Skip to main content

President Barack Obama campaigns in Boulder, Colorado on November 1, 2012
President Obama, campaigning in Colorado on Nov. 1, 2012.

In the weeks prior to the presidential election, I wrote of the rather formidable gap between how President Obama fared in polls where respondents were classified as "likely voters" and how he performed in polls where all registered voters were included.

At the time, I went back to data from the 2004 and 2008 presidential elections, and I discovered the following:

Of the 50 state presidential polls conducting during the final month of the 2004 and 2008 presidential campaigns, the RV result was closer to the final outcome than the LV result in fully half of them. In just 38 percent of them was the LV screen closer to the final outcome than the RV screen. In six of the polls, incidentally, there was no difference between the RV/LV results in a poll.
While there are still hundreds of thousands (if not millions) of ballots remaining to be counted in the 2012 presidential election, we have enough data to draw some pretty firm conclusions on the merits of screening for likely voters.

Exactly a half dozen pollsters, looking at the Obama versus Romney national numbers, bothered to differentiate their "likely voter" screens from their results among all registered voters. Here is how they broke down:

In other words, given the current national margin (Obama +3.2, which is likely to expand with additional provisional and absentee ballots from Obama-friendly states like California and New York), the "likely voter" screen was further removed from the final result than the registered voters screen in all six cases.

Now, in fairness, virtually every national poll in the final week underestimated the eventual Obama margin of victory. Therefore, if likely voter screens create more Republican-leaning results (which has long been a given in polling analysis), if everyone underestimated Obama, it stands to reason that the RV screens would perform better.

That said, some of those likely voter screens were quite a ways off. Most notably, the crew at Gallup has had quite a bit of criticism heaped upon them for missing the fairway, especially since this is the second cycle in a row that their likely voter screen has erred badly in favor of the GOP.

As Steven Shepard, the resident polling analyst over at National Journal, pointed out in an excellent campaign post-mortem:

On Oct. 26, Gallup released a demographic analysis of those respondents classified as likely voters in its daily tracking poll between Oct. 1 and Oct. 24. Of those voters, 78 percent were classified as non-Hispanic white, significantly more than the percentage of white voters measured by exit pollsters, 72 percent.

Four years ago, Gallup also found an electorate that was 78 percent white, an overestimation from the 74 or 75 percent recorded by exit polls. But this year's disparity is of a greater magnitude.

Gallup has also underrepresented younger respondents in its measures of likely voters. Over the first 24 days of October, 13 percent of Gallup's likely-voter sample was younger than age 30. Exit polls show these younger voters made up 19 percent of the national electorate.

Democratic pollster Mark Mellman believes this kind of likely-voter screen is counterproductive. Speaking at the event at the Gallup building, Mellman said that pollsters and the media have "overfetishized this whole notion of likely voters."

"We should not be concerned about finding likely voters," said Mellman. "We should be concerned about simulating the likely electorate."

Mellman has a salient point. Modeling a potential electorate is always an act of guesswork, but a quick examination of presidential exit poll data should've exposed the folly of the electorate that Gallup was expecting to see fill the polls on November 6th.

In 1992, the electorate was 87 percent white. In 1996, the electorate was 83 percent white. In 2000, the electorate was 81 percent white. In 2004, the electorate was 77 percent white. And, finally, in 2008, the electorate was 74 percent white.

You don't need a Master's degree in statistics to see a certain trajectory in those numbers. And, yet, somehow, Gallup was presuming not merely a stoppage of that trajectory, but a reversal of it.

(In the name of complete disclosure, there are alternate theories on the issue of nonwhite populations in exit polling. Read Harry Enten's excellent post-election article in the Guardian on the matter.)

To say that a marked increase in the white share of the electorate seemed unlikely, especially when between 2008 and 2012, there were four million new registered Latino voters, is perhaps the understatement of the year. And it is why Gallup is taking a heap of criticism, quite a bit of which would appear to be justified.

Originally posted to Daily Kos on Fri Nov 23, 2012 at 03:40 PM PST.

Also republished by Daily Kos Elections.

Your Email has been sent.
You must add at least one tag to this diary before publishing it.

Add keywords that describe this diary. Separate multiple keywords with commas.
Tagging tips - Search For Tags - Browse For Tags


More Tagging tips:

A tag is a way to search for this diary. If someone is searching for "Barack Obama," is this a diary they'd be trying to find?

