed above)

Posted by: Zathras | July 12, 2006 4:02 PM | Report abuse

You're right about that, Mike. I only wish more people realized it.

This is off-topic Chris, but I don't know who else to ask--

When will the news media start reporting that the Mid-season [2007] Budget Review released yesterday by the OMB included a proposal to spend $721 BILLION dollars to create private accounts in Social Security and slash guaranteed benefits?

in other words, to spend FAR MORE than it would take to 'fix' social security -- in order to destroy it?

Posted by: Drindl | July 12, 2006 3:56 PM | Report abuse

SurveyUSA and Rasmussen aren't the only autopollers in the business... (here is a local NC company)

Posted by: Justin | July 12, 2006 3:56 PM | Report abuse

SurveyUSA and Rasmussen aren't the only autopollers in the business... (here is a local NC company)

Posted by: J | July 12, 2006 3:55 PM | Report abuse

I was caled, last week in fact, by a polling company with regards to a Referendum on the November ballot: 'Should Oregon ban the use of credit scores by insurance companies for the purpose of establishing rates'. Now, this poll was peppered with questions like "Would you vote for this even though it will result in a 30% increase in your insurance costs?" and "Would you vote for this even though there is a link between people irresponsible with credit and driving?". Polls, and I don't care who conducts them, always show a bias. Some polls, like the one I was subjected to, are nothing more than a 15 minute tirade opposing an initiative and a campaign vehicle for a wealthy and largely unregulated industry and the politcial bums who support it.

Posted by: Mike Brooks | July 12, 2006 3:39 PM | Report abuse

Your stuff is always welcome RMill, no matter how long. Useful info.

Posted by: Drindl | July 12, 2006 3:19 PM | Report abuse

Wow. I didn't realize it was so long. Sorry.

Posted by: RMill | July 12, 2006 3:17 PM | Report abuse

Previewing the Friday Line: Governors

Survey USA June Approval
Riley (R)* 63% (up from 59% in May and up from 52% in Feb)
Governor continues to improve his numbers

This contest has become much less competitive in recent months
Survey USA
June 20
Baxley (D) 40%
Riley (R)* 51%

June 7
Baxley (D) 40%
Riley (R)* 54%

Univ. of S. Ala.
June 18
Baxley (D) 25%
Riley (R)* 53%
Trend: Inconsistent number for Dem Baxley but follows trend for Gov. Riley in low 50's and double digit lead. Outcome: Likely Republican
Survey USA June Approval
Murkowski (R)* 20% (down from 23% in May; down from 26% in Feb)
Continues to slide towards the bottom of the Nations governors, second only to Taft of OH-18%)

Ivan Moore Research
June 7
Dem Primary
Croft 14%
Knowles 82%

Rep Primary
Binkley 27%
Murkowski* 18%
Palin 44%

Head to Head
Knowles (D) 43%
Binkley (R) 37%

Knowles (D) 53%
Murkowksi (R)* 21%

Knowles (D) 43%
Palin (R) 39%

Independent candidate Halcro not included in polling. Trend: none defined other than confirming Murkowski is unpopular even in his own party. Outcome: Toss-up; Knowles is the former Governor but is running within margin of error of unknown Rep candidates other than Murkowski. Influence of independent candidate as yet unknown.

Survey USA June Approval
Napolitano (D)* 58% (same as May; down from 67% in Feb)
Taken a significant hit to approval; may be due to immigration issue.

June Battleground Poll
Napolitano (D)* 48.2%
Goldwater (R) 40.0%

Trend: No other new polls. Continues to enjoy confortable double digit lead over Rep front runner Goldwater. Outcome: Likely Dem


June Battleground Poll
Beebe (D) 49.8%
Hutchinson (R) 43.5%

Trend: Beebe has held a consistent double-digit lead over likely GOP opponent Hutchinson and has hovered just above or below 50%. Outcome: Likely Dem

Survey USA
June Approval
Schwartzenegger (R)* 39% (up from 36% in May; up from 32% in Feb)
Has inched his way up in approval but still in danger zone.

June Battleground Poll
Angelides (D) 44.6%
Schwartzenegger (R)* 44.5%

Survey Polling Institute
July 6
Angelides (D) 37%
Schwartzenegger (R)* 44%

Field Poll
June 2
Angelides (D) 39%
Schwartzenegger (R)* 46%

Trend: This marks a slight improvement from May polling results that showed Angelides tied or slightly ahead. Still within statistical margin of error or tied, incumbent still not approaching 50%. Outcome: Leans Rep

Field has been cleared on both sides. Dem Lindstrom dropped out and Rep Holzman was bounced for lack of signatures.

June Battleground Poll
Ritter (D) 44.2%
Beauprez (R) 36.1%
Trend: A slight improvement over the March Battleground numbers (39.9% - 38.1%) and more in-line with May Rasmussen numbers (43% - 38%). Still within margin of error. Outcome: Leans Dem


Dem Primary
Strategic Vision
June 28
Davis 39%
Smith 33%
Trend: Race tightening from May poll (42%-27%)

Rep Primary
Survey USA
June 12
Crist 52%
Gallagher 21%

July 11
Crist 61%
Gallagher 23%

June 20-26
Crist 49%
Gallagher 21%

Strategic Vision
June 28
Crist 53%
Gallagher 31%

Trend: Crist pulling away; Gallagher generally stalled at under 25% GOP vote.

