« BACK  |  PRINT

RS

MEMBER DIARY

Calling BS on a bogus poll

How a Democrat pollster cooks the numbers

“Poll Finds Obama Would Trounce Palin” reads the headline on AOL News. “2012 Election: Sarah Palin Crushed by Barack Obama” screams the head on Associated Content. In a hypothetical contest, the poll ”found” that President Obama would defeat Alaska’s governor 55 percent to 35 percent.

There are a few problems with the poll, however…

1. The polling firm which conducted it, Public Policy Polling, is a Democrat company which serves Democrat clients and therefore depends on their business.

2. PPP is run by Stanley Greenberg, who was Bill Clinton’s pollster. He also advised the presidential campaigns of Al Gore and John Kerry. Greenberg, in partnership with James Carville and Bob Shrum, co-founded the left-leaning Democracy Corps. Does this sound like a man without an axe to grind to you?

3. The methodology Greenberg used to conduct the survey was biased. Take a look at the composition of the poll sample:

If you are a Democrat, press 1. If a
Republican, press 2. If other, press 3.
Democrat ………………………………………………. 44%
Republican……………………………………………… 33%
Other……………………………………………………. 23%

That’s right, Greenberg sampled 11% more Democrats than Republicans, stacking the poll in favor of the Dem candidate.

4. Here’s another question from the Democrat poll for your consideration:

Do you have a favorable or unfavorable opinion
of Sarah Palin? If favorable, press 1. If
unfavorable, press 2. If you’re not sure, press 3.
Favorable……………………………………………….. 39%
Unfavorable ……………………………………………. 50%
Not Sure…………………………………………………. 12%

These favorability numbers for Gov. Palin are totally out of line with those of more reputable and less biased pollsters. Rasmussen, for example, reported at the end of January the following numbers for the former GOP vice presidential candidate:

Favorable…………… 52%
Unfavorable…………. 46%

An Opinion Research Poll conducted in the week following the election put the Palin favorables/unfavorables at 49%/43%.

5. In all of the non-biased polls, there is an upward trend in Gov. Palin’s favorables; that is, they have been increasing ever since the general election. Greenberg’s numbers, in other words, just don’t add up. Unfortunately, the drive-by media is only too happy to accept bogus poll results at face value and amplify them to an even less critical audience.

6. There’s a long way to go to 2012. The president’s favorabilty numbers have been slowly dropping at the rate of about a point a day in the tracking polls. Most voters have been very patient with him, willing to give the guy every chance to get it right. But expectations for Obama have been very high, and so far he’s given no indication that he knows how to turn the country around. As time passes, that patience will wear very thin if he and his team continue to make all the wrong moves. This Greenberg poll wasn’t designed to measure anything. Rather, its purpose was to diminish Gov. Palin’s image as a leading candidate well before the 2012 election and take some of the focus off of the president’s most recent problems.

Back in MBA school at the Walton College of Business Administration, Dr. Phillip Taylor, who taught statistics, said that the most important thing his students should take with them from the course is that those with ulterior motives can cook numbers to make statistics appear to produce any result desired by the researcher. But a skeptic who knows enough about statistics can usually spot the bias in the methodology.

Mr. Greenberg, you’re busted.

- JP

COMMENTS

  • Rod_Patrick

    King is still riding high before the eyes of his faithful slaves.

    Sarah and the rest of us will overthrow the Fake King when the day of reckoning comes forth.

    • http://beaglescout.wordpress.com LJ “Beaglescout” Miller

      At the present rate of a 5% drop per week he has another three months before he achieves the lowest rate of public approval by a national executive since national polls were first taken.

      • janis

        At the rate that he’s insulting one demographic group after another, we might be looking at another 3-6 weeks before he starts polling in the negative numbers. If only we could catch him in an unguarded moment giving his honest opinion of the whackadoodle leftys, it would go a lot quicker.

  • beltaine51

    No, don’t even respond. Allow succor to lull the sophisticates. They wet their pants when they read this. Did anyone notice the enthusiastic cheer of the crowd when the President of the United States proclaimed his abysmal bowling score on Leno? Months ago he spoke of those people that cling to guns and religion, yet he receives adulation in the minions of mediocrity, then insults those that try to achieve. Don’t you think the release of this poll and the West Coast Cleansing trip are linked? Keep asking of A I G. Prepare for Cap and Trade and all the Enviro bablle designed to undermine the USA.

