Recently Huffington Post has been my favorite political website. They generally have a wide range of unbiased political coverage as well as dozens of progressive blogs. But I must take exception to a recent post by the site's founder, Arianna Huffington.
In a post entitled Say No to Pollsters Ms. Huffington takes the position that political pollsters are destroying civic dialogue in this country, as well as creating a culture in which pols pander to those likely to answer political surveys over the phone- all for the "buzz of a bump in the polls." She then drives her argument home by claiming that, since these surveys often have a participation rate of only 25% ("abysmally low") of the eligible voting public, they certainly aren't an honest representation of how things will go come election day. "Wow!" you say. "Only 25%. That's not representative at all."
But, while it sounds "abysmally low," 25% isn't far below the voter turnout rate for many elections. And, because the people who are most responsive to telephone polls are the people most likely to vote, it generally does provide an accurate cross-section of voter sentiment. I have worked for a polling firm (Dan Jones and Associates) since 2002, and our pre-election results have almost never been wrong. Granted this is Utah, so most races aren't that close, but there have been some- the SLC mayoral race of 2003, for example. To ensure accuracy we even (at our expense) run our own surveys alongside those commissioned. The New Hampshire pollsters have taken a lot of flak, maybe deservedly so, but the fact that so many of the maligned Democratic Primary polls were taken from Friday 1/4-Monday 1/7 is meaningful. The response percentage of calls made is almost always higher on the first couple days of a poll than the last couple, so most of those polls were likely conducted in the immediate afterglow of Obama's Iowa win. And, because there's no reliable algorithm to factor in 'momentum,' Barak's Fri/Sat votes are no less meaningful to the final tally than Hillary's Sun/Mon votes- when she must have been bringing in a much higher percentage.
Polls certainly aren't infallible; no one's making that claim. But the reason the media and politicians buy into their results so much is simply because there isn't a more effective way of gauging public opinion. Huffington claims that, by hanging up on pollsters, "we can force our leaders to start thinking for themselves again." But weren't they put in office to listen to what we have to say? And haven't we had enough (at least 8 years worth) of politicians thinking too much for themselves? Give pollsters your opinions (assuming it's not a blatant push poll), get your voice heard, then vote.
Subscribe to:
Post Comments (Atom)
3 comments:
here here
I sense some bias here from you. For not only do you work for a pollster, but you are giving surveys, which means you get paid more by having people answer your questions. Is this a plead to democratic participation or just a self-serving ploy veiled in a populist tone ...
Seriously, though, I am actually kind of split on the service that polls provide during elections. Now, a poll taken of the general public after an election, which asks questions on public policy and the likes I definitely see as a valuable service to the democratic system.
However, pre-election polls are probably more likely to dissuade people from voting if certain elections are already showing up as pre-determined. Why take the time to drive down to the polling station, stand in line, repeat your name and address four times to the nice elderly person manning the register and try to figure out the new computer voting system when your candidate and/or issue has already been predicted by a public poll to lose?
I certainly understand the value of quantitative research for candidates to understand the issues that voters want to hear about and focus on, and my own livelihood depends to a degree on these polls, but I think overall, the polls that simply try to gauge or predict an election's outcome do a disservice to voter turnout and democracy as a whole.
Yeah, I'll admit that I do have some bias in this area and would directly benefit from more people answering my surveys. But I do feel genuinely good about helping people get their opinions to pols in a form that will actually get those ruling to take notice. It's these type of surveys which, amidst all the meaningless market research we do, give me confidence in the importance of my job.
I do agree with most of what you write regarding pre-election polls, but we both know that they won't go away anytime soon. The most we can hope for is that, instead of driving people away from balloting places, they somehow galvanize people to vote by reminding them that democracy must be participatory, and the only way to truly participate is to vote.
Post a Comment