Head to Head
Strategic Vision
June 28
Davis (D) 41%
Crist (R) 49%

Davis (D) 40%
Gallagher (R) 40%

Smith (D) 40%
Crist (R) 49%

Smith (D) 39%
Gallagher (R) 41%

June 20-26
Davis (D) 39%
Crist (R) 41%

Davis (D) 42%
Gallagher (R) 35%

Smith (D) 35%
Crist (R) 42%

Smith (D) 39%
Gallagher (R) 36%

June Battleground Poll
Davis (D) 39.8%
Crist (R) 42.5%
Essentially the same as March Battleground.

Trend: Mixed results showing Crist leading Davis. Should Smith win the Dem primary and face Crist, the Rep are likelt to hold. In the unlikely scenario that Gallagher wins the Rep primary, then state becomes a toss-up. Outcome: Leans Rep

Survey USA
June Approval
Purdue (R)* 63% (up from 60% in May; up from 60% in Feb)
Trend: Holding steady above 60%

Dem Primary
Strategic Vision
June 26
Cox 42%
Taylor 46%
Trend: Cox had enjoyed front-runner status until recently.

Head to Head
Strategic Vision
June 26
Cox (D) 40%
Purdue (R)* 53%

Taylor (D) 44%
Purdue (R)* 50%

June Battleground Poll
Cox (D) 39.5%
Purdue (R)* 49.6%

Taylor (D) 40%
Purdue (R)* 48.6%
Trend: Troubling for an incumbent Governor to continually battle poor numbers dispite high approval. March and April Rasmussen polls had Cox within margin of error (41%-49% and 42%-48%) and Taylor out by double digits against Purdue. Now Taylor is outperforming Cox against the incumbent and closing to within the margin of error.
Outcome: Leans Rep

June Battleground Poll
Culver (D) 45.7%
Nussle (R) 47%

Trend: After enjoying small leads (all within the margin of error) in Rasmussen polling over prior months, Culver has now been thrown into a dead heat with Nussle. Outcome: Toss-up

Survey USA
June Approval
Blagojevich (D)* 43% (same as May; up from 41% in Feb)
Trend: Incumbent still struggling to get to 50%

June Battleground Poll
Blagojevich (D)* 41.1%
Topinka (R) 37.5%

Trend: A slight tightening from March Battleground poll (42.7% - 36.6%). Outcome: Leans Dem
Survey USA
June Approval
Baldacci (D)* 43% (same as May; up from 41% in Feb)
Trend: Incumbent still struggling to get to 50%

Survey USA
July 11
Baldacci (D)* 41%
Woodcock (R) 43%

June 15
Baldacci (D)* 45%
Woodcock (R) 43%

Trend: Statistical dead heat over past month. Outcome: Toss up.

Survey USA
June Approval
Ehrlich (R)* 44% (same as May; down from 55% in Feb)
Trend: Steady decline in approval is bad news for incumbent.

Field is set for November. Duncan dropped out due to medical reasons.

June Battleground Poll
O'Malley (D) 53.1%
Ehrlich (R)* 39.2%

Washington Post
June 25
O'Malley (D) 51%
Ehrlich (R)* 40%
Trend: O'Malley continues to enjoy double-digit leads over GOP incumbant. Quickly becoming one of the most vulnerable seats for GOP in the nation. Outcome: Likely Dem


Dem Primary
Survey USA
June 19
Gabrielli 23%
Patrick 36%
Reilly 31%

Head to Head
June 27
Gabrielli (D) 42%
Healey (R) 24%
Mihos (I) 15%

Patrick (D) 43%
Healey (R) 23%
Mihos (I) 15%

Reilly (D) 39%
Healey (R) 27%
Mihos (I) 17%

June Battleground Poll
Gabrielli (D) 52.1%
Healey (R) 32.5%

Patrick (D) 55.7%
Healey (R) 33.7%

Reilly (D) 49.4%
Healey (R) 33.7%

Mihos (I) not included

Suffolk University
June 22-26
Gabrielli (D) 42%
Healey (R) 23%
Mihos (I) 7%

Patrick (D) 38%
Healey (R) 25%
Mihos (I) 10%

Reilly (D) 40%
Healey (R) 27%
Mihos (I) 10%

Trend: Despite a confusing and crowded Dem primary field, all challengers are shown to be handily defeating the GOP candidate in all polling by wide margins. Gabrielli has increased polling performance in head to head match ups with Rep Healey, outpacing his primary rivals in some cases. Outcome: Likely Dem (which one?)

