-Caveat Lector- November 6, 2002
The fading accuracy of polling http://www.washtimes.com/op-ed/20021106-14704891.htm Tony Blankley I am writing this column the afternoon before the election, but one thing I feel comfortable predicting: Several famous pollsters will be wrong. For instance, in Minnesota either Zogby (Mondale by six) or Mason-Dixon (Coleman by six) will be left to explain how the dastardly public undercut them at the last moment. Likewise in Colorado, either Zogby (Strickland by five) or Gallup (Allard by two) will grumble about the public not truly understanding the full range of possibilities inherent within the phrase "a 95 percent chance of accuracy within a margin of error of plus or minus 4.5 percent." Properly understood, of course, that phrase means it could be a landslide either way. But pollsters don't get to drive around in Mercedeses and fly first class by emphasizing that their polls are so ambiguously predictive. Americans want to know the future and are willing to pay hard cash for that knowledge. But for how long will we continue to pay for only the illusion of future knowledge? Not that there is anything theoretically wrong with the pollsters' statistical method. The problem arises with the quality of the data input. The changing habits and technologies available to the public are just making it hard for the pollsters to sample a representative group of likely voters. These emerging problems were reported in Tuesday's New York Times in what had to be the feel-good headline of the season: "Cellphones and Caller ID Are Making It Harder for Pollsters to Pick a Winner." The basic problem for pollsters is that cell phones are unlisted, and the increasing use of caller ID machines permit potential polling respondents to screen out unwanted calls - such as polling inquiries. It would probably not be a problem if it were just pollsters who had mastered the art of calling at the exact moment that the weary commuter has finally changed out of his work clothes and poured a drink, or speared the first morsel of his dinner (but not yet brought it gratifyingly to his mouth). But the infestation of telemarketing has driven John and Jane Q. Public to a keen caution, cunning and watchful prudence in detecting and escaping from unwanted telephone calls. This problem has been building ever since cheap answering machines became available about 10 years ago. But with the mass use of cheap cellphones the problem is crossing over into a crisis for practitioners of the polling arts. To compound the problem, a federal regulation prohibits pollsters from calling people on their cellphones without permission. The net result of these developments is that pollsters must spend extra time and extra dollars recalling non-responders. Although the percentage of hang-ups or never-answers is a closely guarded (and deeply embarrassing) trade secret, the New York Times quotes an unnamed pollster as estimating it is up from 10 percent to 30 percent in recent years. I have heard dark whispers that the numbers could be even higher. This factor not only raises the cost of polling, but more seriously, it undermines the methodological integrity of the process. If perhaps up to half of a representative sample self-selects itself out of the sample, the sample may not be representative. If polling continues to devolve from a reliable to an unreliable snapshot of current public opinion, American politics may yet be saved from its current cynical and mechanical state. The proven reliability of polling over the last half-century has tended to drain conviction out of politics. Even naturally honest and principled politicians, when shown the inevitable electoral consequence of their convictions, will tend to find a way to soften or evade such politically suicidal thoughts. As the undecided 20 percent of the electorate has been ever more closely and shrewdly polled by both parties, the messages and positions that both parties target on those soft heads tends to be ever more similar mush. If politicians lose their faith in polling's continued reliability, they will have to fall back on talking to the public, making up their own minds and taking their chances. That will probably result in a somewhat faster turnover of incumbents. Reasonably smart and reasonably honest politicians should still be able to make a decent career of public service. But the cynical robot politicians will be cruelly winnowed out. Praise the Lord. Tony Blankley is editorial page editor of The Washington Times. His syndicated column appears on Wednesdays. E-mail: [EMAIL PROTECTED] All site contents copyright © 2002 News World Communications, Inc. Privacy Policy <A HREF="http://www.ctrl.org/">www.ctrl.org</A> DECLARATION & DISCLAIMER ========== CTRL is a discussion & informational exchange list. Proselytizing propagandic screeds are unwelcomed. Substance—not soap-boxing—please! These are sordid matters and 'conspiracy theory'—with its many half-truths, mis- directions and outright frauds—is used politically by different groups with major and minor effects spread throughout the spectrum of time and thought. That being said, CTRLgives no endorsement to the validity of posts, and always suggests to readers; be wary of what you read. CTRL gives no credence to Holocaust denial and nazi's need not apply. Let us please be civil and as always, Caveat Lector. ======================================================================== Archives Available at: http://peach.ease.lsoft.com/archives/ctrl.html <A HREF="http://peach.ease.lsoft.com/archives/ctrl.html">Archives of [EMAIL PROTECTED]</A> http://archive.jab.org/ctrl@;listserv.aol.com/ <A HREF="http://archive.jab.org/ctrl@;listserv.aol.com/">ctrl</A> ======================================================================== To subscribe to Conspiracy Theory Research List[CTRL] send email: SUBSCRIBE CTRL [to:] [EMAIL PROTECTED] To UNsubscribe to Conspiracy Theory Research List[CTRL] send email: SIGNOFF CTRL [to:] [EMAIL PROTECTED] Om