An Internet hub for moderates, centrists, and independents, with domestic and international news, analysis, original reporting, and popular features from the left, center, and right

Is It Time to Ban Polling?

All of the major networks (for example CNN) have a policy which forbids releasing their projections as to who will win any given state in an election until the polls are closed. The exit polling is supposed to be kept sacrosanct until everyone has had a chance to get out there and vote. Why? The explanation is both simple and valid. If the media announces that a race is effectively over, they depress the turnout and can amplify a win or loss beyond the levels which might otherwise had been registered. If you are planning on voting for candidate A, but the radio announcer declares that candidate B has already been declared the winner while you are driving home from work, why bother going to the polls for a lost cause? In a closer race, this could suppress an eleventh hour surge for the underdog and even change the outcome of the election.

Now for the unlikely but obvious question posed in the title. What is the difference between these projections and the public release of polling numbers and projections? The current presidential race would not seem to apply, as the polls are close and seem to shift by the day. But what if one of the candidates opens up a significant (let’s say 15% or better) lead nationally and jumps to a similar lead in the majority of the swing states by late October? If all of the Joe and Jane Sixpacks across the country are sitting down in front of the television on Halloween evening and the news anchors are saying that either McCain or Obama has effectively won the election already, where is the motivation to get out and vote?

What benefit to voters get from having these poll numbers fed to them every day? I can understand how the campaigns would pay to get poll numbers… it’s a useful tool that can help them craft their message and decide where to expend resources for the best chance of victory. And, obviously, polls are great for journalists, bloggers, radio hosts and television talking heads since they give us something else to chatter about while we wait for the next debate or monstrous gaffe. But are the voters served by having this information? Or might it actually toss an otherwise potentially close election out the window?

Is it time for the media, both new and old, to declare a moratorium on releasing poll numbers?

  • JSpencer
    Yes.
  • DLS
    No, it is not time to ban polling, just to ban (a) their abuse; and (b) their abuse (including the over-posting of polling results by day or by the hour on this Web site). Also, (c) their abuse.

    The Pew polls and reports, for example, have never merited saying "no" to, ever.
  • DLS
    The moratorium that needs to be sought is reporting _exit_ poll data on elections (added to which will be skewing of or invention of pro-Democrat additional results), that is, on the election days. (Uniform nation-wide poll hours would also help. We don't need to hear elections being predicted or conceded before people in western time zones have not yet completed voting.)
  • How do you fairly have synchronized voting hours for the people in NYC and Oahu? Somebody is going to be getting out of bed in the middle of the night to vote in that case.
  • jchem
    You raise an interesting question Jazz. And DLS I would agree that some polls can be abused, but what happens when an honest poll shows the lead to one candidate to be 15% or more?

    I don't see it happening anytime soon. But if it did, you would most certainly see more stories on the news about O.J. Simpson or a missing pregnant wife who's husband is a suspect in her disappearance. After all, the media needs something to talk about...
  • Rudi
    Don't blame the polls for most ills. Voter apathy and ignorance shows up in voter turnout numbers. If Iraqis can wave purple fingers while risking violence, Americans can miss American Idol and vote for something more important than Americas Next Top Model.
  • GeorgeSorwell
    The problem isn't polls, the problem is the way they're reported.

    There have been fairly significant differences in the polls all year long. Yet each polls gets breathlessly reported like it was the latest Hollywood blockbuster.

    When the newest poll gets reported, there should be context about other recent polls that show a few points difference (or similarity, if that were to happen). Some explaination of the differences ought to be included. It wouldn't take that much additional time.
  • Some interesting perspectives on problems with polls, but they don't seem to address the original question. Would heavily unbalanced poll results wind up depressing voter turnout and possibly even skewing the election results? If the answer is no, then why bother having the networks wait until the polls close to predict results? Their projections are no more official than any poll or guess until the results have been certified.
  • Rudi
    Jazz, their is a big difference between polls and exit polls. Polls are dynamic, yet trends and averaging negate the noise. There are no trends or averaging with exit polls on ELECTION day. The difference is months versus on day.
  • StockBoySF
    I think polls before election day are fine to report... and since this seems to be a close race I think such polls will only encourage more people to turn out for their candidate.

    In the 2000 race we saw FL being reported for Gore (which would have given him the WH), but we all know how that turned out. Also Dewey was declared the winner over Truman in 1948. That's still a common joke about declaring the winner before the actual polls close. We've also seen some local races determined by a handful of votes which reinforce the notion that every vote DOES count. If people thought their vote didn't count then they may not be inclined to vote.

    So I think most news networks have learned their lesson about reporting winners and if the networks haven't learned their lesson about early predictions I think most Americans know that the winner isn't the winner until most of the votes have been counted, regardless of any winner the news agencies might announce.

    I think if everyone believes in advance that a certain candidate will win that may skew the election, but who knows? If everyone thought candidate A is the clear winner then few people may actually vote for candidate A (why do something if it's a foregone conclusion?), but candidate B may win because his/her supporters were energized and turned out.
  • Dave_Schuler
    I thought it was time eight years ago.

    Unfortunately, it can't be done without a constitutional amendment. However, I strongly suspect that a ban on professional political polling would be constitutional although hard to enforce.
blog comments powered by Disqus
© 2005-2009 The Moderate Voice | Site design by Elegant Themes | Site customization, hosting, and security by Enxit Group, LLC