Survey USA
June Approval
Granholm (D)* 42% (down from 43% in May; up from 41% in Feb)
Trend: Incumbent struggling to get to 50%

June 7
Granholm (D)* 44%
DeVos (R) 42%

June Battleground Poll
Granholm (D)* 48.1%
DeVos (R) 46.2%

June 21
Granholm (D)* 44%
DeVos (R) 46%

Strategic Vision
June 21
Granholm (D)* 41%
DeVos (R) 48%

Trend: Mixed polls show a general strengthening of the DeVos campaign to oust the Dem incumbent. Michigan is the most vulnerable Dem incumbent seat to defend. Outcome: Toss up.

Survey USA
June Approval
Pawlenty (R)* 51% (same as May; down from 53% in Feb)
Trend: Incumbent hovering dangerously just above 50%
Kelley withdrew from Dem primary when he lost to Hatch at DFL convention. Lourey is continuing to challenge.

Dem Primary

Head to Head
June 25
Hatch (D) 47%
Pawlenty (R)* 42%

June Battleground Poll
Hatch (D) 39.8%
Pawlenty (R)* 45%

Lourey (D) 38.5%
Pawlenty (R)* 45.2%

Trend: Mixed polls but a definite tightening. Previous Rasmussen poll had front runner Hatch up 10 on the GOP incumbent and March Battleground had him within 1 pt. Outcome: Toss up


Survey USA
June 14
Dem Primary
Gibson 43%
McConnell 5%
Titus 34%

Rep Primary
Beers 18%
Gibbons 50%
Hunt 15%

June Battleground Poll
Gibson (D) 33.8%
Gibbons (R) 42.4%

Titus (D) 36.7%
Gibbons (R) 42.4%

Titus (D) 39.3%
Beers (R) 37.7%

Trend: Up and down primary results for Dems makes this unclear. However, Dem Gibson has sunk in performance since the previous Battleground poll (up 42.7% - 39.8% vs. Gibson). Outcome: Leans Rep


Survey USA
June 13
Strickland (D) 53%
Blackwell (R) 37%

June 20
Strickland (D) 50%
Blackwell (R) 37%

June Battleground Poll
Strickland (D) 49.1%
Blackwell (R) 44.3%

Ohio Poll- Univ of Cincy
May 25
Strickland (D) 50%
Blackwell (R) 44%

Trend: Strickland has enjoyed double-digit leads since March and hovered at or above 50% in nearly every poll. Even in the two polls showing a tighter race, Stickland is still around 50%. Outcome: Leans Dem

Survey USA
June Approval
Kulongoski (D)* 32% (down from 35% in May; down from 38% in Feb)
Trend: Worst performing of all Dem governors up for re-election. Winning primary did not help any.

June Battleground Poll
Kulongoski (D)* 47.6%
Saxton (R) 40.1%

Kulongoski (D)* 39.6%
Saxton (R) 36.6%
Westlund (I) 10.1%
Trend: Maintaining a slim lead, Dem incumbent loses ground with Independent Westlund included. Could cause problems considering poor approval. Outcome: Leans Dem

Survey USA
June Approval
Rendell (D)* 58% (down from 62% in May; up from 50% in Feb)
Trend: Steady increase since starting year at 50% with a exception of downturn in past month.

June Battleground Poll
Rendell (D)* 47.7%
Swann (R) 43.4%

June 19
Rendell (D)* 50%
Swann (R) 36%

June 13-19
Rendell (D)* 55%
Swann (R) 31%

Strategic Vision
June 15
Rendell (D)* 49%
Swann (R) 38%
Diamond (I) 2%

Trend: Rendell is enjoying wide leads in most polls. Independent candidate not seen as a major factor. Outcome: Likely Dem

Rhode Island
Survey USA
June Approval
Carceiri (R)* 49% (down from 54% in May; down from 52% in Feb)
Trend: Below 50% for first time. Slight but steady declines should be worrisome for incumbent.

June 5
Fogerty (D) 41%
Carceiri (R)* 40%

Rhode Island College
Fogerty (D) 39%
Carceiri (R)* 44%

Brown Univ
June 26
Fogerty (D) 39%
Carceiri (R)* 44%

Trend: Two recent polls show incumbent edging forward but still within margin of error. Sinking approval could further endanger this seat. Outcome: Leans Rep

Survey USA
June Approval
Perry (R)* 51% (up from 40% in May; up from 47% in Feb)
Trend: First time above 50% this year. Challenge coming from independents wearing off?

Survey USA
June 26
Bell (D) 20%
Perry (R) 35%
Friedman (I) 21%
Strayhorn (I) 19%

June Battleground Poll
Bell (D) 19.7%
Perry (R) 37.7%
Friedman (I) 17.7%
Strayhorn (I) 14.1%

Trend: No changes but remains entertaining. Outcome: Likely Rep

Survey USA
June Approval
Doyle (D)* 45% (down from 47% in May; same as Feb)
Trend: Incumbent struggling to get above 50%.

