Call it the silly season. With the presidential debates behind us, the candidates now spend their days speechifying in their swing-state bubbles, leaving the media little to report upon except, yes, the polls. As a result, the campaign is now as much sport as news. No pundit show is complete these days until it reports the latest “scores” in ESPN fashion, be the tally “Bush 51, Kerry 46” or the precise opposite. There’s just one little problem: This time around, the pre-election polls may prove meaningless, because this presidential election is the first ever to be contested during the Cell Phone Era.
There are an estimated 170 million cell phones in use in America today, up from just a few thousand 15 years ago. Cell phones have transformed our daily life as much if not more than the Internet and, in the process, thrown political-opinion measurers for a complete loop.
That’s because pollsters are forbidden by federal law from dialing cell-phone numbers. Given that fact, trying to decipher how America will vote while only calling land-lines is a little like calling your neighborhood astrologer to pick the winner.
Writing in the Los Angeles Times earlier this week, Philip J. Trounstine, a San Jose State media professor, estimated that only 4 to 7 percent of the U.S. population is CPO (“cell phone only”) and that, as a result, cell phones would have little impact upon polling this year. We respectfully disagree. For one thing, CPOs skew heavily to the under-30 demographic, which group, if it swings one way or the other in a close race, could determine the outcome. For another, large numbers of Americans of all ages and all political persuasions are virtually unreachable except by cell, preferring to make their home phones de facto answering machines.
Anecdotal estimates of the size of this group range as high as 25 percent of the total electorate. And while it may not be politically homogeneous, its members are decidedly different — certainly more mobile, for starters — from the group pollsters are managing to reach out and touch in their homes.
In 1936, the Literary Digest was the behemoth of presidential polling, having used a mail ballot to predict the winners of the 1924, 1928, and 1932 elections. The Digest built its sample base from auto-registration and telephone-directory lists and picked Republican challenger Alf Landon to defeat FDR in the latter’s bid for reelection that year. Meanwhile, an upstart named George Gallup did his polling door-to-door, asserting that the magazine’s lists, skewed as they were toward the wealthier segment of American society, were biased toward Republicans. Gallup picked FDR to win. The rest is history.
We’re not predicting a winner or loser on November 2nd. We’re simply saying that our now-traditional methods of measuring public opinion are as antiquated today as the Literary Digest‘s were in 1936. Caveat emptor.
