The story goes that, on Election Day, 1948, Thomas Dewey told his wife to go buy the best nightgown she could find. She would, after all, soon be sleeping with the president of the United States.
The next day, when Harry Truman had defied all conventional wisdom, most polls, and the headline writers at The Chicago Tribune, to win the election, Mrs. Dewey supposedly said to her Mitt Romney-scale gobsmacked husband, “Well, should I go to the White House, or is Harry coming over here?”
Then there’s the 1936 poll run by a magazine called The Literary Digest. The poll, accurate in the past, said Kansas Gov. Alf Landon would oust President Franklin D. Roosevelt.
This biggest miss in the history of polling — FDR won 46 of the 48 states and 61 percent of the popular vote — was based on an unbelievably unrepresentative sample. Mostly, it was the magazine’s own readers — an upscale, if not altogether idle rich, crowd — plus lists of people who owned automobiles or telephones. In Depression America, that was not an accurate sample of the voting public.
But it was a lot like the sample that has the most to do with picking Utah’s elected officials.
Our unique caucus and convention system proudly eschews the will and interests of the majority of Utahns — even the majority of its majority Republicans — for a system that puts all power into the hands of a few self-selected people who vet the candidates and winnow them down to one or two, who are then handed to voters in what becomes the most undemocratic elections this side of Cuba.
The defense of this system, which has had to be furiously cobbled together in the face of the Count My Vote effort that would toss it out in favor of real primary elections, is that the decisions are better because they are made by a few, highly involved people. Not by the silly rabble whose empty heads will be turned by slick TV ads.
That argument sort of sounds like the one for representative democracy in general. Rather than have all the people vote on every issue and line item, we delegate power to small groups — Congress, the Legislature, the city council.
These happy few will theoretically make wise decisions because, unlike the rest of us, they can take the time to really get down into the weeds of the candidates and/or issues.
The fatal flaw in that thinking, of course, is the same one that destroyed The Literary Digest poll. And the Literary Digest.
A self-selected sample is, by definition, no good. No modern pollster would use one, even one made up of highly intelligent or energetic people, to discern or predict the people’s views.
In Utah, with the system open only to people who can and do show up on caucus night, the hurdles amount to a total denial of the democratic process.
It would be as if the membership of the Legislature were not chosen even by the flawed electoral process we have, but by opening the doors of the House and Senate chambers and seating the first 75 people to go through one door and the first 29 people to go through the other.
Such a sample would be made up of folks who care. Who would study and work hard.
Or who would ignore the wishes of the people to, say, shortchange public education, hold the whims of gun owners as holy writ or put the interests of payday lenders and fossil fuel extractors ahead of the needs of real people.
Or put on their best nightgowns and head on over to Harry’s house.
George Pyle, a Tribune editorial writer, predicted that Alf Landon’s daughter could never get elected to the U.S. Senate. She served three terms.