June Battleground Poll
Doyle (D)* 46.9%
Green (R) 45.1%
Strategic Vision
June 8
Doyle (D)* 45%
Green (R) 46%

June 23
Doyle (D)* 49%
Green (R) 37%

Trend: Slight improvements for incumbent but mixed polls and poor approval still cause for concern for Dems. Outcome: Leans Dem

Posted by: RMill | July 12, 2006 3:16 PM | Report abuse

Since we seem to be ambling off topic anyways-

Previewing Georgia Primary
Tuesday, July 18, 2006

Dem Primary
Cathy Cox (current Sec. Of State)
Mark Taylor (current Lt. Governor)
Bill Bolton
Mac McCarley

Cox and Taylor are now in a real dogfight. Taylor has a slight lead in most recent poll.

Rep Primary
Sonny Purdue (incumbant Gov)
Ray McBerry

Gov. Purdue should not have a problem here.

CD 4
Cynthia McKinney (incumbant)
Hank Johnson Jr. (county commissioner)
John Coyne III

While the seat likely will remain in Dem hands, it will be interesting to see if McKinney's antics cost her with a strong challenger in Commissioner Johnson

Posted by: RMill | July 12, 2006 3:15 PM | Report abuse

'I find it quite strange that a topic of political polling, (and Condi is in most of those polls) is creating so much distress with the Democrats who come in here to express their views.'

You really have an odd take on things, dear. As a Democrat, I feel no distress whatsoever about Condi running. Bring it on!

'But I guess when THE FIX was created, the Wash Post employees were aware that anonymous people can get away with a few nasty remarks without paying any price."

We call it 'free speech' in my country. What do you call it in LaLaLand?

Posted by: Drindl | July 12, 2006 3:07 PM | Report abuse

Tina, The only problem is I believe Condi means it when she says she is not interested in running for president. I do believe she could win the republican nomination if she ran but it will not be given to her, so our best bet is after the dust settles, talk her in into accepting the VP slot for the good of the party. That is unless you know something I don't.

Posted by: bhoomes | July 12, 2006 3:05 PM | Report abuse

The FCC estimates that three to four percent of Americans use cell phones exclusively and have no landline and that this figure might be as much as seven to eight percent.

Other research by telecommunications industry analysts indicates that as much as 10 to 14 percent of Americans ages 18 to 45 use cell phones exclusively or voice over internet protocol (VoIP).

Personally, I used cell phone exclusively for four years prior to getting married and in the two years since have used a VoIP carrier.

Posted by: RMill | July 12, 2006 3:02 PM | Report abuse

This message is for bhoomes

Thank you for your support of Condi. If you want more information, please go to

It has a video of Jessie Jane Duff, the national chair for Americans for Dr. Rice and her appearance on Fox and Friends yesterday. By chance, it is ironic that CNN ran a clip of the DRAFT CONDI movement twice during the Wolf Blitzer spots.

I find it quite strange that a topic of political polling, (and Condi is in most of those polls) is creating so much distress with the Democrats who come in here to express their views. Stranger still is some of the personal remarks which are a bit nasty. But I guess when THE FIX was created, the Wash Post employees were aware that anonymous people can get away with a few nasty remarks without paying any price.

Back to the polls, it is the state by state polls which matter most to test the waters for 2008. So if Sen. McCain was at 50% in the New York area, or at 40% in Iowa, then he would be seen by the media as a major contender. But McCain is tied with Rudy and Condi, and as Glenn Kessler wrote, whenever Condi's name is included in polls, then she is at the top level.

Whether I live in Iowa, or Texas, or Arlington Virginia, I am proudly part of a massive group from across the nation who is promoting Condi for President. If anyone else comes in here, (and there are a few), then we can debate the Republican party. As far as I am concerned, (as a delegate to New York in 2004), hundreds of African-Americans and Latinos were also delegates and also lots of women from across the nation. We all have to pay for our own hotel rooms and airplane tickets, regardless of what the misinformed think.

The polls are a test of the political pie and right now, Condi has a large slice at 20%, right beside Rudy and McCain. That is a fact.

Posted by: Tina | July 12, 2006 2:59 PM | Report abuse

Has there been much data on how polling will need to change as the rates of cell phone use go up, and landlines go down? For example, younger people are probably more likely to use a cell phone as their only phone number, and are therefore removed from the sample. And yes, some of us do vote, and we're only going to get older.

Posted by: Cellphones | July 12, 2006 1:50 PM | Report abuse

yeah, bhoomes, i'm sure blackwell [who's overseeing the election] will swing it for strickland. you bet.

Posted by: Drindl | July 12, 2006 1:50 PM | Report abuse

I was called recently by a live poll. I don't know if I'm a typical responder, but I lied on some of my answers, was truthful on some, and refused to answer others.

Posted by: Ive been polled | July 12, 2006 1:38 PM | Report abuse

No, you would have to move to Massachusetts to live in the most corrupt state in the union. IE: Big Dig that is structually unsound because the mafia got the contracts after bribing state officials to include John Kerry. But one point is well taken, if Strickland wins, we know the process was tainted, so say you.

Posted by: bhoomes | July 12, 2006 1:33 PM | Report abuse

I might add, bhoomes, that every comment you make only reinforces how hopelessly corrupt the republican party has become.

