UPDATED: NBC responds to criticism of change in poll wording

Health Care For America Now (HCAN) argues that a change in the NBC/Wall Street Journal's wording of a key poll question about health care reform produces skewed results.  

In June, the NBC/WSJ poll asked:

In any health care proposal, how important do you feel it is to give people a choice of both a public plan administered by the federal government and a private plan for their health insurance--extremely important, quite important, not that important, or not at all important?

76 percent said it was extremely or quite important to include such a plan in health care reform.

In July, NBC and the Journal changed their wording:

Would you favor or oppose creating a public health care plan administered by the federal government that would compete directly with private health insurance companies?

That produced a much more negative response.  There are indications that the new NBC/WSJ out this evening will repeat that July wording, with similar results.

Here's HCAN's take, backed up by quotes from two pollsters:

These polls are not comparable. The first poll (June) accurately framed the question - should people be able to choose a public health insurance option. The second poll (July and August) pushed them towards an answer by leaving out the essential question of choice and asking a yes or no question.

On Hardball earlier this evening, NBC's Chuck Todd claimed that they changed the wording because the word "choice" "biased" the question.

Todd didn't explain what is "biased" about describing a plan that offers people a choice between a public plan and private insurance as offering a choice between a public plan and private insurance.

Aside from the absurdity of describing the original question as "biased," it is important to note that the first question frames the topic of a public plan in terms of its effect on consumers -- it indicates that they would have a choice between a public plan and private insurance.  The new wording frames it in terms of the plan's effect on private insurance companies by emphasizing that they would face competition.  The new wording is only passingly about consumers.  

It should come as no surprise that a poll question that adopts the insurance companies' point of view yields results less favorable for a public plan than one that focuses on the impact on consumers.

UPDATE: NBC posts a response -- sort of:

NBC pollsters Peter Hart (D) and Bill McInturff (R) released the following statement: "The only agenda that we have is to accurately measure changes in public opinion. To that end, we selected two questions which we think are the best barometers of how and if attitudes on health care are changing in view of the robust public debate that is occurring."

Peter Hart and Bill McInturff have forgotten more about polling than most of us will ever learn -- but their response is, well, non-responsive. First, they make no effort at all to defend the new wording, or to explain why they think it is better than the old.  Second, if you're trying to measure "changes in public opinion," it helps to have a consistent question to track over time.

It should also be noted that NBC's Chuck Todd and Mark Murray misrepresent the wording change in their introduction to the Hart/McInturff statement:

Liberals and progressives -- including Speaker Nancy Pelosi's office and the group Health Care for America Now -- have raised questions why our poll measured whether Americans supported the "choice" of a public/government option in June, while in July and this month it removed "choice" and simply asked whether Americans favor or oppose the option.

The July and August polls didn't simply remove the word "choice," as Todd and Murray claim.  It completely changed the perspective of the question, as I explained above.  The original question focused on impact on the consumer; the new question ignores that in favor of a focus on impact on insurance companies.