Friday, Sept. 29, 2006 | 7:32 a.m.
All pollsters are not equal. But some are more equal than others.
In the Orwellian world of politics, where taxers and spenders boast of being fiscal conservatives and everyone hates illegal immigrants except when speaking to Hispanic groups, it's hardly surprising that measuring public opinion has become less science and more salesmanship.
During Campaign '06, as Roll Call's Lou Jacobson recently wrote, "it's hard to ignore the developing consensus among political professionals, especially outside the Beltway, that nontraditional polls have gone mainstream this year like never before. In recent months, newspapers and local broadcast outlets have been running poll results by these firms like crazy, typically without defining what makes their methodology different - something that sticks in the craw of traditionalists. And in some cases, these new-generation polls have begun to influence how campaigns are waged."
Nowhere has this been more so than Nevada, where the media and campaigns routinely hype results from so-called interactive polls. There is still skepticism about these polls - where automated questions are given over the phone or e-mailed questions are sent over cyberspace - in the same way there is still skepticism about bloggers. But just as some bloggers have journalistic credentials and are not misanthropes pecking away on laptops in their underwear, some of these interactive polls have, on occasion, hit the mark.
The influence has been felt especially in the governor's race. Zogby International, a respected firm, has used interactive polling for governor's contests across the country in partnership with The Wall Street Journal, which has given the surveys more credibility. That rankles a lot of traditional pollsters, but the Journal plays up the Zogby polls, where e-mails are collected and then used over and over again for surveys - with weighting supposedly done to ensure accuracy.
Even though there have been anecdotal reports of folks registering for these polls in multiple states with multiple ethnicities, Zogby believes it can filter out those shenanigans.
Jacobson noted the effect here when he wrote: "Shortly after Nevada's Aug. 15 primary, Zogby came out with a poll showing Democratic gubernatorial nominee Dina Titus with a slight lead over her Republican rival, Rep. Jim Gibbons. This contradicted every prior poll, including Zogby's own ... Subsequent polling, however, has aligned with the previous results."
Until this week, that is. Despite conventional wisdom and mainstream polling data that suggests Titus is behind by nearly double digits, Zogby's latest survey showed her within the margin of error. That is, a dead heat.
You can be sure Titus already has e-mailed those results to prospective donors and a news release surely is imminent. Funny how campaigns will promote poll results from outsiders that they like and deride the ones they don't. They are fickle that way.
Mainstream media outlets often use these interactive pollsters - Zogby, Rasmussen, Survey USA - without explaining the radical methodologies. For instance, KVBC-TV Channel 3 uses Survey USA and frequently touts the results on its newscasts without informing viewers how the poll was done or that the methodology is controversial.
The biggest argument against these new-fangled pollsters is that anyone could be pushing buttons on a phone - my 11-year-old daughter, for instance - or signing up for Internet-based polls (campaign operatives, for instance).
Unlike traditional, experienced pollsters, who use scientific methods to cull voter lists and weight results, these interactive opinion-takers don't have to do as much - and therefore, don't have to charge as much.
But there also is no question that these polls, especially since they are done so frequently, can pick up on trends, even if the data is not always as accurate as mainstream surveys. And it's not as if they don't try to balance for demographics and try to get new names into their surveys.
Even the best pollsters miss the mark once in a while, although the really good ones - Glen Bolger, Peter Hart, Doug Schoen - are much less likely to do so, especially if they have been polling in a certain universe such as Nevada for a while.
So during the campaign's homestretch, when you read about a poll - either online or in a newspaper - or see one on TV, remember that all pollsters are not created equal.
archive