Posted by: Drindl | July 12, 2006 1:22 PM | Report abuse

If Blackwell is the next governor of Ohio, it will because he is the one who is 'supervising' the conduct of said election. Because he refuses to recuse himself, the election will be tainted no matter what the outcome. You can bet that there will be even more funny business than last time.

Ohio is fast emerging as the most corrupt state in the union. It may even surpass Texas soon -- and that's quite a feat.

Posted by: Drindl | July 12, 2006 1:21 PM | Report abuse

Sorry, bhoomes, wrong again. I have voted for my local congressman, Frank Wolf, several times, even though he is very conservative.

You will never get the chance to vote for Condi Rice for president in a general election. Book it.

Posted by: Loudoun Voter | July 12, 2006 1:04 PM | Report abuse

Loundoun sure doesn't sound like he ever voted republican. A lot of conservatives like myself would love to see Condi at the top of out ticket because she is so well qualified to lead this country. Most of us conservatives (I admit, not all) could care less if she is pro choice and a lesbian(liberal rumormongering) to boot. It is why Blackwell will be our next Gov. in Ohio because we are not the bigots you wish us to be.

Posted by: bhoomes | July 12, 2006 12:49 PM | Report abuse

PS Tina: I am in no way either a Democrat or a liberal. I'd let you look at my voting records, but you'd probably start stalking me.

Posted by: Loudoun Voter | July 12, 2006 12:30 PM | Report abuse

Tina: your reading comprehension is very poor.

My post was in no way meant to denigrate Rice's accomplishments. It was meant in every way to denigrate Republican primary voters.

Get it?

Posted by: Loudoun Voter | July 12, 2006 12:28 PM | Report abuse

Wonder if all the Condi Troll Dolls [TM] are the same person. The wording and tonality are a perfect match.

Magnolia: You're right about the abysmal level of awareness in the american electorate. When I did callbanking in 2000, I was astonished to discover that, 2 weeks away from the election, a sizable number of people I spoke to thought George W. bush was his father -- and they were going to vote for HIM because they thought Barbara Bush 'was a good person'.

This is probably the greatest danger to democracy -- an increasingly [and often, willfully] ignorant population that is an easy mark for manipulation and authoritarian propaganda.

Posted by: Drindl | July 12, 2006 12:22 PM | Report abuse


I made no such corrections. I will say that the majority of polls were within the final week (in many cases the final weekend) prior to the election.

In cases where the final poll was further out, these were essentially non-competitive states where further polling was probably considered extravegant and unnecessary, but not in every case.

Exception: Specifically, ARG regularly stopped polling in October for many states, which probably attributed somewhat to their lower overall score.

Posted by: RMill | July 12, 2006 12:13 PM | Report abuse


I have written polls for candidates and for companies. While method makes a difference, that difference is slight when compared to the huge number of no answers. Generally, we use about 500 samples as confirming a general trend, yet as was noted above that number only denotes a snapshot at any time.

So what does the candidate or corporation do when they need information over a long period of time and can't afford (usually candidates) to have polls taken every week or so?

Write a very short survey and pick up the phone book and call every 10th person until you have a data sample of about 30 to 30 people. If the trend is overwhelming over time you can be fairly safe in operating on that. (remember we are talking about a madgin of error of about 20-30 points and some statistitions would say such a small sample is worthless)

If the trend is very close you need to decide on more polling (using one of the methods deliniated in this column) possibly using a focus group, hand out surveys, etc.

Better yet, before all this, read some godo books on the subject and take a course in statistics, Then you can start writing your questions -- which is a chore in itself.

Remember, these days, we are going for a statement and asking how the person feels about the statement. "If Bush were re-elected I would feel comfortable." the rank how they feel on a scale of 1 to
5. Also remember push polling (If you knew that John Smith had a torrid affair with his secretary would you be more likely to vote for him? etc etc etc)

Have fun but remember even 500 person samples (In my estimation because people are not static objects to be measured like flawsd in cars) carry a greater margin of error than a sample of flaws in a car or a tire, for exampe, and some will disagree with that, but mercifully this is a free country and they can do that.)

Have fun.

Posted by: Kurt | July 12, 2006 12:13 PM | Report abuse

A request was made to show a comment or viewpoint by a Democrat or liberal which is trying to downgrade Condi as a 2008 contender, and the above comment by Loudoun Voter is just one example.

Secretary Rice has earned everything she has gained her life, no one handed her a fellowship in a conservative think tank years ago and Brent Scowcroft heard her speak as a young college student. He saw that Condi had the intelligence of a scholar and welcomes her into his foreign policy circle. Every job or political post has been earned, not based on who she was married to.

The people of Iowa have voiced their support, the Quad City Times reported that in a poll, she was at 30% favored by Republicans, Rudy and McCain were in low teens. That is a huge fact, and if people like Loudoun Voter and others want to continue to badmouth Condi, they have the freedom in THE FIX to express it. But I chose to speak FOR a leader, not against a Democrat or a liberal. They are good examples of why their political party is at war within itself, and you can take a look at numerous races for House and Senate primaries. Lieberman is one example of how the Democrats are split with one issue people willing to rip apart the party to show their power to do so.

