The validity of straw polls

Recently, I pointed out recent polling of South Carolina voters in the 2008 Presidential race, and I also commented about it on Sunny Phillip's blog. This data and a recent straw poll held by the Aiken County GOP couldn't be much more different in their findings, where the four top candidates were:

ARG Polling (12/21-23/06):

John McCain 35%
Rudy Giuliani 28%
Newt Gingrich 15%
Mitt Romney 5%

Aiken GOP straw poll (1/10/07):

Mitt Romney 35%
Tom Tancredo 16%
Newt Gingrich 13%
John McCain 10%

In looking at these two polls, we see a large disparity. This isn't the first time - Rick Quinn ran a spirited campaign for Treasurer, worked the party circuit, and as a result, he fared well in GOP straw polls. But party support didn't translate into voter support, as he didn't even make it into the runoff.

While polling is never an exact science, unlike straw polls, it is supposed to be based upon a sample taken which has demographic similarities to the population being sampled. Party regulars are not always representative of those who vote in primaries, and certainly not representative of general election voters.

Not only that, but party regulars have access to information and campaign representatives that most voters don't have. They get far more information via the party grapevines and networks than most voters will see from the usual television, radio, and direct mail marketing campaigns. In addition, while many voters are undecided until the closing weeks of a campaign, most party activists have their minds made up weeks and even months before the votes are cast.

Those who do polling know this sort of stuff and these reasons alone suggest straw polls may be fun, but they don't serve any real purpose of being able to measure voters sentiments or predict voter behaviors in upcoming elections. Nevertheless, campaigns (and sometimes even the media) embrace these numbers as if they're really important events.

So, today's question for discussion is this:
Aside from helping fill empty print space and air time on slow news days, just what are straw polls really good for?

... for fun, here's a link to the American Research Group's Margin of Error calculator. Play around with it and see just how many or few people it takes to complete a survey.

7 Response to "The validity of straw polls"

  1. west_rhino 18/1/07 10:36
    Short answer, nothing.

    Distilled equivocal answer, subject to rebuttal, denial, obfucation and perversion; It does give the nattering nabobs of negativity a nosh for nodding off with.

    I'm incensed (that censer is against the fire code) that the brillaint bombastic blithering Buchanan is a no show on thses polls...

  2. dawber71 18/1/07 16:28
    Earl, could you please decipher what West is trying to say please. I know you and he probably think alike but you know I have the mind of a blonde so sometimes I need a little extra help,LOL. I'll be in MP next Wednesday and Thursday. Hope to see you.
  3. Brian McCarty 19/1/07 02:08
    Straw polls are like watching two mules fight over a turnip. While it might be interesting to watch while it is going on, it really does not mean anything to anyone.
  4. west_rhino 19/1/07 09:12
    dawber, straw polls give the press and folks with too much time on their hands another pile of stuff to play with... is it manure or is it "Play-Do" depends on wheat they do with it, but to use a line of Shakespeare that most of us had to suffer through, straw polls are "full of sound and fury signifying nothing."
  5. Anonymous 19/1/07 10:57
    dawber, stay away. don't buy into his siren song, because behind it all is a liar, a jerk and a creep.

    but if you're the dizzy blonde you say you are, then i can see what he sees in you.

    but i wonder what you will soon see in him, when you get a chance to look behind his false intellectual family guy bullshit.

    if you want to know what you'll find, just look what all the others found who booted him to the curb. two marriages and all his other girlfriends.

    find a few and ask them what THEY think of him, if they haven't run so far away from him that they can't be found.
  6. Darrell ( 20/1/07 13:24
    My gut reaction is that they're good for the same thing that election day polls are good for: manipulation. The media seems to have seen itself as the fourth arm of the US government ever since Watergate. They think that they're part of the checks and balances that keep the nation running. They're happy to try to shape the news ... even manufacture it ... if they think that they can make a change in what they perceive as the current direction of the nation. Granted, I'm a bit of a paranoid right-winger, so take what I say with multiple grains of salt, but that's just my take on it.
  7. Moye 20/1/07 20:21
    I enjoy looking at them but since I have participated in so many and have been over so many that the ones I know of represent the party's hard core regulars and people in the know and not the guy on the street. I believe a straw poll can be used for fundraising on a local party that are conducting them but do not take much else to be considered. We will have one in two months at the latest and I can just about tell you now what they will read here in Clarendon. A year it could change.

Post a Comment

Thanks for taking the time to share your thoughts!

To post a comment without having a Blogger account, select "Name/URL", put your name in, but leave the URL line blank. Email me if you'd like to comment, but need help making it work.