Watch Out for Those Health Care Reform Polls
Atrios links to a Dave Dayen post that links to a report that says, "A new AP poll finds that Americans who think the law should have done more outnumber those who think the government should stay out of health care by 2-to-1." And the poll does say that. However, it doesn't say what "done more" means, and the actual numbers suggest that a more liberal law wouldn't have been any more popular than what we got. Here are the basic results from the AP poll: 30% favor the healthcare reform law, 40% oppose, and 30% aren't sure. Then they asked the 70% who were opposers and not surers a second question:
HC1A. Which of the following best expresses your view of the health care law that Congress passed last March?
I oppose most or all of the changes made by the law 28
I oppose a few of the changes made by the law 20
I favor most or all of the changes made by the law, but I think that law doesn’t do enough to improve the health care system 23
I oppose the law because I think the federal government should not be involved in health care at all 28
So 23% of that 70% thought the law didn't do enough. That's 16%. Add that to the 30% who favor the law and you get 46%.
Kevin goes on:
That leaves 54% who oppose all or most of the law. So you're still at 54%-46% opposed, and this is the best case since it's possible that making the law more liberal might also have turned some of the favorers into opposers. . . .
I don't think this argument holds up as support for Kevin's point which, again is:
. . . the actual numbers suggest that a more liberal law wouldn't have been any more popular than what we got.
Let's look at some other findings from the AP survey:
[ASK IF HC1 = FAVOR STRONGLY, FAVOR SOMEWHAT OR NEITHER FAVOR NOR OPPOSE]
HC1B. Do you think that the health care law passed last March by Congress should have done more to change the health care system, or do you not think that?
It should have done more 61
Do not think that 36
Since the “favors “and “not surers” total 60%, we see that 37% of the total responding to the survey considering these categories alone think that the HCR bill should have done more, which suggests they would have favored “a more liberal law.” Unfortunately, the survey didn't include the “opposers” of the bill when asking HC1B. They just assumed that people opposing the bill don't oppose it because they favor something more liberal. I don't know how large this group is, but I belong to it myself, and many of the Medicare for All advocates I know also oppose the HCR bill, and were opposed to it when it passed. How big is this sub-group of opposers?
What we know from the AP survey responses to HC1A, is that 48% of the 70% in the oppose/not sure total, or 34% of the total were either in the “I oppose most or all of the changes made by the law,” or the “I oppose a few of the changes made by the law.” Possible additions to the 37% group that would favor a more liberal law are most likely to come from this 34%, since respondents in the “I favor most or all of the changes made by the law, but I think that law doesn’t do enough to improve the health care system,” are much more likely to come from the “not surers” and to already be part of the 37% of the total, than they are to be part of the 34% in the above two categories who explicitly oppose the bill.
So, how much of the 34% is “progressive” opposition to the HCR bill? If we assume that it is 50%, that would bring the total of those favoring a more liberal bill up to 52%. But what if it were 80%? Then the total favoring “a more liberal law” climbs to 64%, Is this possibility far-fetched?
Well, first, note that HC1A responses also include the category “I oppose the law because I think the federal government should not be involved in health care at all.” Almost 20% of the sample selected this category. Meaning that 80% of respondents do not reject a federal role in the health care system, and that 80%, of course, must include the possible 64% favoring “a more liberal law.” However, from the responses to HC1B, we also know that those in the “favor”/”not sure” segment who don't think the HCR bill should have “done more” total almost 22%, leaving 58% of respondents who might favor “a more liberal law.”
So, it looks like Kevin Drum's conclusion that a maximum of 46% would have a favored a more liberal HCR bill is much too conservative. On the results of this particular survey the numbers may be as high as 58%, and there is reason to think that 58% is much closer to the truth than 46%. Here are the reasons. First, this survey was incompetently designed. Why weren't people asked, very simply, whether they would have preferred: 1) no change at all; 2) a bill that made fewer changes; 3) the bill that was passed; 4) a bill with a public option everyone was eligible to participate in; 5) a bill providing for Medicare for All; and 6) a bill providing for National Health Care. The responses to that would have settled the question of whether the public wanted “a more liberal bill,” or not. As it is, the questions deliver ambiguous and vague responses, and don't focus in on policy alternatives people know at least a little bit about. It's very hard to tell where opposition to the HCR bill comes from. Second, why decide to restrict HC1A to only part of the sample? Can one really assume that all the opposers would oppose a more liberal bill?
Next, Kevin's interpretation of the results of this survey is inconsistent with previous surveys on health care reform which show, roughly, that almost 2/3 of the public favor Medicare for All in health care reform. Previous survey evidence is reviewed by Kip Sullivan in this series. Kip's results are pretty unambiguous. The AP survey itself, is not necessarily inconsistent with his findings, provided the HC1B responses showing 61% favoring the idea that the Congress should have “done more,” can be projected across the rest of the sample. But the survey design made it difficult to evaluate that.
Finally, the AP survey is another example of one whose design is heavily biased towards the narrative that America is a center-right nation. The possibility that, at least on health reform, the American public is favorably oriented to Medicare for All, or other Government-based solutions couldn't be tested in this survey because of its bias. Designers bias surveys in their selection and construction of question, even more than they do in their reporting of their results. So, all HCR surveys need to be read with a very critical eye.