  • reaganiterepublicanresistance

    There is of course a strategy of pre-emption afoot, with the MSM and online media such as HuffPo all going after Palin in a substained effort, as mysterious “leaks” from nobody/false rumors, persist. If Sarah Palin is so silly and irrelevant, why the obsession?

    Clearly the Left noted her appeal, and are out to eliminate the threat- and her treatment last Fall was pure character assasination. Liberals’ double-standard sexism is appalling; she’s “dumb” because she’s a happy and attractive Christian, who enjoys being a mother? Real feminists need to be angry, wear combat boots, no make-up, short hair, and live alone in a studio apartment in the East Village? She’s not the right kind of example because she doesn’t support affirmative action… but instead went out there and just kicked butt?

    Democrats are plenty afraid of Palin coming-back at them in 2012 with a dynamic and complementary VP like Bobby Jindal… or visa versa. And they have plenty of reason to fear her. Palin is the most popular governor in the country… are all those people idiots? Alaska was a pretty corrupt system until she stepped in. Her reforms took on entrenched politicians (inc. Republicans), a mafioso-style union boss, and Big Oil.

    Wouldn’t it have been nice if Obama had been principled and brave enough to confront the corrupt Chicago Democratic Machine? Or shady political operators like Tony Rezko? Racist preachers? Instead of doing business with every last one of them?

    Even though others campaigned for Saxby Chambliss in GA last December, Sarah was brought in as the heavy-hitter to close the deal… as she did on the way to a vital GOP victory. Chambliss said, re. campaigning with Palin: “We packed the houses everywhere we went. And it really did allow us to peak and get our base fired up.” He went-on to call her “dynamite”-

    Chambliss heaped praise on Palin, saying she has a ?great future? in the GOP, adding that ?I can?t overstate the impact she had down here”. Said the Senator- “All these folks did a great job coming in”… but when she (Palin) walks in a room, folks just explode.”

    So, the DNC’s friends will be discrediting Palin and claiming condescendingly that she is a “joke” for the next couple years- precisely because they know the opposite to be true. She has been highly successful in life while ignoring the left-wing feminist model… this helps to explain the extra dose of venom in the attacks.

    Go get em, Sarah- and don’t mind the press, nobody will be listening to them anymore after the pending Obamamania implosion.

    http://reaganiterepublicanresistance.blogspot.com/2008/11/palinjindal-2012.html

  • azaeroprof

    This is clearly a cooked poll in a continuing effort to thwart Obama’s greatest threat for 2012. The attacks on Jindal after the non-SOTU response were another part. What do you think David Axelcrook has been doing in the White House? He certainly hasn’t been helping with economic policy!

    Contrast the crass insensitivity and arrogance of Obama, clearly indicated by his comment on Leno, with the class of Sarah Palin.

  • harlan

    How many of that Democrat 44% are on obama’s “do call” list?

    • Just_Saying

      Even more telling, how many of these Democrats are paid by Obama’s team?
      .

  • http://groups.yahoo.com/group/republican587/ Elizabeth Christian

    n/t

    • JadedByPolitics

      ….

  • investedinterest

  • wsmith2009

    In the lead up to the November elections, poll after poll showed Obama ahead and McCain/Palin to be stalled. At the same time, essay after essay on sites like this, Free Republic, and others knocked those polls down as “biased” or “cooked.”

    You’re right that this poll is years to early to be meaningful, but unless you can convince Nate Silver to write an article denouncing it, I’m going to believe that it reflects current public sentiment within a few points.

    PPP polls were denounced by right-wing bloggers during the election and they turned out to be accurate then; you’ll have to do more than run through the same set of arguments to convince me (or a lot of other people) that they are somehow wrong now.

    • DONTREADONME

      52%-46% so if I called the election 51%-49% with an error of +/-4% I must be the oracle of Delphi. Expect to make enemies with the Palin fans on Redstate real soon, 10 days on and this is what you replied to?

