Somebody please explain why pres. polls are 2-10 points different

Somebody please explain why polls are 2-10 points different

Yesterday NBC Nightly News reported a new (WSJ/NBC) poll having McCain behind by 10% with Obama at 52%/McCain at 42%. Now a new AP Poll puts Obama at 44% to McCain 43% on the very next day. What is going on? Is any poll reliable? Is this race over? What is causing the swing?

Post to Twitter Post to Facebook Post to LinkedIn Post to Reddit

Posted by at 05:00 | Posted in Measure 37 | 14 Comments |Email This Post Email This Post |Print This Post Print This Post
  • Bob

    Nate at 538 (he’s a democrat, but his statistical analysis is down the middle) posted about this yesterday:

    If you look at Registered Voters, the polls are remarkably consistent (Obama +9-11), with AP/GFK and Pew being the high and low outliers.

    The problem this cycle is estimating likely voters. We know there have been huge shifts in voter registration and party ID, but it’s a guessing game figuring out what that will mean on election day.

    Again excluding the hi/lo outliers, the likely voter margins get a bit bigger (Obama +5-9), but still a reasonable +/-2.

    In other words, it’s probably a 7 point race right now. Depending on turnout, it might be a 5 or a 9 point race. If most of the pollsters are completely wrong in weighting for age, race, and party ID, there’s an outside chance it’s either dead even, or a 400+ EV landslide.

    • Provo

      How could national pollsters be wrong on age, race and party ID? I can’t belive they have not eliminated all parts for errors.

  • Rupert in Springfield

    In general, when you see poll results like this its a form of voter suppression. This last time we saw this was with Clinton vs. Bush 1. At that time we saw the same wild poll results that differed by orders of magnitude, as we have here. Interestingly only one poll was at all accurate in that election and it was the Zogby poll. Zogby had Clinton winning by single digits, the rest of the press had Clinton in the teens. Did voter suppression work in that case? Its hard to say, possibly. However polling suppression is a late game tactic and there were other late game tactics going on as well in this race, the actions of Lawrence Walsh being the most notable example.

    At any rate, that’s pretty much what is going on here. The idea is that is you can put your candidate far enough ahead in the polls, it will dispirit your opponent and thus help the candidate the polls are endorsing. This is done for two reasons. One it can tilt a very close race and possibly help with victory. Two in a very close races it can help a candidate that might otherwise win by a very slim margin increase the point spread and thus claim some sort of mandate in the election. This is why if we look closely at the polls we will often find something like they were conducted with a sample of 900 people, 500 Democrats, 400 Republicans, conducted on a weekend ( a notoriously poor time to poll republicans for some reason ). One suspects that further investigation into the polls would reveal the Democrats just happened to all live on the upper west side in Manhattan, and the Republicans were all a sampling from John Chaffee’s old district in Rhode Island. You can especially suspect this sort of thing when you see nutty result like

    Who is better able to handle the economy:


    Obama: 499 McCain: 1


    Obama: 95 McCain:305

    When you see results like that, you know the Republicans they polled probably live in Rhode Island, probably find paying taxes patriotic and feel a deep sense of shame that in this day and age the NRA still exists.

    Frankly the polls have been wildly wrong in the last several elections, always biased in favour of the medias preferred candidate, invariably a Democrat. Given this, it is hard not to reach the conclusion that wild polling is probably the most common form of voter suppression in use today.

  • Alan

    What makes you think they polled the wrong people. There is ample reason for many Republicans to have confidence in Obama over economics.

    • John in Oregon

      Actually Allen this point is a real surprise. Despite McCain’s history constantly poking his finger in the eyes of Republicans and Conservatives (long list of issues) the polling indicates fairly unified Republican support. Possibly due to Palin.

      On the Democrat side the effect of the Pumas are a real unknown. There is a real push amongst centrist Democrats over the primary fraud issue and from what I can see the fracturing effect is totally unpredictable.

      I don’t expect it but I would not be surprised if a prominent Democrat raised the issue very publicly in the days before the election.

  • John in Oregon

    This particular question is quite pertinent. Following the 2004 election an independent in depth study was done of polling and the election by Bruce W. Hardy and Kathleen Hall Jamieson. One key finding was this.

    “During presidential elections, poll results frequently are presented in the news. Reporters use these polls to tell the public what it thinks about the presidential candidates. We argue that polling results tell the public what it *should* think about the presidential candidates as well.”

    Reporting on that study William Tate observed “In their 2005 article, Hardy and Jamieson detail how the Kerry campaign was able to change the public perception of President Bush by getting the Los Angeles Times to include a question in one of its, supposedly, independent polls. That query reframed Bush’s steady leadership, a plus, as stubborness, a negative.”

    “The Kerry campaign managed to get widespread coverage of that question’s results from its friends in the media. According to the study, “By imposing the negative cultural meaning of stubborn on positive traits such as ‘strong leader,’ the Kerry camp was successful at creating not only a new character trait to assess George W. Bush but also a reassessment of Bush’s trump positive trait.””