Use a person's full name, without any title. Senator Obama may become President Obama, and Michelle Obama might run for office.

If your diary covers an election or elected official, use election tags, which are generally the state abbreviation followed by the office. CA-01 is the first district House seat. CA-Sen covers both senate races. NY-GOV covers the New York governor's race.

Tags do not compound: that is, "education reform" is a completely different tag from "education". A tag like "reform" alone is probably not meaningful.

Consider if one or more of these tags fits your diary: Civil Rights, Community, Congress, Culture, Economy, Education, Elections, Energy, Environment, Health Care, International, Labor, Law, Media, Meta, National Security, Science, Transportation, or White House. If your diary is specific to a state, consider adding the state (California, Texas, etc). Keep in mind, though, that there are many wonderful and important diaries that don't fit in any of these tags. Don't worry if yours doesn't.

You can add a private note to this diary when hotlisting it:
Are you sure you want to remove this diary from your hotlist?
Are you sure you want to remove your recommendation? You can only recommend a diary once, so you will not be able to re-recommend it afterwards.
Rescue this diary, and add a note:
Are you sure you want to remove this diary from Rescue?
Choose where to republish this diary. The diary will be added to the queue for that group. Publish it from the queue to make it appear.

You must be a member of a group to use this feature.

Add a quick update to your diary without changing the diary itself:
Are you sure you want to remove this diary?
(The diary will be removed from the site and returned to your drafts for further editing.)
(The diary will be removed.)
Are you sure you want to save these changes to the published diary?

Comment Preferences

  •  Gallup shouldn't have done an LV screen (7+ / 0-)

    Their last RV screen was actually dead-on accurate

    •  Only about the margin (4+ / 0-)

      and that may not hold up, Obama may go above 3.5 points.

      But 49-46 is not as good as the 51-48 Angus Reid had.

      •  Angus-Reid is tops so far (0+ / 0-)

        Followed by the 3% margins predicted by ABC/WaPo, Pew, Rand and Hartford Courant.   But give it a few weeks and Democracy Corps might vault to the top with the 4% margin.  Heck, how about National Journal's 5% margin?   That might just be within reach.  Ironically, the National Journal is a right-wing rag, so them showing Obama edging Romney shortly before election day by 5% should have been a tip-off to the likes of Rove, Barone, Noonan and Morris that their  "god and guts" driven predictions were hollow and unrealistic.    Oh well, more Schadenfreude for us now, eh?  

  •  Obama won. Skew the pollsters and the GOP. (5+ / 0-)

    Once in a while you get shown the light, in the strangest of places if you look at it right.

    by darthstar on Fri Nov 23, 2012 at 03:45:08 PM PST

  •  To quote Rick Perry: (5+ / 0-)


    It seemed so obvious to me--with zero knowledge of statistics and intricacies of  polling--that something awfully fishy/weird/improbable was happening between the RV & LV numbers. It seemed that there just HAD to be a gap between pundit/pollster land and reality.

    And so it was.

    Ho'oponopono. To make things right; restore harmony; heal.

    by earicicle on Fri Nov 23, 2012 at 03:46:13 PM PST

    •  Registered voters is a fact; likely voters ... (0+ / 0-)

      however that number is concocted, will always be a construct. And we'll always need some kind of construction of "likely" voters/electorate because there's no other way to make polling numbers with any grainularity.

      The RV vs. LV debate, IMHO, is apples and oranges. It's like comparing the number of cars to the total amount of miles driven. There's a "big grain" correlation with cars/miles, but it's not sufficient to give us much in the way of predictive powers. BBB's arguments notwithstanding ...

  •  FWIW Wikipedia now showing Romney at 47.49% (0+ / 0-)

    Sorry if this double-posts:

  •  But really, shouldn't minority voters only be (1+ / 0-)
    Recommended by:

    counted as 3/5's of a vote? I mean, that would have altered the landscape considerably, and fit the "original construction" meme of conservative constitutionalists.

    Gallup, Rassmussen and other pollsters need to come to grip with reality. The percentage of the electorate that is white is on the decline, as is the percentage of the elderly who will vote for Republicans.

    As the World War II generation continues to rapidly fade away to be replaced by baby boomers, I see a decline as well in the percentage of the "new elderly" who will support Republicans.

