Via Political Wire, I learned of a new poll on attitudes toward the Affordable Care Act. The poll was released by a publication call The Morning Consult, with surveying done from September 25-28 by Survey Sampling International. I was not previously familiar with either of these entities. As best I can tell, the Consult is a business-oriented, center-right outlet. Its founder, Michael Ramlet, is described on this webpage as having done some work for Republican organizations and being affiliated with the bipartisan Purple Strategies group.
Regarding TMC/SSI's methodology, the poll's report claims that, "The interviews were conducted online... The data were post-stratified to match a target sample of registered voters based on age, race, gender, education and region." Post-stratification is a form of adjusting the amount of weight given to different demographic groups, to make the sample as a whole conform to known population parameters (e.g., from the U.S. Census).
According to the TMC/SSI report, 48% of respondents approve of "the health care legislation passed by Barack Obama and Congress in 2010" (21% strongly, 27% somewhat). The overall approval reported by TMC/SSI is thus about 10 percentage points higher than obtained by other recent polls. TMC/SSI provide extensive subgroup statistics, so we can look "under the hood" a bit at their recent poll. Partisan composition of the TMC/SSI poll certainly raises some questions:
519 Republican-leaning respondents (26.3% of sample, of whom 15% support ACA)
836 Democratic-leaning respondents (42.3% of sample, of whom 75% support it)
620 Independent respondents (31.4%, of sample, of whom 38% support it)
The TMC/SSI party-ID numbers are clearly out of whack with those of other survey firms. If one looks at HuffPost/Pollster's party ID averages (using the selection tool to limit the results to registered-voter samples, which is what TMC/SSI used), the percent of Americans identifying as Democrats is 37.9% and the percent identifying as Republicans is 33.7%, dramatically closer to each other than in TMC/SSI's partisan breakdown.
Relative party-ID proportions are not the only reason TMC/SSI reports higher approval for the ACA than do other polls. Even looking at self-identified party subgroups one at a time, TMC/SSI shows greater support for Obamacare in some instances than do other pollsters. Take, for example, a recent CNN/Opinion Research Corporation poll, which found 38% approval overall for the law. Approval from self-identified Democrats was virtually identical in the TMC/SSI (75%) and CNN/ORC (74%) polls. However, self-identified Republicans reported greater support to TMC/SSI (15%) than to CNN/ORC (7%). The same pattern occurred among self-described Independents: 38% approval in the TMC/SSI poll vs. 27% in the CNN/ORC survey.
Having said all of this, what first caught my eye about the TMC/SSI poll was response to the item, "What would you like to see Congress do when it comes to the health care law?" Response options were: "expand the law," "let the law take effect," "make changes to improve the law," "delay and defund the law," and "repeal the law." Among the full sample, a collective 67% endorsed the first three alternatives, all of which seem, at the least, accepting of the ACA (even the third category, though not representing a ringing endorsement, conveys a "mend it but don't end it" outlook). The two most anti-Obamacare categories -- delay/defund and outright repeal -- together comprised only 33% of the sample.
One way to put the apparent Democratic slant of the sample aside and escape the partisan lens is to focus directly on Independents. Among Independents, the combined percentage advocating delay/defund and repeal is only 37%. In contrast, 63% favored one of the supportive/accepting alternatives.
It would be interesting to see more pollsters offer the "make changes to improve the law" response option to their participants.