    • Josh Painter

      Costas Panagopoulos of Fordham University studied the accuracy of 23 top polls and ranked them in the order of how closely they called the 2008 election, Here are the results:

      1. Rasmussen (11/1-3)**
      1. Pew (10/29-11/1)**
      2. YouGov/Polimetrix (10/18-11/1)
      3. Harris Interactive (10/20-27)
      4. GWU (Lake/Tarrance) (11/2-3)*
      5. Diageo/Hotline (10/31-11/2)*
      5. ARG (10/25-27)*
      6. CNN (10/30-11/1)
      6. Ipsos/McClatchy (10/30-11/1)
      7. DailyKos.com (D)/Research 2000 (11/1-3)
      8. AP/Yahoo/KN (10/17-27)
      9. Democracy Corps (D) (10/30-11/2)
      10. FOX (11/1-2)
      11. Economist/YouGov (10/25-27)
      12. IBD/TIPP (11/1-3)
      13. NBC/WSJ (11/1-2)
      14. ABC/Post (10/30-11/2)
      15. Marist College (11/3)
      16. CBS (10/31-11/2)
      17. Gallup (10/31-11/2)
      18. Reuters/ C-SPAN/ Zogby (10/31-11/3)
      19. CBS/Times (10/25-29)
      20. Newsweek (10/22-23)

      http://tinyurl.com/58ha9z

      Rasmussen and Pew were the most accurate. PPP didn’t even make the cut.

      In the polling for the Wisconsin Democrat presidential primary, Jonathan Singer wrote, “… as I’ve noted before PPP doesn’t have the best track record thus far in the primaries (in fact it seems to have one of the worst)….”
      http://www.mydd.com/story/2008/2/18/132412/242

      Berwood Yost of of The Floyd Institute’s Center for Opinion Research at Franklin & Marshall College wrote of PPP’s polling inaccuracy in the Pennsylvania Democrat primary, “Three of the four polls conducted by Public Policy Polling (PPP) were biased and all were biased toward Obama.”
      http://www.pollster.com/blogs/predictive_accuracy_of_the_200.php

      Bob Orr wrote, “PPP’s sage, Tom Jensen, active democrat, recently did a series of articles, published by papers around the state giving his analysis of the governor’s race and who was doing what and why. This kid and PPP have a political agenda, and the mainstream media acts like the poll was performed by the twelve disciples. Oh, by the way, the Associated Press doesn’t report PPP’s results because it only reports polls that they consider reliable.”
      http://www.blogorr08.com/2008/01/you-can-fool-all-of-people-some-of-time.html

      - JP

    • ocleverone

      Defending this poll with the criteria of the PPP election poll is irrelevant because the critieria is different, i.e., McCain is not in play on this poll whereas he was at the top of the ticket during the election cycle.

      Having been a ground pounder, telephone voice and an outreach coordinator during the McCain/Palin campaign, I can assure you that more people were enthusiastic about her than him. It was not uncommon to hear comments such as “Can’t we put her at the top of the ticket?”, “I am voting for him because of her.” and other support found for this woman. The rally numbers also speak for themselves. Whenever she was at a rally, the crowds grew. So don’t think she doesn’t have public support.

      I just clicked on this “poll” through the link at AOL and had to chuckle since the response by AOL members discredited the headline in its entirity. My point about reflecting current public sentiment is since when has the media accurately reflected current public sentiment on anything that does fit their own agendas?

      The left can spend all their time screeching about Palin, I hope they knock themselves out doing it. If she chooses to run in 2012 (which I truly, truly, truly hope she does) all the whining and screeching will be old news and they will look like bitter little people rehashing nasty, petty stupidity.

  • Menlo

    What kind of polling firm keeps a blog to post personal opinions and actually responds to this very post. One would have to be a total idiot to take that firm seriously as a credible polling organization.

    What’s more, they are so lacking in confidence in their results that they think to scour the web for criticism of their results and attempt to defend themselves. I guess they figure their readers aren’t smart enough to decide whether to trust their results or the “deluded wingers” here.