CNN, FOX, Washington Post and others have been given fair and accurate reports about the Condi for President efforts. That is their job and the readers seem to be responding since Condi is so highly favored as the most popular current public offical as shown by CNN at 57%, while Hillary is at 50%, McCain at 46%.

The Democrats might not like Condi, but if the Republican party wants to stand by her and support her, that is their right. The Democrats need to think of a better reason to make sneaky attacks against Condi than coming into the FIX to do it.

Posted by: Tina | July 12, 2006 12:05 PM | Report abuse

I would love to see Condi Rice run for president. How embarrassing will it be for the Republican Party when its far-right primary voters go into the voting booth and refuse to vote for a single, "mildly-pro-choice, black woman, regardless of her credentials.

They rejected Alan Keyes even though he is a strict conservative. They'll certainly do it to Condi, who is hardly conservative.

Posted by: Loudoun Voter | July 12, 2006 11:41 AM | Report abuse

Sorry, but Chris is right. You DO have to look at the theoretical basis for pluses and minuses. If there was an accurate base against which we could examine our polls, we wouldn't need polls.

I know people use them to predict election results, but that is not what they do. They are snapshots of the day on which they were conducted. So ABA's suggestion is meaningless. Though RMill has an interesting analysis.

RMill, did you make any correction for how recently the poll had been conducted? After all, people do change their minds. Any of these polls could have perfectly predicted how an election would have turned out if it had been conducted on the day of the poll, and we would never know. Or are these all exit polls?

The point raised by Chris that people lie about the chances that they will vote is very important. This is clear from commercial market research too, people answer according to their view of themselves, which may be different from the truth. This means that more people say they are 'certain to vote' than actually do vote.

In the UK the social makeup of the sample is supposed to reflect the turnout at the last election. There are no 'registered voters' surveys. But that approach would not have helped call the American election of 2004, when the turnout jumped dramatically. Indeed in the UK elections of 1997 and 2001 the turnout fell dramatically, which produced its own measuring problems.

Quentin Langley
Editor of

Posted by: Quentin Langley | July 12, 2006 11:38 AM | Report abuse

bhoomes wrote: But if your a democrat, you gotta love the polling because it is the only time they can fantasize about winning.

Yeah, that's what governor Tim Kaine (R) likes to say.

Posted by: Loudoun Voter | July 12, 2006 11:30 AM | Report abuse


Regarding the survey size, while the success rate is lower with automated calls, the capability to make many more calls is present with an automated system. Also, it is possible that the people who do respond to an automated survey are those more likely to vote, making the survey potentially better than with a live poll, where the person is more likely to answer the questions, whether they are likely to vote or not.

Posted by: Zathras | July 12, 2006 11:23 AM | Report abuse

Regarding RMill's point about sample bias, this is a potentially huge problem for automated surveys. Having worked in a similar business, I can say that automated surveys have about one-tenth of the response rates of live ones (people feel guilty about hanging up on a real person, but don't care about a machine).

Therefore, as the size of the universe being polled shrinks, it's harder and harder to construct a representative sample (or to even get a big enough sample to begin with).

Posted by: ABE | July 12, 2006 11:08 AM | Report abuse

You will find at least one SurveyUSA response rate given in footnote 2 of my article, "Comparing Iraq to Vietnam," in Public Opinion Quarterly, Spring 2006, V.70:78-87.
I obtained the information without difficulty from Jay Leve, of SurveyUSA, and in fact he should have been acknowledged at the beginning of the article.

Posted by: Howard Schuman | July 12, 2006 11:06 AM | Report abuse

I imagine that getting people to respond to polls is more complicated now that telemarketers under the guise of surveys/polls are calling people regularly. I wonder how this will affect the market?

Posted by: ljb | July 12, 2006 11:02 AM | Report abuse

Yes, if you ever want to be depressed about the state of society, be an interviewer for a polling group. The level of ignorance is just astounding. I had no idea until I served on one myself.

Posted by: With Magnolia | July 12, 2006 11:01 AM | Report abuse

As much as I am a political junkie, it is just hard for me to get worked up about polls, because for once, I am in total agreement with Drindl, you can get the polls to say anything you want just by the wording. But if your a democrat, you gotta love the polling because it is the only time they can fantasize about winning.

Posted by: bhoomes | July 12, 2006 11:00 AM | Report abuse

Thanks for this discussion. I received an automated poll on politics last week. I vote every election and I have been writing checks left and write to my party (both national and state). My first reaction was to wonder if the poll was genuine or I was going to end up on some commercial phone list. After I was fairly convinced it was genuine, I answered honestly. I used to wonder who were the people who got called. Now I am one of them.

Posted by: JMC | July 12, 2006 10:49 AM | Report abuse

Slim Girl

Unfortunately, there are many more purposes for political polling than actually garnering the viewpoint of voters.

Polls are released to show a candidate doing well when he/she has been faced with bad news (Ney) or when he/she is not considered on par with an opponent (Space).

