In the various media coverage from left, right, and center, on the Patient Protection and Affordable Care Act (or Affordable Care Act, or PPACA, or ACA), many journalists have a tendency to misrepresent poll numbers. I thought I would highlight two not infrequent mistakes.
(1)Ignore the Difference Between Progressive Opposition and Conservative Opposition
The CNN/ORC polling on the ACA provides something few other polls seem to do: a disaggregation of the opposition into opposition from the right (ACA is “too liberal”) and opposition from the left (ACA is “not liberal enough”). That doesn’t stop media outlets from foregrounding the aggregated topline numbers in their reporting. Counting left-opposition as the same as right-opposition, however, gives numbers more favorable to Republicans than merited by a closer analysis.
CNN/ORC asks the following question first:
As you may know, a bill that makes major changes to the country's health care system became law in 2010. Based on what you have read or heard about that legislation, do you generally favor or generally oppose it?
And it then asks this as a follow-up, if appropriate:
(IF OPPOSE) Do you oppose that legislation because you think its approach toward health care is too liberal, or because you think it is not liberal enough?
In the most recent poll, just looking at the first question, 35% favored the law, 62% opposed it, and 3% had no opinion.
However, if you take the two questions together, you get a more nuanced picture: 35% favored the law, 15% opposed it from the left, 43% opposed it from the right, and 8% had no opinion.
Apparently, about 5% of the people said that they opposed the law but could not provide a reason. Those people are fools.
Let’s look at some demographic breakdowns.
Among men, 35% favored the law, 12% opposed it from the left, and 48% opposed it from the right.
Among women, 35% favored the law, 18% opposed it from the left, and 37% opposed it from the right.
Among whites, 27% favored the law, 11% opposed it from the left, and 55% opposed it from the right.
Among non-whites, 52% favored the law, 23% opposed it from the left, and 17% opposed it from the right.
Among 18-34 year-olds, 38% favored the law, 17% opposed it from the left, and 38% opposed it from the right.
Among 35-49 year-olds, 28% favored the law, 16% opposed it from the left, and 49% opposed it from the right.
Among 50-64 year-olds, 41% favored the law, 12% opposed it from the left, and 40% opposed it from the right.
Among seniors, 30% favored the law, 14% opposed it from the left, and 47% opposed it from the right.
Among Democrats, 59% favored the law, 19% opposed it from the left, and 13% opposed it from the right.
Among Independents, 33% favored the law, 13% opposed it from the left, and 46% opposed it from the right.
Among Republicans, 3% favored the law, 13% opposed it from the left, and 78% opposed it from the right.
There are more demographic breakdowns, but you can look to the results pdf for those.
The support, as well as the combined center and left contingent, has fallen since May, when I analyzed the results from a previous CNN/ORC poll. However, this change has not been even across demographics.
I decided to look at the change across demographics, looking at the combination of support and left-opposition vs. right-opposition. Movement to the left would stay within the former caucus whereas movement to the right would register differently, as departure from that caucus.
There was negligible change (sometimes even negligible change across each category) among the following groups: urban, non-white, college graduate, age 50-64, and Western.
There was very significant change among the following groups: rural, white, Republican, no college, and Midwestern. Republicans were effectively able to solidify their opposition over the course of the year.
The other demographic groups saw drops in support as well, but none so sharply as those.
(2) Conflate Polling Questions That Actually Mean Entirely Different Things
On Friday, WaPo blogger Greg Sargent looked at some recent polling about Republican opinions on health care.
Here is the first question discussed:
Do you approve or disapprove of the part in the 2010 health care law that provides financial help to low and moderate income Americans who don’t get health insurance through their jobs to help them purchase coverage?
When asked that, 56% of Republicans approved, and 40% disapproved.
Another poll asked respondents whether providing health coverage for the poor is the responsibility of the federal government. Republicans said "no" 70% to 25%.
Sargent attributed the huge gap between the two results to the use of the word "government." However, the questions are not asking the same thing at all. The first question refers to the subsidies (or tax credits) available to some people who buy health insurance on the exchanges. This is the voucher concept: the government gives you some money to facilitate a private purchase on the market. The government does not manage anything, and it only partially finances.
The second question, however, asks about provision of public benefits. This question relates more closely to Medicaid, even though Medicaid is still run at the state level (but with the help of federal funding).
If public goods and vouchers were the same thing, then liberals would have no problem with Paul Ryan's plan for Medicare. And liberals would see no difference between public education and school vouchers. That, thankfully, is not the case at all.