    The legacy media also points out another factor this year, the Bradley effect. The media tells us the Bradley effect is that racists vote against the minority candidate. That’s false. The Bradley effect is that after having made a selection unrelated to race, the voter who picks the non-minority candidate lies to the pollster about his choice to be politically correct. The Bradley effect has noting to do with voting due to race based criteria.

    There is also a reverse Bradley voting effect to the benefit of the minority candidate. The voter votes for the minority candidate to be politically correct even tho he prefers or leans to the other candidate.

    Another problem relates to demographics. Nationally voters are split roughly equally into three groups, Democrat, Republican and independent. Often when checking the internals of the poll (if they are provided) one finds the sample was 41% Democrat, 33% Independent, 21% Republican and 5% declined. CBS and NBC are famous for this kind of poll which is highly biased.

    Polls cam be manipulated, intentionally or unintentionally, in a number of other ways, including question wording, the order in which questions are asked, and how respondents are chosen.

    Yet another problem is how the media reports the poll. A recent excellent example is the poll that is reported as showing Obama up by 10. However looking at the exact same poll the likely voters results are quite different, 45 / 43 Obama. Clearly in this case the media has its own bias.

    The organization Public Agenda provides a list of questions journalists-should-ask-about-polls. The first 10 are;

    1. Who did the poll?
    2. Who paid for the poll and why was it done?
    3. How many people were interviewed for the survey?
    4. How were those people chosen?
    5. What area, or what group were these people chosen from?
    6. Are the results based on the answers of all the people interviewed?
    7. Who should have been interviewed and was not?
    8. When was the poll done?
    9. How were the interviews conducted?
    10. Is this a dial-in poll, a mail-in poll, or a subscriber coupon poll?

    To this list I would add the most important question. Who is reporting the results of the poll and what is the bias of the report?

  • Crawdude

    I appears that it is how you asked the question, when you ask, to whom you ask and where you ask. Over the last few elections, polls have been shown to be very inaccurate.

    In Pennsylvania for example: the National polls show Obama 11% up on McCain, but an internal state DNC memo released yesterday put it at 2%. Of course this is why McCain jumped on Penn recently.

    Its going to interesting on election night for a couple hours, McCain will need to everything to fall the right way to win. Obama can actually lose a couple Blue states and still win, the road is a little easier for him if he can keep Biden’s mouth shut, lol!

    2 weeks is a life time!

  • Reper

    Does it matter. All the polls show McCain behind. Shouldn’t the campaign be over?

  • dean

    The 2004 pre-election polling was very close to the actual results. 20 major polls, by the same pollsters doing this election, had ranges (just before the election) for Bush from a low of 46% (Fox ironically enough) to a high of 51.2% (Battleground). Kerry ranged from a low of 44% (Newsweek) to a high of 50.7 (Battleground). Just about all the polls had the race within 2 points one way or the other. Bush ended up winning by 50.7% to 48.3, the lowest margin for an incumbant president being re-elected since Woodrow Wilson. A uniter he was not.

    Read for yourself at:

    To claim that all the pollsters are somehow in the tank for Obama, trying to supress turnout for McCain is sheer paranoia. These outfits are professionals who sell their services in the marketplace. Their only incentive is to be as accurate as possible so they can earn a good living. Predicting turnout is the biggest difficulty, and 04 was a huge turnout election. The pro Kerry vote was higher than the pro Gore vote. He got beat because Rove’s organization got an even higher turnout for Bush.

    In this election, based on everything I read, Obama’s get out the vote organization is as good or better than Rove’s was, and McCain’s is nearly non-existant, which is why in the end the margin is likely to be larger than the average of the polls in favor of Obama. McCain’s last hope is he manages a sweep of the most contested battleground states PLUS picking off a currently stron Dem state. At this point he seems to be putting his eggs into Pennsylvania for some reason unknown to anyone but him. It looks like yet another Hail Mary.

    • dean

      Adding a point, early voting, which favored Republicans by huge margins in 04, appears to have turned heavily to the Democrats.

      Not good news for McCain, especially since there is a lot of early voting in states he needs badly, like North Carolina and Virginia.

    • cc

      “In this election, based on everything I read, Obama’s get out the vote organization is as good or better than Rove’s was, and McCain’s is nearly non-existant…”

      That would be non-existent; you know, like incumbent or suppress – but don’t worry, your deanbonics has become the coin of the realm these days. After all, if you *feel* you’ve spelled it right, that’s all that really matters – plus, since you pay about as much attention to the accuracy of your “facts” as you do to your spelling no one takes anything you say seriously anyway.

