Over at fivethirtyeight.com, Nate Silver has kicked up a bit of controversy over polling done by Strategic Visions. He makes a compelling case here that there is something very suspect about their results. He makes an even more compelling case here that at least one of their polls has results that are patently absurd. In the first instance, he makes his point mathematically, in the second by I guess what you would call induction, or is it deduction?
Meanwhile one of his co-bloggers has piled on a different polling company, Prince & Associates, a company that polls rich people.
The patently absurd poll struck me as fairly obviously made up. It purports to show that Oklahoma high school students are abysmally ignorant about basic civics questions (e.g., what is the basic law of the land; what are the names of the two major political parties). While I’m quite willing to believe many Oklahomans are dumb (after all, both their senators are very dumb), I, like Nate, and unwilling to believe that they are that dumb (only 23% know George Washington was the first president?) and that (according to the poll) none of them are well informed enough to answer eight out of 10 basic questions.
The larger point is that these polls are often taken for legal tender, no questions asked, by those that purvey the results to the public. In the case of Prince & Associates, the Wall Street Journal uses their results. Strategic Visions spreads its product all over the country, like manure on a farmer’s field. It’s hard to fault the press for assuming that these people are not just making things up, but these articles show that in order to do its job, the press must not just pass on the numbers, but must have some sort of peer review process to make sure that bogus numbers don’t enter the discourse. At the very least, any pollster that refuses to be transparent regarding its polling methodologies and raw data should be ignored.
These folks (and, of course, they appear to have a rightward slant) are in one sense merely misrepresenting reality. In another sense, they are shaping it. If, for instance, George Will likes what Strategic Visions has to say, he will repeat it, and it will become conventional wisdom, against which no mere fact can stand.
Post a Comment