They can be used to show a candidate is trailing or leading to help energize a base of voters or spur fund raising.

Then there are push polls, designed to advocate or denigrate a candidate or issue.

Theoretically, the sole purpose of news agency polling is to give a snap shot in time about how a representative sample of a population thinks about a candidate or issue. These, however, make up a fraction of all the polls conducted.

Posted by: RMill | July 12, 2006 10:36 AM | Report abuse

No matter what the conversation is about, a Condi Troll Doll [TM] can be trusted to bleat out something irrelevant.

"Some Democrats and liberals have stated" --really? Who are they? Provide examples, please. Anytime I hear the construction, 'some liberals' I know it will be following by something totally fabricated and generally irrational and slanderous.

Posted by: Drindl | July 12, 2006 10:34 AM | Report abuse

I was involved in a random-digit dialing poll for a class this past spring, and I lost alot of faith in polls. I would say only 2 out of 10 people knew who their congressman/women was, and some even considered Hillary Clinton conservative. That shows you what you are dealing with......

Posted by: Magnolia | July 12, 2006 10:32 AM | Report abuse

I want to go back to the concept of polls to show the viewpoints of voters, and that is afterall the sole purpose of any political poll.

The Glenn Kessler article about Condi Rice being highly favored in doing her job as our top diplomat, 52% favorable in the June 7-11 Wash Post/ABC poll. IF she runs for president in 2008, 10% would definitely vote for her, 44% would consider voting for her.

I also remember a recent poll with only Republicans showed over 70% agree she SHOULD run.

That means to me that by testing the market, so to speak, the foundation has been laid for Condi to step out and announce at some time in 2007, even by September 2007. In fact, she might even be able to announce in January 2008, if the petition drive is successful in putting her name on the New Hampshire primary ballot.

Thanks for the ability to discuss the 2008 race without being reduced to a mudslinging mess. Some Democrats and liberals have stated they think other 2008 GOP candidates will toss mud at Condi, but consider this fact, if Condi was really going to be the next VP, why would any candidate muddy her up?

Condi will be seen as a contender for president or VP by every other serious candidate in the Republican party. She will maintain her decorum and temperment no matter how silly the Democrats or liberals behave or try to undermine her.

Thanks again Cilliezza for bringing the news about Condi to THE FIX. It has been some of the best discussion (over 60 comments) than a few of the other posts. Like McCain on this day has less than 20. I guess it shows a lack of interest in him.

Posted by: Slim Girl in Pearls | July 12, 2006 10:24 AM | Report abuse

Biases with theoretical justifications cannot withstand empirical refutations.

Here are some possible problems with live polling:

1. Non-uniformity. Each person asking a poll question is going to ask it a bit differently, even if that difference is in tone. This difference will exist across the different people polling, as well as across the time one person does polling (the 1st call a poller makes will be crisper than the 100th).

2. The "added layer of randomness" given as a justification for live polling sounds very dubious. Do these pollsters have any objective data backing this up? It seems to me that this might introduce more bias than just polling the person called (e.g. a person being polled who has strong political convictions will more likely ensure the call stays with him/her, rather than passing the phone on).

3. Numbers. The live calling just cannot do the volume that an automated system can do. I don't know if the actual poll numbers bear this out, but the capability to do more polling rests with the automated pollers.

4. Filtering. Perhaps one reason that the automated pollers have been doing better is that a person more likely to vote is more likely to respond to an automated poll, while a less interested person will just hang up. With a live call, a person might feel an obligation to state a preference, whether the preference is accurate or not.

There are just off the top of my head. Do these sound valid?

Posted by: Zathras | July 12, 2006 10:23 AM | Report abuse

Also, none of this takes into account the key problem with polling in general, which is weighting scores to get a representative sampling.

Every polling firm is getting hammered for this. Oversample republicans or democrats and it doesn't matter what questions you ask, whether a human asks them or how it is written.

Posted by: RMill | July 12, 2006 10:23 AM | Report abuse

Biases with theoretical justifications cannot withstand empirical refutations.

Here are some possible problems with live polling:

1. Non-uniformity. Each person asking a poll question is going to ask it a bit differently, even if that difference is in tone. This difference will exist across the different people polling, as well as across the time one person does polling (the 1st call a poller makes will be crisper than the 100th).

2. The "added layer of randomness" given as a justification for live polling sounds very dubious. Do these pollsters have any objective data backing this up? It seems to me that this might introduce more bias than just polling the person called (e.g. a person being polled who has strong political convictions will more likely ensure the call stays with him/her, rather than passing the phone on).

3. Numbers. The live calling just cannot do the volume that an automated system can do. I don't know if the actual poll numbers bear this out, but the capability to do more polling rests with the automated pollers.

4. Filtering. Perhaps one reason that the automated pollers have been doing better is that a person more likely to vote is more likely to respond to an automated poll, while a less interested person will just hang up. With a live call, a person might feel an obligation to state a preference, whether the preference is accurate or not.

There are just off the top of my head. Do these sound valid?