      Shouldn’t that be “get the voters registered organization”? You know, drive them to the registration office, give them cigs, fill out forms for them (especially for Mickey, he’s short a couple fingers), and then do it all again – say 70+ times per “voter”? What you’re ignoring is the FACT that registering “voters” when you’re willing to practically move their lips and manipulate their hands for them (see: puppets) is NOT the same as getting them to actually vote. You and your ACORN buddies have had months within which to register these poor, disenfranchised souls (sniff!), but on election day (with exceptions like Oregon, with its fraud-friendly vote-by-mail) it ain’t gonna happen. True to form, the bums, college kids, and distracted, busy young adults, will forget or blow it off – and no one will be there to pull their strings.

      “20 major polls, by the same pollsters doing this election…”

      You mean like this one?

      It just takes a while for the truth to sink in for many people and it takes a while to cut through the BS barrage from the MSM and see what’s really going on. And I’ll tell you this – NOBODY”S going to be telling a pollster they’re voting for McCain in order to “fit in” or to look or sound “cool”, do ya think, dean?

      As Ann Coulter so brilliantly put it:

      “Political correctness has taught people to lie to pollsters rather than be forced to explain why they’re not voting for the African-American.

      This is how two typical voters might answer a pollster’s question: ‘Whom do you support for president?’

      Average Obama voter: ‘Obama.’ (Name of average Obama voter: ‘Mickey Mouse.’)

      Average McCain voter: ‘I’m voting for McCain, but I swear it’s just about the issues. It’s not because Obama’s black. If Barack Obama were a little more moderate — hey, I’d vote for Colin Powell. But my convictions force me to vote for the candidate who just happens to be white. Say, do you know where I can get Patti LaBelle tickets?’ ”

      Check out those undecideds – they’ll break 75/25 for McCain.

      By the way, Where’s Waldo…

      …I mean Biden?

      Is he being kept from the media because he’s likely to say anything that comes into his aneurysm-riddled brain? It seems to me you were all lathered up about Palin not being available to the media a few weeks ago – I’m sure you share my heartfelt concern about Uncle Joe’s absence at this critical jucture.

      What an idiot – exhibit A demonstrating Obama’s lack of judgement.

      Imagine HIM as president


      • dean

        He hoo laffs last laffs best.

  • John in Oregon

    I post this under the heading SPECULATION. I post it here as an insight into the polling process. HT AJStrata.

    I was having dinner a night ago with a friend of mine who is a statistician for a well-regarded private polling company. They do some work for Republicans in California, but most of the work they do is for Democrats or Democrat-leaning operations (Unions, etc.). Anyway, her shop was retained to do a few Presidential polls for targetted states on behalf of a union so the union could decide where to spend their ad dollars for the last week. They did Ohio, Florida, North Carolina, New Hampshire, Pennsylvania, Colorado, New Mexico, Nevada and Missouri. After mocking the hell out of the voter id spreads used by Rassmussen, Zogby, etc. (and this is coming from a committed Dem who will be voting for Barry O) she said the results of their polling lead her to believe that McCain will definitely win FL, OH, NC, MO and NV. She says Obama definitely wins New Mexico. She said that Colorado and New Hampshire were absolute dead heats. She said she thinks there is a 55% chance Obama holds on in Pennsylvania and a 75% chance McCain wins Virginia. She absolutely laughed at the public polls showing Obama leading Virginia–and pointed out that all of those polls rely on Dem turnout being +4 and as much as +7, when in 2006, Republicans actually had the advantage by +3. She also pointed out that the numbers for Obama in SWVA look absolutely awful and that McCain is running 10 points better then Allen did in NoVa.

    Anyway, her companies conclusion is that the election will come down to Colorado, New Hampshire and the Republican leaning district in Maine, which in her opinion might very well decide the Presidency (apparently the district in Nebraska that Obama thought he might be able to get is now off the table). She said she has very little doubt that the public polling is part of a “concerted voter suppression effort” by the MSM. She said IBD/TIPP was the only outfit doing public polling that was “worth a bucket of warm” urine.

    • dean

      Doesn’t pass the smell test John. Even Fox has Obama up 9. The reason the Dem turnout projection is higher is because of all the Dem registrations, plus the turnout machine Obama has assembled, plus he has way more money to spend during the last lap than McCain.

      New Hampshire a “dead heat?” The poll average is currently 9.3. More like dead meat. McCain threw NH away with the Palin pick. Colorado is closer at 6.5. Nevada also close at 3.3. North Carolina IS a dead heat, as are Missouri and North Dakota of all places.

      It ain’t over till the votes are counted, and stranger things have happened, but a rational person would conclude McCain has long odds getting longer by the day due to early voting.

      I’m still open for bets if you are feeling optimistic.

Stay Tuned...

Stay up to date with the latest political news and commentary from Oregon Catalyst through daily email updates:

Prefer another subscription option? Subscribe to our RSS Feed, become a fan on Facebook, or follow us on Twitter.

Twitter Facebook

No Thanks (close this box)