David McKee

For those of you who are fascinated by the inexact science known as polling, there’s plenty of interesting reading about What Went Wrong in Nevada (and elsewhere). Major news organizations blew off internal tracking polls, to their eventual mortification. They evidently couldn’t bring themselves to believe that, say, that the National Republican Senate Campaign Committee had it right and their preferred buy-a-poll outfits were errant by a country mile. The rinky-dink methods of Mason-Dixon, garbage-in/garbage-out pollster of choice in Southern Nevada are getting particularly unflattering examination. The erstwhile Jedi master of poll methodology, Nate Silver has an interesting mea culpa regarding the undersampling of the Hispanic vote (a hornet’s nest that Sharron Angle insanely swatted time after time after time, even though it represented a quarter of the Nevada population and 17% of the electorate, and broke 9-to-1 the other way), which led to several Tuesday-night surprises, like the considerable underestimation of GOP Sen.-elect Marco Rubio‘s victory margin in Florida.

As for the cell phone vs. land line issue and pollsters’ adherence to the latter, we have reached a nodal point in the history of opinion-sampling. In 1932, George Gallup made his name by polling the presidential race on a door-to-door basis. His rivals simply polled people who had telephones (during the Great Depression, mind you). Thus, Gallup was the only pollster to show Franklin Delano Roosevelt winning the presidency … and the rest is history. I know a number of people whose only phone is their cell and if polling methods don’t change to reflect this irrevocable technological shift, 2012 could be a lot like 1932 — hopefully in a non-economic sense.

Sore losers. Despite a 7,000-vote margin of victory for a casino in Oxford County, Maine, opponents are demanding a recount and muttering dark threats Continue reading →