    "A free society that will not help the many who are poor, cannot save the few who are rich." JFK, January, 1961

    by rontun on Fri Nov 23, 2012 at 03:55:11 PM PST

    •  My parents white 'new elderlies', early 70s. (2+ / 0-)
      Recommended by:
      rontun, IM

      And ARDENT Obama supporters in this election and the last.

      Chances that they would EVER vote for an R between now and when they shuffle off this mortal coil? ZERO. Gallup can put that in their RV/LV model and stick it up their...

      Ho'oponopono. To make things right; restore harmony; heal.

      by earicicle on Fri Nov 23, 2012 at 04:55:35 PM PST

      [ Parent ]

  •  I think Obama got elected by of all things, ACTUAL (2+ / 0-)
    Recommended by:
    chrississippi, Gooserock


    And why the hell in the 21st technological century are they still counting fucking ballots?

    "There's a lot to be said for making people laugh. Did you know that that's all some people have? It isn't much, but it's better than nothing in this cockeyed caravan." --Joel McCrea as "Sully," in "Sullivan's Travels."

    by Wildthumb on Fri Nov 23, 2012 at 03:57:33 PM PST

  •  The likely voter screen (6+ / 0-)

    Was always a transparent method to keep the race close thereby fueling the industry that was the media campaign.  

  •  HUH? (1+ / 0-)
    Recommended by:
    "We should not be concerned about finding likely voters," said Mellman. "We should be concerned about simulating the likely electorate."
    Isn't that what a "likely voter" screen is suppose to do?

    Isn't it premised on the fact that not everyone who picks up a pollsters call and is registered will get their ass to the polls on election day?

    Obviously their assumptions about who would get to the polls were more wrong than right. But "simulating the likely electorate" that was the point of applying the screen.

    Supporter: "Senator, you have the vote of every thinking person!" Adlai Stevenson: "That's not enough, madam, we need a majority!"

    by Scott Wooledge on Fri Nov 23, 2012 at 04:03:00 PM PST

    •  I think a touch of snark in his answer. EOM (0+ / 0-)
    •  I think I can translate, at least (0+ / 0-)

      The idea is that it doesn't matter so much whether your LV screen specifically predicts which people will and won't vote; what matters is that it yields a sample that isn't biased -- or at least is less biased than if all registered voters are included.

      That argument could make some sense as a riposte to studies that show that lots of likely voters don't vote and lots of non-likely voters do vote. It doesn't make much sense when the LV results appear to be more biased than the RV results.

      Election protection: there's an app for that! -- and a toll-free hotline: 866-OUR-VOTE
      Better Know Your Voting System with the Verifier!

      by HudsonValleyMark on Fri Nov 23, 2012 at 07:44:29 PM PST

      [ Parent ]

  •  Messina says pollsters simply have to pay the (5+ / 0-)

    money to actually call and poll registered voters. Using the randomn method, however, and modeling the electorate well (based on the census and other important publicly available data), was sufficient to get PPP pretty close to being accurate in many cases.

    Calling the curring registered voter file and going with that is better than applying these screens.

  •  It seems like the firms built their LV models (1+ / 0-)
    Recommended by:

    in the Reagan administration, and they're sticking to them. At least, they were. Only the stupid among them would start with anything other than a clean slate on that front now.

    Ok, so I read the polls.

    by andgarden on Fri Nov 23, 2012 at 04:10:14 PM PST

  •  I Keep Wondering if 2010 Colored Their Analysis. (1+ / 0-)
    Recommended by:

    I know it was a midterm and shouldn't apply, but on the issue of voter enthusiasm maybe they gave it some weight ??

    We are called to speak for the weak, for the voiceless, for victims of our nation and for those it calls enemy.... --ML King "Beyond Vietnam"

    by Gooserock on Fri Nov 23, 2012 at 04:13:12 PM PST

    •  Double-Countng? (0+ / 0-)

      I keep wondering if the effect of using "enthusiasm" as a factor to enhance one party's percentage of the electorate in any given election cycle and then further putting a thumb on the scale by applying a higher LV factor results in a double-weighting of the prospects for that party.  The voting characteristics of each party member/leaner is pretty well known, so once their distribution in the overall electorate is known, we should pretty much know what the voting outcome is likely to be.

