How Could the New Hampshire Polls Get It So Wrong?



Over the years, polls have gotten very accurate in predicting who will win elections. But obviously, pollsters were wrong - very wrong - in predicting result of the New Hampshire primaries.

Nine individual polls (even Hillary’s own poll) showed that Obama had a significant lead - some even had him with double-digits leads. So how did the pollsters got it so wrong?

There will be a serious, critical look at the final pre-election polls in the Democratic presidential primary in New Hampshire; that is essential. It is simply unprecedented for so many polls to have been so wrong. We need to know why.

But we need to know it through careful, empirically based analysis. There will be a lot of claims about what happened - about respondents who reputedly lied, about alleged difficulties polling in biracial contests. That may be so. It also may be a smokescreen - a convenient foil for pollsters who’d rather fault their respondents than own up to other possibilities - such as their own failings in sampling and likely voter modeling.

Gary Langer, the director of polling at ABC News, explains: Link [ABC World News Webcast, Flash video] | The Numbers, Gary’s blog - Thanks Zach!


Previous Post
Get Neatorama by RSS or email
Next Post
Posted on January 10, 2008 at 5:12 am by Alex
Category: ABC World News Webcast, Politics



12 Comments to "How Could the New Hampshire Polls Get It So Wrong?"

  • Carrington Vanston
    January 10th, 2008 at 8:31 am

    The headline should read “Get” or “Have Gotten” instead of “Got.” Similar mistake in “So how did the pollsters got it so wrong?”

  • Sid Morrison
    January 10th, 2008 at 8:55 am

    Maybe the pollsters neglected to properly factor in the large number of independents who are (oddly) free to vote for either party in NH primaries. I know if I lived in the state, I would register Independent so I would be always be free to vote for the opposing party’s weaker candidate — especially if my own party’s race was not close.

  • Joel
    January 10th, 2008 at 9:19 am

    perhaps Obama backers were so confident in the polls that they felt little need to actually go vote?

  • Rich
    January 10th, 2008 at 9:46 am

    Maybe we need more polling to predict the accuracy of the poll… It’s a shame we can’t be satisfied with the actual results rather than allowing the media to create news by boosting up the soothsayer aspect of their biz.

    As far as I’m concerned the ‘weather team’ has the monopoly on that area.

    Here’s an idea… let’s John Edward and Sylvia Browne to contact the spirit world and tell us where each primary will go…

  • Melissa
    January 10th, 2008 at 12:15 pm

    I think the exit polls showed Obama in a strong lead too.

  • Jacques
    January 10th, 2008 at 12:41 pm

    One word: diebold

  • orionriver
    January 10th, 2008 at 12:44 pm

    Maybe it was a rigged election. Maybe somebody actually holds the controls to the voting machines. Maybe the primary is a better time to fake an election, since fewer people will look at it. The polls are never wrong, (cliche phone polling example aside) Obama won.

  • Tom
    January 10th, 2008 at 4:45 pm

    It is well known that Diebold’s machines can be easily hacked in about ten minutes, so anyone who had access to them for that amount of time (perhaps in the factory) could have decided the results months ago.

    The totals from the hand-counted counties were VERY different from those that used machines, in many cases off by 7%, which is more than enough to give second-tier candidate the win.

    There needs to be a real discussion in this country about this, but most people just dismiss the very concept of a rigged election as tinfoil hat conspiracy talk. All the while our freedoms are taken from us one by one by one.

  • xander
    January 10th, 2008 at 5:51 pm

    Polls and projections seem to have the effect of inspiring underdog voters to vote and/or, in the supporters of the sure-fires, blunting the sense of urgency to get to the polls.

    Why polls anyway?

    Primaries should be conducted nationwide in the same few hours.

  • ted
    January 10th, 2008 at 6:13 pm

    I never did understand the point of your primaries, anyways. Why not just vote for them all on one night?

    I listened to Obama’s speech after the vote - wow, he’s a great speaker, with good writers.

    Hillary’s speech was lame, uninspiring.

    I think people are so afraid of a fiery black man that they’ll vote for the tired-out, wimpy white woman.

  • It's so obvious
    January 10th, 2008 at 8:03 pm

    Instead of looking at the pollsters, look at the ballot paper trail. The vast majority of impartial polling today is unimpeachable. If the polls are “wrong,” you should smell a rat.

  • Dan
    January 10th, 2008 at 10:44 pm

    There are serious discrepancies between the hand count and machine count districts, and there is a single company who is the chain of custody for all the machine counts. The company is owned by an avowed Republican, and a principle in the company has a felony conviction. All of this is circumstantial, but very, very, suspicious. Go to http://www.bradblog.com for the latest on potential vote fraud.


Want your own avatar? Get one for free at Gravatar!



Neatorama Comment Policy
You don't have to register or login to comment, but it's easier if you do so. We don't censor comment based on your point of view but comments that are abusive, use excessive profanity, or contain off-topic links may get edited or deleted. On some posts, it may take up several minutes for you comment to show up.


Stay updated on the comments in this post with Comment RSS