    People who post blogs like that on behalf of a polling firm cannot be taken seriously as professionals reporting any kind of “facts.” The whole thing has no credibility. I guess we shouldn’t be surprised the AP reports this as though it did. Of course, I don’t think PPP ever has been a very reputable polling firm. I haven’t seen their track record for REAL elections, but I’m betting it’s quite poor.

    The question and its timing are also a problem, and that’s where the real “delusion” lies. It’s one thing to question favorability. It’s another to make an “election” after we just had one and to include a candidate who has likely never even thought about running. The responses are no more than an approval rating with no possible basis in reality. It’s not even a remotely realistic question, and it is of no practical use to anyone. How about asking about real candidates in real upcoming elections.

    While it is true the percentage of Democrats among voters is significantly higher than that of Republicans, I doubt that it is as high as they make it. More significantly, I don’t believe for a second that there are that few independents compared to Democrats.

  • Brad Smith

    I hate to rain on the parade, but it looks pretty valid to me.

    1. This is a Democratic pollster. Democrats will quit hiring them if they can’t be trusted to produce good numbers. One thing candidates do not want are cooked polls – those are useless. Now with this poll, where there isn’t any real election at stake, there could be a desire to play around, but still, for the most part reputations for accuracy and good polling are vital to all pollsters.

    2. Stan Greenberg is one of the most respected pollsters in Washington. He’s a high quality professional.

    3. Other recent polls, such as this CBS/NYT poll earlier this month, have shown Democrats with a 10 point advantage in voter identification. A 44-33 spread sounds about right to me, based on history and recent election results.

    4 & 5. You don’t like the results. That hardly makes it bogus. Pollsters tend to get different results, especially on hypotheticals such as this one. Greenberg’s numbers shown Palin’s negatives within the margin of error of both the polls you cite to say Greenberg is clearly wrong. The only real difference is that Greenberg finds lower favorables for Gov. Palin. That doesn’t strike me as so radical a difference to support the claims you make.

    6. Doesn’t mean the numbers are wrong. But you do correctly point up the dangers of letting polling drive candidate selection, forgetting that there will be a campaign, eventually. My own sense is that if an Obama/Palin election were held today, Obama would win handily. I’m a big Palin fan, and I think she was treated horribly unfairly in the last election. But the reality is what it is – she’s got a lot of rehab to do, and given her young age, my sense is she’d do best to win re-election, serve out her second term, and wait for 2016 or later.

    I think this poll has Palin’s present situation just about right. What it points up is that a) Palin may not be our most electable candidate in 2012, and that b) to claim that mantle, she has some work to do. This poll is like any other poll – a snapshot of where we are today. Fortunately, there’s lots of time to change the situation, and the Administration is making a hash of things fast. But if you want to get where you’re going, you have to be honest about where you are. This poll looks pretty honest – part of the mosaic, along with other polls that may be more favorable to Palin in particular or Republicans in general, not to mention all the other bits of info out there.

    • Michael Dugas

      Are like advertisements for the winner. There are people out there who’s opinions(read votes) are heavily influenced by polls that allege
      to reflect what the “majority” think. Many people want to be seen as to be ON the winning side and “fixed” polls can steer that sort of person in the direction the poll fixer wants!

    • Josh Painter

      if you look at the list in my prior comment on this thread, you will see that the accuracy of Greenberg’s Democracy Corps was judged to be only mediocre. PPP, meanwhile, was not even listed among the top 20 polling organizations.

      True, there are more registered Democrats than Republicans. But if a polling sample is a true predictor (which any random sample greater than 900 data points should be withing a reasonable confidence level) of the universe that it should represent, then that difference is already accounted for. In other words, there is no need to weight a sample for that which it should already reflect. Those polling organizations who do weigh their samples to “account” for a greater number of Democrats are actually severely distorting the results.

      The PPP poll is an outlier compared to the more accuate and respected polls I mentioned. And as I pointed out in my diary, it goes against the trend that Gov. Palin’s favorables have followed since November of last year.

      Interesting choice of words – “rehab” – for “a big Palin fan” to use. The governor is doing exactly what she needs to be doing – taking care of business in Alaska and getting her ducks in a row. Would mentoring by someone who knows the issues well (such as Fred Thompson) benefit her? Surely. But doing one’s homework is a far cry from being in need of rehab.

      - JP