Yesterday I noticed a lengthy opinion piece in the Washington Post by Patrick Caddell and Douglas Schoen entitled "If Democrats Ignore Health Care Polls, Midterms Will Be Costly." I thought the piece was a day late and a dollar short, in the sense that its main thread of argument (people will vote against health care and against Democrats in November) was already old hat - political prognosticators have been saying that for months.
One aspect, however, stood out enough that I meant to write a post about it: The authors' claim that "a solid majority of Americans opposes the massive health-reform plan." That, it seemed to me, was simply untrue. Here were two pollsters, one whose curriculum vitae suggests a pro-Democratic tilt, and one who's worked for more Republicans than Democrats, both ignoring the results of countless polls taken over the past year showing that by a wide margin, Americans support most of the reforms in the Democratic-generated plan. This needed to be rebutted.
Luckily, someone beat me to it. Today, Joel Benenson, the chief pollster for the Obama camp,took up that very point in "Most Americans Want Health Care Reform". Benenson also added some evidence I would not have known, i.e. that in follow-up questions in the same poll Caddell and Schoen were referring to, a large percentage of respondents made clear that they disliked the current plan only because it did not go far enough.
It's no surprise that polls, like statistics generally, of which they are a subset, can tell any story you want them to, depending on emphasis and omission. It's a bit more of a surprise to see serious pollsters out in public trying to use their data to sway opinion, rather than inform it. To ignore the many earlier polls indicating support for health care reform, and to fail to describe the caveats introduced by the second-tier questions (after all, such questions are there for a reason) suggests a willful effort on Caddell & Schoen's part to affect how people think. (Of course Benenson too was politically motivated in responding to them, but how could he not have?)
I think the truth is that neither of them really knows with any degree of certainty how this issue will color people's voting in November. Each has ahold of a different part of the elephant. That's why personally, I'd prefer to see the actual poll questions and answers than read the ramblings of pollsters masquerading as shapers of opinion.
Comments