Posted by: Zathras | July 12, 2006 10:22 AM | Report abuse

Automated polling has distinct advantages in avoiding the inherent bias of human questioners, especially if the voice is mechanically generated. The human voice always has emotional shadings that the speaker himself may not be aware of.

Having worked for years conducting focus groups and crafting wording for questions, I can state unequivocally:

[1] there is so such thing as 'neutral wording' -- all language has an inherent bias

[2] it is possible to lead people to say anything you want them to say, no matter how they think they feel when you start

[3] anytime there is more than one human being involved, a relationship dynamic is created which distorts the nature of responses.

Posted by: Drindl | July 12, 2006 10:14 AM | Report abuse

Come on Chris. What about results? Aren't they more important than the general pros and cons? Which firms more accurately predicted the outcome? You need to go back to school. Times are changing.

Posted by: ABA | July 12, 2006 9:59 AM | Report abuse

How damn stupid do those pollsters think we are? The very first poll ever taken may have been nearly valid. Since then it has become a game.

Posted by: Clayton Davis | July 12, 2006 9:50 AM | Report abuse


In reviewing state by state polling by various firms in the 2004 Presidential race, I set up a quick scoring system to judge accuracy.

Taking the final poll done by each firm in a state and matching with actual results, I scored 1 pt for each correct call of the state, 3 pts for the exact margin of victory/1 pt if within margin of error (3%) and 3 pts for calling the candidate vote total by exacft percentage / 1 pt. within MoE.

I also tracked misses (states called incorrectly) and quality picks (where the state was called and the margin and candidate percentages were all within the MoE). No points were awarded for misses:


ARG (American Research Group)
44 states polled
131 pts. (avg. 2.997)
5 misses (11.36%)
10 quality picks (22.73%)

10 states
19 pts. (avg. 1.9)
4 misses (40%)
2 quality picks (20%)

Mason Dixon
21 states
92 pts. (avg. 4.381)
1 miss (4.76%)
7 quality picks (33.33%)

5 states
11 pts. (avg. 2.2)
1 miss (20%)
0 quality picks (0%)

30 states
101 pts. (avg. 3.367)
1 miss (3.33%)
13 quality picks (43.33%)

Research 2000
11 states
39 pts. (avg. 3.545)
0 misses (0.0%)
6 quality picks (54.55%)

Strategic Vision
11 states
24 pts. (avg. 2.182)
4 misses (36.36%)
3 quality picks (27.27%)

Survey USA
30 states
116 pts. (avg. 3.867)
1 miss (3.33%)
16 quality picks (53.33%)

20 states
55 pts. (avg. 2.75)
3 misses (15%)
6 quality picks (30%)
1 perfect call- Missouri

The state by state polling data was compiled at:

The point system may be worthless as I just concocted it myself but it allows a comparison since all firms did not poll in all states.

As far as accuracy, Mason Dixon, Rasmussen and Survey USA came away with only 1 miss and had the most net quality picks minus misses

Posted by: RMill | July 12, 2006 9:13 AM | Report abuse


I have to agree with Jonathan. Considering that the key to polling is accuracy, I cannot see a reason not to cite Survey USA.

And in light of the opinions of Mssrs. Blumenthal and Yang, which I have previously stated myself and which I would apply to any poll, I do not see a reason for your citing out-of-date polling data in lieu of results from these companies.

If you are indeed, like I am, a poll junkie, then providing the breadth of relevant information available would allow your readers to look for such trends and formulate their own opinions.

Based on my post yesterday- and I fully admit that the data I used was found online and not compiled by myself personally so I cannot vouch for its completeness- it would seem to bear out that Survey USA and Rasmussen appear to be as accuarate or more so than many traditional polling firms.

Also, in my opinion, removing the possible human bias from the equation makes up for any possible disparity in response rates.

Posted by: RMill | July 12, 2006 9:11 AM | Report abuse

I must admit I'm confused. You begin your entry by reminding us of your longstanding bias against automated polling. You then suggest you'll take a "closer look" at the issue. Your "analysis" then lists plusses, all of which are lifted from the automated pollsters' own marketing, and minuses, 2 unattributed weaknesses devoid of any substantiation. Not exactly insightful.

You finish by pointing us readers to 2 sources of objective data. One is an academic paper from Oregon that, unlike your entry, actually looks at the data for accuracy, and finds: "as much as academic survey researchers may have wished to see SurveyUSA under-perform the field, they clearly did not, and may have actually done better than average."

Your second source is himself a pollster, which seems like a potential bias. Nevertheless, he too seems to find that automated surveys have as good or better track record and deserve a seat at the table.

After reading the very sources you cite, I must admit I'm confused why we're not reading an outright retraction of your bias against automated polling. At the end of the day, all of the assertions about bias or error rates are background noise: the real show is in the results, and everything I've read (in part thanks to you) suggests automated polling is as accurate as live polls. Indeed, I'm convinced knocks against automated polling are primarily initiated by entrenched incumbants fighting to maintain their way of life.

Jonathan Goldstein

Posted by: Jonathan Goldstein | July 12, 2006 8:53 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company