Views on a Public Option: Let the Fur Fly

Fur flew last week over public opinion on the proposal that’s emerged as the single most contentious point of debate in health care reform – a “public option,” or government-created alternative to private insurance. It’s an argument on which proponents of reform have been losing ground – and trying to claw it back.

Attacking polls usually is a sign of a worried politician, and that’s what we saw: A news release from House Speaker Nancy Pelosi jumped on an NBC poll question that found 43 percent support for a public option, criticizing its wording and analysis alike.

Pelosi cited other polls finding majority support for the proposal; she quoted Democratic pollster Mark Mellman as saying the idea “generates barely a ripple of controversy… When different pollsters, using different methods and different wording, all converge on the same answer, you can bet the public really does support a public option.” Mellman’s piece was titled, “Voters settle on a public plan.”

But there’s evidence to suggest otherwise. Public opinion on health care reform long has been highly malleable – as we reported back in June, it’s an issue on which pushback works. Given that reality, the variability of polling data this summer and the lack of specifics in reform itself, “settled” is about the last thing I’d call public opinion on this issue.

While we found 62 percent in favor of a public option in June, that dived to 37 percent if it would put many private insurers out of business because they couldn’t compete, as critics charge. In another question, 78 percent were concerned – 51 percent “very” concerned – about government bureaucracy in a reformed health care system. Even more were concerned about a potential impact on the deficit. And 58 percent disbelieved the administration’s assurance that they’d be able to keep their current plan if they wanted to.

How has this all played out in the hot summer of debate over health care reform? In our latest ABC/Post poll, released Friday, we found support for a public option down to 52 percent, a significant drop – led by a 17-point decline among independents. Others also have shown a decline; the Kaiser Family Foundation found support for a public option down from 68 percent in June to 59 percent in July, then remaining there this month.

Still, Kaiser’s 59 percent is more than our 52 percent – and other polls have ranged much more widely. How so?

There are a few possible reasons, many relating to the movable nature of attitudes on reform. One, Kaiser’s poll was completed Aug. 11; ours, a week later. Another, it describes a public option as “similar to Medicare,” and support for a public option in its data is 19 points higher among seniors (52 percent) than it is in ours (33 percent).

In July, a CBS News/New York Times poll similarly described a public option as “something like the Medicare coverage” seniors receive, and found 66 percent support. A Time magazine poll completed the same day didn’t mention Medicare, and found 56 percent support; a poll Pew poll two days earlier also didn’t mention Medicare, and found 52 percent support. The gap between seniors in the CBS/Times and Time magazine polls was 13 points; in the CBS/Times and Pew polls, a narrower 6 points.

Kaiser did a split-sample test in June and found no significant overall difference when it mentioned Medicare. It could be that an effect has arisen since. It also could be that entirely other factors are at play. Question structure can matter; some of these polls ask about a public option within a list of items while others ask it as a stand-alone question. Several also have different levels of undecideds, a function of polling technique; last week’s ABC/Post and NBC polls had nearly identical levels of opposition to a public option (46 and 47 percent, respectively), but 9 points of difference in support alongside 8 points of difference in indecision.

Then there’s question order. NBC asked about a public option immediately after asking people if they were concerned about health care reform “going too far.” Fifty-four percent said yes, which might have cued lower support for a government-administered plan. On the other hand, its July poll had a similar result with a different preceding question.

Polls have described a public option as “government-run” (Fox, 44 percent support in July), “government-sponsored” (Time, 56 percent support six days later), “administered” by the government (NBC, CBS/Times and Kaiser). Our ABC/Post question says it would be created by the government, but doesn’t specify how it would be run, since this isn’t set – and when we tested it in June, we found 2-1 preference for its being administered independently ("with federal funding and oversight") rather than "by a government agency." Kaiser, though, tested the phrase “government-administered,” in April, and again found no effect.

These approaches underscore the challenges in polling on health care reform; it’s tough to come up with wording that precisely portrays proposals that themselves haven’t been clearly defined.

There are, though, clear indicators of the problems facing reform proponents. Approval of President Obama’s handling of health care dropped from 57 percent in April to 46 percent in last week’s ABC/Post poll; disapproval rose from 29 percent to 50 percent. (Disapproval rose from 26 percent in April to 43 percent in July in Pew polls, and from 28 percent in April to 43 percent in July in AP polls.) In our poll, views that reform will “do more harm than good” rose 7 points from June. And when we asked people if they supported or opposed the proposed changes being developed by Congress and the Obama administration, it was 45-50 percent, with “strong” opponents outnumbering strong supporters by 13 points. (We rotated the words “Congress” and “the Obama administration” in this question; no effect.)

Pelosi did score one point in her complaint with NBC. Its analysis described its finding on a public option as “a shift from last month.” At 46 percent support in July vs. 43 percent in August, that change was within the poll’s margin of sampling error, as Pelosi rightly noted.

Another complaint, though, was far off base. Pelosi noted that an NBC/Wall Street Journal poll in June asked people “how important” they think it is “to give people a choice of both a public plan… and a private plan,” (76 percent “extremely” or “quite” important), while in July and August it asked people if they favor or oppose a public plan. The questions are very different ones, and the word “choice” in this context, without any counterbalancing term, is rife with positive attribute bias.

Pelosi’s handout accused NBC of “bad analysis, worse question.” In fact its question was a perfectly reasonable one – “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?” It appears that the speaker’s real complaint was not with the wording – but with the result.

Click here for a list of all the questions and results on a “public option” we’ve compiled for this memo.