Here's how MSNBC's David Shuster just described a new Politico poll:
Here's the latest Politico poll, and it shows that public support is slipping for the public option. On the idea of whether it would make it worse, 42 percent now say a public option would make health care worse, 33 percent say it would make it better, 25 percent say it has no effect.
That description is wrong.
First, the Politico poll did not measure a shift in public opinion -- this was the first time the poll asked the question Shuster cited. You can't look at a single data point and declare a trend, as Shuster did when he asserted that the poll showed "support is slipping."
Second, the question Shuster referred to is an awfully blunt tool for assessing public support for a public option. The question asked whether "adding a government-managed health care coverage option would result in better, the same, or worse quality health care in the U.S.?" It is not at all difficult to imagine respondents who think the quality of health care available in the US would remain the same with a public plan, but who support such a plan, either because they think such care would be available to more people, or for other reasons.
It simply isn't responsible to look at a poll question assessing expectations about a specific outcome of a public plan and use it to assert overall support for such a plan. But that's what Shuster did.
A better way to assess overall support for a public plan is to refer to poll questions that ask whether people support a public plan, or think it would have an overall positive effect, or think it is necessary. Like this one, from McClatchy:
One of the points being debated is whether or not the government should create a public health insurance plan as an alternative to private insurance plans. Which of the following is closest to your opinion? It is necessary to create a public health insurance plan to make sure that all Americans have access to quality health care. Access to quality health care for all Americans can be achieved without having to create a public health insurance plan."
That poll found that a majority of Americans think a public plan is "necessary."
Or this one, from CBS News:
Would you favor or oppose the government offering everyone a government administered health insurance plan -- something like the Medicare coverage that people 65 and older get -- that would compete with private health insurance plans?
That one found 64 percent support for a public plan, and only 29 percent opposition.
Or this one, from Quinnipiac:
Do you support or oppose giving people the option of being covered by a government health insurance plan that would compete with private plans?
That one found 69 percent support for a public option, and only 26 percent opposition.
All three of those polls were conducted within the last month. All three of them actually assess the level of support for a public plan.
The Politico poll Shuster used does not. The data may be useful in other ways. It may well indicate opportunities for opponents of reform, and challenges for advocates. But it simply does not assess, as Shuster claimed, public support for a public plan. And it certainly does not say anything about the change in such support, given that it is merely a single data point.