      "Love the Truth, defend the Truth, speak the Truth, and hear the Truth" - Jan Hus, d.1415 CE

      by PrahaPartizan on Fri Nov 23, 2012 at 04:49:43 PM PST

      [ Parent ]

    •  I thought about this, too. (0+ / 0-)

      I know that the turnout for young people (under 30) goes down in midterms and up in presidential years. Plus, the party out of power (POOP -- in 2010, the Republicans were POOP) tends to get a boost during midterms. I don't know if the Hispanic or female or African-American turnout goes up or down during midterms.

      If they modeled likely voters on who voted in 2010, it might explain things a bit.

      But the angle said to them, "Do not be Alfred. A sailor has been born to you"

      by Dbug on Fri Nov 23, 2012 at 09:07:40 PM PST

      [ Parent ]

  •  i believe that (0+ / 0-)

    those that overestimate likely and therefore white voters might be trying to adjust for the voter suppression that has been going on for many cycles, just as the thieves of elections like to discredit exit polling because exit polls don't take into account votes lost because of voter manipulation by those that refuse to lose no matter the means, suppression is her to stay as long as there is a republican party and its fascist agenda.

  •  is the title right? (1+ / 0-)
    Recommended by:

    did screening for likely voters lead to INaccurate polling?

    "Politics is like driving. To go backward put it in R. To go forward put it in D."
    Four More Years! How sweet it is!!!

    by TrueBlueMajority on Fri Nov 23, 2012 at 04:46:24 PM PST

  •  Thanks for the analysis, Steve Singiser n/t (1+ / 0-)
    Recommended by:
  •  How they developed their screens is my question (2+ / 0-)

    I was monitoring results for African Americans. Most polling was showing still high numbers for Gallup's screening questions. Blacks were showing the same numbers for items like plan on voting, closely following the campaigns, enthusiasm for voting, etc as in 2008. The only number I saw a drop for was knowing your voting location. For populations that move more frequently (poorer) of have their districts lines changed because of redistricting this is normal.

    Gallup and Mason Dixon created the LV screens DESPITE their data not because of it.  Go a head and google these polls from African Americans. Gallup though Obama would only bring as many blacks to the polls as John Kerry, really?

    The numbers for Latinos were different. Before Obama sign the mini-dream act executive order Latino enthusiasm was low (Markos himself wrote about this) after than their numbers exploded upwards.  But Gallup and Mason Dixon never adjusted their screens upwards DESPITE polling showing Latinos more enthusiastic.

    I could understand a "bad model" but the problem is their model couldn't have produced these results unless someone changed it because they either didn't believe/like the results it was giving. That is the real scandal, who did this and why.  

    -1.63/ -1.49 "Speaking truth to power" (with snark of course)!

    by dopper0189 on Fri Nov 23, 2012 at 05:06:24 PM PST

  •  just patient (0+ / 0-)

    The  l-v scheme by gallup consisted of 7 questions. Whomever lasted through those questions werent l-v but just patient ones. People with nothing better to do then listen to a poller. Would u trust them? No wonder they were wrong.

  •  Just goes to show.................. (0+ / 0-)

    that there is a big difference between constructing a poll seeking the truth as opposed to a poll based on wishful thinking. Gallup's likely voter model was wishful thinking on their part.

    The mass of men lead lives of quiet desperation--HDT

    by cazcee on Fri Nov 23, 2012 at 07:05:52 PM PST

  •  Gallup's RV polls better 3 times in a row & more (0+ / 0-)

    As I noted right after the election in my diary, Gallup has now had THREE consecutive elections in a row where their Registered Voters polls have been more accurate than their Likely Voters polls: Gallup wrong 3 times in a row but with *

    It's obvious that Gallup has some serious internal issues. They, along with their twin Rasmussen were the ONLY two polling firms that consistently had Romney at or near the lead, and both had a final poll that was more than 4% off AND with the wrong winner to boot!

    I would also add that Presidential years are usually much different than off-years (a BIG reason Dems better be wary of 2014). I call them 'Christmas & Easter Voters' who turn up in Presidential years like clockwork, but, are very inconsistent in off-year and local elections. That firms like Gallup and Rasmussen really believed that 2010 was a better model for their Likely Voter screen than 2008 was a fatal mistake.

    Christmas & Easter Registered voter phenomena

  •  It doesn't have to be guesswork. (1+ / 0-)
    Recommended by:

    There are voter files, goddamnit. Which actually tell you who votes in which elections. Which you can match up with Hispanic surname data to actually figure out how many Hispanics voted or whatnot.

     Why the hell does everyone refuse to use actual data and shoves it aside for their voodoo shit?

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site